|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
198.54.202.234
cPlay The open source high-end audio player using ASIO May 2008 cPlay delivers high quality audio playback using ASIO 2. Playback is achieved using any ASIO compatible soundcard. cPlay is a minimalist audio player using the latest high quality SRC resampler (Best Sinc SNR 145.68db or 121.33db) or SoX (VHQ or HQ). cPlay's design offers state-of-the-art ASIO-only playback and caters for touch screen users. Installation, setup and use is easy. cPlay is built in c/c++ and operates on Windows XP SP2 Professional (32 bit).
FEATURES
- Resampling is sourced from LibSampleRate (version 0.1.5) and SoX 14.2.0 under GNU GPL license. LibSampleRate is aka SRC (Secret Rabbit Code) and supersedes the version as used in foobar2000. Best Sinc converter now offers a SNR of 145.68db or 121.33db (versus 97db). SoX VHQ offers better than 170db SNR. Resampling is bypassed when input rate matches output.
- Supports Steinberg's excellent ASIO 2 and is backward compatible to prior ASIO versions (as required by ASIO drivers).
- Offers high quality 64 bit double precision digital volume control (in 0.5db steps). This can be bypassed.
- Playback is achieved through .cue, .wav or .flac files. cPlay loads entire .wav or .flac (decoded) into RAM before starting. Playback is done directly from cPlay's internal buffer. Cue playback requires .cue files as created by EAC (single or multi file standard).
- Ensures efficient CPU resource utilization allowing for low specification processors or high levels of upsampling. This means CPUs can be underclocked / undervolted.
- Supports up to 3 ASIO soundcards with each having up to 100 output channels.
- Advanced optimizations are applied (if available from ASIO driver) during playback.
- Best results achieved when using cMP (i.e. cMP²). This allows for low level Windows optimizations. Use cMP release 1.0 final or later as this allows for bypassing RAM load in cMP (set RAM Load in cMP Settings to No) otherwise wav file is RAM loaded twice. cPlay allows for both svchost and lsass to be suspended during playback thus reducing the Windows footprint. Only exception is EMU's ASIO driver which requires both (svchost and lsass) to be operational. Set cMP's Optimize setting to Critical.
- Full remote control is achieved with cMP: offering volume control, track navigation, next/previous and stop/eject via (wireless) mouse.
USER MANUAL
Visit cMP² website (http://www.cicsMemoryPlayer.com) for more details and setup.
Screen Shot
GETTING STARTEDDownload cPlay's installer here (1.3MB). Installation and startup is straight forward.
If you don't have an ASIO compatible soundcard, use ASIO4ALL. Note that ASIO4ALL does not support channel mapping (use default) and rarely handles above 48k sample rate.
Your feedback will help guide cPlay's future development. Source code (4.1MB) is available via email.
Edits: 05/05/08 07/12/08 09/03/08 05/28/09 07/10/09 09/08/09 03/20/10Follow Ups:
Hi folks,
It appears cics Memory Player site is now fully back up:
http://www.cicsmemoryplayer.com/index.php
This means that I may in the future take down my copy of it at:
http://www.tasteuk1.webspace.virginmedia.com/cicsmemoryplayer/index.htm
(I could do with the web space back!)
However, for those who want an offline copy (which can be used in cases of no available internet) or to safeguard should the site go down again, I will leave the entire site archive zip file available still from here:
http://www.tasteuk1.webspace.virginmedia.com/cics/CICS_website.zip
I've been using cplay for long time. It's great. However there is no update for several years. And the home page and forum seems to have no more activity. As the cplay is open source, could any body have the source code of the latest version so that I can made some enhancement
(typically UI, e.g. the current UI window is non-sizable and not big
enough to show the song title if it is too long).
Thanks and regards
This is a link to an updated cMP shell made by Alexander, a Russian friend and cMP2 user, eight monhs ago.
It includes:
1. Complete keyboard support.
In the archive file there are some examples on the key commands for cMP.
It's particularly important for users with no mouse and PS/2 support.
2. Horizontal scroll in the list box.
3. No need for Winspool.drv.
Perhaps I should ask him if there have been any other developments, as he wanted to include support for [Date]
Perhaps I should ask him if there have been any other developments
I seemed to have missed that one - got it now. Many thanks for posting it.
D
http://db.audioasylum.com/mhtml/m.html?forum=pcaudio&n=103693&highlight=resource+riboge&r=&search_url=%2Fcgi%2Fsearch.mpl%3Fsearchtext%3Dresource%26b%3DAND%26topic%3D%26topics_only%3DN%26author%3Driboge%26date1%3D%26date2%3D%26slowmessage%3D%26sort%3Ddate%26sortOrder%3DDESC%26forum%3DALL
There actually is a way to change the window size and the font size so that more of titles will show. Search the thread as I try to recall what I did years ago. It involves modifying the resource files the application uses, but I haven't yet remembered how. Maybe someone younger around here can before I do.
Don't know who maintains CMP website but something happened and it is empty. I always go back here for ideas.
I am one of them who have been greatly supported by this site. I learned a lot here, and have been reguraly checking any updates. I hope the site will be back soon, or this is a sign of commencement of another new project.
Fear not my fellow enthusiasts. Last time this site went down, after being hacked, I went out of my way to create an archive of it (using cached files). I never bothered publishing the pages properly online because the issues got resolved and the site got restored. But as most of the pages are now blank I've posted my old archive pages for you onto my webspace.
On revisiting it some of the graphics look unreadable (displayed too small), so I've removed the resizing from all of them. To see a page graphics in its original size, scroll to the bottom and click the 'print' link. A new page will open set to original sizes. If the original site comes back up I'll check this problem out as I can't remember if some of the graphics were resized that small?
I am in no way trying to steal any credit for this work- hence the rather garish (in colour scheme) disclaimer page. If for some reason Cics wants me to remove it then I will comply. If the original site comes back up I will take it down again also.
In the meantime you can use my copy if it helps.
I could also zip up the entire site so you can put it on your own HDD to keep if this is useful (not posted that up though due to web space- if this would be useful too let me know).
The link to the archive is below:
Finding your kind message and your archive, I am very much relieved. I thank you very much for your help. Could you send the zip file of the entire site to me? This is just for in case of.
Thanks for the kind messages everyone. I hear you!
I'm putting up a zip file of the entire site. The coding structure is a bit different regarding the source pages. On the original site the pages were served by php server scripting. Apart from the search feature, which I have not tried to emulate, the entire site has been converted to pure html. All links have been made local (apart from the ones to external web sites). This means that you will be able to unpack them to any folder on your hard disk and they should all work. Just double click on index.html or point your browser there (or double click any htm file). index.htm is my front page explaining the archive, index.html redirects you to the Cicsmemory player front page (your one -not the original site).
The pages will display initially with the original shrunken down sizes for some of the graphics. If you find some too squashed and unreadable on your monitor (as I did), then go into the folder, you have unpacked the web site in, and run the file called 'size_large.bat'. This will rename the .LARGE.htm files and name the current ones .SMALL.htm (this will only happen on some of the files which need this doing). The graphics on both normal and 'print' pages will now have 'some' of the images much larger. To convert them back to small again run 'size_small.bat'. (Note you can still open up any htm file with the SMALL.htm or LARGE.htm name in them and they will still open up okay, there are just no hyperlinks to them from the other pages in the archive if that makes sense?
The archive includes all the executables and a couple of large PDF files (included on original site), which is why the file size is 41MB! I hope quite a few people download this so we have a plentiful supply should anyone need a copy as it looks like the original site has died a death or got hacked / corrupted?
If anyone can work out why the SMALL (i.e. original structure) pages graphics look too small on some of the screen shots please let me know? Maybe they looked like this originally or maybe I made some error converting the html? The graphics are large, but for some reason they seemed to have been scaled very small in some cases on the original site pages (from about 1400 width to 640!)... This is why I've added the extra html copies (with different scaling) and included the two batch file converters (mentioned above).
Good job I kept this archive... Took quite a while to compile it up. Maybe the original site will re-appear one day hopefully!
Link to zip file below:
Thank you very much. I have successfully downloaded and opend it. It works nicely, and the site is back for the first time in three months. I greatly appreciate your effort and concern.
Thankyou!
Hi folks,
I know Theo was wondering if this is possible and it is (see title).
If you want to see a guide on how to make your own batch file (without myself needing to distribute minlogon.exe or the completed batch file containing the minlogon.exe code) then just click on the link below.
Thanks.
All the best,
Mark
Hi folks,
While trawling through the cPlay archives I stumbled on a recommended way to install Minlogon, cics suggesting a post by Seger:
http://www.audioasylum.com/forums/pcaudio/messages/4/40869.html
There is a link on Seger's post for the Minlogon download:
http://thuun.boot-land.net/index.html
Now if you download it and open up the cabinet file you see 'minlogon.exe' 24384 bytes. This is exactly the same version that Microsoft distributes from here:
http://www.microsoft.com/downloads/details.aspx?FamilyID=9bdf1dea-a37e-4d25-83df-aabbaa78914f&displaylang=en
However, if you look inside the boot-land cab file (use Winrar) there's another cab file called 'minlogon209170.cab'. If you extract / open up this cabinet file there's another 'minlogon.exe' that's only 17408 bytes in size!
The question is, what is this version? Does it still work and is it better for being smaller? Has anyone else tried it...?
Now onto the other part of my post- the 'naughty script'!
I've written another batch file. With this you run it, within 5 seconds it installs Minlogon, disables WFP (including the registry changes needed for both files). Then after a reboot all is ready, it will then boot to SYSTEM account and WFP will be turned off. No extra files are needed- not even Minlogon itself! This is because it is actually encoded into the batch file text, then the file gets created on the fly. Again, it uses Debug and Reg to do its dirty work (these are built into XP, so unless you've deleted them all it should work). Nothing needs to be installed, no temp files should be left behind. Once the batch file completes, you can delete it as it's not needed afterwards. XP Service Pack 2 or 3 only.
Now the dilemma (and why I say 'naughty'). I'd love to post this here to help others, it's all printable text so you could copy and paste it again into a notepad and save as minlogon.bat (or maybe a link to it would be better, as it's quite big at just under 100k). BUT, I guess as it effectively creates a M$ file that I'd be breaching some kind of legal rule book by doing that? It doesn't seem to stop other people distributing it though, like the bootland site (link above).
Any ideas about this? Anyone want to try the script out?
I just checked my minlogon and it is the 17kb version.
Thanks for the info Theo, I might have to try this other version to see what differences (if any) there are...
Interestingly enough, if you hover over the files (or right click and choose properties, then choose version tab- then choose 'file version'), you get the following information (larger file first):
File Version: 2.0.1024.0
Size: 23.8 KB
File Version 2.0.917.0
Size 17.0 KB
So the 17KB one is an earlier file. Makes me wonder why the later one is nearly 7KB larger? Could it be that the smaller one uses less resources when starting the system, or loads up less of Windows? I wonder if these are the only two versions around, there may well be dozens of different builds of Minlogon file then? I'm assuming yours is the 2.0.917.0 file too then?
Just as an aside, I can alter the batch script easily to install the 17KB version, if this proves to be the best one to use.
Hi fellow cMPers!
Have to say this is quite an amazing forum. Been trying to read every post here but taking a very long time. Fascinating though how cPlay has developed and evolved over time, especially the introduction of gapless playback and the teething problems which were sorted along the way.
My name is Mark. I would like to write some software over time which may help others using cMP2. Either utilities or ways of automating set-up. Thinking of using Auto-it for most of it. Why that language you may ask? Well it's free, no restrictions on license, creates stand alone executables (ok if you've stripped every single dll out then I can't guarantee these projects will all work? After-all, Windows applications are all basically a load of calls to various dll modules). Also Auto-it is powerful enough to manipulate other Windows software (like clicking a button for example). I am very wary of sound quality issues software could create, so these will mainly be 'sleeping giants' so to speak. Most of the time you probably won't run them, unless you need their function, so it won't get in the way of that sublime cMP2 sound.
As an example, Serge (steppe) found a way of stripping out resources from cicsMemoryplayer.exe creating the infamous 1.exe! This had an improved sound and removed the need for winspool.drv module. The unfortunate side-effect is that we lost our options page. Well I've managed to re-create this in an Auto-it gui. This directly writes to the mysterious cicsMemoryPlayer.ini file. If you want to know the structure of the ini, pretty much worked it out. It is in binary so you can't edit it in a normal text editor. The options can be adjusted outside of cMP (XP mode) too if you wish with this gui program. Still need to finish all this off, but if anyone is interested or wants to know how the ini file works please let me know.
Another thing I intend to do is automate some of the reduction methods (software strip-down) others have come up with. Doing this manually proved horrendous! The pc can do much of this for you with the right script ;). Finally was working on a little script to do all the cMP setup. Initially this did mainly registry changes, but I was always left disappointed with the end sonic results. So the new version actually stripped down XP, as if you were clicking all the gui buttons yourself, including the re-boots! Took about 20 minutes but you could just leave it going... So maybe another project there for the future (still some bugs to iron out there).
Anyway, a final taster- and on to the topic of this post at last!
Disabling WFP (Windows File Protection) has always been a bit of a pain. I've always thought that a nice little batch (.bat) file would be the best way, if possible? So spent a while trying to research it. Well hopefully the solution is below!
Okay here's what you need to do:
1. Open up notepad.exe
2. Copy and paste the script below- everything inside (in between) the two - - - - - lines (but not the - - - themselves!) into the notepad window
3. Choose Save-As from the notepad file window. Make sure you change save as type to 'All files' as we need to save this as '.bat' and NOT '.txt'
4. (Choose a folder you can find again.) Type in the Filename as: Disablewfp.bat and click 'Save'
Then you can either double click this file to run it, or preferably run this from a command (cmd) prompt so you can see the visual feed-back. (You can add a 'pause' line to the bottom of the script if you want to use the double-click method though).
What this will do is disable WFP permanently after the next reboot. Okay, we've all done this one way or another, but this is an easy way and might help someone new to setting up a cMP machine and configuring it?
The script uses two command line programs which are built into XP. reg.exe and debug.exe. Script should work okay, but I strongly recommend you run this on a clean machine (with nothing to lose!) to test. If you want to know how it does it, I can explain all (big believer in supplying source code to projects, so others can edit or improve on the work).
Just a final note:
This is for Windows XP with Service Pack 2 or Service Pack 3 only. It will exit without modifying anything, if run on any other versions. This has not been tested on anything other than Professional, though it should work if sfc_os.dll is the same file on those versions- proceed with caution! If your XP is a different language to English, the checking part of the file (at the start) may need to be edited, as it's looking for English text (like 'Service Pack 2' for example). Please let me know if you find any issues.
Hope this is useful for somebody anyway. I apologise if somebody else has already done this in the past, but not got to that part of the forum yet as starting at the beginning working towards the present (and reading only the cPlay forum currently).
All the best,
Mark
Copy the text below to create the bat file:
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
@echo off
:: Disablewfp.bat written by taste in September 2013. Freely distributable
:: Function: To disable Windows File Protection system for use in a cMP environment
:: Only for Windows XP Professional SP2/3. Not tested on XP Home or other XP versions!
:: Creates backup copy of %windir%\System32\sfc_os.dll stored as: sfc_os.dll.bak
:: Patches sfc_os.dll so this registry key\value will operate correctly:
:: [HKLM\SOFTWARE\Microsoft\Windows NT\CurrentVersion\Winlogon]
:: "SFCDisable"=dword:ffffff9d -Disable WFP, "SFCDisable"=dword:00000000 -Enable WFP
:: Reboot required to take effect when setting above value, initially set to Disabled
:: Use at your own risk. No liability for loss or damages accepted!
setlocal
for /f "tokens=*" %%a in (
'reg query "HKLM\SOFTWARE\Microsoft\Windows NT\CurrentVersion" /v ProductName'
) do (
set productname=%%a
)
if not "%productname:~-10%"=="Windows XP" goto incompatible
for /f "tokens=*" %%a in (
'reg query "HKLM\SOFTWARE\Microsoft\Windows NT\CurrentVersion" /v CSDVersion'
) do (
set servicepack=%%a
)
if "%servicepack:~-14%"=="Service Pack 2" (
set patchaddress=EDE9
goto startprocess
)
if "%servicepack:~-14%"=="Service Pack 3" (
set patchaddress=ED89
goto startprocess
)
goto incompatible
:startprocess
if not exist %windir%\System32\sfc_os.dll.bak (
copy /b %windir%\System32\sfc_os.dll /b %windir%\System32\sfc_os.dll.bak /y
)
copy /b %windir%\System32\sfc_os.dll /b %windir%\System32\sfc_os.bin /y
echo E cs:%patchaddress% 90 90 90 > %temp%\disablewfp.tmp
echo W > > %temp%\disablewfp.tmp
echo Q > > %temp%\disablewfp.tmp
type %temp%\disablewfp.tmp | debug %windir%\System32\sfc_os.bin
del /q %temp%\disablewfp.tmp
reg add "HKLM\SOFTWARE\Microsoft\Windows NT\CurrentVersion\Winlogon" ^
/v SfcQuota /t REG_DWORD /d 0x00000000 /f
reg add "HKLM\SOFTWARE\Microsoft\Windows NT\CurrentVersion\Winlogon" ^
/v SFCDisable /t REG_DWORD /d 0xffffff9d /f
reg add "HKLM\SOFTWARE\Microsoft\Windows NT\CurrentVersion\Winlogon" ^
/v SfCScan /t REG_DWORD /d 0x00000000 /f
reg add "HKLM\SOFTWARE\Microsoft\Windows NT\CurrentVersion\Winlogon" ^
/v SFCDllCacheDir /t REG_SZ /d "%windir%\system32\DLLCache" /f
reg add "HKLM\SOFTWARE\Microsoft\Windows NT\CurrentVersion\Winlogon" ^
/v SFCShowProgress /t REG_DWORD /d 0x00000001 /f
reg add "HKLM\SYSTEM\CurrentControlSet\Control\Session Manager" ^
/v AllowProtectedRenames /t REG_DWORD /d 0x00000001 /f
reg add "HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Control\Session Manager" ^
/v PendingFileRenameOperations /t REG_MULTI_SZ /d ^
"\??\%windir%\System32\sfc_os.bin\0!\??\%windir%\System32\sfc_os.dll" /f
echo.
echo DONE. Please reboot computer now for WFP to be disabled permanently
echo Do not try renaming/deleting any files in System32 folder before the reboot!
goto cleanup
:incompatible
echo.
echo Aborting- This is only for Windows XP with Service Pack 2 Or Service Pack 3
:cleanup
endlocal
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
I have source files for both cMP and cPlay (latest version) in case you need them. There are a few features that could be implemented in either of them.
Yes please Bibo, I'd love to see the source code, mainly to see how cics implements some of the Windows trimming and footprint reduction on cMP.
Does the cPlay source have the four versions, or is this just a compiler directive to create the sse2, sse3, ssse3 and sse4 varieties?
I'm assuming this is in C or similar language?
I can send you a PM with my email address.
Thanks.
All the best,
Mark
Please, send me code cmp and cplay.
I love cplay 2.36 sse2
Edits: 09/30/13
I will send code to both of you. I just have to put my backup back online.
Oleg,
if you are interested in AMD optimized cPlay, I have cPlay_2_0b7 which is the last AMD SSE2 optimized version.
Thanks again Bibo01.
I also will try the AMD optimized version (I have an Athlon 64 processor). Hope the features of 0b39 are mainly present (at least gapless playback)?
Many thanks,
Mark
Thank you Bibo01! An important goal is to transfer control to the keyboard cmp (very important to turn off PC). I will take you to any (all) of the information
and then you can get rid of lots more useless stuff.
Hi Rick,
Yes I agree, Nlite sounds very interesting. I've tinkered with it in the past but never quite worked out how to create the best stripped down image. One time I removed so much stuff, that once the image was installed, it just seem to sit there with a blank background screen and nothing else :(
Only problem is obviously a stripped down image file containing Windows could not be distributed directly. I'd be very interested if you've got some set-up scripts for Nlite configuration though?
This forum is huge so my apologies if this has all been done to conclusion in the past...
All the best,
Mark
and what you expect and I would be glad to give you my advice.
One can strip almost everything with nLite. And once you get everything running you can strip out A WHOLE LOT MORE.
But it depends on how much of a purist you are and what you are using for a sound card. At the moment I am using the WAVE-IO usb card instead of JULI@.
Well I have tried a full specked cMP machine, pretty much what cics recommended, apart from I had 2 fanless PSUs. Now it may have been to do with the PSUs, but I could never get it to match the sound of an AMD machine I had used previously.
The AMD one just outputs to a USB dac, Russ Andrews DAC 1, using AQVOX USB ASIO drivers. This is the machine I still use.
It could have been that the DACs input was more geared up for USB than COAX / Optical (has those inputs too). I also had some issues with the Juli@ and that awful breakout cable, the weight of this cable used to sometimes cause a bad connection, resulting in a sound that was a bit like Theo's distortion / metallic sound problem I suspect?
So the AMD PC is the one I still use. Even after preparing the recommended cics board and using it's USB output, I could still not match the AMD board for some reason, no idea why though! I still have both machines available to experiment with.
So the build I would need would contain no network, would use internal graphics card, no sound card, USB output to dac and I use a wireless USB keyboard with a adaptor that plugs into a USB port. Also I seem to get better results with XP SP3 than SP2. AMD unit I have can only use the first sse2 version of CPlay due to it being Athlon 64 CPU.
I do use my own software to do the software modifications, but I get variable results. Occasionally it really sings, often doesn't though. The sound also seems to 'go off' and sound a bit sloppy and vague after 5 days or so, not sure why either!?
My amps are about 12 years old though now so probably all needing a service, I think that this is the main reason the sound isn't fulfilling most of the time (using Naim amplification, bi-amping, this system can be very revealing, which has its good and bad points).
Thanks for your assistance. Please let me know if you want any more specifics about my system. I can probably find the model / type number information of the PC components if that helps...
All the best,
Mark
While this is remarkable work I must say this approach would be invaluable for migrating to minlogon. I did the hard way 2 times before I got it right...had to reload windows and reset everything.Well done!
Edits: 09/17/13
Hi Theo,
I believe this is something you wanted all in one script- minlogon install and WFP disable without the need of additional files or manually setting up.
I have now managed to come up with a batch script that both installs minlogon and does the WFP disable all in one go. Just needs a reboot afterwards. Unfortunately it will need building as below, as I am wary of giving out anything directly that creates the file.
TO CREATE THE BATCH SCRIPT: Just place my createminlogonbatch.exe file in the same folder as a minlogon.exe file (it must have the name minlogon.exe NOT renamed to winlogon.exe). Run it and it will create the batch file in the same folder.
After this you can just store the batch file and forget about the other files (minlogon.exe gets 'built in' to the batch file and will be created just by running the batch file).
Run this and it will install minlogon and do the WFP disable- including all registry patches needed. Then re-boot.
See the link below for where to get the files (all you'll need is the createminlogonbatch.exe file as you have minlogon.exe already- just see step 3 and 4).
While I appreciate your effort, for me it is unnecessary. I am already there plus I have many copies of my OS on Snapshot files which always allows me to recover after an attempt to 'process improve' my OS goes awry. For newbies though this is a very useful thing to have.
Yes, I thought you'd be sorted regarding this, no worries.But you can take the credit for coming up with the idea. Only ever thought about just making a script to disable WFP. But this one is much better.
For anyone tinkering with their cMP builds though, I suggest making yourself a copy of 'The Ultimate Boot CD for Windows'. [Not my work this time.] You just need an XP installation disk and the software from their webpage (link below). It builds an ISO which can be burnt as a bootable CD.
Whenever any tweeks cause the cMP machine to not boot up, and safe-mode fails, boot to this disk (takes about 5 minutes) then you can rename files and even edit the registry without needing a working hard disk image. Saved me from re-installing many times while developing tests. Also contains DriveImage XML (I think it's called) for making / restoring partition backups.
Just one caution though: You need to have your PC set to IDE mode as AHCI will cause a blue screen on booting to the CD.
Edits: 10/07/13
Anything that eases the task of getting the various optimisations under way for cMP has got to be good news. A set of scripts that ensures that the right steps are taken in the right order is something that I for one would love to have. Just one suggestion. It wouldn't be a bad thing to have complimentary scripts to undo the optimisations. Just a precaution!
Not directly connected with the sonic optimisation, but something that I think is badly needed. Has anybody figured-out how to have the monitor switch off until it's re-called by mouse action. I find it a big distraction to have the monitor constantly glaring at me.
You can minimize the cplay window for the monitor to minimize the effect. Also there were some posts by jack wrong who showed how to reset display colors to simulate a totally black screen. This also improved sonics by minimizing emi/ rfi.
Edits: 09/17/13
Thanks for these pointers. It would be nice, though, if the screen blanked during inactivity and then woke-up at the touch of the mouse.
Thanks Theob and Andy222 for your positive feedback! I'll discuss a few of your comments below.
Andy, I've been looking into your 'screen always on' dilema. It started out as a rather mind-boggling problem. Even if I set power management to 'turn off monitor' after one minute, it stayed on all the time. At first I thought it may be to do with disabled services so I re-started a few (of what I thought might be) relevant ones back up again. No luck!Then I started looking at Device Manager and re-enabling some system devices, still no joy. Even in Windows XP mode the power management still would not turn off the display! I even tried sending commands to user32.dll and still the display remained lit.
I tried, on one of my spare PCs, a fresh XP build with no drivers installed and this is what I've discovered. It's all to do with us 'disabling' / not having display drivers installed for the graphics card in use. With no driver (or indeed with or without if the device is disabled) the system call to turn off the monitor is effectively ignored, Windows cannot communicate that command to the hardware.
So here are your options:
1. You could re-enable and re-install all the graphic card display drivers. Then set up power management to turn off the display on inactivity. Now, if you wanted a shorter timer, like say 10 seconds, I can find you a way of achieving this (usually one minute is the quickest time as an option). But the downside may be that the sound quality suffers by having the graphics devices installed? You'd have to test it and see?
2. Look into Jack Wong's method. I haven't seen these pages so have no idea what Mr Wong did. But if it was changing palate colours that may be a great solution, without adding any extra system load which may degrade sound?
3. I am in the process of writing a solution for you. I'd be interested if you wanted to try it. It will be an application that starts up when you boot and runs in the background. It will mainly wile away doing nothing so as (hopefully) not to add load. This will (every 1/4 of a second by default) look at the 'inactivity timer' in Windows (Kernel32.dll). After 10 seconds of inactivity the screen will appear to go black. What actually will happen is a window with a black background will cover you entire display so making it look like the screen has switched off. Bear in mind that the backlight will still actually be lit though as there seems no-way (without display drivers) of turning this off (I could write similar software to do this otherwise, or leave it to 'power management'). On moving mouse or touching keyboard display will re-appear as normal (for a key press, first press will make screen come on, 2nd will issue the command to cMP/CPlay).
Now the downside to no. 3. This may or may not effect sound quality with the program running, though it has been made to be low on resource use. Something you'd have to assess maybe? Last thing I'd want to do is upset the sound in any way! Maybe it won't?
If you want to give it a go, I'll make sure it will have a method of clean install/uninstall and it will activate in cMP mode, in XP mode will exit, as otherwise will drive you mad going blank so quickly when you're adjusting something! Of course 10 seconds is just an idea of how long I'd have it lit for, if it was for me. Could do it for any length of time you like, also the amount of 'inactivity checks' per second could be adjusted. Bear in mind that if it's not checking often, then when you move the mouse it may appear sluggish to bring the screen 'back to life'?
Theo, the 'disabling WFP' batch file could easily be modified to do the minlogon swap (though for licensing reasons the user would have to obtain 'minlogon' file as that can't be distributed legally by one of us). I could update this if you want? But the minlogon file would need placing in the same folder as the batch say (or in System32 folder as another example), then the script could be made to do the swap.
Andy, On your point of making automated cMP set-up scripts, the software I put together does do this but before releasing, I'd have to polish it up and make it more functional.
The problem is two parts:
1. A 'one item fits all' dilemma! Some people are using juli@, some USB dacs, some others ASIO sound cards etc. All these would need to be accommodated for. Some have system boards which blue screen if some devices are disabled, others are fine. Then there's the people who use networks, the people who don't etc. I think we could safely create something where it works with XP or nothing, so to speak. As it seems the general consensus that XP sounds better in a fully stripped down set-up?
2. Some of the files needed to achieve the results (e.g. the command utility that disables devices) are freely obtainable, but are Microsoft products, so cannot be directly distributed with the program / script files. These could though be maybe automatically downloaded by the setup software, assuming that user's pc is set up to access a network and the links to the items remained consistent. My cMP machine though was never put on the network, to keep it as clean as possible, so this would need to happen on another PC which was!
Anyway, both those points I think are pretty minor, and it would be good for something to automate the, let's face it, tedious process of all the Windows adjustments that need to be made. I think last time I did it manually, it took something like about 4 hours of grief lol.
Just a question too... Does anybody find that the sound of their setup goes off after a few days using cMP/CPlay (this includes re-booting / powering on / off over that period of time)? I find after about 5 days a rebuild / set-up 'freshens' the sound up again (starts sounding a bit loose in timing / grainy). May be my set-up, or something to do with using USB? Also, think my amps are needing a service desperately, so may all be to do with that? I'm using the AQVOX ASIO usb drivers. Not cheap, but they sound amazing! There's a free trial on their site for anyone who wants to give them a go (you just have to put up with it making an annoying 'beep' once every minute while in trial-mode, goes if you purchase license key).Anyway, thanks again for the feedback. I'm a bit of a hobbyist on the programming side, but find it so enjoyable none the less. Is the cMP2 project still 'open source' as I've never seen the source code to any of this? Wonder where you get it from? Looks like cics hasn't been posting here for a long time too... Still, his (or is it her?) legacy lives on! This forum and the cMP one is huge and still growing!
All the best,
Mark
Edits: 09/19/13 09/19/13 09/19/13 09/19/13
Thanks for the explanation for the monitor not turning off. I am slightly reluctant to get the facility back by re-enabling the display drivers because I fear for the sonic impact. Minimising cPLAY and playing with the palate colours is an option, as Theob has highlighted, but a big problem for me is that minimising only works once following the launch of a file. Once the window is maximised again (to take a quick peek at the track listing, perhaps) it stays that way until the file is closed and re-opened.
Your possible solution sounds pretty interesting and I’ll be more than happy to try it. Just as soon as you’re ready.
I fully appreciate the difficulty that differing hardware poses for automated set-up scripts. I myself have deviated a little from the “classic” cMP hardware. I started with the juli@ sound card, but I’ve moved across to asynchronous USB because I thought it sounded a little nicer.
If you do get around to producing some scripts, please count me in as a willing body for trials.
Hello All!
Is this anything the communities sound quality could benefit from?
It has shielding, ELNA caps and such.. all on the mother bard..
http://www.asus.com/Motherboards/MAXIMUS_VI_FORMULA/
Gaston
With apologies to Varese ...
Having been without any digital since my third JULI@ died and having no desire to buy another one I started reading so much good about Lorien's WaveIO (not to mention Jack Wong saying he had got one) I decided to get it and see what could be done with a minimal XP Pro install.
Jack's ISO was missing too much USB stuff to get this to work and as much as I love Jack and his brain I remain slightly doubtful there is that much difference between XP and HOME when you get it down to almost nothing.
I am in the midst of assembling Doede Douma's DDDAC1794 so all of my minimizing has been done silently. I have no idea how it sounds! My premise remains that the smaller the OS the better so I see no need to listen to it along the way.
One can get rid of most everything we could get rid of before. Of course, there are the USB entries and since USB has dependency on PCI there is still much of that remaining. The registry ends up being larger. At this point I think I have removed all that I can. I am at 44kB.
I did find, and have no idea if it is something we missed the "first time", that isapnp.sys can be removed from the drivers folder and all mentions of it in the registry. Not enough to balance out all of the USB related stuff that must remain.
In deference to Serge I used NTFS for this install. He feels strongly it is superior to FAT. The NTFS.sys is significantly larger than fastfat.sys which leads to a larger OS "size".
So my windows folder is 17.2 mB mainly because many of Jack's special .dlls and .syss will not work with USB or something. Luckily the micro-shell works just fine.
Into the project I was contacted by Dave Bruce who has been working with WaveIO since the spring with a minimized windows install and he is very pleased with what he is hearing.
SO, if anyone is still looking at this thread I hope this will inspire others to use cMP within the USB "craze" and due to devices like WaveIO it is most likely the way to go.
Hope to have my DAC working this weekend and be able to tell how good it is. Too long ago to be able to make comparisons to JULI@/cMP but hard to beat the LP reference for judgement.
Hi Rick
Looking forward to hearing your sound impressions. I was one of the early adopters of WaveIO... first on CMP, and then on MPDPUP. I continue to be very happy with WaveIO. Run it with good linear power supply.
I really love my WaveIO driving a TDA1541A with tube output. Next step is to rebuild the DAC, WaveIO, and ALIX into an integrated music server to keep signal wires short. All power supplies fully tricked out with salas shunts on the digital and all film cap and choke dual mono tube section. All electric parts collected and now building the chassis. keeping a build blog at http://www.diyaudio.com/forums/blogs/wlowes/
We are still listening :)
Only reason I will switch from cmp^2 is for dsd playback capability. If cics or Serge ever updated cplay for dsd playback I would be done.
You will have to have two set-ups.
Guys, I have been worrying a lot about about some day in the future when I won't be able to run my cmp..even if it is 10 years from now. I assume the only cause would be that pci asio sound cards or motherboards will fail and new ones won't support xp. Is this a real problem and if so, how many years away is that?
Can you send some links to other forums where the other main contributors went so I can see what they are tinkering with now? I'm sure I won't defect, mostly because I don't have the time. I just don't want to be lost someday and not be able to find the people I already trust and have all the expertise I've benefited from (leeching is another word)over the past six years or so. You just can't find a new forum and believe someone who says, "This is the best computer based audio set up I've ever heard".
I also have a bunch of dsd albums but have yet to make a cmp install with the only software I have to play it. jriver. It requires directx, which I removed with XPlite. I know you've all got your xp installs really small, and used other applications to get there, but after doing xplite I was only able to do some slimming. I'm sure directx would be a big no no. For those doing dsd, is there slimmer software other than jriver I can use until the software theob has been asking for magically appears? I'm also a bit sad not to have something that can recognize the sample rate and set the output to match.
Lots of questions in here that may not have an answer. Just thought I'd post while those of you that are left happen to be looking.
Mike
I understand that there is a juli@ pci-e card available that is very similar if not identical functionally to the juli@ pci card. So don't worry . I also hope there is some white knight out there that can adapt cplay to dsd.
Edits: 08/10/13
Anybody try to substitute a dsd capable player into cmp? I doubt it will work with uber slimmed os's but there seems to be more and more dsd stuff available. So it would be interesting to try. Better yet dsd capability with cplay would be great.
Just a shot in the dark question.
T, it is not a shot in the dark, or at least you are not alone. I am beginning to worry about this also.
Well, almost.
Hello, everybody. It is finished. It is in pdf format and will be uploaded shortly. Some inmates are already getting it.
Serge.
Any updates on this, steppe? Been a while, nothing's been posted.
I just discovered this project -- and am ready to start building. However, it seems that many of the suggested components are discontinued (e.g., that beautiful Zalman HD160XT home-theatre PC case). I have build a couple of PCs... but would like some direction as to how to build a like-sounding system in 2013... keeping to the same under $1000 price. Can someone please help me? Thanks so much!
I think it was wrong to get one. Its choice was explained by the idea of using a touch-screen display to control the server, this idea, though elegant, died, so don't pay such money for nothing.
Hello, and welcome to cplay-cmp.
First, go to corresponding site and download all materials on building,
next - hardware - I am sure many inmates will answer as for current hardware., but find here on AA the cmp thread - it is mostly on hardware.
All I can eagerly recommend is a simplest gigabyte micro itx mobo, that accomodates the Core I3 5** processor of the second (or is it third) generation. These processors, though faster, use less power and coldstart from lower voltages - both are critical.
Get Yourself such memory - Kingston HyperX KHX1333C9D3B1K2/4GB - it is a lit of two 2Gb modules, that unlike the recommendations, sound best of several I have tried, though they are 2 GB. It turns, it doesn't matter much - the size, that is.
Get Yourself the SIMPLEST and OLDEST pci express videocard you can get among the used video cards. It won't be more than 20 dollars (usually less.
The perfect example is Asus Nvidia g7300 le or any other based on 7300 chipset - passive cooling, no need of drivers, absolutely stable and perfect for our purpose)
See the upcoming tweak post, I shall post it shortly. There You will learn more.
Try to get a English only version of xp 32 bit, preferably the vl version - less troubles when slimming down.
Get a good power supply - the recommended Zalman Heatpipe is very good, but Thermaltake is better - They had a now oop fanless version for 350w - more than enough. The main thing is to have a psu with active cable changing capability or whatever they call it - it has receptacles for cables, that You can plug and unplug as needed. it is easier, and cleaner inside the case this way and after installing windows and settling on Your final configuration, You will want to unsolder a lot inside even such a psu.
Very important and convenient thing is to get an EXTERNAL box for hdd that has both usb AND ESATA-SATA connectivity. - Saves a lot of trouble when slimming and allows to externally power Your HDD. I use Thermaltake Silver River II and cannot be happier - about 35 usd.....
Decide on the capacity of your hdd. If You are going to have less than a thousand cds on it, get a 500 gb Seagate Barracuda ES.2 - it is expensive, but You don't want to waste your ripping efforts. Rip only to cue and image.wav, no ripping multiple wavs for every song - just an album image (to be explained).
Ask often, inmates will help.
Serge.
Get a simple mobo with a PS/2 port, or, even better with two - such modern mobos still exist and it will save You alot of trouble when tweaking after You disable the hateful USB inside BIOS. Get the best pci sound card possible. Try to get a used Lynx l22. It, the hdd and a PSU will be Your biggest investments.
Edits: 03/16/13 03/16/13
Omg you had my heart thumping...I thought this was the big one but alas it is not. Anxiously awaiting your next post. You ought to be in advertising Serge. You know how to deliver.
I do not resample when I listen to Cplay but because I used to upsample with SRC that box is checked on Cplay settings. When I checked the SOX box (but still not up sampling) and Alias box (which allows aliasing)the sound changed for the better. More extension on the highs. Since I still do not upsample I am wondering if I am getting aliasing. What does Aliasing mean? If I had to speculate aliasing means no digital filtering. No digital filtering means no brick wall filter effects. Can anybody confirm this or said another way, why do the sonics change? BTW try it I think you might like it...whatever it does.
Edits: 03/08/13 03/09/13
Hello All,
Just received the Juli@ PCI card yesteday and installed it. There is audio output using the ASIO4All driver. With it, I can see the 1/2 and 3/4 output seections as well in cPlay settings. However, cPlay does not see a juli@ ASIO driver from ESI software CD.
Let me preface by saying, the juli@ card took 3 weeks to arrive, so I began using cPlay and cMP with the on-board sound card. It sounded very nice. The Juli@ with the ASIO4All driver sounded much better. Anyway, before installing the Juli@ drivers on first boot, I went into the BIOS and shutdown the integrated sound card, then removed the ASIO4All driver.
XP Pro 64-bit found the Juli@ card, then proceeded to manually install the two drivers: Juli@ Controller and Juli@ Audio (I think it was Juli@ Audio.) I did not install any other software to avoid anything unnecessary from going into the system. Device Manager show no warning of missing drivers.
Reading through this thread, I still need to check if the Juli@ card has its own dedicated IRQ. Is there anything else I can check to see why cPlay is not finding the Juli@ ASIO driver? Maybe the juli@ ASIO driver needs to be dropped in a folder somewhere? Are there any other software requirements?
Thanks in advance,
Vince
try toggling between 1/2 and 3/4...one id digital one is analog.
Theob,
Thank you for the reply. I will try it tonight. Didn't see this message earlier. Email notification was off.
Do you have any recommendations on moving the sound card IRQ?
Juli@ shares IRQ 19 with the SATA controller. The Gigabyte MB only has 1 PCIe and 1 PCI slot. The card is PCI. Had to wait a month for the PCIe Juli@ card.
While using the on-board sound card, there was break-up in the audio playback. It doesn't happen with the Juli@ card, but that doesn't mean the SATA controller is not adding noise to the bus. Might try to move IRQ via BIOS, but read the OS takes over anyway, regardless of the BIOS setting. Might also try moving the SSD SATA drives to another connector, but that might not work because of boot requirements. The BIOS had some adjustments, so it may work. If it works, I can shutoff the offending SATA controller.
Might look at a more "advanced" motherboard. The Gigabyte board I'm using allows for under-clocking, but it's missing some fine tuning.
The CPU temp in the BIOS reads between 28C-35C. Most of the time, it's at 32C. Using an i3 Ivy with 22nm technology.
Thanks for your help.
Vince
I think you can reset irq in bios but follow all the instructions in the cmp web site. This will give you all the tweaks necessary for initial sonic benefits
Gentlemen,
I am new here - and hope someone can help me.
I have been using cPlay for a couple of months very happily, and then added 2gig of ram to my XP 32bit system for other programs.
All of a sudden, when trying to play 24/96 files (ie large) I have "memory allocation failed" in cplay diagnostics.
The message shows 3g physical memory, 2.5g available and 780m required!
As usual windows error message are mystifying!
I can't be the first - can anyone help with this?
Thanks!
Is the message from Cplay diagnostics? Are you using ram load option in Cplay? If so unchoose this option and try again.
Thanks!
As far as I can see this option is available only on cmp not cplay. I have been so happy with cplay and had so many other issues to resolve with the whole system that I have not taken the plunge to cmp. I suppose its time!
Thanks for your help!
...full CPLAY/CMP route.
...you should create a separate installation to host cMP/cPlay. It will repay huge dividends in better performance(s). (sorry, couldn't resist the pun)
If you cannot manage to build a dedicated machine, then I urge you to consider rebuilding your present computer for dual operating systems - one for general needs, the other strictly for hosting cics's software. That is how I began when I discovered these threads years ago, and was curious enough to "check it out". Later I built a machine dedicated to music playback - and cMP^2.
While there are a few ways to organize dual-boot on a Windows computer, my time-tested method is based on a handy bootup utility called Smart Boot Manager (available at no cost -> http://sourceforge.net/projects/btmgr/). It allows you to define more than one primary partition on your HDD (ie: multiple "C:\ drives") and choose which one to run with when you boot up or restart.
The advantage is, your OS installations are completely independent from each other, and can be backed up and restored without impacting anything else. The rest of the drives on your system can be seen by both copies of the OS, so access to the rest of your stuff is unchanged. This leaves you free to slash away at your cMP copy of XP, safe in the knowledge your "regular" computer still functions normally.
As you can read here on AA, the more you reduce your XP OS in size and complexity, the better the platform you provide for cMP^2 to do its thing.
It is a testament to the quality of cics' programs, that something so obscure as clearing out data values in selected "(Default)" fields of the Registry, should cause a change in SQ. This is indeed a high performance player which responds to fine tuning of the environment.
Welcome to the cMP Game michaelagiles, and I hope you can catch up with us soon !
Cheers,
Grant
That's not a Toy... IT'S A TOOL !!
Hello, everybody! This is not one of the promised tweaks that will follow very soon, but this thing came unexpectedly and seems SO GOOD, I decided to break the order of things.
So, here goes.
Open
1. system hive
2. default
3. software
4. SAM
5. SECURITY
and do the following:
at the very top of every entry You will find a STRING VALUE (REG_SZ) and it has a name smth like (Default) and a Value Not Assigned or closest to it. Well, see the pic and sure, You won't mix things, as this thing is everywhere and it is the same.
So doubleclick on this value to edit, type 0 and then press backspace to remove it, say OK
You will get this field EMPTY.
Perform it on ALL values in system exept
HKEY_LOCAL_MACHINE\000\ControlSet001\Control\Session Manager\Environment
HKEY_LOCAL_MACHINE\000\ControlSet001\Services\MountMgr
HKEY_LOCAL_MACHINE\000\ControlSet001\Services\Npfs
HKEY_LOCAL_MACHINE\000\ControlSet001\Services\Npfs\Aliases
Do it everywhere in default
then comes software
Well, You can remove this value everywhere, however, I suggest for sonic reasons not to touch ASIO, Classes, CMP and CPLAY. I thought that removing this value in
HKEY_LOCAL_MACHINE\000\Microsoft\Windows NT only gives better sonics.
SAM - there are only three such values, - empty them all
Security - There are NO such values.
The picture shows how it looks BEFORE You empty the value.
Next post shows the same string value EMPTIED.
Serge.
In an effort to move this along I imported a key (thanks Grant) back into Control\Enum\Pci which I had deleted over the weekend. Doing so allowed me to edit the defaults in all my PCI entries in Enum. Unfortunately after I additionally editted the ACPI key I got a no boot black screen. So slowly but surely I am making progress.
I was able to edit down through Control the Steppe stuff except for Session Mgr and got a 256kb sized System hive. Even though Serge believes it does not make a difference I compressed it anyway and it compressed back to 32kb and sounded better IMO.
Edits: 12/05/12
Wish we knew someone with an intimate familiarity with WINDOWS so we could know what is happening with this stuff!
As soon as I get working amplifiers and digital system ...; I can't wait to hear it.
Are you still anti-regopt? I wonder if this results in a smaller size of the hives?
Take care,
Just heard from him.
I think he stayed up all night doing the tweaks.
Is this becoming the way we spend our Christmases? Seems to be.
Can't wait to hear it for myself.
Take care,
I was hung up on Control\Enum\acpi so I skipped that and did all the
others. In collaboration with Grant I did not do the DEFAULT hive but
did edit the one element in Security and it boots fine. I also
refrained from doing Session Mgr\Environment and the others in
Services that Serge recommended not to do.
Bottom line....Wow...sounds great...richer, less edgy, soundstage is
killer. This cmp^2 version is a game changer. I have one Joe Lovano
piece entitled Rushhour with Joe on alto sax and he is backed up by several
other brass players...all of which are in their own distinctive
space...now. The sense of acoustic space around vocalists and percussion is phenominal.
Serge was absolutely brilliant on this tweak.
WELL, ONLY (do) CONTROL, ENUM and SERVICES, do not touch select, setup and mounted volume - the system won't boot.
... I got all the way through Control but not including Enum, Sessions Mgr and the rest. Funny I did Enum yesterday and it worked but today it hangs.BTW so far it sounds scary good!!
Edits: 12/05/12
Got all the way through Control but could not get it to boot with any of Session Mgr edits even while leaving environment. Should I stop there?
Hello, inmates, Read this with care:
After You are finished with CONTROL and ENUM, do the session manager and only AFTER that go to the services.
read, please, my How to do it post once again.
Serge.
I did do control then enum then I tried session mgr and still got a no
boot. Still have not done services or the other hives. So I was asking
should I not do services, then default, software, security, and sam? In other words do the rest depend on session mgr being editted 1st or are there some others that can be done w/o session mgr?
I found 4 such (Default) Reg_sz (value not set) 'values' and editted as you indicated in regedit above and got a no boot. I wonder if this does not apply to the G31M mobo?
Hello, Ted! Please, send me Your default hive and, please, read, the order of hives to be processed in the OP.
Serge.
So just that I am clear:
1) Do System hive 1st
a) do Control bottom up (dont do session mgr yet)
b) then do Enum
c) then do Session mgr in Control
d) then do Services
2) then do other hives Default, Software, Sam, Scerurity
Should I compress hives along the way after each step?
Also let say I do the System hive in 4 steps a), b), c) d). Should I boot after each step?
Qestion
1) Do System hive 1st
a) do Control bottom up (dont do session mgr yet)
b) then do Enum
c) then do Session mgr in Control
d) then do Services
Yes, except all entries in control, enum, services should be done from TOP to bottom.2) then do other hives Default, Software, Sam, Security
Yes
Should I compress hives along the way after each step?
No need
Also let say I do the System hive in 4 steps a), b), c) d). Should I boot after each step?
I did it in 50-60 steps. I'd boot and check after every 5-7 entries...
Do not be in a hurry, inmates! It takes full day or two.
About the SIZE of HIVES
as soon as we reach 256 kb limit for system and software and 24 kb for sam security and default it's NOT THE SIZE, but the remaining strings that matter, because let it be known, that our hives swell to those limits during a cmp session anyway, and the size of memory allocated for them is always 2 mb, so it is what's inside, that matters....
Serge.
Edits: 12/04/12
Well, inmates, You should consider this order with care. I do not know why it has to be that way, otherwise the thing won't boot.
It is relevant only for SYSTEM hive, all other hives can be done in one jump.
So
First empty in the LOWEST ORDER entries of CONTROL, like
HKEY_LOCAL_MACHINE\your hive\ControlSet001\Control\Video\{23A77BF7-ED96-40EC-AF06-9B1F4867732A}\0000
Go from top to bottom, but do not touch session manager at all
next go one level highier, like
HKEY_LOCAL_MACHINE\000\ControlSet001\Control\SystemResources, also, go from top to bottom
next go even highier, like
HKEY_LOCAL_MACHINE\000\ControlSet001\Control\Arbiters
DO NOT YET TOUCH SESSION MANAGER
Next go to ENUM, get to the deepest level, like
HKEY_LOCAL_MACHINE\000\ControlSet001\Enum\ACPI\PNP0A03\1\Device Parameters
Perfort from top to bottom, then get a level highier, like
HKEY_LOCAL_MACHINE\000\ControlSet001\Enum\PCIIDE\IDEChannel
And so on.
After You are finished with CONTROL and ENUM, do the session manager and only AFTER that go to the services.
It will take a day or two, but a hundred entries is a lot of good sound.
See the order of hives in my original post.
The whole thing sounds now, as if not a veil, but several woolen blankets came off, seriously.
Don't be in a hurry, save Your hives first, do not stop midway.
I am sure, that other inmates with a lot of experience in slimming the registry will help anyone, who has doubts.
Serge.
P.S. Will be back in a couple of days.
only a pic
Well, removing all these values from our super-slimmed registry means removing MORE THAN A HUNDRED LINES.
Try it for yourselves, I am sure the sonic resolution will attain a level literally "unheard of"
Serge.
After this long one can only assume that I am writing this post to no one but myself.
All good things come to an end.
One would think there was something to add to this but I cannot think of anything and I guess no one else can either.
Short of a perfectionist audio motherboard and a sound card with the same mission this probably has gone as far as it can.
Take care,
Hello, Rick! Hello, everybody.
The first hour back to AA and I see that the mood is bleak indeed.
Well, small wonder, for we were silent for so long, however...
There is one more tweak, one more hardware change (very easy, cheap AND IMPORTANT) and one more bios setting to be published in a day or two. All produce very big change for the better.
For now only two "confessions" -
first = I HAD TO LISTEN TO YOU AND GRANT more attentively
second - At the current stage, after the tweaks announced above to be "voiced" , any doubt and experiments with kernels ARE OVER, - native XP kernel IS BETTER.
A teaser now -
READ this http://blogs.technet.com/b/markrussinovich/archive/2005/07/24/running-windows-with-no-services.aspx
and You, inmates, will know what I am working on now.
It's a hard thing and expensive (for me) as I am not a coder myself and had to find, hire and PERSUADE a man to try it for us. No definite promise as yet, however, my experiments with UNSLIMMED windows show fantastic results. The work is slowed much due to the absence of cPlay source code.
Will write SOON.
Serge.
P.S. I apologize for being silent, but I was off internet for some personal reasons, so I will start answering Your letters only tomorrow, please
the one that is lost and gone forever! (hope you are familiar with this bit of Americana). I think I have used it before when you disappeared. I stole it from Harry Pearson of THE ABSOLUTE SOUND magazine. From one of the early issues when it was a very exciting magazine like this thread has been at times! He was speaking of some distortion that was ameliorated by some bit of magic but more likely a circuit refinement.
Grant and I have wondered aloud many times - WHERE THE HELL IS SERGE?
Not that we wanted more stuff from you, just to enjoy the interaction on many subjects.
Hope things are working out better for you.
I had feared you were in the cell next to PUSSY RIOT! Knowing you like cats so much ...
Take care - I will send you a letter this weekend.
You are the light @ the dawn. Cannot wait for your upcoming posts.
http://6moons.com/audioreviews/jplay/1.html
I have never heard what this can do and my skepticism it about equally matched by my curiosity.
Seems like the reviewer experiences a few revelations due to this software but has not yet made any attempt to minimize his WINDOWS.
I am not too interested in buying a copy of SEVEN so I could try out JPLAY but if I had nothing else to do I could be tempted.
One wonders what this reviewer (whom I must admit I have not read many of his reviews so I have no idea of his biases, etc) would think of a WONG/Serge cMP set-up? He did make the step to eliminate the bells and whistles and is on the path to a minimized interface.
If the thing did not require SEVEN I would give it a try just to see where it is in comparison to where WE are.
ADMISSION - I do not have a working computer at the moment so I have not had the pleasure of hearing Serge's new cMP. I have been listening to LPs only through a phono preamp I just finished assembling.
There are important differences, as we all know, between the two approaches. Both are noisy in their own ways. Many of my LPs are almost as old as I and some can be crackly even though the music shines through the noise. With digital, as Jack Wong has brought to my attention, there is this "stuff" that is everywhere which is what we have tried to minimize. I do not think it can eliminated any more than LPs can be pop-less. But there is no question that you can learn to ignore the pops and crackles where the digital noise is so interwoven that it preys on the subconscious at all times.
I find myself listening to whole LP sides and the time flies by. Was it the convenience of skipping around that made me change selections with the computer? Well, in my setup I still had to get up and walk across the room to get to the computer to make a change so I do not think a lack of convenience is at work.
There is something about LP's.
When the computer is back up (JULI@ died and I have to re-mod the new one to get back - too busy with the preamp) a comparison will be easy to make.
There are just as many things digital does well as analogue. Both formats
are needed for the music lover.
Amazing how much one has to pay for an LP these days - they do seem to be of a better and more consistent quality, though.
Do you remember Marcin_gps? I'm pretty shure that he's the poland author of jplay sw.
He wrote here and in AA Computer Audio until last year but no mention to cMP2 in his interview.
Well done marcin! Ah money money......
What I thought was interesting about the review was the reviewer beginning an understanding of what has been described and implemented here.
Hoping this same reviewer might find it worthwhile to listen to a minimal WINDOWS/cMP setup.
JPLAY sounds like something for the more casual audiophile but might begin a path for those more inclined to find something better. It all sounds too EASY to be true to my way of thinking analogous to adding amplification stages to an amplification chain and finding it sounds better - that doesn't happen very often either.
Hi guys.
-dear (happy) cMP´ersThanks for your recommendations. Much appreciated. -and always helpful.
Really !However : Since the cMP² venture began with the slimming DLL`s, drivers and step 1-20.... things have become much more fragile and wobbly.
Especially our (limited) way of communication of following suit on our forum structure.
It´s almost impossible to recapture what´s going on when one gets behind.I miss (and I´m sure a lot of cMP followers too)that the development get´s summarized from time to time.
Problem is, that when one has missed out a detail, further "tweaks" become un-manageable because a detail might not be implemented from a previous tweak.
We have to address this somehow, and improve our communication skills in this development.
Otherwise it will get obsolete very quickly and only in the hands of very few only...
-This can hardly be in spirit of cics, who gave way for this incredible democratic high-end project.Any suggestions on this ??
kind regards Leif
cMP2 Computer w/digital crossovers & FIR filters > Lynx Aurora 8 FireWire /192kHz all channels. 2x AcousticReality Ref. 202 & 2x AcousticReality Ref. 601´s ICEpower. Magnepan MG3.3R beechwood frames & custom stands. Miller chokes
Edits: 08/15/12
Hi Guys,
After almost 6 years of AOB and cmp(2) i decided to take another route and would like to thank you all for all i've learned.
Like Walter i changed from PCI to usb-i2s. I could not follow the last series of tweaks. I had great results with Jolida's iso and bitchbatch (@Jolida: i didn't respond because i used Juli@), but didn't manage to create time for further tweaks and experments.
This summer i changed to a usb-i2s/spdif interface, and PuppyLinuxMPD (mpdPup) as OS. I experimented with Alix, but chose an Asrock Atom board. Nice thing about mpdPup is that it loads via usb and than functions in memory. When it is started up, i pull the usb-stick. No HDD/SSD/whatever (well yes, there is one: in the NAS). Great sound, similar to Jolida's iso and Bitchbatch, BUT: streaming from NAS (very convenient) AND: i can play -yes- mp3 files!
A Raspberry Pi is waiting for further experiment this winter, when i planned to learn more about linux.Again: very very much thanks to you all!!
@ Greg in mississippi: since i do not use my Juli@ anymore: PM me your address and i will send it by mail.
Douwe
Edits: 09/10/12
Hi Douwe,
Right now I’m considering the same but for me it doesn’t feel as ‘Changing direction’. More like ‘moving on’.
Thankx too Cics cMP + cPlay project and the improvements suggested by other inmates I learned a lot. But I don’t like too severely cripple my XP installation close too dis-functionality.
So I’m now also looking at (puppy) linux distro’s geared towards audio. Unfortunately this only means: fully packed with software that is somehow related too recording and playback. So quantity over quality.
Untill now all these (puppy) Linux audio distro’s sound less too my ears than a fully ‘Steppe slimmed’ cMP+cPlay machine. But I still haven’t tried HQplayer from Signalyst on Linux yet
May be these linux audio distro’s can also be further optimized too equal of even pass the sound quality of a crippled cMP+cPlay machine.
Unfortunately untill now I still haven’t found a linux-audio forum on which inmates strive for real audiophile quality.
Do you know such forums or inmates on the net?
Mark
cMP2 PC > ESI Juli@ > Van den Hul Optocoupler MkII > Lavry Black DA10 > XLR Vovox Direct S > Klein & Hummel O300
Hi Mark,
Have you tried MPDPUP? Pretty well slimmed down, plus low latency core. I have posted some settings that to my ear take it to a good place.
If you have tried comparing linux .. specifically MPDPUP or if you have never tried it..
Take it for a test drive. The author just put up a new release 9.3. OMG!!
Really nice pop in quality. On my system a level never before contemplated.
My Alix/WaveIO/TDA1541A rig loves this new release. I am ready to take on vinyl challengers. Bring it on!
Hi Walter,
yes I'm almost ready getting there.
For half a year I already have a GA-D525TUD Atom mobo lying arround.
I used it too test if the hardware and chipset had better SQ than the standard socket 1155 cMP Gigabyte MoBo's.
I could not detect any differences. So I continued using the 1155 socket mobo with with Celeron processor.
Seeing you, Douwe and others moving on this summer, made me curious how this GA-D525TUD Atom mobo would sound on Puppy-Linux.
But I don’t want too highjack Cics cMP thread. So I’ll start a new thread ‘How too get best SQ from Puppy Linux’.
http://www.audioasylum.com/cgi/vt.mpl?f=pcaudio&m=114482
So I continue in that thread.
Mark
cMP2 PC > ESI Juli@ > Van den Hul Optocoupler MkII > Lavry Black DA10 > XLR Vovox Direct S > Klein & Hummel O300
Sorry to se you move on (not for you but for me). Pls come back and let us know how you're doing from time to time. Good luck.
Hi Theo,
I’m not planning too leave the audioasylum. I just need a new challenge. There is no progression in the cMP thread any more. I expect not many new software tweaks. I also don’t t like too cripple XP untill it almost breaks down too get good SQ from it.
Reading Wlowes and Douwe10NL moving on too Puppy linux inspired me too do the same.
I now notthing about Linux. It’s al new too me. But may be I learn just as I learned from Cics cMP project.
I notice a gap between the Puppy Linux’s newbie forum and the audioasylum PCaudio forum. The newbie threads on the Puppy Linux are still way too complicated for me. I don’t understand what is writen.
May on the audioasylum I can start a thread too brigde that knowlegde gap. Kind of: ‘How to get best sound quality from Puppy Linux for dumm’s’.
I already got an enthousiastic reply from Wlowes. But (again), like on the Puppy Forum - it’s much too complicated / cryptic for me too understand it.
I plan too periodicly report what I’m doing and what my progress is. I hope other inmates will respond or follow into experiments with Puppy Linux. This way there is chance we will all learn how too setup a Puppy Linux Distro and get very good Sound Quality from it.
WLowes, Douwe10NL, and I believe also Ryelands, already toke the plunch this summer. I follow now. May more inmates like too follow if there a thread on the audioasylum.
Mark
cMP2 PC > ESI Juli@ > Van den Hul Optocoupler MkII > Lavry Black DA10 > XLR Vovox Direct S > Klein & Hummel O300
hfavandepas,
Have you explored mpdpup? It is designed to get you up and running without really learning linux. You will need to go through a relatively frustrating learning curve if you want to modify the various config files. mpdpup is a good way to get started without making changes and then read everything you can about beginners guides.
http://208.109.22.214/puppy/viewtopic.php?t=70052&sid=3d93dcb742a75f80aa0c8624ad187360
I did find a good manual describing the basic concepts of file system which i'd be happy to email.
My own experience with MpdPup is one of mind numbing frustration. I think that you probably need a very special mindset to figure it out. It’s supposed to be the simple route to linux-based audio, but I certainly couldn’t get to grips with it.
I’ve kind of given up on Linux as a possible direction because most of the stuff I’ve looked at is just way too flaky and seemingly dependant on the user being able to master a whole new language in order to comprehend the cryptic screen messages.
Cmp/Cplay still seems like a nice approach to me. What I think it needs is further development in areas like the user interface.
Douwe,Thanks for the bounty of a Juli@! I've responded separately. I'll hold it in reserve for someone who needs one (perhaps after a botched mod!).
Please don't be a stranger and keep us updated with what you're doing, especially your innovative HW mods.
I personally am also interested in what you're doing from a player perspective with PuppyLinuxMPD... and also your experience with an Alix board. I've been interested in trying some different player setups and have been looking at an Alix 1D (with PCI so I can use a powered-Juli@) to try as a headless NOS cMP player as well as a server for the Squeezebox Touch I recently bought and also as a PuppyLinuxMPD player. Maybe post your experiences in the 'Tuning Linux/MPD/Alsa for best sound on ALIX processor' thread? Or if you already put them somewhere else, point us to them.
Greg in Mississippi
P.S. I forgot to mention that with the latest slimming tweaks, the slimmed version that works with Juli@ is STUNNING!
Everything matters!
Edits: 09/15/12
Hi Greg
If you play with Alix, I'll point you to a couple of key tweaks that bring out the best. One inmate suggests the 1d over the 3d2 I use as it has bias control you would enjoy. I had good success with replacing a couple of caps and of course had to add some blackgates. Sounds like you have CMP really perfected. Interested in your comparison. Of course, stay clear of Alix if you want to upsample in real time. Now way it has power for SOX or SRC. It is strictly NOS.. fine by me as that is my preference but each ear is different. You should consider WaveIO with the Alix. You could reserect your 1541 dac + WaveIO as an alternate to current rig?
Hi guys,
I red your idea, I completely agree.
I have myself problems following some evolutions because I just don't have the time to be here every days and to test all the new tweaks.
In order to better communicate, I think It could be great to have a structured web site, like a wiki.
The project could be seen like an open source dev project where everybody can contribute on what he like.
I have in mind http://www.rockbox.org. This site is very well structured, with a lot of contributors from the rockbox community.
This is just a idea, but I hope you will succeed in it. For Myself I don't have the time to contribute, but why not a small financial contribution in order to help things.
Nicolas
Unfortunately, I saw this downhill slope coming many months ago. I warned that this slimming process had to be "organized".
Primary importance: cMP2 must remain a rock solid system working with ALL asio devices out there - PCI/e, FW e USB - therefore both in synchronous and asynchronous mode.
It is NOT possible to produce versions on different OS, or for one sound card only, or a cMP shell that can work in particular conditions only, no LAN, no USB...and think that most users are going to fall that path. All this process has extremely reduced the already scarce usability of cMP2 without a certain proof of increased SQ in controlled conditions.
OK, do you want me to follow that path and test the system?!
This is my suggestion: I install OS (WinXP SP2/3) and sound card (any one I like - PCI/e, FW o USB), you provide an EXE that installs the right cMP/cPlay, AND that makes all slimming and variations to the system providing exceptions for things like LAN or anything else I care to leave untouched. The system must work and be stable.
The EXE can then be developed and refined with further versions following people's input.
I had already made that EXE long long ago under the title of the "Bitch-Batch", but none seemed to provide me the exceptions of the files needed for their particular sound cards. All I got was just one or two responses with regards to Juli@ ( which is also what I used), & Lynx (if I remember ). So I gave up. I even remember sending the EXE to one inmate to test it, & funny he dint reply either :) so I gave up the idea of fine-tuning it to the various specs.. I had made it interactive, which asked the user about the soundcard & network usability & then proceeds with deletions etc. Too bad..
Junaid
Hi BiBo01,We certainly have to come up with a strategy on this.
My own feeling is that we ought to set off from a nLite reduced XP.
nLite offers the possibillity of define exactly which components can be taken out. This is done via an .ini file (easy to distribute Notepad-file)so that we at least all have the same set-off point.
The idea of an .ISO file has leagal problems, so that´s a no go.
We would also need to agree on some kind of a standard as to how the harddisk´s are to be configured.What I would like we could do, is to enroll the various steps in maybe 4 chapters.
Each communicating how the hives are changed and what elements (DLL & drivers) are to be moved/deleted.Maybe we should also look into a different webspace than the cPlay forum, as I could imagine a more convinent platform to operate such undertaking.
And we´ll need to find a webmaster for this (Serge ?).I really feel we ought to preserve this grandiose system and the development that started with some "belgian beer".
-despite that cics might be on a different chip/trip...kind regards
cMP2 Computer w/digital crossovers & FIR filters > Lynx Aurora 8 FireWire /192kHz all channels. 2x AcousticReality Ref. 202 & 2x AcousticReality Ref. 601´s ICEpower. Magnepan MG3.3R beechwood frames & custom stands. Miller chokes
Edits: 08/16/12
With all due respect there are folks on this forum with differing talents. You cant just assign responsibilities to others who may not share your priorities. For example Serge is undeniably the 'spear of the arrow' type. He is the leading/bleeding edge of unbelievable tweaks. I think he likes doing that and will probably continue to do that. I also hope Serge continues in this effort. Jack is very good at a completely new iso based system that I like to call Jwindows. Jack does what he wants to do and will continue this way. I also hope Jack continues in this effort. Others are very good at communication/clarity of language in describing the steps necessary to achieve certain levels of tweaking . In fact I know of at least 2 efforts to start a coordinated documentation of all this cplay/cmp stuff. But in both cases as far as I know both efforts never got out of the gate. Why this never happened is open to debate but I think some folks dont like being given assignments by others that are not aligned with what they like/want to do. So this will always be a difficult task. We are forever beholden to cics and others who participated in the creation of the cmp website. Now that was an acomplishment given all the potential roadblocks discussed above.
So what should an audiophile do? In the absence of an overall bullet proof website I submit one should ask questions on this forum to help get a discussion started to 'get at' the clarity we all seek.
Hey theob,
...I was not trying "to assign duties" in my post,and I very well aware of how difficult it is.
What I was trying express was that I think some of these "steps" are findings that others deserve to understand, even if someone is joining in at a later point, or not neccessary spend 5 hours a day on his cMP rig.
Plain: a summary
I think the way to get there is possible if a certain structure is agreed upon. We deal with DLL´s and drivers; we deal in changing/deleting stuff from the registry; and there are certain "tricks" in changing difficult-to-get-rid-of modules, right ?
Certain things are obvious stuff, and other things do alter the functionality.
If we had a platform and a structure, I´m pretty sure we could work it out. Maybe not to the fullest potential, but at least to gain a bit of ground.
Let me suggest to do a "copy&paste" document of the steps 1-8 and edit it to my understanding. Then proof-read it by a volunteer with some experience in this.
Often things get stuck in the complexsity and a too wide a scope.
I´m happy to put in some effort, but I can´t do it on my own....
kind regards
cMP2 Computer w/digital crossovers & FIR filters > Lynx Aurora 8 FireWire /192kHz all channels. 2x AcousticReality Ref. 202 & 2x AcousticReality Ref. 601´s ICEpower. Magnepan MG3.3R beechwood frames & custom stands. Miller chokes
No offence was meant ... so with that aside, sure I can help too. But somebody has to be the organizer. Are you that person? I know I will not be the organizer but will help where I can. But remember there are approaches taken here that some do not believe in. For example some have expressed not agreeing with the totality of what Serge recommended. I think it was Jolida who only believes in a subset of Serge's 22 steppes. How do you reconcile that with Serge, Jack, Serge Mihaylov approach? If this were a business and I was the program manager of this task I would know how to approach it. But for a volunteer, 'we dont really know each other' and 'sometimes disagree' this is tough. I still think that someone seeking info should ask questions right here on an 'as needed' basis. Generally those here are usually very helpful in that way.
Edits: 08/16/12
" For example some have expressed not agreeing with the totality of what Serge recommended. I think it was Jolida who only believes in a subset of Serge's 22 steppes. "
As in what..???
Junaid
Based on my memory you indicated that you felt stopping @ Steppe 6 was best for you. Also you had mentioned that your cd player was better than cmp^2. If something has changed I was not aware of it.
I believe Steppe intends to come out with a step-by-step guide of how to achieve a super slimmed cMP2.
I hope that it will be taken into account the possibility to mantain USB, FW or LAN!
Hi Play-mate,
Months ago when I asked to organize the tweaks I also offered the Italian cMP2 forum I represent. As a matter of fact, there I posted both the nLite XP Jolida version and a few "super tweaks" by Steppe which many people are happily using.
However, here a few users wanted to continue along the "wild" experimental route, others wanted to organize but did not deliver.
I personally felt that a lot of effort was spent by Steppe and others on not verifiable results as real improvements and not just simple side steps. For example, the use of nLite XP version gives different results with different sound cards when used as DACs.
With no measurements of any kind on noise, jitter or differences on the analog outputs, it's all too empirical!
That's why I said - give me an EXE. A before and after situation can be measured. A tweak no.69 that is good only when tweak 4-8, 25-31 and 45-51 are implemented, has no use to me - period!
Sorry if I sound too aggressive. I care for this project and I put a lot of effort in helping people to know it and appreciate it.
I would still like to see effort put in the development of cMP shell, in cPlay's better asio compatibility and DSD playback.
I agree totally!
Dear Forum,
Have I missed something ?
By installing the new cMP program and putting "winspool" drv (and the other one) into the DLL Folder (with other deleted stuff).....everything worked fine.
Significant sound-improvement.
BUT : when setting configurationmenu into "Start cMP" -Mode, the bloody computer responds with a "Black Screen".
Can´t get anywhere. The entire system is BLOCKED.
Any idea what to do ???
Please help.
Leif
cMP2 Computer w/digital crossovers & FIR filters > Lynx Aurora 8 FireWire /192kHz all channels. 2x AcousticReality Ref. 202 & 2x AcousticReality Ref. 601´s ICEpower. Magnepan MG3.3R beechwood frames & custom stands. Miller chokes
Hello! This is a nasty situation. This new cmp was meant to be used AFTER all settings were made. It should only be put instead of the old file. Can You access the cmp hdd from another computer? Maybe this way You can manually edit the ini file, though I don't know where this start xp-start cmp info is stored. Can You overwrite NOW the new file with the OLD one? Maybe this will help?
Serge
Hi Serge,
I´ll have to see what can be done regarding changing the cMP files.
The cMP .ini file does only display stange symbols, where as the .pth file is displaying
"EXPLORER_KILL "c:\windows\system32\taskkill.exe" /F /IM explorer.exe"
-which is probably the one to replace.
As for now, it seems that the new cMP.exe doesn´t allow the "START XP / cMP change function.
kind regards Leif
cMP2 Computer w/digital crossovers & FIR filters > Lynx Aurora 8 FireWire /192kHz all channels. 2x AcousticReality Ref. 202 & 2x AcousticReality Ref. 601´s ICEpower. Magnepan MG3.3R beechwood frames & custom stands. Miller chokes
You can also try to load your
WINDOWS\system32\config\SYSTEM
registry file from another (working) windows install, and look for this key:
HKEY_LOCAL_MACHINE\SOFTWARE\Microsoft\Windows NT\CurrentVersion\Winlogon
It should point to your explorer exe if you want to start XP, and to cicsMemoryPlayer.exe ("1.exe") if you want to start CMP. If you need help, please ask.
I assume that you ran into problems, as most (?) of us happy CMPers :) are using super-slimmed installs by now (correct me if i'm wrong), with no remaining explorer.exe etc... so this issue w/ black screen did not occur.
Hello, everybody. Here is a nice discovery. As some inmates know, I am using a cplay.exe that was extremely heavily modded in resource hacker. In fact this is all I see, after I double click on an album in cmp: see pic - no controls. even to stop I have to click on close window x
So it happened that today I decided to check if it is shlwapi.dll that is needed by cplay and NOT needed by cmp, and after I deleted and booted, everything worked. I mean there was no usual jump-out-of cplay without this dll. I REALLY do not know if it is a by-product of the new cmp file or the hacking of cplay, but if You want to try, just e-mail me and I will send You my cplay.exe (2.exe) that works with the new cicsMemoryplayer.exe and allows to delete shlwapi.dll. Without it and the Winspool.drv (I hope You already deleted it) the Windows folder is only 15,5 mb.
Serge.
Hello Serge (and all you other DLL Exterminators)
What a pleasant surprise when I deleted SHLWAPI.DLL from my cMP... the effect of removing WINSPOOL.DRV has been amplified by this change. It seems as though dynamic contrasts are laid bare with even greater precision - mostly a good thing (I'd rather know when a recording has "too much sauce" than not be able to tell, for lack of playback veracity).
However, "be careful what you wish for" has already occurred... in listening to a new favourite tune, it's obvious they hit the compressor(s) a little too hard in the chorus(es) - bed tracks and verses breathe just fine, but everything slams into a brick wall when they go into the refrain. Too bad, but now I know (...their previous album did not suffer from this choice).
Of course, these things are most apparent in the middle of the night, when the local power lines are calm and quiet.
Funny Serge that you suspect a link between 256-color mode and dropping SHLWAPI.DLL. The JW-Home recipe recommends 256-color setting, and so I've been set that way since long ago. If only I'd known...
Thanks for the tip !
Cheers,
Grant
That's not a Toy... IT'S A TOOL !!
It may be we can delete this dll now, after setting all files to use 256 colors? - I have read today somewhere that this dll is somehow responsible for color settings.
Serge.
Did you try deleting with a bit more functionality in cplay such as vol/phase/pause? I have determined I need these but would certainly like to rid my system32 of shlwapi.
Thanks again Serge
OOps I think I like shlwapi back in for sonic reasons. Deleted it sounds a bit edgy. Sonics improve with it placed back in system32 folder.
While eliminating shlwapi can produce some edgy highs it is primarily with files tending that way. What is not in dispute is that eliminating shlwapi produces a stronger overall sound imo. I had not yet eliminated one of my 2 krnl's in system32 so I decided to try that with shlwapi deleted...everything seems to fall in place now...so far.
I should know better than to question Serge!!
Hi folks,
-I was just going through the german website again, and it struck me :
most "services" are recommended to be set to "manual" and so I´ve always configured cMP² machines, but would it possibly make a difference to "de-activate" them instead.
-what´s the difference in techno-speak ?
Any ideas ? what do you do ?
kind regards Leif
cMP2 Computer w/digital crossovers & FIR filters > Lynx Aurora 8 FireWire /192kHz all channels. 2x AcousticReality Ref. 202 & 2x AcousticReality Ref. 601´s ICEpower. Magnepan MG3.3R beechwood frames & custom stands. Miller chokes
Hello, everybody. It finally happened. We have a cicsMemoryplayer.exe file, that does NOT REQUIRE winspool.drv to boot.
Those of You who have renamed/relocated files according to Jack Wong's recommendation (like myself), should rename the file to 1.exe. Everybody else, who use the old file under its original name should simply add .exe to the new file. Just e-mail me to get Your copy.
Serge.
Yes, You can reboot, shutdown and so on. Thw winspool.drv can be deleted.
Yes, it makes the difference. I hope, You will like it.
Hi, Does anybody know where I can get the 'cicsMemoryplayer.exe' from please, which does not need winspool.drv to boot?
Although we were asked to send an email to obtain this, tt looks like Serge Mihaylov has got his email settings to not allow contact from forum members.
Anybody else have a copy or know where I can download it from?
Thanks.
PM sent.
May I have directions to get this new cicsMemoryplayer.exe file also, please? Tks!
Thanks, the new CMP file improves things tremendously. Sound quality in a word- amazing!
I assume that the bit stream must be cleaned up no end on the timing side (less jitter maybe)? What this does on my system is tighten and deepen all aspects of the music. Treble becomes delicate, clear and subtle, the integration and power across the whole audio range is much stronger. Syncopation, timing and tautness is a lot better. There is a greater relationship between instruments (including vocals) and I'm also noticing some inner details I've never heard on my system before. Bass lines are much richer, the whole thing just has more of that smoother, analogue sound.
Makes me wonder what this 'winspool.drv' does to cause such a downgrade. In comparison, the sound I was getting before was, sort of, watered down, a bit fizzy and less focused, sloppy with a glassy treble and less weight.
There is just a couple of strange things about the new file (they're livable with- and no-way would I want them fixing if it may degrade the sound).
1. The button at the bottom left, which used to bring up the cmp settings (things like 'folder' locations, 'RAM load', 'Start CMP' etc.) now just brings up an 'ok' box with an exclamation mark in it, when you click 'ok' , another of the same 'ok' boxes pops up, then clicking this goes back to the main song interface.
The only way I got the new file to run, to use it in cmp mode, was to set my options with the original 'cicsmemoryplayer.exe' file running (including 'start cmp' button). Exited the settings page and rebooted. Then using 'explore' in the Cics song list interface, located the file 'cicsmemoryplayer.exe', renamed it and then altered the other file name (1.exe) so it became the new 'cicsmemoryplayer.exe' (I had already copied this file into the same folder already). After re-booting it then uses the new version.
2. If you run the new version of the 'cicsmemoryplayer.exe' then exit back to XP (in other words, if it was NOT running in 'cmp' mode), then it goes into an endless loop opening (I think) 'My Documents' folder up, over an over again, for some reason?
Well apart from those two things, the sonic benefits are well worth the slight issues. I could possibly write an interface to do the settings page screen, maybe as an additional little app in Autoit? All it would need to do is alter the config file I think?
Thanks again, this has been a real eye (ear?) opener.
The button at the bottom lefthas, as I understand it, been disabled by design in steppe's version of the cMP shell. Your suggestion is one workround but others include:
1. Editing the registry (fiddle about with the shell installed on a different machine to see what to change and how);
2. Restoring Resources that steppe deleted from the file, thereby restoring functionality he's happy to do without but that you might want to keep.
I've used #1 before but had to try #2 because the new file didn't work at first. I could log on to the system (I use headless mode) but got only a blank screen. After deleting steppe's dialogues and importing the one remaining dialogue from my own "known good" but heavily hacked version of the file, all was OK. It's easy to do that in Resource Hacker.
I agree with you that steppe's file sounds significantly better than the "stock" version but I'm a bit puzzled as to what you mean by "inner" detail. How is it different from "outer" or "middle" detail?
Edits: 08/09/13
Yes, I thought it may be by design, as well as the removal of the CMP graphic logo (for resource reasons). I agree that the lack of settings is livable with. May look into this on a separate pc, as no-way do I want to risk messing this new sound up.
Sorry if I was using hi-fi babble terms when describing the new sound, got a bit enthusiastic! Guess what I mean by 'inner detail' is hearing more of what is inside the mix, the multi-layering, instruments that were there before, but vague and indistinct now sounding clearer, new nuances or sounds that were before undetectable, that sort of thing... It's like, you can hear the whole, but also more of what's making it up.
I've tried other Cmp tweaks from here before, stripping down Windows even further, but I think this is the first one that has genuinely made much of a sonic difference. Do you know of any others (or links to pages here) that improve the software side? Not sure how much better it can get though?
A problem I used to seem to get, is that the sound quality of Cmp / Cplay after a few days (from fresh) seemed to 'go off' (get worse), loses its tightness and clarity? No idea why, suspected it was something to do with Windows doing its thing in the background? Re-installing Windows, doing the software optimisations and re-setting up Cmp used to bring it back to life again... Anyone else have this issue? May have even been due to 'winspool.drv' being used, who knows?
Steppe (is Serge also involved? -whoever created this file) have done a superb job on these improvements. Bit confused why a printer spooler driver was ever needed as a dependency in the first place?
Please PM me also.
Thanks
Well Serge... thank you AGAIN for pushing the boundaries of cMP^2 !!
This is a transformation of no small consequence... at least in my JW-Home_Uber-Slimmed_XP-OS it is !
I really like the way the system sounds so much clearer in complex passages than before the change - as if some low level noise has been washed away. New details have revealed themselves, and everything occupies its own space, distinct from other elements in the mix. And yet, this does not deconstruct the song, diluting the musical intent, but rather lets me hear the care and finesse the producers put into making my favorite albums.
I'm going to enjoy getting accustomed to the new ("expensive") sound of the system.
Highly recommended to all cMP-ers, but most emphatically for those on the Ultra-Slim-XP diet.
Thanks again.
Grant
That's not a Toy... IT'S A TOOL !!
Hi serge, I also experience a fan-tastic sound improvement using your cicsMemoryPlayer.exe (1.exe)The soundstage improved right away, and rythm and pace is even better!
Seems you found a friend who could "remove" winspool requirement from the source code! Impressive. I also had a fellow who assumed he could do certain things with the CMP source (Bibo... I am sorry!) about half a year ago, but canceled the idea (things got too complicated w/ requirements of CMP as a shell replacement) With things now über-slim for best sound, the idea seems not viable anymore, as certain big (deleted) dlls required for i.E. jpeg decoding are no longer hooked by CMP anymore. At last. As we know, the design goal of CMP/Cplay was sound over visuals. Even if it would be nice for some to have display of cover images, which I understand perfectly.
BTW, I have to say that using the SPDIF-out of Juli@ which I had until 1 month ago, did "hide" many of the improvements that I made upstream. With the posts of GStew and Theob I managed to change to the I2S interface with the DAC (AK4396) inside my CMP case. I had to pick my jaw off the floor w/ the improvement, no kidding. (Although the transport-dac combo is not galvanically isolated anymore, and inside the case there is much RFI which i plan to shield against using crystals and ERS paper if money allows)
Again thanks to Serge & JW & team, best wishes and happy listening!
Edits: 08/15/12 08/15/12
It seems, that it won't.
Hello, everybody! It was a long time, since last tweaks were posted, so, here goes.
This tweak is Big, it's Long, BORING and tedious, but I hope, You will like it.
First:
If You remember, we were changing files' attributes, like removing the indexing for fast search, setting exe files' attributes for 256 colors, disabling additional text services? for them and so on. I translate from my memory, and as You know my working windows is Russian, but, here goes:
1. right click on ntldr and rename it to ntldr.exe ! Set the attributes for it just the way You did to regular exe files, next rename back to ntldr.
2. Rename ALL the remaining dlls from system32 to exe, like, for example, basesrv.dll to basesrv.exe, and set the file attributes the way You would to normal exe files.
See the pics. DO IT ONLY ONE BY ONE, doing it to a group of renamed dlls will end in NOT setting the file attributes to them all, only to some. Don't ask me why. Don't forget to click APPLY and the OK.
AND A WARNING! After You rename the files back to dll, they will automatically require that you untic File ready for archiving and allow indexing for fast file search crap. see pic 0.jpg
Hi guys
I must admit that this steppe's trick (original> .exe> .com> original) is really incredible!
Disclaimer: I don't use a nlite cMP2 (I've found the regular version sounds better in my setup) nor slimmed version. My system get some fishy secrets, formerly maxmem (most important IMHO).
In brief, I've found that:
- any modified file works just right
- we can apply this one to .exe files of cicsmemoryplayer and cicsplay
- if you treat all .exe, .dll and .drv of system folder the sound becomes gorgeous, the dynamic improves like the timbric refinement, all the music developement appears more clear and better coherent.
So I've trated all .exe activated in process of taskmanager, without any problem
After these files I've trated all files (but shvapi.dll) listed here (from a previous jack wong's thread )in the system32 folder
1) ole32.dll
2) shell32.dll
3) setupapi.dll
4) netapi.dll
5) advapi32.dll
6) comctl32.dll
7) comdlg.dll
8) cryptdll.dll
9) lsasrv.dll
10) mpr.dll
11) msasn1.dll
12) msvcp60.dll
13) ntdsapi.dll
14) rpcrt4.dll
15) samlib.dll
16) samsrv.dll
17) scesrv.dll
18) secur32.dll
Result is incredible. Try to believe!
Thanks a lot steppe for your efforts in the cMP2 developemnt
Hey Serge,
1. right click on ntldr and rename it to ntldr.exe ! Set the attributes for it just the way You did to regular exe files, next rename back to ntldr.
Do you have a link to what you are talking about that was done to the regular exe files?
Also for multiple picts in one post just hit the preview option and then browse for a new pict. Repeat
Afterwards we discovered faith; it's all you need
It really opened up after I did the process on ndetect and ntdlr. Way to go on this tweak Serge.
that get really very very clean and smooth.
Hello, Ted! Hello, everybody! I STARTED this tweak with ntldr and ntdetect.com and clearing System Volume information, so in my cmp it became evident immediately, then I started working on all other files. They all contribute a little, and the cumulative effect is getting bigger and cleaner after several days, but it is very probable, that the biggest part of the secret is in the first two files and Sys Vol Info folder.
BTW, with highs so clean it sounds now like THE LOWS and Lower Mids are cleaned in fact, but it IS the Highs, You are right.
Glad, that You liked it.
Serge
...Ole32 and lsvr(sp?) and it worked like a charm. Picked up 3 mb of memory and its hard to tell if it sounds significantly better but certainly no worse and maybe just a little better. But this is testing only 2 of the dll's.
Now I will do rest of dll's.
For those who try this here is my process. I copied the cmp dll's into a folder, changed them there per Serge, renamed original cmp dll's to *.dllbak like ole32.dllbak then copied reworked files into cmp. This way there is always a backup to easily go back if something does not boot.
Serge your genious/creativity never ceases to amaze me!! Well done.
Did another 20 or so files. So far so good.
Nice to see you post again Serge! I'm trying to remember how to set attributes. Is it right within Windows or is it through Resource Hacker or a similar program?
Its amazing that you have found yet more stuff to reduce memory.
Welcome back.
Just Right-click on a file, like lsass.exe, and You will see the options, I have posted in the tweak (see the pics). Just the same way, when you RENAME, say, kernel32.dll to kernel32.EXE, You will see the same options. Just follow the tweak guides, First set the attributes to a file, while it is an EXE, then rename it to *.com and set attributes according to part 2 and three. I will meanwhile try to rename and tweak all *.sys files. I HAVE tweaked this way winspool.drv, but I HAVEN"t YET tweaked the nls files and win32k.sys, but They WILL follow!!!
The effect is dramatic, I'd say, knowing more or less YOUR setup. With 16 mb windows everything matters, and such a tweak, is a BIG thing. Also, I think I should post that:
I use JW's micro shell32.dll with ntfs wo a problem
I have tried both fat32 and ntfs AND I prefer ntfs for LESS noise, if any
My system hive is 24 kb with ntfs, and it DOESN'T matter what file system You chose
All these things were done on a system with ONLY ONE hdd, that contains windows AND all wavs-cues on the same partition.
I boot in less than 1,5 minute with appr. 900 cue files, the single piece of a song not split is more than 5 min 14 sec. with / maxmem set for 500Serge.
Edits: 08/07/12 08/07/12
Hi Steppe,
thanks.
Can you please give us a description of what this tweak should bring.
with setting file attributes to the real exe files, renaming and setting file attributes to DLLs (that are, in fact, programs-read exe modules) will give a very clear and airy presentation. More, I think that data from restore points' folders in system volume information folder is to some extent kept in memory. The effect of clearing a large amount of memory is akin to setting very tight timings, I mean it's immediate, but will bloom and surprise within appr. 24-48 hours.
The overall presentation is now with a profoundly deep and wide scene and with lower mids much better defined, however, I'd attribute it to clearer Highs' presentation. MUCH clearer. The attributes we set will remain with the file even after renaming back to dll. I will start renaming and setting these attributes to *.sys files as well. I know, that this work is very dull, especially, as we need to do it only one file a ta time, and I understand that there is little explanation (from such an ignorant-in-computers person, like me) but the tweak works, no doubt. In fact there ara only about 40+ dlls that remain, so... If You remember, many folks were doubtful about removing debug info from the files, however it worked brilliantly. I want to point also, that when we rename dll to a com, we set the file to PREVENT using High Memory Area, and ttf fonts. It seems to work as well.
Just try cleaning system volume info and setting attributes to ntldr, ntdetect.com, kernel32.dll, ntdll.dll, bootvid.dll and framebuffer.dll. I am sure the effect will be very strong.
Serge.
It's all about cleaning more memory, after all.
I am aware that you are using an ultra slimmed cMP2 system.
However, can your tweak be used safely by people with a "classic" cMP2 (from XP SP2/3)?
here is Ted's discovery comes handy.
Hello, J-L!
Ted pinpointed, that the biggest effect comes from tweaking ntdetect.com and ntldr. I would add, that clearing System Volume Information folder seems to be another big thing. BTW, if You are going to post this tweak on Italian site, I can send You my complete windows (16 mb) with registry and all even with tweaked cmp-cplay programs so that You can post it there as a reference material for people who would like to go the slimming way.
Serge.
Before we do it, let me try to tweak drivers, also, I will report by saturday.
Can't wait to give it a try.
When I can ...
It´s always worth considdering that the leaner the overall system becomes (and the systems involved here, are indeed), every little change provkes a substantial & audiable change.
I often think that people that are not following the cMP approach, and thus are fiddling around with huge standard platforms, always report that they can´t percieve any change when they delete/deactivate things.
Yeah right : of course one cannot expect any sonic improvement by removing 50KB of "overhead", when the entire system comprises 700MB....
Agree ?
kind regards
cMP2 Computer w/digital crossovers & FIR filters > Lynx Aurora 8 FireWire /192kHz all channels. 2x AcousticReality Ref. 202 & 2x AcousticReality Ref. 601´s ICEpower. Magnepan MG3.3R beechwood frames & custom stands. Miller chokes
pic 6
Now to the tweak part three
If You take a look at the System Volume Information folder on Your windows (CMP)hdd, You will see, it has zero size and The folder is empty,
Now, IT'S A LIE ! ! ! It contains about 10 mb of data in restore09765454767--- something like that subfolders. Including copies of the intact registry, btw.
Open this hdd as a seconday one (the way You did when tweaking the registry) and Take posession of the folder:
It is a Yelling folder, and I had to perform it several times, but to take posession You have to make all hidden, protected and system files visible, then reboot and rightclick the System Volume Information Click on Properties
Click on Security tab
Then add the user account on which You want to access the System Volume Information Folder (administrator or Administrator). Now click OK and it is ready to explore.
Delete there everything You can. You will only have a tracking.log there remaining. Perform all You did to dlls, to the last wonderful file, namely ntdetect.com. (be sure to back up ntldr and ntdetect.com before You do this tweak.)
Now reboot into CMP and listen to the music.
The complete tweak sets in 24 hours, make several reboots.
Serge.
Could not find the security tab here...does not show up. I can see the restore file about 15mb of stuff. I guess I can just delete it all. Right?
Works fine,sounds good!Thanks again Serge!
Edits: 08/11/12
pic 6
pic 5
pic 4
pic 3
pic 2
This is part two of the tweak and it's even longer, even more boring and much more difficult, as it requires attention.
After You set the attributes to dll files that You had renamed to exe-s, don't hurry to rename them back to dll-s, rather, rename them now to *.com, so that the original basesrv.dll that is now basesrv.exe would become basesrv.com, Yes a DOS extention!
One-by-one method ONLY! Now set the attributes as follows:
1. open general tab, at the bottom You will see attributes with a MORE option, click and set everything as in pic 2.jpg
2. open program tab, click on more (or advanced, I don't know how it goes in normal windows), a dialog appears, just delete there info about autoexec.bat and config.nt, SEE PIC 3.jpg, You cannot miss.
3.Open Font tab and set the file to use ONLY raster (screen) font, remember those tiny 4-5 kb fonts, that have a *.fon extention? There they go! Also set the font size to the smallest - 4x6 pixel. See pic 4.jpg
4. Pic 5.jpg shows two more steps. open memory tab and uncheck Use HMA (High Memory Area), next open Screen tab and remove (uncheck):
Restore parameters at startup and Emulate RAM Function (Fast ROM), see pic and DON't touch Dynamic memory assignment
5. Last But Very Important! Open Miscellaneous tab and set uncheck Screen saver, all hot key combinations, everything, except Background regime , set priority while waiting to Lowest. SEE PIC 6.jpg
This is the end of part two. Part three to follow now
just a pic to the initial post
Edits: 08/06/12
There are also a younger crowd that has discovered cMP²
:
http://www.youtube.com/watch?v=LqroXbLJmUQ
cMP2 Computer w/digital crossovers & FIR filters > Lynx Aurora 8 FireWire /192kHz all channels. 2x AcousticReality Ref. 202 & 2x AcousticReality Ref. 601´s ICEpower. Magnepan MG3.3R beechwood frames & custom stands. Miller chokes
Nice vid.I really dont see the need for the fans. If they follow the recipe the "case" fans can absolutely be removed. Same for the psu too, and that IN a case.
But on second thought maybe this is a HT rig too????
Afterwards we discovered faith; it's all you need
Edits: 08/04/12
Hey Dawnrazor,
Well. The guy has (with a dual-boot) combined a cMP² music player AND a video rig. Therefore the fans.
My aim was to highlight that there are cMP² enthusiast out there that belong to a younger generation, and that there actually is some cMP² stuff on youTube.....
-and I find the Banchetto 101 is a very cool (albeit impractical) case.
L.
cMP2 Computer w/digital crossovers & FIR filters > Lynx Aurora 8 FireWire /192kHz all channels. 2x AcousticReality Ref. 202 & 2x AcousticReality Ref. 601´s ICEpower. Magnepan MG3.3R beechwood frames & custom stands. Miller chokes
Hey Leif,
That is good news and I am thrilled that the "younger" generation is catching on.
I agree about the case. Isnt there an rf nightmare with a case like that with no covering??
Afterwards we discovered faith; it's all you need
Very nice
Yeah....good to see that also the youth (he is about 20ys. old) is gettin´on to cMP².
(-but it´s NOT from me....:-))The guy has an entire site on his music-/video computer:
http://home.tele2.de/my-audio-pc/
The Banchetto 101 case is COOL - isn´t it ?L.
cMP2 Computer w/digital crossovers & FIR filters > Lynx Aurora 8 FireWire /192kHz all channels. 2x AcousticReality Ref. 202 & 2x AcousticReality Ref. 601´s ICEpower. Magnepan MG3.3R beechwood frames & custom stands. Miller chokes
Edits: 08/04/12
Dear Forum,
I´m about to add a manual on cue-sheets to the new german cMP² Website, and have made a little YouTube Clip to show how it can be done.
I´ve used the GENRE tag in the cue-sheets´s to divide sections 0-9 for rythmical music, and organize all the classical albums in order of the composer.
Have a look at it here :
http://www.youtube.com/watch?v=M0ZT2BeuDPU&feature=g-upl
kind regards
Leif
cMP2 Computer w/digital crossovers & FIR filters > Lynx Aurora 8 FireWire /192kHz all channels. 2x AcousticReality Ref. 202 & 2x AcousticReality Ref. 601´s ICEpower. Magnepan MG3.3R beechwood frames & custom stands. Miller chokes
Looks cool although I understand very little german. Your music library looks alot like mine although yours is more apropriately organized.
Dear community,
Allow me to announce that a cMP² Website in German language is now avaliable and online.
www.freewebsite-service.com/cicsmemoryplayer
In all modesty & kind regards
Leif
cMP2 Computer w/digital crossovers & FIR filters > Lynx Aurora 8 FireWire /192kHz all channels. 2x AcousticReality Ref. 202 & 2x AcousticReality Ref. 601´s ICEpower. Magnepan MG3.3R beechwood frames & custom stands. Miller chokes
Well done playmate, very nice!
Another european place to meet cMP2's friends.
Italian forum, thanks to bibo01 and others friends, grows up gradually but costantly.
Now I feel the strong request to sytematize the great number of variants of the original cics' project because the structure of audioasylum it's not the best one for this use so it's impossible to rebuild each steps, valid or not, of each proposal ( steppe, jackwong and so on)done until now.
Best could be to leave some complete manual for these new upgrades....
Hey Audiodan,Thanks for your acknowledgement and comment.
When I structured the German site (and previously the danish cMP² site) I was also thinking how to make the most of it.
I´ve done about 6-7 cMP² machines for friends and myself and neither of them are done exactly the same.
I actually think the nLite way, is the best way to setup a contemporary system (eventually with the nLite configuration file to download), and then apply the other changes as in normal order.The offset from a slimmed (nLite)system would seem most obvious for new/comming fans of cMP² to understand.
In respect and regard to the original cMP² site, I refrained such strategy on the german and danish ones.
Another issue for newcomers could also be, they might like to take smaller steps at a time and not "ruin" their computer from scratch...
It´s a difficult one....
kind regards Leif
cMP2 Computer w/digital crossovers & FIR filters > Lynx Aurora 8 FireWire /192kHz all channels. 2x AcousticReality Ref. 202 & 2x AcousticReality Ref. 601´s ICEpower. Magnepan MG3.3R beechwood frames & custom stands. Miller chokes
Edits: 07/28/12
Hi playmate
I've done a different experience about nlite, because after an year of tests on more or less slimmer versions ( mihaylov and steppe with or witout nlite jolida version too) at the end I've found the original cics' setup the best sounding of all. Each slimmer version is more ethereal, very clear and dry without the beefy character that a realistic musical reproduction must have, IMHO.
Moreover after a while I've lost the way in the developement of the various step of each version, so my opinion could be wrong. I think it could be better to order each proposal in clear manuals from start to the end for every version of them.
Ciao
Daniele
Hi Daniele,
I think that everybody who started/experienced the various "steps" of deleting sys.drivers and DLL´s was surprised over the different sonic impact certain meassurements made.
Suddenly it was (not always) : "less is more....."
It will be difficult to explain these empirical findings in technical terms, and I´m not sure there is not a synergy to some of these "steps".
-but there is definitely some magic !
I´m pretty sure that there is a sonic difference between cMP2 machines when they are configured with the same steps, but implemented in different order....
-at least I´ve experienced that machines sometimes behave irregular to modifications done in different order.
Can we even be sure that a german, danish, italian or english XP sound all the same ?
I find nLite intriguing and I´m sure we could find a configuration that will yield a near perfect platform for further cMP optimisations.
The big question is : how !
kind regards
cMP2 Computer w/digital crossovers & FIR filters > Lynx Aurora 8 FireWire /192kHz all channels. 2x AcousticReality Ref. 202 & 2x AcousticReality Ref. 601´s ICEpower. Magnepan MG3.3R beechwood frames & custom stands. Miller chokes
I have just built a cMP2 based on Asus P8B75-M LE motherboard, Intel Core i3 2120 CPU and Prodigy HD2 sound card (with 2xLT1124 + 1x2xAD797) and am thrilled by the sonic result. At 192 kHz output with SRC 145 dB resampling the sound is shocking good, with transparence and emotion that probably surpass most refined analog gear. I would like to address a thousand thanks to the skilled and innovative cMP2 constructor(s). You have seriously furthered and made available high end digital sound to everyone who is a little handy with computers.
I only have one little problem with the cMP2. That is, I can’t use my freestanding 7” USB touch screen I bought for the player because it is widescreen (16x9) and cMP2 is apparently made exclusively for 4x3 aspect ratio (or am I wrong?). The 7” widescreen cuts off the lover part of the cMP form so the lower buttons are out of reach. Had there been a (standard) window maximize function in cMP2 there would have been no problem. All small freestanding 7 and 8” touch screens now seems to be widescreen, so is there any chance that a revision of cMP2 to fit also widescreen can be done? If this is not the appropriate place to put the question, please tell me someone where or to who I can address it.
Thanks
Ingvar
is there any chance that a revision of cMP2 to fit also widescreen ?
Sorry to report that there seems to be little or no chance of any revisions to cMP.exe which is, as you note, stuck with a 600 x 800 display. It should be possible to work round that limitation by using ResourceHacker.exe to edit cMP's dialogue 102 (and the equivalent in cPlay.exe) to change the shape of the window and so bring the missing buttons into view. It's not what I'd call elegant but it is at least a solution.
HTH
Many thsnks Ryelands for the great advise. It solved the problem. For guidance of others with the same problem, I have in the picture above demarked the hacked settings with red rectangles. cMP and cPlay now fit well into my 7" 16x9 monitor (Liliput UM-70). The settings reflect that it was necessary to reduce not only the height but also the width a little to fit the screen. For cPlay I therefore narrowed the field showing numerical information immediate left of the close button. This is ok for me but alternative solutions are possible of course if one wants the original readability of the numerical field retained. The settings forms are also to high for the screen and it is necessary to move them upwards as much as possible to see just the upper edge of the buttons at the very bottom, but the most important buttons are in that way reachable. I can live with that, but if one wants to hack these forms too, look after Dialogs 131 for cMP and 142 for cPlay.
Regards/Ingvar
Hi Ingvar & Ryelands,
Figured out how to use Resource Hacker and change the layout. When saving, Resource Hacker asks about to "compile" the new settigs. When OK it ask me to "save under"....so what to call this file and how to implement new settings ???
kind regards
cMP2 Computer w/digital crossovers & FIR filters > Lynx Aurora 8 FireWire /192kHz all channels. 2x AcousticReality Ref. 202 & 2x AcousticReality Ref. 601´s ICEpower. Magnepan MG3.3R beechwood frames & custom stands. Miller chokes
Hi play-mate. Change the setting(s),compile, and if you are happy with the result, save it. The original file (cicsMemoryPlayer.exe or cicsPlay) is overwritten and you also get a saved copy of the "original" file (i.e., the status of the file before you saved). To be sure the changed files will work, let them retain their original names.
Regards/Ingvar
I can add that there was quite a discussion on this topic that will be buried deep down in this thread. You can move buttons around to suite your screen layout. Some inmates reported an improvement in sound by removing buttons that you do not use or need. Might be worth digging through the last year to find the reference.
I have not used it for some time and cannot remember the file save details. I just remember it as saving as you do in any of the other resource hacks. It makes a backup for you in case you need to back out your edits.
I just remember it as saving as you do in any of the other resource hacks.
It's slightly different when editing the likes of a window layout rather than just deleting crud - you need to recompile the file first then save it.
As you suggest, I thought there was a slight improvement to be had from deleting stuff from cMP.exe and cPlay.exe though I didn't do it for sonic reasons, didn't listen with great care and might have been imagining it. Whether others found the same, I can't say.
Dear Forum,
I can report that the adjustment of screen size works fine with Resource Hacker, but discovered a problem with the alphabetical order of the album list.....it just shows albums in (random) order.
Any ideas to correct this ?
kind regards
cMP2 Computer w/digital crossovers & FIR filters > Lynx Aurora 8 FireWire /192kHz all channels. 2x AcousticReality Ref. 202 & 2x AcousticReality Ref. 601´s ICEpower. Magnepan MG3.3R beechwood frames & custom stands. Miller chokes
Resource Tuner works without causing this problem...as previously reported.
I encountered this problem which I reported here:
http://www.audioasylum.com/forums/pcaudio/messages/10/106923.html
Any ideas to correct this ?
I don't get the problem because I don't use cMP's UI but some others did and, IIRC, got round it by using alternatives to ResourceHacker to edit the cMP exe file. I think it was asylum member Riboge who reported both the problem and the solution. Try a search or hope that he sees this . . .
Has anybody been able to revise cPlay to provide dsf file playback? Recently I have been able to convert some of my sacd's to 24/176 files which sound great in cPlay but I'm told dsf played back natively with other programs is even better.
Would be great if cPlay could support this format now that there are affordable DAC capable of DSD playback.
I wanted to say a final thanks to the innovators on this site. I have taken a different direction, inspired by the extreme slimming of xp and the hardware by the leaders here.
Early on I chose not to use the Julie@ card and the reference MBs. I used a USB dac. As a result it was not possible to use JackWong's amazing iso. Following on the progression to use a minimalist MB and OS, I have gone the route of replacing xp with linux, and the CICS hdw with an industrial embedded processor called ALIX. I first used VoyageMPD, but now have gone to a very simple to use minimalist OS called mpdpup. Music is on a NAS, the Alix processor ($90) does not even need a heat sink as it has no video, sound, hdd or anything but cpu eth0 and usb. ASIO is replaced by alsa. The entire OS with player is 60mb and runs in memory. It boot from a CF card. The gui for the player is on yet another machine at your chair. Can be anything from a pc to your iphone. The sound is the best I have experienced. It may or may not duplicate what Jack Wong has created on XP, but its likely right up there at a fraction of the effort and price.
I post this here not out of disrespect, but quite the opposite. I think this forum has pushed the limit of great sound at great price. But it is not for everyone. I hit a wall, and others may as well. I mention this as I think there is another avenue for those that are, like me, locked into USB dacs to continue the theme that CICS has created and this group has perfected.
I am keeping a couple of CICS machines to plug in from time to time.
Thanks again to Serge, Rick, Jack and all the innovators that helped me get such amazing sound.
hi, your Alix device sounds interesting, do you have more details about how to put it together or a link to somewhere for more info ?
cheers
Thanks for sharing.
I suppose that the only disadvantage of the device you suggest is computational power which restricts playback to bit-perfect...also, perhaps, interface.
A similar route would be to have a network audio adapter between the computer of your choice and an Alsa recognized DAC. The NAA device would be similar to the one you suggest acting as a fifo buffer, is attached through ethernet to the PC, comunicates asynchronously and is galvanically isolated. It would run a no-gui version of Ubuntu Server with LAN-in and USB-out.
The computer can be Win or OSx based with player and, eventually, DSP of your choice.
Bibo01
You are correct. It is limited in power and best suited to NOS playback (which is my preference). Doing bit perfect playback of any source material, the cpu idles at 1%. Resampling 44.1 to 96 using Fastest Sync is possible at 90% cpu. > Best Sync (Secret Rabbit code) is not possible.
NOS (or bit-perfect to the source) playback is fine. With the most common sigma delta converters, oversamplig is then done at chip level.
I also like NOS DACs. However, they have some limitation in the highs at redbook resolution.
With the Network Audio Adapter method I described earlier all computation is done by the PC source. Undoubtedly, there are some advantages to separate it from audio adapter. PC source does not need to be a dedicated machine anymore with undervolt/underclock. It can also be in a different room and easily remoted.
If it's required - especially with NOS DACs-, the PC source can use oversampling and employs extra DSP (like DRC or XO).
Ultimately, it depends on user's requirements and audio chain structure.
Actually a very interesting idea. I believe there is merit in splitting up the processing chain to isolate the functions. For example, to gain experience I put everything on one dual core machine running Ubuntu. (GMPC/MPD and the HDD with music). Sound not nearly as good. Even after killing superfluous processes, prioritizing threads and allocating mpd on one cpu the sound was very detailed but not this natural breathing reality that you get when you strip it down to just mpd on a bare bones processor.
Your solution would be a very good alternative for those looking for the real time DSP. Personnally I am doing the dbpoweramp approach to reformat before it hits the playback chain. Took some 192 high quality material and reformated to 96 to accomodate my elderly 1541a DAC. Wonderful.
especially for the folks using USB.
Wish I could give it a try. My curiosity is killing me but I am kind of committed to what I have at this point.
I think this is part of the same project and I am glad to read of your experiment.
Congratulations on your success!
Sounds very interesting. Would like to learn more.
Why not start a separate thread?
There are a couple of threads on this.
The mpdpup thread is currently experimenting with linux/mpd/alsa settings for sound quality.
http://208.109.22.214/puppy/viewtopic.php?t=70052&start=165&sid=5db937f03461866b66f653f12b8c1691
I started a thread in AA to discuss tuning on Alix for sound quality.
http://www.audioasylum.com/forums/pcaudio/messages/10/109926.html
One of the things that really sounded great right after implementing the Jack Wong system was the impressively strong dynamics and bass. However I had no idea what my system was capable of in the bass until I started implementing better feedback isolation techniques. Isolating every power supply not only helps but helps a lot. But isolating power supplies in a high quality way really opens up the soundstage. I have 3 Hammond 193L transformers in my system one on my dac, one on my crossover, one on my mid bass amp. When I repositioned the bass 193L to the input of my power supply chain (Running Springs ---> Shunyata Hydra and an assortment of Innovolt devices) I broke a bottleneck. Now when I added more high quality isolation devices the dynamics and bass ascended to unheard levels. There is a lightness to almost every big classical file I play until the real thing comes and when it comes it is really staggering. Surpringly the highs take on more reality too, that clear eyed ever soaring effect. This is not trivial. I think the moral of the story is when you move the needle in the system performance (and the JW system certainly does) you need to address all prior system performance attributes/assumptions. Going to a live performance of a world class orchestra also helps to highten the senses in the pursuit.A big thank you to Jack Wong for making this system performance possible.
ps: Isolating the hammonds also yields improvements.
Edits: 05/31/12
So when I upsample a 16 bit 44.1khz file to say 88.2khz, does the data also get converted to 24bit, or does it stay at 16 bit.
I am referring to the data on the spdif stream. Does anyone know for sure what happens?
Thank you.
Erin
Hey Erin,
Whenever you rip a CD, and thus converting it into a file on a computer, the file is stored at 24 bits.
The resolution is however still only 16 bits.
This applies to all computers and has nothing to do with upsampling.
This gives you the possibillity to use the digital volume control in cPlay util about 20-25 dB attenuation without any loss of resolution.
kind regards
cMP2 Computer w/digital crossovers & FIR filters > Lynx Aurora 8 FireWire /192kHz all channels. 2x AcousticReality Ref. 202 & 2x AcousticReality Ref. 601´s ICEpower. Magnepan MG3.3R beechwood frames & custom stands. Miller chokes
Thanks for that information. I did not know that.
I have the JW XP setup and with 44.1 to 44.1 playback I find that the resample settings in cPlay make a big impact to the SQ. Apparently the DSP is still processing the playback even though the sample rate is not being changed. I thought this would be bypassed. Is there a way to set up cPlay so that the DSP is not processing the file or is this just part of the way cPlay works?
Hey orediger,
What is your concern actually ?
Popular said, the DSP is the software engine that turns the file into buffer packages to be sent to the ASIO device (juli@). This cannot be circumvented, and DSP is in essence tha core of any playback software.
The preparation/processing that cPlay also performs, is the loading of the file into RAM....is this what you mean by "processing" ?
You must understand that a ASIO memoryplayer like cPlay, is NOT just streaming samples away in a continous flow, but rather transfering packets of data (buffers) to the juli@, for further "assembly" into a time critical signal flow.
When not upsampling, cPlay does not alter the data. In this case the unique upsampling technique of cPlay is circumvented.
Does this clear your question ?
kind regards
cMP2 Computer w/digital crossovers & FIR filters > Lynx Aurora 8 FireWire /192kHz all channels. 2x AcousticReality Ref. 202 & 2x AcousticReality Ref. 601´s ICEpower. Magnepan MG3.3R beechwood frames & custom stands. Miller chokes
Thanks for the reply, that does clear things up. I also went back and reread the CMP2 homepage discussion which explains the settings in detail. I still hear a difference in the sound with changes in the SRC/SOX settings even though I'm not upsampling, subtle though.
Under bios you will find 'On Chip Frame Buffer Size' and there are 2 options. Default is 8mb and the other option is 1mb. Does that mean you free up 7 mb of memory? I think you free some up. When I went the 1 mb option and clicked on settings for cplay I show more ram available for system than with the default setting. Immediately it sounded a lot stronger. For those who believe in burn in...after listening for 7 hours it seemed a lot better (yea I know maybe I'm adjusting to it). Either way a nice freebie tweak for those of us that have these or similar mobo's with this bios option.
On my new nLite install everything seemed fine.-Made the cics optimizations; done AWE (had to allow more user assignemts for that in mmc)....but cMP is not performing it´s "kill explorer" duty correct.
Booting XP and clicking the cMP icon, the XP-mode-background does not disappear. First when the system is set to boot into cMP mode, it works (sort of).
When rebooting I get the msg:
"cMP.exe DLL initizaltion failed.
The application failed because window station is shuting down"
The running of the blue cMP statusbar (below right) does seem rather shortened too....Any ideas of what´s missing ?
Thanks for your consideration.
L.
cMP2 Computer w/digital crossovers & FIR filters > Lynx Aurora 8 FireWire /192kHz all channels. 2x AcousticReality Ref. 202 & 2x AcousticReality Ref. 601´s ICEpower. Magnepan MG3.3R beechwood frames & custom stands. Miller chokes
Edits: 05/08/12
...when you run native sr (44 at 44 or 176 at 176 for example)but choose sox with the alias box checked does that mean you are eliminating the digital filter?
http://www.cicsmemoryplayer.com/index.php?n=CPlay.Guide
Its explained here.
It only stops aliasing in frequencies above the passband of audio.
I check the box because on my system this sounds better, but for no other reason.
Hello Intrepid cMP-ers !Here's a call-out to those still living in The Land That Time Forgot ... with your Gigabyte G31M motherboard, and "DRAM Timing Selectable" parameters still stuck on Auto in your BIOS (phew... how's that for a room clearing qualifier ?)
This is a message of Hope and Help , offered to you few who're still riding the Dinosaur mobo (G31M) from the Jurassic Age of cMP.
I am one of you - and this is the Tale of How I Tamed The Memory Timings... and made everything sound better !
It all began when Serge (Steppe) posted his suggestions for BIOS manipulation ("New Tweaks part 21 CORRECTED", http://www.audioasylum.com/forums/pcaudio/messages/9/97480.html ), especially the recipe for memory tuning. I must confess I've no real idea of what changing these numbers actually does to the operation of the RAM (except to make it perform "better"). To compound the confusion, he referenced settings in his H55M-UD2H BIOS, which have few corresponding controls in the G31M BIOS (similar function but totally different label). What's a Boy to Do ? Nuthin...
Without understanding of how these things relate to one another - and no guide specific to the older BIOS, the safe thing was to leave it as AUTO, and know it would work according to spec.
Chicken, Chicken, Chicken.
Recently I asked Serge off-forum to recommend settings I should try with the G31M, and based on my initial positive results, he encouraged me to continue in the same direction (...till it wouldn't boot - and then back off a little - having empirically found the limits for my particular samples of RAM/CPU/mobo). During these email exchanges, he remarked the numbers were getting "insane".
I am very pleased at how the faster timings transformed the sound of my cMP^2 playback. I've been running trouble-free (and best-sound-ever ) for two weeks now at the 6th level, and it's time to share the final results with the community.
Here follows a chart of the performance plateaux I passed through on the way to The INSANE G31M Memory Timings !
(numbers in square brackets are values I changed myself; the column under SPD shows the default settings indicated by the memory module itself; the "LowVal" item shows the bottom limit specified by the G31M)
Remember to hit SHIFT+F1 to gain access to the advanced settings in the "MB Intelligent Tweaker" page of the BIOS setup.
I don't expect everybody's G31M will behave exactly as mine - many other factors could get in the way - so it's safe to promise YMWV (your mileage WILL vary !) - however, it is definitely worth the trouble & reboots to get the most out of your particular RAM.
I encourage you to try it for yourself and report back your impressions. Perhaps those other secret beta-testers will chime in with their findings.
Enjoy the New SQ !
Cheers,
Grant
That's not a Toy... IT'S A TOOL !!
Edits: 05/05/12
Hey Gjw,I played it most of the day and then turned it off. FWIW it booted with no issues, so the timing seem stable at least on my system.
I'll try the voltage. On this box I can get to 960mhz but it often resets.
Oh and are you using just one psu? Or 2 with the 2nd working the drives? Or the 2nd one doing the drives and the processor??
Afterwards we discovered faith; it's all you need
Edits: 05/06/12
Hi DRI'm glad this works for you... thanks for reporting back.
Are you using just one psu? Or 2 with the 2nd working the drives? Or the 2nd one doing the drives and the processor??
Single Antec 430 PSU (P4 & P24 only), Granite-type power brick for SSD (3 x OS + 1 FAT32 local Music library), external eSATA/FW/USB drive enclosure with dedicated power brick (2TB Master Music Library).
Typical clockspeed is 900 MHz, thought I can get down to 800, if I'm prepared for glacial response from mouse & keyboard. I haven't listened to 800-SQ recently, must put that on the list of "Things to ReVisit".
Vcore will not drop below 0.75000 V - if I want the thing to start !
Drives are connected to primary IDE controller (master & slave) - second controller is disabled in BIOS.
cPlay settings -> Output: 192 KHz; Buffer: Tiny; Resampler: SRC/145 dB
Juli@ settings -> Driver: v1.10 (XP, 2004-05-20); Latency: 64 samples (stubbornly, this driver will not do 48 without severe distortion !); S/PDIF out (copper) via custom BNC connector > DIY cable ("SUNHER SWITZERLAND RG 302/U 75 OHM" proclaims the outer jacket)
Cheers,
Grant
That's not a Toy... IT'S A TOOL !!
Edits: 05/07/12
Hello, Grant!
I am really happy, that You have posted it, for these settings ARE VERY IMPORTANT to many CMP-followers.
I can well remember how helpless I felt, when inmate Dawnrazor, whose experience is respected here, asked me about details on his mobo, and I couldn't help him much. Now we have the comprehensive way.
You are the only one, who had courage to start this timings thing with this mobo, really in the fog, and I am SURE, inmates will use these settings.
It's not only helpfull to those of us, who have the "precedent" mobo and followed all our tweaks, but also, and that is of UTMOST importance, to the inmates, who are just going to start tweaking.
Brilliant work and a lot of patience was requiered, I know.
Thank You for pushing the thing further and upwards.
Serge.
Hey Grant,
I just plugged in the numbers you have on column 6 and it worked.
My voltage settings are different and I wanted to keep that the same so I was just testing the memory settings.
They work great. Though I cant comment on the sound. This is my headphone rig and I dont claim that it is all that good or resolving. It does sound nice though with your timings.
Thanks for doing this. Mine were just set to auto.
Afterwards we discovered faith; it's all you need
Hello, everybody.
Just a pic. It's so empty and cool, well, ... and dusty inside my cmp with an external drive and no fans.
Serge.
I thought you were going to refer to this forum!
Well, many cMP2 systems are similar to that one.
However, I sincerely hope that you will be able to provide the community, together perhaps with other dedicated users, with a final routine installation which allows "hooks" to different cards and requirements (lan, fw, usb).
Otherwise, it will really remain "empty and cool".
Dear Inmates,
It´s very comforting to find more and more people having genuine interrest for cMP2. Now that I´ve done quite some advertizing for it, people come and need help (which I´m happy to do to my abillities).
So far it´s been about 6 or 7 mashines, that I´ve configured from scratch, so eventually I need a bit of a shortcut.
Doing an configured cMP2 image on a foreign SSD does not work...
I´ve understood that one can boot a "virtual" drive via CD/DVD by the means of an ISO image, but there is no way that this can be installed on a clean SSD partition, right ?
I´ll start doing a nLite installation disk instead, but how much more than the obvious can be configured via nLite ?
I guess AWE, .ini-file and minlogon has to go in afterwards ???
I´m not sure if I´m ready to deleting hives (yet), but maybe someone could summarise what can be done within nLite, and what would be the most logical steps afterwards....
Help & suggestions most welcome !
kind regards
cMP2 Computer w/digital crossovers & FIR filters > Lynx Aurora 8 FireWire /192kHz all channels. 2x AcousticReality Ref. 202 & 2x AcousticReality Ref. 601´s ICEpower. Magnepan MG3.3R beechwood frames & custom stands. Miller chokes
Playmate!
I have used Paragons migrate OS software to successfully migrate to a clean SSD. I first installed the hotfix for XP2 pro for ssd on the original HDD. I have had zero problems for over a year on an H55 mobo, just updated to Serges update 21 corrected. I am assuming you are using identical motherboards.
T.
No worries!
Hello, play-mate! (please, excuse me, I do not yet know Your name).
if You are going to try the thing to produce a DISTRIBUTABLE iso, write to me, I will tell You how to obtain a VL version of xp pro, that will allow to delete the OOBE from the start. Also, I will send You my ini files for my nlite disk, that allows trouble free AWE, unlike the isos from Rick and Junaid. No problem here.
I'd like to focus the inmates' attention on the possibility of using a esata-sata cable with an external esata-usb hdd box. It will allow to power the sata hdd externally, but at the same time to avoid using esata on the mobo, just plugging the hdd into the sata slot master 0. This allows for even deeper registry reduction. Also, though I don't have JW's ISO, I'd try to contact with him, as I believe, that he has some how-tos about manually editing ini files in the iso. He will explain what I mean.
Ant mobo will fit, as long as it uses NO hdmi and other alternative video outputs. We all should experiment with battery powered mice.
To Dawnrazor, sir, please, wait a little, as I think we will soon see the post from one of our active cmp-users about memory timings on GA G31 MS2L mobo.
Also, about the timings, as far as I know, other folks experience very good results with modern mobos that allow to coldstart the cmp with EXTREMELY low voltage...
Serge.
Play-mate, sir, thank You for Your great initiative!!!
Hi Playmate,
You CAN put a "HDD-configured" CMP2 image on a new SSD. What is important is that you should align the ssd right. You can do it in a few different ways.
This is the way i did it:
- google for GParted and download (alternative: Parted Magic, it's a tiny OS containing GParted)
- google for GParted on USB and download the programm which makes a USB stick bootable and boots GParted
- start GParted by booting from USB
- when you find the ssd make the desired partition(s).
- BUT you should leve EXACTLY 1 MB unallocated (free) space BEFORE the first partition. Now "HDD-cmp" can find the firrst sector of your disk and boots.
- second action BEFORE creating the partition is to leave the box "Round to cylinders" checked. (i don't know exactly why, but i did it this way and it worked).
- now you can put the "hdd"-image on it; do not forget to copy the MBR
Another way is to put the ssd into a Windows-7 machine and partition the disk. Win7 recognizes an ssd and will do the right allignment itself.
I didn't try this method.
HTH
Douwe
Best thing you can do is run one of those programs (Russinovich Process Explorer?)that tells you what system32 files you need for your vst routines. Then delete all files/ hives per Jack and/or Serge.
Hi Theob,
so, what you´re saying is, that working with nLite, AWE, minlogon is no longer the way to go ??
I haven´t looked into the registry hives yet, as I missed the start of the discussion and the fishy parts.....I could probably catch up, but where to begin ? I´m sure you can imagine confusing it is to assemble all the bits, comments and threads on this matter....
This new configuration is not direcly for myself, but I´m naturally keen on staying "afloat" of new tweaks. This new setup is standard with Juli@.
For my own system with digital crossovers, FIR filters, FireWire, VST etc. I will have to do some more experiments with "Process Explorer", and see what works and what not.
That´s a different story.
Thanks for your help.
cMP2 Computer w/digital crossovers & FIR filters > Lynx Aurora 8 FireWire /192kHz all channels. 2x AcousticReality Ref. 202 & 2x AcousticReality Ref. 601´s ICEpower. Magnepan MG3.3R beechwood frames & custom stands. Miller chokes
No not at all. By all means do N-lite, awe, minlogon then do the Jack and Serge stuff.
Hey theob,Thx. for your input.
I´m about to asseble a kind of recepie on this, by going through this forums entire tree of comments and advice.....uh, gettin´dizzy....Since I´ve started the work on the german cMP2 website, and sort of taking a new approach on the guideance, I better get it close to right....
I will configure a completly new hardware along with re-writing the steps configuration.I found this nLite advice from rickmcinnies.
Does anyone have a comment before I perform these steps ? :
"Make an nLite install disk with the only the following remaining:
DRIVERS
IBM PS/2 TrackPoint (for obvious reasons)
HARDWARE SUPPORT
CPU Intel
Intel PCI/IDE Controller
OPERATING SYSTEM OPTIONS
SMBus contoller
Certificate Management
Group Policy Mgmt Console
Local Security Settings (this one might be done without. I will try eventually)
Out of Box Experience (if you delete this the machine will not work)
SERVICES
RPC Locator (figure cMP needs this)Make sure you disable Windows File Protection and Prefetch.
There are other things to do that should be obvious. Look at all of the choices after the main deletion menu. Many are personal taste choices and others disable or eliminate stuff you do not want. Do not rush through here the first time. Think about what you want.This will result in the fastest WINDOWS install you have ever experienced.
Install the chipset drivers
Implement MINLOGON
Install JULI@ - be sure to go ahead and turn off the undeletable AUDIO CODEC and to check "do not map through this device' and "do not use the mixer".
Implement AWE - when you type in ADMINSITRATORS it will tell you it does not recognize the name. Disregard the warning, cancel out, SAVE and restart.
Install cMP2 software. make sure all is working and check to see if "AWE successful" - if not go through the MMC process again.
Do the optimizations - you will find many of them moot due to this minimal installation.
On my current build I disabled the SMBus Controller in HARDWARE and now it is tricky to restart into STEPPE's BIOS. Next time I will leave this enabled. Anybody know how to RESTART without turning the PS on and off?
Disable the SERVICES. You will notice the list is much smaller and many are already disabled.
Do AUTORUNS. I took a chance (all worked out well) and checked a few of the VIDEO CODECS.
Use the Steppe REGISTRY mods all at once. Junaid's batch files make this quick and easy."
I will probably need further advice on the Registry Mods after this.
It would be great to recieve everones support for this undertaking, as I think the cMP2 project has become ripe for a serious update....Kind regards Leif
cMP2 Computer w/digital crossovers & FIR filters > Lynx Aurora 8 FireWire /192kHz all channels. 2x AcousticReality Ref. 202 & 2x AcousticReality Ref. 601´s ICEpower. Magnepan MG3.3R beechwood frames & custom stands. Miller chokes
Edits: 05/03/12
I am looking for the batch files of Junaid, can somebody help me to find them?
I can send u the Batch files if u need them. Let me know..
Junaid
All of the above plus "Multi-Processor Support". Make sure to retain it as well. Its very simple to remove unwanted dll & sys files by editing the 'textsetup.oem' file that goes with your nLite install. By doing this, you would no longer need to opt for file deletions, as you would already omit them before install, by means of the textsetup.oem file. Same goes for adding drivers as well as executables like Cicsmemoryplayer.exe or Cicsplay.exe. Edit the file with Notepad & remove unnecessary files that u do not need...
Junaid
You will not be able to access GROUP POLICY EDITOR in MMC but it does not seem to matter.
Once you get to cPLAY and check AWE, DIAGNOSTICS will say AWE SUCCESSFUL - so who knows when and if AWE is ever really successful! It was written that if you asked for AWE and it was "not there" cPLAY would exit. Who knows ...
None of the rest is needed unless you just want it.
The system will work fine.
One nice thing about an nLite XP Pro install is that you can usually delete all of the unnecessary stuff at one time.
Hey Jolida,
I´ll be okay with the mulithreading in BIOS, but what do you mean by "textsetup.oem" ???
Cannot find this file; nor "txtsetup.oem".....
Can you explain further, please.
Do we (ayone) have a complete summary list of unwanted DLL and drivers (in this context)?
Can´t wait to get it all up and running....
kind regards
cMP2 Computer w/digital crossovers & FIR filters > Lynx Aurora 8 FireWire /192kHz all channels. 2x AcousticReality Ref. 202 & 2x AcousticReality Ref. 601´s ICEpower. Magnepan MG3.3R beechwood frames & custom stands. Miller chokes
When u create an NLite disc, opt to create the installation in a folder on the desktop instead of burning the image immediately to a cd. The folder will contain a sub-folder named i386. It's inside this folder that the file txtsetup resides. It may be named as txtsetup.oem or txtsetup.sif.
When windows starts installing, it creates a list of files to be copied for installation. It is from this txtsetup file that it gets info of what to install & what not to. You can edit this file using notepad & remove or add any file you wish to. At first, it may seem a bit confusing. Just open the file with Notepad & look at its contents. You will, in no time, understand how the data in it is written to instruct windows to install which file in what directory. You can customise your xp install just by editing this one file. But before that, u need to be certain as to what set of files actually need to remain for your cMp installation to work without throwing up a missing file error. Just a bit of homework & you are good to go.. After editing the file, u can then burn the image to a cd to start installation as usual...
Serge & Jack have done an outstanding job of taking cMp to a level which might have seemed impossible for even a software techie to accomplish. Great job ;)
Junaid
.
cMP2 Computer w/digital crossovers & FIR filters > Lynx Aurora 8 FireWire /192kHz all channels. 2x AcousticReality Ref. 202 & 2x AcousticReality Ref. 601´s ICEpower. Magnepan MG3.3R beechwood frames & custom stands. Miller chokes
Hi Leif,
I think Jolida is referring to the file LastSession.ini (or .inf) where you can find details on the installation and edit it through Notepad.
I highly recommend you not to restrict your cMP optimization to a classic one (by cics notes), otherwise SQ will be compromised - bass is missing. It seems that one has to go all the way with slimming optimizations to enjoy a full sound again.
You can keep your "normal cMP2" and compare it to "super cMP2" all the time...and then report back. :-)
It would be interesting to know from Steppe and other users who have implemented the change at what stage "sound is back", to know exactly what it made the "final" change for the better. Registry slimming?!
Hello, everybody!
First of all, I'd like to note, that Junaid's warning about multiprocessor support is extremely important, as it assures CORRECT hal.dll and correct initial kernel will be renamed to real hal.dll and ntoskrnl.exe DURING INSTALL. Yes, this is the windows' trick!
Next! It's much easier NOW to tell what files SHOULD remain after ALL, because the files to be deleted are LEGION.
I think that the difference between Julia and Lynx and Firewire-usb options will be 3-4 files, while their total number is less than a hundred.
I can once again post the pic of my system32 and drivers and fonts folder, or, if anyone needs, can write me for details. Warning! this is an NTFS-XP-PRO-LYNX version.
Serge.
Junaid! Welcome back!
Hey Leif,
I will configure a completly new hardware along with re-writing the steps configuration.
Great project and I cant wait for the write up.
Though I doubt it will help (me at least). The mobo you mentioned is not available here and no one seems to be able to recommend one that IS available or at least give any kind of guidelines I can follow.
So what does one look for in a mobo these days outside of a pci slot?
Here is a link that shows the current mobos that are available. Any thoughts?
Afterwards we discovered faith; it's all you need
Hey Dawnrazor,
Yeah, it´s a bit of a puzzle to find out how to give the project an easier approach, but putting together the last 6-8 month of discussions on hives, AWE, registry etc. Wat works and what not
...is indeed a challenge.
Of your list of MB´s I´ve found one that I think would be good :
GIGABYTE GA-H61M-S2H
You may expect that the BIOS settings are different, especially on the clock- and voltage settings in the M.I.T. section.
But since there are fewer "knobs" to screw up, it will be easier to find an acceptable balance.
kind regards
cMP2 Computer w/digital crossovers & FIR filters > Lynx Aurora 8 FireWire /192kHz all channels. 2x AcousticReality Ref. 202 & 2x AcousticReality Ref. 601´s ICEpower. Magnepan MG3.3R beechwood frames & custom stands. Miller chokes
I am thinking to build a new Audio PC and thinking about the choices:
- Atom 32nm Board vs
- ITX vs
- uATX
I am leaning towards Lynx with PCIe (have as well the Julia, like the Lynx better).
Must be a bios which can be adjusted to undervolt etc...so Z77 most likely.
I am thinking to get a 22nm CPU...question there: i3 ? or I7 and then switch off the unneccesasary cores via bios ?
Choices over choices. One question: has anyone an opion about the GIgabyte 2 oz copper boards ? Will they sound better ?
Ironically the thinner sheet would actually be better for high frequencies which is not an issue for us at the speeds we ask of our boards.
So I figure it is only for ruggedness.
One could argue that the thicker material might further slow down the MB. But that is pushing to outer limits of tweak obsessives!
The best sounding MB I have ever used has the standard weight copper but I am sure that aspect has nothing to do with its performance.
I am using the board Jack Wong recommended - GIGABYTE H61MA-S2H.
When i changed from GA-H55M + i3-530 to a Atom-based Asrock board, it was a huge SQ-improvement. I think because of lack of (unneeded) functionality on the Asrock, like hdmi, dvi, usb3 chips. The Atom can upsample to 88.2, which is ok for me, my dac upsamples to 172. So i would ssay pick the smallest board with the least funcionality. Also count the number of coils (don't remember the right enlish wrd), it can give some info about stability of powerlines on the board
of i3 or i7, I think?
Hello,
I think it is the choice between i3 and i7. If it is i3, it means less problems with "intracore timmings", but no videocard, and the processor using memory for video, while with i7 you NEED an individual videocard, get highier, IIRC, voltage from the start, but will free memory.
Also, keep in mind, that going ultra slim registry way, You will in the end see, that You have to choose, if it's removing Lynx mixer (hardware) settings from the registry, or keeping them, (which add 32 kb to the system hive). If You remove the mixer settings from the registry, Lynx will be working in +4bdv professional analog output regime, and You can damage inputs of the home (not pro) preamp, If You will be using a dac, I think it makes no difference, but isn't Lynx set for 48 not 44,1 by default?
Serge.
Hello, everybody. It turns out, the proof is in thr fat32-ntfs booting routine, and though I was wrong, thinking that kernel changes ntldr, It does change , but the BOOT RECORD
The complete article is here (I and I suspect this man may know about Windows boot proceess more than the collective knowledge of those hackers):
http://thestarman.pcministry.com/asm/mbr/NTFSBR.htm
Here follows what leads me to insist that inmates cannot boot into Fat32 with neptune.exe:
... We will point out below that the NTFS "Backup Boot Sector" is not part of the NTFS Volume, yet their paper incorreclty states: "Hidden at the end of the volume, is a copy of the boot sector (cluster 0)."; it's actually 1 sector beyond the end of the Volume).
NTFS Boot Record is often described as being a single sector (see the proof below!), the truth is that it would never be able to boot up a Win 2000/XP OS without the next six sectors (so far) that comprise most of an NTFS partition's Boot Code. So, even though an NTFS partition's “Boot Record” is technically only 1 sector in length, we should never forget about the necessity of the “Bootstrap Code” which follows it. Furthermore, Microsoft states that the OS itself “allocates the first 16 sectors for the boot sector and the bootstrap code1.” Thus, Microsoft could have created a bootstrap code that was more than twice the size it is now.
So, the "Bootstrap code" really is just that: It's only used to boot up ANY existing OS in that partition! Even though the phrase "Boot Record" isn't very descriptive for a sector that's only used to access files in the partition, there's an historical precedent for using it this way: The first sector of every floppy disk is also called a "Boot Record" even though it will never be bootable unless you add the correct system boot files to it! You can, of course, remove most of the boot code from either the NTFS or a floppy's Boot Record, since their BPBs are the only thing that an OS needs in order to access their files!!
The NTFS Boot Record's “Backup Sector”
The Win 2000/XP OSs make a "backup" of each NTFS volume's Boot Record which they store in the very last sector of its partition!
[ Note: I said "partition" not volume. This is why an NTFS partition's Total Sectors count in the MBR/EBR's Partition Table is always 1 sector more than the "Total Sectors (in Volume)" count found in its Boot Record. Although the words partition and volume are often thought of as being synonymous (we may even use them as such here!), they are definitely not the same thing.
" Count of hidden sectors preceding the partition that contains this [NTFS] volume. This field is generally only relevant for media visible on interrupt 0x13. This field should always be zero on media that are not partitioned. Exactly what value is appropriate is operating system specific." [ Although I'm not exactly sure what all the implications are of this quote from Microsoft, I can tell you that this value is definitely used while reading the boot sectors into Memory!
According to most documentation, this area is not supposed to be used by NTFS, but after examining the code in the Boot Record, it seems clear that the first byte of these four bytes is used to set the drive number! (See offsets: 7C7B and 7CAF).
For those, who want details, please, read the original (detailed and well illustrated) article in full.
So, the culprit, as I understand is the DIFFERENCE in the way the boot process is executed in ntfs and fat32.
Serge.
Hello, everybody. As I started all this mess with kernels I think I must Disclaim it now.
None seems to be able to load it, and what occasional boots I have, I think are normal boots with win xp kernel, with the /KERNEL=neptune.exe syntax in my boot.ini either wrong at times with multiple retries, or just being ignored.
The Idea is NOT good and wrong from the start, as is explained here:
http://www.osronline.com/showthread.cfm?link=224660.
So, I apologize for leading some of You into this dead-end. I must have been wrong from the start.
I think I should avoid posting any new and not BULLETPROOF tweaks in the future, especially if they are not yet preliminary confirmed to work 100% by other inmates through personal circulation.
Serge.
P.S. I am sorry, inmates.
P.P.S. Many thanks to GRANT for clarification. It will save us a lot of unnecessary frustrating attempts.
Edits: 04/27/12
NT
cMP2 PC > ESI Juli@ > Van den Hul Optocoupler MkII > Lavry Black DA10 > XLR Mogami Gold > Klein & Hummel O300
I would try anything Serge suggested.
No need to feel bad about this. You are still a hero...big time. So there is one tweak out of ... what maybe several hundreds that did not work. I'll try your ideas every time. Don't stop...chalk it up as experience and go on. Some of the comments on the thread from windows experts(and kudos to Grant for a very well written problem description) show me how far and how intelligent some of the ideas borne on this site have progressed. I know how good they sound. When you listen to music 4-6 hours a day you dont need a-b double blind tests.
No question there probably are great ideas in the thread posed by windows experts(linux any one?) but they probably dont know that the ideas in a minimum registry/os have developed over several years and have been confirmed all over the world.
I say there is nothing to be ashamed of.
Hello, everybody. I must confess, that the first set of directions (sent with the neptune pack) was wrong. When I first introduced the neptune.exe, it sang immediately, but then it would refuse to reboot. i have made ALL SORTS of manipulations, including removing / slash before /kernel=, also, only today I have finally found that the switch itself is a one session only thing and that it should be put after fastdetect.
Now, Grant introduced the neptune.exe as I did at first, and he was very enthusiastic at first, but the thing refused to reboot, naturally, and we spent several hard hours to restore his system, with mine playing "perfectly" but WITHOUT /, so, without the true kernel. After we (me and some other inmates) all put the kernel= without a switch, the systems worked well, and what differences and improvements came from other files in the neptune pack were interpreted by us as the kernel improvements, which they were NOT. It's a simple placebo effect augmented with our expectations + real improvements that came with other files' substitutions. So, this is the truth, which I felt I should not conceal.About other boot.ini switches.
You might have read that on our level of registry tweaking boot.ini parameters are not relevant anymore... This statement is incorrect, because boot.ini is used by windows loading (booting) process BEFORE the registry is read, and the boot.ini parameters override any registry parameters. (Read wiki for details on win xp boot process).
next, UsePMTimer setting is not for kernel, it is FOR THE HAL, and we all use the same hal.dll, be it xp pro or home. The timeres swithc set to 2441 is a division of 9766 by two and then by two again (rounded, like rounded are the official MS figures) and it is for hpet timer, that should be enabled in boot.ini. I hope You remember, whenI suggested to return to it and the discussion. Query Performance Counters on a nonstripped machine is documented in the net to work with hpet in xp and EVEN Win2000.
For those, interested to know, neptune.exe is the ntoskrnl.exe from NT 3.5.1.
Serge.
Hope, You will forgive my mistakes, and will enjoy Your system equipped with a new donor heart.
For those, still in doubt, Yes, this big tweak is worth it absolutely.
Edits: 04/23/12
Hello, everybody!
First of all, I must apologize for sending INCORRECT INSTRUCTIONS before with neptune pack, concerning the neptune.exe introduction.
This is the right thing now, with all due switches, a-la MS, booting, rebooting, rebooting to tighter memory timings, everything works, even after unplugging the powercord from the wall outlet.
0. Make sure that Your cmp works, as usual, wo a problem.
1. Copy the neptune.exe file to Your system32 folder.
2. Open Your boot.ini with the notepad and enter this switch AFTER
/fastdetect
/KERNEL=neptune.exe
(please, note, that original file was neptune.exe, not Neptune.exe or NEPTUNE.EXE)The following is my WORKING boot.ini string:
multi(0)disk(0)rdisk(0)partition(1)\WINDOWS="Microsoft Windows XP Professional" /noexecute=alwaysoff /fastdetect /KERNEL=neptune.exe /maxmem=512 /UsePMTimer /timeres=2441 /nodebug /pcilock /nopae /onecpu /noguiboot /MININT
Please, note, that there are NO double spaces and ANY ADDITIONAL spaces between or before the switches.
Make sure that notepad is not set to auto break strings or auto word wrap or any other comand in Format menu.
3. Copy ntkrnlpa.exe file from Your iso and don't forget to decompress it to system32 folder.
4. Rename Your existing ntoskrnl.exe to ntoskrnl.old (never mind the warnings)
5. Reboot. If You have made everything carefully and correctly, You will hear this magic immediately, not after 3-4 days, as I wrote before (explanations later)
6. Now, You will probably LOVE what You hear, but if You reboot now, You will get a nasty ntoskrnl.exe failed notice. WHY SO. That was the thing I faced with w2k kernel, and we have experienced with Grant and Rick. The problem is, that the /kernel= switch is for debugging of experimental kernels and this feature seems to be a one-time only. On reboot, our kernel itself MODIFIES ntldr and ntdetect, preventing You to boot it again, unless, ... You change the SYSTEM DATE IN BIOS. Every wise man has a lot of simplicity in him, as we say in our country. So, to make the change permanent (is 12 years enough for You?)
7. Close cmp, and when it asks, if it should reboot after the system shutdown, say yes, the system will start rebooting, so immediately ENTER the BIOS SETUP and change the system date to the earliest possible. Mine is now at 01.01.2000. press F10 to save changes to bios and let the system reboot. You ARE THERE!!!
Enjoy.
Serge.
P.S. I wish to express my gratitude to Rick and Grant for being as obsessive with finding the truth as myself and for expressing healthy doubts to focus my attention to the problem.
Edits: 04/23/12
Hello, everybody. I must confess, my instructions were wrong AND the neptune.exe files themselves BOTH TIMES were WRONG.
I was making dozens of reboots and substitutions and simply got lost in all these boot.ini-s and file versions.
That time I had written about everything playing well, my neptune kernel was called neptunee.exe (smallest difference) to confirm to 8.3 standard in an attempt to somehow fool the ntldr-ntdetect.com. While in the boot.ini I had /KERNEL=neptune.exe. So in fact the kernel wasn't working at all and all my time-date tweak is false, for it doesn't work.
Now, the file that really works, is here, it's really a NT 3.1 kernel, not NT 3.5.1, - another thing that worked to mislead myself.
Well, only about two hours ago I found this thing. I am sorry. The file would boot only once, certainly, though You can easily boot it any time, Simply, it requires a certain procedure.
First, the neptune.exe should be in system32
Edit Your boot.ini, just like MS says we should (after fastdetect) /KERNEL=neptune.exe
It boots wo a problem, plays very well, it's much more resolving and smooth.
When You need to power down, just power down, but to boot with normal kernel next time (and You can only boot with a normal kernel next time), You need to do the following exactly and only in the followind sequence:
1. Edit boot.ini to DELETE the /KERNEL=neptune.exe switch
2. Delete win.ini file
3. Overwrite Your ntldr and ntdetect.com with a fresh pair from Your iso or, like me, from a reserve folder.
Reboot.
If You do this correctly, You won't get the ntoskrnl.exe error.
This way the whole procedure of using neptune, powering down and rebooting with normal kernel takes appr. 5 minutes.
I know that it's a PITA, but this way we can use it, at least, for some more important and longer sessions.
Serge.
Anyone who is STILL INTERESTED, can e-mail me to get the right one.
Hi Serge just a question or 2:
1) put the new neptune in system32 (leave both ntoskrnl and ntkrnlpa in sys32?)
2) amend boot.ini with kernel = neptune.exe
3) boot & enjoy great music
4) When you have to shut down amend boot.ini, delete win.ini, overwrite ntldr and ntdetect and boot into normal kernel
5) Go back to 1 to get the good kernel
Do I have it right?
Hello, again, Ted! You do not need TWO original kernels. Keep just the one You were usually listening to say, ntoskrnl.exe or ntkrnlpa.exe
Everything else is right. Just be sure to observe the sequence of deletions to reboot to "normal" non-neptuned state wo an error.
Serge.
I usually power up once a day, so for me the whole procedure is not a problem, exept it doesn't allow me to reboot to enter the cmp-extreme memory timings, but it's another story.
One more question for you Serge (and btw I'm nutty enough to live with this if I can get it to work and do I appreciate your efforts in making cmp a great sounding pc system) but what if I Snapshot my neptune kernel system and my regular kernel system and save those SS files. Then on an ongoing basis why could I not just restore my CMP disc with one or the other prior to rebooting? In fact why not restore the initial OS with neptune everytime I reboot? What tells the pc that the boot into it is the 1st or 2nd boot? With Jwindows so small I can Snapshot it back up in under 7-10 seconds. This I can definitely live with if I can get it to work.
I think, exept, You will be fragmenting Your drive very quickly.
Serge.
But I am not sure now. What if ntldr and ntdetect.com won't snapshot well? They are VERY TRICKY.
On the run up to where I am right now I have SS my ssd probably 200 times since last year. Always worked well. Think I'll try.
I feel I am not getting the whole picture here...
Where is this kernel file?
Have you been experimenting with other users?
What Iso?
Is it safe to change for ANY cMP user?
is availible in different places. E-mail me for details now.
Serge.
Folks! This tweak isn't yet tested with Your Home windows, so, I suggest that You
do not use ntkrnlpa.exe.
Instead, leave Your ntoskrnl.exe intact,
DO NOT rename it and
do not introduce ntkrnlpa.exe to Your system32.
Everything else remains the same.
Serge.
Be attentive! If Your neptune file is Neptune.exe, the syntax in boot.ini should be /KERNEL=Neptune.exe
If Your neptune file is Neptune.exe . . .
It's probably an age thing but I'm getting confused here. If "Neptune.exe" is the NT 3.5 kernel (BTW and IMHO, NT 3.5 was possibly Microsoft's best ever OS), why are we using boot.ini's Kernel switch? Why not just rename Neptune.exe to ntoskrnl.exe and replace the XP version?
I ask because, when I tried the Kernel switch on my Intel D510 cMP system, everything seemed OK though it wouldn't work on my Fit-PC2 setup. Fiddling with this and that to get to grips with the issue, I renamed Neptune.exe and deleted the switch on the Intel board: the system booted and allowed VNC access over the LAN but would not logon to the music "server".
I'm wondering therefore if the replacement kernel was in fact not loaded when invoked by the boot-dot-ini switch but was when the XP file was replaced. In short, how can one be sure which version one is running - if at all?
I have tried renaming the neptune file to ntoskrnl.exe with ntkrnlpa.exe present in the system. The system will boot... ONCE. It was then that my suspicions became certainity - the ntldr doesn't only have a hardcoded feature to recognize the name ntoskrnl, but each version of nt, xp or w2k has its own ntldr + ntdetect.com that work only with their native kernels. others will boot once for debugging session. I have compared sizes, and the ntldr and ntdetect.com sizes are different for all versions. I have tried using nt3.5 ntldr in xp, The system won't even start, which is easily explained, because not only kernel, but also, HAL.dll is involved in the boot process, and while kernel is the same for all versions (the size is different because of unnecessary crap), the hals are different. So, the only supposition was that the date-time stamp on ntldr and ntdetect.com are the culprit. Also, any ntfs.sys or ntoskrnl error is "spoiling" the nice pair - ntldr and ntdetect.com. We greately underestimate/neglect the role of these two. In fact, any burnt down Windows system can be easily repaired with a little manual editing of the registry and overwriting ntldr-ntdetect.com pair with a fresh copy from the iso. So it goes.
You can say that the new kernel WAS working, if the system doesn't boot immediately after the experimental session, if the sound quality change is not enough.
Serge.
Certainly, I haven't tried the thing in Your setup configuration...
Edits: 04/23/12 04/23/12
Hello, everybody. You know how I was all time at war with drmk.sys. Here's the way to kill the thing.
I describe here the procedure for Lynx (including Aurora, I presume), but I have small doubt it will work for anyone, who still has portcls.sys in his system. It's this file, that won't initialise without drmk.sys. In order to, just:
rename your current portcls.sys to portcls.old and
insert portclis.sys from WINDOWS 2000.
Rename drmk.sys to drmk.old. Reboot.
If the system plays, delete the drmk.old
Listen to some free music at last.
Serge.
I have this stripped and ready w2k file. Just e-mail me for "details"
This tweak (yes even for us Jwindows types) is a revelation. Try it and although Serge indicates it takes a while I noticed the effects right away. Highs become more articulate with a loss of egdiness I did not know was there until I tried this. Another masterpiece from Serge!
Dear Forum,
For your information :
It may come to no surprise to you (what does in cMP regime) that a recent upgrade from the GA-H55M-UD2H with i3-540 to Gigabyte GA-H61M-D2H-USB3 with i3-2120T has made a really significant improvement in especially micro-dynamics and solidity of the bass performance.
Needless to say that all the other hardware remained the same (RME 9632 with AES/EBU output).
kind regards
cMP2 Computer w/digital crossovers & FIR filters > Lynx Aurora 8 FireWire /192kHz all channels. 2x AcousticReality Ref. 202 & 2x AcousticReality Ref. 601´s ICEpower. Magnepan MG3.3R beechwood frames & custom stands. Miller chokes
I recently ran across you post and found the Gigabyte H61 board still available at Amazon. I also installed the 2120t because of its 35w low power and the Seasonic fanless 400w PS. Next I minimized windoze and yes it sounds way better that my previous 2 servers.
Many thanks!!!!
Hi play-mate,
Why not the GA-H61M-S2P-B3 ?
That one has even less hardware options like DVI and HDMI.
And only 2 mem slots.
And is cheaper.
kind regards
HUH? I thought you used Lynx. Maybe this is a quote from another post?
Anyhow what current mobos are there that will work for cmp2? The one in your post is discontinued, and I cant seem to find ones that have onboard video.
Afterwards we discovered faith; it's all you need
Hi Dawnrazor,
I do indeed use the Aurora, but we´re a number of cMP2 users here.....:-)
And we configure/compare a few mashines for a growing "community", so I thought it would be useful for others to know.
The improvement is significant !
BIOS sttings are a little different, but we have captured the settings that was published by JackWong recently.
In the new i3-2120T CPU the graphics are performed directly in the processor, so the motherboards do not have a graphics chipset anymore.
The GA-H61M-D2H-USB3 is currently avaliable in Europe.
kind regards
cMP2 Computer w/digital crossovers & FIR filters > Lynx Aurora 8 FireWire /192kHz all channels. 2x AcousticReality Ref. 202 & 2x AcousticReality Ref. 601´s ICEpower. Magnepan MG3.3R beechwood frames & custom stands. Miller chokes
This happened during slimming of the software or system registries. Selection of All in cMP shows all files are not in alphabetical order. Anyone know the cause?
When I used Resource hacker I had this result, too. It damages as it removes things. Use Resource Tuner, if I recall the name correctly, which can do slimming of cMP without doing this sort of damage. (This all has been posted previously.)
THANKS again for the reminder.
Amazing how quickly one can forget!!!
GLAD you didn't.
Resource hacker also messed up the text in the title bar which I had managed to restore.
Now I am back to the original back-up installation to resolve the menu order. At least can go through it faster than when I first started.
I remember posting something to the effect that YOU were correct in your assessment of the benefits of PE vs RS!
Thanks for the reminder!!!!!!
Hated the idea of having to restore PartMgr ...
and have just got used to it.
Grant and I were talking about this last night and the thought occurred that PartMgr, if restored to the registry (Services) and the .sys file to drivers.
Did not have a chance to try this last night so I cannot say it is the cause but I will try it tonight.
Now that I have achieved a Jackwindows system (16mb windows and 32kb system hive) this tweak adds significantly to this already great sounding system. It is simply putting all the cues in a folder (call it cues) on your data disk(s) then setting your cmp to only look at this folder within your data disk(s). With this arrangement cmp is only storing cues (very small in size) versus storing cues and wav files (very large in size). This tweak was originally identified by Ryelands and is still an easy one to implement. The only work is putting an absolute path into the cue file on the file line. So for example lets say you have a cue file with this line and your data disk is the D drive.
File 'amos lee.wav' wave
It simply needs to be amended to
File 'd:\amos lee.wav' wave.
Try it on a few cues then amend your cmp settings to only point to the cue folder...you may be surprised.
this result. It has been said that all the cues addressed in the cMP library window are retained in memory. My library comprises 2221 albums. The size of my separate cue folder is 5.6MB. So that is the most that could be occupying RAM during playback. But from what I observe it appears to me that only the cue file names are kept in memory not the entire content of all the cue files. I have this impression because sometimes a new cue file won't play in cPlay due to some error in it. I edit this cue file on my network drive from my desktop computer while that cue is already loaded into cMP with all the others, then I double-click the same entry in cMP and now it plays in cPlay. Doesn't this have to mean that when selected in cMP the cue contents are called up anew by cPlay from the changed file not from the original content of the file in memory, meaning the original content was not in memory?
. . . it appears to me that only the cue file names are kept in memory not the entire content of all the cue filesI tested this by pointing cMP to a cuefile on my "server", editing the cuefile then launching it from cMP without hitting "Refresh". The edited file played fine and the changes I made were displayed correctly. IOW, it appears to me that you're absolutely right - thanks for the correction.
Edits: 04/11/12
You are welcome. So, "how do you explain" the difference in sound knowing that?
And what does IOW stand for?
how do you explain the difference in sound . . .
Not having used cMP's UI for selecting files for many years, I only have the word of others that there is a difference. As to why, I haven't a clue.
what does IOW stand for?
In Other Words
You may be right but how do you also explain that putting all cues on the ssd c drive is better sonically than all the cues on a data drive?
There may be a simple rationale but in the interim I'll leave all my cues on the ssd.
That was my point. It doesn't seem to have to do with how much ram is occupied with cue data. I can't explain either why a single cue sounds better or why many cues on the ssd sounds better than many on an external drive, which I also find to be true and which clearly does not differ as to how much ram is occupied. I have used them on the ssd for a long time but actually have reverted for most of the time to keeping them on the network drive with the music files because it is far more convenient--no continual updating of the cues on the ssd as music is acquired and ability to edit cues while in cMP without transfers.
I have never done it any other way and have described this way here many times. It is faster and better though to put the folder containing all the cue files referencing the data drive on the cmp drive. This way when cmp starts it loads much faster and Ryelands first reported that it sounds better this way than having the cues folder on the data drive, though it means updating the cues folder frequently as more albums are acquired. Keeping it on the data drive means it is up to date always as cues are created.
Also, Al Jordan's recursive cuesheet creator can be set to automatically create cues with absolute paths for your whole library in a separate cues folder--Steppe take note. You can leave the relative path cues in place with each album, if you want, or search the library for cue files and delete them all at once and use only the separated cues.
Whether the music files are in memory or not (the logic is strong that they are not) it still makes a difference apims (as perceived in my system)if the cues only are in a separate folder and cmp only points to it.If you only have one cue being pointed to it makes a bigger difference yes. But in my current setup I have anywhere from 250 -500 cues per disc (discs are 160-320 gb)with 3 different discs. Since in my current setup I can only have 1 disc at a time in cmp it is good ....apims. When I was running windows xp pro with a 350 kb system hive the system sounded better when only one disc was loaded. So for me it is no problem to have only one disc at a time loaded. This is one of the advantages of having a network whereby you can truly minimize how many cues/discs cmp works with. I guess one could reload explorer then do it per the ryelands method. One has to assess whether adding explorer and whatever else may be required to do this degrades sq less than the added benefit of the 'one cue at a time' method.
Edits: 04/09/12
Well maybe you were the 1st then. Kudos to you.
This tweak was originally identified by Ryelands . . .
Hmmmm. Not by this Ryelands, it wasn't. Not quite.
I've never advocated having all of a library's cuefiles in one folder though my original motivation for not doing so that was that I couldn't work with cMP's UI. I still can't but I now also think that browsing for data should not be performed by the music player or shell for "sound quality" reasons.
1. To clarify, by "absolute cuefile" I mean one whose text includes the full path to the target data (e.g. M:\Jazz\Duke Ellington\Duke Ellington - 1960 - Piano in the Background\01 - Duke Ellington - Happy go lucky Local.flac) and by "relative" I mean one that has the filename only. Obviously, the latter must sit in the same folder as its data;
2. As we all know, when "cMP" is selected in cPlay's settings dialogue, the player loses its browsing capability and is launched by clicking on a target cuefile in cMP's main window;
3. cMP sees cuefiles only; it does not load music data into RAM though it does load every cuefile in the "watched" folders selected in cMP's Folder > Add dialogue. The watched folders are scanned at boot time and when the "Refresh" button is hit. If there are many relative cuefiles, this can, as riboge notes, take several minutes. If the files are absolute and all in one location, it is significantly faster;
4. That this is how the two programs work is shown by cuefiles that "point" to data that are in fact missing. The error is recognised only when cPlay is launched, not before. It follows that cMP does not load music data or check cuefile integrity (there being no compelling reason why it should);
5. However cMP is configured, "watching" a large number of cuefiles eats up significant RAM, in my case (a medium-sized collection of a little over 6,000 albums), probably about 30 MB. Even for a small collection, it's unlikely to be less than an MB or two.
Given the heroic efforts of pioneer "slimmers" to reduce OS footprint, it seems to me worthwhile to see how this footprint might also be reduced. Ideally, cMP should load the target cuefile (typically < 10 KB) only;
6. The easiest way to do that is, it seems to me, to copy the target cuefile only to the one folder that cMP is configured to "watch", to use cMP's "Refresh" facility to recognise when the file is changed and to delete the file once play is finished. Obviously, the files have to be absolute but such are easily written with AJ's "CueFileCreator";
7. It's certainly easy (better, in my view) to work this way in a networked cMP setup without loss of SQ. I'm not sure that it's quite as simple on an isolated cMP2 box where Sonic Trappists have long since sacrificed File Explorer in pursuit of SQ (OK, OK, I'm one of them). If anyone's interested, I'll happily explain how I do it but, if no-one is (fair enough) I'll not bother.
Dave
Hello, please tell us.
Serge.
That might be misunderstood but I'll assume it's audio we're discussing . . .What follows describes how I configured my "headless" cMP2-with-networking setup to ensure that cMP never needs to load more than one cuefile into memory.
I developed the scheme (which went through several different permutations but which I haven't changed for over a year now) to get round cMP's restrictive UI which IMHO is simply not up to handling a medium-to-large library. It was only later, after a series of exchanges with theob, that its sound quality advantages became clear. Well, to him at least - I don't really have anything to compare it with.
My music library resides on a shared drive on a desktop "server" which I also use to run my two cMP2 boxes. It should not be hard to adapt the routine for different scenarios as it relies on nothing more than cuefiles with absolute paths (which many already use in any case), a few shortcuts and a little batch file.
++++
1. If you don't already have one, create a library of cuefiles with absolute paths separate from main music library (the boring bit).
Tip 1: use Xcopy /T in a command window quickly to replicate the source library's directory structure and the "Set Output Folder" facility in AJ's CueSheetCreator to create cuefiles with abolute paths.
Tip 2: Use TextCrawler to edit cuefiles in bulk if e.g. you move them to a different drive or just screw up.
Tip 3: use the same character (typically M:) for the desktop's drive letter and the cMP2's network share so that the cuefiles are valid on both.
Tip 4: be sure to make regular backups of the new cuefiles.
2. Create a folder called M:\00. You later set cMP to "watch" this folder, i.e. to load its contents on boot or after a "Refresh".
3. The cuefile library is not seen by the cMP2 box and can sit anywhere on the desktop box. Browse it to select an album and copy its cuefile to M:\00.
Tip 1: for easy browsing, create a series of shortcuts to your library and put them in a separate folder called (e.g. and with imagination) "Music". This folder acts as a proxy for the cMP UI with the benefit that it doesn't depend on accurate tags to maintain structural integrity, is under the user's control and, crucially, doesn't gobble up resources on the cMP2 box.
Create a shortcut to it, give it a distinctive icon (so you can find it quickly) and put it on the QuickLaunch bar. To browse, click on the icon, select your cuefile, right-click on it and use SendTo to copy it to M:\00. Pictures H/W. You can then close the Explorer window.
Tip 2: create a batch file called "00-ClearList" (the 00 bit ensuring it appears at the top of the window) that reads "Del M:\00\*.cue /s". Put it in the same folder as your shortcuts - see above. Double-click on it to remove cuefiles from M:\00 for albums you have finished playing.Tip 3: create a shortcut to M:\00 and put it the "Send To" menu - set Explorer to show hidden files and put the shortcut in the user's SendTo folder in "Documents and Settings".
4. In cMP's "Settings" dialogue, clear the "watched folders" list and add M:\00 to it.
Tip 1: to avoid any error messages on caused by cMP not seeing any cuefiles (e.g. before logging on), create a dummy cuefile, put it in C:\ and add C:\ to the watched folders list. To make a dummy file, copy these lines into Notepad:
REM GENRE 0
REM DATE
PERFORMER " "
TITLE " "
FILE "Theme.flac" WAVE
TRACK 01 AUDIO
TITLE "1"
PERFORMER "2"
INDEX 01 00:00:00and save the file as Dummy.cue. At about 150 bytes, it can safely be assumed to have no effect on SQ.
5. Back on the cMP2 box, even after you replace a cuefile in M:\00, cMP still reports it (demonstrating again that it does indeed hold cuefiles in RAM). Click Refresh to lose the old cuefile and make the new one visible; play it as normal.
I quickly got used to this routine and find it faster and easier than messing about with cMP's UI. Aside from any SQ benefits (for which I have only theob's word as I haven't used the cMP UI for years), I do not intend to return to the old way.
Anyone interested in trying it is welcome to come back with questions where my notes are not clear.
Now, off to Evensong . . . Trappists have no time to gossip.
Dave
Edits: 04/08/12 04/08/12
Thank You!
Serge.
Tell, please.
Dying of suspense here.
Expecting your normal clever, lucid directions... don't disapoint!
Greg in Mississippi
Everything matters!
hello, Ted! Hello, everybody, Just like You had suggested BEFORE, I tried to do the following:
I renamed all my 5 folders, containing my library and made an experimental 001 folder with only one album folder inside:
D:\001\Diana Krall\The Look Of Love. Started cmp with a single cue.
The quality gain was great.
But now to Your current tweak. Do I only have to point to drive C: or do I have to change the line for EVERY ONE of my 10 000 songs like below:
FILE "D:\RIP\CREEDENCE CLEARWATER REVIVAL\1970 Cosmo's Factory\06 Run Through the Jungle.wav" WAVE
If that's the matter, I'm afraid this tweak is not for me, unless I'll reRIP all my library...
Serge.
But the tweak WILL WORK, especially, if You have cue files pointing to smth like Image.wav
I'm afraid this tweak is not for me, unless I'll reRIP all my libraryYou don't need to re-rip any data to implement this routine though you do need to re-create your cuefiles to use the same routine as me. How long that would take obviously depends on how many albums you have but, for a thousand or less, I reckon it'd take a couple of hours including getting the hang of Al Jordan's CueSheetCreator.
But I really do owe grovelling apologies all round. There is another perfectly viable routine which I'd completely forgotten about which would at least allow you to try the idea at length without changing anything and may well be (as the adverts say) all you ever need.
Using cMP's Settings > Add/Remove Folders dialogue, temporarily clear the watched folders list then browse to and Add the folder - and only the folder - containing the target cuefile (and, of course, the music files though cMP doesn't see those). Repeat this every time you select an album.
If you don't have a relative cuefile in each album folder, CueSheetCreator can create them in, literally, minutes with almost no intervention. Point it to your music library's root folder, deselect "Use Custom Output Directory" and then hit "Process Library". You're done - a realtive cuefile in every folder. (To remove them, open a command window in the root directory and run: Del *.cue /s.)
Unlike the first routine I described, where you are copying the cuefile to where cMP expects to find it, you are instead pointing cMP to the cuefile. The effect is the same - cMP loads one and only one cuefile. The difference is that the second routine works fine on non-networked machines. Hair-shirt ones, even.
I'd be most grateful if someone could give it a try and report back any effect on SQ.
I used it to browse for music for at least a couple of years after cMP was published, having first spent several fruitless hours trying to make cMP's UI work with my library.
A friend for whom I built a cMP2 box has been using it ever since I handed over his system nearly three years ago and sees no reason to change. (His music is stored locally whereas mine is stored on a networked drive.)
HTH
Dave
Edits: 04/09/12 04/09/12
Hi all,
I did some tests on my (Atom-based, no-network, usb-soundcard) machine.
I used to load all cuefiles from the data-drive. Now i load individual cuefiles from CMP-drive, this turned out to be the best SQ, thanks guys!
in order from best sq to less good sq:
- single cuefile from cmp-drive
- all cuefiles from cmp-drive,
- single cuefile from data-drive
- all cuefiles from data-drive
I keep a backup of all music files in my desktop computer. The musicfile-drive has the same driveletter as the datadrive of the CMP-machine. I make the cuefiles on my desktop, and after that copy the cuefile to the cmp-drive.
Douwe
Just to explain my background on this saga, I started out putting all my cues onto a folder in my music drives. It sounded better than leaving all the cues on the music drive. Then I migrated to loading only one music drive per listening session after I took one of them off line (to load new music) and found out that sonics improved. So then after Douwe indicated that putting cues in a folder on the cmp drive was better yet I tried that and confirmed that yes this was better.So never one to leave well enough alone I tried organizing my cues into sub-folders then loading only 1 sub-folder onto my cmp drive per listening session. I did this with my favorites at 1st to see how having only 8 or so cue files on my cmp drive would sound compared to the the 300-400 cues I used to have when I loaded each of my music drives (5 all together). Doing this made a significant sonic improvement yet again. Maybe not as good as one cue on the cmp drive but certainly better than the 350 or so. Yes this means everytime I want a new group of 8 cues I have to shut down, pull out the cmp drive and load new cues. But since all of my drives are literally outside my cmp box this is not an issue. If anyone wants to try this remember you have to have cues loaded into another pc so you can access them at will. But the data disc since it contains 350 music files does not have to be reloaded.
Yes Dave your idea of 'putting all data on another computer then network everything' while loading one cue at a time is the best . My way is more complicated but results in approaching one cue file at a time on the cmp drive. And at least for me the loading of 8 cues at a time is not too cumbersome.
Why this makes such a big difference sonically is still not clear but it sure works good.
Edits: 11/03/12
Just tried it and it does improve sq significantly yet again. Thanks!
Dear Inmates,
Before attempting to install the JW-XP1 ISO and performing the nessesary USB reinstallation/adjustmenst, I wonder if it would not be easier to re-distribute the corrected OS on ISO again....
I should think we could end up with a few different ISO files for different cMP2 applications.
kind regards
cMP2 Computer w/digital crossovers & FIR filters > Lynx Aurora 8 FireWire /192kHz all channels. 2x AcousticReality Ref. 202 & 2x AcousticReality Ref. 601´s ICEpower. Magnepan MG3.3R beechwood frames & custom stands. Miller chokes
Now that explorer is not available in the 16mb JW Home Edition SP1_2, new folder cannot be added from cMP.
In cMP mode it would not be able to regconize a new disk, and further more pnp functions are disable.
Hello dear forum members,
Great to see you have solved a problem I have as well.
Just too bad that to me about everything you discuss could as well be Chinese, which I don't understand as well.
SwissKnife, Jackwindows, shlwapi, original shell32, edit Your cmp registry and reboot immediately, NTREGOPT , 16mb JW Home Edition SP1_2...
All this means what exactly?
I also have my music on an external HD, waiting to be transferred to the internal drive.. but why does my PC not see the drive? It did before..
Thank you for your extra help!
Sincerely,
Gaston
Hello! Get Yourself this thing - It's an external ESATA-USB hdd box from, say Thermaltake ($35-40). It allows You to attach Your hdd to cmp machine as the main or single hdd and AT THE SAME TIME attach it to USB of another machine - the one You rip or edit registry on. When You listen to the music, Your ripping (working) machine won't see this hdd, but soon as You power down Your cmp, the working machine immediately recognizes this hdd as another drive.
ALSO IMPORTANT: get Yourself a ESATA to SATA cable ($6-7) It will allow You to attach Your now externally powered cmp hdd to a sata receptacle on Your CMP mobo, which is important to further slim down the registry.
This combination works like a charm and is VERY convenient, as not only You will be able to add files to the library, but edit Your cmp registry and reboot immediately. Also works with NTREGOPT very well.
Serge.
Thanks Serge.
I have a similar one of this for my music files connected via eSATA to the H55M UD2H mobo(Drive D:\). cMP is in another SSD(Drive C:\).
Drive D is now not detectable after files deletion and substitution.
I believe your cMP and music folders are on the same external drive. Is that correct?
Regards
Try putting back in original shell32, shlwapi then go to cmp screen settings, add discs/folders. Then add D:\. Verify it plays files off of D. Then put back slimmed shell32, shlwapi.
Thanks for responding theob.
Did as you suggested. Can see the drive D in cMP setting but when attempt to click on the expand(+) icon, it said "Drive not formatted".
Can add drive to the list but it would not detect it.
Is it formatted the same as c drive? Are you in Jackwindows?
It's a different format. C is FAT32 and D is NTFS. I was trying to format to the same but Windows doesn't allow that. SwissKnife only format about half of my 1TB in FAT32 and leave the other half unformatted. I am now using command prompt to format it to FAT32 and it's darn slow.
Yes I am in Jackwindows.
Let's see when I am home this evening..
Best Regards
And it was!
Finally use Easeus Partition Manager to format my 1TB external HDD to FAT32 format for the music folders in a couple of minutes after spending the whole day with dos command line which doesn't work.
Once again thank you theob for the guidance.
This nice piece of software is available at the link below.
.
And now can have a go too on the idea of having only cue files in cMP drive.
I am planning to avoid esata. I have finally persuaded our local computer store to bring in several esata-to-sata cables and I am going to attach today this hdd to the onboard sata (blue socket), not to esata external terminal.
Serge.
Also, as I have a single hdd, the cmp case is very "empty" and is perfectly ventilated, and the hdd itself is the ms-dos boot device, which is necessary for it to be recognized by the registry.
that it can only be a single drive machine!
Am I correct to say this?
BTW, I had unknowingly bought a SATA to eSATA cable adapter for a few bucks some years back. Good for my H61M S2P mobo.
becoming typical, no matter how many drives. The number of files in windows\system32 and\drivers and fonts remains the same.
Sorry to answer only now.
I have finally got my esata-sata cable, The box works perfectly as a channel 0 master, So I'll start to experiment with JW ISO these days.
Serge.
Could had just reinstate ntfs in the registry to see whether it would works, but had since reformatted my external drive to FAT32,and with some hints from theob, managed to add the drive to cMP.
My Windows folder have an additional folder(Cursor)with just the arrow_rl file. I have problem with mouse pointer visibility on my screen even with the original cMP Windows.
Glad to hear that you have got your cable. Your method is definitely offers a lot more convenience than having to move the disk around.
Regards
Hi, I downloaded the sse4, when I opened it, there was a message:
No valid ASIO driver found.
I am using Windows 7, USB output to Peachtree audio decco2.
Anyone has the ideal how to set it up properly/
Hi jimxing149@gmail.com
I guess the Peachtree is not an ASIO capable DAC, so you have to install asio4all to get it working.
kind regards
cMP2 Computer w/digital crossovers & FIR filters > Lynx Aurora 8 FireWire /192kHz all channels. 2x AcousticReality Ref. 202 & 2x AcousticReality Ref. 601´s ICEpower. Magnepan MG3.3R beechwood frames & custom stands. Miller chokes
I hope someone is. I'm sure there are a good number of people now unable to use it without at least one of these added. It is too much for me to take on given that I will still have to adapt the result of that for firewire and Weiss dac driver.
I second riboge's request!
I believe that JW is perhaps working on it. I hope that both Serges can help him and achieve an ISO perhaps by using a XP SP2 (or whatever is necessary)...
Normally, one should be able to substitute all files in system 32 in one attempt and then all files in drivers.
If this can't be done, in the first attempt I recommend putting the following set of substitution file in system 32:
Once you've successfully substituted the bulk set for the micro-shell32, the rest is simple:
- authz.dll, basesrv.dll, bootvid.dll
- csrsrv.dll, csrss.exe
- the rest in alphabetic order.
I am able to substitue all the dll and exe files in system32 all at one go. Doing only the files associated would generate an unknown hard error either for lsass.exe or cicsMemoryPlayer.exe.
But I am unable to do SRC45 @192kHz. Even using sOX192kHz would generate pops and clicks. I suppose these are latency issue.Ok if I am at natvie resolution.
What would be the PCI latency setting using Ltfcy2.3? I am at 128.
Regards
I use H61M mobo and need not use Ltfcy to set latency. In addition,I do not use 192K and have not experienced the problem you encountered.
If it will work at 192K at native resolution, what's the problem? I recall that H55M-UD2H should be set to latency of 128.
You may want to download from the link. It contains settings for Juli@ service, Enum\PCI entry and Class\Mixer setting, where the latter (Mixer09) is necessary for Juli@ to run at latency of 48 sample. Perhaps this may be the cause.
Thanks for the files.Just a small problem. Had latency set to 128 earlier when I was still in XP mode.
Now I only get a lazy mouse and a slow screen refresh.
Edits: 04/03/12
It will get better in time.
Without processr.sys in System32/drivers/ cPlay was unable to function properly-like there isn't enough processing power.
Do we need this processor driver?
after deleting this file. Do You use flack? I use only wavs, but I think, it's not the cplay problem. After You delete all unnecessary files You will get a 16 mb windows that will fly like a butterfly.
Serge.
There can be something in the registry, that is "holding" to processor.sys
search and try to delete this entry
Hello Serge
Thanks for the response. I am using flac as well as wav. I believe it is not cPlay problem too, It is the CPU problem. Anyway, I am keeping the processor drive in the system for the moment.
Will try deleting it upon completion of registry slimming-still got system, sam and security left.
Now is at 30.8 mb so still got more work to do.
Regards
I could not use Jack's ISO as I am USB based, but am using it as a best practices template to create an XP SP3 Home platform on FAT32 with USB.
I am well down the path. Sound improvement of Home on Fat32 nothing short of spectacular. Couple of questions on System Hive slimming.
1. In Jack's guide, #53 removal by Name and Type:
I can safely remove:
Class
DisplayName
Description..
Can I also remove:
Group
ErrorControl (assuming value 0)
Mfg
Service
LocationInformation?
Anything else?
2. I seem to get a lot of things reappearing under Control in System hive even tho I have /mininit specified. A whole host of things like Watchdog for example reincarnate on demand. What am I missing here?
I am down to 56kb for the system hive, but still can't delete controlset002/control/class.
What could be the problem?
Hi JackLatest config folder that works sent.
Regards
Edits: 04/04/12 04/04/12
add "Everyone". Then delete from the bottom up for both entries.
That's exactly what I did but no success till now.
Can you send the hive to me?
Thanks Guys, very helpful as always
Jack, thanks.. I am using your hive as a best practice reference. These additional tips are helpful.
I find that System stays at 512 for long time and then drops to 256 once enough is removed. I'm using NTREGOPT as the compression. Is this normal?
Rick, I have removed the rest of the folders and all the deletions in steppe 1-17 etc but so far have had alot more trouble getting to your level of slimming. I had no trouble in XP Pro, but getting lots of BSOD in XP SP3 home. I'll keep at it, but it looks like many things can kill it. Also very easy to lose the mouse cursor with USB mouse.
Did not realize importance of System Volume Info. Luckily its empty.
Ryelands.. Not a typo.. I had the /minint wrong. Worse, somewhere recently on a recovery of a partition I lost this from Boot.ini and did not realize it. This likely explains the reincarnation of lost entries.
BTW look forward to compare notes on WaveIO. I think its awesome. It too likes good power. I now use a choke and a big can cap as the first filter with BGs on the board. Last nite I exchanged the choke and got to new levels. What was the choke? I used the primary winding on the 1.5kv transformer from a cast off microwave oven. Basically a huge chunk of iron with very low resistance primary. Free and awesome upgrade. I was about to order a Voyage MPD to drive it when Jack came up with this new XP approach. So I want to perfect this first and then compare to minimalist Linux.
Never come across such drastic drop in size. If you have not check each and every subkey to remove the system account, you must do so. It's the only way to stop system intervention.
Thanks Jack. With all the subkeys with "everyone" in control and /minint specified correctly the respawning of keys has stopped. I have System hovering around 160k (config backup #62). Have not yet cracked SAM/Security without hitting 'hard error' on boot. I can't seem to get micro shell32 in, but have used resource tuner to strip everything possible. Have used some of your collection per Serge earlier recommendations. Will take another shot at the full collection.
The sound? Quite incredible actually. Just plain natural with great extension in high and low, crystal clear 3d. Just very natural. Find myself just listening to the music, not listening to the system.
Thanks for sharing your approach. I could never have gotten to this stage without your help and samples. You, Rick and Serge have been so helpful.
I have to figure out an issue with my I2S converter. On exiting CPLAY, it loses its ability to sync up. I have to power cycle it between cue sheets. It is as if it expects some kind of clean stop/restart that it does not receive in this config. No problem in other CMP configs. Likely an issue with Theyscon USB ASIO driver. It came at some point in registry slimming. I'll keep at it. Not possible to reinstall driver. I might try deleting and reimporting all the keys from the hive of the work partition.
Hi Walter,
I've post some recommendations on file substitution.
At 160kb, you are only scratching the surface. Once get down to sub-100kb level, he would not want to listen to a 128kb system hive.
I recommend using Resource Hacker rather than Resource Tuner. The latter offers beautiful presentation on what it is actually not capable of - it hides things rather than removing them.
I always say the credits must go to Serge(Steppe). He is an inventor, I just rode on his wheels to try to improve things. Rick has always been selfless. He is always innovative with his unknown but effective methods in doing things, which helped get pass a lot of hurdles.
I too experienced frequent lost of Musiland USB driver in earlier setups other than SP1. You may want to import/retain the entries for PnP, USB, and the sound card and try deleting them carefully at the end.
Deleted entries reappearing is normal in a regular Windows.
The most important thing in successful registry slimming is to get rid of the system account. You need to go into each and every key, sub-key and sub-sub-key to remove "System". In the meantime, retaining "Everyone". Do this not only to the system hive, but also to all other 4 hives.When you arrive at a certain point or sufficiently small system hive, the system will give up struggling. How small would that be? I believe when the system hive is near 100kb.
Yes, you can remove:
Group
Mfg
LocationInformation
ErrorControl (except for PlugPlay. For USB device and your USB soundcard , you need to check whether the entries can be removed, these cannot be removed from my PCI and soundcard serivce)More, you can remove:
tag
all entries inside GroupOrderList
all entries inside ServiceGroupOrder except Video SaveYou should not remove Services inside CriticalDeviceDatabase
There are much more that you can remove but I don't know what state your system hive is. Therefore, I've uploaded my config folder for reference by those who are engaged in registry slimming.
Edits: 03/30/12
My guess is you still have those folders in there that restore stuff.
Make sure the System Volume Information has nothing in it but logs.
Remove all of the folders in WINDOWS but config and drivers. Many of those folders contain what is needed to replace what you delete.
Have you done this?
What am I missing here?
Just in case it's not a typo, the correct syntax is /minint, not /mininit.
Out of interest, why SP3? As you know, fairly soon, I'll be using the same USB-to-I2S card as you so I'll be following your efforts with considerable interest.
Someone has warned me about offering pirated software over this forum. He is right, I will cut off file sharing in eight hours.
Moreover, I have amended the first post to reflect the limitation of the ISO and the difficulties involved in the implementation. Sorry to have caused such confusions.
Mouse went missing after booting into cMP Windows. Any help would be deeply appreciated.
Oh Well! The answer an be found here!
http://www.audioasylum.com/forums/pcaudio/messages/10/105820.html
Amazing! even with just the fresh install. Set no paging file. Disable error reporting. Adjust for best performance.
Many thanks to JW.
Sadly after replacing the 2 amended files still got just black screen and no mouse. Start Menu is there though.
Any thought on this?
Are you using PS2 mouse? Try this.
Download http://dl.dropbox.com/u/64313985/CriticalDeviceDatabase.reg
Load your cMP System hive in the name of "Test" and then execute CriticalDeviceDatabase.reg.
Don't know the cause of no mouse. Might perhaps due to massive removal of CriticalDeviceDatabase.
Thanks for the prompt response.
I had check the contents in the downloaded file CriticalDeviceDatabase.reg and compare them to my cMP system hive. All the entries are already there.
Let's see how it goes with the two new amended files.
Best Regards
Did you make any progress with your PS2 mouse issue? I have exactly the same problem at the moment and I’ve had to re-load my original cMP/cPLAY installation while I search for a solution. My hardware is Gigabyte GA-G31M-E2SL /Intel E7300 which I think is fairly standard-issue for cMP.
For myself, I think that a compromise between ultimate sound quality and usability might be needed. Restoring the USB capability might be my only way forward.
Hi Andy
Yes! Mouse problem resolved. Your mobo have both the P/S2 sockets for mouse and keyboard so I don't see any issue with your installation. Devices need to be present in the installation for them to work later in this case so you have to plug in the p/s2 mouse during installation.
My H55M UD2H board have a shared socket for the p/s2 mouse & keyboard so I have to enable USB in BIOS and make use of a USB keyboard.
But I am now having problem with the substitution of the micro shell32.dll. Error msg. ordinal 644!
Good luck.
One should always stay in cMP mode after the first boot. One should be able to substitute all files at once in cMP mode and with the software hive reasonably slimmed.
The error described will appear when the micro shell32.dll is substituted and some of the conflicting entry points relating to shell32 are still present in the CLSID of the Software hive.
The error described may also appear when the micro shell32.dll is not substituted together with some associated files.
Thanks Jack
What are the other associated files to be substituted with the micro shell32.dll? shlwapi.dll?
.
.
.
.Can't describe the files, it consist of multiple tiers, just too long to list. If interested, go check for yourself with PE Explorer. All-at-once substitution is recommended.
.
.
.
Edits: 03/30/12
Your information about the need to connect the PS2 mouse before installation points me to where I probable went wrong. I connected it afterwards. I’m going to try again.
Thanks for your help. It could save me from going around in circles!
Another thing-get into cMP mode(Start cMP) before you star deleting folders and files in Windows else you would have lost the start Menu.
Edit: I have under estimated the complexity of this project and the difficulties this would present to beginners, In order not to give any false hope to inmates, I have amended this post to such effect. Sorry for the inconvenience caused."After switching to XP Home few months ago, I've started creating my own XP installation, not NLITE though. After some learning process, I have successfully incorporated registry slimming and tweaks into an XP HOME ISO. The ISO offers a big jump in reduction of registry size upon first boot. After implementation of tweaks, I am extremely satisfied with the sound quality produced by the current installation.
During my lone journey, I have invited a couple of experienced inmates to try the installation. After they did they unequivocally agreed to the supreme sound quality offered by the ISO and have parted with their beloved NLITE XP PRO.
Sadly, the current ISO is custom built for exclusive use with Juli@ and possibly LYNX L22 & AES16 series sound cards, there is no NETWORK or USB function either. The Windows does not Plug and Play in a traditional manner because too many functions have been removed. As such, it may not work for non-Juli@ sound card.
Due to complex nature of the current project, it is considered that only those who have good experience in computing can successfully achieve the desired results. As such, I am only inviting interested inmates who have sufficient knowledge in computer to join the trial, preferably Juli@ users. Upon request, I will send him an invitation to join “Dropbox”.
Please bear with me that I cannot walk one through or spoon-feed beginners to do such complicated task. This ISO is not for everyone, in particular beginners.Although I have tried my best to make things clear, the installation was produced during my leisure and is by no means professional. As such, the ISO is offered as is without any guarantee that one will successfully install or implement.
To safeguard my interest while offering the download, I will only make it available for a limited period of time and shut off accounts to access my Share Folder."
Edits: 03/20/12 03/20/12 03/20/12 03/20/12 03/23/12 03/23/12 03/23/12
Inmates have questions about the recomended Work OS.
I am suggesting to run a dual boot PC with:
C drive = cMP Windows with my modified XP iso installed
D drive = Work OS with a regular or Nlite XP(no specific requirement but system restore function should be turned off.)
The above requirement is not mandatory. One may use another computer to perform the optimiztion on the cMP Windows, which mostly involves file substitution and registry deletion.
ask why you have chosen SP1 and not SP2 or 3?
Apart from SP1 and SP2, I have also auditioned SP0 and some earlier beta versions of Whislter Personal. In fact, Cicsplay.exe sounds best in Whilster 2457 > Whilster 2526 > XP HOME SP0 > HOME SP1 > HOME SP2 > PRO SP2. However, Cicsmemoryplayer does not work in SP0 or earlier verions because it does not meet the requirements of Dynamic Link Library for the eariler OSes.
Not only does the earlier Windows verions sound better, I found that the earlier hives (config folder) also sound better. By saying this I mean I have migrated config folder from eariler Windows versions to SP1 and SP2 to make comparison.
Each service pack upgrade added additional security features and updated Windows with more features, in particular security, and software compatibility, all of which are not needed but detrimental to music playback.
Hey Jack,
I always used sp1 until the recent hardware with the i3 or whatever cpu and mobo. But I recall that it wouldnt install with sp1 and I had to install sp2 with the new hardware.
Are you using the latest hardware or the older hardware and will I have a problem with installation?
Afterwards we discovered faith; it's all you need
Hi Douwe,
It installed successfully on H61(Rick and mine), H67(mine) and G31(Ted & Grant) mobos without problem. i3-2120T, G620T and Celeron G440 were used.
Why not clear your doubts by installing on an unused HDD?
Is it possible to rstore usb functionality? They can be disabled individually later.
To fmak and inmates using USB soundcard etc:
The current Windows installation lose so much of the functional abilities that it no longer PnP or respond to changes. In fact the system account, which is always in chagre of Windows, is no longer present in most areas of the registry. That is why one can comfortably delete the registry without undue struggle or facing a lot of dead screens.
Despite I've successfully restored USB ports in an recent attempt, the Windows was incapable of PnP or any installation action, not even a HID mouse was detected. It is apparant that a USB device needs to be attached during set up. However, even when I did, my HID mouse did not work smoothly. One may recall that there was a major update on USB from SP1 to SP2, and I also recall that the usb in SP1 did not respond positively if chipset driver is installed. The USB on SP1 is not mature.
There are two ways to make a USB soundcard work:
(a) Inject the working USB registry entries to the current registry, same as what I did to revive a ESI MAYA soundcard after instllation;
(b) Compile a new installation capable of PnP and USB.
A new instllation will take time to compile and test, but I will certain try one. With network? Most likely!
Injecting registry entries is a viable option that one may try with the current installation. To this end, I will to try to revive a Cmedia USB converter during this weekend and let you know the results.
Dear super-inmates,This all looks like a milestone in the history of cMP2, but for those of us, who have not followed the hive/.dll mods in all details, it seems that we´re close to what we could call a : "ultimate cMP2 installation disk"
There are of course those who do use other DACs and have certain specific requiremets, but maybe we will (in time) assemble a number of versions catering for different setups : USB, Firewire D/A converters, VST capablillity etc.
Just a few questions :
The ISO file needs to be copied on a DVD/CD disk, right.
Will this be enough to run a proper installation upon inserting the DVD/CD in an attached drive ?
-how and why the disk media, and not just "imaging" the ISO ?Does the ISO have specific CPU/MB requirements, or can it be installed on any PC...?
Sorry, but there are just a few basics I need to get my head around before digging into an attempt.
Maybe someone feels for writing up a summary on all these facinating developments.
I´m sure this is necessary to let the advancements apeal to a wider group of cMP2 fans.Kind regards
cMP2 Computer w/digital crossovers & FIR filters > Lynx Aurora 8 FireWire /192kHz all channels. 2x AcousticReality Ref. 202 & 2x AcousticReality Ref. 601´s ICEpower. Magnepan MG3.3R beechwood frames & custom stands. Miller chokes
Edits: 03/22/12
Thanks for your appreciation!
Here my reply to your questions:
The ISO file needs to be burned to CDR disk with software, such as Nero, Deepburner free or Easy CD Creator. This will allow a proper installation upon inserting the CD in an attached CDROM drive which is set to boot in BIOS.
"how and why the disk media, and not just "imaging" the ISO"
Not quite understand this question, would you please clarify?
The ISO have been successfully installed on H61 and G31 mobos. I have no reason to believe it would not work on other modern mobos or a regular PC?
Perhaps the trial users can help writing up a review of the current XP installation in order to make a wider appeal to cMP2 fans.
Hello inmates,Sorry for my late turn up and my mistakes made during preparation of the ISO! I tried without success reviving USB when I found that the Intel Chipset driver was not properly installed and that I've put some over-hacked font into the ISO.
Both issues have been fixed. I've uploaded an update version of the ISO, named "JW_HOME_SP1_2 (Jack Wong 2012-03-21).iso", and put two original fonts into a folder named "XP Original Fonts". Please don't hesitate to download the update version.
I must apologize to Richard (riboge) for give him false hope about the installation, as I didn't make it clear in my post that there will be no USB function (I only mentioned that in the Installation Notes). I did start with a Musiland 01US USB converter but gave up as it needed SP 2 to work with. As such, I switched to Juli@ and chopped off all USB, leaving such function available for use only during setup.
Thanks Ted for responding to the questions of Dawnrazor. I would like to add that, USB will be functional only during the installation period and it was my personal choice to remove the function because I want to remove all unnecessary things including devices and registry.
Yes, if one decide to start the installation, be prepared to switch back and forth. How many time? It must have been well over a thousand times for Serge, Rick, Ted and myself, can't remember. I have over 130 successful hive backups, some are accumulative and the unsuccessful rate is much higher! Fortunately, the current installation will ease the efforts. One can slim Default hive entirely in one single step and Software in 3-5 steps to 12kb. When one starts with the system hive and first deleted ControlSet002, the hive will become less than 250KB, a value once some pioneers cannot achieve even after months of work.
One can use any Windows as the Work OS, but there are benefits installing the Work OS on the same HDD as the cMP Windows and on the same mobo. Firstly, no jumping up and down. Secondly, it's a must to look into the Device Manager in order to identify devices that are useful for retention and those unnecessary that should be deleted. One must be able to read the cMP hive when he compares the details(long list of number!) in the Device Manager! Do one really want to plug another HDD onto the mobo to do that? One may partition the first partition into 2 viable ones, e.g.1st partition 300 mb for cMP and the second take up the rest of the spaces.
Edits: 03/21/12
Just a warm reminder. Now that about 18 inmates have joined the Dropbox, and some still pending, to access the Shared Folder, I should remind inmates to take care when downloading the shared files, as they are needed by everyone.
Thanks
This may just be an issue for me using a laptop instead of a cMP box, but what connection are those running this jw os using for their music library--perhaps,sata or esata? I have recently been using network to nas and earlier external usb drive so am wondering how I could use this os without either. I use firewire to connect to a Weiss dac, so would not want to use firewire for external drive simultaneously--to say nothing of problem of adapting jw os for firewire to begin with.
Thanks for your work. You have my most sincere compliments.
One question:
do you know if asyncronous USB can work (normally it needs a .sys from SP3)?
and to Bibi01s question, what about usb mice? It says in the instructions:
There will be no CD ROM and USB function for the current installation
So are you using a keyboard for control? And if so I thought cics preferred usb powered separately.
Also when you say "Burn the JW_HOME_SP1_2.ISO on a CD as one would normally do." I am assuming that that means making an iso disk with something like image burn (http://www.imgburn.com/) and not just putting the file on a cd??
And the concept of a "work" os is new to me. If I already have a partition with xp cant I just install the iso on another partition? Do I have to put the iso on c where the current install is? I am assuming I am going to be switching back and forth right??
When you write: It is important that “System Restore” is turned off in the Work OS or removed entirely during Nlite process, so as to avoid creation of “System Volume Information” in the cMP² HDD. Is that where you use Nlite to turn off WF "something" that you need for minlog??
I am seriously playing catch up here and that doc was written for someone who was paying attention.
Anyhow thanks for your work!! This is the kind of thing that will be great for the ho hum inmates like myself.
Afterwards we discovered faith; it's all you need
"So are you using a keyboard for control? And if so I thought cics preferred usb powered separately."
No basically you use a ps2 mouse for control."Also when you say "Burn the JW_HOME_SP1_2.ISO on a CD as one would normally do." I am assuming that that means making an iso disk with something like image burn (http://www.imgburn.com/) and not just putting the file on a cd??"
Yes just use a good burner, I think Jack has a recommendation in his dropbox.
It took me a few days to get the system to play music well and a about a week or so to get it to the best front end I ever heard.
"And the concept of a "work" os is new to me. If I already have a partition with xp cant I just install the iso on another partition? Do I have to put the iso on c where the current install is? I am assuming I am going to be switching back and forth right??"
All that is meant here is have another pc (preferably one with xp on it) so you can do the hive editting.
"When you write: It is important that "System Restore" is turned off in the Work OS or removed entirely during Nlite process, so as to avoid creation of "System Volume Information" in the cMP² HDD. Is that where you use Nlite to turn off WF "something" that you need for minlog??"
No much simpler than that. Once launched and you start the hive editting process turn off System Restore in your 'work pc'.
Edits: 03/21/12
Hey Theob,
Thanks for the answers.
On the mouse thing. Are you using a wired mouse? Or wireless??
Is there an advantage to shutting off all usb???
Afterwards we discovered faith; it's all you need
It is my understanding that this is set up for a wired mouse (ps2 version). Yes there is huge advantage to shutting off usb but that is not peculiar to this setup. It was true of the xp pro set up too.
Hi Jack,
Thank you for all the hardwork and creativity.
Can't wait too try your version.
I send you an email too get the download instructions.
Mark
cMP2 PC > ESI Juli@ > Van den Hul Optocoupler MkII > Lavry Black DA10 > XLR Mogami Gold > Klein & Hummel O300
Jack has pulled all of the major threads of cMP/Windows tweaking with major innovative insights into how WINDOWS works that results in an extraordinary package that with patience one will be able to hear good computer-based audio.
It was easy for me but then I have been playing around with this and had got a pretty good feel for the tasks. It does look like Jack's new instruction sheet does a much better job of guiding the installer than the notes that came with the early release.
This is not for the squeamish. It will take time and you will be making multiple trips back and forth between the cMP machine and your "other" machine. Nonetheless, Jack's disk will reduce the number of these trips significantly. Most importantly, you will not run into as many dead ends as you would if going at this blindly.
This is a reference work that gives us the new baseline to work from.
Time does not stand still and there will be suggestions on how to improve this. But for now this is the benchmark and a fine benchmark it is.
Even from the first listen (in the rawest state) I was amazed at how different the atmospherics were. Before I was listening to a 18mB XP Pro installation that I thought sounded spacious. Even in its raw state it exceeded my old install in this quality.
As one goes through the optimizations for this install the tonality will increase and quickly exceed that quality one could get with the XP Pro install. At least until someone comes up with a better way to use XP Pro.
So this is a gift. In the tradition of cMP/PLAY ...
If you are looking for the silver platter, this is not it since you will have to work to bring this to its highest state. But, as with all things cMP this is a good excuse for learning more about computers and this helps you to understand what to do when something goes wrong. Do not shy away from this. I have found this process addictive. It is actually fun once you get going!
One is left awed by Jack's global knowledge of WINDOWS and the fact that he has very finely tuned ears. Both are in great evidence with this disk.
Hello, everybody! This is a great leap forward for everyone, who has gone this far and is seeking for a perfect way in computer audio! The first major step I am going to take. Jack Wong Has DONE A GREAT JOB!!! I really say, I hope it will help many other inmates to join us, as it GREATLY simplifies things, and we are getting closer and closer to CMP3.
Serge.
This is the one....I know you have all hreard this before. One thing I have learned through help from both Serge's, Rick is that slimming the registry is a key enabler to better sound. When I was using XP PRO I was able to get down to a System hive of about 350kb which sounds great. With Jack's system I (and remember I am not a computer expert) have been able to achieve a 28kb system hive. All other hives are 8kb. This is coupled with files from prior xp levels mostly I believe from XP Home early versions. Remember when we all slimmed shell32 from 9mb down to 2mb how great it sounded. This system has a 45kb shell32!
Enough tech talk. What does it do for the sound. It deepens the soundstage, improves dynamics and imaging. The tone is perhaps the best part. It is warm but detailed, crystal clear in the highs and not bright. When you use the default hive provided by Jack to get a black, black screen when you minimize the screen during cplay playback you simply have the best highs in the business, period!!. It is the best I have heard ims and perhaps all systems. As many of you know I use Martin Logan Monolith 3's with very good (but not distinguished) electronics. Recently I heard vinyl through stellar electronics playing through the latest Summit X's (for anybody local it was @ Alma's in Birmingham. Mi)and I'm sorry this front end (I'll call it the Jack System) smoked it!!! Not even close.
There are a few pioneers in computer audio. We have Cics, Serge (Steppe) and Serge Mihaylov and now we have JackWong96. I worked in the auto industry for many years and we had an expression in powertrain that when something was so designed and implemented well it reached the 'end of the box' of its cycle. WE have the end of the box with respect to cplay/cmp. I am sure of it. Yes I know some of you out there will beat it with other tweaks but in the here and now, this is the state of the art!! Well done Mr. Wong you have have spun a masterpiece!
Hello, everybody. If You have problems inserting watchdog.sys from Jack Wong's collection, You can now easily do it this way: Insert it TOGETHER with his vga.sys AND videoport.sys.
Also, if anyone wants, You can e-mail me to get the smallest and very good sounding winspool.drv and! drmk.sys. PLEASE take attention, these files will be slightly renamed but do NOT need to unpack them after extracting from zip.
To Wlowes: I AM sorry, I was trying three times to send You the registry settings, but I had been inattentive and missed You alternative mail address.
Serge.
Edits: 03/18/12
Serge
winspool.drv and! drmk.sys are fantastic. they remove a lot of clutter and really open up the sound. big improvement thank you. try again to send jack's registry setting. now I am totally intrigued.
many thanks from canada
walter
The google search results for the cics cmp2 web site warn that it is contaminated with malware, and going to the web site yields a 404 error.
This has been reported a number of times. Especially the InternetExplorer is prone to this warning on the cMP site.
cMP2 Computer w/digital crossovers & FIR filters > Lynx Aurora 8 FireWire /192kHz all channels. 2x AcousticReality Ref. 202 & 2x AcousticReality Ref. 601´s ICEpower. Magnepan MG3.3R beechwood frames & custom stands. Miller chokes
.
That's not a Toy... IT'S A TOOL !!
Works Fine For Me
Not for me, last night or this morning. Sure you're not getting a cached hit? I've dropped cics a line though I've no idea if he'll respond.
D
...or as absolutely sure as can be when you:
1) update Firefox to current release
2) clear cache via Options panel
3) shut down machine to stone cold, inanimate lump
4) restart machine, Firefox and navigate to http://www.cicsmemoryplayer.com/
5) upon arrival, display several pages using the index of topics in the left side column
It seems as though it's there. Unless my ISP has it cached on their servers, I can't think of any other explanation than... it's there.
However, I've been wrong before (just happened yesterday, in fact). So that's the report from this corner of the world-wide-wiretap.
Cheers,
Grant
That's not a Toy... IT'S A TOOL !!
Bucking conventional wisdom, I gone with USB rather than the Julie etc sound card route. Recently switched to USB2 Asyncronous with really good results. Asyncronous protocol puts makes the PC a slave to deliver the bits to the converter. For sound, this is the way it always should have been. I chose WaveIO, XMOS based card. Beautiful design and execution. http://www.diyaudio.com/forums/digital-source/188902-xmos-based-asynchronous-usb-i2s-interface.html It also brings up to 24/192 resolution.
To get the drivers loaded, I needed a clean install. This card completely removed the USB conversion as the weak link from the system. It made the recent sound quality improvements outlined by our inmates via XP slimming even more apparent.
Starting with a base XP Pro Nlite install and minlogon + CMP optimisations the improvement from the WaveIO card was apparent.
I did the Steppe 1- 15 + Mihaylov tweeks. At this stage, sound was very clean and crystal clear with great sound stage etc. But it was not something I wanted to listen to. Compared to my well modded Rotel CDP I far preferred the Rotel. Simply more involving and natural.
I thought the WaveIO card might be the issue. However, for completeness I carried on first doing the Resource Hacker slimming of the remaining DLLs in Sytem32. Unbelieveable!! Now this rig exploded with natural texture and toe tapping involvement. Then I added all the registry slimming. Again a huge jump into that 'I am there' and involved with the performers who are real breathing humans. Instruments are made of wood, brass and other natural elements.. not some digital reproduction.
Now this thing is truely high end. Now no comparison with the Rotel CDP. It is useless by comparison.
Summary. USB2 is a viable alternative. If you have not ventured into the ultra level of resource hacking and registry hacking, you have no idea what is possible in PC based sound. I am no where at the level the core team has achieved. I think I am hovering around 23mb in Windows with a 600k registry. Even without the extremes the CMP based system can beat anything out there.
Thanks again to the trail blazers that have made this possible. Now its time for me to get back to building my OTL 6c33c monoblocks and see where CMP2 can take me.
There is a comparison of USB2 boards taking place. WaveIO wins first round,
http://www.diyaudio.com/forums/digital-source/208834-usb-i2s-m2tech-vs-xmos-vs-open-source-qnktc.html
Too bad they don't have an ultra slim CMP2 rig!
Hi wlowes,
Very interested in the XMOS card, Can you give more detail in your implementation? e.g.
-- do you need a power supply for the card? or
-- do you just plug it in the USB port at one end and the dac at the other?
-- does the ASIO driver bundles with the card? or just the demo version (with the beep)?
Thanks
KC
KC
You don't need an external PS. There is a jumper that sets power from USB or from external. On board handling is very good. In my system external was an improvement, and for the ultra in transperancy, you want that supply to be clean. With USB power it was still a big step up in SQ over a very good pcm2706 implementation it replaced.
You plug in USB at one end. I2S at the other. 2 choices for I2S. mBNC cables for non isolated signal, or a multi pin header for an isolated signal. USB also has choices. On board USB socket, or a header to allow you to run USB to a socket on your case away from the bd. Allows you to place the board close to the DAC. Your main challenge is to ensure you can connect I2S on your DAC. I just solder to the pins on the tda1541a.
The builder has customized the Theyscon driver. It is fully licensed without beeps. I paid as much just for the AQVOX asio driver on my last rig.
This thing is a steal. Allow 6 weeks for delivery as this is another DIYer with a day job, and he is a perfectionist.
Recently i stepped over from Juli@ to USB-async using a modded M2tech Hiface. This unit gets 5v from USB-bus and 3,3v from Lifepo4 batteries.
CMP2 setup: nLited small install of XP-pro, minlogon.
Hiface beats Juli@ in this setup, and equals SQ of my Marantz CD63 cdplayer.
Don't have time (yet) to slim xp, but your results make me eager.
I'm very curious about your results if you would try on a Unix-based SO. As i read Voyage-MPD and MPDpup produce great SQ.
http://linux.voyage.hk/voyage-mpd
http://208.109.22.214/puppy/viewtopic.php?t=70052&sid=a211e2cab717aebc13cd5c51416918e9
If SQ on MPDpup is equal/better than CMP2 it could be a great experiment to run it on a very small ARM-based unit like Raspberry Pi (700MHz ARM), or (not so small but very tweakable) Alix-boards (500MHz AMD-Geode). And for me it could be interesting to change to XMOS (Hiface doesn't come with linux-driver)
Please keep us informed! tia!
Douwe
Douwe
It was in fact the Voyage Store that caught my eye. It strikes me that a small ALIX board acting as a headless network appliance serving up the bits is an interesting proposition. Also very simple to build a single 7V linear power supply and it is completely isolated from HDD. Will be a while before I take this step but will post results when/if I get to it.
Walter
There is a new HiFace Two with ASIO and Linux support that will be out soon.
You can find info at their site...
Hi wlowes,Congratulations to your achievement.
Is the WaveI/O card using ASIO ?
- I have the feeling you´re tricking the system to run async via WASAPI...(This will compromize the dedicated ASIO/buffer design of cPlay)
And to an extent, explain that your interface protocol gains quality from the software mods on Windows/cMP side.To run cMP2 with async ASIO protocol, it will always require an ASIO driver software to use the full benefits of bulk mode data transfer in an external DAC.
It is my experience that true external (async or iscochronous)clocking via ASIO does (almost) completly eliminate efforts on the Windows OS.
-so if you experience improvements, my guess is that some sort of jitter is allowed to influence the DAC via the protocol on the WaveI/O.
In other words : the protocol you´re running is the "weak" point, and could possibly gain even more quality if it could run ASIO.kind regards
cMP2 Computer w/digital crossovers & FIR filters > Lynx Aurora 8 FireWire /192kHz all channels. 2x AcousticReality Ref. 202 & 2x AcousticReality Ref. 601´s ICEpower. Magnepan MG3.3R beechwood frames & custom stands. Miller chokes
Edits: 03/12/12
I was kind of expecting async to be immune to our tuning efforts. Not so in my case. WaveIO uses ASIO 2.1 drivers from Theyscon. per the author's site.. "Since this board is entirely based on XMOS / Thesycon work (SW speaking) you can see details directly on XMOS web site: "USB Audio Class 2.0 is natively supported by Apple OS X version 10.6.3 and above. Support for Windows is provided via XMOS partners, Thesycon and Centrance. The drivers are configurable and provide support for WDM/Direct X and ASIO 2.1."
Cplay sees it as an ASIO driver. There is no trace of Direct X left in my slimmed XP.
I am toying with the notion of trying running it from a unix appliance as an experiment, for even slimmer hdw/softw than we achieve with CMP2.
WaveI/O card uses ASIO...
As it output I2S I would be interested to know which DAC you connected it to.
Bibo01,
My Dac is a DIY TDA1541a based DAC with a 6n2p tube output stage. I started with the audio board from an Arcam Alpha CDP. There is not much left from Arcam but the box. Power supply is all BG and Oscons with Burson regs. Tubes have choke supply on B+. DEM reclocking per Grundig design. Separate Plitron transformers for DAC and tubes. I use mBNC connectors for the I2S signal. Not perfect, but seems to pair well with WaveIO.
Dear Bibo01
hmmmm....see, ASIO is a software protocol, and I2S is not a protocol at all, ey ?
One can start wondering why cics implemented tiny, small, medium & large communication buffers, and drew so much attention to buffersizes, as in theory, when it comes to pure playback, long buffers should be as good as small ones.....
I´m not really convinced that just because USB is async and derives its clocking from the DAC, that everything is as straightforward simple as that....
I always come to think of how much difference a new Lynx WDM driver can improve sound quality and how much time & effort cics put into cPlay´s ASIO/DSP optimization.....
I seriously believe that driver software is the most underestimated part of any digital enginering and that it is too easy just claim that async is the blessing of the holy grail.
kind regards
cMP2 Computer w/digital crossovers & FIR filters > Lynx Aurora 8 FireWire /192kHz all channels. 2x AcousticReality Ref. 202 & 2x AcousticReality Ref. 601´s ICEpower. Magnepan MG3.3R beechwood frames & custom stands. Miller chokes
ASIO is a software protocol and I2S is not a protocol at all, ey ?Correct. ASIO is a soundcard driver protocol; I2S is an interface standard, not a protocol.
The Thesycon drivers used by the Wave I/O device and other USB-based ASIO drivers (e.g. the Ploytek/AQVOX driver) bypass the Windows USB Audio drivers to transfer data straight to a device (using the PCM270x, that XMOS something-or-other or the TAS 1020) which in turn pass the data over the I2S bus to the DAC. They do so asynchronously (Thesycon) or (if I have the term right) adaptively (Ploytek).
This ground has been covered before - see link and the replies, esp, as ever, the one from John Swenson.
I seriously believe that driver software is the most underestimated part of any digital enginering
I'm sure you're right but reports are that the Thesycon drivers used for Wave I-O device are particularly robust. IIRC, Thesyon supplies drivers for some of the much-respected Ayre products. Its people do know what they're doing.
it is too easy just claim that async is the blessing of the holy grail.
Also correct only I'm not sure that anyone does claim it's a Holy Grail. Again, reports have it that the protocols work well in many implementations - including cMP^2/cPlay. If you take a few moments to visit the thread wlowes linked to, you'll see that a number of cMP2 users are already successfully using the Wave I/O device.
. . . the protocol you´re running is the "weak" point, and could possibly gain even more quality if it could run ASIO.
It doesn't "run" ASIO, it conforms to the ASIO standard.
Edits: 03/12/12
Hi Dave,
Pardon me, but I did read the two links you´re refering to, but I´m still not convinced that it´s irrelevant how cPlay connects to the DAC.
It´s however not a question of that an async USB DAC can work with cPlay, but rather how.....
Apart that I´m very sceptical to how well all these newish async DACs actually "eliminate" the (jitter-) influence from the computer.
The theoretical thesis is that when cPlay (i.e. the PC) is slaved to the DACs masterclock, jitter becomes irrelevant, right ?
So how can wlowes experience a difference then ?
-is it only ground-noise that makes the difference ?
According to cics (and many others) it is jitter that alters soundquality (bit-perfect transfer granted), so it does not make sense that there should be any perceptible SQ difference.
When I first switched from internal Lynx Two-B (fully optimized cMP2)to external Lynx Aurora via isochronous FireWire, I was somewhat surprised that switching from cMP2 to regular XP, did hardly make any difference at all....I did expect at least a little.
Second, I´ve witnessed several software upgrades on the Lynx drivers and a lot of us have marvelled about cics work on cPlay´s DSP and ASIO software.....
My point is that software is an essential part of a devices performance and most manufacturers have a very "casual" attitude to this.....
Surely, it´s cheaper, easier, faster and less complicated to licence al sorts of "asynchronous USB" circuits and advertize it as "immune to jitter".....
Infact I have the impression that manufacurers just pick´n mix design features from a catalogue. And where this is perfectly acceptable in order to implement well-enginered technology at a price-point, I feel it´s problematic to claim that only because it´s "async" it eliminates all jitter issues.
-it seems that there are ways of "async" that do not capture the benefits entirely....
kind regards
cMP2 Computer w/digital crossovers & FIR filters > Lynx Aurora 8 FireWire /192kHz all channels. 2x AcousticReality Ref. 202 & 2x AcousticReality Ref. 601´s ICEpower. Magnepan MG3.3R beechwood frames & custom stands. Miller chokes
CMP WITH THUNEAU ALLOCATOR INTERACTION
Cics, I am a newbee,have cmp2-cplay from the very start and want to use Thuneau allocator as final touch.
What causes the problem when using CMP with ALLOCATOR (must to restart), is is Cplay or CMP.
I am asking this because if it is Cplay than maybe using FOOBAR as a player
will solve the problem
Thank you in advance
Zoran
Edits: 03/06/12
Hi Zoran,
What is the problem ?
Can you please explain more detailed.
I´ve used Allocator with cMP2 for years and it runs absolutely perfect.
Please describe your hardware too.
kind regards
cMP2 Computer w/digital crossovers & FIR filters > Lynx Aurora 8 FireWire /192kHz all channels. 2x AcousticReality Ref. 202 & 2x AcousticReality Ref. 601´s ICEpower. Magnepan MG3.3R beechwood frames & custom stands. Miller chokes
Hello,
I am new here, not sure I drop the question in the right place.
I have a question about Allocator and Cics player as VST host.
Everything is working fine exept for one thing.
With a restart of Cics player the last frequency settings change as if the frequency gang in Allocator is on.
In the folder Cplay the Allocator file does show the right frequencys and the frequency gang is off.
This occurs in Windows XP Pro SP2 and Windows 7
I don’t know how to solve this, can anyone help?
Thanks in advance.
Jeroen.
I’ll try to upload an image.
Hey Jeroen,
I´ve used Allocator for some years in cMP², but I´m not sure I understand your question ?
What do you mean by "gang" ?
Are you using saved presets ?
Try to explain a bit further.
kind regards
cMP2 Computer w/digital crossovers & FIR filters > Lynx Aurora 8 FireWire /192kHz all channels. 2x AcousticReality Ref. 202 & 2x AcousticReality Ref. 601´s ICEpower. Magnepan MG3.3R beechwood frames & custom stands. Miller chokes
Hello Play-Mate.
Thanks for Your response.
This is how Thuneau describes “gang”.
“Gang Freqs ” This switch when in active “state” gangs together the frequency values of all filters in the section.
By activating this switch I found out that the frequencies in all sections get the same value.
In the folder Cplay there is a (hidden) Allocator file. I found out by using the diagnostics in the CPlay settings. I guess this is where the last settings are stored.
You can open this file in notepad.
In the screenshot You can see the frequencies for High for the high pass filter. Also You can see if a section is active.
The values are from left to right . First section 1493Hz and active. Second section 3706Hz and active. Third section 500Hz not active Fourth section 500Hz not active.
It also shows that the “Gang freqs” switch is off.
For all that, with a restart of Cplay the frequencies of all sections get the same value, 500 Hz. As if the “gang freqs” switch is on. Not good for a tweeter.
This happens as well using saved presets or non preset filters( Dial in a filter, close Cplay and restart).
I can use Cplay and after a restart load a saved preset filter, it will work fine again.
I hope this explanation helps.
Kind regards.
Jeroen.
I din`t yet install Allocator,but on the Thuneau forum there is a guy who installed CMP with Allocator and announce there is a small bug that every time he changes the album he have to reinstall again Allocator.
That`s all.
My hardware is Gigabyte moth. Intel e7500 Kingston 2.5 ssd for software
and 2.5 wd hdd for music storage,linear psu for cpu, and linear over pico
for motherboard,granit digital psu for each ssd and hdd separate.
Monitor is 17 inch LCD classical spdif over Juli@ to upgraded Behringer DCX2496 ( Frank Oettle input,akm4396 jensen jt-11-emcf to Poor Sebian Man Attenuator-http://www.diyaudio.com/forums/diyaudio-com-articles/168796-poor-serbian-mans-optical-attenuator.html)over Neotech silver cables to Korato ONE for high/mids and Korato KTA100 for low.Than over Neotech coax braided speaker cables to DIY made 60 inch wide range ribbon and Diy Woofer in closed box.
If you have Win 7 Starter you can install gpedit.msc from URL below so that AWE can be enabled in CPLAY.
First you need to install & run gpedit.msc the follow CPLAY/GUIDE 'how 2 set lock privaledges' steps 7 - 16.
Then in cplay setting enable AWE
Chevron
With a bit of work it is possible to implement AWE in Win 7 Home Premium.
It's explained here but it needs to be Google translated:
http://www.nexthardware.com/forum/cmp-cmp-cplay/76567-privilegi-di-lock-win-7-home-premium-sp1.html
Thanks for that info, the gpedit.msc URL I posted also claims to work for Win 7 Home editions but I have I tried it with Starter.
The cplay instuctions set Lock Privaledges for Adminstrators so you would need to login to the PC as a user with administrator right for AWE to work in cplay.
Chevron
...and while it was a great time I visited another old friend who is an audiophile of super pedigree. We listened to high end digital, analogue and surround sound on his wonderful home theatre. I had a great time. His 2 channel system is uber high end with no degree of system tweaking left undone. I can post what I am about to say here because he does not read AA. What was suprising was the degree of shortfall his front end exemplified versus a fully maxed out CMP^2 system. We listened to many discs I have and it wasn't even close not even the vinyl (about $10K worth). So for those who have labored through the CMP machinations of the Serge's, Jack et al... you have the best front end I have ever heard. I believe computer audio has now surpassed all other front ends. Its no longer a convenience feature ...its a performance must today.
Cant seem to find any.
Though I think I prefer the small OCz for a little bit more.
The Kingston gets unusually hot in service.
Thanks just ordered one. How's the 47kb shell doin'?
In fact, better than ever thanks to you know who.
8 Gb or smaller ssd
I've just Googled "Kingston 8GB S100 SSDNow" - it seems they're to be had over here.
HTH
D
NEW VERSION: Now plays all four formats supported by Recursive Cue Creator, using Cplay, by automatically converting APE and MP3 files to WAV in the Windows Temporary Folder and automatically editing the CUE file to match, and then starting Cplay with that cue file. (This feature requires a .vbs file, which is included in the download link. It also requires the free Monkey's Audio program to convert APE files, and/or the free Lame program to convert MP3 files, and links for those are at the bottom of this same post.)This will work without cue files, but then only with one file at a time, as cue files continue to be the "playlist" of cplay. (It also supports Shorten SHN files, but those are not currently supported by Recursive Cue Creator, so any cue file would be supplied on your own.)
On my PC, conversion to WAV takes about one second per minute of audio files, YMMV.
For those who are unfamiliar with Cplayrun batch file, the following is the entirety of the earlier post explaining it:WHAT: A Windows Batch File that integrates cPlay for flac and wave files, with another player (foobar2000 v0.8.3) for all other music files, and provides automated rate switching for cPlay:
MOTIVATION: Recently I discovered that a DAC that 1) generates its own timing, 2) generates its own regulated power, and 3) isolates the digital audio chain from the analog chain, removes most of the defects of digital audio that makes it sound worse than analog audio. (The HRT Music Streamer II at US$150 does all three things, but there are others as well, but perhaps not as inexpensive.)
I then became interested in comparing the sound of music playing software, using this DAC, and found that the smaller and lighter the program, the better it sounded (in other words, the least distortion it applied to the original audio). The free program "cPlay" - developed in this very Forum - continues to produce the best sound quality of any player I have tried. (Amongst the others, currently MusicBee is notable for combining stability with a lot of fun features, such as lyrics scrolling in sync, album art, display of web information on the artist, etc - but it seems those features come at a sound quality cost.)
My music playback PC is also my living room video PC, so it runs Vista, and so I have been unable to run the "cmp" system that replaces the Window shell, since it does not work with Vista. However, I did build that PC from scratch using "quiet home theater PC" principles that mirror the ones in the "cmp2" hardware system. And I had already disabled all unnecessary services and software for better video playback, so that also mirrors some of the cmp principles as well.
DETAILS: There are two difficulties with using cPlay that I have sought to overcome:
1) It has a fixed output sample rate. cPlay was written when virtually all music was 44.1 khz and so upsampling to a much higher rate had some benefits. However, now we have hdtracks providing 88.2, 96 and 176.4 files. My DAC is limited to 96khz, and upsampling from 88.2 to 96 definitely sounds worse than just leaving it as 88.2 khz. So, I need a system to set the sample rate based on the file's rate.
Fortunately, Bibo01 posted a batch file to this Forum a few weeks ago that changes the cPlay .ini file's RATE setting based on the user's keyboard input to a prompt. I modified that batch file to instead look for a text file in the same folder - "44.txt" to use 44.1 khz, "48.txt" to use 48 khz, "88.txt" to use 88.2 khz, "96.txt" to use 96 khz, "176.txt" to use 176.4 khz, and "192.txt" to use 192 khz. (The files can be empty, zero length, text files.) Since you can right-click in a folder (using Windows Explorer or a replacement) and select New File or New Text Document and name it, then you specify the output sample rate very quickly and easily. And this can be more sophisticated than "use the file's sample rate" if, like me, you have files with rates greater than the DAC's limit. So "88.txt" in a folder with 176.4 khz files will downsample in a clean way by tossing out every other sample, rather than intensive computation to go from 176.4 to 96.
2) cPlay only plays WAV and FLAC files. I have many lossless audio files in other formats, including APE, SHN, ALAC and WV, and also some lossey files in MP3 and WMA format. If you start cPlay and try to use it to play music, you find that sometimes you want to hear music and it is in APE or some other format unsupported by cPlay and you have to close cPlay and then find the music again to play it with another player.
So, I wanted a system that will play all my music files, and use cPlay for all WAV and FLAC files, and another player for all the other files. In my personal testing, I found foobar2000 0.8.3 to be the best sounding player outside of cPlay - better sounding than the current foobar2000. It starts up instantly, showing that it is very small and light (smaller than current foobar which has added a lot of code over the six years).
HOW TO USE: The simplest way to accomplish my goal, was to use a file explorer, and associate all audio files with foobar2000 0.8.3 and then associate WAV and FLAC files with the batch file mentioned above that sets the sample rate. In order to have the most minimal setup for best sound quality, I chose a popular small, light, free file explorer and included a line in the batch file to kill its process before playback. The batch file also runs cPlay in realtime priority using the same lines from cPlay's included batch file.
However, that only allows you to play one FLAC or WAV file at a time. In order to play a whole album, you also need to associate CUE files with the batch file. You can then use Alan Jordan's Recursive Cue Creator to create CUE files for every folder in your music library (available from alanjordan.org). (By the way, you don't need CUE files to be associated with foobar, because if you click on a full-album APE file, it will look for a CUE file in the folder for the track information.)
A link for the cplayrun.bat batch file is at the bottom of this post. The provided version includes code that supports rates up to 192, but is intentionally limited to 96 - since that is my DAC's limit, and may be yours. If you have a 192 DAC, open the batch file in a text editor, and there are comments explaining the lines to remove to allow it go all the way up to 192.
The current version of the batch file in this cPlay thread is the second version 0.2.0, which includes two new additions: (1) added support to automatically use cPlay for cue files that it supports, and foobar2000 for all other cue files, and (2) support to add a folder context menu entry that will choose cPlay for folders that can be played by cPlay, and foobar for all other folders. (For those who cannot figure out to add that support to the registry, let me know and I'll provide a .reg file to do it.)
Also, the batch file should be put in cplay's folder, and it is hard coded in the batch file to be:
C:\Program Files\cics Play
If you have cplay in a different folder, you will need to change the location inside the batch, in several places, using a text editor.
You can also edit the batch file if you want to use a different file explorer than xplorer2_lite. Note that if you try to kill Windows Explorer, in most versions of Windows, it will kill the whole shell, so it is cleaner and simpler to use another file explorer for selecting your music.
The free file explorer xplorer2_lite can be downloaded from dev's site:
http://zabkat.com/xplorer2_liteTB_setup.exe
NOTE: you'll want to unselect the toolbar installation to keep a minimal setup.
foobar2000 0.8.3 can be downloaded from File Hippo at:
http://filehippo.com/download_foobar2000/55/
otachan's audiophile ASIO plugin for foobar2000 0.8.3 can downloaded in a dizzying array of variations from:
http://personales.ya.com/angel49/foobar2000_otachan/intro.htm
the fourth choice of ver 0.51 seems to be the most generic and works fine.
Monkey's Audio can be downloaded free from the creator's site at:
http://www.monkeysaudio.com/download.html
Lame MP3 encoder and decoder can be downloaded free from:
http://www.rarewares.org/mp3-lame-bundle.php
Shorten 3.6.1 encoder and decoder can be downloaded free from:
http://etree.org/shncom.html
NOTE: Use of any software provided in this message, is done at your own risk and no warranty of any sort is implied. It is proper practice to backup any important data from your computer before installing anything new.
Edits: 02/23/12 03/02/12 03/03/12
Original post's download link was just updated with bug fix version 0.3.2
(The previous bugfix exposed some syntax errors that can cause problems for certain combinations, for example if parentheses are in a folder name. I also corrected the syntax of the foobar2000 start line. )
Original post's download link was just updated with bug fix version 0.3.1
(The bug is that the folder was not correctly set for an individual file that was chosen to be played with cplayrun.bat )
Hi Ken,
thanks for this new effort.
In the same fashion can this be extended to other formats as well, like AIFF and ALAC?
I would like to point out that under Seven some problems with your batch. Seven has VirtualStore so it saves a copy of cPlay files in another directory which get involved in the .tmp creation. The answer is to change Ownership of cPlay folder by going to C:> Program Files> cics Play, then Property> Security and change user's account authorizations.
Yes, it could be extended to any other format, just two things are needed, which is 1) installing a program that will decode the files into WAV files using a command line, and 2) having a cue file (unless the there is only one file). The latter is the major problem.
In fact, I am thinking that the one biggest problem with cplay was the decision to use cue files as its playlist file format. There is virtually no software to deal with cue files, since they are only designed to burn CD-Rs.
If cplay had been designed to use .m3u or .pls playlist files instead of cue files, the program might have several times the user base by now.
PS Concerning the ownership of the cplay folder under Win7 - is that merely a suggestion to other users, or is it something that ought to be part of the batch file itself ?
Re. the ownership of the cplay folder under Win7 -
It is a suggestin to other users otherwise the batch file simply does not execute properly.
Don't know if this will work for anyone else but maybe there is something that will help.
I decided to follow Mr. Wong's use of FAT instead of NTFS so I began with a fresh minimal nLite install.
I should say that I intially (just for fun) tried to insert the micor-shell into my exisitng install with no luck. This did give me a hint as to what to do next. I was told that WINDOWS would not load because c_950.nls was missing. I copied this from my main machine and made sure it was in the registry (where I had previously deleted it). This was not successful.
So I began the new install.
It is amazing as many of these as I have done that I still mess them up. This is due to my incessant obsession with looking for short-cuts. After a few of these I decided to take my time and be methodical.
Shortly after getting minlogon, cMP/PLAY and JULI@ installed I did Mr. Wong's SECURITY and SAM deletions. I also did what remains to be done of the the original cMP optimizations. Most of this stuff is taken care of by file and registry deletions so much of it is now a waste of time.
I had done this a couple of times and then tried to insert the micro-shell and it would not work. The last time (the one that worked!) I remembered Serge's protocol for inserting files from his BIG FILE SUBSTITUTION tweak and thought I should try that.
I did and it worked. I substituted Mr. Wong's files in the first group specified by Serge which happened to include shell32.
It works well. The only slightly strange thing is when you minimize and mazimize the cPLAY screen there is a little stuttering of the playback. Big deal.
The strange thing is how much trouble I seem to have going further. All I have done to the system registry so far is to remove controlset2, root out of ENUM and most of the media/audio/video entries. I have done very little file deletion. When I try to go further (which is still taking away big chunks) I cannot boot. Back to SNAPSHOT ...
I know this will be overcome but it is frustrating having this huge WINDOWS isntallation at the moment. (no jokes riboge)
Tonight I will do further file substitutions per Serge using Mr. Wong's files and see if I can use the minimized SOFTWARE and DEFAULT hives.
If you can't get it smaller in over 4 hours...
I have tried again and again and can't get these to work even with the other substitutions. You are amazing being able to keep this up (here I go again) through such labors as starting over FAT and then the frustrations that are following.
How can the file/disk format make this difference? And if it does, shouldn't there be some "translation" application out there that converts between NTFS and FAT in a way that maintains all functions?
This is supposed to do it without data loss--we'll see about function loss. It's demo/free version works on converting partitions up to 8GB.
http://www.aomeitech.com/n2f/download.html
I will try it and hope a couple of you will, too. Who knows if this will bridge the gap between nLite installs in ntfs and Jack's shell32 et al?
The converter's site says "Aomei Ntfs2Fat32 Free Edition can work on Windows 7/Vista/XP/2000/2003/2008, but it only supports converting the NTFS partition maximum 8GB to FAT32 for testing"
After conversion, which was very fast, all that was required was to repair boot files with EasyBCD. It ran fine as it was but with Jack's shell32 and shlwapi substituted I got the same error message as I had gotten when I was NTFS. My nLite installation has all the mods and tweaks but only limited slimming, especially to the registry. So that may make a difference but it is hard to see why it should. I have substituted most of the other files with no problem.
You made sure you included the c_950.nls file and made sure this shows up in SYSTEM/Nls - can't remember what comes next but it is easy to figure out once you are there.
Did you include, at the time of substitution ALL of the files Serge recommended changing at the same time as shell32?
I think that was key.
Wish I had known about this YESTERDAY!!! The converter that is. It never occurred to me to LOOK for such a thing! Good to know ...
sounds to me it won't do this neat trick unless you buy a copy.
Could be worth the money!!!!
Dear community,
A friend of mine has problems with his cMP2 setup after he tried to backup his music library via an external USB 3.0 disk.
This back-up took extremely long time.
He reports that cPlay now makes digital noise for about 20-25 seconds at the beginning of playback of each track.
I have told him it sounds like a memory error, and I´m awaiting the diagnostics file soon, but maybe someone can identfy the problem....
kind regards
cMP2 Computer w/digital crossovers & FIR filters > Lynx Aurora 8 FireWire /192kHz all channels. 2x AcousticReality Ref. 202 & 2x AcousticReality Ref. 601´s ICEpower. Magnepan MG3.3R beechwood frames & custom stands. Miller chokes
I was doing Jack's slimming using Psexec. Can someone tell me what should be inside this key? Is it standard or machine specific? Can I recreate this key or do I need to substitute new copy of security hive and start this tweak all over again?
Also, using pskill as described disconnected network. Which of smss, lsass or svchost should not be killed to preserve network connection?
Well, if you can still boot without it that would be great! If not, please load your last security hive, export the SecDesc key and import to your current hive.I forgot to mention that the "Services Killing" thing was done in a non-network environment, it hasn't been test in a network CMP.
Edits: 02/19/12
This suits normal CMP users with a regular Windows more than those using a slimmed Windows.Download PSTOOLS, extract and copy Pskill.exe to system32.
http://technet.microsoft.com/en-us/sysinternals/bb896683Create a Pskill.bat in C:\, with the following entries:
@echo off
pskill -t \\cmp smss.exe
pskill -t \\cmp winlogon.exe
pskill -t \\cmp lsass.exe
pskill -t \\cmp svchost.exe
Assuming the domain or computer name is "cmp"
In cicsMemoryPlayer.pth,Amend:
RIPPER #M "c:\program files\exact audio copy\EAC.exe"to read:
RIPPER #N "c:\Pskill.bat"
Boot into CMP and RIP to kill services.Csrss.exe and Service.exe are the only processes left.
Please note that Service.exe can't be killed as it's needed by cicsplay.exeNote:
(1) You will lost the ability to shutdown or reboot, after killing the services!
(2) It hasn't been tested in a network environment.
In theory, users of slimmed Windows can put in cmd.exe and usp10.dll, and with the necessary registry entries, services can also be killed. Please don't ask me how, as I haven't got time to try it yet.
Edits: 02/19/12 02/19/12 02/19/12
Without the latter you cannot connect and without the former you cannot get cMP to call cPlay even with the former running.Edit: Hours later I tried returning the killing of lsass.exe to the pskill.bat and now network play does work. Puzzling--especially since either way lsass.exe is shown running in task manager. Doing this, however, eliminates normal shutdown and requires hardware shutdown. Makes you wonder about cMP's suspend of svchost and lsass.
Edits: 02/20/12
Ok expert registry guys. I have an issue wrt to keeping revised hives. I edit them load, boot, play music. Then I come back for more reg edits and the hive seems to have grown again. For example I editted software from 28 kb to 20 kb (deleting persistent handlers etc etc per steppe)and when I come back for additional edits its back up to 28 kb. Yes I ntregopt it always. In my system hive I get it down from 512 kb to 384kb load it play music come back later for additional edits its back up to 512 kb. Anybody else have this issue?
Edit the file boot.ini, adding the key /MININT into the command line (last line) to prohibit any registry changes.
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2 Ru, GA-H55M-UD2H, i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S FAT32, ESI Juli@, full linear PSU, NAS - WD My Book Live
I have it in boot.ini but not the last entry. I'll move to the last item. Thanks Serge.
Moved the MININT switch to last item in boot file and it does not seem to work for my enum root key ... it keeps coming back. While doing key deletion I asked for permision (even tried advanced permission once) and my editted system file goes from 384 kb back to 512 kb over and over again.
I'm giving up for now.
Hello, everybody! Here is a nice and easy thing to do, while we are looking for the way to adapt new shell:
WARNING: THIS TWEAK NEEDS ABOUT 24 HOURS and several REBOOTS to shine.
Go here:
HKEY_LOCAL_MACHINE\000\ControlSet001\Control\Windows
change the marked settings as You see it on the pic.
set ErrorMode to 2
set NoInteractiveServices to 1
set ShellErrorMode to 0
Yes, and DON'T HESITATE to remove Watchdog from registry, It's a great tweak.
Also, go here
HKEY_LOCAL_MACHINE\000\ControlSet001\Control\Video\{23A77BF7-ED96-40EC-AF06-9B1F4867732A}\0000 and set InstalledDisplayDrivers to only
vga framebuf.
Serge.
Edits: 02/19/12 02/19/12
I initially did this but somehow managed to miss the last part on the '...set InstalledDisplayDrivers to only vga framebuf...'. When I did the last part it resulted in a very very strong positive sq pop. The way I would describe it is strong strong dynamics, bass, better extension top and bottom. More coherence.
Thanks Serge.
I don't know whether inmates can see this when you load the Security hive. There is a copy of Sam inside the Security hive. The 2 Sams make changes simultaneously, if data in Sam hive is changed.
I can't see the Sam inside Security either, when using load hive and log on to system account. In order to see all data and make changes simultaneously, one need to log on to the system account in the CMP Windows.
Delete the data inside Sam hive and the deletion will be reflected in the Sam inside Security hive. You'll hopefully get a 8kb Security hive.
The Sam hive remains the same as per my last report - "Please see Screen Shot". Please don't touch anything inside Account and Builtin. There is nothing inside RXACT but the folder is needed.
I've deleted more from the Security hive, as follow:
1. The remains of Security hive as per the screenshot. The presence of Account\SID\Priviledges does not improve sound;
2. Delete the Key/Dword inside PolAdtEv, PolAdtLg and PolMod respecticvely. The folders are needed; and
3. In RXACT, delete content of the log only.
Please don't touch PolAcDmS, where the machine SID and backup of Sam reside, and SecDesc.
There are some good improvements after the registry deleletion. Music is definitely more 3-Dimensional; there are slight but noticeable improvement in minor details. Less policies in the hive perhaps reduces interactions between the Security hive and Isass.exe, hence, less noise and overhead.
The deletion is done on XP Home SP1.
The secret was in leaving all the other policies in (as per steppe) then I was ok.
Hello! Ted, I have just restored my cmp. I have tried new shell + shlwapi and even + Jack's gdi32.dll, still no sport. Jack! If You read this, where is this shell from, really, just for interest?
Serge.
I believe they(shell32 and shlwapi) are from xp sp1 home and they did not work in my system either. My thought is that, like your set of files, they probably will all have to be substituted for it to work. Jack indicated that his file set is from a whole bunch of different sources.
Mr. Wong is using this instead of NTFS.
Since I assume we have deleted the FAT files and entries in the REGISTRY this can cause a problem.
I am making a new install with FAT on all drives and see if I have success.
You are spot on check your email.
I replicated your security hive screen above and got a bsod. Upon looking at my hive before I did your edits I noticed that I had a few more entries under Policy in parallel with PolAdtEv, PolAdtLg and PolMod that I deleted. So I am going to try again leaving in those other entries. I am on another computer away from my cmp box so I cant copy them here but there are 3 or 4 of these.
Edits: 02/17/12
from one direction, bottom up or reversed. Jump to next if unsuccessful. Keep the Account\SID for the time being, that's the last one to be deleted.
After emptying some of the key/string, try deleting the folders.
If you have no success deleting a string/dword, try emptying it's content.
You said: '...The presence of Account\SID\Priviledges does not improve sound...' I just listened and you are right no difference maybe even a slight degredation.
Its interesting that I have no sam within security. I do have a sam within sam but that is all, There are differences as well.
Going to try your new registry edits now.
As always thanks Jack.
The Sam in the Security hive(not in Sam) can only be seen when one boot into its own Windows and view directly in Regedit. It's not about loading the Security hive using another computer, the "Sam in the Security hive" can't be seen this way.
Here is how I did it using Snapshot.
I opened an earlier image of my os.
I pulled my Security file out of it and placed on my desktop.
I used the new program Jack mentioned to open and hack it.
I used Ntregopt to compact it (while still in regedit mode)
I rebooted.
Loaded my ssd into my docking station on my desktop.
Replaced the Security file with the newly hacked/compacted Security file.
Rebooted and now I am listening to see (hear) if I can notice a difference.
For those of us who can no longer do ANYTHING on our cMP installs?
Do we load an "older" version and work on it there, save it, and place it in out minimized installs?
For me, I would use the first ghost image or without it install a new Windows using the same Nlite disc. On 1st boot, I'll log on to system account and do Sam & Security slimming simultaneously. Thereafter, copy the Sam and Security to the config folder of the slimmed CMP Windows.
I am lost with all the changes to Sam and Security changes (which ones to do).
@jackwong96: can you tell me (maybe via a link to the changes) exactly which changes you make in Sam and Security after a fresh install so that I can try them?
This is how it looks, so far, after booting into Windows.
Actually,I found that load hive & delete data in SAM only, will result in corresponding changes mirrored in the SECURITY when boot into Windows.
For SAM, it is 8kb.
The Security hive is also 8kb. I have not yet decide which Policy\Account to keep. S-1-1-0 is "Everyone" and S-1-5-32-544 is "Administrators". The Privilege etc are still inside. I have copied the Privileges granted to "Administrators" to "Everyone" in case they are needed.
The PolRevision & PolSecret... were deleted and no longer present when I load hive. Nevertheless, they kept coming back in Windows(only) despite further deletions.
One thing, add read and write rights for Everyone and delete System and anyone else in all keys and sub-keys after load hive. Right click "Permission" to do that.
Edits: 02/17/12 02/17/12 02/17/12
I installed Jolida's nLite xp and did mods, tweaks and some slimming all on a hdd. I made an image which worked to restore to the hdd when I goofed with further slimming attempt. But when I copy this image to either of two ssds I tried it will not run. Boot begins and gets just past the XP with progress bar then stops on a blank screen. Seemingly it is cMP that will not start. I have tried this on two computers several times with two different applications for copying the partition, one by image and the other direct disk to disk. No go. I am baffled. Any ideas or suggestions?
This aging thing sucks memory-wise. I later recalled I had a year ago used this software to set up my first ssd(no longer used). It worked just now to migrate from the hdd nLite cMP drive to another better ssd drive.
Paragon Migrate OS to SSD was offered originally free from Paragon. That is how I got it a year ago. It's second version is now licensed by them for $19.95. Other software I suppose offers this migrate function as well. I don't know whether I can legally offer the first, free version to others. If anyone can advise about this and it is okay, I can make it available.
I had taken Mr. Wong's advice to install on a HDD and then transfer to my SSD and it would not work.
Anyway, I think the SSD fragility stuff is more than slightly overstated.
Hello! You have installed on hdd1, and installation process has created and modified the registry according to your THEN current hardware set, namely in software and especially SYSTEM hive. You cannot just copy and attempt to run the same windows, hacked or not, on another hdd, unless it's just the same hdd model and capacity, and, I suspect, even batch.
Serge.
Actually, I have done so quite freely up until now. I have copied from one drive to another many times both between the two drives in my laptop (one is removable and goes in cd slot adapter)and to external drives which then were configured to boot. These are various brands--Hitachi, WD, Kingston et al--no two the same. Also there has been discussion here previously about working on the system on hdd until satisfied and then transferring to ssd at the end to avoid too many rewrites on the ssd. So clearly others have done this.
I can't imagine how it can be done. There's device enumeration in the system hive that simply has to cause bsod as the drive type for the system position gets changed, though, I may be wrong, especially with laptops. Also, there is a thing in system hive that marks which control set is default and which current and which, if any, is considered failed, look there, It may be that the failed one is the one You need, then just change the failed value.
Serge.
Had played with the files long ago with reshack and when I read riboge's report I started playing with RT. But I was modifying the already reshacked files.
Last night I went back to my original files and RT'ed them.
Riboge is right - reshack screws up cPLAY and it is not just cosmetically.
There is quality to be gained with starting over and using RESOURCE TUNER.
I have made two versions of each. One minimized to only one dialog screen and the other with the other dialogs for making adjustments. I think there is something to be gained with this.
I do retain (in the minimized version) phase/volume/balance and X in cPLAY. Nothing is retained in cMP. I went back and forth and settings are retained since they are kept in those .ini files.
Out of curiosity I thought I would see if one could see any change in the MEMORY MANAGEMENT registry when AWE is implemented.
On my host computer I looked at memory management and took a picture of it then implemented AWE (with the necessary reboot) and looked at it again.
There is no difference in the values shown. So I have to figure there is nothing we can do THERE to adjust AWE.
I did not look at the whole registry since I had no idea where else to look. I would be glad to do this again if someone knows other places that might be affected by this setting.
Now to find a way to view the sub-registry since there has to be something like that.
It's stored in binary inside HKEY_LOCAL_MACHINE\SECURITY\Policy\Accounts\S-x-x-x-xxx\Privilgs]
Where the SID:"S-1-5-32-544" is the BUILT-IN\ADMINISTRATORS.
The below is captured from my working computer where multiple privileges, including AWE, were assigned to the Administrators:
[HKEY_LOCAL_MACHINE\SECURITY\Policy\Accounts\S-1-5-32-544\Privilgs]
@=hex(0):15,00,00,00,00,00,00,00,08,00,00,00,00,00,00,00,00,00,00,00,11,00,00,\00,00,00,00,00,03,00,00,00,12,00,00,00,00,00,00,00,00,00,00,00,0c,00,00,00,\ 00,00,00,00,00,00,00,00,13,00,00,00,00,00,00,00,00,00,00,00,18,00,00,00,00,\ 00,00,00,00,00,00,00,09,00,00,00,00,00,00,00,00,00,00,00,14,00,00,00,00,00,\ 00,00,00,00,00,00,16,00,00,00,00,00,00,00,00,00,00,00,0b,00,00,00,00,00,00,\ 00,00,00,00,00,0d,00,00,00,00,00,00,00,00,00,00,00,0e,00,00,00,00,00,00,00,\ 00,00,00,00,0a,00,00,00,00,00,00,00,00,00,00,00,0f,00,00,00,00,00,00,00,00,\ 00,00,00,05,00,00,00,00,00,00,00,00,00,00,00,17,00,00,00,00,00,00,00,00,00,\ 00,00,19,00,00,00,00,00,00,00,00,00,00,00,1c,00,00,00,00,00,00,00,00,00,00,\ 00,1d,00,00,00,00,00,00,00,00,00,00,00,1e,00,00,00,00,00,00,00,00,00,00,00,\04,00,00,00,00,00,00,00,00,00,00,00
When AWE is removed, the first value changed to @=hex(0):15,00,00.... and the values at the end "04,00,00,00,00,00,00,00,00,00,00,00" were removed.
I guess you've already removed it from your CMP computer registry.
I probably have. Not at home to check.
Easy enough to restore. I have plenty of old config files laying about.
We'll see.
The question does remain: just how important is this? If I have already disabled it (which is likely) and am enjoying what I perceive to be improved quality I wonder if this is still important in our minimized machines.
The more I read about AWE, and considering I am using a 512mB memory I really wonder just what it could do. Seems intended for large value memory installations. Seems like it could be confusing to the machine when the value of the memory (as in my case) is so small. Any opinion?
I tried the 1gB Kingston Value Ram without success. Could not get anywhere close to Serge's settings, would not run on low voltage and did not sound very good. Maybe the fact it was new and I was not patient enough to let it break-in? Maybe the version you are using is different than the one I got? I bought it from NEWEGG so I feel sure it was a real KINGSTON product.
And while I am rambling on - I have yet to attempt running the memory stick from a battery. What battery voltage are you using? I assumed a D cell. Does the MB still regulate that voltage down to the 1.2V?
Thanks!!!
So Gentlemen, riddle me this:
If I have deleted the entire BUILT-IN\ADMINISTRATORS account (ie: "S-1-5-32-544" no longer exists in the SECURITY hive), why does cPlay's Diagnostics screen still report "AWE successful" ?
Surely it knows what it's doing. On my main desktop (where cPlay runs solo - without cMP) I've not implemented the AWE function in XP Pro (though the box is ticked in cPlay's settings window), and in this case cPlay faithfully reports "AWE failed, reverting to standard allocation".
Could there be other places in the registry directing the kernel to use AWE memory management ? It would seem so on my cMP^2 rig.
To Rick's question (and I'm hypothesizing here with little deep research to back it up...), I think one of the real benefits of AWE is it reserves a chunk of RAM for exclusive use by the program - in this case cPlay - and is therefore spared the chore of requesting/obtaining memory space "as needed" during operation. One of cics's guiding principles is to reduce any OS activity that can be avoided, and AWE helps go in this direction.
I don't think the size of the RAM plays as big a role as the activity overhead of managing it. Does anybody know better - or differently ?
In any event, I'm happy I still (seem to) have AWE working for me in the slimmed-down, Franken-OS world of cutting-edge cMP^2. It sure continues to sound better and better as we discover new tortures to put our systems through.
Cheers,
Grant
That's not a Toy... IT'S A TOOL !!
it does not say that is what it is doing.
Their literature makes it sound like it has more to do with utilizing large amounts of memory.
Of course, the 3gB switch in boot.ini mystifies me the same way with a 512mB memory ...
Hello! Grant, Hello, everybody. It is also an assumption, certainly, but I think that we have long overlooked this file - ntldr, that resides in the root. It is very important to the whole system, and, I believe, it's getting modified as time goes by. At least it surely is modified by cmp-cplay, - there are strings about cmp-cplay among the usual program gibberish inside the file (opened in the notepad).
I think that the behavior of all: system, default administrator or guest, is described in one and only account, or rather pattern - the god's mode, with our accounts representing only differences (limitations), assigned to the god mode. In my ramblings across the net I have come a couple of times across refferences to developer's and god's mode in windows, but I don't remember where or when...
Serge.
Edits: 02/16/12
other Privilege Rights in one of the Security Configuration Templates, named DEFLTP.INF, in the XP PRO SP2 installation disc. It reads "SeLockMemoryPrivilege = "
Adding *S-1-5-32-544, or *SID, to it will result in AWE being enabled for the named SID, after installation.
All other privileges written in the same Template for different SIDs will appear in Local Computer Policy\....\User Rights Assignment in the exact manner as it is written in the Template. And they will be stored in the Security hive. One can see the changes in value each and every time the user rights assignment is changed. To view one's own Sam and Security hives in real time, one needs to logon as the System Account. In this connection, Psexec.exe from the System Internal can be put to good use.
However, the situation is different in XP Home SP0, SP1 and some earlier Beta versions, where there is no mention of LockMemoryPrivilege amongst a few Privilege Rights present in the Template. Nonetheless, Administrators will be assigned AWE privilege automatically. This is not the same case for XP Home SP2 though. The relevant entries for Privilege Rights can be found in dpup.inf in SP0 and SP1 etc.
I know too little about AWE and privilege assignment, and have no answer to all the questions. But what's presented, is solid.
You may aware that I've only 4 entries inside Winlogon and the UserNames/DomainNames were gone. I did so because the Administrator account was removed and there was no point logging on as such SID. I successfully logged as no one! And still got AWE. Perhaps, I've logged on using the Everyone account, i.e. S-1-1-0, which is the only account left behind after the deletion.
Back to the question of AWE in cMP. Although it says AWE successful, it also Ram Load each and every time, even when told not to do so.
And to echo Rick's comments, the effects of AWE is diminishing as we move on. I guess our Windows have become so efficient and quiet that it now closely resembles loading music files in Ram or alike. Just a guess though.
By the way, Rick, if you put NiMh batteries in, there will be no voltage regulation. The voltage will slowly drop from 1.35V to the lowest working voltage of your memory.
Using minlogon is one not always logged on as the system account?
.
.
.
It should be used in conjunction with shlwapi.dll.
Shlwapi.dll also caused a no boot. I was able to use this substitution with mini XP previously but not here. What I haven't tried yet is putting in both your shell32 and shlwapi at the same time. I will report back if that happens to work.
Quoted from MS "How the System account is used in Windows."
"The system account and the administrator account (Administrators group) have the same file privileges, but they have different functions. The system account is used by the operating system and by services that run under Windows. There are many services and processes within Windows that need the capability to log on internally (for example during a Windows installation). The system account was designed for that purpose; it is an internal account, does not show up in User Manager, cannot be added to any groups, and cannot have user rights assigned to it. On the other hand, the system account does show up on an NTFS volume in File Manager in the Permissions portion of the Security menu. By default, the system account is granted full control to all files on an NTFS volume. Here the system account has the same functional privileges as the administrator account."
Some up until now have said minlogon puts one as the default user, others more lately have said it is the system account, now you--who seems to know what you are talking about--say it is "an" administrator account. Is it "the" (real, fundamental) administrator account? Is there such a thing?
More importantly, how DOES one logon to the system account?
download a set of very useful Pstools from Windows Sysinternals.http://download.sysinternals.com/Files/PsTools.zip
Unzip and copy Psexec.exe to your system32.
In CMD, type "psexec -i -d c:\windows\regedit.exe" and enter.
A powerful Regedit will then pop up under your service.
You can also load and view other Sam and Security hives without using any 3rd party software.
How the System account is used in Windows:http://support.microsoft.com/kb/120929
"Some up until now have said minlogon puts one as the default user"It depends on what account is logged on. E.g. in SP1, after minlogon, one logon as Owner, this one is logged on as Default user, and the Default User folder will be created in C:\Documents and Settings. If logged on as Administrator, then the Administrator account.
Edit: after minlogon, one would use the Default folder in C:\Documents and Settings, as a computer users, perhaps a super one.
Edits: 02/16/12 02/16/12 02/17/12
WOW, you are speeding ahead and flying high - I cannot wait to see/hear what you come up with.
Thanks for your efforts and your generosity.
if excluding cPlay, cMP files, 1 Chinese font and C_950.nls, The config folder is now 164kb.
I want to do more file substitution before letting you know. I now use a micro-shell32. It sound so much better!
I might have found a way, as my registry slimming went exceptionally easy and smooth.
.
with .dll substitutions and registry mods you still end up in the same place (if you want to).
config at 164? Incredible!
I have not been trying much lately because I thought when I got below 300 I was just about done AND NOW you post this.
I get the feeling there must a combination aspect to further deletions since I have tried numerous "single" items with no success.
Will TRY to wait patiently for the micro-shell32 but it is NOT going to be easy.
Congratulations and thanks again.
.
There is something else. No question.
But whether we can access it and more importantly have any idea how to manipulate it - that is the challenge.
unattainable spheres of security and default hive or probably also SAM, that are dealing with policies.
I asked this question to myself some time ago while experimenting with nLite, remember, I couldn't make awe with those ini-es?
Then I became more attentive and retained in nLite iso Group Policy Object Editor and some other group policy smth, and when trying to move forward, I was prompted, that some more items were needed for it to work, so I retained them also.
Serge.
if more of DEFAULT is shown?
I was doing some searching and came across something to the effect that winver is involved in this.
Can a .dll be modified within a WINDOWS installation or do they always stay the same?
Oh, well.
The Drive Snapshot, a little simple proggy from Germany? With S letter and white gear and a drive on the desktop Icon?
Serge.
Don't want to start over again without a backup.
It's this one
Anyone tuning computers on the cMP level, should be familiar with backing up a disk image. In this regard SNAPSHOT is brilliant. Easy and straight forward.
Get it !
cMP2 Computer w/digital crossovers & FIR filters > Lynx Aurora 8 FireWire /192kHz all channels. 2x AcousticReality Ref. 202 & 2x AcousticReality Ref. 601´s ICEpower. Magnepan MG3.3R beechwood frames & custom stands. Miller chokes
As there has been a lot of talk on RAM management in recent months, I have been considering the implementation of Page Size Extension (PSE), that is using 2-4MB large pages to transfer between RAM and CPU instead of the usual 4KB chunks.This is a method used in servers and it increases RAM speed. In our case it would be a further extension to our lock pages and AWE.
What do you think about it?
Is there a way to implement it in kernel?"Enabling PSE (by setting bit 4, PSE, of the system register CR4) changes this scheme. The entries in the page directory have an additional flag, in bit 7, named PS (for Page Size). This flag was ignored without PSE, but now, the page directory entry with PS set to 1 does not point to a page table, but to a single large 4 MiB page. The page directory entry with PS set to 0 behaves as without PSE."
A further insight:
http://www.rcollins.org/ddj/May96/
Edits: 02/13/12
Hello, it's a thing I have seen several times and always wanted to achieve. But where in registry, what is exact syntax, I couldn't find it.
It would be great, even bigger, say, 16 mb pages, it means less run to and from for processor and a tighter, read - more precise, timing, I think.
Serge.
Following upon Ryeland's example of changing size and font in cPlay via resource hacking, I took on cMP. Using reshacker I was able to enlarge the window and reduce the font in the same way. It turns out cMP does NOT truncate the cuesheet listings, so much more information gets shown, especially helpful with classical recordings comprising several works. However, the listings became non-alphabetized. With some experimenting I learned that this was due to the way reshacker compiled the changed dialog. I found a better program Resource Tuner which compiled it correctly. This is offered with a free 30day full-function trial at:
http://www.restuner.com
I found a better program Resource Tuner which compiled it correctly.
Thanks for the tip. I did try Resource Tuner earlier but, though I could delete resources easily enough (and, helpfully, in bulk), I couldn't find a way of deleting elements within a dialogue such as buttons. Was that me or is it what the programmer intended? (I can delete them with Res Hacker.)
I'm glad Resource Tuner solved your problem but I can't reproduce what you report with my "Resource Hacker"-hacked (sorry) UIs. For me, cMP lists alphabetically by album title. No idea why there should be a difference.
Dave
It is puzzling, but also your way of doing file names and constructing cuesheets is highly individualized. I have tried to follow your description of how you go about it with only partial success. It brings unusual outcomes. For instance, ordinarily cMP does not list by album titles at all. It has alphabetized primary lists of genres, artists and all(starting with performer/artist). Album titles are only alphabetized secondary to those, so I still don't understand how you get a display of all album titles alphabetized. Seemingly, this sort of difference could explain the different behaviour of the two hacker programs as well.
In any case, it is great for anyone desiring it to be able to modify the size and font of both parts of cMP2 to suit his display type and taste, to say nothing of slimming made possible.
Hello, everybody. I was just messing around with the system hive and again deleted smth wrong. The system started to a BLUE SCREEN, so, as usually, I booted into another windows, deleted everything in config folder and repopulated it from reserve folder. Started cmp machine and... it won't boot, telling me it misses config file.
So, I deleted everything again, DELETED WIN.INI, opened working system in regedit just to create a log file and rebooted into cmp again. All went smooth.
So, It's either the log file or deleting win.ini. And I think it's the latter.
Serge.
default 8, security 8, sam 12, software 16, system still 350 kb!!!
Edits: 02/11/12
Inspired by your kunacy I have been making my way.
At 20kB and the only things I have left to delete are the MICROSOFT and POLICES entries - I am assuming the entries that mention ASIO and my soundcard must stay?
But who knows?
My system is at 252kB.
Sorry I am not good at making clear screenshots. Grant has given me the tool but I have not found the time to learn how to use it.
What more have you removed, Serge?
Hello, Rick, Hello, evertbody!
The Goog thing is that Software can be reduced to 16 kb (see the pic of what remains).
The Good thing is that Mr.Wong has found even smaller versions for some files, and their substitution, especially one of *.nls, csrsss.exe and cssrv.dll give brilliant results, however small the difference in sizes.
The Bad Thing is that not all his files I can use and most of xp-pro users, especially basesrv.dll, what a pity, really, as it's one of the most dirt-important.
The really UGLY thing is, I never had my system backed-up like some of you do with some image, I don't know what You do and how, but this is what happened. Mr.Wong sent me his watchdog.sys, that is smaller, but warned me I have to substitute among kernel and vga.sys, and videoport.sys (a long mortal friend of mine) also hal.dll. And when I did it, the sun stopped shining, my mermaid girl has abandoned me, four horsemen rode across my sky... Well, it all BURNED F***ING DOWN, with no restore. No config substitutions, no total windows folder rewrite, nothing would help. It seems, that NTLDR and NTDETECT.COM are working in conjunction with hall.dll. I am with no cmp again. But I don't griev, as it will allow me to start again from scratch and, documenting every step, to "pave" the way for other NEW inmates to achieve our ULTRA setups easily.
Serge.
See the pic for software. I have deleted WINDOWS, but left windowsnt. Also, delete all persistent handlers in class.
DON'T ASK ME FOR THE JW FILES.
Not only can you get rid of persistent handlers you can get rid of all of the file extensions.
Now to see if I can combine our work!
Sorry about the start over, even though there is always something to be learned.
I would certainly keep my config folder since it should work just fine in your new installation.
You need to get SNAPSHOT. It is worth the money!
Good evening (It's late and very fine and snowy evening here) to everybody.
Surely, I'd invest in this thing, but what's this. Please excuse the wild boy. I am used to surgery without anaesthetics...
Joke. I really would like to know about the ways to restore fast. Can anybody tell me, please.
Serge.
BTW, MR.WONG has just sent me another set of files and a possible way to a tweak. HE IS A JEWEL!!!
Will post it in the morning, just let me restore
OK, "Editing resources in cicsMemoryPlayer.exe and cPlay.exe" would be a more accurate heading but it doesn't somehow sound as slick as "Slimming cPlay" seems to.Both programs carry some crud - icons, bitmaps and the like - that can safely be deleted. There isn't that much but, as another inmate has confirmed, the resulting improvement in sound quality is noticeable. Of course, those who use the track navigation buttons on the bottom of the cPlay window are restricted as to what they can delete but, even so, it might be worth at least trying both programs on a rigorous diet.
A little care needs to be taken with cMP. It seems that some buttons (e.g. the "Explorer" and "Startup" buttons in the top RH corner and the "Genre" ones on the bottom) are needed for it to load though I'm not yet certain exactly which ones.
If you alter settings from time to time, you need either to leave the settings dialogs in place or be prepared to pop the original executables and/or dot-ini files back while you do so. Might KS's recent batch file be suitable here?
What I also found was that I could edit various window parameters to get a display more to my liking AND get round, at least partly, cPlay's notorious Big Letters, Small Message limitation.
Do this by opening cPlay's Dialog 151 and using Edit Dialog to change the font size (takes a bit of finding). You then need to alter the dialogue's Width and Height inverse pro rata to the point size you select to get you roughly where you want to be. For example, to go from size 28 to 18 type (as pic - I presume it's points but it doesn't say), I changed the dialog size from 151 by 106 to 226 by 159:
28 divided by 18 = 1.5; 151 x 1.5 = 226 and so on
The result is not perfect but it's good enough to get started - you can always "fine tweak" later. As the pictures below suggest, I'm not finished yet. I plan to try a horizontal scroll bar as well.
Why I don't yet know but unusually long track title (> about 35 characters) are cut off before the end. See e.g. tracks 8, 9 & 13 in the pic - there's nothing wrong with either the tags in the files or the cuefile derived from them. As few track titles are as long as that, the bug may not matter much in practice.
You can do the same thing with cMP's UIs. If anyone does try anything on these lines, do please let us know how it goes. It might even be worth a try with Foobar . . . if you promise not to tell on me.Finally, many, many thanks to whoever first thought of using a resource editor in our context - it was clever. nagual19? (Sorry if I have the wrong name but you know who you are.)
Dave
Edits: 02/10/12
I did some minimum cplay hacking and it was good. But now I have now replicated the above clean cplay look with Dave's help. Yes I lost some function (volume, balance, scrolling within a track, polarity control ...well maybe a bunch of function) but my oh my the sonics are way better. I mean way better...dynamics, strong tight bass, extended highs, very very upper midrange grunge free. You gotta try this. I now have 3 sets of volume controls to adjust but hey why should it be easy when it sounds so good. I wouldn't want it any other way.
Ted,
If you are not looking for the best "polarity" of a recording I wonder how you can hear the subtleties of this cPLAY slimming.
The file is not reduced that much!
I slimmed the settings dialog to just do what I want it to do. All of the other buttons and text are gone. I got rid of the diagnostics screen since I do not need it anymore (and since I deleted its button). Same with the add disks dialog (more buttons. Personally I think this is a better way to slim.
All is a compromise but being able to switch phase should be important to you. I know I could not live without it.
If you are not looking for the best "polarity" of a recording I wonder how you can hear the subtleties of this cPLAY slimming.
It's 4am in a sleepy suburb of Smallsburg, MO. All is quiet - not even audiophiles can hear cats prowling in the back yards. Three large cars pull up outside Theob's house; twelve officers of the Audio Enforcement Agency crowd round the front and back doors, all wearing flac jackets and carrying night-wavs.
Suddenly, all hell breaks loose as the doors are rammed. Lights ablaze, Sargeant Dwight B Test (known to colleagues as DeeBee T) barks orders, 'Officer Cable, listening room! Officer Tohnarm, basement!'
But it's for show only; they know what they're about.
A tense hour later, DeeBee asks, 'What have you got?' Cable lists six mini-ITX mobos. 'All bar one of them dead! What business has any honest citizen wrecking five of the things? He's up to something, I know it.'
Tohnarm agrees: 'Two Heathkit 60s-vintage tube amps, neither working, two two-way Miranda Wrights still with electrolytics in the crossovers and an old Quadraphonic Converter still in its box but not a phase button in the place . Never seen the like in twenty years in the AEA.'
'Arrest him. Read him his Magneplanars and take him to the station. I'll faze him soon enough. Good work, men; the lab boys will be round shortly'.
The gear you mention would probably not reveal the difference so there is no need with that set-up.
Once you hear the difference it is clearly audible.
It is not immediately obvious when you first experiment but with time and spending time with each polarity when listening it will become clear that there is a best position for each recording and sometimes for tracks within an album.
I find the difference to be, at least, as audible as many of these other mods we have used.
Plus without a remote it makes for a good excuse to get out of the sofa.
If you can't hear it, I guess, you can't hear it. But I think that is impossible ... unless you never tried!
Oh oh busted again!!! LMAO!!!
It is all a matter of compromise. Yes I do miss the phase switch. But cplay sounds way way better w/o it because of the great sonic impact of the tweak. You really have to try it. But you are right we all to individually decide which way is better.
Edits: 02/13/12
The listings are too large but the title--artist and album at top--is too small. I used to enlarge this via Display> appearance> advanced> Active Title Bar, but with nLite and slimming, especially hacked shell32.dll, this is undone. Anyone know where one can set the title bar font size in the registry directly?
Hi Riboge
The "WindoMetrics" key controls this. You can edit/ADD it in the DEFAULT hive... look for [HKEY_LOCAL_MACHINE\DEFAULT\Control Panel\Desktop\WindowMetrics], but the job is not as easy as setting "10 pt." to "14 pt." and your font increases.
There is a nice, detailed description of these items here -> http://www.virtualplastic.net/html/desk_reg.html
I suffered the same fate when I slashed my registry hives, and some settings reverted to the Windows defaults (just as you, I previously increased the title bar font for the very same reason).
My solution was to go back to a previous image of my cMP_OS (when the desktop settings were still as I has customized them) and export the entire WindowMetrics key. This was imported to my current DEFAULT hive - and BINGO !! - colours & font sizes appear just as before. In my case, since the DEFAULT hive had been mostly emptied (toward the goal of best SQ), I chose to ADD these few entries in return for the GUI I am comfortable with.
If you also have a custom colour scheme, the same technique can be used, where you export/import this key: [HKEY_LOCAL_MACHINE\DEFAULT\Control Panel\Colors].
I hope this gets you back to where you want to be.
Cheers,
Grant
That's not a Toy... IT'S A TOOL !!
.
not just the frame dimensions and font size. This wasn't clear but became apparent in trying to reproduce this.
Thanks, Ryelands, I have been hoping for a way to do this all along and asked cics about it several times. Of course, he wanted none of it.
Nice sq pop...thanks Dave. Another item that maybe some of you may not
have done is the compacting of the Default hive provided by
Steppe/Jackwong96. When I 1st got this hive the results were so good
right from the go that I never looked at it. But if you do compact it,
it will go from 256 mb to 12 mb.The tweak above is very good. Ims I get louder, more dynamic, more
transparent sound all around.
Edits: 02/10/12
Nagual 19. Sorry that he is away from the thread so long.
Your Idea about horizontal scrolling is good, though I was against any "interference" before, now, that I have abandoned the tiny in-built touchscreen display in my zalman case, it seems MUCH more sensible. Also, don't know why and how to change it, but my cmp won't get into full screen no matter what I did. It was from NLite installation onward. Cplay can (by modifying ini file adding #M), but the "informative" part of the window will remain the same though it will jump top left, with the "canvas" expanding to the right and to the bottom. It doesn't matter musch, as I use it now mostly minimized.
Serge.
P.S.Have done some tweaking on both with reshacker some time now, good result, so I hotly support Your advice to other inmates.
P.P.S. I am looking for the way to attempt running a sequence of cmp-cplay being loaded at boot time as WINDOWS PROCESS with maximum priority. I hope, if achievable, it may bring a jump in performance, with most other services deleted or defeated.
My thinking is: would it work if I changed the stuff in the SOFTWARE hive to look in system32 instead of PROGRAM FILES and simply put the files in system32.
I think the problem is the .pth aspect.
I placed it in the system32 folder but I get the idea WINDOWS does not look in there for such a thing.
The machine would get to the point of the cursor showing but no cMP screen.
I should have tried it by leaving the .pth in the PROGRAM FILES folder but my premise was to attempt to get rid of the folder.
I did find that one cannot simply put everything back where it was and expect it to work. I had to copy the image back onto the disk to get back to a working machine.
So the question is: where to put the .pth folder and have it recognized?
One can get rid of more stuff in the folders. And delete stuff out of the .pth file. The last two items I remember for sure. I am writing from memory at work.
I suspect there is a way to do this and make the machine even more of a dedicated cMP machine.
Maybe find a way to put the .pth information in the REGISTRY?
Your Idea about horizontal scrolling is good
Except that the cutting off of long tags makes it a bit superfluous . . . there's nothing to see if you do scroll to the right. I should have thought of that!
cmp won't get into full . . . It was from NLite installation onward
Er, don't use nLite? Others may like it but I don't trust it.
attempt running a sequence of cmp-cplay being loaded at boot time as WINDOWS PROCESS
That would be excellent though I've no idea where to start.
Fellow cMP community,
I wonder if any of you have wider experience and/or tried out the optical (Toslink) transfer method on your systems.
cics reports excellent results on this due to complete galvanic isolation.
I´m playing with the thought of going optical on my FireWire as I have discovered that one can get glss-fibre extentions for FW.
It seems that these extension-devices support the full range of conectivity up to FW800.
Any thoughts before I try it out ?
kind regards
cMP2 Computer w/digital crossovers & FIR filters > Lynx Aurora 8 FireWire /192kHz all channels. 2x AcousticReality Ref. 202 & 2x AcousticReality Ref. 601´s ICEpower. Magnepan MG3.3R beechwood frames & custom stands. Miller chokes
I use a 4.5m firewire cable. Do you really need longer than that? Looking around at the topic of glass-fibre extensions, I see they are meant for lengths of 20m or more and can get very expensive if you use repeaters. Aren't there easier ways to get galvanic isolation, e.g., by using 4-pin at one end or both (with or without adapter)? Is there any indication that glass-fibre only connection is doable and in some way superior?
Hi riboge,
It´s not the length that triggers my consideration.
-it´s the galvanic isolation.
I´ve thrown up the question on the Digital Asylum and AbeCollins has severe reservations.
I was only trying to snap up some thougths.
Thanks for your reply
kind regards
cMP2 Computer w/digital crossovers & FIR filters > Lynx Aurora 8 FireWire /192kHz all channels. 2x AcousticReality Ref. 202 & 2x AcousticReality Ref. 601´s ICEpower. Magnepan MG3.3R beechwood frames & custom stands. Miller chokes
WHAT: A Windows Batch File that integrates cPlay for flac and wave files, with another player (foobar2000 v0.8.3) for all other music files, and provides automated rate switching for cPlay:MOTIVATION: Recently I discovered that a DAC that 1) generates its own timing, 2) generates its own regulated power, and 3) isolates the digital audio chain from the analog chain, removes most of the defects of digital audio that makes it sound worse than analog audio. (The HRT Music Streamer II at US$150 does all three things, but there are others as well, but perhaps not as inexpensive.)
I then became interested in comparing the sound of music playing software, using this DAC, and found that the smaller and lighter the program, the better it sounded (in other words, the least distortion it applied to the original audio). The free program "cPlay" - developed in this very Forum - continues to produce the best sound quality of any player I have tried. (Amongst the others, currently MusicBee is notable for combining stability with a lot of fun features, such as lyrics scrolling in sync, album art, display of web information on the artist, etc - but it seems those features come at a sound quality cost.)
My music playback PC is also my living room video PC, so it runs Vista, and so I have been unable to run the "cmp" system that replaces the Window shell, since it does not work with Vista. However, I did build that PC from scratch using "quiet home theater PC" principles that mirror the ones in the "cmp2" hardware system. And I had already disabled all unnecessary services and software for better video playback, so that also mirrors some of the cmp principles as well.
DETAILS: There are two difficulties with using cPlay that I have sought to overcome:
1) It has a fixed output sample rate. cPlay was written when virtually all music was 44.1 khz and so upsampling to a much higher rate had some benefits. However, now we have hdtracks providing 88.2, 96 and 176.4 files. My DAC is limited to 96khz, and upsampling from 88.2 to 96 definitely sounds worse than just leaving it as 88.2 khz. So, I need a system to set the sample rate based on the file's rate.
Fortunately, Bibo01 posted a batch file to this Forum a few weeks ago that changes the cPlay .ini file's RATE setting based on the user's keyboard input to a prompt. I modified that batch file to instead look for a text file in the same folder - "44.txt" to use 44.1 khz, "48.txt" to use 48 khz, "88.txt" to use 88.2 khz, "96.txt" to use 96 khz, "176.txt" to use 176.4 khz, and "192.txt" to use 192 khz. (The files can be empty, zero length, text files.) Since you can right-click in a folder (using Windows Explorer or a replacement) and select New File or New Text Document and name it, then you specify the output sample rate very quickly and easily. And this can be more sophisticated than "use the file's sample rate" if, like me, you have files with rates greater than the DAC's limit. So "88.txt" in a folder with 176.4 khz files will downsample in a clean way by tossing out every other sample, rather than intensive computation to go from 176.4 to 96.
2) cPlay only plays WAV and FLAC files. I have many lossless audio files in other formats, including APE, SHN, ALAC and WV, and also some lossey files in MP3 and WMA format. If you start cPlay and try to use it to play music, you find that sometimes you want to hear music and it is in APE or some other format unsupported by cPlay and you have to close cPlay and then find the music again to play it with another player.
So, I wanted a system that will play all my music files, and use cPlay for all WAV and FLAC files, and another player for all the other files. In my personal testing, I found foobar2000 0.8.3 to be the best sounding player outside of cPlay - better sounding than the current foobar2000. It starts up instantly, showing that it is very small and light (smaller than current foobar which has added a lot of code over the six years).
HOW TO USE: The simplest way to accomplish my goal, was to use a file explorer, and associate all audio files with foobar2000 0.8.3 and then associate WAV and FLAC files with the batch file mentioned above that sets the sample rate. In order to have the most minimal setup for best sound quality, I chose a popular small, light, free file explorer and included a line in the batch file to kill its process before playback. The batch file also runs cPlay in realtime priority using the same lines from cPlay's included batch file.
However, that only allows you to play one FLAC or WAV file at a time. In order to play a whole album, you also need to associate CUE files with the batch file. You can then use Alan Jordan's Recursive Cue Creator to create CUE files for every folder in your music library (available from alanjordan.org). (By the way, you don't need CUE files to be associated with foobar, because if you click on a full-album APE file, it will look for a CUE file in the folder for the track information.)
A link for the cplayrun.bat batch file is at the bottom of this post. The provided version includes code that supports rates up to 192, but is intentionally limited to 96 - since that is my DAC's limit, and may be yours. If you have a 192 DAC, open the batch file in a text editor, and there are comments explaining the lines to remove to allow it go all the way up to 192.
The current version of the batch file in this cPlay thread is the second version 0.2.0, which includes two new additions: (1) added support to automatically use cPlay for cue files that it supports, and foobar2000 for all other cue files, and (2) support to add a folder context menu entry that will choose cPlay for folders that can be played by cPlay, and foobar for all other folders. (For those who cannot figure out to add that support to the registry, let me know and I'll provide a .reg file to do it.)
Also, the batch file should be put in cplay's folder, and it is hard coded in the batch file to be:
C:\Program Files\cics Play
If you have cplay in a different folder, you will need to change the location inside the batch, in several places, using a text editor.
You can also edit the batch file if you want to use a different file explorer than xplorer2_lite. Note that if you try to kill Windows Explorer, in most versions of Windows, it will kill the whole shell, so it is cleaner and simpler to use another file explorer for selecting your music.
The free file explorer xplorer2_lite can be downloaded from dev's site:
http://zabkat.com/xplorer2_liteTB_setup.exe
NOTE: you'll want to unselect the toolbar installation to keep a minimal setup.
foobar2000 0.8.3 can be downloaded from File Hippo at:
http://filehippo.com/download_foobar2000/55/
otachan's audiophile ASIO plugin for foobar2000 0.8.3 can downloaded in a dizzying array of variations from:
http://personales.ya.com/angel49/foobar2000_otachan/intro.htm
the fourth choice of ver 0.51 seems to be the most generic and works fine.
NOTE: Use of any software provided in this message, is done at your own risk and no warranty of any sort is implied. It is proper practice to backup any important data from your computer before installing anything new.
Edits: 02/08/12 02/08/12
Hey Mr. Stuart,
Welcome to the phenomenal world of cMP2.
I´ve read your contribution and your motives with the automated switching feature. This is all very well, but I can´t supress my feeling that you have not quite understood what cMP2 is all about.
By all means, everyone can tweak/configure software so it fits into their expectations and needs, but I´m quite certain you´re compromizing the very core of the cMP2 strategy : the unique upsampling technology of cPlay; the quality enhancing cMP shell AND the efforts of getting rid of overheads/ processes of the OS.
On top of this, you´re claiming that : "My music playback PC is also my living room video PC, so it runs Vista, and so I have been unable to run the "cmp" system that replaces the Window shell, since it does not work with Vista"
-of course cMP2 works on Vista (albeit XP is the better OS for cMP2)!
I do aknowledge that cMP2 is difficult to grasp fully on the first attempt, but for those who are interested in cMP2 (but have maybe not taken the plunge yet), you´re conveying the impression that cMP2 is just as versatile as one fancy....I´m sorry, but it is not !
No one will get to the very core performance of what the system can achieve, when newbies get the idea that all the effort is irrelevant.
kind regards
cMP2 Computer w/digital crossovers & FIR filters > Lynx Aurora 8 FireWire /192kHz all channels. 2x AcousticReality Ref. 202 & 2x AcousticReality Ref. 601´s ICEpower. Magnepan MG3.3R beechwood frames & custom stands. Miller chokes
Thanks for your post !
You wrote:"I can´t supress my feeling that you have not quite understood what cMP2 is all about."
There are hundreds of pages about cMP2 for 6+ years, and its own web site. It is pretty well documented.
This post was about using cPlay, clearly in a non-cMP2 environment, simply because it is better sounding than any other music player.
You wrote:"of course cMP2 works on Vista"
You must have missed my three recent posts asking for a fix to the Vista problem. When one engages "cmp mode" under Vista, and reboots, it hangs at the Welcome screen. The fix offered by cics on the web site for that very problem does not work in Vista. There were posts 3 years ago by riboge that he had fixed the problem, but I emailed him and he has forgotten what he did. There is the second problem with Vista that the cmp internal explorer to pick cue files does not work at all, there was a sub-thread about that, 4 years ago IIRC, and it never ended up with a fix either.
You wrote: "...the efforts of getting rid of overheads/ processes of the OS."
You must have missed where I stated:
"However, I did build that PC from scratch using "quiet home theater PC" principles that mirror the ones in the "cmp2" hardware system. And I had already disabled all unnecessary services and software for better video playback, so that also mirrors some of the cmp principles as well."
You wrote: "the unique upsampling technology of cPlay"
First, I would like to point out how the upsampling technology is exactly the opposite of cMP2's "efforts of getting rid of overheads/ processes". If small Windows services produce an audible degradation of the sound, then surely SOX or SRC produces a significant degradation.
Again, I wrote in my post: "My DAC is limited to 96khz, and upsampling from 88.2 to 96 definitely sounds worse than just leaving it as 88.2 khz."
When I first installed cPlay, I was curious about the upsampling idea, so I did extensive comparisons, and I had come to the conclusion that 88.2 files definitely sounded better at 88.2 than upsampled to 96. Then, a few days ago, I was testing new headphones with files played by cplay, and the headphones were sounding better than I had expected, and then, on the last files they were sounding worse. Puzzled, I happened to glance at the top of cplay and noticed " 88.2-> 96 ". When I played the last file again without oversampling, the headphones clearly sounded better. It was unexpected and obvious (and thereby cannot involve placebo effect).
The main audiophile principle has always been "less sounds better". I understand the theoretical advantages of upsampling 44.1 to 192, but these days, most audiophile music files are 88.2 or 96, and many audiophile DACs are still 96 rather than 192.
Note that 96 files played in a 96 DAC cannot be upsampled (and 192 files cannot be upsampled to a 192 DAC), so how can upsampling be a crucial part of playback, when current audiophile music files are at the same rate as the DAC ?
That is probably why a number of people have requested automated rate switching, and why it was my main motivation for cplayrun.
You wrote: "...newbies get the idea that all the effort is irrelevant."
You may have missed that I first posted cplayrun as a new thread in this Forum, and the response was a very loud chorus of quote "I can't be bothered with cplay and cMP2".
The cplayrun batch file provides a way to use cplay that a) is convenient and simple (by using the Window Explorer and file associations that most nerds and DIYs are already familiar with), and more importantly b) plays all music files, not just flac and wave (I would be willing to bet that a majority of audiophiles that use Windows happen to have mp3 files in their music library, and probably Apple or Ape files as well). When one is running cplay or cmp, and selects music that one wants to hear and gets the "file format not supported" error message, the inevitable reaction is frustration and the thought "this is not a practical way to play music files". The cplayrun batch file solves that problem.
Whereas you think that providing this easier path means that many will use cplayrun instead of cMP2, instead I think it far more likely that would provide a path to using cplay instead of not using it at all, and then the positive experience will lead them to investigate cMP2.
Hi Ken,
Thanks for your fine response to my "comments".
The issue with Vista is acknowledged and so is your argument about peoples wish to playback different file formats.
There is no doubt that cMP2 is not perfect and enhancements are always welcome. Despite I have no mp3´s.....:-)
Upsampling is a native feature of cPlay and I can comprehend that users wish that cPlay would playback files at the samplingrate as they come. But cPlay is designed for the sinc-filter upsampling by Secret Rabbit Code, which (in an ideal world) does not compromize/gets deterred by a slightly off samplingrate.
The SoX feature came later in the development of cPlay, and I presume as a reference to the sinc-algorithm upsampling.
SoX is entitely a different approach to upsampling.
The sinc-function of Secret Rabbit Code tries to calculate what was analog in the first place, and is not lead by even(double/quadrupple)coefficents as normal upsampling in D/A converters do.
Note also, that most 88.2 kHz files you buy, are converted DSD files from SACD format releases.
-those are reportedly not very easily/well converted into FLAC/WAV...but that is entirely another story....
By all contoversies. Welcome to the cMP2 way....
kind regards
cMP2 Computer w/digital crossovers & FIR filters > Lynx Aurora 8 FireWire /192kHz all channels. 2x AcousticReality Ref. 202 & 2x AcousticReality Ref. 601´s ICEpower. Magnepan MG3.3R beechwood frames & custom stands. Miller chokes
Hello, everybody! When we think it must by now be probably over, new possibilities for enhancement present themselves.
Load Your SYSTEM hive go to:
HKEY_LOCAL_MACHINE\000\ControlSet001\Control\PnP\PciIrqRouting
and
1. delete there instances for USB controllers (asuming You don't need USB, hopefully) and
2. empty also there IrqRoutingTables. You will see there scary vendors name, off they go. See the pic to leave only the needed irq miniports and empty irq routing tables entry.
Serge.
Edits: 02/09/12
Not just there.
I had tried deleting all of the IRQRoutingTables but had probably done something else at the same time and thought THAT was why the machine would not start.
Now to give it a try again!
I just started using cPlay tonite and cannot figure out how (or if) to get the software to automatic change the sample rate to match the selected track.
Hello, sir. I have been visiting computeraudiophile site, namely jplay thread and saw there Your? comments. You were telling other people there, that You are evaluating cplay and jplay and have only just started the process. May I ask if
1. You are using cplay as a standalone player or in conjunction with cmp
and
2. Have you done any/no/all of the tweaks that are proposed here (not only by me, but by many other inmates).
If I may be so bold, I'd recommend first to "perform" all these tricks and tweaks, and then to compare. I think the difference is very big.
Also, I'd like to point out, that many things about processor-ram interaction, that were discussed there and advocated by Joseph and Marcin, are also my point of view, and, unfortunately, they were not receiving sufficient attention on that thread or were met with complete misunderstanding of what jplay authors were implying.
With my respect, Serge.
I wrote a batch file to automatically change the rate to match. See the link (ironically, it is based on a batch file written by Bibo01, so that shows you how messed up this Forum Software is, that he did not see I posted it):
Of course I did - I just didn't want to ask you for royalties ;-)
Simply - it is not possible!
You do not have automatic switching sample rate in cPlay.
I'm evaluating it against foobar, Jplay, and JRMC. Without auto switching, it's gonna need to be clearly better (sonically) than the others for me to use it.
What is the point?
Someone tell me.
It has some importance as a timestamp bearing file, so when I have to restore registry from reserve, I always delete win.ini as i remember that oncve I really HAD to delete after some crash, but probably it was kernel? I don't remember. I never had it reappearing until I started doing mass registry deletions.
Serge.
and had copied and re-inserted conif many times with no ill effect.
I did always copy and re-insert the whole folder.
Hello, everybody!
1. This new set of tweaks comes in part (THE BETTER PART) from Jack Wong. He has sent us all the new slightly modiefied default hive, that contains only one entry with settings to SEVERELY MINIMIZE video interference on the motherboard level. This hive can be used BY EVERYONE, who has emptied his own. Just overwrite it and reboot. BUT BEFORE you do it, check that Your cplay-cmp can minimize-maximize. The tweak works in any case, but with cplay set to minimize, it's SIMPLY GREAT. So, anybody can e-mail me and get the hive. I also encourage and even ask inmates, who already have this new hive to distribute it to others. In case you get stuck in minimized mode, just power down your cmp, open its hdd as a slave from another machine, edit your cics play. ini file to delete MINIMIZE string, reboot.
2. Open Your system hive, control set 001, services and IN EVERY SERVICE set the ERROR CONTROL value to 0, don't touch other settings.
3. In RPCSS and SAM services entries open Security subentry and delete inside, see picture
4. In Services-vgaSave open Device0 and set InstalledDisplay drivers to vga, removing everything else. (Don't do it in any other place)
Happy Listening.
Serge.
The PICTURE SHOWS registry tweaks 3 and 4 from this set - vgasave, sam-security and prcss-security, hope it's not too confusing.
Edits: 02/05/12
Long ago I eliminated AHK & Cicsremote from cics memory player. I use a ps/2 mouse. Now that I am going dark screen to listen it would be convenient to have the right click-- next track function-- back. So I restored the above files amended the .pth file...still does not work. Anybody know what windows files are required?
Edits: 02/11/12
Not very sure, but you can try restoring wsock32.dll
Thanks it was worth trying but it did not work.
...the magnanamous (SP?) graciousness of Grant. The ability to limit folders/discs in the music library does affect sq. I think the, what I call the Ryelands routine (loading only the cues into the folder only), is another example of this same principle. Some how or other I believe that the CMP library (not the files themselves but mapping to the files) is loaded into memory and that when you limit its size you make CMP sound better. Its just like getting rid of windows overhead not required for functionality.
. . . the CMP library (not the files themselves but mapping to the files) is loaded into memory
As we all know, when cMP is launched, it scans the folders listed in the (wait for this . . . ) "Folders" window in the Settings UI to look for cuesheets. Double-clicking the title of the displayed cuesheets launches cPlay.
The paths and names of the "watched" folders are stored in the .ini file in cMP's folder in "Program Files" but the content is not. The captured data reside in memory until the user either reboots or hits the "Refresh" button, at which time the folders are rescanned.
If you have a large collection of albums with all of them "reported" (as it were) to cMP, scanning can take a considerable time; a slowish machine scanning a large library over a LAN can take several minutes.
As no-one is ever going to tag every cuesheet perfectly and as cMP was never designed to cope with large libraries, it can be very hard to find a given album and the whole process is in any case impossibly slow.
I get round that by copying the cuesheet for an album to be played to a folder on the server that is "watched" by cMP then hitting the Refresh button in cMP. I've configured the server using shortcuts and the "Send To" menu to make this quick and easy - it's nothing like as cumbersome as it may sound.
The result is that cMP typically never has to "remember" more than a couple of cuesheets. As my 6,500-odd cuesheets come to over twenty MB of text, there is obviously a significant reduction in memory demand as well as in time wasted scanning thousands of cuesheets that will definitley not (for obvious reasons) be played this session.
I don't see off-hand how you can avoid superfluous cuesheets hogging memory where a system is controlled locally but Explorer is no longer available unless, of course, you use cMP's "Add/Remove" facility to browse the library. (This is certainly viable - a friends of mine does just that as he hates the cMP UI even more than I do.)
It should certainly make it easy to see if minimising the number of scanned cuesheets affects sound quality.
Could you please give the specifics of how you configure your server to do as you describe. It seems you must use another computer, the server I guess, to do this then return to the cMP computer to play the transferred cuesheets, yes? Or is this workable with a NAS only networked to the cMP computer?
Could you please give the specifics of how you configure your server to do as you describe.I'll try. This describes how I currently do things; it got simpler over time as these things tend to do.
Setting up:
1. The library sits on a 2TB HDD on a desktop computer organised in folders and sub-folders by genre and artist, pretty much like everyone else's. Call it the M:\ drive. Thus, there's M:\Classical, M:\Jazz, etc at the top with the usual sub-folders beneath.
2. There's a folder called M:\00 which I call the "watched" folder. The drive is shared at root level so that the cMP^2 box can access both the "watched" folder and the data. (Using M:\ for both local and network drivenames makes the cuesheets valid for local and remote use.) I made a shortcut to M:\00, moved it to the "Send To" folder in my Docs & Settings folder (which is normally hidden) and called it "00".
3. Cuesheets sit in a separate folder (where doesn't matter, whatever suits) whose sub-folders mirror the structure of the main library. (There's a DOS command that does that for you.) The cuesheets need to contain the full path to each music file but, again, AJ's Cuesheet Creator does that for you if you ask it nicely. Only the user needs to see this folder. Cuesheets have this form (with the absolute path underlined).
FILE " M:\Classical\18th Century\Mozart W A (1756-1791)\Concertos\Wind\Mozart - Clarinet concerto in A, K 622 - de Boer, NS Amsterdam, Markiz \01 - Mozart - Clarinet concerto in A, K622 - Allegro.flac" WAVE4. I made shortcuts to all the main or favourite folders in my library and put them in a folder out of the way but accessible. I created a shortcut to that folder, gave it a distinctive icon and put it in the Launch Bar. Call it the Browser icon. Next to it I have an icon that logs me on via UVNC to the cMP^2 box. (The shortcuts are not essential but they do save a lot of time.) See pic.
5. The cMP^2 box logs onto the shared drive on bootup using the Net Use command.6. cMP is configured using cMP > Settings > Add to "watch" M:\00. (I also tell it to watch a small folder on the local drive that has one cuesheet in it nearly all of whose data has been deleted. That helps to prevents irritating messages from cMP if, for whatever reason, the cMP box is not logged on and so can't see any cuesheets.) See pic.
To play an album or albums:7. To select an album, I click on the browser icon and get a basic view of the library.
Say I want to play some Chopin. I double-click "Nineteenth Century" then Chopin and then select a cuefile. I right-click on it and select "Send to" > 00. See pic.
The cue file is copied to M:\00. I usually shut Explorer down at this point as relaunching is quicker than navigating backwards.8. I go to the UVNC window tracking the cMP^2 box and click "Refresh" to have cMP display the selected album. Thereafter, it's as normal.
Edits: 02/07/12
Since like many I have lost a lot of functionality on cmp. So I rip new music on my desktop. When I have 10-15 discs ripped I shut down cmp, take out one of my data discs, load into my docking station and use copyto software to copy all the files onto the data disc. Thanks to Jack Wong for the tip on Copyto. While still in my docking station I create a folder on my data disc. Into the folder I load my cues. I name the folder Cues.You have to revise cues so they have an absolute path to the file. In my case the data disc is D: on cmp, so in the cue on the file name line (say the music is Taj Mahall Live) I revise the file name line to this
File name "D:\Taj Mahall Live.wav" wave
Now when I load the data disc back into cmp I can go to the ADD / REMOVE text and click add...then D:\ will show up along with the folder named Cues. So if I remove all other discs (yes even D) I only have those cues in the Cue file showing in CMP. Hit play and you get a nice sq pop. To test it load D:\ back into the CMP library and play Taj Mahal Live again and you will notice a sq degredation slight but noticeable.
I just did it and will now load all my cues into subfolders henceforward. A bit of work but isn't that we are all about in the pursuit of sq?
We can all thanks Ryelands for this easy tweak.
Edits: 02/07/12 02/07/12 02/07/12
... of ADD / REMOVE discs into cmp. I did this before by restoring Dialogue 1079 (as per Mihaylov...thanks again Serge)and Strings 267 1893 1907.
However when I tried loading user32 from either Pro or Home versions of XP sp2 I get a no boot error. I am using all the files from the Big Tweak and I presume that user32 came from xp 2k. Can you please send me user32 from xp 2k (or whatever its source for Big Tweak) or post it so I can get some the above functionality restored?
To everybody. DO NOT change (substitute) USER32.DLL and WIN32.SYS
I have written about it in Big Tweak.
Serge
E-Mail never got here.
I had no error messages, but I have sesent it with a different extention *.fix instead of *.dll this time, just rename it to its original name.
Serge.
..
I will resend it under absolutely different name and from another e-mail address. I wonder if Macrohard is watching their file distribution?
But this file is freely downloadable from dozen sites. BTW You have it in Your Xp disk, but it has to be unzipped.
Serge.
Hard to believe it can get that much better.
Yet another step forward in transparency - better definition extends further to the back of the image. I guess yet again a lowering of the noises that obscure.
Thanks, yet again, to Serge and Mr. Wong. Waiting with anticipation for Mr. Wong's next chapter.
I have a file---Gregorio Paniagua - La Folia De La Spagna that was a Harmonie Mundi France recording that, while another old fave of mine, must have been recorded on a warm spring/summer morning because you can hear birds chirping presumably through an open window in the recording studio. You can hear them throughout the recording but they are very faintly heard in the 1st cut. Well now you can darn near count the little rascals they are so very clear in this 1st cut!! Very nice. Resolution is incredible. I agree with Ric for those that have lost this functionality somebody has to find a way to restore it. I guess if you have a network and dont use the cmp screen or mouse maybe you dont need it.
I know I have no idea since I lost the ability long ago.
Before I knew it was something good to retain. Missed that one entirely!
Restore an unhacked user32 and the ability will return.
I think the key is STRING TABLE #57.
The rest can probably be done away with but finding out will have to wait until tomorrow.
I would have thought Grant would have posted this and seeing that he did not and that there are maybe two or three others out there who over-hacked this feature away I decided to post this myself.
Nonetheless, all credit must go to our favorite Canadian alchemist, Grant.
No more squeals.
Loads files quickly.
There is no question some of that stuff (I had deleted) is important!
Thanks, again, to Grant!
You are using now completely unhacked version?
Don't know if I should try, but I have no squeals ano problem loading files at all. I have taken fresh user and tweaked it a la Grant, only probably one or two dialogs less - I left native cursor and strings with minimize-maximize. Working perfectly. The problems You say about may be due to excessive registry deletions. I cannot make my system hive less than 412 kb, it bluescreens constantly, and I know that Mr.Wong has his config folder under 300 kb total.
Serge.
P.S. His SAM is 8 kb!, I think I will ask him to send it for me to see what else can be done. Also I have deleted more from software, will send later for You to see.
I had the problem start when I FINALLY installed the files you had sent instead of my hacked versions of the files I thought you were using.
I could only tell there was a difference by the size of completely hacked versions - yours were still smaller than mine.
I did the swap as an all at once operation and maybe I should have done it in stages as you recommended. But now the machine is running happily so I will leave it alone for the time being.
But now with Grant's minimally hacked user32 all is well. I iwll see how much more can be taken away and he is doing the same. Neither of us claim that this is the FINAL version but it did point the way for those who want to be able to minimize the cPLAY screen - for which I remain skeptical but all small things add up!
I would also like to hear the reasoning on the black screen. Being a cics loyalist I would like to know why this is better.
Of course, I was stuck with the default blue screen with nothing in DEFAULT before and must say I like the black one better!
Maybe the XPHome SAM allows one to see more stuff to get rid of? I do not think if we deleted EVERYTHING out of our SAM files would we have that same result.
I exchanged some emails with Jack and he said '...I've been dealing with electronics and noise which is the by-products. I love listening to music with absolute dark background so I've remove display chips and use batteries to achieve such effect.
After I saw this.
http://www.audioasylum.com/forums/pcaudio/messages/6/65390.html
I tried and the sound was promising. However, it needs a graphic driver to work with. I then explored other means and eventually used the background color to shut off the display...'
A black body in thermodynamics is a minimum radiation (video interference in this case) source. So the color simulates a shut off display. Very, very clever!!
I have never tried it since I worried my mouse function has been minimized!
But the way I read that - this is very different.
That procedure, seems to me, turns off the monitor - disables it and then re-enables it. Which I would think would be ideal.
Has anyone every got it to work?
I do think good things are happening with the tweak. Since I did them all at once and I am skeptical about the black monitor thing I think the security entries being turned off is the big reason it has made an improvement. Plus my monitor looks like it is glowing black, not exactly turned off black!
I am skeptical about the black monitor thing . . .
As I understand it, monitors typically draw more power to display black than white. Before I stopped using the wretched things in an audio PC, I did measurements that tended to confirm that although the difference was a couple of per cent only. What difference, if any, it makes to the PC side, I've no idea.
Dave
...I am in the process of un-hacking some stuff in dialogue/strings in USER32 so I can ADD / REMOVE folders & disks in cmp. I found in the past that when you have multiple disks or folders it sounds better when only the one necessary to run is loaded.
I would insert the orignal, do what I needed to do and then go back to the hacked.
Now, I have not had a need to do this since installing Serge's shell32.
My issue is a little different. I have 3 data disks. When all are loaded it works fine. Also I can remove 2 so I can just listen to one. However when I try reload (the Removed discs or folders) nothing shows in the CMP screen...like there are no others. I want to do this all the time back and forth to get the sonic benefit. My CMP sounds better when a minimum of discs/folders are loaded in CMP.
Easy enough to modify a copy of the original registry.
Where this total tweak reveals itself is in large classical orchestral pieces. My all time fave of Suppe Overtures on Mercury by Paul Paray and the Detroit orchestra is simply wonderful. More space, more distinctiveness, more texture to the sound. On another old time fave that I had on Lyrita vinyl, then cd, then ripped file the Malcolm Arnold Dances recorded by the composer and the LSO never sounded closer to the vinyl as now. The strings & horns (difficult to get right) are so much better now. All in all a very good tweak.
Thank you Steppe and thank you Jack.
Hello, everybody!
1. Are Your CMP and CPLAY running maximized (full screen and no blue or black background?)
2. Can You minimize and then restore the window of cmp - cplay?
If both answers are yes, can somebody, please, send me their cmp and cplay ini files:
cicsMemoryPlayer.ini and cicsMemoryPlayer.pth
from cics Memory Player folder
+
cicsMemoryPlayer.pth and cicsPlay.ini
from cics Play folder.
Serge.
You've got mail
Thank You, now, check Your e-mail, too.
Serge.
I have just discovered two things:
1) Outside of this thread, there is a lot of hatred for cPlay out there. People actively dislike it. Am I missing something - perhaps some history ?
2) Everyone who posts in this thread seems to be good people who are actively trying to help each other. Everyone who posts in this Forum outside of this thread seems to be jerks, bitter losers who have nothing better to do than belittle others' efforts.
Why did cics choose this Forum ??
for letting this thread continue to grow and grow.
It was one reason I made a contribution this year to AA. I should have given more and this should not have been my first year.
All in all AA is the best of the forums in my unhumble opinion. There are jerks everywhere, why be surprised?
Many people seem to think cPLAY should offer the same idiotic bells and whistles of the software that shows pictures and all other kinds of superfluous stuff. It kills me since I can think of no dedicated transport on the market that shows pictures while playing the album. There is always someone here making the same comment; they cannot be happy with cPLAY because it is so basic. One figures they would be happiest with a BOSE home theatre system.
cics wanted this system to compete on an absolute level with top level CD players/transports. I hope he is playing with the OS like many of us here are. We do it in the spirit of cics/cMP2. Who cares if one loses some convenient function of the computer if the result is turning it into a first class transport? It is meant to be a transport and nothing more.
For a very reasonable amount of money and varying amounts of effort, depending on how far you want to go, this can be a good to very, very good way to play CD's.
So one should add a suspicion of laziness for those who also "hate" the effort that goes into cMP. Oh, well ...
My favorite is the fellow (who owns dCS gear, he used to never miss a chance to remind us) who finds fault with cPLAY because during its (shared) evolution there were so many versions. He continues to repeat this and my mind boggles each time I read it.
Whether one can call it hate is up for discussion but it is strange that people have refused this gift. If they looked this gift horse in the mouth they would find very healthy gums.
Just found this while trying to get more information on watchdog.sys and was led to the site because they had posted Serge's WATCHDOG post on their site.
The site is run by a dealer ODYSSEY AUDIO - http://www.odysseyaudiohk.com/cgi-bin/yabb/YaBB.pl?num=1325421658
What do you think of this recommendation:
/FASTDETECT /noexecute=alwaysoff /CLKLVL /timeres=9766 /nodebug /pcilock /use8254 /nopae /noguiboot /INTAFFINITY
Seems like a WAY above average store. Maybe Mr. Wong can tell us something about them?
Anyone have any comment about these sitch choices?
/INTAFFINITY This new switch directs the multiprocessor HAL (HALMPS.DLL) to set interrupt affinities such that only the highest numbered processor in an SMP will receive interrupts. Without the switch the HAL defaults to its normal behavior of letting all processors receive interrupts.
/USE8254 This switch is intended for systems with older BIOS's. It instructs the NT HAL to use the 8254 timer chip as its base timer.
See Microsoft KB Article Q169901 for more information.
/PCILOCK Stops Windows NT from dynamically assigning IO/IRQ resources to PCI devices and leaves the devices configured by the BIOS.
See Microsoft KB Article Q148501 for more information.
Why would these be good? /use8254 seems particularly inappropriate. I know /pcilock has been discussed here before. /intaffinity is the one most worth finding out about.
I had read the explanations which are opaque to me.
Was wondering if anyone had used something like this line or individual pieces of it.
Thought it was neat that an AUDIO STORE would have a forum with such a discussion. Very nice ....
Wonder if INTAFFINITY might be more useful now with all of these "old" files?
HarryET uses similar boot.ini parameter as shown in the link.
I tried those a while ago on my XP Home SP2 system and I am unable to boot into cMP.
I tried the CLKLVL switch when Serge 1st recommended and did not like it because of a soft impact on my sonics. The others are probably worth a try.
Edits: 02/03/12
Well, inmates, believe it or not, but with YOUR HELP I have it done now. Windows is only 18,5 mb. It became possible after I had castrated the ttf fonts. Thanks to Serge Mihaylov for his idea!
Anyone, who wants, can e-mail me to get the fonts. (3 files - arial, micross and l_10646 I haven't touched marlett - no use.)
Now, security and crypt dlls won't let go anyway, sadly, But to THINK that Rick managed to kill SVCHOST!!!!!
I am happy to know such people as You are!
Serge.
Edits: 02/03/12
I am finally at 19.5 mb!! But root (with all the legacy stuff)
within enum keeps coming back. Dang!
minimizing the FONTS, as absurd as it sounds that such a thing would make a difference, makes a difference.
Now to figure out how the HELL Serge has 600kB less WINDOWS than I have!
Serge has asked me to to assist in distributing these so if you would like them send an email to my "moniker" and I will gladly forward them.
I still am trying to get to your level but I seem to recall setupapi is about 400-500kb.
I compared your file sizes and quickly realized there was something I had missed.
Went back to your files and found most of the difference.
I am ashamed I did not use them as soon as I got them. In the interim between you bringing this idea up and actually RELEASING them I thought I had found the same ones.
Also, had no idea we could get rid of vga and vga256 until I tried it. There is no need for those things. Now I want to see if this means something more can be removed from the registry.
Thanks very much. I should have known ...
Was there are prerequisite for deleting the vga's? I tried and got a bsod. FYI I shrunk security but not sam.
Edits: 02/05/12
As Mihaylov says every system is different and inexplicable different at that!
Very strange.
Nonetheless, I was SURE those were needed files!
.
That's not a Toy... IT'S A TOOL !!
Hello, everybody. I have already posted that I don't have setupapi in my windows folder, and it doesn't reappear, nor it is required to work. I can attribute it only to Lynx-Juli@ difference, as I can't remember any problem removing this file from regular retail version of XP or the VL version I am using now. VL version, however, allows to remove out of box experience (oobe)from nLite install. i do have win.ini file reappearing each time I delete it and it's a good practice to delete it if you are restoring the "burned" registry. I have a folder called config recent where I keep the latest working hives. If I have rendered my system unbootable by too brave registry deletions or while experimenting with kernel file, I delete the real config folder, delete win.ini, repopulate the config folder from config recent AND LOAD SYSTEM HIVE into another Windows' reg editor to create a log with a time stamp in it. Then I reboot, and it works perfectly. This way I have my cmp windows just like an empty shell, a cadaver, if You allow such comparison, with a registry of my choice to be its mind.
About the files' sizes. I AM SURE, that while giving all attention to registry, many have forgotten to delete EVERYTHING from EVERY file in reshacker. I have done it even with drivers, as sys files are mere dlls. Even my Lyns driver is hacked free of almost all strings. I don't have any code texts inside my file and even version info. All this crap is mapped into memory, btw. And this is how we come to fonts.
Fonts are (in computer) not just a table that allows You to print by pressing corresponding keys. There are 1674 glyphs (bitmaps in the form of a letter or sign) in every Unicode otl-ttf system font, add to it apart from bitmaps, however tiny, their VECTOR (real) incarnation. All this crap is constantly being held in memory, as these are SYSTEM fonts + their screen doubles from screen fon files like 8514fixed, sys or oem. Reducing what was possible, and I left only barely enough for normal lettering+ computer conventional glyphs like {\+=!? <> , we have a reduction comparable, if not exceeding the shell32 and user32, thus the effect is so prominent even now, with every, even big tweak, bringing less and less in sq - there is practically (almost) nothing to improve.
I think, that should we have our fonts reduced at say tweak part 10, everyone would cry WOAHHH!!! (though, I may be wrong, it's too late to check now).
So, this is my opinion and my folder.
Serge.
P.S. In order that Your screenshots be more sharp, I'd recommend to reopen the bmp files, that You create in MS Paint and save As jpg. They will be 5-8 times lighter and won't cause problems with AA.
JULI@ and LYNX require different files.
He has to keep some stuff we do not need.
Dont know how you and Rick do it. I am tapped out at 21.9 mb.
With Serge's font files I am at 19.1 mB.
Serge I would also appreciate if you sent me these.
:)
Please send at your earliest convenience!
Will have to return the WATCHDOG to the registry and hear the difference.
It was no trick deleting svchost. Just did it and the machine worked!
So what is Mr. Wong doing to rid us of WATCHDOG?
Check Your e-mail, sir!
besides, removing a 1 mb+ of raster (bitmapped) gyphs works like a cleaned shell32 to me (well, not really That strong, but the effect is audible).
Serge.
Hello, everybody! this is a new thing, I hope, and all credits should go first of all to Jack Wong. He is working on a serious project now and he was so kind as to tell me about security and sam hive encryption. Unfortunately I was unable to use his hives, but he helped me to find this program - it's free and very small, but effective. It cannot decrypt SAM but it loads SECURITY as a normal hive, also, after one-two loading and asking all kinds of permissions like you do with default hive (read Rick's post) the SECURITY hive starts visible in normal regedit.
So, download this program, unzip and install on your hdd
http://users.csc.calpoly.edu/~bfriesen/software/hiveloader.shtml
2. Run the program and tell it Load, a dialog appears that allows you to browse and find your cmp security hive, open and assign it a name, like 000. Say load once again again and press REGEDIT button WITHIN HIVELOADER.
3. Edit your security hive like the picture attached.
4. unload it from regedit in the usual manner.
5 Load it into REGEDIT once again and compact with Regopt.
it should become 8 kb
PLEASE NOTE THAT ACCOUNTS ENTRIES ARE EMPTY
all other entries have only one dword in them. Leave them be.
Please, try to tweak this hive IN A SINGLE SESSION, as I had to restore my registry many times before I achieved this, and the order was nonlinear!
Serge.
I think this is as far as one can go. It is as far as I can go. I tried many combinations and this is what works
I also found I could delete svchost.dll. Whether this has anything to do with "this" or not I do not know. No need to "suspend" that one any longer!
Tried to delete isass and watchdog but to no avail.
As I posted before one can delete WINTRUST and WATCHDOG from the registry.
The machine is working very well.
If you try this ignore the momentary squeals that you might hear. I heard them upon first re-installing the drive but after a second boot I never heard them again.
Before I finalized this the machine would take longer than usual to load a file. But now it is back to loading quickly. Sometimes it will turn itself off when asking for shutdown and sometimes it hangs after doing the typical cMP bar zipping a few times and just hangs there. Nothing to worry about.
I do think it sounds even better.
Tomorrow more experiments. Wondering if hex editor is the key to minimizing issas and watchdog? They might just need to be there even if they are rendered impotent. I hope so.
Hello, Rick! I have just trimmed down SAM and removed from security all accounts but 1-0-0-1 or whatever is its name. compacted. Sam is 12 kb. Security - 8kb.
About svchost.dll - IT IS A VIRUS - there is no such dll, I have checked it here - the place I look up the info about *.dlls
http://xpdll.nirsoft.net/
Or was it svchost.exe?
I think I will try removing now authorisation and security dlls. As for Watchdog, as it's still in the system, it must be envoked by lsass or even ntloader, so, removing it from the registry may cause it to function at default, and this is smth, we don't need. So, probably we should let it remain "sedated"?
Serge.
it was .exe
Really! I marvel at Your ability to do such things, Rick!
Please, tell me, exactly, do I load SAM under HK_USERS?, and then I will see it in two places?
About svchost - it's unbelievable, yet, it's true...
However, it's svchost.exe that is suspended together with hatefull lsass.exe (but they, yes, they call upon svchost.dll and lsasrv.)
I wonder if you are so tired that you don't listen to the result now?
I know what security deletions would bring, now for SAM...
Serge
P.S. Thank You!
.
I'm having difficulty getting Hiveloader to show contents of sam or security.
So these are the steps
run hiveloader...click load...browse to security...assign name in lower box...then click regedit or load again?
This is a bit confusing for me..so I tried both ways and I can see the name I assigned in hiveloader within regedit but no contents within. What am I doing wrong?
This also happened to me at first. Do as steppe says but I found you may also have to change the permissions in loaded Security hive by checking the lower box in Advanced as Rick described to get its contents to show.
Edits: 02/02/12
SECURITY under HK_LOCAL_MACHINE (you can select it in the same dialog where you browse for your hive, and, according to Rick, you load SAM under HK_USERS).
And then press REGEDIT (note, that regedit itself should be closed and onlu called upon by hiveloader)
Serge.
Edits: 02/02/12
Who knows what I did not do the first time I tried to load into LOCAL MACHINE.
Funny I still use LOCAL USERS when I work with SAM even though I do not have to.
Ted, make sure you allow unlimited permissions for the entire LOCAL MACHINE hive (checking the boxes in ADVANCED while highlighting HKEY_Local_machine, this should allow you access.
The first time (fro some reason) seems to be tricky.
(to quote the DAMNED a late '70's punk/pop band)
Can't wait to give it a try.
Thanks to you and Mr. Wong!
Thanks to you and Mr. Wong, again.
I also found that you can access SAM - you have to place it in HKEY USERS and it will appear. I use ___wer as my cMP files I.D.. Hence samwer.
Like the above. Could not open it up completely and get in the screen.
Now to figure out what to do with it!
no success yet in deleting anything.
Can actually be loaded into both headings.
Since it worked first in local users I have kept going there.
so far
tweaked Security? I think, it should be done, as it allows to delete and empty most of the accounts like 1-0-0-1 something. and SAM is security accounts manager, so when tweaked security "interacts" with sam, trhere's gonna be some inevitable change in the SAM, after which it may become wore pliant..., Just thoughts.
Serge.
This is where I am at the moment.
I was able to delete WATCHDOG and WINTRUST from the regsitry after this.
I tried to delete the watchdog.dll but the machine would not allow that.
Who knows what this has allowed (the registry deletions)?
I finished late and did not have enough time to really listen since it always takes a little while for the machine to settle down after these things.
Was going to see if there is more to remove from SECURITY and yes, they do seem to interact. The rxact entry which you deleted in DECURITY cannot be deleted in SAM. Of course, you can delete it but the machine will get to the default blue screen and a the cursor will show but it will not go on to the cMP screen.
I have the feeling that USERS might be able to be deleted and will try more deletions tonight.
Since the machine did not recognize ADMINISTRATORS when implementing MINLOGON I think we might be able to get rid of the whole thing. Might be we can only delete NAMES.
With SAM I had no problem making additional deletions from the same file.
Compacted the above comes down to 12kB.
Hello, everybody! I am sorry for having been away. But it happens. I am happy to see that discussion continues with more and more people getting in. My best wishes to Play-mate with his site.
I had some time to think and here is a modest and dubious result of this process:
We perform minlogon tweak and are left with only default user. Only? No!!! We ARE NOT THE DEFAULT USER, we are Administrators, and our user hive is "hidden" in the file ntuser.dat in documents and settings\administrator folder. Just load this file in YOUR REGISTRY and you will see it's a normal hive in itself and UNTWEAKED. Do it. A conservative tweaking and compaction with regopt will shorten it from 256 kb to 20. The sq change is here, strong and for the better. Unlike the default hive this ntuser.dat is more personal and contains info that is not present in default. Also, there you can set EnableTrueType setting to zero and try to boot without ttf files in fonts folder (I have thought of it just now, haven't tried this font thing).
Glad to be with You again.
Serge.
Edits: 01/31/12
is none at all.
Just deleted the whole thing and my machine works faster.
Also, configured the RECYCLER (the first one, the BIG one, to delete stuff immediately so no space is allotted for it.
I had had D&S down to about nothing anyway and this discussion got me to think just what is it doing anyway.
Nothing useful.
not sure if it is connected but I was able to delete everything else in Default.
If you delete Default entirely the system does not seem to be able to locate ntoskrnl but the system works fine without anything else.
I haven't tried this yet, but I believe if one maps a network drive that mapping is in DEFAULT so one can't just empty this hive. Anyone know different?
My fear was unfounded.
The batch file on bootup method for establishing network drive has never worked consistently on my computer. I don't know why. It works once then not again on most attempts to keep using it. One can map a drive once that persists on restarts. One simply needs to open the directory on that drive--e.g., by clicking cMP icon and then add button then + by drive letter-- to make it accessible to cMP.
Anyone know different?
If you log on with a batch file, esp if you use Serge Mihaylov's routine (which works very well), you don't need to keep drive mapping details regardless of where they are stored. Whatever, I managed to delete everything in the default hive. (IIRC, it's only what the OS uses as the default settings for a new user.)
Some stuff kept reappearing but, by using Rick M's "permissions" routine and/or deleting tree by tree from the bottom upwards, I eventually got shot of everything. Like steppe, I'm down to 8KB, still networking and liking the sound.
I've also now deleted the Docs & Settings folder. I heard no difference from doing so but it was all but empty in any case.
I did the same on my second machine but here, though the Default hive definitely had nothing in it, it was reported as 256KB. I therefore copied the 8KB file over from the other box and restarted. It's now working a treat (inc the network) and sounding very fine, better than I've had before.
Top marks to the Reg guzzlers . . .
which is one reason why I remain skeptical that networking will result in a superior cMP machine.
There is so much one has to leave in that in my opinion cannot possibly compensate for eliminating the mouse and the monitor.
Compromise is always there and each of us has to make the one that suits us best.
I have never tried nor heard a network installation so I have no idea what the difference would be. I do know that I thought there was much to gained with getting rid of all USB stuff and that is what I base my opinion upon. Whether the 1394 stuff is as polluting is a mystery.
Serge Mihaylov did the original registry slimming for his network installation and his hives pictures are available a little further down.
An excellent place to start. It is where I started and is the true pioneering work on slimming the registry. His pictures are of an active registry so you will have to keep that in mind when working with hives.
To differentiate: Serge of the Steppes registry work was finding items whose elimination improved the sound. Mihaylov took registry slimming as an end unto itself as the path to improving sound and took a wholesahe approach to deletions - get rid of anything that is not needed for the machine to work and the end result will be the best sounding installation.
Needless to say Mihaylov was inspired by Serge's minimization of the .dlls and such. Talk about hyper-threading ...
I much prefer working with the hives. Much less complicated; not as much duplication!
I accuse myself of being a naysayer in hopes I will be proved wrong.
Another case of that, here.
The result is super!
I have
1. deleted documents and settings folder with ntuser.dat in it, rebooted into cmp - works.
2. Rebooted into normal windows, loaded default hive, deleted everything inside, compacted to 8kb, rebooted into cmp - works.
3. Booted into normal windows, loaded system hive, asked for permissions to delete ControlSet02 and changed the permissions as You described. Won't delete at first because of lynx, but I opened inside and deleted Lynx inside ControlSet02 manually, then I was able to delete ControlSet02 itself completely. Rebooted into cmp - WORKS GREAT!
The added clarity in the middle of the stage (deepwise), more pronounced and drier (in the good sense of the word) bass, tighter and better defined. Longer sounding instruments in the mids A very good deletion tweak! Recommended FOR EVERYONE!
Serge.
Hey Steppe,
loaded default hive
Sorry to be so daft, but how does one do that???
Afterwards we discovered faith; it's all you need
I have another computer, that has "normal" windows (untweaked) on it. I boot into this windows with my cmp hdd attached as a slave, run regedit, place my mouse on HK_local_Machine (see picture) It MUST BE THE HKEY LOCAL MACHINE, then say - file - load hive and browse my cmp hdd to find windows\system32\config folder. There You will find system, software, SECURITY, SAM and default hives. Select default and say yes (or open) The dialog appears to assign a temporal name. I usually assign the name 000. say ok. Click now on Your HKEY_Local_Machine and under it You will see 000, That's Your (default in this case) hive. Please note, that default means default user, not some general default hive.
Please note for convenience that in my setup I have an external sata/usb box with a hdd that is a cmp system disk, It is attached to cmp machine with esata cable and simultaneously attached to the working normal machine with USB cable. The power is from a wall outlet via 220 v adapter. When I listen to the music, but when I switch the cmp machine off it automatically is recognised as a USB flash drive)
Hope this is not confusing, but read and perform everything with care, and no harm will come to Your system. It's much easier than it reads.
Serge.
a pioneer in exterminating all the registry crap! And I am at awe at your courage, as You delete the things I would otherwise not dare touch! So, do I understand correctly, that
1. I delete documents and settings with ntuser.dat in it completely and
2. after that I load default in another windows regedit and delete EVERYTHING inside it, leaving an "empty shell" of a hive, just name...
Do you compact it afterwards?
Serge.
But you see how simple it was!
There is nothing to fear! Make images or copies of the pre-deletions config folder and give it a try!
I tried deleting DEFAULT entirely and the machine could no longer find ntoskrnl. Since the compacted DEFAULT is 8kB there is obviously a few things left in there we cannot access. Undoubtedly there is a reference to ntoskrnl in there.
Also, go ahead and delete RECYCLER. You will first have to go into RECYCLER/pick one of the bins/PROPERTIES/ADVANCED/ GLOBAL/click do not move stuff to the recycle bin. Then go to the drive in question (within the same box) and say not to send stuff to the recycle bin. Then you can delete the folder entirely. If you have two recycle bins do this for both.
Can't say it makes a difference but part of the cumulative effects that we believe minimizing WINDOWS will culminate.
Hi Serge - and Welcome Back !
Once you go to MinLogon, XP uses the SYSTEM account (all-powerful, all-seeing - even more so than an ADMINISTRATOR account), and the settings for it are stored in the DEFAULT hive. It is nicely summed up here ( http://www.mp3car.com/vbulletin/winnt-based/37724-using-minlogon-to-speed-up-booting-and-decrease-footprint-26.html#post1175620 ) as follows:
"minlogon uses the System account and cuts out a lot of logon actions and security stuff going on in the OS when using accounts. It's meant for truly embedded devices that don't need any of that. It cuts down boot time significantly, but it's not for all scenarios."
Long ago I removed all instances of "ntuser.dat" from the Documents and Settings folder. Boots just fine without them. Recall that I run cMP standalone (no network stuff), so I can't comment on how this affects networked set ups.
Try cleaning out all ntuser.dat's you find, and have a listen to that state.
You could try the TTF fonts trick by exporting the EnableTrueType settings from ntuser.dat and importing them to the DEFAULT hive... might work - or not. Please DO let us know.
Cheers,
Grant
That's not a Toy... IT'S A TOOL !!
You are absolutely phenomenal...cant wait to try. Gotta get cmp to boot again. I try again tonight.
I thought we all deleted that long ago.
Unless you want it to do something, it is not needed.
It is completely redundant.
I have found sonic improvements with all redundancy eliminations in the registry. Cumulative, that is. Though there seems to be something abut ENUM/PCIIDE/ IDE Channel and deleting the second entries for Primary and Secondary channels, the bottom two. This one seems to make a difference all by itself.
My WINDOWS folder is now at 20.1 mB.
Glad you are back.
You said:
'...I have found sonic improvements with all redundancy eliminations in the registry. Cumulative, that is. Though there seems to be something abut ENUM/PCIIDE/ IDE Channel and deleting the second entries for Primary and Secondary channels, the bottom two. This one seems to make a difference all by itself...'
I had 3 sets of these and I eliminated the bottom set only and whoaaaa...I'll say it made a change. Nice call Rick. Think I'll go after the 2nd set too. Got any more of these gems?
I was down to 23 mb before I blew it up.
I was simply rebooting my cmp box today and got the dreaded HAL error. HAL = hardware abstraction layer, the cpu hardware drivers if you will. One can get this error by either having a corrupt/missing hal file or some hardware failure. Given that I had my last HAL error only a month ago and solved it by reflashing my bios...I reflashed again but this action did not solve the hal error this time. I tried reloading earlier versions of my operating system to make sure my hal file was good but that did not work either. Relative to hardware ... I measured the psu voltages and they are all good going to the mobo (12v, 5 volt supplies) and the cpu (12 v). I can enter bios to reflash or reset optimized defaults so that infers that mobo, cpu and memory hardware are ok. I ran a chkdsk on my ssd (on which my OS resides) while it was in a docking station hooked up to my desktop and it showed no errors or bad sector. Something is obviously not right so any ideas out there on what to test?
Turns out I had my Master Boot Record on my ssd corrupted. Only way I could fix was by reloading Windows. During that process one of my 2 switching power supplies failed (it was supplying my cpu and one of my data discs). So this may be at the root of my downtime. I put my cpu on the one good psu and finished loading xp from scratch. Then reloaded my 25 mb Windows with Snapshot (thank goodness for SS). So I'm good & will probably leave p4 and p24 on one PSU and drive data discs with separates. Ah the sweet sound of music!Now why does my switching psu's keep failing? Beats me. I used to run my psu's on the low side (on the pot inside I set p4 to 11.92 v). After 2 failures I adjusted my latest psu to a tad over 12v. Well now that one failed. So for a while I will run p4/p24 off of one psu to see if that helps. Who know maybe running low cpu core volts and low speed contributes. I was running at 150 core speed with 6 multiplication factor so thats 900 mhz. I run v core at a tad under 1 volt.
Now all have to do is figure out how to back up my MBR.
A big thanks to Douwe01nl and Ryelands for all their help.
Edits: 02/01/12
I do not think that turning the voltage down does it. I think it is the minimal work we give them to do and maybe the added caps.
I think the less you ask of a switcher the harder life becomes for the thing. I think they like to work around 80% capacity. We are asking for something like 20% with the EA430, if that much. I think this is as hard on it as working at full power.
Does it have to sink the extra/unneeded power within itself? I have no idea and am lapsing into my bad poetry explanations again.
I also wonder if adding all of the these capacitors does something that the switcher does not really like?
Ted, you should make yourself a linear supply for the P12. Not hard at all. If you can solder I can give you an easy list of parts that will make it easy, including circuit boards. P24 is just three of the same supplies when you are ready to have some real fun!
Rick yes I would appreciate if you sent me copies of schematics and parts list for the linear psu's you built.
Give me a day or so to make it easy for you.
,
http://pcsupport.about.com/od/findbyerrormessage/a/missinghaldll.htm
Thanks but yes I read that and did all of the mentioned actions...still not booting.
I sometimes got the HAL-error when i edited the partition-sizes, or installed dualboot OS's. Errors occurred especially when using windows disk management (no promblems since i use gparted).
In this cases the MBR (Master Boot Record) was corrupted. I never succeeded in repairing the MBR. I use True Image 2011 for imaging the OS-partition(s); it can also recover the MBR. Very nice feature, just put back the former MBR and ready to go again.
I don't know which imaging program you use, but this could maybe a solution.
Other possible direction:
If the OS was build on a HDD, then copied to SSD, you could try this.
- make backup image of OS
- using gparted, move the OS-partition 1Mb. So free space preceding the OS-partition will be 1MB.
- restore the OS
Maybe MBR can now find the first Windows-sector again.
link: http://lifehacker.com/5837543/how-to-migrate-to-a-solid+state-drive-without-reinstalling-windows
Another idea:
- make a new, fresh, nLite install (only 15 minutes). A new MBR is set up.
- reload minimized image
I would try this one first.
I hope this can be of any help, HAL-errors are ugly.
Good luck!!
'...In this cases the MBR (Master Boot Record) was corrupted. I never succeeded in repairing the MBR. I use True Image 2011 for imaging the OS-partition(s); it can also recover the MBR. Very nice feature, just put back the former MBR and ready to go again.
I don't know which imaging program you use, but this could maybe a solution...'
This certainly sounds plausible. I use Snapshot and have reloaded OS images many times with success. But after this hal error nothing seems to work. Can you say more about how you reloaded a MBR image? I take it that the MBR is on the HD ? If so why wouldn't just reloading an OS image do the same thing? Anyway this is my only lead. Thanks again.
according to specs Snapshot can restore MBR, so you could try this before doing a new install.
http://www.drivesnapshot.de/en/restdos.htm
MBR is the first physical sector of a harddisk, which functions as a content-table of the harddisk. It contents all info about the partitions on the disk(partition names, where they start end end, etc), and points to the first sector of each partition, so your computer knows where to go for startup instructions (boot.ini). So when it cant find this, due to corrupted MBR or corrupted first partition-sector, it won't boot. XP then reports very confusingly that hall.dll is missing.
Reloding an OS image would not reload the MBR, as this is not in the physical partition. You'll have to do this separately.
hope this helps, good luck!
Very helpful I believe this may be the answer. One more question is there another way to reconstruct the MBR on my ssd? I can certainly access it via a docking station on my desktop pc.
I re-read the Snapshot website on restoring an image and apparantly you can re build an MBR or whole sector by a simple right click in the restore process. I did this for the MBR and the whole sector separately and after both attempts I still got the Hal error during boot. I will try the n-lite version but am skeptical now.
it should be possible. I assume the dock is SATA.
But i think, if you have a small XP iso, it is faster and easier to do a fresh install & reload your own image.
I looked at snapshot: difficult dos commands and all. True Image it's just ticking a box. Easy.
I looked at snapshot: difficult dos commands and all.Not so. Snapshot has a perfectly good GUI. I've no idea whether it's still a better program than TrueImage but it certainly was when I first bought it some years ago. (We won't even talk about Norton's "Ghost" which, dammit, I paid good money for.)
True, I have in the past configured Snapshot to perform an automatic restore using a bootable floppy and a batch file but that was a facility for the non-technical, not a limitation of the program. That was way back in the days of the floppy disk and is no longer necessary as the program can now restore a boot partition.
More recently, I've seen me dozens of times get locked out of a cMP^2 machine and be back up and running in two or three minutes, ne'er a DOS command in sight.
D
Edits: 02/01/12 02/01/12
Hello Gents,
Just wondering if those who have "highly tweaked" operating systems would be kind enough to post their Pagefile size, threads and handles as displayed in task manager upon CMP startup. I think this would help myself and others determine how far we have to go to improve our systems.
Thanks.
Would give you that if I wasn't so 'highly tweaked' that I dont have that functionality any more. In the absence of that my tweaked windows xp system is 22 mb in size with config @ 0.52 mb in size. Not much left to do anything except/boot/play music.
Hi, yes that is what I was after. My stats are below:
Pagefile = 25.5MB (before loading an album)
Handles = 896
Threads = 92
Processes =11
Still hoping to see some more responses from other people.
I have noticed that every time I reduce handles and threads the sound gets better, without exception, but I am using a USB spdif converter.
I have noticed that every time I reduce handles and threads the sound gets better . . .
Not every time in my case but generally I agree. Meanwhile, the linked piece helped me to understand what Task Manager is reporting. Sort of.
I'm not so far down the road as Rick, Ted and others and can still launch Task Manager - see pix.
Note that winvnc.exe is a "headless" control utility. It is believed (though, obviously, unverifiable) that Isass.exe and svchost.exe do not normally run when suspended in cMP's setup dialogue but are invoked by Task Manager. Well, it's what I like to believe.
If I'm right, you're looking at five processes plus cMP & cPlay.
As I do not like the sound of (and have had issues with) AWE-enabled pre-loading of music data into RAM, I don't know what figures are returned for cPlay when so used but, as you can see, cPlay clearly pre-loads data into RAM somehow or other regardless of setting. (You can also see this by looking at the networking tab.)
The setup here is an Intel D510MO dual-core Atom mini-ITX board driving a USB-connected DAC without upsampling. Data from my main system, a slooooooow single-core Atom, are pretty similar but I can't show them as it no longer has TM access. When I could, the main difference was that CPU load was a bit higher, especially at the start of a track.
Hope that's what you were after,
Dave
Hi All,With help of Douwe01NL I finally also succeeded in getting an ‘Jolida’s nLite Disc’ and installed it.
The Installation process is also an unattended process.
So one can sit back and watch it happen just like with Mini XP.
I only had too install network drivers and the ESI-Juli@ drivers.
When I installed Foobar I got and error because dsound.dll was missing.
After I copied a dsound.dll-file from the Mini XP installation, Foobar installed without any problem.I have 3 (2,47 Gb) partitions on my little 8 Gb Kingstone SSD, so comparing SQ is really easy this way.
I Just reboot into another partition.Already with the first music tones coming from Jolida’s version it was very clear too me: Jolida’s standard version sounds MUCH better than the standard Mini XP install.
Much more micro details, better imaging and better sound quality.
Especially in the upper-bass and lower-mids regions SQ improved very much.Roughly checking the registry it turned out that Jolida’s version had more tweaking done geared towards sound quality than the Mini-XP version. Mini-XP still had 3 performance counters working. In Jolida’s version all were diabled. Checking the audio-codecs: Mini-XP still had them all in, Jolida’s version only had 2 audio codecs left (MP3 and WAV). Also unnecessary hardware like WAN Mini Ports, Direct Parallel, ect where not present.
In the registry I only had too ad: disable CD autorun and the disabling the wathdog bug checking.So all in all Jolaid’s version is much better optimized for (networked (internet)) audio playback than Mini XP is.
Which is immediately and very easy audible with the first music tones played (with Foobar).I also swapped in an Steinberg MR816x converter from a friend of mine.
And again: the SQ differences where immediately audible between Mini XP and Jolida’s version.
Also on the Steinberg MR816x, Jolida’s version has much better audio SQ.So very recommended this nLited XP version from Jolida !.
And if you are not able too create Jolida’s slimmed XP version yourself with nLite (like me) ask somebody too do it for you or help you with it.Mark
N.B. I’m not using the stock Juli@-digital part. In my setup the 3.3 volt and 5 Volt too the ESI julia digital part are separately powered from a 12 Volt linear bench PSU. The 12 volt DC is first filtered in with various L-C filters (ferrites, coils and caps) and than post regulated too 3.3 V and 5 V which than again is post-filtered before being fed into the Julia digital part .(Julia stock 3.3 regulator is unsoldered/removed and the 5 volt PCI power lanes from the MoBo are cut)) This tweak ‘quantum-leaped’ SQ in my setup.
cMP2 PC > ESI Juli@ > Van den Hul Optocoupler MkII > Lavry Black DA10 > XLR Mogami Gold > Klein & Hummel O300
Edits: 01/27/12
Would it be better if SP2 is slipstreamed to SP3 in Jolida's nLite install?
How do u think it would be Advantageous?? I guess not. Unless u have hardware which need sp3 support, u would only be populating the install more..!!
Junaid
I am asking because this file Multiprocessor LAST SESSION.INI I downloaded courtesy of Bibo01 shows the ENV and Target to be SP3.
Edits: 01/30/12
Mark,
I'm glad u liked the nLite. Yes, it does sound quite good straight out of the Box. When I first installed it, I felt absolutely no necessity to make tweaks or running my Batch files. I am still at the stock nLite state with just a few registry mods & the boot.ini switch which serge suggested. All in all, it sounds clean, & as I put it "more-balanced" throughout the music spectrum.
Certain cMp set-ups respond different to file deletions & reg tweaks. Which is why even after all these months, we still have a few who are happy with the steppe 4 to 4.5 segment.
NLite by default sounded very promising so I let it run & now am very carefully tweaking areas which show potential. With so many flavours of cMp, some with Linears, some with batteries & some with the stock smps supplies, one needs to custom-tailor their set-up according to the rest of their line-up. At the end of the day, it's all about getting things right for YOU..!!!
Junaid
Hi Jolida,
I am still at steppe 4.5. But over the last copule of months I had installed all the Fishy Secrets, the Shell32 slimming, the automatic Registry slimming, and boot.ini tweaks (got to thank Serge for all the wonderful finds). I had now starting to manually slimming the Registry-massive deletion according to a slimmed down sample that is currently flowing around. I had just started last night cutting down the SOFTWARE hive by 40% (still huge comparing to Serge's but it's a start). The effect is quite dramatic. With immediacy and blackness of background that I had never been able to achieve on a digital setup. One has to try it.
I will definitely try nLite on the new server.
KC
The less stuff there is at install the less stuff there is in the registry.
nLite does allow you to make a few setings that are effected by the registry like disabling SFC and pre-fetcher, balloon tips, etc..
If you want to take the next step I would be glad to send you a minimized config file that you can use to truly tweak the registry.
Most likely you can simply insert default and software since you are using JULI@. Are you using the H55 MB?
Getting rid of superfluous stuff in registry makes every bit as big a difference as a minimal nLite installation.
Hi Rick,
I checked some more registry settings and you are right.
Only some registry values are changed and other registry values are simply missing.
Never the less, the standard Jolida’s nLited XP, sounds really very good even without all these registry tweaks.
I’m not using an H55 MB.
I’m still using G41M chipset MoBo and dual-core processor on 775 socket.
But and a H61 Mobo and a dual core Celeron 1155 and DDR3 Dimms are on its way.
PCI slots are disappearing so I ordered a new MoBo as I expect in a year there will be no more new MoBo’s with PCI slots (which would mean I can not use my tweaked ESI Juli@ card any longer)
I would be happy and thankful if you are willing too send a registry config file too minimize my registry too see what that does for extra sound quality. Thank you for that.
Mark
cMP2 PC > ESI Juli@ > Van den Hul Optocoupler MkII > Lavry Black DA10 > XLR Mogami Gold > Klein & Hummel O300
Funny, though I can get the H55 one EVEN smaller than the H61 which is purely academic and cosmetic.
I really like the H61 and think it is better than the H55 just as Jack Wong reported.
Consider removing the chips per Mr. Wong. I had removed the chips from the H55 so I am able to compare the two fairly.
I hope we do not lose the ability to use JULI@ any time soon. No ohter card allows easy access to I2S and that is what I am committed to.
Obviously, you have got over any fear of the non-native PCI support. I sure have. No doubt it made a difference at one time but time has brought improvements.
Send me an email to my moniker and I will forward the H61 folder to you.
Something to consider, if you want your config folder to be as small as possible (and why not?) use /minint in the boot.ini file to eliminate all of the useless logs. Can make a difference of 100kB in the size which becomes significant when the folder goes below 500kB.
There are logs within each hive along with the logs you see when you open the folder.
You will be surprised how good JULI@ can sound when all of the superfluous junk is deleted in DEVICE CLASSES.
You still will need to use your head, especially with SYSTEM. Repeating myself you will likely be able to insert SOFTWARE and DEFAULT as is as long as you are willing to deal with no features at all, though no different than getting rid of EXPLORER. The computer becomes a dedicated transport when this is done.
I see it is not there and wonder if there is something less intrusive in win2k that will work in its place.
Without dxg.sys the system will boot and gets just to the point where the cMP screen would be displayed but just stays there, can't load the screen.
Any ideas?
This seems to work very well in my system.
It is really easy but it took me until now to simply check the box to do it.
While in regedit - select ControlSet2
right click PERMISSIONS
then click ADVANCED
the second unchecked box REPLACE PERMISSION ENTRIES ON ALL CHILD OBJECTS ...
check the box
then you are able to delete!
Make sure /minint is in the cMP drives boot.ini
Then compress your file again and you might come up with this if you have done other deletions
Thanks again worked like a charm.
Great find!
Dear cMP2 Forum
I´ve established a danish cMP2 Website (still not quite fished)
Have a look and leave a comment if you wish.
http://www.freewebsite-service.com/cmp2/forside.php
Kind regards
cMP2 Computer w/digital crossovers & FIR filters > Lynx Aurora 8 FireWire /192kHz all channels. 2x AcousticReality Ref. 202 & 2x AcousticReality Ref. 601´s ICEpower. Magnepan MG3.3R beechwood frames & custom stands. Miller chokes
Very nice Play-mate, congratulations!
.
cMP2 Computer w/digital crossovers & FIR filters > Lynx Aurora 8 FireWire /192kHz all channels. 2x AcousticReality Ref. 202 & 2x AcousticReality Ref. 601´s ICEpower. Magnepan MG3.3R beechwood frames & custom stands. Miller chokes
Grant and I were talking last week about all things cMP when the subject turned to how and why minimizing the available memory would be useful.
At the beginning cics wanted to use a small memory to minimize power usage. But when we do this I suspect the memory still uses the same amount of power.
Well, the only way to find out is to do it.
I took it out of my boot.ini and I must say it does make a difference. Taking it away has reduced noise not unlike removing the video and audio chips on my board.
Previously I had noticed a strange electronic noise when cPLAY was engaged. While the cMP screen was showing there was silence but as soon as I selected something to listen to, and especially with orchestral pieces where there are a few moments of silence before things get started this noise would begin. When the music began it was relegated to the background but of course it was still there.
With it removed so is the noise and I notice a general quieting in the noise floor. I was hearing some random soft pops - I listen to LPs so this was not bothering me but now that is gone.
I am upsampling at the maximum level. I am using a 512kB memory and I am using JULI@ so I wonder if what Serge is hearing with this is very different than what I am hearing. If he is not oversampling maybe this has a different effect with his system. We know he is not using JULI@
I am hearing a more robust sound. The sound is not as recessed which is an effect I liked but now I wonder if that was a sound effect instead of accuracy.
Serge, please explain what this is supposed to do.
I get the feeling the system must actually have to work harder with this in the system. Obviously the memory is and I do not think it a bad thing for the CPU to be given a load. I wonder if the CPU like a class AB amplifier might sound better working harder but there is more to it than that. It seems to me this is an obtrusive instruction that gets in the way of the machine doing what we want it to: simply process the information with as little interference as possible from unneeded files and registry entries that must be polled (I assume in my poetic/idiotic understanding of how computers work).
I tried taking out maxmem from my boot file and while I did not hear less noise I did hear many of the things you mentioned. Its not subtle. I hear soundstage more up front (but less depth). I hear a fuller midbass (but less deep bass). I hear more distinctness with maxmem back in. I think my reaction is the difference between our systems. My system (electrostatics and cone drive bass) is very inefficient (maybe 85 db) whereas your system (horns) is much more efficient. Bottomline I prefer maxmem back in.
But this is yet another way to balance one's system. Its just like, imo, reducing cpu frequency in bios...I tend to find the value that sounds dynamic/lively (not too low a freq) but not hard in the upper mids, highs (not to high in freq).
like I said I wonder if it is sound effects?
I did like the effect that maxmem gave, too, but the reduction of noise is something I think I will prefer over time.
There is the possibility that maxmem will work better after being re-inserted. I have noticed this many times with other tricks.
I still cannot figure out WHY this should do something worthwhile and I wonder if it has something to do with upsampling v. no upsampling?
You said: '...like I said I wonder if it is sound effects...'.I think its all about sound effects, that's what voicing is all about. Here's another item that has an air of absolute science about it but is really pragmatic in a sense ... 'lowering the voltage or cpu frequency' to lower rfi. That may be true (that it reduces rfi)but it is also just modifying the power supply. For example we're all told to lower the voltage and cpu speed as much as possible (to reduce rfi). But taken to an extreme if you lower both to zero and what do you get...no music right. Raise them both till you get something to work and get music...but it may be real weak music... soft, no strong bass, no high frequency extension. Raise it too much and you get strident music. So you try to set it just right...not too high not too low. There is much about audio design that is really just about voicing something just right. Yes new technology begets new electronic devices to produce music but they still have to be voiced. I believe since we have so many variables in computer audio to play with ... we should play with them to voice our system. Why not? I also believe those that have tried computer audio and dismiss it have just have not really tried to 'voice their system'. Look at it from an analogue perspective...you need to adjust cartridge vertical/horizontal tracking angles, stylus tracking force plus electrical loading (47k ohms or 10 ohms). If you don't do all that (plus a lot of other stuff) you don't optimize/tweak the system to get it...just right.
I bet if any who claim computer audio is not up to analogue or even good cd disk playback ever heard your cmp fully optimized as I know it is they would be instant believers in computer audio. I know I am.
Edits: 01/23/12 01/23/12
Adjusting a cartridge's load is not neccesarily about voicing though one could do that if they wanted. There are actual electrical parameters that need to be met due to the cartridge's characteristics.
All I can say is that I no longer am getting the blue screens and squeal and it is not as if the sound I am hearing is not attractive.
It is a more robust sound.
When someone loads a MC cartridge with too low resistance one will get a bassy bloated sound and lose output voltage; one can like it but it is not (really) what the designer intended.
One can take the voicing thing to the point where fidelity to the sound that was actually captured has little to do with what is actually ultimately heard. Sometimes this is good for a really bad recording but usually will introduce distortions (euphonic to some) that will mask what is there. This can be advantageous on some recordings but usually makes a mess of many others. I guess I want as much information as I can get and that is my bias. Neutrality is my goal. Sometimes you have to live with the warts.
I would like to know the reason for /maxmem since there is no reduction in power used by the memory when implementing this so the argument about reducing RFI does not apply here.
I think it is something that worked before and has been made superfluous with later developments.
Just my opinion.
For me the same experience: without the maxmem it sounds just a bit more musical, more body, more detail. Very subtle though.
I implemented it in a former setup (H55M), and continued this in the current atom d525 setup. So for me the lesson is every new setup needs its own check of the tweaks.
As for upsampling: i tested:
A. songs 44.1 upsampled to 88.2 by cmp
B. same songs upsampled 'by hand' and stored in a 88.2-file, no upsampling by cmp.
B sounds better, little more music, stage, detail
So in this case: the less work for cpu, the better
Other than the massive additional storage needed I do not doubt it is better to do this before but I have been concerned that there is something about SRC that is not available from, say, dBPOWERAMP which will upsample. I am sure there are others that I am not aware of.
My DAC upsamples everything to 192 and seems to me SRC/cMP does a better job so I leave cPLAY with 192/highest resolution.
What are you using for upsampling? I would like to give it a try.
Thanks,
For upsampling i use SoX (like cPlay and Foobar). Ik like it more than SRC. These are the settings i use in batch-resampling:
mkdir H:\MUZIEK_ORIGINEEL_3288\A_sox
FOR %%A IN (%*) DO sox %%A -b 32 -t wavpcm "H:\MUZIEK_ORIGINEEL_3288\A_sox/%%~nxA" vol 1.0 rate -v -s 88200 stat -v
pause
I don't recall the meaning of all settings (did it years ago).
I think 'stat' gives report which i use for checking on clipping. If clipping occurs, i set volume to 'vol 0.8' and retry.
32 stands for 32bit lots of 'zero's', huge files but better sound
My DAC upsamples to 176.4. Thats why i let cPlay/myself upsample to 88.2 In my setup this gives the best results.
Douwe
When I did upsample I also preferred sox. Now I listen to 44k sr. But the idea to restructure the file to 32 bits is very clever. I should try that.
I did not realize/remember that.
I think my problems with it are entirely due to upsampling.
I used to halve the duties in a similar way but eventually settled on full on cMP.
On cPLAY I seem to prefer SRC to SOX though I have no way to describe the differences I hear and wonder if I really DO hear a difference.
But upsampling the file does interest me.
Thanks for bringing it up!
I have been away, and missed the registry slimming discussion. I have been looking for an example of what is possible to delete from the registry. Looks like Mikhalov had posted an example that is no longer available. Is there any guidance that anyone can provide to help me understand what I can delete before experiencing the BSOD?
thanks much
Walter
See link.
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2 Ru, GA-H55M-UD2H, i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S FAT32, ESI Juli@, full linear PSU, NAS - WD My Book Live
My SOFTWARE and DEFAULT are not much different from what you posted. Those screenshots showed the way. I remember first looking at them and thinking there is no way that will work!
SYSTEM being the most individual required lots of trial and error since your system retained network stuff and I had to find my own way there BUT STILL using your model was and is the key to true registry minimizing.
No matter how hard I try I cannot remove all of ControlSet2.
At this point I find screenshots to not give a clear enough picture of all that is going on plus it so much easier to load both hives and compare.
I would be glad to send you what I have done if you would send me an email.
I do not doubt at this point you are probably already at the same place.
Rick, I longer read (listen) than write (say) :).
I sent you email but it was blocked (it's unfortunately happens sometimes :(). Send me please the hives to my email via the Forum. Thanks.
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2 Ru, GA-H55M-UD2H, i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S FAT32, ESI Juli@, full linear PSU, NAS - WD My Book Live
plus I do not think I can send attachments through there, anyway.
Did you use my moniker email address? I have no blocks on it at all.
Please try again.
Thank you for the link. I can't make out the writing, but looks like Rick will send a copy of his which will fix me up. Thanks again.
I assume you would want the H55 version. I have one for the H61 board, also.
If you are using JULI@ and H55 you can substitute DEFAULT and SOFTWARE.
SYSTEM might work if your HDD's are basically the same arrangement other wise you need to load your and "mine" into REGEDIT and toggle back and forth to see what to do.
I recommend adding /minint to the boot.ini while doing this.
After you are done use the registry compactor.
Worked like a charm. Thank you!
Config file will be put to good use.
Hi
I'm using cPlay for the first time. Got this error message. What should I do?
ASIO start aborted. Only 32 bit ASIO format supported. DSP does NOT support current ASIO output format (only 32 bit supported).
Only 32 bit ASIO format supported.
There's a Settings button in the bottom LH corner of the main cPlay window. Check that your output device is set for (supports) 32-bit ASIO. What the settings dialogue displays varies with the driver but it should be easy enough to work out.
What device are you using? It's just possible that it will not work with cPlay though most do.
I can't access the ASIO settings button. It's shaded. Do I need to download something?
My soundcard is an ASUS Xonar Essence STX and is supposed to be ASIO compatible
On the right hand bottom corner right click on the Xonar icon to change ASIO settings. Latest Asus software supports 32bits.Make sure you have the latest drivers.
Edits: 01/21/12
Thanks for the tip
I downloaded one of the latest drivers 5.12.8.1792, but when I open the ASIO settings, I cannot choose the bits. It is fixed at 16 bits and the button is shaded.
I can't open the newest driver from the website as it is a .rar file. How do I open this?
OK, I found the 32 bit tab and changed the setting.
Now I got "ASIO start failed"
Install Foobar and its ASIO extension.
Open ASUS ASIO driver, change resolution to 32 bit.
Now you can open cPlay and it will work.
Here are threads on the problems encountered on the Xonar SC.
I don't know whether the exchanges there would help, but no harm in checking the registry to see if the driver is there on your computer.
Hello,I've decided to revive what I've considered a dead project by making some changes to CPlaylist Editor that will hopefully make it more useful to CMP setups.
The first change is that it no longer requires Java to be installed on the PC. I've compiled the application into a native executable that can run without the Java virtual machine. Along with this, the application comes as a standard Windows setup package, with an uninstall feature as well.
The second change was suggested by Ken Stuart. When set, CPlaylist Editor will quit after launching CPlay, so no other programs will be resident in memory. Obviously, it would be a bad idea to set this option of you are using CPlaylist Editor to control CPlay over a network.
I am hoping someone could test this program on a CMP box, as my CMP partition is currently hosed and randomly freezes whenever I boot into it.
I am in the process of updating the help file, but the program can be found at the link below.
Alan
Edits: 01/20/12 02/17/12
Hi Alan,
Your link is not working for me.
Hi,
I updated the link in the original post. Sorry about that.
Alan
Very nice idea,
I'm quite busy but I could try to make some test
Nicolas
Hello, everybody. Knowing that many are really impatient, I post now this tweak, hoping You, inmates, will find the way to finish it. (Please take note that I am posting only things I managed to achieve).
It started with Mario's brilliant idea of using reshacker to remove icons and stuff. He posted a link to a forum where folks were discussing it, and one of them suggested using in win XP a shell32.dll file from win2k. Mario wrote to me he wasn't ready to create such a monster, but I was.
Shell32.dll is the biggest file. I have managed to subsyitute it with w2k pro version (no sp). after appr. 7 attempts. It's current size iz 1,17 mb. I DON'T KNOW HOW I managed it as I started adding more and more files from w2k. It still remains a mystery unsolved by me and Rick. I was waiting if he would suggest some final version of the registry slimming, and meanwhile Jack Wong and afterwards Riboge and others started working with XP Home edition. As not all the files were fitting from w2k, I tried to substitute others from xpHome.
The following is the list of files from both OSs that can be substituted. The resulting windows folder (with registry deletions performed according to Rick) and with registry compacted according to Ted is 19,6 MB. Though I must confess, it doesn't work, but because I have deleted too much from the registry. FILE SUBSTITUTIONS WORK.
Please note that the files are to be substituted in the order they are suggested (THE CAPITALS are for w2k others are for xpHome)
SHLWAPI.dll
WINSPOOL.DRV
USERENV.DLL
COMCTL32.dll
COMDLG32.32
SHELL32.DLL
WLDAP32.DLL
lsass.exe only together with lsasrv.dll
samlib.dll only together with samsrv.dll
rpcrt4.dll Only after two sam files
services.exe
svchost.exe
FRAMEBUFF.DLL
SCESRV.DLL
advapi32.dll
ole32.dll - you know how I "love" OLE
netapi32.dll
ntdsapi.dll
MPR.DLL
MSASN1.DLL
WINSTA.DLL
WS2_32.DLL
sxs.dll
umpnpmgr.dll
crypt.dll
secur32.dll
msvcp60.dll
NTOSKRNL.EXE (if you retain ntkrnlpa) This file can be substituted by w2k no sp version but rename it and add in the beginning of boot.ini string /KERNEL= yourname.exe. add also /nopae and /minint. (a must)
Best recommended to use xp-pro kernel - it really sounds better, but with ntkrnlpa.exe removed.
That's all.
Serge
P.S. So, if anyone can find the way to substitute shell32 from w2k, will really help others. I can't remember How I did it, But I did it.
Don't touch user32.dll, win32k.sys.
Hi Serge,
Is there a way too check which kernel is actually loaded at startup?
(Situation A)
It is my impression that if one retains the ntkrnlpa, the boot.ini string “/KERNEL= yournamekernel.exe “ is ignored and the ntkrnlpa.exe is still loaded (despite the boot.ini string).
(Situation B)
If I rename the exe-file extensions from botch ntkernelpa.exe and ntoskernel.exe too *.bac and put the ntoskernel.exe (w2k-version) in place, I get an error message at startup there are DLL’s missing.
If in situation A the ntoskernel.exe from w2k was loaded through use of the boot.ini string, than I should have gotten that same error.
But I didn’t got that error.
So an alternative kernel is loaded in situation A.
Probably the ntkrnlpa.exe because it is retained.
Mark
cMP2 PC > ESI Juli@ > Van den Hul Optocoupler MkII > Lavry Black DA10 > XLR Mogami Gold > Klein & Hummel O300
to load yourname with ntoskrnl.exe present or with ntkrnlpa present. In both cases you need at least one native kernel to be present. Nt loader checks to see that you have at least one. To, say load w2k kernel with ntkrnlpa and to be sure it is w2k kernel, that is at work, specify /KERNEL=yourname.exe in the beginning of the string AND /nopae closer to the end - this comand FORCES nt to start with either ntoskrnl.exe or with alternative one. If you retain ntoskernl.exe and delete ntkrnlpa, then use /kernel=yourname.exe, but at any time ONE NATIVE should be present, thus any time with alternative kernel there are TWO kernel a native one and alt. one, But to use and load in memory ONLY ONE kernel, use your native ntoskernel.exe from xp. This way, though it's bigger you assure the best memory-cpu interaction and absence of the secondary kernel in memory. It seems to me it sounds more robust, vivid but without harshness, while with w2k and ntoskernel(native) together the sound is smmother but slightly dull.
Serge.
P.S. Hope it's not confusing.
I tried as you suggest: "If you retain ntoskernl.exe and delete ntkrnlpa, then use /kernel=yourname.exe" taking it to mean I rename ntoskrnl.exe, say to ntriboge.exe, then remove ntkrnlpa and in boot.ini up front put /KERNEL=ntriboge.exe. Is this correct?, because it didn't work causing cycling of startup just like simply removing ntkrnlpa. My computer may somehow be different because I was also able to substitute win2k ntoskrnl without /kernel and renaming.
I also tried leaving in /KERNEL=ntriboge.exe while not renaming ntoskrnl.exe just to see (since your instruction is a little ambiguous) but no surprise that did no better.
Any suggestions?
rename xp native ntoskernl. It should remein unchanged. Once you do it You get a black screen on reboot and you will have to remove the file and paste it back, also remove all logs that were created by registry hive.
in othe words you have 4 kernel options
1, two native kernels
2. one native kernel. but only ntoskrnl.exe and /nopae switch
3. w2k + ntkrnlpa + /nopae switch
4. w2k + ntoskrnl
Leave at least one native kernel unchanged
5. option to use only native ntkrnlpa can be possible but you need 4 gb physical ram, /PAE and /3GB switches.
Hope, this is it.
Serge.
When I try to do your 2., as I said, I do not get a black screen but a restart before anything shows up over and over until hard shutdown. This is with renamed ntoskrnl.exe and no ntkrnlpa and with /nopae. Nonetheless I will go on to cut and then past ntoskrnl.exe and remove all hive logs and see what happens.
That it should be THE NATIVE kernel. Don't give its name to alternative one. ntoskrnl.exe should ALWAYS be the real native one.
Serge.
As for e-mail, some things were made there that make a lot of cics' optimisations done or unnecessary, but the services were untouched, and some othet things, so You better treat it as a minixp. It should install easily and without oobe, serial, regs and so on. Video was already set, most things removed. When installing weiss or other things, be patient, please, it takes looonger. Awe and minlogon should work like a charm as they were described. WFP was already removed. There is also a session.ini file that you may see what was done, if you have nlite, just load the ini files from the iso. It's very easy to navigate.
Serge.
Without ntkrnlpa I have, of course, only been trying with the native ntoskrnl. I tried it after cut and paste and removal of hive logs both as "ntoskrnl.exe" and as "ntriboge.exe" with /KERNEL=ntriboge.exe but neither got anything but recycling restart as before.
The nlite xp allowed Weiss install normally as well as Intel ethernet install and that for cMP/cPlay, but it did not recognize the net command in command window despite network running.
I thought you had stated previously that ntkrnlpa was insufficient by itself. If that is so "to be sure it is w2k kernel, that is at work, specify /KERNEL=yourname.exe in the beginning of the string" would seem unnecessary if one has removed native xp ntoskrnl and is using win2k one with ntkrnlpa.
If that is not the case, then why not use just ntkrnlpa on purpose?
hfavandepas, there IS a reason: you might prefer the sq even though steppe doesn't. I like it that way but have not so far gotten a successful try at using the original kernel alone to see whether or not I find it better myself.
nt
cMP2 PC > ESI Juli@ > Van den Hul Optocoupler MkII > Lavry Black DA10 > XLR Mogami Gold > Klein & Hummel O300
...all I can say is that your voicing ability to tweak an os for audio purposes is phenomenal. All the tradeoffs between too crisp or too soft I mentioned in earlier posts are gone. This is the best. I am only somewhat sentimental in that this sadly may be the end of the journey. But oh what a finish you have brought us to dear friend Serge. You are a master tweaker Steppe: very, very well done. As I type I am listening to Diane Krall...your kitty would have been proud of you and proud of what DK sounds like right now.
I know I know you are working on something else now...the next tweak. But if you want to take a break this is the right time. I just wish Cics could hear the heights to which you have brought Cplay/cmp.
all this, that I am listening to her "Girl In The Other Room" too, right now. And a new kitty is right here. She is still young but also likes music. But her faves are Can from Germany and Steely Dan. I was listening to the French Military marches yesterday you've mentioned before. A very educational thing to listen to, soundwise. As for new tweaks, yeah, I KNOW what else can be done, but it NEEDS a serious programmer. I'd like to rewrite cmp as a Rootkit, only in the best sense of the world. Also, for Windows xp to work and load rootkit at boottime, it doesn't need anything but 5-6 files. no lsass, no svchost, no nothin'
Serge.
P.S. I am happy, that you find the sound good.
P.P.S. I now know what the writers mean when they write it's a sad thing to have lost a good enemy.
Why, But THERE ARE drivers yet to try to substitute, like videoport.sys and dxg.sys...
...you guys are going to hate this but the sound tightened up significantly but now may be too zippy. I'm thinking I may have to do all the file substitutions now or go back to my original registry (for session manager /memory). Funny thing this computer audio stuff.
Rick was your session manager /memory registry like Segre's or like mine? I checked my desktop which is a win xp home and its registry was like mine and not Serge's. So I'm not sure which is right.
.
No remote viewing here ...
I showed in this post. Look at bottom 4 or 5.
of substitutions should com from the following files:
shlwapi
ole32
shell32
framebuff.dll
lsass-lsasrv
scesrv
sxs
authz
and especially from services.exe, svchost.exe, secur32, msvcp60 and crypt.dll
So, It's not the size that matters! But presence in memory and service strings/threads or whatever they call them that are EXPORTED by the files.
My registry session manager snapshots were initially for Rick only, but if You find them helpfull, they are for everybody.
Also, Rick, If you use 3rd stage extreme bios settings, be sure I sometimes also get a bsod while listening and the soundcard starts stutter-hammer like a small pneumatic hammer.
Serge.
Edits: 01/21/12
everything to do with BIOS.
I got a new message today: DRIVER_IRQL_NOT_LESS_OR_EQUAL.
Now whether that is due to my registry experiments or BIOS will remain a mystery.
No stutter, though I used to get that from using batch files which is when I lost interest in them, but a squeal.
Starts right back up.
What is neat about the H61 board is that it will start with "third stage" memory settings. No need for the intermediate step.
Not driving me crazy - it can go hours before it does it.
I changed my session manager memory settings to Steppes. It booted ok will listwen and report my on my sonics. Rick that is a great offer but I dont even have that mobo. Still have the old G31-s2el for the 45 nm hardware.
Here is my registry. Look at bottom 4, some not even shown on your's Serge.
Edits: 01/21/12
Access to the network was denied with these substituted and returned when I reverted.
I got shlwapi.dll substitution to work with version supplied by steppe which was slightly smaller but also from win 2k pro source.
Who knows what the problem was?
This time no complaints at all.
Substituted all in three operations. No problems at all.
I do think it sounds better. I was becoming skeptical due to my frustration with the win2k shell32 refusing to work. But now I can see what you are talking about. Yet another increase in clarity and good tone.
Found the reason I could not do some of the registry deletions - I had had /minint in the boot.ini before and took it out when you said I should. I put it back in since it was the only difference in the two set-ups (between h55 and h61) and deletions stay deleted.
My WINDOWS folder is now 23,076,605 bytes. Config - 416kB.
I still have some RESOURCE HACKING to do but I see no way I will get the folder below 20mB. A few more items to be removed in the config, too.
I do, and had been, getting a blue screen that can show up at any time. It could be fifteen minutes or hours, no rhyme or reason. It could be (still) because of my BIOS settings or some other setting, Says something about paging area - I will write it down next time. Not a big problem. Starts right back up.
I did 4 or 5 files from w2k and the result was very good sonically. Then with the Steppe files I basically substituted only those files which showed a large (> 50kb) size reduction for everything in my system32 folder. I did it in 3 steps. Only major files I did not do was shell32 and winspool. The sonics progressively got softer (not in a bad way), almost as if a tube stage was introduced into my system. The negatives in sonics: deep bass lost a little punch and extension, soundstage and depth shrunk a little. The positives in sonics: just like in tubes the mids and upper mids and highs became mellower/sweeter (more sweet?). So it depends where you are in the spectrum of on the one side of tight/crisp all the way through mellow/sweet on the other end of the spectrum that in my opinion will determine what and how much of these substitutions you might like. I found a good balance for me/my system. I probably did all but maybe 4 or 5 files including those mentioned above. All in all a very worthwhile tweak...thank you Serge!
The first was session manager - memory. This one is session manager itself.
Serge
I will try these registry settings. I have a very different session manager - memory. I forgot to mention that I got no BSOD's as Rick did. Rick did you get these while booting or while listening to music?
accompanied by a squeal!
Booting is never a problem.
I think I just finished getting the last of what I know I can eliminate from the registry (thanks to Mihaylov on both counts due to /minint and his original road map) and after using the compressor my config is 398 kB.
Now to see if the problem re-occurs. It is annoying but not alarming, other than the squeal! Starts right back up after an incident.
By the way, WHERE is Mihaylov?
As far as sonics I suspect my complete happiness (for the moment until the ear/brain wants more of something to keep the charade of music reproduction interesting) is my use of horn loudspeakers. Softening is never a problem with horns!
There may be softening but unlike some tube softening there is more information making its way through the electronic maze.
Also, at this point I think I can safely say with no qualms that the H61 board Jack Wong recommended is truly superior to the H55. For those wanting to get a little closer this should be procured.
Ted if you want the chips removed I would gladly do it.
I increased the IO thing to 20000 and there was one more that was different on the second suggestion.
Now to give it a try!
Thanks!
nt
Rick! When I was doing deletions according to your hives I have noticed a setting, I thought was strange - that is clear page cache or smth on shutdown. As cmp is using rpc, the system may have too little time to clear this cache and it is trying to restore the cache settings (not lost completely in the previous session) prior to booting. I think it's the legacy of nt 3,5 or 4. I remember from my printshop business times, if ever you abort a print job and force reboot, nt4 would restart with the spooler full and retrying a print job. Try to find this setting and remove (set to 0) in the system hive session manager, IIRC.
Serge.
It's here, I hope
ClearPageFileAtShutdown
It is in MEMORY MANAGEMENT.
My experience: I did win2k substitutions down to WLDAP32.DLL on your list. Shlwapi.dll caused a halt just before cMP saying cicsmemory player could not find dynamic links in shlwapi.dll.
Shell32.dll reshacked down to about 1.6MB from 2.2 loaded fine the first time.
With xp kernel and no ntkrnlpa.dll bootup kept auto-restarting from a very early point so was literally a non-starter. So I am back to win2k kernel with ntkrnlpa.dll.
This is all with a mini-xp installation, which is not getting the attention it deserves. It sounds the best I've heard and is much easier to setup and get great without slimming.
Hello, Have you substituted shlwapi.dll after all? Also, very impostant, have you substituted shell32 that was hacked in another folder? It seems, I have done smth. similiar, then hacked it to the end. What is now left inside your shell32? At what point in hacking shell32 other inmates should stop to insert it in place wo a problem?
Please MORE info, as I cannot recreate it myself, too.
About mini-XP... But You are wrong. It deserves indeed and is getting all due attention. It's nothing other than just somebody's well thought through universal NLITE installation. Mine is 98 mb and assures awe and minlogon and everything. The difference is that you are using a side product instead of using what is suggested here and in plenty of flavours specifically suited for our purpose. If You want, I will send you my ini files. Download nlite. Making an iso with it is MUCH easier than substituting files, kernel, reshacking them.
Also, yes such installations sound very good from the start, read Rick's and Junaid's posts. But after that we tweak such installations TO THE LIMIT, and this is another story soundwise. I only suggest that you delete all unnecessary files - see my pic. You will surelly hear the great difference. But, please, first give us more details on shell32,sir.
Serge.
I am glad that You find these tweaks usefull.
Edits: 01/20/12
There is not much to tell about shell32.dll. With reshacker I deleted all the animations and bmps and many of the icons. I refrained from taking on the dialogues and menus though clearly going a good bit further wouldn't be difficult. I did this to an shell32.dll extracted from a win2k .iso on my desktop computer then substituted the result into my mini-xp install on the laptop I use in lieu of a cMP box.
I passed on substituting shlwapi.dll since I don't see how given what I encountered. I can't try reinstalling cMP to work with the substituted shlwapi because I can't boot up with it.
I have read just about every post in the cPlay thread for many months so am familiar with all you write of. I have tried almost all of the tweaks and all your and Mihaylov's and Junaid's steps. The only 'further' step I haven't taken is the xplite very minimal install or rick's latest radical registry stripping. I had trouble over and over with attempts to reduce the registry before so hesitate at this. I also feel that at some point practicality comes in for me. If I can get 95% of the improvement and still retain some useful functions that are all foregone for the final 5%, which is where I thing I am now, I start questioning the worth of it beyond the 'science' of it, that is, the exploration and defining the limits that you and a few others have productively pursued to the great benefit of all of us. I care about every last bit of sound quality but not only that and not about having the least number of MB to my install per se.
I guess I would be willing to try the install you describe from things you send me, if you do, to see if perhaps it brings more than 5%. riboge@gmail.com
With shlwapi.dll and then usernv.dll I got a blue screen on startup saying the header checksum didn't match the computed checksum. I could not load these into reshacker either. Both were said by it to not be "win32 executable files". This also happened trying to reshack the ntoskrnl.exe that I have running already in my mini-cp install. What am I doing wrong and/or what can I do further to make substitutions? I am getting the win2k files by extraction from a win2k pro no sp .iso.
After you extract a .dll or whatever from the win2k .iso you then have to decompress it with winrar or the like to get the unpacked form. This reshacker will load and, I presume til I try it, will work as a substitute. This is puzzling a bit since I used for substitution ntoskrnl.exe I extracted previously without this unpacking step.
so we can be sure we have found the same files.
Then I think: you have have removed VERSION with RESOURCE HACKER!!!
I am thinking with file substitutions we might best keep VERSION attached to the files!!!!
I noticed you did not send frambuf and the .iso I found did not contain it OR it has a different name in there. All of the other files are quite obvious so this is a bit of a mystery.
I (thought I) went through all of the FOLDERS thinking it might be hiding in one of those.
Send it, please.
Serge,
If I sent you the folder with GRANT in the name that was as far as I went with it - I think it is CLOSE to being minimized for the H55 board. There might be one or two more things that could be found. It would takes hours of trial and error to find them. Since then I have been working on the H61's registry which seems more guarded as to what it will allow to be deleted.
Ted found a few additional items he could delete with his G31 system. I feel sure he will post what he found.
As Mihaylov cautions - each system is different (unless it is EXACTLY the same) so some might be able to do things others cannot.
The DEFAULT and SOFTWARE files I sent you seem to be universal for JULI@ systems - they drop right in. But as you found LYNX is another story. I should have mentioned that DEFAULT is absolutely universal if one is willing to accept what it does, which is not much different than what happens when one deletes EXPLORER. You no longer have the screen color you selected - seems MICROSOFT BLUE is the default color, which does not other me. I think it dubious that the power the monitor requires to produce a color has any effect on the sound quality.
Each SYSTEM is unique because of different HDD/SDD's even if the motherboard is the same.
I have been thinking that it would be interesting to compare a substituted win2k shell32 with an xpPRO shell32 with a minimized SOFTWARE registry. Yes, the unneeded junk is stll sitting in memory (which I think at this point is not terribly concerning) but most of what is DOES has been negated by the registry no longer allowing the functions to be accessed?
I will try again (and again) to get the win2k shell32 to work but I have a feeling it is likely cosmetic. I, too, gain some pleasure in seeing how small the WINDOWS folder can become but one wonders if at this point the real work has been done and we are removing straws from the camel's already broken back.
hello, everybody! This short set of tweaks is independent, but is best used in conjunction with Fspart4, if Fs 4 is stable in your machine.
So,
1. HKEY_LOCAL_MACHINE\SYSTEM\ControlSet001\Control\Session Manager\SubSystems delete: Optional and Posix
2. HKEY_LOCAL_MACHINE\SYSTEM\ControlSet001\Control\Session Manager
find EnableMCA and EnableMCE, set values for both to 0
Serge.
.
FS parts 4 and 5 are only pecursors to it. The Big one is dealing with the basics of the OS we use, and it's really Big. It's perfectly stable now exept in one thing, that I am striving to settle. BTW, the reaction of Grant to only a single part of it shows, what it may bring, or may not...
Serge.
P.S. To Grant: yes, sir, if You read this, I use the file fom before the sp1. That is pro version with no service pack. So the tweak works, and it's audible. But I must Thank You for sharpenning my attention to the whole thing. I now know what IS /nopae in reality and how it works for us. The file version is 5.00.2195.1 from December 7 1999.
Edits: 01/18/12
I gather the tweak 4 page lock limit change won't work for use with later versions of xp with a later ntoskrnl.exe, so can we substitute the earlier version you have and can you supply it or suggest where we could obtain it?
Edits: 01/18/12
...take your time.
Hi Serge,Many (but not all !) of the ‘Fishy secrets’ registry tweaks and cMP registry tweaks I also found in the Mini-XP Registry Tweaks as done by eXPerience.
But.. the list of registry tweaks done by eXperience in his Mini-XP is much longer.
It’s an endless list of tweaksYou might want too take a look at that list of Mini-XP registry tweaks too check if there are extra tweaks inside which can be added too your list ‘Fishy secrets’ tweaks.
As also riboge stated in his post (mini-xp) “is really sounding wonderful, better I think with just tweaks and fishies than xp home with all tweaks, fishies and steppes too.”
Also in my setup Mini-XP is sounding SO MUCH BETTER (no hyperbole !) than a standard tweaked cMP XP installation + a lot of fishies and Steppes.
Low level details are really stunning.
Imaging is also much better.
Same for mids and highs.So have a look at the list of Mini-XP registry tweaks, you might be able to recognize more ‘hidden treasures’ useful as extra ‘Fishy secrets’ tweaks.
Or even better: install Mini-xp and have a listen yourself.Mark
List of Mini-XP Registry Tweaks as done by eXPerience.
see http://www.mediafire.com/?xtga9i917c96vq8
cMP2 PC on Mini-XP by eXPerience > ESI Juli@ > Van den Hul Optocoupler MkII > Lavry Black DA10 > XLR Mogami Gold > Klein & Hummel O300
Edits: 01/17/12 01/17/12
...very nice Serge. Thank you! When I went to a < 1 mb registry everything got very dynamic but a tad dry and etched. This tweak fixes that very nicely..
I'm a user of cMP / cPLAY for quite a long time, this is not a perfect solution, but to me it's the best in sound quality.
Reading "the absolute sound" from January 2012, page 40 to 53, there is a very interesting article on computer audio. It's talking about many software, from cPLAY to JRMC, foobar, etc and a new comer for me name JPLAY. From what there say, it's even better than Cplay !!!
There not enough detail on the configuration of the PC to have a good idea of what they did. But there is enough to try to benchmark both JPLAY and CPLAY on the same computer with all the tweaks from here. Sadly, I will not have the time to do a serious benchmark, but I will try to report in the future about this new software and cplay.
Does someone already test it ?
Nicolas
If you search this Forum for "jplay" you will find previous posts about the program.
You can download a free "trial" version of jplay which allows you to verify the sound quality, but not use it for enjoyable listening because it mutes the sound every so often (clever).
I tried it and I found it to be close to cplay, but not quite as "analog" sounding as cplay. In other words, cplay is still better, and is free.
Note:
- I am running Vista 32-bit (so whoever said in this thread that jplay is 64-bit-OS-only is wrong, I think they are confusing that with the fact that it IS Vista/Win7-only).
- My PC is a Home Theater PC for both video and audio, which is why I use Vista. So, prior to discovering cPlay, I had already turned off many Windows services. But I still have more things running than most cmp/cplay uers.
- I use an asynchronous USB DAC which avoids a lot of the jitter that the cmp optimizations are designed to remove. (Many people have found this asynchronous link to make the biggest difference in computer audio; the V-Link product does the same thing for SP/DIF DACs)
- I do not use cMP shell, since cMP mode freezes in the Vista Welcome Screen (and all the prior posters who successfully used cMP mode in Vista no longer read this Forum). But, jplay has a "hibernate mode" that I also did not use, so that is equal. :)
- jplay does sound better than foobar2000. This is mostly due to the fact that the jplay authors have previously used cmp & cplay, and because the foobar creator is a "sound quality denier" (and in fact, posting about sound quality differences on the foobar forum can get you banned). So, for example, you cannot use foobar2000 with an asynchronous USB DAC, because he refuses to allow a buffer latency of less than 50ms.
- I use the free player MusicBee for everything other than flac and wave, because the developer is friendly and willing to make changes (it has an .ini file parameter for the butter latency which you can set to zero - hurrah!)
It was my mistake about 64 bit OS, certainly I meant vista/7. I have read everything that I managed to find about jplay, including a too warm review from Poland, with nice photos and everything. The hibernate-and make-the-processor-run-at-extreme-at-the-same-time idea seemed interesting.
Serge.
I think that this is not the place for such topics. This topic about cPlay! Better start a separate topic.
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2 Ru, GA-H55M-UD2H, i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S FAT32, ESI Juli@, full linear PSU, NAS - WD My Book Live
Nicolas! We cannot test jplay on the same machine (I mean os), as jplay is 64 bits only. Marcin Ostapovich and Joseph Piri - the authors of jplay have taken a lot of ideas from foobar forum and from here, I think, and probably their own to produce a player. Even the name is nothing short of imitation. I have VERY STRONG doubts that jplay can even get close to the sound, one can get from cplay WITH FULL TWEAKING, so why waste 100 euro? My doubts are based on the fact that jplay is a commercial product (and even that being evolving), that has to work in a normal unabridged OS environment. While, probably, a good product, it doesn't cost its money, I am sure, as no simple player does cost such a sum.
Now, cplay is more than a stellar and splendid player. Its a hobby, a rite of passage and assention.
BTW, Can you imagine anyone paying 100 euros for the player and then starting to tweak his windows7 to reduce it to 25mb?
Serge.
I agree completely.
I became aware of this a few months ago and thought the claims made were bordering on ridiculous (considering a full WINDOWS installation) and the imitation did not have any of the graces of flattery.
There is nothing wrong with someone reporting their experience especially in comparison to cMP. Just as it is interesting to know how our evolvning cMP systems compare to CD players. But there is no question this Forum cannot become a JPplay forum, also.
I did not get the feeling that n_vincent had any intentions of that happening by his hand.
=Just as it is interesting to know how our evolvning cMP systems compare to CD players.=
System: a source - cMP2 classical optimized (one year ago) w/o network with Lynx Studio L22 upgraded and with full linear PSU (see my website, black Zalman); the amplifier - ART AUDIO LAB m25, acoustics - Klipsch La Scala 2. A reference source - Accuphase DP-75V (http://www.stereophile.com/cdplayers/249/).
At first the various jazz and classical material was listened to get used to system sound, first of all Klipsch La Scala 2 (personally I listened the horn for the first time). The record of Miles Davis and Gil Evans "Sketches Of Spain" and the record of Anna-Sofi Mutter "Carmen-Fantasie" were listened attentively. The system sound on these records made strong impression to me - live, natural sound. Anna-Sofi stood in the room and played for us. A reference source and cMP2 were compared on the record of Bucky Pizzarelli "Swing Live" (Chesky).
Summary of my friend Michael Nikolaev:
"Both compared sources has similar sound in sense of a musical involvement. But Accuphase gave more gathered elastic bass, in my opinion, more correct. May be Accuphase has the slightly higher resolution but not considerably. cMP2 has more analog sound and sometimes it seemed that sound of cMP2 is more preferable for music listening. It is possible to tell that the source cMP2 was gained worthy, involving the listener in music. And as a whole the system can be described as a true Hi-End".
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2 Ru, GA-H55M-UD2H, i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S FAT32, ESI Juli@, full linear PSU, NAS - WD My Book Live
Is that you or Michael in the picture?
It is my friend Serge, owner of system. I and Michael were on visit to him. Michael is audiophile with great experience.
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2 Ru, GA-H55M-UD2H, i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S FAT32, ESI Juli@, full linear PSU, NAS - WD My Book Live
There is not a perfect source.
I compare cMP to an analogue setup.
All in all I find myself enjoying cMP with minimized WINDOWS more than analogue.
We all are sensitive to different aspects of sound. I guess cMP aligns with my biases?
Hello, everybody! This is not the Big tweak, promised for nearest time, but these settings may bring Your system to sound better
1. HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Control\Session Manager\Memory Management. Create DWORD IoPageLockLimit (REG_DWORD): 1000 is for 4096 hex - (This is real AWE working parameter). You can set the value for 1000, 2000, 4000, 10000 - the one I use now is 10000
2. HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Control\Session Manager\Executive Find there AdditionalCriticalWorkerThreads set the value to 4
3 HKEY_LOCAL_MACHINE\SYSTEM\ControlSet001\Control\FileSystem Create there dword DontVerifyRandomDrivers set value to 1
Reboot, listen to some music, please.
Serge.
Edits: 01/14/12 01/14/12
It is 753KB rather than 2.03MB for xp sp2's version. It works great, at least in my mini-xp installation From observing task manager processes it seems to reduce the overall amount of RAM used more than that so perhaps it calls on fewer functions while still playing beautifully. Those adept with process explorer and the like can determine this. The IoPageLockLimit increase should now be functional. I will proceed to do a sound comparison. I hope others do too.
I am using that and was when I installed the earlier ntoskrnl.exe. Perhaps not using this is the cause of the puzzling difficultes for others when I had no problem at all. I have not slimmed my mini-xp a la either Serge but have done all other cics mods, tweaks, boot.ini changes, etc.
Steppe seems to have made the substituion: what version of xp were you using, Serge, and in what state of tinkering?
using this substitution my internet connection went down. Well, from the start.
1. The kernel manipulation is only one part and THE ABSOLUTELY LAST part of the big tweak that is coming. It is exactly this and one more thing that remain unstable.
2. You can use another kernel. Mine is named neptune.exe, it's a w2k kernel from a no sp version of w2kxp pro. To use it one has to copy it into system32 folder, reshack it and rename, next open your boot ini and type /KERNEL=neptune.exe (or whatever name). It will substitute ntoskrnl.exe for ntldr - the small os loader program which sits in the root directory, e.g. C:\
This substitution is possible as a permanent one and even allows you to completely remove the file ntoskrnl.exe, it will even produce2 entries in the registry system hive. Smth funny, like JustWorking neptune.exe and modify the registry entry cotrolset001-control
However, THERE IS NO NEED TO DO IT. This substitution was made obsolete soon after Grant expressed his interest. Matter is, there is another kernel file, that we don't pay attention to - ntkrnlpa.exe - it's a kernel, that's loaded into memory but not used unless you have specify it with /3GB switch in boot ini and /pae setting. Now whenever we restart our cmp machines, BOTH kernels are in memory, though one is passive. I have made countless takes on deleting the second kernel - the result IS GREAT, but it will kill the machine on the next start up with BLACK screen of death. Problem is that w2k kernel is insufficient alone. while xp ntoskernl.exe IS SUFFICIENT. Moreover, it sounds better alone than a lighter w2k kernel+ mtkrnlpa.exe. So the best thing to do is AFTER ALL THE TWEAKING AND WITHOUT EXPLORER to add /nopae and /mimint to your boot ini and then remove ntkrnlpa.exe. This leaves us with our native xp kernel ALONE. It doesn't only sits well in its own place it is self sufficient and sings better that w2k+ntkrnlpa, due to its better memory-cpu routine. The substitute w2k kernek is lighter and it ALONE sings smoother, but it is a little bit dull in the highs. The native eats more memory, but is vivid without harshness and more dynamic. So, thanks to Grant for his interest and sorry for this late warning. So, kernel is a component itself.
Serge
P.S. If You managed to roast your registry with this substitution, you only have to replace it, delete all logs, recopy all native kernel files from installation disks (as there is a timestamp on them) and reboot.
and it was no big deal to get rid of it.
Has not been on my list of files for a long time.
I thought this was part of the standard deletions though I have not gone back to see.
Had no problems deleting it.
Maybe this is why I have been so pleased with the sound I am hearing?
I have not renamed ntoskrnl so maybe that is why I have not seen anything like what you wrote in my registry.
deleted it and are living good with native kernel, better don't attempt to substitute. It may work for as long as the system is on. Once you reboot, you get a black, not blue screen and it will ask you to replace the wrong kernel. As for the shell, try first to substitute common controls+ common dialogs + shell lightweightapi. Don' remember their names correctly, and after that - the big shell.
Serge
I have tried in every way (which I thought I had already told you) to insert that, even going back to the original installation to see if it would be accepted then.
I am starting to think since it seems impossible, to not worry about it.
Maybe some new inspiration will strike and I will see a way.
Waiting, impatiently, to see the new regimen.
Unlike Ted I do not say take your time! Let us try and see what happens.
according to your method and hives. default is like yours 16 kb, sowtware - only 80, but mine is different than yours because of lynx. With system, I was doing and laughing - there are man even madder than me (in the best sense of the word). Really, what You deleted, I would never even dare to touch... I deleted everything and compacted. Was happy, as my windows folder with unhacked yet fonts and system became only 19,6 mb in size - it is just as I hoped for. Unfortunately, it didn't boot - got a bsod, but I think I know why - ennums and especially acpi devices. Looks like I managed to delete a coprocessor or motherboard - I was in a hurry. Will try to this weekend. As for the big one, You and others already know part of it - it's file substitutions from w2k and xphome, yes, also from xp home. The list of files and THE ORDER of things will be posted.
Serge.
and nothing happens.
I kept getting confused with one of those things - kept deleting it!
Thanks to SNAPSHOT I was able to recover. I find nothing wrong with re-writing a previous "version" - are you saying you do?
With the H55 board I was able to get rid of everything ROOT everywhere.
ENUM was down to nothing except the ROOT entry - all "beneath" it was gone.
The H61 board for some reason will not allow as large a deletion.
I hope I sent you the H55 version - did it say GRANT in the title? If it did that was the H55 version.
I made MANY deletions one at a time. Got lots of exercise going up and downstairs to the other computer.
Now to look at your post.
Steppe seems to have made the substitution . . .
Having to hand a copy of Win2K from way back when, I extracted the file and got it to run successfully on my main cMP2 rig. I even managed to "Res Hack" it from 1.7 to 1.5 MB (compared to 2 MB for the "hacked" Pro version).
My system is sounding better than it ever has though how much of that is down to the different version of ntoskrnl.exe I cannot say (as I've been making other changes).
But, Alas!, another inmate couldn't get his setup to boot using the same file. IOW, the problem is not necessarily version-specific.
Damn!
Dave
I do not think minlogon has anything to do with it since I always have implemented minlogon.
Maybe the fact that your OS has what is required for networking has left enough "in" for the change to be made?
Did you have an old install disk laying about or is there somewhere one can go?
I had tried, unsuccessfully, a week or so back to make this substitution. Maybe the version you are using works and others do not.
Any direction would be appreciated/
Hi Riboge?
What is the trick you used?
It won't work when I substitute:
The ntoskernell which is in mini-xp:
NT Kernel & system
5.1.2600.2622
14-11-2005
2,07 MB
with the one I extracted from a W2K disk:
NT Kernel & system
5.00.2195.1
7-12-1999
1,56 MB
At boot up I get a message it is missing DLL's too startup correctly.
(I disabled Win file protection)
Any idea's why it may not work in my setup?
Mark
cMP2 PC > ESI Juli@ > Van den Hul Optocoupler MkII > Lavry Black DA10 > XLR Mogami Gold > Klein & Hummel O300
I extracted ntoskrnl.ex_ from an .iso of Windows 2000 Pro (no sp). This was dated 9/3/02 and is 967KB in properties, no version number given. Is this in fact the whole of "NT Kernel & system"? I don't know. I simply put this as .exe in \system32\ in place of the ntoskrnl.exe that was there, which is just like the one you have. There was no problem. The one you tried to substitute is larger and the same version steppe said he used. You are likelier than me to be able to explain what is happening.
Hi Riboge,
Thank you for your quick repsonce.
Could check which w2k version you use?
Version info is available with:
right click on ntoskrnl.exe > properties > version > file version.
Mark
cMP2 PC > ESI Juli@ > Van den Hul Optocoupler MkII > Lavry Black DA10 > XLR Mogami Gold > Klein & Hummel O300
Perhaps you missed that I said there was no version number given in properties. I did give wrong details, though, which were for the ntoskrnl.exe in xp home which I considered substituting but didn't. The correct details are that the file(an archive) is 744KB (unpacked 1,640KB) and is dated 12/7/99, still after the date of the version you used by 5 mo. Perhaps the difference is that mine came from a pro version?
Hi Riboge,
Thank you for the info.
My w2k version is pakked 744 kB
and unpakked 1603 kB.
May be other inmates can shed some light on it, why simply subtituting in the ntoskrnl.exe in my mini-xp setup with a w2k-version won't work.
Thank you for responce.
Mark
cMP2 PC > ESI Juli@ > Van den Hul Optocoupler MkII > Lavry Black DA10 > XLR Mogami Gold > Klein & Hummel O300
Where did you get the win2k ntoskrnl.exe?
"1000 is for 4096 hex" is hard to parse. Do you mean 1000 in hex equals 4096 decimal, or 1000=4096 hex? You gave a screenshot in which you used 10000 hex which is 65536 decimal. What is the meaning of that number? One might suppose it means 65.536 Kb, but that seems all wrong here. 65MB would also not seem suitable.
In any case, using 10000 as you did, I experienced none of the softness theo reports and perhaps some benefit in liveliness and detail. This is obscured by my making other changes concurrently in refining my try out of mini-xp. This is really sounding wonderful, better I think with just tweaks and fishies than xp home with all tweaks, fishies and steppes too. It is the smallest installation I have since I have not gone the xp lite route (?yet).
Is widely known as one of the most obscure tweaks.
it started probably with NT and Windows 2000 REALLY benefits from it, it's not only attested by me, as I use in my cmp a w2k KERNEL, but by many people on the net. It is used for securing ram solely for i/o operations. BTW, I think the values are in bytes, so that 1000 stands for 4 kbyte - the standard page size in w2k and xp and 7 for x86 processors. In w2k retail registry there was, in session manager, an additional entry, but it was removed from w2k sp1 and onward. Many people tried to detect the effect of this tweak on xp machines, they attest that the xp kernel understands this command but... and then all the info is deleted by MS. I am now using the value 20000 - the effect is even better, and the mouse movement is smooth like upon polished glass covered with oil, if I describe it not too dramatic. Try for yourself, IT WON"T HARM. As for additional critical worker threads, you cannot add more than 16!
Serge.
Hi Serge
You've been holding out on us - using a different "ntoskrnl.exe" ! Which version is it ? I ask because ever since Win2K-RTM, all Windows kernels have not recognized the IoPageLockLimit parameter. This can be (almost conclusively) verified by running the Sysinternals utility "strings.exe" and searching for the keyname in the text output file.
** see -> http://technet.microsoft.com/en-us/sysinternals/bb897439 to download strings.exe **
This string search technique is referenced here -> http://www.wxpnews.com/archives/wxpnews-039-20020820.htm.
Read the section called Debunking the "Speed Up your Windows XP Box with IRQ Prioritization" Myth, (about half way down the page). Interestingly, Jamie mentions the "IoPageLockLimit Hack" in bullet point three:
The same "strings" output will show that another long-written-about registry hack, IoPageLockLimit, is also missing in action. This one at least used to be there, in Win2K RTM and earlier; in Win2K SP1 they kept the mechanism but the limit was then spec'd by a new value, IoPageLockPercentage; in Win2K SP2 and later they dropped the whole thing.
There is also a thread on MSFN Forum discussing this issue - a concise summary is found in the last post -> http://www.msfn.org/board/topic/25684-registry-myths-1-iopagelocklimit/page868768#entry868768.
Sooooo... I wonder which Win2K edition your kernel file came from ? This might explain the reports of other cMP-Slimmers, with regard to using IoPageLockLimit.
I must hasten to add: THANK YOU Serge, for your tireless efforts to squeeze the most out of the cMP^2 concept!
Cheers,
Grant
That's not a Toy... IT'S A TOOL !!
I'm ok with the
HKEY_LOCAL_MACHINE\SYSTEM\ControlSet001\Control\FileSystem Create there dword DontVerifyRandomDrivers set value to 1
tweak. The other 2 seem to excessively soften my sonics. When I look up the
HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Control\Session Manager\Memory Management. Create DWORD IoPageLockLimit (REG_DWORD): 1000 is for 4096 hex - (This is real AWE working parameter). You can set the value for 1000, 2000, 4000, 10000 - the one I use now is 10000
tweak some links are unclear whether to input the value in megabytes or bytes. So I'm not sdure what it does. One post believes that the 10000 value is equivalent to the default value of 512 k . So very unclear. The other tweak on adding critical thread workers just messes my system up sonically.
Who knows the difference in perception (I believe you Serge that it is a positive in your system) might be in systems, computer h/w or whatever. But we're getting down to the short strokes with a 25 mb windows where everything does impact the sound one way or the other.
Here is my registry.Thanks Rick for schooling me yet again.
Edits: 01/16/12 01/16/12
This is 1st time that a Steppe tweak went wrong for me sonically. I have gone back and forth and conclude that I prefer 'not doing this tweak'. At 1st I did not do the IoPageLockLimit tweak but did do the other 2 tweaks. These made my system softer, pleasant but softer and not as dynamic. So after 2 days of listening I decided to try the IoPageLockLimit tweak, 1st @ 2000 then 4000, 6000 and finally @ 8000. This brought back some immediacy but it still did not sound right. So I went back to 'none of these tweaks' and prefer it. What's troubling also is when I compare my registry to Steppe's as shown I am different in many ways not all but many. For example my Systempages is something like 30000 (how do you do those screen shots) not zero, PhysicalAddressExtension was 0 not 1 (aren't we using nopae in the boot.ini file anymore). Anyway maybe these other items is the difference, maybe its new 32nm hardware versus my 45nm stuff. Believe me I did not knowingly change any of these items.
are easy.
I recently learned how to do it.
First press the PRINT SCREEN button.
Open up PAINT (START/ALL PROGRAMS/ACCESSORIES/PAINT
EDIT/ Paste
and there is the screen you "printed".
Save it with any name you like.
I will have to go back and forth a few times on the new tweak though so far I like it.
/
I cannot remember what the setting in:
2. HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Control\Session Manager\Executive Find there AdditionalCriticalWorkerThreads set the value to 4
was before.
I am hoping returning to how it was will get the drivers back.
Please let me know.
Thanks,
I was playing with BIOS settings and everything came up as if there was nothing (???!!!) wrong with the BIOS.
I tried to play something and it said drivers would not load.
I had never had this experience with BIOS settings previously so I blamed it on the TWEAK.
The H61 board does some interesting things compared to H55. That one would either work or not. This one will go into cMP but will not function.
Mea culpa, mea culpa, mea mea culpa ...
I should have know better.
I see there were a few other changes I must have missed.
I felt sure there were!
Additional critical worker threads concerns cpu, not drivers, Rick. The allowed values are up to 16!
You could have lost juli@ drivers when performing deletions in ENUM, I had had the same problem with Lynx before.
Ted and other inmates! Setting lock pages size starts from 1000, go to 2000, 4000 or 10000. I have a 1gb stick. I used 1000, 2000 and went for 10000. Works perfectly. More than 24 hours, with fresh logs created. Several reboot. Random and consecutive.
Serge.
Rick! it should have no direct effect on driver, only on memory mapping. default value as Ted is right to say is 0, see that it's not the command in another tweak - to disable unknown driver check - this command prevents windows from constantly polling drivers to the add-on cards like lynx and juli@.
I have been listening with this settings and rebooting even after registry (software) complete slimming several times.
To Ted, try a smaller value like 1000, then go to 2000, to 4000 see for yourself. I think 10 000 stands for page 65 kilobytes. I. as You know use a 1gb stick.
Serge.
Serge in this tweak
1. HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Control\Session Manager\Memory Management. Create DWORD IoPageLockLimit (REG_DWORD): 1000 is for 4096 hex - (This is real AWE working parameter). You can set the value for 1000, 2000, 4000, 10000 - the one I use now is 10000
Does it depend on memory size? I use a 256 ram sshould I still use 10000?
More immediacy. Another step in transparency.
Cymbals have more character.
Those are my immediate first impressions.
Are these in conjunction with your earlier recommendations for memory or do these replace those?
a screenshot of the Memory Management Screen would be helpful.
here it is. Don't remember if anything's done here. But YES IN ADDITION. I am making deletions now according to your hive. Stuck with CLSID - Lynx is apparently under a different number, sh*t!
Have to read all this heap of entries!!!
Its pretty easy though.
It will still be a good guide. Not too hard to find the LYNX entries.
AMAZING amount of junk there.
There is no question that all credit for the CLSID files goes to Serge Mihaylov. All of my registry deletions were both inspired and guided by his work.
Hi All,Slimming an XP installation disk with help of N-lite is still not so easy for most of us. And also it is A LOT of work and also i'm not sure what too keep and what not too keep.
So I started looking around for XP ‘distro’s’ who already have been slimmed by others.
A very popular slimmed down XP-‘distro’ seems to be ‘Mini-XP’ slimmed by ‘eXPerience’I’m not advocating software piracy, so be sure too also have a legal XP license for your cMP machine. Microsoft will not like the fact when you are running XP on a computer which you did not pay them for. So with this cleared, now let’s see what this N-lited XP version by eXPerience brings for SQ.
Here is the information text that is included with ‘eXPerience’ N-lited XP-distro.eXPerience Proudly Presents...
"Mini-XP"Windows XP Pro SP2 Unattended
Thank you for your interest in this stripped out 125Mb Edition of XP. I clocked mine in Task Manager using only 49Mb of RAM. Windows XP usually uses 300Mb of RAM at the best of times. This "Mini-XP" takes up only 500Mb total space on your system's hard drive.
By using only 50Mb of RAM, this allows your PC to run fast, I mean VERY fast!
All important hardware drivers and other essentials have been kept in this release - no more problems with those stupid Alcatel stingray modems. You have joystick support, printer support, ISDN, and other bare essentials. This Windows XP Edition CAN have MS Office installed, it is also supported! You still have your calculator, Wordpad and of course, MS Paint.
GONE is that nasty Internet Explorer! It does not exist in this Edition. You will have to install Firefox or Opera or any other browser to access the internet. ALSO GONE is Windows Media Player. You will have to grab some kind of player from somewhere if you want to play stuff. GONE also are all Windows noises like startup and shutdown noises, but audio does work. However, there is no support for scanners in this Edition.
This XP Pro SP2 Unattended is tiny @ just 125Mb!
Windows XP Pro Custom Added Features...Custom eXPerience© Wallpaper
NT Registry Tweaks for Performance, Gaming, Internet and much moreIntegrated XP "Post SP2" Windows Updates (Hotfixes) 23rd June 2005...
Enjoy Mini-XP by eXPerience
Build Date: 14.11.2005
Release Date : Winter 2005
I decided too install this XP-‘distro’ too see what it would do for SQ of my cMP setup.
Installation of XP-mini is no different from any other regular XP installation.
You have too point too what partition XP should be installed.
After this is done, the installation process can be continued unattended.
You can sit back, relax and watch it happen.
When the Mini_XP installation was done, I only had to ad the LAN-network-drivers ( I store my music files on a NAS) and the ESI-juli@ drivers.* Sound Quality in one word: EXELENT!!
This slimmed down MINI XP by eXPerience sounds really stunning.
No hyperbole. The basic install right-out-off-the-box sounds very much (!) better than a standard cMP!It turned out that most cMP XP optimazations where already done:
But these still remain too be done:Chapter 7.1 - Optimising Windows using Control Panel
* step 3: I did not install any video drivers but video traffic still has too be minimized
* step 3.3: Disable Visual Effects
* step 3.5 Set the video mode
* Step 5 tune performance (some tweaks still need to be done here)
* Step 6 disable or suspend services (half of the services is already disabled. But still 15 are running)Chapter 7.3 - Optimising Windows using Registry Editor (regedit)
* Step 2 - disable windows prefetcher
* Step 4 - NTFS disable last acccess update
* Step 5 - change sheduler time slice
* Step 7 - Reduce USB Polling (when using USB)- Disable Watchdog
- Disable Performance Counters
When I checked with the exctrlst.exe tool, only 3 counters where still present.
So eXPerience already had some serious slimming down done
- install slimmed down shell32 (stop Windows File Protection first)
(I don’t use minlogon)Chapter 9 - Optimising the Windows kernel
After having done these final cMP based tweaks, the sound quality improved somewhat.
But, a basic standard MINI XP installment (XP version slimmed by eXPerience) already sounds really, very good! Much better than a standard tweaker cMP XP-setup.So if some doesn’t like too go through the hassle of slimming down there own XP installation disk with N-lite. Than I can recommend an already ‘N-lited’ XP version done by eXPerience (better know as Mini-XP)
Very much recommended !!
Mark
fully optimized cMP2 PC -> ESI Juli@ -> Van den Hul Optocoupler MkII-> Lavry Black DA10 -> XLR Mogami Gold -> Klein & Hummel O300
Edits: 01/13/12
I'm very surprise by the following "I clocked mine in Task Manager using only 49Mb of RAM" ! I did myself a nlite windows XP2 installation with full optimization for cPLAY cMP (It was on my first system), it was under 30 Mo of RAM, if I remember right, it was at 23 or 24 Mo of RAM in task manager.
I think the best we could do is with windows XP embedded, but this is another story, as I don't have the knowledge neither the time to learn IT.
Hi Vincent,You wrote: “I don't have the knowledge neither the time to learn IT.”
I wrote: “Slimming an XP installation disk with help of N-lite is still not so easy for most of us. And also it is A LOT of work”
So, this is exactly why I pointed other inmates at Mini-XP.
No hassle with first learning how too use Nlite, etc, ect.I found the web is flooded with copy’s of this popular slimmed XP installation disk by eXPerience.
Inmates only have too download it and burn it too CD-disk, too try it.Too my surprise: a standard mini-XP install, sounds much better than a standard tweaked cMP XP setup.
So for inmates with only basic computer software skills (like me), this is good news.
Without much hassle, they also can have a slimmed down XP version which sounds excellent.Mark
fully optimized cMP2 PC -> ESI Juli@ -> Van den Hul Optocoupler MkII-> Lavry Black DA10 -> XLR Mogami Gold -> Klein & Hummel O300
Edits: 01/15/12
Just wondering.
Hi Rick,
I forgot to mention AWE.
I apply the AWE function, since it is very easy to do.
But... I can not hear a SQ improvement/difference when using AWE or not using it.
Same for Minlogon and using LAN
I also can’t hear any SQ /improvement/change with these being used or not used.
So I don’t use/implement minlogon
but do use LAN.
Mark
fully optimized cMP2 PC -> ESI Juli@ -> Van den Hul Optocoupler MkII-> Lavry Black DA10 -> XLR Mogami Gold -> Klein & Hummel O300
I do it but I do not hear anything.
Some say they prefer it NOT installed.
What is LAN?
I find MINLOGON to be essential, though.
The best part of these minimal installs is how much junk is left out of the registry. You do not have to take it our yourself.
Easy to remove the unneeded files and folders. A full fledged WINDOWS registry is a lot of work and time to whittle down.
Did you also try Jollida's nLite version?Jolida's is 98MB and uses only 32Mb of RAM in Windows XP.
Firewire and USB Audio are included in the following Multiprocessor Last Session.ini.
In case you intend using "cplaychg.bat" I recently published for cPlay resolution change in cMP, you can delete "Command-line tools" from Multiprocessor Last Session.ini.
Edits: 01/13/12
nt
fully optimized cMP2 PC -> ESI Juli@ -> Van den Hul Optocoupler MkII-> Lavry Black DA10 -> XLR Mogami Gold -> Klein & Hummel O300
.
Hi Bibo01,
I think I lost you somewhere this quickly already.
I was expecting an *.iso (image-file) with an Jolida nLited XP version on it, so that I can burn it too CD-rom too get an installation disk.
The link links to an ini-file.
What should I do with it?
Mark
fully optimized cMP2 PC -> ESI Juli@ -> Van den Hul Optocoupler MkII-> Lavry Black DA10 -> XLR Mogami Gold -> Klein & Hummel O300
(and supposing You have nLite installed - it's free) run Nlite, tell it where the files (you supposedly had copied already) from original xp installation are, then on the third screen say import ini, after that ony yes and continue till the last page, there You will be suggested to create an iso. Approximately 50-75 seconds' process, from starting nlite to getting a new iso (including NLITE's WORK on substracting files).
There You are.
Serge.
P.S. In case You have any problems setting AWE from Junaid's install, e-mail me, I will send You my ini file, that is 100% AWE compatible supporting cics' routine (Just like Junaid's, AFAIK)
P.P.S To Run Nlite You need to install some Net framework, that can be downloaded from nLite's site for free.
Edits: 01/13/12
With all the deletions of files in system32 and keys in the registry this apparently results in a very fragmented registry. Before I started reducing my registry its size was 16 mb. After some manual deletions (shown the way by Rickmcinnis) I was able to reduce it to 6 mb. And it was a significant sq pop. But I could not get it reduced further. I found a wonderful registry compacter Ntregopt (thanks to help from Ryelands) and it works like a charm. What you need to do is:..Download Ntregopt (google it...its freeware)
..Open regedit on your desktop
..Load your CMP os disk into your desktop (via a docking station or through a usb/sata converter)
..Load all your cmp hives into regedit (Sam, Security, Default, Software, System). Load them one at a time.
..Run Ntregopt...it will ask if its ok to start, say yes and its off to the races defragging or compacting your desktop registry as well as all the hives you have loaded. Its important and time efficient to do them all at once. When I tried to do them one by one I got to System and tried to boot it in cmp ... it would not work. After Ntregopt is finished it will ask to reboot ... say yes.
..After reboot go into your cmp registry and delete all bak files and all logs...Put cmp os disk back into your cmp pc/reboot. When I did this I ended up with a < 1 mb registry size.
This process was heavily contributed by Rick, Dave as well as Steppe/Mihaylov for pioneering registry slimming.
As you may go further in reducing cmp registry manually... defrag it again.
Forgot to say this is a significant sq pop... dynamics, timbre, mid and upper mid purity.
This tweak is especially good for those of us who did not want to reload Windows with a light footprint, do all the CMP / Cics tweaks again, redo all the Steppe/Mihaylov tweaks ... to end up with a tight registry.
Edits: 01/13/12 01/13/12 01/13/12
almost done with registry deletions.Can't seem to get rid of root subentries in ENUM this time. Much more stubborn than the H55 install.
Edits: 01/14/12
...and far be it to comment on my 1st (and only tweak) but this is mega significant. I played one of my faves ... a Paul Paray Detroit Symphony Orchestra recording of Suppe overtures and it was so good I could not believe it. This recording is bright but very lively and my kind of bombastic power music. Its basically military type music so there is alot of macho' in it. I could not believe the power, the dynamics, the scale/sense of space, the density of the musical fabric and inner detail. I was disturbingly overwhelmed in a positive way by the playback.
I believe that anyone (no matter where you are in the Steppe/Mihaylov journey) can benefit from a registry defrag. Iirc the registry is located in memory so any efficiency moves made to it have a direct impact on music from a memory player like Cplay/cmp. Another benefit I picked up from the cumulative effects of registry slimming/compacting is that it doesn't swap data until at least 2:23 min/sec into the music whereas it was 2:13 or so. So this is an all around efficiency move but my oh my what it does for the music.
I have compacted my registry (it was not really very empty before). Thanks to You I now know which way to go and I'm beginning to see the light at the end? of our toils.
Serge.
P.S. Now almost everything is ready for the FINAL TWEAKED VERSION easy for everybody.
Edits: 01/13/12
Thanks Serge just happy I can give back a little. You got another tweak coming? Hope so. I always look fwd to your stuff.
Congratulations, Theo, with your registry tweak!
And Serge (steppe) keeps saying riddles :).
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2 Ru, GA-H55M-UD2H, i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S FAT32, ESI Juli@, full linear PSU, NAS - WD My Book Live
Thanks Serge with this tweak maybe some of us can start catching up with you.
nor a secret. Hello, Serge, and Happy New Year, ar last. (Other inmates will think, I have finally grown mad, but You know what I mean). Yeah, not a secret, it's just that I managed to achieve a very interesting and significant result soundwise, but cannot yet recreate it so, that other inmates would have a no-brainer tweak. This one is big and single wrong step brings a BSOD, that's why I said it's dangerous. Please, be patient, it's coming. BTW, what size is your winlogon (read minlogon) file. Is it 17 or 24 kb?
Serge.
Hi Serge! Happy New Year!
My winlogon (minlogon) is 24KB.
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2 Ru, GA-H55M-UD2H, i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S FAT32, ESI Juli@, full linear PSU, NAS - WD My Book Live
and, most probably, most inmates have the same. I had initially big problem with minlogon, so even deleted initial archive, but then downloaded again, next I have found an older one. Know what? Well, they are DIFFERENT. I use a 17 kb version (totally hacked - 16,5 kb). Shall I post about it at the top, You think? Also, how can I make it availible to the others, if they are interested? I can send it via e-mail...
Serge.
You are very very cryptic.
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2 Ru, GA-H55M-UD2H, i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S FAT32, ESI Juli@, full linear PSU, NAS - WD My Book Live
Hey we have Snapshot or the equivalent let us be the judge of risk to take.
If I fail to create a safe routine, I promise to post it tomorrow night Moscow time (it's 21 00 here now)
Serge.
No problem take your time.
Thanks Ted for the concise write-up... with any luck I can use it this evening. Looking forward to another (addictive) SQ-Boost !
I've used NTREGOPT before, but it never occurred to me it could act upon anything other than the live registry (of the system it was run on). Loading the off-line hives and performing the compaction is "just too easy". I hope it works out that way for me. (...I can hear Rick laughing already...)
Cheers,
Grant
That's not a Toy... IT'S A TOOL !!
It will work for you. I'm sure Rick was laughing at me too because he heard it before I did (so did Ryelands).
My config folder is approx 852 kB
I have been looking all over the thread, and I can't find where it is described what is safe to delete.
I am at 16mb after Nlite and compacting. I would love to try this but prefer to benefit from your trial and error and not re invent.
Good, Rick! Could you upload the screenshot of SECURITY hive?
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2 Ru, GA-H55M-UD2H, i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S FAT32, ESI Juli@, full linear PSU, NAS - WD My Book Live
So a screenshot would be just IT sitting there as an entry.
I think the key to those files being small is doing a minimized install and then compacting.
In my case that is all I can do.
I have trimmed another 100K out of the config folder since that post.
I had not finished deletions out of SYSTEM when I made my first "caompaction"
Strange. Well then maybe you show a screenshot of the entire registry as I showed?
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2 Ru, GA-H55M-UD2H, i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S FAT32, ESI Juli@, full linear PSU, NAS - WD My Book Live
There are things to see on both sides of the screen.
I have sent you an email on your AA account with my home email address.
Write me back and I will forward a zipped file to you.
Much easier to deal with, and more informative, than a screenshot.
That's what I get for writing before trying.
Send me an email to my AA account, same as my username you do not have to use the AA buffer, and I will send you the files.
I assume you want the H55 version?
Could not have done this without you...a big thanks again.
This ia a very simple batch file to change cPlay audio resolution from within cMP.
Place the bat file in cPlay directory, probably C:\Program Files\cics Play.
Associate cPlayCHG.bat to a cMP button and use it as you like it!
When batch file is used, it creates a tmp file in working directory.
In this way you will be able to change cPlay starting audio resolution without enter/exit/enter cPlay.
This is particularly useful, for example, for users who normally keep cPlay at red book resolution and want to change it from time to time.
I would like to point out that, in case you use Jolida's nLite intallation,
you must delete "Command-line tools" from his LAST SESSION.INI in order for cPlayCHG.bat to work.
Could you add to the batch file that it display the samplerate for the files referred to in the highlighted cuesheet in cMP (knowing this only helps with cuesheets and not playlists of differing rates)? Who can remember with hundreds or thousands of cuesheets/albums? That way one could choose an even multiple.
I'm not able to do that in a batch file.
That's why in my cue file I keep information on resolution in album title: "my album title (24/xxx)".
Thanks for the batch file.
It may be the basis for the most viable workaround, which is for a batch file similar to yours, to read the sample rate from the header of the audio file.
Less elegant, would be to require the user to put something in the cue file, or file name/path.
riboge - check your email for email from AA from me on Vista, thanks.
Ken, I replied on friday to your email and did so again just now.
I tried cMP shell on Vista SP1 32-bit and it hangs at the Welcome Screen.
I followed what everyone calls "disabling the Welcome Screen in Vista", which is to use only one account, and to untick the checkbox to ask for a password logon. But even when you do that, there is a screen in the boot process in Vista with the word "Welcome" and it hangs at that point when you try to startup in cMP mode.
Has anyone managed to fix this in Vista ?
Thanks !
I got no response posting under Jack's thread so...Seeing what can be achieved without it, I am doubting that trimming per se is worth it. Once you have done all cics optimizations including minlogon, boot.ini tweaks and fishy tweaks to xp home cMP you end up with a much more natural, rich and balanced output with all the great detail, soundstage and localization of instruments that trimming xp pro brought while maintaining many convenient functions with no seeming sacrifice to sq. At least, that's how it is seeming to me after a couple days of playing a xp home cMP.
Do others agree? Or are there those who have trimmed xp home and achieved even better sound, that is, better without the imbalances,'thinning' and somewhat artificial 'air' of the highly reduced xp pro setups? Or are there those who prefer the sound of the reduced xp pro?
I can supply the files to anybody that wants them. And all should know (I can now say) that it was Rickminnis that came up with this. It is brilliant.
Mihaylov's inspirational registry screenshot gave me what I needed to know where to begin this project. That is when the project began
Few, if any, will be able to use them as is. I offered them to Ted and others on Jack Wong's XP Home reg list as an example of what a minimal registry would look like after seeing how huge the XP Home registry was and as a guide to what can be deleted and the machine still work and to my ears work quite well.
One can start with either the Pro or the Home registry and end up in the same place. My suspicion is that the HOME version sounded better just because it was simpler. Not as simple as my minimal nLite registry but vastly less complicated than a standard XP Pro install. I never inserted the HOME registry into my machine.
There is lots of audio stuff left in even after STEPPE's deletions. Vast amounts of video stuff and what I call multiple entries - you will see the same thing listed numerous times which would make sense for a computer used for multiple tasks but with our (assumed) single purpose machines these do nothing but add confusion to the sound.
I started deleting these items towards the end of my trials and errors and it was if the best was saved for last.
USB has been completely eliminated since the PS/2 input is less intrusive (thanks again to Grant for making the case for this). ControlSet2 is as small as I(or you)are allowed to make it. There is a bare minimum that WINDOWS will not let go. There is a way to get rid of it all but at this point there is no installing SW required to do this into my machine. Something to try with my next install.
I can tell you Serge/Steppe has not stopped his explorations and there is much excitement coming from him in the next few weeks.
Just when you think it is DONE ...
thank you for reminding me. Of course Mihaylov is a pioneer on registry stuff.
you said
'I worked on registry slimming in my xp pro cMP but didn't get too too far before giving up going further due to too many bsods. Might you have a list of items that can be deleted, or can you send a copy of your software and system hives for me to explore and work from? My setup is clearly different enough that I could never use them directly but only as a guide.Also, do you or Mihaylov know for a fact that the differences between home and pro are trimmable, that is, that they are only re additional items in pro and not ever difference in the items themselves like different versions of essential dlls, drivers, etc, or contents of untrimmable registry '
Maybe you are already there. Remember as of a week ago I did no registry trimming...just the tweaks. If you used n-lite or some registry defragger or cleaner you may already be there.
Edits: 01/06/12 01/06/12
Cant do the registry (its proprietary) but the system32 files are what others have reported (Steppe, Rick, Mihaylov theob).
Prprietary means given to me by another unidentified inmate who wishes anonymity.
...like Mihaylov says if you skinny it down on xp home or you start with xp pro once you get down to the bare min windows including registry you are there. Maybe it was easier for you with xp home and yes that is certainly worth something. If I was starting a new I would take your route with n-lite. In reality in all honesty, if I was starting anew I would never take this journey. I had a steep learning curve as an audiophile who knew zero about computers just launching cmp/cplay. Hell I didnt know what a driver was let alone all this stuff regarding the registry. And I believe a lot of 'would be computer audiophiles' not yet into it will not take the journey because its too difficult. And that would be a shame. This stuff blows everything away that I have heard as a front end. Not arguing just expressing a different perspective.
Have you slimmed down your registry? With my recent slimming including this am my windows folder is now down to 32 mb.
I worked on registry slimming in my xp pro cMP but didn't get too too far before giving up going further due to too many bsods. Might you have a list of items that can be deleted, or can you send a copy of your software and system hives for me to explore and work from? My setup is clearly different enough that I could never use them directly but only as a guide.
Also, do you or Mihaylov know for a fact that the differences between home and pro are trimmable, that is, that they are only re additional items in pro and not ever difference in the items themselves like different versions of essential dlls, drivers, etc, or contents of untrimmable registry keys?
I have not xp home so I can not to say anything for a fact. But I know for a fact that versions of some .dll and .sys are the same in xp pro and xp embedded ;).
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2 Ru, GA-H55M-UD2H, i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S FAT32, ESI Juli@, full linear PSU, NAS - WD My Book Live
I have never regreted any of the Steppes. Yes some were better than others but as Steppe mentions after Steppe 16 it was all good. The fish tweaks, resource hacker stuff plus watchdog et al even better. But where it really enterred the sacred zone for me was when I started registry slimming. About a week ago my registry was 16 mb. I, through the invaluable help of an anonymous inmate, have slimmed this to about 7 mb. The results are simply staggering. I know I have further to go and will do it but its a long and painful process, lots of no boots. Others have reached simply mind blowing levels of registry size and while this is not goal in and of itself it is a rough indicator. For me the biggest gains were from eliminating unused sevices entries and software stuff in the registry. I really thought that the tweaking was done but no it goes on. The open source nature of cplay/cmp is the only way this could have been accomplished. I just wish cics could hear his creations fully tweaked as Steppe, Mihaylov, Mario, and others have shown the way.
You don't say if this refers to use of xp home with all these trims. I have done all save your latest registry radical trimming with xp pro and liked the results but prefer xp home with boot.ini tweaks, reduced shell32.dll and fishys without trimming.
tweakings on XP home, truly, I am somewhat tired with all this long search in a away, but I can say from my own experience that really good and balanced sound comes to you at the end of part 15,, then after part 18, then at the end of part 22+ fishies, that is now. I don't think it's artificial, I rather prefer to call closer to studio reality, though it is by no means an "audiophile"-smooth or comforting. I don't know what music you are listening to, but if you put the following cds on your player, you will immediately hear what you have achieved with tweaking
Deep Purple 1970 In Rock
Rainbow 1978 Long Live Rock'n'Roll
Ozzy Osbourne - Ozzmozis
Serge.
BTW I am using several xp-home files in my setup...
I have been using Aspire One as the music server ever since I started the cMP2 project. Since XP Home came with the netbook, I have no choice but be a XP Home user since day one. I have never had the opportunity to have a XP Pro setup to compare with XP Home side by side on my system, but have always been extremely satisfied with the performance of cMP/cPlay on XP Home. In fact, I am more than satisfied with it, the current implementation way surpassed anything I could have expected and I considered it to be the best source I have ever achieved in my long hifi journey.
I do not go all the way in my Windows trimming. I stopped at steppe 4 + jolida 2. I had ventured way up to steppe 8 but found the sound thin with artificial high and had to retract back to steppe 4. Though I did found the latest steppe (and others) tweaks extremely beneficial (maxmem, watchdog, performance counter, and sliming of shell32).
I am planning to build my 'real' server soon. May be by then I can compare the full slimming of XP Home with what I have now.
KC
=I do not go all the way in my Windows trimming.=
I think this is the reason. In fully tweaked system as my cMP2 there can be no differences between pro and home.
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2 Ru, GA-H55M-UD2H, i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S FAT32, ESI Juli@, full linear PSU, NAS - WD My Book Live
I compared cPlay to MusicBee to MusicBee playing the same 24/96 file from Ramdisk.
Ramdisk made a mild improvement, but cPlay was noticeably better yet.
So, what does cPlay do differently from other players - other than RAM Memory playback ?
(Or is it some sort of secret?)
as the saying goes. Hello, everybody. You have probably forgotten, that cplay appeared at first as a supplement player to the cmp shell. CMP was designed earlier to, mainly, stop or kill or suspend other processes that a pc is running commonly while playback takes place. I have seen a diagram of the windows audio process - its TERRIBLE. Before data reaches kernel and after it passes through "fire and water and copper pipes". Then asio itself, its driver uses several other dlls and drivers like portculs.sys and f*ng drmk.sys that portclis is dependent on, but there are many many stages, where echo canceller and audio splitter and sysaudio and redbook and codeks all take place (sorry for punctuation). Cplay was specifically designed to avoid many of the nasties and it is and was to exploit the benefits of cmp. Also, if you read the logs with care, you will find that mostly Cics wrote of ASIO refinements, not DSP. especially at first. Note that cplay is an asio-only player, no foo-out dlls and crap. It's the shortest and most minimalistic player, that I know of. Even video interference was partially taken of by cmp tweaks.
I bet, cics would explain better. But it's ASIO changes that matter most, IMHO.
Serge.
Edits: 01/04/12
is it some sort of secret?
I don't think it was ever a secret though it has never been formally written up, the nearest thing to a write-up being the Release Notes, esp (obviously) those for the last version, v 39.
Though noting that this or that release features "DSP improvements" is a bit cryptic, it does suggest that the program was continually honed to make the code more efficient rather than, as is more common, taken to the point where it worked well enough then inflicted with "features". The approach is in line with the general thrust of the cMP2 "project".
cPlay is of course feature light with a CUI (Clunky User Interface) rather than a GUI. Those who followed the program through fifty-odd releases can testify that its sound quality did steadily improve over the two years (Sept 2008 to October 2010). Try comparing (say) v 2.1 with 2.39 and bear in mind that the only difference between the two besides minor GUI improvements is the efficiency of the code.
Interestingly, the developer of Pure Music, an affordable player for the Mac, seems to be taking much the same approach with, if users are to be believed, much the same results.
HTH
Dave
Thanks for your reply.
You stated:
"Try comparing (say) v 2.1 with 2.39 and bear in mind that the only difference between the two besides minor GUI improvements is the efficiency of the code."
Okay, my question then is, "efficiency in doing what ?"
cPlay is doing something differently than other players, that causes the audio waveform to have less added distortion in the playback process.
At first glance, I thought this was due to RAM memory playback, but the comparisons with other players using RAMdisk confirmed that this is only half the story.
It's not related to timing, because I am using an asynchronous USB DAC that creates its own timing internally.
So, some other usually found distortion has been eliminated in cPlay, and it would be quite worthwhile to everyone to pin down exactly what it is.
(Perhaps what is needed is a very detailed change log, if one exists ??)_
my question then is, "efficiency in doing what ?"
In crude terms, the number of instructions the program executes to perform a given operation. I'm no programmer but I recall working with programmers who would write a program so it worked after a fashion then spend a long time refining the code so it ran faster. Compare e.g. "the square root of the product of the square of four and the square of two divided by four times the square of one" with "one plus one". (I think I have that right).
One particularly skilled programmer I worked with would get his routines working in Fortran then re-write them in Assembly language. I've seen differences of one, maybe two orders of magnitude in runtime.
cPlay is doing something . . . that causes the audio waveform to have less added distortion
As I see it, no waveform is passed to the DAC - the DAC creates one from a stream of data but has to do so in real time. (The meaning of "real time" has been disputed here but it's clear enough IMHO in the USB audio context.)
It's not related to timing, because I am using an asynchronous USB DAC.
An asynchronous transfer protocol reportedly reduces USB timing errors (I say 'reportedly' because I've never heard a device using one, let along compared it with others) but developers on this forum at least are adamant that it is not a panacea. I don't think that anyone seriously suggests that a correctly-configured Foobar (e.g.) is not "bit perfect". USB cables apparently affect asynch devices. And so on: the effects are subtle.
If I may say so, it is refreshing to read a post that starts with perceptual evidence and queries the orthodoxy rather than the other way round. A common response is "I don't see how it can . . therefore it doesn't".
"As I see it, no waveform is passed to the DAC - the DAC creates one from a stream of data but has to do so in real time."
Yes, "audio waveform" is talking about the eventual analog. In terms of digital, it would be alterations of data values.
Now, I can certainly see how electrical noise, voltage spikes, etc. could cause alternation of data values in a PC, so all the hardware mods make sense (with the only variable being the actual degree of benefit of any particular mod, which varies with each listener's ears and audio setup).
And I can see how any process running the same time as an audio player could cause a disruption (skeptics clearly do not understand the difference between a real-time streaming digital process and an error-corrected "offline" process).
But I don't understand how any non-faulty code in a player, can sound better than non-faulty code in a different player, both of them running on the same PC, so the above factors of hardware and OS generated noise and spikes are the same for both players.
You state: "I've seen differences of one, maybe two orders of magnitude in runtime." An audio player has to process one second of audio data within one second, since it is real-time. I don't see a benefit for completing it in a tenth of a second.
I'm not using upsampling, so that is not part of the equation, and it's a 2.80ghz dual core processor, so it is not being taxed at all by any audio player.
PS I'll table the USB/jitter issue for now, except to agree that it is not a panacea, because timing jitter is certainly not the only distortion.
hey Ken,
I certainly dont know for certain and Ryelands and steppe have given some good explanations.
But I don't understand how any non-faulty code in a player, can sound better than non-faulty code in a different player, both of them running on the same PC, so the above factors of hardware and OS generated noise and spikes are the same for both players
Not sure it is that simple. And I definitely DONT think that all players create the same hardware and OS noise. I know this for a fact.
Let me explain. There was a time when none of my outlets were grounded (it was an apartment). There was a something similar to hum, but it was different in that there were just noises that ranged up and down the audio spectrum. Some players sounded like a growl, some like a delivery truck backing up, some a bit of both. I think that alot of the sound might be from the HDD or cpu, etc.. The sound was soft and with the volume down you could hear it but it didnt change with the volume. The computer noise was definitely different among different players.
The main things that people complain about cplay are IMHO some of the reasons it sounds so good. For instance the black and white no nonsense display. That cuts down on video noise. The ASIO only capability. As steppe mentions that bypasses a bunch of badsounding os stuff. I recall cics talking about "period jitter" and how asio could control it. See the link below. I know he spend a bunch of time on asio refinements and i can tell you from my Lynx card that asio drivers make a huge difference.
Finally cplay is the only player I know of that was designed around a comprehensive hardware and software approach and I think it shows.
No one here remembers the bending of our minds
Hey Ken,
More on the async usb
No one here remembers the bending of our minds
Fortran...I used to program in that many years ago as well as Basic and APL. We had a version of Fortran developed by Nasa called Nastran that did highly sophisticated stress analysis (for its day) of Vehicle parts. For sure this was a long time ago when the computer was humongus in size (maybe the size of todays c class car (focus)). Ahh the good old days.
I would appreciate someone's help in activating XPHomeSP1 on a non-network environment. It started checking the licence when I load it config folder to XPProSP2. I've no such problem in CMP mode when minlogon is applied.
Thanks
Why would you not want to remain in MINLOGON?
I would think your only choice other than MINLOGON would be to call MICROSOFT and get the key.
I would stick to MINLOGON myself, AND cMP mode.
A couple of questions:
- Does the difference between the versions (SSE2, SSE4) only affect sampling ?
and
- Is there a way to set cPlay to output the exact same sample rate as the file's encoding ?
Many thanks !
Does the difference between the versions (SSE2, SSE4) only affect sampling ?
As I understand it, it's upsampling where the differences matter. (My earlier point that versions other than SSE4 would not even do 192KHz seems to be incorrect.)
Is there a way to set cPlay to output the exact same sample rate as the file's encoding ?
Not automatically, no. Sample rate switching is not, to put it politely, one of cPlay's strong points.
HTH
Dave
In order to complete Bitch Batch program, Junaid needs your help!!!
He requires the list of exception files (files that cannot be deleted) for Juli@ card and, possibly, for EMU 1212.
He also requires the list for firewire interfaces.
Ribogee (Play-mate or anyone else), please we need your help.
I remember you saying all 1394 dll/sys/exe files, but you had to look further for ASIO.
You can all post your reply here. Thanks
Could you please supply list of files for Juli@ that cannot be deleted?
Thanks
JulaASIO.dll and Jula.sys.
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2 Ru, GA-H55M-UD2H, i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S FAT32, ESI Juli@, full linear PSU, NAS - WD My Book Live
I've been using XP Professional Edition SP2 for Cplay/CMP upon recommendations made on cicsmemoryplayer.com, ever since I started the CMP project long time ago. Recently, when I found you guys slimming your Windows and Registries, I have this in mind - Why not use XP Home Edition? It's less sophisticated.
The Home Edition is perhaps a simplied version of the Professional one and has a lot of features trimmed down. I think that's what we are lookng for. You may refer to the below links for the differences.
http://www.annoyances.org/exec/show/article10-002
http://www.windowsnetworking.com/articles_tutorials/wxpdifs.html
The only issue for not being able to use XP Home Edition for good CMP playback is the absence of "lock page in memory". Nevertheless, I tried installing an Nlited version of XP Home. Upon first boot, it's about 50mb less than the Pro verion of XP. With basic optimization and registry tweaks applied, it sounded very good, and in my view even better than the XP Pro platform with AWE enabled and minlogon applied.
Somehow, I was able to enable AWE in the XP Home by installing a third party software called "Local Account Manager". You may download it from the link. You will need to enable AWE for all users and groups. You may delete unnecessary users before uninstalling the program, after confirming that AWE is working in Cplay.
I did registry tweaks and minimun trim down on audio and video codecs only, feeling this is more balanced.
I feel the AWE enabled XP Home CMP platform sounding better than its Professional sibling. It's more musical, more airy, more presence; everything is better.
Just try it to see if you like it or not.
*What is missing in Home Edition:
Only supports only one CPU
Does not include the IIS Web server
No System Preparation (Sysprep) utility
Does not support Dynamic disks(instead, support Simple Disk type)
Does not support Encrypting File System (EFS)
Cannot be used to logon to an Active Directory domain
Group Policy Objects not available in Home Edition
cannot be used to logon to an Active Directory domain
Home Edition does not support RIS deployments
No user interface for IPSecurity (IPSec)
No SNMP
No SAP Agent
No Network Monitor
No Multiple Roaming feature
I have just been comparing
1)the result of fully cics-optimized xp home with all boot.ini tweaks, codec disables, and minlogon but not yet with Jack's files to
2) fully optimized xp pro(except no bios refinements possible on my laptop) after 19 Steppes and all boot.ini tweaks, minlogon, reduced shell32.dll and all fishies. For unknown reason, on this second system I can't get Mikhaylov tweak 1 to work as it did in the first system so didn't do any of his.
Even before I just did minlogon, which worked first try, 1) sounded fuller and had deeper, more apparent bass. 2) had more 'air' by a bit. After minlogon, 1) is even better in these ways plus air with a clear liveness and richness advantage.
I will now try to use the reduced shell32.dll and fishy reg changes to xp home and see what that does.
The first 3 brought greater clarity and detail.
Shell32.dll supplied to us that works in the trimmed stepped cMP with the xp home system produces a startup halt with error: cics Memory player: procedure entry point AssocGetPerceivedType not found in shlwapi.dll.
I took on editing with Resourcehacker. My first try was to edit shell32.dll copy I had from earlier xp pro version. This produced the same error. Then I edited shell32.dll from the xp home installation down to 2.66MB from 7.95MB. This worked with some overall improvement. Statement made earlier by someone, ?Jack, that the shell32.dlls are the same appears incorrect, but I am don't know how or why.
I found that some times you have to hack your shell32 or other OS file yourself (with Resourcehacker) and then it will work versus using a supplied file from elsewhere.
Hello, everybody! In most cases you can delete everything inside your file with reshacker. In some cases you can substitute your file with a supplied one or !!! with the one from another OS!
There are two conditions that should be observed:
a. The file should already reside in your system32 folder
b. Your config folder should contain hives and logs to them (remember, I said remove /minint from boot.ini), dated from a time AFTER the file appeared in your system32 folder. AND YOU SHOULD HAVE A SPARE COPY of your config folder with THE SAME hive files
NEVER try to substitute kernel32.dll, ntdll.dll, gdi32.dll and ESPECIALLY USER32.dll. Files samsrv and samlib should be substituted only in pair, FOLLOWED by rpcrt4. But before You do it, substitute lsass.exe and lsasrv.dll. Oh it's a jungle! If You will get a BSOD and after that you 1. restore original file, reboot and get a message, that your system32\config file (no name given. It's a system and software hive that is broken) is missing, open your hdd from another windows, 2. delete everything inside config folder, ALL HIVES, all logs! and 3. populate it anew from a reserve (spare) folder. Reboot. It should restore, reboot again to create new timestamp on the registry. All the crucial files are registered by system and software hives every 24 hours. I am now finishing a big tweak, that directly concerns xp pro, xp home and some other things. Will post it soon, but need to recreate it several times - very dangerous. WORK ONLY WITH YOUR OWN FILES!
Serge.
:).
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2 Ru, GA-H55M-UD2H, i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S FAT32, ESI Juli@, full linear PSU, NAS - WD My Book Live
Yeah, I thought I might get this suggestion. I have been daunted by the prospect of taking this on. Is it easy and straightforward or tricky and requiring many trial and error cycles?
Always tricky for me...I dont know about trial and error cycles. What I do is load my os disk into a docking station on my desk top and hack it from there.Forgot to mention have your Snapshot or Ghost program ready to use you'll need it.
Edits: 01/02/12
Using XP pro, I have done a few of the registry tweaks, minlogon, and removed some but not all drivers, noguiboot etc etc....
I also added the /safeboot:minimal switch to the boot ini file.
I also removed a few services from the safeboot registry.
I use a USB SPDIF converter using the ploytec driver.
The sound is better using the safeboot option. But I know my os is not as tweaked as some of you guys. Perhaps someone else can boot into CMP running in safe mode, on home or pro, and see if it sounds better for you also.
Now when I start CMP taskmanager says I use about 34Mb of RAM, about 900 Handles and about 90 threads. I can give exact numbers tomorrow if interested.
.
This is useful for no tweaked non-cMP2 systems. cMP2 users already have disabled or removed (in registry) unnecessary services.
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2 Ru, GA-H55M-UD2H, i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S FAT32, ESI Juli@, full linear PSU, NAS - WD My Book Live
This is useful for [ untweaked ?] non-cMP2 systems.Agreed - but it also suggests a very quick way of demonstrating that "slimming" is, despite claims by many a guru on this forum and elsewhere, a worthwhile exercise.
Folk who find that the tweak offers no change need bother themselves (or anyone else) no further.
Dave
Edits: 01/02/12
You said '...I also added the /safeboot:minimal switch to the boot.ini file. I also removed a few services from the safeboot registry...'
Just so I understand add /safeboot:minimal to the boot.ini file? Is that right? Also which services did you remove from the safeboot registry and how do you do that?
Yes you are correct, just add safeboot:minimal to your boot ini file.
To edit your safeboot registry entries:
Start> run > type regedit hit enter
find and open the folders: Hkey_local_machine > system > control > safeboot minimal.
I intended to delete; Help SVC, SR Service, Crypt SVC, RPCSS, Dcom Launch, winmgmt. But actually there were at least two of these that I didn't delete. Sorry my memory isn't able to tell you which ones exactly I deleted.
Perhaps someone else can advise if its safe to delete all of the above?
And perhaps any others that can be deleted.
Before deleting, I right clicked each service and exported (and saved) it as a .reg file so I could double click and re-instate the service in case I made a mistake.
Hi inmates, happy New Year!What are your wishes for 2012? Expecting to receive New Year gifts?
OK. Back to XP Home Edition. After having some hard time installing XPHomeSP1, I compared its sound quality with XPHomeSP2 and XPProSP2. Each OS underwent the same basic optimizations,including the latest registry tweaks, AWE and minlogon.
I mentioned in the last post that the SQ of CMP on XPHome SP2 is on top of XPProSP2, the HomeSP1 is one step forward. Greg, you are right! The differences between HomeSP1 and ProSP2 are profound. For example, less harshness, better minor details and focus, lighter but more real highs in HomeSP1 than ProSP2. A note lasted a bit longer in HomeSP1 and there were more minor reverberations. More importantly, the sound is more musical and appealing.
The HomeSP1 brings a jump in SQ, which I we believe is difficult to improve with the trimmed XPProSP2. However, I experienced difficulties in making HomeSP1 work with CMP, e.g. it couldn't shutdown in CMP and often couldn't reboot into CMP at all, occasionally no AWE. I didn't have such problem with HomeSP2. By chance, I made a few working installations, out of more than 12 attempts on 3 different computers. That said, however, I think it's still worthwhile to do so for the sake of SQ improvement.
In addition to the differences in SQ , I have the following observations:
1. The files in the 2 versions of SP2 are of the same size, but smaller in the SP1 version. This would potentially offer better chances to manipulate dlls to smaller size with Resources Hacking. There may be fewer AVIs, Icons, Strings and Dialogues.
2. HomeSP1 does not work well with CMP. It could hardly reboot into CMP mode. According to my experience, there was only one chance to do it right, i.e. CMP-Reboot-CMP. Thereafter, one cannot reboot into XP mode or shutdown. Should one missed the golden opportunity, will have to reinstall again.
3. After minlogon, User's ID was no longer recognized and the OS sometimes behaved erratically - settings were changed, I know, but some were very difficult to optimized because sometimes explorer wouldn't run in CMP mode.
4. The Registry of HomeSP1, and predominantly its simplicity rather than the size of files I believe, accounts for most of the improvements in SQ.How did I find out it's about the Registry? Well, I think you should take the opportunity to explore and experience this. I have some instructions, if you want, to try this out on your XPProSP2. Therefore, I urge you, my friends and brothers, if you are interested or want to experience upgrade of SQ to another level without installing XPHomeSP1, please send me an email. A New Year gift is awaiting your collection!
Edits: 12/30/11 12/30/11
I've been living with the in-ability to easily switch between cMP-mode & XP-mode since first starting to use cMP 3 years ago or so. And since I've learned to get around that limitation, I have sorta forgotten that it was a problem caused by use of SP1.
I originally asked for help in this sub-thread about the time of cPlay 2.0b18: http://www.audioasylum.com/cgi/t.mpl?f=pcaudio&m=45174
There's a lot of extraneous content in that sub-thread, but bottom-line is that the suggestions did not help get me out of cMP-mode. Of course, I didn't realize then the issue was mode-switching not playing well with SP1.
Grant had a similar experience over a year later when he was playing with SP1: http://www.audioasylum.com/cgi/t.mpl?f=pcaudio&m=78335
The suggestions that he received also didn't help, I did respond to him with how I get around the limitations of being stuck in cMP-mode.
I also had some direct communications with cics on this. First, he pointed me to the registry entry that was changed to switch the modes:
" At the key 'HKEY_LOCAL_MACHINE\\SOFTWARE\\Microsoft\\Windows NT\\CurrentVersion\\Winlogon\'
/v Shell /t REG_SZ /d Explorer.exe
That is, at above key, the value for "shell" string is Explorer.exe for XP Mode."
But just changing this did not switch modes.
Cics also suggested "From Process Explorer, try resuming winlogon & smss processes. Then only shutdown."
Doing this, then selecting 'XP-mode' or 'cMP-mode' in cMP would sometimes switch it. I never spent enough time to totally troubleshoot and make it work consistently as using the 'Run' command from Task Manager to bring up 'Explorer' or 'Regedit' would suffice for about anything I wanted to do. The main exception to that is that the Juli@ control panel GUI only seems to be accessable when Windows is the shell... At least I haven't found another way to get into that GUI. But it hasn't been an issue for me so far, so I haven't pursued it further.
Jack, I am very intrigued by the SQ improvement you've reported using XP HomeSP1 and before I setup an installation of that, I'd really like to try your registry tweaks if you'll send them to me... I'll send you an email requesting via the normal channels.
Then to throw a couple more wrenches into the mix... Discussing this and a few other things via phone with Grant, he reminded me that SP1 was the first upgrade of Windows XP and that going to the original installation might provide further benefits. Now personally, I'm a little cautious with that because I remember SP1 making WinXP fairly stable and a lot more bug-free. But OTOH, I suspect most of the bugs fixed won't bother an optimized cMP setup even without the extreme slimming-down, so it might be another good direction in looking for the ultimate in SQ from windows.
And of course, if you can get access to them, the pre-release Beta versions of XP might even be better.
And in that vein, I don't know if any of you saw this thread on DIYAudio.com discussing making your own SATA cable and SQ improvements from doing that:
http://www.diyaudio.com/forums/pc-based/202369-hand-made-sata-cable-cat.html
In that thread, a post by Wombat from Germany says:
"Don´t waste your time with Seriel ATA!
The best sound without a doubt you get from some well engineered IDE drives. German senior audiophiles have heard this as 3 times more dramatic improvement!
Most hunted for are well matured Samsung 320GB Spinpoints drives jumpered as SLAVE, running on an Intel ICH2 IDE Controller rev. B0. Beware! Later revisions sound edgy!
Finetuning with endless listening sessions for the best 40 pole cable are still in progress but you can expect a real winner. Rumors intensify around some UDMA/33 IDE cables delivered by a still *secret* Taiwanese manufacturer.
All this you best pair with the Intel chipset beta hotfix driver released at the end of January 2003. We talk about Windows 98SE of cause, all other OS´s are way to overloaded with all their tasks to ever be in the audiophile league we talk about here!
More surprises to be announced soon here!"
So the ultimate Ops System is Windows 98SE?
GACK!!!!!
Realistically, while I have copies of that around still, I doubt cMP or cPlay (or most other players around) will run on Windows 98!
Greg in Mississippi
P.S. Ultimately, if sticking with SP1 or even going back to the original WinXP distribution is deemed to be the hot ticket for SQ, I suspect we could spend some of our collective brain-power and figure out how to fix the 'stuck in cMP-mode' problem for SP1 and earlier.
Everything matters!
as I'm using this device. My 120mb SSD as boot device does not sound as good as this small DOM(Thanks to Mihaylov!!), so I have to attach the big SSD as a secondary. And I'm moving over to network, as the size and quality of HDD/SSD won't affect SQ too much during data transfer over the network. I have a WD Book Live with 2TB HDD. I can't tell which config sounds better!
Thank for your advice on SP 1 startup. Never mind, I can live with it.
Regarding IDE HDD sounding better than SATA, I think we can't just look at the drive alone, rather it should a bird eyes view on the overall setup. I will no doubt IDE is better sounding, sure it was, at one stage when all SATA drives were using IDE-SATA convertor chips. They were erroneous in data transfer. In fact, I have used IDE drive for CMP in the early stage, but I didn't find them sounding better than SATA ones.
Even if IDE drives would sound better, I'm seriously doubtful if mobos with Intel ICH2 IDE and those old switching regulators would be worth for
use in CMP! No to mention the old CPUs generating high current, heat and noise.
Let the audio guru do what he think is right. We have to compromise and use everyday products. I find this change of hardware giving one motivation, as one switches to a new mobo/cpu and hears the improvement.
I would bored to death to stick with an old mobo, would you not!
but I did not buy yet because at first wanted 128 MB, and already 64МB or even 32МB, and such was not present :).
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2 Ru, GA-H55M-UD2H, i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S FAT32, ESI Juli@, full linear PSU, NAS - WD My Book Live
Edits: 01/01/12 01/01/12
The info on the Memory Depot & Innodisk websites suggest the DOM SATA drives are powered from the SATA data connector, but the info I've been able to dig up on the connector pinout shows:
1 Ground
2 A+ (transmit)
3 A− (transmit)
4 Ground
5 B− (receive)
6 B+ (receive)
7 Ground
— Coding notch
(From here: http://en.wikipedia.org/wiki/Serial_ATA)
No power there. What am I missing?
Thanks!
Greg in Mississippi
Everything matters!
How I realized there are two versions of the SATA DOM:
1. SATA DOM D150S http://www.memorydepot.com/ssd/listcat.asp?catid=satadomD150SV
2. DOM SATA D150 Vertical Built-in Power (pin7) http://www.memorydepot.com/ssd/listcat.asp?catid=satadomD150SVF
"- Does the Disk on Module come with the power Cable?
- All the Disk on Module that requires to get power from the power supply are coming with Power Cable adapter, The USB DOM and 44pin Disk on Module, that get powers from the data connector, doesn't come with any power cable since it is not required. "
(http://www.memorydepot.com/ssd/service.asp).
Dave correctly told about the second version.
A quote from the site just speaks about the first version with power cable. I suppose so. Jack Wong would be able to say exactly.
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2 Ru, GA-H55M-UD2H, i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S FAT32, ESI Juli@, full linear PSU, NAS - WD My Book Live
No power there. What am I missing?
According to the web site (provided you look hard and get lucky with it), there's a +5v power line on Pin 7 with this connection type. I'm not sure how it's done as Pin 7 is normally, as you note, Gnd. But it seems that conventional SATA interface devices are available - see link.
Best
Dave
What a great innovative product! Thanks Jack/Mihaylov.
Who needs to go back and forth?
Just connect the drive to another computer to do what you want to do.
You might be pleasantly surprised by what happens with a minimal installation. There is no comparison to where you are (per Grant) and where this can be with all unnecessary files and folders deleted and corresponding removal of references to these files in the registry.
To each his own. But one wonders how to compare these old WINDOWS versions to the minimal setup when the folks who are experimenting with them (as far as I know) are not comparing them to a fully minimized XP PRO SP2.
I would have thought you would be one of the first to forego convenience for sound quality. I suspect greater gain is there than making one's own SATA cables though that is something worth trying. My concern is where to get the connectors since I don't think cutting off the ends and soldering some new wire inbetween is a good idea. The bad wire is still there and soldered joints can be a problem for digital signals if one has not done it perfectly which is, of course, impossible. We have to do the best we can.
You have this great wire with two bottlemecks.
Now if you want to make it REALLY good you would directly solder the good wire to the HDD and the MB doing without the connectors. That would probably do something worthwhile.
IDE is out of the question for me since I have unsoldered the chip!
I'm a big fan of the work you, Steppe, Jolida, Theob, Grant. JackWong96, and others have been doing in developing the 'slimmed-down' WinXP installation. Honestly, the only reason I haven't applied all of the tweaks to my setup is that I wanted to keep the software side constant while I've been tweaking on the Juli@ digital section to make it easier to hear the differences. Once I get to a good settling point on the Extreme Juli@ mods, I'll go down the slimming-down route. I will say that applying the first set of slimming tweaks to my system (Steppe 1-5, the early Jolidas) made the largest system/software-related improvement in my setup's SQ & was on par to any one of the major power supply upgrades (Antec-> Hybrid/Linear w/PicoPSU, Hybrid/Linear-> Fully Linear, Bus-powered-Juli@-> separate power). Having heard that improvement, I'm clear on the magic that I'll hear when I perform the Slimming-Windows tweaks.
But one assumption I've held (and as far as I know, as yet untested) is that starting with an SP1 or Home system that starts out with a smaller file and installation footprint and sounding better than the typical by-the-instructions SP2 system will result in an even better sounding final setup after doing the Windows-Slimming. This may not be true... A lot of the work done in the slimming effort will minimize the footprint differences between the various versions of WinXP. But the remaining components will likely still be a bit simpler if they come from an SP1 or Home initial install (or earlier). Will it make a difference? Dunno... And honestly, I probably won't take the time to setup that comparison, got too many other interesting areas to explore.
But it was good to hear confirmation from JackWong on the SQ benefits of using an earlier, simpler Windows version over Pro/SP2 or later.
About the link to the SATA cable thread on DIYAudio.com, I really was only pointing out the post on the German audiophiles using Win98 for their audio systems... Makes even the most extreme cMP-Windows-Slimming efforts look a bit mainstream by comparison!
But on the subject of SATA cables & sounds, I'm not sure where I'm at on that. I do wonder about the suggestions from Mihaylov & others about small, simple IDE-connected drives for the OpSys drive, but I suspect that's more a matter of IDE processing vs SATA. These also come in SATA-connected versions, but I got the impression from their comments that IDE had some benefits. But back to the actual SATA cable, I'm open to it making a difference (remember Peter D's comments on that 3 1/2 years back?), but would tend to go with an ultra-short exotic-wire ones like Peter used... And soldered at one end like he did (and you also suggested).
On the connectors, I found some cables in one of my local computer stores that have snap-apart end-housings. So one can pop them open and solder directly to the end-connectors instead of retaining any of the original wire. If you ever care to experiment, if you can't find this locally, I can send you some.
But I think that power to the drives, SATA or IDE, is more important... and when I have a few spare minutes, I've got a 3.3v undervolting SATA power regulator to put in and try on my SSD, inspired by Douwe01NL's posts.
And finally thanks to you for being one of the 'beta-testers' on chip removal. Another great mod direction!
Later!
Greg in Mississippi
P.S. You do have me pegged as a 'hair-shirt' kind of guy with my audio stuff. Two of my not-too-distant past digital sources were a Peter-Daniel-inspired Shigaclone (bare-bones home-built CD drive) and before that a modified Sony PS-1 (not even a track display although I was able to come up with a remote for it!).
Everything matters!
It seems to me that you are trying something analogous to deciding how to voice a preamp using what you know is an amplifer not as good as it can be.
The transparency that results from getting rid of lots of sonic clutter could only make your task easier. I feel certain what your mod's effect would be more clearly apparent with the best cMP installation.
Tell me the brand name of the SATA connectors. I still think the way to mazimize it would be to solder one end directly to the board. I have found connectors to be one of the worst offenders when it comes to acheiving good sound. I am thinking for these frequencies the Allen Wright silver foils might be really good. I see the connectors are far apart on the standard SATA cable which would reduce capacitance at the expense of inductance which does seem kind of strange for a high frequency conductor but is this because the buffer/drivers (?) are not up to driving a higher capactitance? Was this discussed?
Simple is always better until it isn't. My most recent experience with this idea was installing a non-acpi version of XP Pro. I thought it sounded really good especially after it was "minimized". I fought with Junaid who said the acpi version sounded better. I did not want to do another re-install this quickly and I assumed the archaic had to be better. In this case it did not turn out to be so. Of course no rule holds fast for everything and it well could be that a more archaic version of XP Pro or Home could well be better. I would be glad to give it a try.
To close this brings me to what I consider a discontinuity that I cannot resolve - why would someone try the archaic OS without having first tried the minimal Pro since both seem to offer the same level of inconvenience which I know (damn well!) that you are a not a slave to. I know, that like myself, you take pride in making digital audio as difficult to use as a well set-up turntable!
If it were tricky to get to this current end point I would understand your reluctance but it is actually simple to get there. Just connect your drive to another computer and get rid of the files. No need for batch files, just delete all of the files between the ones you need. Just make sure you leave the OS in cMP mode before doing this (which seems to be the case, anyway) Especially if you get rid of EXPLORER which NEEDS to be got rid of.
I tried deleting JULI@ Controller in the registry and it no longer worked. I do not think one can get rid of this.
and I feel sure WOMBAT is being as facetious as a human being can be.
I think I will wait this one out!
I thought I would go ahead and download the GROUP POLICY EDITOR and see that the machine must be connected to the internet to work. I have never connected my music computer to the internet and suspect I will do without AWE.
RYELANDS has mentioned that he does not like what AWE does and I have noticed from numerous installs; a few nLite installs that would not allow AWE (I had gone too far) and I could not honestly say what was missing when AWE was missing, also.
One wonders just how important it really is at this point?
Thanks for another discovery. Just when you think cMP is done ...
Hi Rick,
Your emails were received. I'll send the gifts to your personal email.
Apparently, there are compatibility problems with SP1 and CMP, or perhaps associated with minlogon. I didn't delete any file at all when CMP was installed and I try to reboot.
You can download the program with a working computer, copy to your CMP machine and install therefrom. My gift is for use with XPPro, there is no need for any installation, a few basic steps will do the job. It's magic!
As for AWE, SP1 works well without it, though it is a bit better when enabled. XPHomeSP1 beats XPPro even without AWE.
Looking around for a cheap copy and there does not seem to be such a thing.
I see copies with SP2 selling on EBAY for sixty dollars without the COA which I guess means no key. So what are supposed to do with it without that?
From legitimate retailers it is around seventy eight dollars. For one dollar more you can get PROFESSIONAL. I thought the price difference used to be greater than that.
There is no need to buy one. Please wait for my report.
PRO down to 28 mB?
I doubt any of that stuff us still in use but I have o real way of knowing.
I know my registry is quite small at this point.
PLUS - I do not think I am ready to do it again! (today)
Windows components and utilities in XP home. Even with Nlite, things are written in the registry anyway. I believe one cannot manually trim down XP Pro registry to equal that of XP Home. The comparison I did was fair, both OS underwent the same tweaks and optimization. Assuming the trimming effects on both OS are the same, as they have the same structure and files, in the end one may get a better sounding Windows with XP Home. May be more files can trim down in XP Home too. Who knows the outcome? But it is worth exploring. We have installed Windows hundreds of times recently, would you mind doing one more on this one?
One thing I know, both OS use the same shell32.dll and kernel etc. And from what I saw during resource hacking, XP Home is not going to use most of the dialogues and menus etc, therefore perhaps we can delete more things without adversely affecting on XP Home.
The "size" of the registry never changes. The space that is required to build the registry at installation seems to remain no matter how much stuff you remove. It can grow but does not seem to shrink.
As it is, the XP Home registry is only about 200 kB smaller than what my folder contains but there is no comparison with what is inside mine and what is inside the XP Home config folder.
In fact, it looks like none of the STEPPE registry mods have been done.
Beyond that there is Mihaylov's suggestions per his screenshots.
I have gone beyond that with lots of trips up and downstairs to see if it still works and, if it does, taking out more.
What was sent does not look much different than what I had after the nLite minimal install. Not to say this basic registry is not a good starting point but there is much work to do to get this into top condition for listening to music.
There is so much extraneous audio and video stuff left in there.
I have found the best approach is to as closely as you can (or dare) have the registry match what .dlls, sys's, and exe's are left in your WINDOWS. Should not have left out the .nls's.
I would be glad to send my config folder as a guide to anyone who would like to see where it is at the moment. I doubt it will work, as is, but I feel certain if you made your config folder look like mine it will work.
Now this is with the understanding that this is a registry based on a FINAL STEPPE machine with no Explorer. All of the files removed per his lists (and a few more). Using the Gigabyte H55 board, PS/2 mouse.
LIST OF REMAINING FILES
config folder
drivers folder
advapi32.dll system32
authz.dll system32
basesvr.dll system32
bootvid.dll system32
c_1252nls system32
c_437.nls system32
comctl32.dll system32
comdlg32.dll system32
cryptdll.dll system32
csrsrv.dll system32
csrss.exe system32
ctype.nls system32
dnsapi.dll system32
FNTCACHE.DAT system32
framebuf.dll system32
gdi32 system32
hal.dll system32
juliaasio.dll system32
kbdus.dll system32
kdcom.dll system32
kernel32.dll system32
l_intl.nls system32
locale.nls system32
lsasrv.dll system32
lsass.exe system32
mpr.dll system32
msasn1.dll system32
msprivs.dll system32
msvcp60.dll system32
msvcrt.dll system32
ncobjapi.dll system32
netapi32.dll system32
ntdll.dll system32
ntdsapi.dll system32
ntoskrnl.exe system32
ole32.dll system32
oleaut32.dll system32
oledlg.dll system32
olepro32.dll system32
rpcrt4.dll system32
samlib.dll system32
samsrv.dll system32
scesrv.dll system32
secur32.dll system32
services.exe system32
setupapi.dll system32
shell32.dll system32
shlwapi.dll system32
shutdown.exe system32
smss.exe system32
sortkey.nls system32
sorttbls.nls system32
svchost.exe system32
sxs.dll system32
umpnpmgr.dll system32
unicode.nls system32
user32.dll system32
userenv.dll system32
vga256.dll system32
vga.dll system32
watchdog.sys system32
win32k.sys system32
winlogon.exe system32
winmm.dll system32
winspool.drv system32
winsrv.dll system32
winsta.dll system32
WLDAP32.DLL system32
ws2_32.dll system32
ws2help.dll system32
acpi driver
atapi driver
classpnp driver
disk driver
dxapi driver
dxg driver
dxgthk driver
ftdisk driver
hidclass driver
hidparse driver
hidusb driver
i8042prt driver
Jula driver
kbdhid driver
ks.sys driver
ksecdd.sys driver
mouclass.sys driver
mountmgr.sys driver
npfs.sys driver
ntfs.sys driver
partmgr.sys driver
pci.sys driver
pciide.sys driver
pciidex.sys driver
vga.sys driver
videoprt.sys driver
volsnap.sys driver
wdmaud.sys driver
wmlib.sys driver
FONTS
8513fix
8514oem
8514sys
arialbd
l_10646
marlett
micross
tahoma
tahomabd
trebuc
trebucbd
vga850
vga860
vga863
vga865
vgafix
vgaoem
vgasys
WINDOWS
fonts folder
system32 folder
win.ini keeps showing back up!
Documents and Settings
Program Files
RECYCLER
System Volume Information
WINDOWS
boot.ini
NTDETECT.COM
ntdlr
There is the problem with drives being different and depending upon how much you have removed or hacked you might not be able to make it work.
I have found when needing to reconfigure the drives within cMP if you will replace the hacked shell32 with the standard one you can make the changes.
Jack,
I LOVE your lateral thinking! Great direction!
One thing you didn't mention was if you tried to get Minlogon to work with XP Home. While the conventional wisdom is that it shouldn't work, the way we do it (with registry & file hacks) might work. If you've got a good backup of your XP Home setup, it's worth a try.
Back in the day, I did quite a few XP Home-> Pro upgrades, mainly to use the PC's on a more sophisticated network. Since we generally AREN'T using domains & such and do little account management, these features are useless to us. BTW, watching those upgrades, they end up replacing a large portion of the Windows files.
In a similar vein, I've stuck with XP Pro version SP1 for my cMP. Unless you are doing advanced networking (and likely just wireless networking) and/or using USB (as I remember, there was significant upgrading of USB functionality with SP2), then SP1 should result in a smaller and likely cleaner (and hopefully better sounding) Windows. But honestly, I've never compared.
It's funny... with the development of cPlay halted, one might think cMP is in it's twilight. But with all the Windows-hacking that's being developed, it's like a new and better setup all over again! While I don't think we'll be able to get it down to the same footprint as a clean Linex-based player (or firmware), the flexibility of being able to use a wide variety of hardware (MoBos, processors, soundcards, HDDs/SSDs, etc) and configurations (networked, headless, stock, etc) due to the flexibility of Windows and availability of drivers is just too beneficial to pass up!
Again, THANKS! And also, belated thanks to all the slimming Windows pioneers & brave beta-testers!
Greg in Mississippi
P.S. I should put the latest evolution of my Extreme Juli@-modded digital section into the system today for break-in and audition. It features very tightly-coupled power-bypassing upgrades and a combination of organic-polymer (Oscon-type, but not all Oscon-branded), low-inductance high-value ceramic (on the deep-digital processing chips ONLY) and SMD film capacitors (PPS film, also low-inductance/high-bandwidth). I'll post an update in the 'Further Juli@ Follies' sub-thread in the next several days.
Next steps after this one are to try undervolting (especially the 3.3v rail) and clock upgrades (including a try at replacing the 24.576Mhz clock with a low-jitter 49.152Mhz)!
Everything matters!
if you tried to get Minlogon to work with XP Home
I suspect minlogon will work fine with XP Home, it's AWE that is available only in XP Pro. (I find the sound in my system better with the feature disabled.)
Unless you are doing advanced networking (and likely just wireless networking) and/or using USB
Networking should also be fine in that as much as possible is switched off. The USB audio side I can't comment on but proprietary drivers typically by-pass the MS stuff anyway.
I look forward to Jack's report - as you say, it's a clever idea.
I admire your insight on the use of SP1. Simple is the best for CMP. I will try to compare the sonic differences between service packs and w/o it.
Can't wait to see your development on Juli@ modifications. Because I'm building a second machine using Juli@ and network. I'm an eager audience waiting for your update.
In addition, I've just bought a few cheap ESI MAYA44 @$22USD each. I want to use them as alternative to Juli@, as I believe they share similar digital section.
Now to find an old copy of the thing laying around the office!
That's cool. I presume nLite does take off most of the stuff from the pro edition though, making it as simple as it could get. Never tried an nLite xp home edition. Good to see u found it interesting...
Junaid
Hi Junaid,
Do you remember what you included/excluded in your minimal nLite?
In my setup (dualcore atom) your minimal nLite recognizes two cores, but not the hyperthreading. Despite of this it is very musical (see The Sining Atom in the HW-forum (http://www.audioasylum.com/forums/pcaudio/messages/10/100814.html).
I want to try to make a new minimal&musical install which can use two cores and hyperthreading (virtual 4 cores). Much thanks in advance!
Douwe
Email me. Will reply with the details, as I quite don't remember what I left in. It's been a while since I looked back at my nLite ini file. Two things I quite remember leaving in are MultiProcessor support & Internet Explorer Core.
Junaid
Hello, hope you're enjoying holiday season!
I've been following most of the tweaks by amazing inmates, especially Steppe, Mihaylov, and Jolida.
I'm currently using a laptop and usb digital-to-digital transport to play music (completed all modes up to resource hacker tweaks).
Recently, I've found some more switches for boot.ini file.
They are,
/USE8254 ( use8254 timer chip as its base timer),
/MAXPROCSPERCLUSTER= ( MAXPROCSPERCLUSTER : max cpus to use in cluster),
/INTAFFINITY ( INTAFFINITY : sets interrupt affinity).
(there are explanations on parameters)
http://www.cspv.nl/info/boot_ini.htm
I took a different approach for tweaking boot.ini.
1. When I apply /maxmem=256, it seems music files are divided into small pieces, (2 or 3 spilts for 3-4 mins song). And there is abrupt mute for a half second during music playback. I guess it's where a music file is spilt and cPlay's loading the other parts into the ram (or it could be converting left over parts of flac file to wav pcm).
2. Also, I don't want to leave out computing power (not so much for atom processor though), which is actually benefiting the way the processor was built - using two cores for more stable performance. I don't use any upsampling method.
3. While, I keep things simple as possible, which is the main goal of most tweaks so far.
So, I took out /onecpu parameter and raised /maxmem size, not to have many spilt parts (I have 1GB ram installed physically).
I also replaced /usepmtimer (ACPI controller's) with simple 8254 Programmable Interval Timer /use8254, which is about 30 yrs. old :).
(I'm not sure whether XP supports HPET, but my laptop doesn't support HPET in BIOS and I deleted HPET from hardware manager long ago.)
/INTAFFINITY for directing interrupts to CPU 1. Process Explorer tells me that cPlay takes entire CPU 0 during playback. (So, why would I interfere CPU 0?)
My parameter list is the following
/FASTDETECT /noexecute=alwaysoff /CLKLVL /timeres=9766 /nodebug /pcilock /use8254 /nopae /noguiboot /maxmem=512 /MAXPROCSPERCLUSTER=4 /INTAFFINITY /minint
I'm doubtful what /MAXPROCSPERCLUSTER= actually does. But, I set it to be 4 for more organic sound.
It would be much appreciated if someone tries and gives us feedback!
Merry Christmas!
I do not experience any pause during playback using the maxmem=256 switch. I am using a 1gb ram too. I have tried various combinations in the boot.ini file & I liked the one suggested by Serge (except the onecpu switch). Don't know why but it did not make an improvement in my case. The maxmem & pmTimer did improve the sonics, especially in the detail & liquidity of the mids, & also separation became better. I will try the ones suggested by u above & post a feedback.
At the moment, I'm at nLite with just the Registry tweaks done, along with the boot.ini mentioned above. I love the way this sounds as it is.
Tweaking the shell32 using reshack or Restorator served like fine-tuning my set-up. Though at one point while deleting quite a bit of stuff from the shell32 ruined the bass. So I guess it has to be done in stages & analysed carefully.
I strongly believe its all about getting the correct Balance in Ur system, more than reducing the size of windows or it's files.
The sound behaves strange while tweaking certain files so I am doing a lot of listening at every single stage, so as to not slip in the wrong direction...
And I'm glad that Rick finally agrees with me that nLite with Acpi does sound a lot better :)
Junaid
I'll give some of these parameter changes a try, but first:
I do not experience the half-second or any interruptions/mutes using maxmem=256 on my laptop. I have not seen any report from any one else about experiencing this either. You may well have some other issue to resolve.
and it doesn't come back.
Something was said to this effect, that the registry would rebuild itself.
Hasn't yet.
Whether that has anything to do with adding the switch /minint (per Mihaylov) I have no idea. But that did allow deleting setupapi.log and the .logs within config and they stay away. I have noticed that those config logs will come back after working the registry through LOAD HIVE in my "other" computer which is interesting but not surprising.
I deleted everything that could be deleted. There are some items within CLASS that even with asking for every kind of permission will not allow deleteion. Same with ROOT within ENUM. But everything else can be made to go away.
Whether this has resulted in sonic nirvana I cannot tell yet. If there is a change it is subtle but empirically to me it has to be better (doesn't it?).
Serge you said: '...Take care to hack win32k correctly...' What does correctly specifically mean?
"Twas the night before Christmas and all through the house
not a creature was stirring not even a mouse
(thats because we deleted the drivers in Steppe 16...)
So much for my lame joke...guess it does not translate well.
Theo! And Everybody! Wish You All The Best, Inmates!
You can see in the right top corner of the pic the piece of a screenshot that shows win32k.sys. It's a mistake. What I mean is, certainly, winsrv with the described changes - only fontdir and fonts resource remains. Sorry. A typo.
Serge
Edits: 12/25/11
with setup/api? I have deleted it "ages" ago see the pic with all such files found on my 3 hdds. It went off at part 19 and never came back.
Also I have loaded system and software hives and deleted CS02, got a 1,5 mb hive and the Lynx security cab won't let me start cplay. The thing that causes the registry to rebuild uses cset 02 ar a refference set, seemingly. I believe it may be this:
HKLM-000 (the loaded hive) -microsoft-windows-current version-reliability-ShutdownDoStateSnapshot the value is (2).
You must be reading my thoughts. I have posted a similiar thing yesterday, but deleted as I wanted to make sure once again, trying to delete it different ways. The system crashed after all. Rebuilding now for the third time in 24 hours. I have, btw the same problem with shell32 after hacking, but I don't care. Files that Serge Mihaylove said he'd failed to hack, are easily hacked with the same program, namely hall 130kb, ntoskernel - 1,98 mb, kernel32.dll - 553 kb, advapi.dll - 496 kb. The trick is to have a /noguiboot in your boot.ini.
I have hacked also all *.sys drivers - the are just the same DLLS.
Serge.
I envy your prowess with soldering things...
Edits: 12/24/11
You said: '...ALSO Kernel files contain background bitmap THAT ARE USED by logonui, so they can be removed only with /noguiboot option, there is also rle animation code and winlogo rle inside and a lot A LOT of text. Stripped from this sh*t the sound of ks, and kernel-hal files is akin to a minlogon tweak in it;s power...'
By kernel files do you mean only kernel32, ntoskrnl, hall?
And just to confirm you delete everything accessible in these, correct?
setupapi.log would always return after I deleted until I used Mihaylov's suggestion of /minint. So would all of the .log files in config. I am surprised you did not have this happen.
Your RESOURCE HACKER files seem impossibly small in some cases. I deleted everything with the exception of VERSION and those entries with all of the numbers and mine is 100kB larger than yours. I tried /noguiboot on my "other" machine where I modify my cMP install and this makes no difference here as to what one can do with RESOURCE HACKER. Did I misunderstand what you meant? Also, are you deleting EVERYTHING from these files? Even the entries with all of the numbers?
I did not realize until your note that one can "hack" the .sys files. They do not come up when you ask for the drivers folder. Today I looked and see there is no way they will come up when the default file types are .dll and .exe. A HA!!!
So tell us more of what you are taking away.
And I no longer think shell32 is responsible for my slow start up (I assumed you were referring to that). I had to reinsert the full sized shell32 to get cMP to recognize extra hard drives and it was just as slow with that. I must assume my assault on the default file is responsible for this now. I left almost nothing.
Next week I will try to figure out how to do a screen shot and show my registry - default, software, and system.
Greg Stewart was wanting to get rid of the JULI@ controller and this is easily done within the registry.
It seems I haven't shown the a-n files here they are. NO SETUPAPI. I have a folder Hacked windows, that I use to poulate EMPTY system 32 folder from another machine.
I install nLited (my own version of) vl windows with no OOBE, install cmp-cplay, awe it, minlogon it, run registry bats around, run registry mechanic and while it's opened I run deletion bats. Then I reboot, enter cmp, set it to start cmp, reboot but from another hdd. Exterminate all remaining folders and files, except System32\config, populate with files from hacked windows. Voila!
I want now to redo everything, because not all registry tweaks were done by the bats. Service group order and such, Media, media properties smth, device ennumerations, that I remember only visually...
Will report.
Serge.
USER32.DLL that causes the disappearance in cmp restart yes-no button and a slow start. I have worked with it long, even tried to get the cmp buttons back by reimporting resource, but it won't so, be it.
ALSO Kernel files contain background bitmap THAT ARE USED by logonui, so they can be removed only with /noguiboot option, there is also rle animation code and winlogo rle inside and a lot A LOT of text. Stripped from this sh*t the sound of ks, and kernel-hal files is akin to a minlogon tweak in it;s power.
Serge.
P.S. I remove EVEN VERSION. my files are naked, just the way I like mermaids to appear.
...and it would not boot or should I say it just hung for 30 seconds during the boot process. So I retreated. Is this boot delay normal? If yes how long does it last?I did add noguiboot to the boot.ini file but at the beginning and not the end of the list.
Edits: 12/27/11
not sure how our files compare but as I mentioned to Serge, my machine, after leaving the BIOS screen will first give you the LOADING OS screen then go black, it blinks a couple of times and then after 25 seconds or so I get the blue screen, about five seconds later the cursor appears and then momentarily later the cMP screen.
The first time I thought for sure it was not going to work. I walked away for a moment and came back and THERE IT WAS!.
Ok I will retry. Thanks rick
...but for only kernel32 bitmap hacking. I could not get advapi and ntoskrnl hacked versions to boot.I did get all bitmap stuff out of ntoskrnl but only that.
Good sq pop.
Edits: 12/27/11
in all files, even the entries with all of the numbers. I only left untouched entries in the following:
advapi.dll - RCDATA - 102 as I don't know what is this mumbo-jumbo
user32.dll -Cursor - 52-1033 and 54-1033 and cursor group - 100-1033 and 102-1033
winsrv.dll - FONTDIR-1033 and fonts 1001-1033 1002-1033
BTW ks.sys is 137 kb.
See the pic of all system32 folder with all files hacked and their corresponding sizes. Take care to hack win32k correctly. Also on the pic expanded completely. (top right.
Serge.
I am not hiding anything. To tell the truth, I was scared at first to do it, waiting for smb to be so kind as to send me already hacked files. Nobody did. So... Also about Mihaylov's idea to hack the fonts IS GREAT. The fonts ARE residing in memory always. I know it from Adobe Illustrator and CorelDraw I work with. Why do you write it's too expensive? You mean effort or the price of the program? As for the price of the program, I can do it myself and you will get the files.
Serge.
1st the bad news I was not able to replicate Steppe's strategy to delete (almost) everything. Every attempt resulted in a no boot or gross instability (boot screen goes crazy). I am talking kernel32, ntoskrnl, advapi32 and shell32. So if any of you are going to attempt to strip these files have Snapshot or SS equivalent handy because you will use it. The only thing that worked for me was Jack's recommendation to eliminate all menu stuff in ntoskrnl and all kernel32 bitmap stuff. Good news is that this is zero short of fantastic ... sound quality wise. The soundstage is enormous, bass is real, highs pristine. Airy, airy and airy. I have not yet attempted user32 stuff (per Jack) but that is next. WRT shell32 I have bottomed out at 2.497 mb.
So as noted above I was able to get shell32 down to a 2.497 mb level but realized I still had all the Dialogue stuff still intact. Recall that initially while hacking shell32 I took out all the Dialogue stuff (and it booted fine) but I lost the function to ADD / REMOVE folders from the cmp screen. So with Mihaylov's tireless effort to help he recommended reinstalling Dialogue 1079. Ok so now after reinstalling 1079 (along with all the Dialogues) but after other shell32 hacking I realized I had an opportunity to go back & hack away at all the Dialogue (except of course 1079) so I did. And it would not boot. Almost as if these things were interacting and some combinations work and some don't. So its really a crap shoot when you get into these system32 hacks.
Edits: 12/30/11
I would greatly appreciate a gift of minimized FONTS!!!!
That program was seventy euros.
Now to your RESOURCE HACKING comments - my WIN32K already looks like your screenshot before I do anything so I am not sure what you are recommending to do.
I just went through and deleted almost everything out of my files and got a HARD ERROR at start up. Restored back to where I was before and just played with ntoskrnl and WIN32K by themselves and got the same message. I deleted the unneeded language stuff from WIN32K on the compiler side and this must not be how to do it. I will keep trying.
I am in the midst of concetrating on registry so I think I will hold off on this until I am satisfied with where that goes.
Finaly got my OS installation to where it was before doing the removal, and with the new board having to settle in, I am able to hear/not hear what this is doing.
I was expecting a tonal change, even though Mr. Wong made it clear in his posts that this was about extraneous noise in the background.
This morning I am able to appreciate what this does. There is great increase in nothingness between sounds and a seemingly for the moment disappearance of hash within sounds. Of course, the ear is relentless and will become sensitive to some new gremlin in time but for the moment this is perceived as unimproveable.
While I did not find it as difficult as I expected to be it is still scary screwing around with what could end up being one hundred dollars of junk.
For those who might be interested here are some notes of how I did it.
I initially tried to unsolder with the heat gun the connectors at the back fo the board. I removed all but the monitor jack and the PS/2/USB jacks to the right of this. I knew I would never need them and the space this gives made it easier to work.
Since heat wasn't getting them to move I forcefully took them apart. This is crude as it can be but I worried the massive heat needed to remove them elegantly would likely take a toll on the board. Just kept tearing away stuff till I got near the MB and then cut the wires to the traces (most of them are wires that go "up" into the connector assembly.
After this I did use my soldering iron (turned up very high) to clean off the traces. Many of the traces simply disappear. (I did this with all of the devices removed.)
Then to the heat gun. I did find one with a nozzle 0.75 inches in diameter. I set the heat as high as it would go and used the low fan speed. I removed them in the order Mr. Wong recommended This post is linked below.
Sure enough the sound chip actually POPPED off of the board. Made noise and leaped into the air. The others required picking up to get out of the way. I made a heatsink to protect capacitors when they were close with some copper foil, the kind they make crossover inductors with.
I did use the soldering iron to remove the MOSFETS. High heat was required.
Removing the jacks took quite awhile so I did the component removal the next day. I think I needed the time to recover (!) and to plot my course of where I would point the gun. While doing the jack removal you get plenty of time to burn the board's image into your brain which allows you to meditate on it for the next 24 hours. Having never done anything quite like this I wanted to be confident. This is nothing like woring with the typical through-hole PCBs I have worked with for years.
Since it was a new board, though basically the same as my previous MB, it was a later revision which made it different enough that my OS install was not allowed. I had done lots of work on the previous install thinking it would "bolt right in". Oh, well, got to do it again and I tell myself I do it better each time.
I have done so many lately that I think the process has become a transcendental meditation mantra, though a very long one. Only the registry slimming gets old and RESOURCE HACKING; modifying all of the .dlls can get old. Of course, some of what I did was likely superfluous but one figures why not?
With my seriously slimmed .dlls there is a 30 seconds delay between the message LOADING OPERATING SYSTEM and the appearance of the cMP screen. Slightly inconvenient and who knows what I deleted to get that to happen. I can live with it easily.
Also, when you want to turn off the computer or restart the little screen has the buttons but no writing. I think I deteted almost all of the STRING stuff.
SO, it is very worthwhile to do this and while tricky not daunting if you have good soldering skills.
Thanks again to Jack Wong for his ingenuity.
You are thorough. Did you also remove the com port ic and small transistors of the USB ports? And the crystal for the sound chip. The last thing, if you dare, the 1394 controller. I'm just joking, don't take this seriously.
If there is something to be gained I am ready to give it a try if you would be kind and show me where these things are.
One can't be satisfied once they know of the existence of the next step.
I hope you are serious.
friends! I don't want to persuade anyone to try something risky like I did. That's all.
I've been working last night and today auditioning CMP on XP Home Edition. Will post soon on this.
I'll show you the components to be removed, but be reminded that it's more risky this time.
There has never been a doubt in my mind that you are as SERIOUS MINDED about audio as one can be.
No one could accuse you of baiting anyone into taking your advice.
But for those of us wanting a challenge ...
In [HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Control\FileSystem]"NtfsDisable8dot3NameCreation" Set DWORD=1
"NtfsAllowExtendedCharacterIn8dot3Name" Set DWROD=0
Faster loading of cue/wave files. 8.3 filenames (short filename) is a filename convention used by old versions of DOS. I'm sure we don't need this.
Edits: 12/24/11
And I can't try it because I use FAT32 for the system drive.
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2 Ru, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Liv
disable Windows checking Fat32 integrity?
I use fastfat.sys instead of ntfs.sys in drivers folder. That's all.
I did not disable "Windows checking Fat32 integrity", I don't know what's this.
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2 Ru, GA-H55M-UD2H, i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S FAT32, ESI Juli@, full linear PSU, NAS - WD My Book Live
Anyone tried these Registry settings
I'd disabled eight-dot-three filenames long ago but I tried your second suggestion earlier today. The effect is to make Explorer (which I use as a music browser) much more responsive when opening folders on my music server: any with large numbers of sub-folders (> 100) could take several seconds to open. It could be slightly irritating.
Why it should speed things up given that "AllowExtendedCharacterIn8dot3Name" merely allows you to use an extended character set when using a feature I don't use, I cannot even guess - but it does. Many thanks.
and what happens with longer file names?
I'm sure you can apply the registry changes without missing a single file file. 8dot3 filenames are those used by old versions of MS DOS. It's there so that NTFS is backward compatible with legacy products. I put it wrong in my post, causing misunderstanding.
I tried and indeed cmp opens files quicker. And when you are listening to one track and you scroll fwd a bit, it does it much quicker. So in that regard a nice benefit.
What surprised me though was the sq pop. Everything got stronger, you can hear more into the music (inner detail improvement). Seems a bit louder. Clearer all around ... at least in my system. Maybe not as good as a skinnied down shell32 but sq moves in that type of direction.
A nice unexpected one Jack!
I will when I get home. Have you tried it? If so what was your reaction?
In normal Windows, folders with large numbers of files opened much faster, including individual folders. The same for CMP mode, faster loading of cue and wave files.
While I believe all the Steppe/Mihaylov tweaks have directionally made for purer music (I will never go back) I have to admit the slimming of shell32.dll via resource hacker, to me, exceeded the improvement made by going from a 1 gb windows to a 40 mb windows. Other certain registry tweaks(especially those asociated with irq, polling...) are also very very good. But slimming via resource hacker ( and to a lesser extent deletion of audio codecs) are the winners of the year. And remember Mario would never had thought of resource hacker stuff w/o Steppe's and Mihaylov's 'five month slog through the details' effort. Maybe this aligns with Mark's theory of where to concentrate.This leads me to believe the next level of tweaks are probably some 'out of right field' type of thing. Maybe something we're not even working on or are aware of yet.
Edits: 12/22/11
A friend of mine uses a dual-boot Atom board (standard cMP^2 config on the second install, minlogon, AWE but no "Slimming") with an Asus Xonar soundcard (as recommended here, big success, no tweaks, internal DAC). He has has good (Spendor) speakers and a Linn pre-amp & amp but limited interest in the technical side of the project.
I sent him a slimmed-down version of shell32.dll this morning along with instructions on how to replace the version on his cMP install and on deleting logonui.exe. Without, for obvious reasons, telling him what to expect, I asked him to report at his leisure if he hears any change. Trust me, if he heard nothing, he would thoroughly enjoy telling me. Here is what he has just sent me:I began rehearsing my words about how difficult it is to come to a conclusion about hi-fi, comparisons etc. As soon as I sat down to listen though - I chose what I thought was a fairly modern recording, Larry Willis [The Big Push] (turns out to be 2005) - all that went out the window.It's not proof of anything, of course, but I do think it is encouraging anecdotal evidence in support of The Mermaid Project, given that they are such simple tweaks to do.
Sounded like a big advance to me - indeed, I don't think I'm exaggerating to say the best hi-fi I've heard. It is difficult to imagine how it can be improved upon anyway. Same with the description - I usually use "clarity" and "detail" - but it seems to me there is only one way to describe it - as though they were in the room, playing. It sounded real.
I tried a lesser quality FLAC recording - Lars Gullin with Chet Baker from 1955-56 - and sure enough that would need a comparison of the two systems to tell. Switching back to ordinary boot, I played Willis again and it is good I think - just did not sound as real. With the new bit, I get a stronger stereo picture too, which is perhaps surprising considering my set-up.
So very pleased indeed. Thank you. A great stride forward.
Dave
"I sent him a slimmed-down version of shell32.dll this morning along with instructions on how to replace the version on his cMP install and on deleting logonui.exe."
Can somebody tell me how to find slimmed-down version of shell32.dll and the instructions? I cannot find anywhere to download. Thanks.
Can somebody tell me how to find slimmed-down version of shell32.dll
Here's how I do it.
1. Download "Resource Hacker" - see link - and install it on your desktop box.
2. Delete resources from the file. Take a few moments to get the hang of the program; it quits w/o saving or warning unless you go to File > Save. You cannot edit the files that matter on the cMP^2 box as XP won't allow you access. The easiest thing to do is to copy the target file from \System32 on your desktop to somewhere else, edit it there then copy the "slimmed" file to the cMP^2 OS partition by moving its drive to the desktop box.
3. If you e-mail me, I'll happily send you a known-good, UK English copy but, as always, YMMV.
4. Be warned, this is addictive. On this ward of the Asylum there are cases in serious need of extra treatment.
HTH
Dave
Thanks for the files. I used resource hacker to slim my shell32.dll. Problem is that I need to use "Explore" button in CMP to copy files (e.g. copy music files from USB memory stick to the harddisk). With the slim shell32.dll, the file explorer is not working properly. That make me hesitate to go for the hacked version.
Any suggestion?
With the slim shell32.dll, the file explorer is not working properly.
The version of shell32.dll I sent you works fine with Explorer at this end (though some of the icons look a bit funny). I got some loss of functionality - e.g. batch files that used to work now don't - but the improved sound quality makes that a small price to pay. It is, at least in my setup, a quite extraordinary difference.
FWIW, my suggestion would be to delete a few resources at a time and try the file at intervals to see what resources you need to keep and to find a compromise between "slimmest of the slim" and what is, for you, a viable setup.
That said, there are others here better informed about this process than I am.
HTH
Dave
Dave,
Thanks for your suggestion. I'm making deletion bit by bit. First AVI and bitmap are quite safe. Will work on icons tomorrow.
Barry
I finally deleted icon group by hand and keep around 20 icon groups which looks familiar in file browser to enable copying file which is all I needed. The size is now 3,542KB.
It sounds great to me, vocal is now more natural and organic. I used to have problem playing a 24/192 very complex record. Now I can very comfortably play that record and the instruments never mess up again.
Highly recommended. Thanks Dave.
Hi Barry
Can you share which icon group you kept in the deletion?
Regards
Its all in the link
= Sounded like a big advance to me - indeed, I don't think I'm exaggerating to say the best hi-fi I've heard. It is difficult to imagine how it can be improved upon anyway. Same with the description - I usually use "clarity" and "detail" - but it seems to me there is only one way to describe it - as though they were in the room, playing. It sounded real. =Strange. But I would write the same word for word in describing my impressions from the sound of my system now after all the tweaks.
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2 Ru, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Liv
Edits: 12/23/11
that again at least for me were fantastic tweaks:
..using ferrites and pi filters on all the pc powerlines (mark's idea). Does great stuff for noise reduction, increased dynamics and soundspace.
..using a small 8 gb ssd for os
..not discussed much but real cool tweaks involved using a mechanism to load music files onto my ssd for playback rather than leaving them on hdd's (ryeland's idea). As a simplification/improvement of this process in terms of time to load files is (again a ryelands idea) to load just the cue files onto the ssd for playback. This last idea requires that you define the absolute path to the music file in the cue. I really liked this last one it sounds great, and is very fast(loading cue and music onto ssd takes anywhere from 10-30 seconds). Almost as good sounding as loading cue and file onto ssd. Once I skinnied down windows I no longer have this ability...too bad.
Hi Theo,
IMO you choose the right moment too reflext on the XP OS-software tweaks from this year.
It contributes too Bibo01 call for getting organized and also might contribute too a better insight in what is effective for better SQ and what is not effective for better SQ.
Ranking the XP OS-software tweaks as I have implemented in my setup I get this list as ‘must do’ tweaks for any (!) cMP user, because the software tweaks listed below have not wrecked my cMP system. With these tweaks my cMP still functions as a computer with a working (wired) LAN connection too access a NAS and a internet-router.
My ‘non-cMP-wrecking’ XP software tweak-list:
* disabling the audio-codecs in the device-manager
* disabling the Watchdog feature in the registry
* disabling the Performance Counters with the exctrlst.exe tool and in the registry
* slimming the shell32.dll (in my setup from 7,99 Mb to 4,87 mB) with help of the Resource Hacker program
I find it difficult too rank these tweaks because all give a real substantial SQ improvement. Each individual tweak listed above can be compared too a SQ improvement as when implementing ASIO or WASAPI. Some even gave a bigger SQ improvement.
Disabling the USB polling frequency as suggested by Ryelands also gave a worthwhile SQ improvement which gave rise too abandon use of USB all together. I disabled USB in the Bios and moved too a headless setup with remote control through use of the VNC protocol and an Android Smartphone / tablet as remote.
Reflecting on the XP software tweaks for my cMP in 2011 also learned that using LAN in my setup does not harm SQ in any way. My setup lets me hear every minute (!) software tweak and every little power supply improvement. But no matter how hard I try: using LAN does not negatively impact SQ in my cMP setup. So my conclusion is: contrary too ‘general audiophile wisdom’: using LAN does not effect SQ (in my setup). And even if it does impact SQ a ‘teeny weeny little bit’ it’s probably largely out weighted by the positive effects of the possibilities too use a little SSD (for OS) + NAS for music-files and the possibility for going headless (which allows for disabling USB and disconnecting the VGA monitor)
I also did not do any of the Steppe deletion steps ‘number 1 too number X’, as I lost count and overview. Both on SQ as also on the ration in the approach. But the good thing is: the Steppes experiment made it possible for Mihaylov too publish a list of files which are being used by his cMP setup. Too some extent both our cMP setups are comparable. Mihaylov is also using LAN and a headless cMP setup,
So, in January 2012 I want too delete the files that my cMP setup still is using (according process explorer), but which files are not being used by Mihaylov’s cMP setup.
Process explorer lists much more files (about some 20 to 30) being in use on my cMP setup, than there are being in use on Mihaylov’s setup. So my setup is using DLL’s and is running processes that are not strictly necessary. See if I can delete these files without wrecking my cMP.
fully optimized cMP2 PC -> ESI Juli@ -> Van den Hul Optocoupler MkII-> Lavry Black DA10 -> XLR Mogami Gold -> Klein & Hummel O300
I may have missed something. Can you elaborate a bit on the "registry tweaks asociated with irq."
As for "polling", are you referring to USB poliing?
I'm talking about basically all the Steppe registry tweaks since July of this year, maybe I mischaracterized them. The last item Serge recommended was basically suggested years ago on this forum...had to do with how much irq priority cmp or background processes gets. Ryelands also suggested one that was very good...had to do with polling of some sort. I meant these.
.
You are welcome ... pls see link
Is it worth deleting setupapi.log AGAIN?
In the midst of a new build and have not had a chance to place that in the boot file yet.
I am going to try it tonight but won't know until tomorrow evening.
Plus I thought it was a good opportunity to bring it up again!
No more setupapi.log or the ones in the config folder.
As you may know I had a hard down for a few days but finally got the system to boot after I installed a 1 gb memory plus other stuff. I went through the massive file deletion/registry changes of last few months in a one day marathon today. Then I tried to replace the 1 gb memory with a 256 mb memory but of course it would not install properly for cplay. I should say it booted but cplay said zero memory so it would not play. Any ideas? Anybody know which files are needed?
Put back all stuff mihaylov said was required for cplay/cmp and I still get a ' Not enough ram for playbackmin required 160 mb
physical ram available 139 mb
physical ram total is 0 mb'as a typical error in cplay. I played with the boot.ini file with my 1 gig ram in and it greatly affects available ram for playback. But with it same as before I dont have enough ram with a 256 stick but it used to run ok before. Puzzling?
Edits: 12/19/11 12/19/11 12/19/11
These are not for you smart guys out there (you know who you are) but for the rest of us.
1) Copyto is a very easy to use sw and can be used to goto your ultimate 'slimmed down xp' with a few clicks. This is very useful for going back to some 'useable xp level' to do registry tweaks. In my case I go back to a 2011-9-14 level os, make registry changes, then get back to a 40 mb windows with copyto.
2) If you do the above make sure you leave your earlier level os in cmp mode (with all of its own registry changes) other wise you cant reboot. Experience is the best teacher here.
3) When you get a HAL error make sure you try reflashing your bios before giving up. It was my key for getting out of a noboot condition.
4) Rebooting is necessry for implementing registry changes.
5) Use of an imaging program like Snapshot or Ghost is an absolute must to diagnose/solve issues. If you dont have one get one if you want the ultimate is computer audio operating system. Everybody makes mistakes.
It does work like a charm...thanks Jack. Serge (Steppe) you have zero to apologize for...we all do these tweaks knowing that something can and will go amiss ocassionally. Given that all your other, what maybe 1000+ tweaks went well, I'd say you have a pretty good track record.
Happy holidays to all.
Edits: 12/21/11
I went back again and reloaded a 2011-9-14 version of my os. It played well with the 256 ram. I am going to save my file deletions for last and do with copyto. So I started doing registry tweaks for Steppe 5 onward...checking play works after a reboot. I got all the way up to resetting non paged pool size to 4194304 and when I tried to run cplay it said error no ram available. I reset it back to zero and rebooted and it plays ok. Why it worked before I dont know. Maybe I never did it right, maybe I dont know what I did. Did anybody do this while using 256 ram? Should the number be different fo 256k ram?Oh yes it should be different. I just read rick's comments wherein he had to half the number for his 512k ram. So I guess I have to use 4194304 divided by 4.
Edits: 12/21/11 12/21/11
Memory size of 4194304 bytes (4 mb) is very small size and will certainly be depleted . Why this setting?
to advice it. I was hoping to enable large pages for awe - not 4kb, but 4mb in size - so that processor wouldn't have to pace to and from memory all the time. Thinking that awe locked pages in nonpaged pool I suggested this setting, but didn't find the answer to the second part of the equation. So it remains in my and other folks' cmps. Those, who made the change are probably advised to set the non-paged pool to 0 (default)
Serge
your courage and innovation have made all these big changes possible.
I'm giving up on this path. I got an old cmp image back into my ssd disc (back 3 months), loaded in my 256 mb ram and it booted and played cplay fine. Then I tried to synchronize my latest os (the one I redid yesterday manually) to the disc via copyto and it loaded the files. But when I tried to boot it booted but would not play for lack of ram. Was I supposed to use as a source my ultimate os that I had before the swap of mobo/cpu? I'm a little confused now.
I guest you did.
Do you have another stick of 256mb Ram for testing? Sometimes, Ram can go wrong. To save your time and efforts re-installing a new Windows, but to find out later that it wouldn't work either, you should make sure your Ram is healthy.
If you install with a new Windows and still can't get it working on different set of 256mb RAM, there's a problem with the mobo.
Here is what I did
1) Reloaded a 3 mo old version of my ultimate os in my cmp pc with the 256 ram installed. Ran cplay and it worked ok.
2) Tried to synchronize via Copyto my ultimate os (all file deletions) that ran with this same hw but with 1 gb ram. This booted but when I tried to run cplay I got the error message 'not enough ram'.
I like the Copyto software. I will likely use again for other stuff but I think as Mihaylov suggested I must have something occupying big chunks of ram. Maybe when I went through my marathon files deletions manually I screwed up something in the registry.
given that you already knew what files should be kept/removed.This works for me, on the condition that the same hardware are used for re-installation:
1. Reload your ultimate Windows image to the C drive. Boot from another PC and copy everything in the C drive to a folder, e.g. Windows Backup.
2. Do a fresh Windows Installation, with all the optimizations, including all registry tweaks, on your G31 mobo with the original ram. Make sure Cplay and CMP are working with your 256mb ram. No file deletion yet. For the registry tweaks, to save time you may load hive and export different parts of tweaked registry and load them into the new Windows.
3. Download CopyTo from http://www.kish-d.com and install on the other PC.4. Run copyto on the other PC. Compare the "Windows Backup folder" to the newly installed Windows on "C drive". Run synchronize to delete unwanted files all at once. Don't worry, it will not delete any wanted files.
With the presence of a Windows backup folder, I am able to reinstall and restore an ultimate Windows in less than one hour.
This works charmingly for me and I no longer run batch files. However, things are different with a different sound card.
Edits: 12/20/11 12/20/11
This is a cool way todo it. Let me understand
'...2. Do a fresh Windows Installation, with all the optimizations, including all registry tweaks, on your G31 mobo with the original ram. Make sure Cplay and CMP are working with your 256mb ram. No file deletion yet. For the registry tweaks, to save time you may load hive and export different parts of tweaked registry and load them into the new Windows...'
I take it I could load an earlier snapshot of cmp os (with the 256 mb stick installed) instead of doing a fresh install. Correct?
If yes cant wait to try this.
Of course, you can. If it works with your 256mb ram, that's OK.
Edits: 12/20/11
One more question will Copyto work with an image provided by Snapshot which is what I use to copy my ssd os every week or so? If not how do you make an image of your ultimat c drive?
Copyto compares real files and doesn't work on images such as Ghost. As such it takes 2 steps, 1st to load an image to C drive(booting up is not necessary), and 2nd to extract the windows files, in order to conduct comparison and deletion.
Copyto is a only tool to delete unwanted files under this circumstance, image of your ultimate C drive was already taken by Snapshots or Ghost etc during different stages of tweak.
The process mentioned will only work, if and only if, a list of Windows files is known or present.
When you sysnchronize your hd os with a backup windows (ultimate file deletion) does it just do the files and not reset the registry?
No. If a file exists, the synchronization will not touch or remove it. Only extra files will be deleted. If you choose "update", older files will be replaced by newer ones.
If you worry, there are many ways to preserve the config folder. You may synchronize folder to folder, e.g. system32/driver(backup Windows) to system32/driver. You will need to move all .dlls etc to a folder on both windows/system32. Just synchronize this folder without tampering with the config file. The simplest way is to backup the config folder and put the backup back after synchronization.
ok thanks.
I am not at home so I cant run my pc but did anybody find which files task manager need? I'm sure mihaylov is on to something. Also jolida I will try the .rom files first thing.
Here:
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2 Ru, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Liv
Thank you Serge!
At you it is banal there is no free memory for cPlay's start. It is necessary to find out where it got to. For this purpose it is necessary to start somehow the Task Manager and to look at what process eats off a lot of memory. Somehow so.
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2 Ru, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Liv
remove /maxmem=256 switch altogether, to allow windows to see full stick and to find addresses for it. I had this insufficient memory warning with cplay but the deficit was only 1-3 mb and it was memory timings' dependent. Are the old and new sticks otherwise totally identical - manufacturer, speed... Are You putting it in the same bank (white slot). Also, when You were making deletions, was there a 1 gb stick still on the mobo, or you put 256 and started the deletions? I would guess You had 1 gb, deleted everything and then put back 256 mb stick. For windows it is a hardware change, and as all services and files are deleted, it cannot properly initiate plug-n-play/install new hardware automatic routine.
Serge
'...remove /maxmem=256 switch altogether, to allow windows to see full stick and to find addresses for it...'
I tried that ...did not work.
'... I had this insufficient memory warning with cplay but the deficit was only 1-3 mb and it was memory timings' dependent. Are the old and new sticks otherwise totally identical - manufacturer, speed...'
Same sticks I have used for 4 years.
'... Are You putting it in the same bank (white slot). Also, when You were making deletions, was there a 1 gb stick still on the mobo, or you put 256 and started the deletions? I would guess You had 1 gb, deleted everything and then put back 256 mb stick. For windows it is a hardware change, and as all services and files are deleted, it cannot properly initiate plug-n-play/install new hardware automatic routine...'
Same slot, yes I started with 1 gb did deletions then put in the 256 mb. Looks like I have to have yet another do-over. Put in hardware 1st then do the deletions.
Thanks Serge
Anybody know which files need to be restored for windows to do this?
I doubt you would have any problem using the 256 from the beginning.
Especially if you do an nLite install ...
Just make sure you use default voltage settings for the installation. I would leave all other settings (for your board) the same as cMP.
You are right. In the process of debugging my problem I swapped in a new stick then forgot about it until I noticed I was getting 8 minutes + before a memory swap. So unless I can pull a rabbit out of the hat by substituting a few files I'll have to do it over anyway.
Have u tried restoring the bios.rom @ other '.rom' files & checked???
Junaid
Do you mean all the files with a rom extenesion?
Yes... Can give it a try !!
Junaid
I think I deleted this on my non-ACPI install thoughI am not sure.
From what I read it reappears if you delete it so my thought is with the RESOURCE HACKER doing its work on the .dlls will this cache be filled from the .dlls (with their mising icons) or from some mysterious source deep within WINDOWS?
I know I will find out tonight but thought I would see if anyone has already found the answer.
Shouldn't have this one in the first place, as it is created by photo viewing softwares. You should have checked the "don't cache thumbnail" box in the folder option before deletion. Shouldn't have anything to do with RESOURCE HACKER.On second thought, it is mysterious. Windows fax viewer only create a thumbs.db but not the one you have. I don't have this problem though.
Edits: 12/19/11
I did not know what that meany with my Nlite installs so I left it alone.
NOW I KNOW!!!!
Going to try some chip removal tonight. Did some practice and I think I can do it.
I am going to remove some of those caps around the video chip that are so close to be sure I do not cook them. Might replace with something better but not exotic like the low ESR Panasonics (the blue ones).
Can't wait to give it a go!!! With thanks to you for the inspiration.
Offers high quality 64 bit double precision digital volume control (in 0.5db steps). This can be bypassed.
Just noticed this at the top of the page but have no idea how it can be done.
Can someone tell me?
Don't know how to bypass this, but sure can be removed.
Edits: 12/19/11 12/19/11
I am 99 percent sure that cics has designed cPlay as such, that when You set the volume to 0.0dB, then all Volume Control Code would be bypassed. I think at last that would be the nearby Idea when one designs the graphical interface of such a great software as cPlay is. (to not spend another GUI button somewhere)When one selects -0.5dB or other settings, then the code kicks in. I really think it does that in the least intrusive way from all worlds' audio software (32bit float etc)
#been away for some days, trying to comment on some of Mark, Bibo, Theo.. questions below. Many good points made these are brilliant developements :)
Edits: 12/19/11 12/19/11
I feel sure you have it right.
Hello, everybody! Please help. I have Gigabyte GA-H55M-UD2H motherboard, just like many other inmates. Here is the problem:
After the crash, that happened with /CLKLVL setting in my boot.ini, I, just like Theo, lost my bios settings. They were reset to default. That in itself isn't a problem, exept, that my bios was COMPLETELY CLEAR. No cmp settings, that I had previously saved with f11 key were present anymore. This is also not a problem, but after I restored Cics' recommended settings, the reboot time is VEEERY LOOOOONG. It practically takes a minute or more to reboot after exiting from bios. Any switch on and switch off takes as long. Moreover, the more agressive settings, even slightly different from original, lead to a unbootable machine, it requires to shorten the cmos, restart everything again and so on... What can be done? Voltage doesn't get below 0,7, though video is 0,612. No memory timings shorter than 5 and auto. But what could have happened?
Please, help!
Serge.
For what it is worth, I have the same thing. It followed using resouce hack to go through every dll in system32 and remove all resources.
I run a laptop with none of the extreme bios settings, and made no bios setting changes prior to seeing this.
I had just started listening to this config. I switched between songs and all of a sudden all went silent. The USB DAC was no longer recognized, and refuses to be recognized by any winxp image. I think I got a voltage spike and zapped the pcm2706. I think this is in no way related to the resource hack, but it means I'm offline for a month while I research asyncronous USB2 options and get it working. I've wanted to take this step so no problem. Meanwhile back to spinning cds.
Hello, trying to find out for myself, I learned, that such problems are more often with asus laptops and lenovo smth. It may denote a problem with BATTERY. I have reinstalled a virgin untweaked windows on a cmp machine on another hdd and, as OS is interacting with bios, THE NORMAL OS with most functions of bios in Normal position seem to have partially settled the problem. At least it allows now rebooting with risky timings. As for the time of reboot, I don't care much, now that I know it Will reboot. Otherwise it was sorta uncanny, with a fanless psu everything seemed dead... For me it is everything goes, as long as I don't return to those "nasty round thingies" - (crooglakh - a rustic Russian word for anythind round, midsized and not really pretty)
Serge. Anyway, thank you all, inmates.
Edits: 12/18/11
Serge had a similar problem. I had a problem trying to put in a granite digital ps on my ssd. Got a boot error ...HAL corrupted. Had to reflash bios which got me past hal then I had a boot.inf file missing so I had to reload a Snapshot image of my OS prior to the deletions. That finally worked. Maybe this can help othersSome have helped me in this and I offer my utmost thank you to them. Dont want to id them. he knows who he is.
Edits: 12/18/11
If it's a bios issue, flash your Bios once more to clear every thing.
Jack can you please email me?
Can you reload another ghost or snapshot image of your os from a prior level? I dont have that mobo so not sure what to do. Maybe just try another 'load default settings' in bios.
Edits: 12/17/11
Well, inmates, if you can tell, tell me please, why would Cics write in his optimizations' papers, that we should switch priority to background services, not to programs?
VERY INTERESTED, not idly.
Serge.
,,
Who knows what that means but this was another thing I did not have to do with the nLite install.
My recollection is that it improved the IRQ priority of cplay. But yes its easily checkable by going back.
this. This setting was supposed to let the cpu work less, thus less noise. But it is wrong. The number of priority threads remains constant, no matter what is given the priority.
Here comes a BIG tweak - in bios open MIT, Advanced cpu settings, where we set the number of cores, Third from above - enable the cpu to enter e3/e6 support - ENABLE IT (I was holding it long), If you have the earlier mentioned bios capability ENABLE VDROP DIRECTLY (against Intel specs), get into the registry,HKEY_LOCAL_MACHINE\SYSTEM\ControlSet001\Control\PriorityControl , Doubleclick Win32PrioritySeparation and set it from 18 to 26. Reboot. Give 1 hour, enjoy. It was the last big fish from the mermaid because she is now off to see her mother and organize things.
Serge.
Can't wait to try the last big fish from the mermaid! But its late night now, will have to try tomorrow.
We just did what we had been told to do :)
At that stage, it was assumed that Cics, or for that matter the author of the AOBCT had done enough research to reach to such a conclusion. And I'm sure it had a reason..!!
No harm in swapping the priorities in those settings to see what change it makes.. After all, it's just a setting..
Junaid
Sorry I cannot find such claim in cics papers - I reread "07 - Optimisations".
Where is it?
in the AOBCT v0.3 page34. right on the topof page- paragraph 6 - Performance settings:
a. switch to system cash and prioritise to background services
b. Set virtual memory to 0mb
Serge. I find it strange, why? CMP-CPLAY are programs, so, shouldn't we, rather, give priority to them? especially, as no other programs are running?
Serge
OK, I see.
That document is old (I do have it) and it has been superseeded with cics site. Anyhow, the information is the same.
Explanation for System Cache switch is...well, obvious.
Backgroud Services switch specifies that all programs receive an equal amount of resources. I am not sure, but perhaps it is to remove some of the "logic" control from OS and to make more a "balacing act" in the way files are played, passed to RAM and sent out...
I would like to express that I am rather perplexed at this turn that the cMP community is taking.
I am rather proud to belong to this community, not just as a simple user, but as part of a project - cMP2 - to help music lovers around the world to listen to PC Audio music in the best possible way.
I realize that together with this democratic approach there is a search for innovation. However, it seems that we are getting into a frenzy without putting a fullstop to amything. A new chapter is started without finishing the previous one almost every time. Don't get me wrong - I am grateful and thank anybody (Steppe, Jolida, Mihaylov...) who is involved in searching for a better quality system. However, a stepping stone has to be placed and experimentation cannot be constantly shifting.
For me, slimming of the OS is a very worthwhile approach - no doubt about it - but it must reach a point which it has to be reasonably easy to be employed by the rest of the community at large. This is part of the goal itself. It must be made systematic, otherwise the real sense of cics project would be lost!
We are not here to make a sort of "world record" of PC Audio reproduction!
That is why I was very much in favour of Jolida making a sort of installation disc with questions about each own PC system for making a proper "stepped cMP2 or a cMP2s". This stage must be taken to completion.
After such stage there can be obviously further development, but it cannot be our only concern. Most of all, we cannot constantly shifting our testing. Otherwise we would be loosing the community, rather than enlarging it. There are so many testing "threads" which are left unfinished. Or maybe the original proposer has reached a conclusion, but it was never presented in a proper finalized manner. Research must be channeled in to something realistic.
Let's not forget that usability is as important as quality. For example, Mihaylov's cicsRemote.exe was simple but useful; also, cMP interface would require a freshen up...
Furthermore, this latest hacking approach of kernel files worries me. Although it might be useful in terms of sound reproduction, it verges on illegality. I think we have to be careful threading this path - for our future. It's a concern of mine, that's all.
Perhaps it is time to organize the developments with proper chapters: Chapter 1 - nLite installation, Chapter 2 - drivers; Chapter 3 - minlogon;...and each chapter must have a step by step clear explanation for reaching its completion. That particular forum thread has to be reserved to those chapters only.
Obviously, a cics site update would be better, but he is not around and we better get organized.
then to organize things for ourselves.
Hello, everybody. This is unprepared post, so it may seem structurally illogical, but I'd like to express certain thoughts and ?proposals.
1. I understand perfectly well that a long pause in development of the project, that was caused by the absence of Cics, cooled many of the inmates who had been active before May this year. When it all started to develop again, it started without an "announcement", which led to the fact, that worthy and interested folks came back in the middle of the process, some almost close to the state, the process is now in. Certainly, it is very difficult to get inside once again, just because the structure of this forum is linear and we are limited by as many strings as would fit on the scareen, while previous pages are psychologically "outdated".
2. A lot of different things were proposed by inmates to improve the sq both soft and hardwarewise, and I don't think we should cut cmp thread. IN NO WAY!!!.
3. As I see it, and I may be totally wrong, we are getting very very close to an end of a stage in development, so it's a period of chaos before the natural settling down.
4. I am not a lawyer but I don't think we are doing anything illegal as long, as we do not openly and for the profit distribute files, that are not our own. I think we may easily distribute nLite ini-session files with settings for anyone to create his own iso, but not isos themselves. We can modify windows files as long as we do not touch the code and only play with gdi features and we can distribute our OWN experience, but, I am afraid, not the modified files.
5. In order to protect our developments IT IS NECESSARY to brush all the thread (both threads, in fact) in search of tips and ready-made receipts of both hardware tips and software tweaks and to preserve this experience in view of inevitable shift towards either new Microsoft or any other platform. I am afraid XP is getting a little bit to old and will be hardware obsolete in another 5 years, while our know how may last much longer.
8. I have serious reasons to believe that Cics WILL update his site with our tweaks included, but they, the tweaks, should be MORE USER FRIENDLY. Junaid, here is working on it. More people are invited to develop a ROUTINE for tweaking,starting, possibly with windows'install.
9. A proposition to copy our tweaks to another site is welcome, even for the sake of preservation and as a kind of pool whereyone can EASILY find ta. I. for example, have my tweaks ON SHEETSOF PAPER! WRITTEN BY HAND! And in order to rebuild the cmp again had to shuffle through them together with Junaid's batches.
10. We need to keep all lists (there are 3-4 variants) of the final windows folders. WE HAVE TO GIVE JUNAID OUR NOT_FOR TOUCH FILES'LISTS!!!
Junaid, did I give you mine? Or did I mess everything up again?
11. The strength of Cics' project was in ROUTINE meticulously written. From beginning to the end. We need to do it. And we need to find a most EASY WAY for those who will come to join. That's all about organization, as I see it.
MOST IMPORTANT WE NEED to find the source code for cmp and cplay. I think that Serge Mihaylov has it (He translated the GUI to Russian) and probably other folks. Somebody created a fine cue file generator. Lets find the man! He may have ideas of his own.
As for the tweaking, I think it's not long before we run out of raw material, if you see what I mean. As always the FEEDBACK on tweaks is most precious. I have just posted several tweaks but have got only 2 posts for feedback. Mario recieved few feedback posts for a BRILLIANT TWEAK... It discourages, because it... well, discourages. Your feedback, even negative is necessary.
That's about all. Not very systematic, I am afraid.
Serge.
Steppe, you haven't given the list of your "Lynx exception files" to Junaid yet.
I do have cMP source code and I passed it onto Mario for a possible interface revision through a friend of his.
I am all in favour of more user friendly!
Re. your Fishy Secrets, you can organize and post them in the Italian cMP2 site, if you feel necessary. Or you can send them to me and I will organize and post them there.
I am sure Mario has received more feedback privately, me included, than over here publicly.
I am all for it!
Hi Gianluca!
Are we waiting for new year's gift - a new version of cMP as the good old days? ;)
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2 Ru, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Liv
Hi,
the cMP Cover Project has been a bit of an odissey.
I went through 4 different programmers (one was even my cousin) who "did not find the time to do it", although it would take only one day of work.
Right now there are two hopes. One is a friend of Mario's (nagual19).
So, it will not be a Xmas present but I am still actively pursuing it...
...and I will get it! :-)
I've heard in a long while. I also have some ideas concerning the facelift...
As for Lynx exeption files... Well, I AM stupid. I don't really understand, what is needed. I have written the files that Lynx uses that are not needed for Jul@ - its portclis.sys and drmkaud.sys and/or drmk.sys, don't remember on the spot. I can simply send or post here all my files left, they are not many. Please specify exactly what is needed as if to a child. I am not joking. I have posted ks.sys, three files above, wdmaud.drv. Lynx works without any other specific files, if cmp-cplay and, yes, Windows start at all. I should note, however, that splitter, aec, drmkaud, drmk sys files are extracted from windows'cab during Lynx installation together with lynxWDM.dll and LynxWDM.sys. I even asked on Lynx forum what these macrohard' files are needed for, but no reply yet.
Serge.
Hi BiBo01,I fully agree.
And even more important: don’t skip ‘rational thinking’ at any time during the process.These two Steppe threads are a perfect example that some reflective and rational thinking is very much needed.
Thread #1: cplay 2.40 or better? New very important tweaks. (VERY LONG)
Thread #2: Tweaks reconstructed+ some newsIn the thread #2, Steppe reaches the conclusion that the strong effect on SQ that came from deletions and registry tweaks as where done in thread #1, where caused by disableling the Codecs. All other deletions in thread #1 had caused no effect.
Unfortunately, these 2 threads where the last threads where Steppe (and other forum inmates) asked themselves what deletions and tweaks had caused the effect on sound quality and which did not have any effect.
To me thread #1 and #2 also show how easy it is too draw false conclusions and how easy it is too mess up ‘causes’, ‘effects’ and ‘symptoms’.
I could not detect any learning curve or knowlegde development in the whole series of batched tweaks that followed. When inmates noticed an improved sound quality, this improvement was attributed to a curtain batch number of the Steppe tweaks. “After Steppe tweak number 14, my cMP sounded better”.
To me this is like ‘shooting in the dark, hoping you hit something’. And in case there is anything hit (better sound quality), we than still don’t know what caused the effect . We only know: after batch number 22 it sounded better.So unfortunately, besides better sound quality, IMHO the whole process also gave birth too a whole lot of mis-conceptions and false (underlying) assumptions.
- Why should deleting files that are not used by XP anyhow, give a cMP setup better sound quality?
- Why would slimming XP give a better SQ? (besides ‘the shooting in the dark’ effect)
- Why would slimming the registry give a better SQ ? (besides ‘the shooting in the dark’ effect)
- Why would storing 50 music files on a cMP disk, give better SQ than storing 500 music files on it.
- etcDon’t get me wrong: Steppe definitely brought my cMP setup a MUCH (!) better sound quality. The ‘codec disabling tweak’ and ‘the watchdog`registry tweak’ gave my cMP setup a major boost in sound quality. And I thank Steppe for that.
But I agree with BiBo01, getting organized is very much needed !
As it will almost surely also dramatically improve the quality of the process and the quality of thinking Which could lead too the conclusion: that we do not need too wreck our cMP too get a much better SQ from it. (which could unravel another underlying assumption: a wrecked cMP sounds better than a non-wrecked cMP)Mark.
NB I don’t think it is neccesaary too move too other forums, split up things, ect, ect.
The cPlay-thread is perfect for ‘software’ and the cMP-thread is perfect for ‘hardware’
fully optimized cMP2 PC -> ESI Juli@ -> Van den Hul Optocoupler MkII-> Lavry Black DA10 -> XLR Mogami Gold -> Klein & Hummel O300
Edits: 12/14/11
You said '...Why would storing 50 music files on a cMP disk, give better SQ than storing 500 music files on it...' has bothered me also. I use 3 major discs/folders for my music files: D, F and G. With all my stuff loaded I think there is about 1400 music files. If I load my smallest disc/folder there is about 60 or 70 music files. So listening to the same music file @ the 60-70 level sounds a lot better. Today I think I noticed a possible answer. When you ask for diagnostics on cplay it always tells you how much ram you have available. @ the 60 file level I have about 6-7 mb of additional ram. This also answers my own question as to why I was swapping data at a level 15-20 seconds earlier just recently. I had not yet restored my ADD / REMOVE functionality in shell32 so I always had 1400 music files loaded and less available ram. Now that I have restored it I get a lot more ram available and longer times 'before ram load' in a cplay run. The size of the music file also affects available ram.
taking up the space.
Very interesting, Ted.
try to remove /maxmem=256 setting from boot.ini and try again, please.
Serge.
Tried it (deleting maxmem=256) and no difference in swap times or sonics. Big sonic improvement must have been from nopae and onecpu then.
ok
Hi Mark,
i think you are right, we can use some ration at this moment in our thread. But i think we would never got here without the creativity of the Serge's and others. Steppe could never have reached the watchdog-tweak without what you call shooting in the dark. When you shoot in the dark and hit something, you can go there and see what you hit, and then try to figure out where it came from, at which speed, maybe together with other things you didn't hit, etc. And from there start a rational process to the next step. That is what i like very much here. My learning-curve was very steep the last months: i learned more about xp than in the 5 years before!
Besides that i think Steppe's mermaid is a very rational one. She always comes up with very logical tweaks, which must be based on deep understanding of xp.
I think the main underlying assumption is "less is more".
We try to reach the sweet spot of size, design and sound quality. With both rational and unconventional methods.
So again, thanks Bibo01 for bringing up this issue and for offering space on your site to put the batches and 'evidence-based' tweaks.
Douwe
Hi Douwe,
Don’t get me wrong.
I’m very pleased with Steppes creativity and that Steppe came up with the idea too explore and tweak the XP software. And probably most important: that he was willing too put his energy and devotion in it.
“you can go there and see what you hit, and then try to figure out where it came from”.
This is exactly what I miss. I would welcome this very much but unfortunately this is rarely done. (As I argued in my previous post when referring too Steppes thread #1 and thread #2)
I fully agree that the ‘Mermaids concept’ is a rational thought/concept/idea. As Steppe nicely puts it: “leave our KS beauty almost totally naked and busy with sound reproduction exclusively”.
I also don’t see any other way than to do Steppes project through the trail-and-error concept. But for me, BiBo01’s call for getting organized, came at the right time.
As in any innovation project, Steppes project benefits most when many inmates participate and many inmates give their feedback and ideas on the project.
But right now, only just a few inmates are still on the project. Many other inmates lost track or interest.
And lastly (as in any innovation project) Steppes project doesn’t benefit when inmates suppress idea’s from other inmates or when they are displeased when other inmates make suggestions for improvement.
So my take is: Steppes project will greatly benefit in many ways from BiBo01’s call to organize.
Mark
fully optimized cMP2 PC -> ESI Juli@ -> Van den Hul Optocoupler MkII-> Lavry Black DA10 -> XLR Mogami Gold -> Klein & Hummel O300
Hi Mark, I won't get you wrong, i know you appreciate creativity and know your postings on the HW-forum. I agree we need some 'mass' to get enough feedback. Balancing act is nice put.
So let's hope the roundup will bring new directions/downloads on the cics or Bibo forum and attract some new inmates.
Cheers
Douwe
One can wait for a ready to go recipe or decide this is interesting enough to try to learn something about it.
I have no technical skills other than what I have learned from following what others have discovered and looking further because of their inspiration.
As Mihaylov has said many times everyone's OS is different. There are many possible approaches. There will not be one best for everyone anyway.
I think that if one reads the posts it is not hard to come up with something that will work well for you.
No one here has promised a turnkey solution to getting the best out of cMP. I suspect that if someone does it will not work for everyone, anyway.
Who knows where this effort will end? It might come to a halt next week or it might go on for years. One can wait for the easy go or jump in and
you may contribute something to the process. Even if it is simply reporting what something does is worthwhile.
So, who is going to organize the various aspects? I do not think anyone has had just one area of interest. Does someone tell that person to just concentrate on one thing?
If one wants convenience you best leave the whole thing alone. Those of us in the pursuit of best possible sound quality have no interest in convenience. I know I don't. Like wanting a turntable that adjusts stylus tracking angle and azimuth for itself. Not going to happen.
One cannot have it all - compromises/consequences.
We know what files are needed to have a working cMP. We are still exploring ways to minimize registry interference. Hacking the files illegal? Are you serious? I would not think a program that has been downloaded thousands of times would put us in jeopardy. If that is the case there are plenty of other things that have been suggested that would probably be illegal, also. Another reason for people to proceed with this on their own taking cues from posts instead of waiting for the "kit".
As with all DIY audio there is the opportunity, when combined with some effort, to produce something that suits your needs. That is what is happening here. Not something made to appeal to as many as possible.
If you want the best possible sound quality it is not that hard to read through the posts and try what has been suggested. The problem for you will be that the Serge's, who have led the way, are not interested in convenience. Would you expect them to decide to change their mission to suit those looking for convenience? Somehow I do not think that is going to happen. It might, anything CAN happen, but I would doubt it.
The obvious solution to this, for those who are interested in what you are wanting from cMP, is to find like minded folk and start experimenting. To expect someone else to do it is ... well, I won't say.
In actuality, to get to a very good sounding minimal WINDOWS/cMP is pretty easy. Quicker than setting up full WINDOWS and doing the cMP optimizations. But if you want to retain convenience, which for everyone will be different, that is another thing entirely.
Maybe we were all spoiled by Cics who gave us a gem and a kit (i.e, website instructions). But these 'post cics' tweaks are just that...tweaks that somehow have become more than 'just tweaks'. I cannot believe that anyone would complain about the chaos here....imo its great. Out of the chaos comes the creativity. If its hard to follow and difficult to implement...well welcome to the asylum! Out of the chaos comes a wonderful sounding front end that if I bought commercially would cost way way more than $ I have sunk into CMP. And how many of you have paid $600-3000 a pop for post purchase upgrades for small performance pops while cmp just gets better and better weekly sometimes daily with significant breakthrough performance for FREE.
Yes I get frustrated. I need to go n_lite to fix my registry but so what. What else would I do. Maybe I have the time and many of you do not. You can always do what I did when I worked 12-14 hours a day...spend $5-10K (or more now a days)for a front end. Or you can slog through the chaos and just keep learning/doing and making wonderful acquaintances with the brave creatives that share their work here ....for free! Let me repeat...for free!
Where else can you (for free) use an idea for a computer program from a brilliant physics/IT expert Cics (location tbd), have it tweaked by the ever creative Serge brothers (Russia), made easier by Jolida (India), Rick (US) and hardware perfected by a manufacturing/electrical experts from Scotland (Ryelands), US (Gstew) and Hong Kong (Jackwong). I'm sure there are others I cant remember right now but I don't know about you but I'm delighted just to have the opportunity to listen to them share their ideas through this forum and maybe even sometimes have some dialogue with them. I dont want to to give any of them an assignment to make it easier for me. I don't want to scare them away.
Hi Theob,
perhaps my idea of cMP2 is different from yours. I think the original cics project is about sharing ideas on PC Audio reproduction, not tweaking away your personal toy.
As I said before, experimentation is as important as sharing knowledge among music lovers. So much so that lately I have been promoting the idea to go beyond forum borders and bring cMP2 to the people, among the people, by the people.
That is, I have been organizing cMP2 Days in Rome, next Jan 15 in Milan and possibly soon in Naples. It's a workshop explaining liquid music and cMP2 approach, where we listen to music, compare systems and share ideas in a proper setup audio room. In Rome we had over 100 people.
I'd like to point out that I am doing this for free, like my whole Italian forum activity (including reviews). I am not promoting any brand, personal or otherwise. I do not work in the audio business.
Hi Theob,
I get what you are saying and probably would feel that way too.
I certainly dont mind the effort and messing with things. Please look at my comments not as me saying "hey I am too lazy and I just want someone to do the work" That is not what I am saying and not what I think Bibo01 is saying.
I AM saying that things are so muttled and confused that I cant figure it out. If you could even tell me the overall steps then I could figure it out.
Something like:
1. Do steppes steps 1-20
2. Nlite
3. Registry tweaks
Honestly I dont know if you guys are using nlite to do registry things or to accomplish steppes steps or something else. Or if the batch files should be used and when, etc.
And if I went back and started at step 3 where are the pitfalls? I recall seeing some revisions and some dead ends. I dont really want to go down that path if I can avoid it.
I would settle for a thread that just gives a general overview and links to the posts that tell you what to do.
Anyhow I read posts about missing things and going back to the original cmp2 build and finding things that were missing and it is easy to think we are chasing our tails. I am guessing that things are better now and that the improvements are worthwhile these days??
No one here remembers the bending of our minds
I have been through all of this from the beginning. Some points left me confused, and some ended in dead ends that required a complete rebuild. Along the way I learned a lot and a this point could build one very fast. I can offer a couple of pointers that might make this easy.
1. take the time to set your hdd with partitions. I have 3 key partitions.
C: is the fully optimized CMP2 player
D: is a full function XP with network with network
E: is where I store data. 3 key folders - music, all install files & image backups. There will be a lot of these.
2. set up the system to dual boot. later when cmp becomes nothing more than a transport, you can just reboot to windows and tweek it some more, or simply reload an image backup over C: and keep going
3. start cmp with nLite. Do the cmp optimizations and take your base image backup.
4. you can apply all the recommended file deletions pretty much in any order. I like to use a base system that is more or less nLite with steppe deletions up to 12, the Jolida tweeks and in my case full Bold Fortune. I added the lost registry tweeks and watchdog stuff. This is a very sweet sounding base that remains fully functional.
5. from there, registry cleanup and compression is really worthwhile.
6. I found it quesitonable if I liked the sound with the deletions up to step16. It was clearer but maybe not as real. But if you take frequent image backups (takes about 1 minute from the full function xp, you can easily go back to any combination and savor the flavor. take win_SXS stuff out, put it back in until the sound is what you like.
7. If you want to go for broke, delete the rest of the files and from xp, do the hacks on shell32, and perhaps all the remaining dlls. At this point my system came into a beautiful state. The clarity, sound quality, sound stage are all precise and top quality. Most important, the texture came back in spades. There is separation and air around everything that just does not exist at base cmp, step 3 or 4 or 12.
So, you can do a little or a lot. Until you master multiple partitions and frequent image backups, you will be frustrated by the threat of BSOD and a rebuild.
Hope that helps a bit. enjoy.
Wlowes' is a very helpful post. Now let's build on it.
Wlowes has illustrated 7 points.
What needed is a clear organized explanations of points 3-7.
Can someone provide links for each point or, even better, detailed info from each tweak author (you can email me privately if you like)? I will then organize it and repost on my site in a user's friendly manner (well...as much as possible).
concept of the development I see, with the way I understand how and why the tweaks work, how I see the possible future development and so on. I only warn that it's not "scientific" enough. Just the concept where logical thinking is mixed with dreams, like Bertone cars from the 70-s.
Serge.
Shoot away!
Registry tweaks, boot.ini...
For deletions, I think Jolida is providing batch files and, hopefully soon, Bitch Batch.
Let's assume that deletions are carried out to the last stage and resources kept as required (i.e. network and USB), it would be interesting to know how to, for example, upgrade a sound card driver and, in general terms, how to reimplement a resource without restarting all over again.
THANKS!!!!!
That does help a bunch!
Funny you mention image back ups. In the beginning I made the and found that they were kind of a pain and then I think once I tried to use one and it didnt work for some reason. Anyhow since them I have just chanced it.
Looks like it is now time to get going with them.
Thanks.
D
No one here remembers the bending of our minds
Well here is what I learned going from a full CMP^2 to Steppes 1-22. It locks you out of making registry compacting and makes it difficult to do registry tweaks. You can still do tweaks but either by reloading files into system32 or by using a host computer to tweak loaded registry hives (which also is risky...was for me anyway). If you do an nlite install much of the registry is already compacted. So you start out cleaner but then you have redo cmp^2 and all its settings per the CMP site (not easy or at least time consuming). So after you reload cmp^2 probably an easy way to do the file deletions (not the registry tweaks though) is to start with the files you need and eliminate all others. But do the registry tweaks 1st. Then with a 30-40mb windows you can do resource hacker (through a host computer and a docking station or network). Then you are done until something new comes up and you have to retrace anyway.
Hey I'm a victim too because I havent done the registry compacting. So what I'm going to do is buy another 8 gb ssd nlite it and slowly do it the right way. Why do I want to do this? I just read that the registry is loaded somewhere in memory so I have 10 mb excess baggage in mine till I compact it.
I also think we all have to organize our own work because our needs/skills/knowledge base is at different levels. I'm not against a site for this like as been suggested. But it takes people willing to do this.
I agree that we should't try to assign creativity, and i think we can learn a lot going through all the posts, it's just great!
Bibo01 just offered to put the batches & mermaids on his site, in a sticky topic, which is a wonderful offer!
Now we have both: creative chaos and an abstract of the most important CMP-knowledge on his site!!
Douwe
In the past days, I am working hard in making an Interactive program that executes the Bitch Batch in a very organised fashion. Somehow managed to get it done. The only thing that remains is the list of files I had repeatedly requested inmates to email me for their respective peripherals, which need to be retained for their functionality. I have got only Mihaylov's list for Network & probably Juli@. What about Lynx users?? And Asus Essence?? USB dacs?? EMU sound cards?? Nothing has been sent to me so far !!!
To maintain a web-page or whatever at this moment is a big question mark, as it would need frequent updating. Mods keep coming up every week with regards to Registry, Boot files, Shell modification, and the list goes on. I presume we need to wait a little more till we are Clear about the final picture. It's Hazy at the moment. Once the picture gets completed, it's quite easy to lay down a kind of Blue-print & then organise things in their respective segments.
I strongly feel its ABOUT TIME that Cics shows up !!!
Are u guys thinking of obtaining a separate Web-space to list the tweaks ? Or is it not a wiser idea of Cics integrating it on his own cMp website???
Junaid
Hi all!
I think Bibo01 is right, and have a few thoughts about this issue. The fact that this issue pops up means we are growing up as a development-forum.
What i like about this forum is that everybody is developing something and directly giving feedback in a very positive and encouraging way. That's very important for innovation, and it makes hanging around attractive.
I think our goals are good sq and fun. For some inmates sq is more important, others have more fun in tweaking.
imho growth of community is not a goal in itself, but is important to create sufficient 'mass' for feedback on the tweak-proposals.
About 'to get organized'. I think we should keep our chaos, and at the same time organize 'backwards'. I'll try to explain.
I have in mind that the first step of our last big improvement (steppe 1-3) was not the result of professional planning, but the result of the combination of Belgian beer and Serge's brain. In this mix the idea emerged "what the hell, lets delete whatever and see what happens".
And it brought us lots of sq-improvement, new knowledge and even more fun!
So lets keep that, and the chaos.
On the other hand, we should convert the chaos of the past months into some organization, i agree with Bibo01.
So, in order to enable ourselves in continue innovation, we need both perspectives.
We need inmates like Steppe, Mihaylov, Rick, and others who have technical insight and (sometimes crazy - in the positive way) ideas.
We need inmates like Junaid who takes lots of time to make batch-files and put much effort in findig the right nLite-ini.
We need feedback on the ideas.
And we need some organization, which enables others to take part in innovation (in stead of spending time scrolling through 100s of posts trying to follow the discussion).
Jolida did great work in the batches, and his idea of BitchBatch is more than great! I agree with Bibo that we need some space to store our knowledge (in the sense of 'where we are now'). A place where inmates can download batches, ini's, tweaked files (not iso's - could create legal stuff). We could share this space with the hardware forum, so newbies can be informed effectively. Maybe Mihaylov's site (if Serge wants to maintain this)?
In this way we can go on posting and feedbacking on tweaks and ideas. And once in a while, when somebody feels like summarizing or batching the development/discussion, this can be done in the 'knowledge-space'.
I wouldn't vote for splitting up in diverse forums. One software forum and one hardware forum is good for 'keeping the community in one space'. In many tweaks different 'chapters' can be involved.
I think splitting the forum may lead to splitting the knowledge and to less ideas and less inspiration from posts you would miss if they were posted in an other forum. Remember, the best ideas emerge when you are not focused on the subject.
Still here? Great!
Bottom line: let's keep the current posting-method, and let's create some space for 'backwards organizing'.
Happy tweaks!
Douwe
I am glad that a few people feel the need to organize.
Experimentation and organization have to go hand in hand.
I would like to hear from as many inmates as possible on this matter.
As some of you may know I am in charge of the Italian forum for cMP2, which can be found here: http://www.nexthardware.com/forum/cmp-cmp-cplay/
It is also "transparent" to naviagate the forum through Google translate:
http://translate.google.com/translate?u=http%3A%2F%2Fwww.nexthardware.com%2Fforum%2Fcmp-cmp-cplay%2F&sl=it&tl=en&hl=&ie=UTF-8
Therefore, folIowing Douwe's proposal, I can offer such cMP2 site - it's already there! Eventually, an extra "knowledge space" can be created, for example in the form of a sub-section of the present forum which could be called "cMP2 Advanced".
As it is now, discussions started om the forum can be saved and made "sticky".
What do you think?
I knew about your site, but totally forgot. It's a great site!!
It would be very nice if you could just make a new sticky topic, and put the batches, ini's, mermaids etc on it.
Thanks for offering to do this!!!
Douwe
Jolida, you can post your nLite.ini file, the batch files and the upcoming Bitch Batch on the Italian cMP2, if you like.
I will then organize it, if it's required, and make it into a sticky.
The same invitation is to you, Steppe, and to your Fishy Secrets. :-)
...and to all.
I am ashamed, but I don't know the site's address and how to post there. Does it imply, that I can go on posting here and elswhere as well?
Serge
A few messages before I posted yje address of the Italian cMP2 site.
Anyhow, here you find it again at the bottom.
If you go there, you will realize that to open a discussion it's quite simple. Just register and then you can post. If it all comes to the worst, you can always message me and then I post it - and SAVE IT!
A "sticky" is a message or a discussion thread that for its particular relevance a forum moderator plaees at the top of the page and it remains there. It is marked in red and it doesn't get lost. They can be considered as "Chapters".
You will see that those "stickies" corresppond to a revised cics Guide to cMP2. For example, also suggestions for the new H67/Z68 platform are included.
PS: I just discovered that the Google translated version does not include the 'New Topic" button, which is essential to open a discussion. I will try to have it fixed tomorrow.
on the Italian cMP2 forum one cannot log in through Google Translate.
Unfortunately, Google Translate does not allow it.
However, one can log with the Italian version and open topics/post messages in English - it's rather intuitive. We had it before with some German cMP users.
Those posts, which probably will become "stickies" as part of the main Guide, will obviously remain in English.
As an alternative, you can still message me directly and I will organize an Englsh Guide to nlite.ini, registry tweaks, deletions, batches...
One can still use Google Translate to simply browse the site.
HI Bibo01,
Well said, and I agree for the most part.
One of the things that has kept me from going past steppe 2 or 3 is that there is no real recipe. This was the genius of cics approach. Simple easy to follow steps.
What I have seen is some great attemps for which I am grateful. However there seems to be a rush and some things done in previous steps have to be corrected. This has kind of soured me. And the other major factor is that one has to go to every, EVERY thread and try to dissect what is going on.
I got lost. Is it steppe 100 then some registry thing then an nlite disk?
Or is the nlite an after thought (yeah I know what nlite does), a quicker way to do all the steps?
One nice site would be helpful. If you skip a day it is easy to miss what is going on like I have missed. I am ASSUMING that all this talk about nLite makes the old steps unnecessary???
And then what happens if you have to upgrade a driver. Is that going to be something that can be done after we remove so much. A recent example from my non music computer. I put in an ssd and had to re-register windows. It simply wouldnt work. It kept saying I had to call in, but there was no number listed and I couldn't activate windows. Turns out I had turned off the Telephony service, and that was needed to get the MS telephone numbers.
Anyhow I am not being negative. I agree with Bibi01 and want to make this the community it was. I appreciate that some of you are taking the lead and really running with this, but if there is no vehicle for others to follow then this whole endeavor is diminished IMHO.
Bibi01 has a great idea about starting new threads. Make sure we put them on the main computer audio site so they are more responsive and easier to search. Dont tie them to the existing cmp or cplay threads. Kind of like the old and new testaments.
No one here remembers the bending of our minds
with FontCreator. For example now arialbd.ttf is 67KB, l_10646.ttf is 36KB and micross.ttf is 54KB.
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2 Ru, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Liv
Are you serious?
H'm... of course!
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2 Ru, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Liv
Open your registry here:
1. HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Control\Session Manager\Memory Management
in the window to the right find: DisablePagingExecutive, rightclick, edit ener 1 to disable.2. Find this: HKEY_LOCAL_MACHINE\SOFTWARE\Microsoft\Windows NT\CurrentVersion\PerHwIdStorage Exterminate this s*t.
One entry above:
HKEY_LOCAL_MACHINE\SOFTWARE\Microsoft\Windows NT\CurrentVersion\Perflib
add a new DWORD Value named Disable Performance Counters, set it to 13. HKEY_LOCAL_MACHINE\SOFTWARE\Microsoft\Windows NT\CurrentVersion\Windows
and set GDIProcessHandleQuota 500(1280) and USERHandleQuota 16000 decimal.
Hope You like it.
Serge.
Edits: 12/13/11 12/13/11
This seems THE RIGHT WAY to put it:
HKEY_LOCAL_MACHINE\SOFTWARE\Microsoft\Windows NT\CurrentVersion\Perflib
add a new DWORD Value named Disable Performance Counters, set it to 1With blank spaces NOT DisablePerformanceCounters see here
http://www.pcauthorities.com/pc-speed/how-to-disable-performance-counters-in-windows-xp/
see here also
http://technet.microsoft.com/ru-ru/library/cc737243(WS.10).aspx
Edits: 12/15/11
Thanks Serge,
I had make exactly that mistake. Corrected it, and very noticeable improvement in sound.
KC
"Disable Performance Counters" was already present in the registry and was set to 1. It was an Nlite installation. I confirm there are blank spaces between the words.
add a new DWORD Value named Disable Performance Counters, set it to 1
According to the link below, the correct name for the D-Word (at least the EN version) is DisablePerformanceCounters. Can anyone confirm?
Hope you like it
Indeed I do. Definite progress here.
As there is with using ResHacker though the "standard" file needs work for use with networking at least in my setup.
Many thanks.
=According to the link below, the correct name for the D-Word (at least the EN version) is DisablePerformanceCounters.=
And all already wrote as with Disable Performance Counters well sounds. It is gained that no difference is present. Strange, isn't that so?
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2 Ru, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Liv
me, if you see what I mean. This set of tweaks is subtle but definite. I have first put it with no blank spaces, then saw it on a microsoft techsupport written as I have put it. Can it be that some keys allow such syntax?
Serge
Yes, I see. I think that often the changes in sound only seemingly.
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2 Ru, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Liv
Yes, I see. I think that often the changes in sound only seemingly.I agree - it takes time to evaluate subtle changes. Given the flood of suggested changes in recent weeks, probably too long for reports always to be reliable. But there's absolutely no doubt that there's been big improvements overall.
Whatever, Serge is right here. I tried the Reg Edit on a standard PC: if the D-Word is labelled DisablePerformanceCounters, Perfmon still runs (Start > Run > perfmon.msc). With the spaces in the label, it doesn't, i.e. the Reg Edit works as described. Apologies for the false alarm.
Now that "Resource Hacker" is Tweak-of-the-Week, inmates using cMP^2 headless over a LAN [i.e. properly :> )] might want to try slimming down WinVNC.exe. I brought it down from 1.5 MB to just under one MB. Whether that or the perfmon thing is the cause, there's definitely been an improvement IMS.
In a day or two, I'll restore the old VNC file and try again.
Dave
Edits: 12/15/11 12/15/11
;)
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2 Ru, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Liv
this time. I was also surprised, as I saw two variants in the net, but microsoft techsupport says it has to be three separate words. BUT THEY MAY LIE? Try, please both variants.
Serge.
Hi Serge,
Tweaks had been arriving fast and furious, not that I am complaining, and I want to be sure of what I am hearing before I post. But this 3 mermaids tweak is another great discovery from you. The sound is now more rounded, a fuller body with an organic bass, and the sound stage is now more balanced front and back, left and right. Together with the slimmed shell32, the sonic is now very continuous with much finer grain.
Keeps the goods coming and thanks.
KC
Thank you so much for what you, and the others, have done. The tweaks never ends!
Hi Jack I have a couple of questions on the G31 mobo memory power supply tweak. See link.
http://www.audioasylum.com/forums/pcaudio/messages/9/99863.html
.
Thank you I also replied in the other thread.
Will try 1 later then 2.
*
Just did all 3 initial impression is positive...less upper mid grain and hardness which increased just a little wee bit after paring more files with resource hacker. Now it has a very liquid high frequency grain structure. Very nice...I always know that when you tweak you have the same sonic goals as I and I don't really have to worry and this time you did not disappoint. I like it!!
Edits: 12/14/11
hello, everybody. Folks, please tell me if I can remove the *.rrr files from config that are left by registry mechanic? Can anybody post updated list of registry files to absolutely remain in the config folder. I remember that some were possible to delete, like userdiff or smth.
Help, please.
Serge
default
sam
security
software
system
that's all,
and their log files. You can delete them but they come back.
Tried deleting sam and security but you cannot and it screws up your BIOS which you will have to enter again. With your BIOS settings this is something one would rather avoid!
Add /MININT in boot.ini and they will not come back ;).
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2 Ru, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Liv
You've been holding back on us!!!!
How long have you been using this trick?
About two hours ;).
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2 Ru, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Liv
see the post on kernel and hal? What do you think? Is it possible?
Serge
with "/HAL=" and "/KERNEL=" options :(.
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2 Ru, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Liv
I guess the following can be done the same way as AWE:
Would anyone try it with cmp-cplay? what other programs are running undercover?
A common question is how can I restrict which programs can be run on Windows? With Group Policy (or Local Group Policy on a standalone machine) you can do this, though it takes a bit of work. Here's how it works:
In a domain, open the GPO linked to the container holding the user accounts you want to restrict (or on a standalone computer use Start --> Run --> type gpedit.msc to open the Local GPO).
Expand User Configuration \ Administrative Templates \ System Open the policy named Run Only Allowed Windows Applications Enable the policy and click Show
Click Add and type the executable name for each program you want to *allow* the user to run
Once the policy is applied, the user will only be able to run the programs you specified and no others. Unless you have allowed them access to the command prompt cmd however, in which case they can run pretty much anything if they can find it.
If this can be done from a host computer it would allow all kinds of neat things to be done.
Other than the obvious (cMP and cPLAY) what else has to be put in there?
Another thing to play with tonight.
No computer expert am I, as you know, but if we can use the host computer's mmc to change things within our cMP disk drive ... that would be great.
Hello All!
I have been sticking with 44.1, aka no oversampling for the 3 years or so i have been using cmp2. I have read lots of bios and registry hacks lately, but am wondering a few tings, and not much turned up doing a search, so i thought i woul simply ask:
what basic settings do you use, e.g. svchost, awe, priroity, buffer size...
more importantly, what settings do you use, if any for oversampling, and what sonic changes do you hear when you engage or deactivate upsmpling?
do you use an internal soundcard to do the d-a conversion, or output i2s or spdif? and what external dac do you use?
i have been comparing jriver via ethernet w/ cmp2 via spdif into a number of new network dacs; ps audio pwd/bridge, marantz na7004, and the musical fidelity clic.
any comments would be appreciated.
I too am running the G31 / 7200 h/w. But you did not answer whether you did all the cmp optimizations...these are key to sound and a bitch to do by the way. The stuttering indicates to me that you did not do everything. Not trying to embarrass or harrass you just trying to get you better sound.
I guess I should ask you what tweaks you have done to Windows if any. While the settings are certainly very important more important is doing all the Cics recommended windows tweaks... slimming down the OS ,eliminating features, very important is minlogon, windows kernel stuff. These are all described in detail on cics site. Do these then the other tweaks called out on AA the last few months. I'm not saying you cant appreciate CMP/Cplay w/o these but these will definitely optimze cplay. IMO CMP is more important that cplay ...do you use cmp?
theob,
i am using an earlier hardware recipe, with the ga31 mobo, 27200 cpu, in an antec case. soundcard is an essence stx, with a swenson psu running the card via the molex. there are caps added to other psu rails internally. power cord is a shunyata, and it is plugged into my shunyata talos conditioner. i did many of the bios optimizations, and do minlogin.
my questions were more concerned with the upsampling, and other settings. i've played with buffer size a bit, (currently using large)to avoid stuttering at the beginning of playback.
until yesterday, i was Mr. Native resolution all the way, but when i started trying various upsampling options, i liked 192 the best, even though 176 is an integer multiple of cd flac files.... the sound was fuller, more natural, and the soundstage depth expanded pretty dramatically.
i am currently playing with the ps audio perfect wave dac, with the bridge installed, so i have lots of permutations to try. with the ps set to native, cmp/cplay won hands down over jriver. with the upsampling options either within cplay, or via the dac itself, i am able to more closely level the playing field, but i think cmp/cplay may still sound a bit better.
anyhone else want to chime in on this upsampling thing?
Mark
I use no upsampling. Relative to svchost(suspend and also for lsas), awe(yes but its not that easy you have to do some initializing in windows---do a search under cics) , priroity(critical), buffer size (tiny). I was into 192 upsampling but now prefer going native. When @ 192 I get more highs, less bass and dynamics vs native. Although this a personal preference type of thing.
Inmates! After Mario's tweak I got positively heated and suddenly remembered, that nLite installation removes Volume shadow copy. So Why I am having those two dirty files in my system?
I have removed (about 25 minutes ago) volsnap.sys from drivers folder and vssapi.dll from system32. I think the thing is worth doing.
Serge.
Try it, please.
OK, volsnap.sys can be removed. vssapi.dll is absent in my system.
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2 Ru, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Liv
Hello, Serge! Can you tell me when did you remove it if you keep track of the events?
I couldn't until I have made a fresh install with nLite without Volume Shadow Copy.
Serge.
vssapi.dll was removed earlier in My Tweak 2.
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2 Ru, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Liv
Edits: 12/12/11
Studying Mihaylov and steppe screenshots and implementing steppe's last boot.ini tweak (OMG this sounds good thank you!),
files like shell32.dll are indeed one of the 'core' data sets Windows requires for operation. With the superb sounding /maxmem=256 switch it happened to me that some cue playlists could not be loaded, too many split loads. So either make new cues, OR try to squeeze the last bits of memory :)Friends, it is possible (using a resource editor) to eliminate unneccessary Icon, Bitmap and AVI (!) resources that are embedded in files like shell32.dll and explorer.exe
And this
http://www.msfn.org/board/topic/87943-replacing-shell32dllwas actually my starting signal :) and made me wonder how much RAM savings this could lead to. Being that cMP is a replacement shell with uses the same 'core' dlls as Explorer shell; but DOES NOT use all the desktop, file browsing, printing dialog stuff it also makes sense.
With a bit of trial-and-error I could safely reduce the sizes:
explorer.exe size is now 388 kB, original 1008 kB
shell32.dll size is now 2686 kB, original ~8000 kB
taskmgr now shows explorer.exe eating up ~5800 kB, previous was about 10500 kB RAMFirst I looked how big shell32 and explorer from the Windows XP SP2 Embedded pack are (remember, we have our minlogon.exe from this pack). Of course, the are the same monsters as in our XP (shell is even larger). Then just trying reshacker (Resource Hacker), its funny to see that for example the animations in Your Windows file copy/move/delete dialogs are some AVI(!) files that are played back when you see the dialog... Which of course does never happen when we are in cMP shell, but is nonchalantly eating up RAM which we want for MUSIC.
Second, all system Icons are stored inside in 16, 256 and Hi-colour resolutions. Of course after we switched to 256 colours in Windows you only ever, ever see the 16 colour versions. Remember, even the tiniest things are cumulative. To some this very idea may seem over-the-top. Rough assumption, but shell32.dll seems to be 20% program code and 80% multimedia resource if you open it up.
--SNIP-------------------------
To make thinks clearer now: You DONT GET TO SEE the mentioned '20% program code' at all in ANY of these resource editor applications. Theyre just there for editing resources, there is really no reverse-engineering, decompiling etc involved here. So don't be afraid you could mess up the program code along the way, this is much more deep inside :) (where monkeys can't see a thing)
--SNIP--------------------------After deleting AVI and (many) ICO and bitmap stuff, plus editing the dialogs that use them (which is easy, I could explain in next post if you like ), as stated shell can be made about 25% of original size.
Please dont be afraid of this tools' name, in the right hands its (quite) safe to use (actually uses a Windows API to access embedded resources, not program code from the designer) Looking the web it seems also common for freaks to edit their shells (usually to make it more PRETTY, but we audio nuts care for the opposite)
I also like the connection which seems to exist between Mussels, Shells and Mermaids :) For sound quality I cannot really say if better (or just different) as I have new 'boutique coupling caps in my DAC which are breaking in.
If Youre past steppe 22 (use no explorer) and like the sound I can send you my ATM super-tweaked shell32 for trying out etc. Keep in mind its probably safer to edit your own files, because of language and file version differences. I have SP2 ENU.
respects,
Mario / nagual19
btw, haven't tried if using Win2k dlls (as suggested in link) works and I dont want to make anything instable, the sound is just so fantastic.
Edits: 12/11/11 12/11/11 12/11/11 12/11/11
Should we delete all lines except for those that say 256 colors? Or something else?
Edits: 12/15/11
I was able to get a limp along version of ADD / REMOVE discs into cmp. Restore Dialogue 1079 (as per Mihaylov...thanks again Serge)and Strings 267 1893 1907. What you get is the ability to work it but when you press add / remove you get a folder tree that omits some of the names but if you click into spaces where the folder names should be you see a little dark shaded area, click on that and you are ok.
Also for those of you still using setupapi you can delete about half of its 950kb size by deleting AVI and other stuff within it by resource hacker.
I substituted Nagual19's shell32.dll in my xp sp3 based cMP setup with no problem and clear improvement. Then comes Fishy Secrets 3, so I wanted to use regedit again within the system if possible. So I substituted back the original shell32.dll. Then I found when trying to start regedit I get an error saying there is no associated program for this .exe file and to set it in folder options. The same happened trying to start cmd.exe. So in the substitution and resubstitution that association was lost or cancelled. Anyone know how to restore it with things at this stage?
sorry I cannot remember the name (these days it takes to long to load a page within this thread I have lost my patience).
OH YEAH, PEChecksum. I bet that would make a difference.
I have not used it myself and the link in the post did not work but I am sure you can find it somewhere.
Did anybody ever figure out how to renable add / remove discs from cmp screen?
Here so:
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2 Ru, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Liv
So I understand Dialogue number 1079 but which String number is it...1079?
It doesnt look like it. Should I try to identify a string table with 13 lines in it? God I would love to learn Russian (my late Mom could understand some russian).
I do not have English file. Therefore identify 13 lines in String under their numbers: 4162, 9092, etc. These are inscriptions in a dialogue window 1079.
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2 Ru, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Liv
thanks
I have not tried it though.
http://www.droopy.narod.ru/BootIni.htm (in russian)
and here is the best explanation I have found so far:
/KERNEL=/HAL= Enable you to override Ntldr's default filename for the kernel image (Ntoskrnl.exe) and/or the HAL (Hal.dll). These options are useful for alternating between a checked kernel environment and a free (retail) kernel environment or even to manually select a different HAL. If you want to boot a checked environment that consists solely of the checked kernel and HAL, which is typically all that is needed to test drivers, follow these steps on a system installed with the free build:Copy the checked versions of the kernel images from the checked build CD to your \Windows\System32 directory, giving the images different names than the default. For example, if you're on a uniprocessor, copy Ntoskrnl.exe to Ntoschk.exe and Ntkrnlpa.exe to Ntoschkpa.exe. If you're on a multiprocessor, copy Ntkrnlmp.exe to Ntoschk.exe and Ntkrpamp.exe to Ntoschkpa.exe.
The kernel filename must be an 8.3-style short name.
Copy the checked version of the appropriate HAL needed for your system from \I386\Driver.cab on the checked build CD to your \Windows\System32 directory, naming it Halchk.dll. To determine which HAL to copy, open \Windows\Repair\Setup.log and search for Hal.dll; you'll find a line like \WINDOWS\system32\ hal.dll="halacpi.dll","1d8a1". The name immediately to the right of the equals sign is the name of the HAL you should copy. The HAL filename must be an 8.3-style short name.
Make a copy of the default line in the system's Boot.ini file.
In the string description of the boot selection, add something that indicates that the new selection will be for a checked build environment (for example, "Windows XP Professional Checked").
Add the following to the end of the new selection's line: /KERNEL=NTOSCHK.EXE /HAL= HALCHK.DLL
Now when the selection menu appears during the boot process you can select the new entry to boot a checked environment or select the entry you were using to boot the free build.
/HAL= BcdOSLoaderString_HalPath
/KERNEL= BcdOSLoaderString_KernelPath
Serge.
Edits: 12/13/11
I used to get 2 min and 14 sec before a swap. Now after I have done some resource hacking I get a swap as early as 1 min 59 sec. How did I lose 15 sec? I thought shrinking these files from within added memory. Anybody else notice this?
Sounds great but what physically is happening?
Do you mean the size of a chunk into which a song is split? It is dependent on the amount of ram availible, With /maxmemory=216 on my machine a typical song of 4 minutes is split into 3 segments (chunks). Also, the shorter are the timings of the memory, set in bios, the more physical memory cmp needs. At the slow and slack rate of standard cmp optimisation it needs 214 mb, while with most agressive timings - 217.
Serge.
By memory swap I mean music elapsed time at which the 1st memory swap of new data(music) occurs. So in my system I have 256 mb physical memory, I run 44k sample rate, tiny buffer. I use onecpu/maxmem=256/nopae in the boot.ini file. I used to see no memory swaps before 2:14 now they occur at 1:59 (after Resource hacker) which is opposite to what I expected. So my question is very simple why is memory swapping 15 seconds sooner than before?
But if it sounds better, let it be?
Serge
I do agree it sounds better but I'm just curious about what is really happening.
Here is one I took Mario's shell32.dll and eliminated all Dialogue stuff and got 2.595 mb....cant tell a difference yet on sonics. Anybody else?
Edits: 12/12/11
I prefer to remain passive and wait for other people to finish their work, as I continue with memory settings. It's good that we are splitting now our activities. This way we can work on several problems simultaneously. My report You could have read already. Most positive. But I would be even more happy, if we manage to reduce the impact of winmm.dll as well. Did you remove the volsnap driver, btw?
Serge
No will eventually try it though.
...I get what Dialogue is but what again is a string?
*smile* a string (in computer speak) means just a 'character string', so
strings == the TEXT that gets displayed in dialogs or elsewhere
As you know, things like text documents are (even if they may contain many pages) quite a bit smaller than Videos, or Pictures on your harddisk) Thats why I wrote, might be overkill. But we want to squeeze every BIT :)
Thanks for the info...I'm starting to delete these too.
This is quite extraordinary!.
RESOURCE HACKER is simplicity to use but I am concerned about the PEChecksum aspect since I have no experience with working at the commnad line.
Hoping someone will supply a step by step for this? I have minimized many .dll's with the obvious things to delete but am afraid to insert them just to see a blue screen.
Personally, within shell32 I deleted everything since I never see any of that stuff - figure I cannot go wrong, but will have to wait to see.
Makes one wonder just how much further we can go? Just when you think we are reaching the end of the line AND THEN something like this gets posted.
Thanks, again,
I deleted all avi; all icons except the cd s; all dialogues except file open related. Size is 2,431. It works perfectly, and sounds amazing.
I will likely take more out. I have dual boot so its easy to toggle between full xp ans cpm lite.
Which string did you leave for CMP ADD / REMOVE discs? I see Mihaylov's post that says to leave Dialogue # 1079 in but I cant figure out from the russian which string number to leave.
I did all dialogues and all is well with respect to cmp^2 functionality. But thats all I did for now, sonics I'm still evaluating.
Just using Mario's file is a significant sq pop.
is how I perceive it.
As with this continuing evolution beginning with Serge's idea, there is much to be gained by minimizing WINDOWS.
Went back to my non-ACPI version because the ACPI one was doing strange things. I think there must have been a problem with my install disk. Will have to try again.
Adter I got my confidence up withe RESOURCE HACKER I went through every file in system32 that had something to delete and basically deleted everything except "version" and the one's with all of the numbers. I left very few of messages and the strings and will go back and see if I can eliminate them all. I do not think they have anything to do with what happens with a cMP machine in cMP mode with no EXPLORER.
Whether being able to get rid of everything makes much of a difference in file size remains to be seen. More importantly whether that will make any sonic difference at all (after the inital slimming) is doubtful but IT MYST BE DONE!
I figured it would not work but it did. I thought I had already taken too much away.
Who knows if manipulating the other files have the any impact like shell32 since they do not show up on TASK MANAGER (when you could access TASK MANAGER) but there has to be something worthwhile to making all as small as possible.
As we near Serge's goal of a 20mB WINDOWS ...
as this process proceeds, how easy it is to tell the condition of the tape on older recordings.
Not that anyone WANTS to hear these artifacts but they do illustrate how fine the resolution has become with this decoder.
Luckily, the noises are not obtrusive, as in LP playback they are there but do not overwhelm the sound. Surely an indicator of how well behaved this system has become.
Yesss! Just this morning while listening @ 3:00am (insomniac) I noticed this and made a mental note of it.
Rick thanks for reporting this. As You wrote 'I left very few of messages and the strings and will go back and see if I can eliminate them all.'
Yeah, at least for our beloved shell this should directly lead to more and more bits of RAM savings. Because from OS engineering standpoint it makes sense, all GUI related things have to be kept in RAM, so that they can get quickly rendered when the user clicks somewhere on the screen. (and not loaded from some dll on hdd)
With other files that get loaded alongside shell and explorer it's quite the same I think. Except the dlls that are not kept always in RAM (assumption)
Mario, if You haven't deleted the registry entries for known dlls, please, open with notepad cmp/cplay exes and check, first, all the dlls that the 'perfect pair' uses from this list. Try, please, to hack them as brutally, as only You can, and next proceed to the others. Am I asking too much? I know it's not simple, but please, if you have time, do it.
Serge.
BTW, Serge Mihaylov was on a bright way to try hacking hal and kernel and, possibly gdi32 and ntdll.dll. He reports that they are being checked for integrity by windows on start up, however, hall and kernel CAN BE SUBSTITUTED for different ones with an appropriate setting in boot.ini. That's for sure, I have read it. If You are ready, I will re-find the corresponding switches for boot.ini to load different (read hacked) hal and kernel.
Serge.
Edits: 12/13/11
=I will re-find the corresponding switches for boot.ini to load different (read hacked) hal and kernel.=
Serge, what's this?
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2 Ru, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Liv
In my ramblings along different articles on boot.ini and its possible settings I have come in several places on switches that ALLOW to use a different hal and kernel for debugging? and for different "experiments" I don't remember exactly, but it's something easy, like /hal=yourhackedname.dll and /kernel=something. As it is dangerous, please, give me some time to find these switches. I also think they are at wikipedia, where I got the illbegotten /clklvl.
Serge.
Mario in the spirit of your encouragement to try stuff I'd like a little guidance on the method you used to generate such a great sounding (great is not an understatement) shell32.dll. Do you just delete whole folders of AVI, ICON and DIALOGUE files? Or do you leave some stuff in each folder? I listened intently last night (at my low powerline noise hours) and the sonics of the slimmed down shell32 is nothing short of fantastic. The highs, inter space silence, soundstage (depth and width) is the best I experienced in my system. What a find!
This is a breakthrough!!
Best SQ ever. I went through Shell32.dll fairly quickly. Killed AVI, icons and even tons of dialogues. Went on to look at a few of the other larger dlls in sysem32 and removed dialogues and error message strings.
There is a lot to experiment with here.
Thanks Mario for that you reminded me of such possibility to tweak files ;). Now I tweaked 39 files in system32. The Windows's size is 29.1 MB.
The size of shell32.dll is 2194KB. "Add folders" in cMP settings works fine! To upload english shell32.dll and I will tweak it this feature work.
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2 Ru, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Liv
Dear Serge(s) and co-workers!
It really makes happy that this tweak gets so well received here. I now think about it as a natural progress of the 'Art of Windows tweaking'. The improvement this shell32 reduction seems to make is in the good tradition of minlogon, because these very files are used by Windows all the time and are not 'peripheral' components like many files we have deleted in the meantime. This was also what made me look inside shell32 from SP2 Embedded Pack to see if any smaller, but it wasn't.A few ideas:
I want to get somewhere the Win2k SP0 shell32, open it up and analyze how it's smaller than XP version. Because MS always has to think about compatibility when developing new versions. And this brings some structure in tweaking. Remember I have posted a link where some nice tweakers talk about using W2k versions of these files. But I dont make to want a frankenstein Windows (at least YET)Now Serge (Mihaylov), did you just delete Icons, Bitmaps and AVIs or also Dialogs? I had already the feeling, that we can make shell even smaller by deleting almost all dialogs (except the ones that get called from cMP plattform, namely via the add-remove library buttons. Of course, that many if not all Windows files could be opened with resource editors and made smaller (Your are at 29 and counting? :))
Another idea, if You open up the section -Accelerators- These are used to define Windows Shortcut keys (We all know the Windows key Strg+C Strg+V etc). It could very well be that Windows is constantly polling (our deleted) keyboard drivers, if any of these keys are pressed - another disturbance in shell! Possible!
I want to look into this further. To all Friends asking for my current version of shell32, I will of course send it. But You are encouraged to try yourself, that way we may discover more than a single person could. It is really quite easy. I am amazed at the speed Mihaylov is at... which lets me wish I had more time for this -
Mario
Edits: 12/12/11
=did you just delete Icons, Bitmaps and AVIs or also Dialogs? I had already the feeling, that we can make shell even smaller by deleting almost all dialogs (except the ones that get called from cMP plattform, namely via the add-remove library buttons.=
Mario, your feeling did not trick you ;).
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2 Ru, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Liv
Wow that's fantastic. Is it possible to make some of these files available to us even at the risk again of maybe not working in our XP?
These are russian files (with russian language) and they are not suitable for English xp. Upload the English shell32.dll and I will tweak it.
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2 Ru, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Liv
tweak winmm.dll in such a way that it will only work with wave audio just remove midi and the rest of the crap inside?
Also, do you need an already tweaked or intact english shell32 file? I can send you both through e-mail. Can You, please, post the list of 39 files to tweak? Did You work with reshack or resource tuner?
Serge.
Hi Serge!
I work with Restorator.
I need the intact english shell32 file.
The list of 39 files to tweak will be later perhaps (these are largest files from system32).
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2 Ru, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Liv
Don't try to hack following files:
hal.dll
ntoskrnl.exe
user32.dll
win32k.sys
because apparently the system checks their integrity.
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2 Ru, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Liv
Hi Mihaylov,
Somehow, I was able to remove all the Bitmaps inside ntoskrnl.exe. I also removed all Menus from user32.dll, others couldn't be removed. They were working on 2 computers respectively.
You may wish to use other hacking software to achieve the results.
Hi Jack!
OK. I will try again.
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2 Ru, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Liv
Works!
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2 Ru, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Liv
What software do you use which displays multiple dlls?
Jack, look here:
http://www.AudioAsylum.com/forums/pcaudio/messages/10/100161.html
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2 Ru, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Liv
using alternate switch in boot.ini, but the tweaked hal and/or kernel should go under different names. If interested, I can possibly re-find these switches.
Serge.
Edits: 12/12/11
Unless you have all of your library loaded you cant add to it from the cmp screen. At least I cant. Shoot this is a biggy for me.
Add folder button in cmp? But what if you simply copy the folder with wavs and cue into, say, general RIP folder on another windows, then reattach your hdd to a cmp. Will upon loading cmp rescan and add new item to the library?
If Yes, I don't think it matters to me, as I have 3 general folders RIP, POP, CLASSIC. Inside these folders I have subfolders, like DEEP PURPLE and inside it: 1970 In Rock, 1971 Fireball, 1972 Machine Head. Inside each I keep my corresponding wavs by tracks and a cue. I often keep a subfolder Covers or Sleeves here as well.
Serge.
Another workaround for me is to go back to the 8mb shell, load all my 3 discs, reload the 2mb shell and wala I'm into all my music. But it would be nice to be able to remove / add discs from the cmp screen.
...but I got there. Yes it is a nice big sq pop ... very nice. I agree with Steppe.
Thank you so much Mario.
Oh, YESSS!!! She is. We had a good listen for 45 minutes, and I can say that the sonic change is most outstanding and is getting better, Better, BETTER!. The highiest mids are so clean now that it affects the lowest counterpart. It became very multilayered, textured, rich and polytonal. The overall change is for the best. The stage is high, very deep, in fact it's abyssal. the vocalist now stands 3,5 - 4 meters from the frontline of speakers. But the best things are happening in the background. It is rapidly becoming the arena of many musical and sonic events. Happily, there is no sonic hole between the front and the "backest" back now. The sound is getting lighter, meaning less dark. More sounds of the singers' throat and lips, when he (it is she in my case) is trying to help herself in a complex place.
Technically it was a very easy thing to do - just remove old dll from another windows installation, put the tweaked one inside and return my esata hdd to cmp.
the Mermaid seems so satisfied with the change that she asks her permission to bestow upon You the title of The Father Of Pearl, if You will accept it.
Serge.
It inspires, inspires me.
As I read Your message I cant help but smile and also being fairly proud... I really like Your descriptions of sound qualities.
Of course being The Father Of Pearl is... well... thanks :) Btw this encouragement is what I need ATM in my life, I have to get a job soon. My last one was not good and left me almost burned out. I believe if one is happy inside it will automatically lead to the right People and events.
Spread Your wings and get ready! It won't take long, if you feel really inclined to find a new occupation/job. I felt I was "growing a layer of fat" on my brain, it's the closest to a russian saying, and found myself a job immediately. Take a rest for some time and when you feel, you're ready, start searching, and it will find you.
Serge.
I d/led the reshackers and I follow your posts but still feel somewhat gunshy in attempting any of this. Maybe I just need to study it a little more. Also resetting checksums discussion is a bit confusing to me. I think you understate your level of computer acumen. But I will try to catch up.
Theo, thanks for posting. It may seem a bit much at first yes. Please read
http://www.thinkdigit.com/forum/tutorials/31112-tutorial-all-about-resource-hacker-brief-tutorial.html
if you like and have some time. After removing resources using your reshacker, U just open up a command prompt, go to the directory where you have both PEChecksum.exe and your newly tweaked DLL, and run the command
PEChecksum -c shell32.dll
This is it. Btw I have just found out this step MAY not be necessary as the checksum is only verified if you have WFP enabled (we all have disabled Windows File Protection already by implementing minlogon)
If you just happen to have Windows XP SP2 English-US (United States) You could use the files I have sent. ^^
try it and report in an hour.
Serge.
Thank you too for all your hard work! Take time because I also had some learning curve to do this stuff just a few days ago.Eventually U will stumble upon a Dialog named 'Windows Default CD Writing Wizard' in shell32-Dialogs:
http://computertruths.blogspot.com/2007/06/interesting-hacks-using-resource-hacker.html
(interesting read)I do think - for the sake of making our shell cleaner - these dialogs (-> 1125 to 1138) could be terminated w/out ill effects. U get the ideas... We have deleted the CDR burning components a long time ago, and I just think these Dialogs could therefore never ever show up anywhere, because there is no possibility these dialogs get 'called' from any program routine anymore.
Edits: 12/11/11
You said '...To some this very idea may seem over-the-top. Rough assumption, but shell32.dll seems to be 20% program code and 80% multimedia resource if you open it up. After deleting AVI and (many) ICO and bitmap stuff, plus editing the dialogs that use them (which is easy, I could explain in next post if you like ), as stated shell can be made about 25% of original size...'
It would be great if you posted exactly how you size reduced shell32.dll. Also did you do from a host computer? Does Reshacker work that way? Any further deails would be appreciated. Here is an idea: give one simple example of something you did successfully to free up memory.
a charm on any computer. I have just downloaded the prog, unpacked and loaded dufferent dlls and exes from several hdds. The reshacker eats them eyes closed. It doesn't seem to care. But How and What to remove, this is what scares me. I would tweak this way many big files that remain. And, certainly cmp-cplay (not much there, in fact) just to get rid if possible of the oles and winspool.drv
Serge
1. This shows shell32.dll remaining resources in reshack:
http://i43.tinypic.com/2cqi2xu.pngSerge, yes just copy your existing explorer and shell32 files from cMP to another computer for modify, then copy back. Please also make some backups of these files. If you compare my screenshot with yours You can figure out what can be deleted. But TESTED this only with deleted explorer.exe (I use multi-boot for this stuff). This screen shows maximum for now. I did not want to go further.
2. And this shows the tedious work to do with the DIALOGS (at last, if you remove most Icons)
before
http://i44.tinypic.com/zmia9x.pngafter
http://i41.tinypic.com/21l4a5k.pngYou see a rendering of each dialog and a script window showing the used Bitmap, Icon or AVI resources (they have IDs pointing to). You point-and-click the Icon or 'sysanimate32' (AVI) control and hit delete. If you try, it seems more difficult than it really is. Remember deleting even strings (I know, make a facepalm!) as I did is not necessary at all. And MAYBE most of these dialogs could be terminated completely but I did not dare. Being only text there isnt much to be gained, most is graphical stuff.
3. FINALLY you have to run a tool called PEChecksum.exe to rewrite some checksum that has been modified.
http://www.materialordner.de/kyE5JHblwgs4QO43lmdwOrXCKb4vZPH.html
click on "Downloaden von pechecksum.zip" password please use my nickThen open command prompt, go to folder where your files reside and type
PEChecksum.exe -c shell32.dll4. COPY back to cMP and done! To those interested I could send you my shell32.dll that you can directly open and compare. Would need too many screenshots but I did my best!
Please also see attached link (post #40 is important) and happy reading.
Mario
Edits: 12/11/11 12/11/11 12/11/11
Yes you are on the right way, give me some time I will prepare a post to show how and what I have removed w/ success :)
can cmp.exe and cplay.exe be tweaked in the same way - what I mean is to remove certain parts of the script that use ole32 or oledlg and ESPECIALLY winspool.drv?
Serge.
Good question.. IMO this would only be possible with either decompiling the exe or analyzing cics' source code (don't find a source). AND having a much deeper understanding on how dlls like OLE are used in Windows 'global hierarchy'. Btw I am also no programmer but have experience in IT servicedesk using our BELOVED Win XP :)I really believe cics has coded already in a way, that the fewest possible dlls etc are getting used.. and We can only do our best to streamline all these components to suit it 'best'. This is what iE minlogon is also about for long times now.
What you can do using resource hacker is replace GUI related things in cMP and cPlay. Respectfully tweaked B39 SSE4 executable is shown above, I put in oldschool icons and black and gray (I dont want cics take this modification as disgrace. but Respect for his great player!) Also, cics has used much simpler (IMO prettier) icons in old builds. When I get a hold of these, I extract its icons and 'paste' it on the B39.
Keeping the spirit and only dare to mod icons keeping the same size and colours :)
Edits: 12/11/11 12/11/11 12/11/11 12/11/11
I would be disinclined to monkey with cics programs generally, but I would love to know how to reduce the font size so perhaps whole titles would show. Better yet in my case would be if someone else did so and posted the resultant programs.
I know this may initially seem like monkeying around (please get the pun with my shown Nilsson playlist and track title!)
o.O
But the answer is No. Because One can only replace resources that are designed to be just that, resources for the GUI. They get loaded along with program code when one starts a program. The font rendering of track titles in cPlay is 'hard-coded' and hidden in program code (so cics design choice), and can never be changed with tools like reshack. They are not accessible. Better that way, I happen to think too :)
Too bad.
No deregation intended. Monkeys are a higher order of animal than fish. Mermaids are harder to classify, of course. Still I wouldn't want all of us to go ape over reducing system/file size for its own sake.
Hello! This is what I was dreaming about - to remove the shell32.dll
-At the end of this post - http://www.audioasylum.com/forums/pcaudio/messages/9/97960.html I was expressing anger at not being able to.
Now You have found the way to practically neutralize it. really, working like a group we can move ever further! Please tell me, if the tool you are posting about is simple enough for the likes of me. Can I download it, copy my file from cmp machine to the working one, tweak it and replace? does the prog have some kind of user interface? Can you, if possible post some pics with explanations? I feel kinda scared...
Also can you try to tweak (if there is anything tweakable? WINMM.DLL?
Serge.
What is SP2 ENU is it english uk? I would be happy to try, if I may get Your version. I have original us version of vl windows. Hope my mermaid will eat the shell with pleasure...
Edits: 12/11/11
I see in Mihaylov's latest registry screenshots that he has only 1 numbered control set and current control set. I had already posed that one should be able to do without the other set. Now I have tried but my fully reduced cMP partition won't allow it to be deleted in regedit. I exported a hive and imported into another cMP less stepped partition but regedit wouldn't allow ControlSet 002 to be deleted there either. Booting to a PCRegedit CD and trying it there also wasn't possible. How can it be done? Mihaylov, how did you do it?
First to try make to be so:
However I find it difficult to tell precisely as I made it.
See my latest registry list (link below, password: Serge).
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2 Ru, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Liv
I could not remove the other user S-1-5-20 either.
Worse, I know see that the registry has been restored and that the recent deletions guided by your screenshots from HKLM control, enum and services have been returned from somewhere. Where could that be and how? I have no .SAV files in system 32/config. Could windows file protection do this--especially despite its having been turned off long ago and showing no signs since of reappearing? I made the deletions directly with restored regedit function in the same cMP install.
This is making me crazy. It has undone hours of work. HELP.
Hello! Yes, Windows IS restoring part of the registry. There is a dissertation on this and similar subjects by a woman, who is one of the spokesman for Macrohard Qs&As site. She wrote, that registry entries are different in a switched-off computer than in a working one. This feature is used to extricate data for criminal investigations, for example. Control set 2 is used in case current and set 1 is broken or changed as they are The Working sets, so the secret Hydra (search and you Will find it in the registry) is relying on set 02 data for restoration of itself. It is written in a highly and scientifical style but the essence is primitive enough. This pdf can be found on the net. I have asked several times What SQ change such reduction brings. No reply as yet from anyone. Removing unnecessary files and entries that are relevant to the services that we delete anyway, is one thing, but to delete a selfrestoring part of the registry just for the sake of champion's size is another. Thw WHOLE registry is read on start into the non paged memory and only when you say good night, this memory will be released and the data stored back after you can do nothing else to it.
Serge.
Edits: 12/10/11
since it all gets put back, or are you saying one can slay the Hydra? I now see that huge amount of deletions to HKLM/software have been restored. These are mainly for programs no longer installed or not used. Why would these be restored? And the restoration in the system hive seems to happen for deleted or blocked services, too.
Hello, again. No the Hydra is possibly immortal, the name speaks for itself, but, though Internet explorer, some basic system and network services or even program entries are restored, the things audio and secondary services are not restored. Read, please carefully my suggestions on registry tweaks. Those things don't get restored, as well as most things that you delete about hardware interrupts, the hardware resources and so on, I may be not exactly true to the end, but mostly true.
Read Jolida's txt for his registry tweaks batches, there he says which of my tweaks is accompanied by registry changes. Read and print out these tweaks and do everything exactly. It won't get restored. Also, You will note that I recommend to make deletions in registry in all sets.
Serge.
I have done all your steps and used Jolida's registry batches, too. I will check to see if any of these items has come back. But in my case anyway, many clearly non-essential elements in control, enum, software, etc, that I recently removed also came back, so I don't have the impression Hydra can be given all the responsibility for what is happening.
BTW, search of registry shows hydra only once in security\policies\secrets. It is a key that starts L$hydraenckey_[and a bunch of numbers and letters]. Its subkeys have very little in them none of which reveal anything to go on. Searching windows 7, on the other hand, shows no hydra at all.
Can you identify the article you refer to better for finding purposes, please.
First thing. Are you editing your registry or you are editing a cmp registry loading hives?
If the second is the case, are you sure you are editing ONLY within the loaded hive? Aren't you. by chance, editing the registry of another - nonCMP machine? such thing can happen in "heat".
Second About the dissertation. Well, no I cannot point exactly, I was disinterested after first 5-6 pages of real text, page 16-17 and onward. The woman can be found on microsoft technical service questions page.
Just google smth like awe in windows 2003 and see what page directs to microsoft site. Answers will be answered by some guy, mostly, but time after time that woman appears. Then google her name, probably... Sorry... I know it's vague. The hydra is the one I was referring to, but, also. I am not sure that this is the real service that is answerable for our troubles. I make a blind asuumption, I should say, but the name is not just "fishy", it's "snaky" if you see what I mean.
Serge.
No, the heat came after it all was undone.:) I have already said I did the editing directly in my cMP install having restored regedit. I made some other registry changes booting to a cd with PCRegedit that works from outside. Its changes were undone as well.
I tried editing an exported hive on a different computer but when I went to reload the hive it was refused with a message about having signs of a different product. I think I have a golum rather than a hydra.
I was loading and reloading my cmp hive to another (2 different - russian and english untweaked) windows' regedit several times and have done it through esata and usb and under different hive names. Never a single problem. I, for example have only today removed my own nonpagedpoolsize setting and it holds after reboots and even shutdowns. So go, check your golem. It has, probably, eaten smth.
Serge.
As for dissertation, really, I never kept any path for it in favorites, sorry.
Edits: 12/10/11
The Device Manager no longer work after Batch 10. After Batch 16, I was somehow able to reload the Device Manager and saw a number of exclamation marks on some devices, including the Processor. Anyone tried re-installing the CPU driver and the results?
It sounds much better without it. These drivers that are deleted are for very old processors.
I have deleted them and forgotten. Never a single problem.
Serge.
with such a "scary" name I never tried deleting this one.
When I read about it it is an obvious one.
As they say looks can be deceiving!
My list will now become shorter.
Rick, and everybody! Lately we hve been compacting, reducing our registry, including yours truly (Registry Mechanic said, mine is smth like 4,9 now) but noone reported any sonic change or result. Is there any reason in reduction?
Serge.
occupied with some other tweaks at the time. crusoe, amd-something, processor and intelsmth sys and dlls were suggested for deletion by me at the time of part 11-14, can't say exactly, don't really remember.
This is the picture of my folders: http://www.audioasylum.com/forums/pcaudio/messages/9/98163.html See my other folders beside that post, Now I have also removed (thanks to your suggestion of ps/2 mouse) several usb sys files. What about VOLSNAP.SYS? Can it also be deleted, as we don't install volume shadow copy?
I have managed with AWE, but had to remake an NLite installation myself. Tried Batch files, but easiest way was to open windows from another windows, just exterminate every flea inside and copy into empty directory files from my OLD WINDOWS directory. no batches, no nothing... Piece of cake. Certainly, You have to first come all the way of deletions to have such a folder. I don't mean You, I mean other folks who may join later on. Also I agree with you that registry is best tweaked manually. This is what I'm going to do tomorrow. Start with a new installation and make everything manually.
Serge.
Edits: 12/09/11 12/09/11
I need to quit relying on my memory.
Sorry about that.
Of course, it was on the LISTS!
because I thought it was necessary but looking around I see it just might not have anything to do with a machine using the INTEL CPUs.
You have found another useless file that we can do without IF the machine still works which from what I read I think is likely.
As with the intellppm.sys, which the processr.sys file is mentioned with; is not needed, either.
Of course,t he easiest way to get it back, if it is needed, is just to copy it from another WINDOWS installation and put it back into the drivers folder.
Can't wait to hear if your system works without it!
once you familiarize yourself with the registry you see all of the things that the CONTROL PANEL manipulates.
Working with the registry from another computer simplifies the registry; an advantage.
But I need to be sure which option will offer better SQ. If it is the same with or without the file, I will delete it.
I suspect it will make no difference in SQ by itself.
There is the possibility of a cumulative effect from a less cluttered installation, though.
Seems to be needed by AMD processors from what I read.
Another brilliant mistake!
.
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2 Ru, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Liv
The driver was "intelppm.sys" removed during Steppe 11 Batch. I searched the Registry on another computer and found some entries relating to this file. The CMP machine is not with me now, will look at it tomorrow and take screenshots.
I recall some other devices with exclamation mark:
Firmware Hub
Microcode Update Device
System Management BIOS Driver
Numeric Data Processor
Programmable interrupt controller
Our goal is to delete unnecessary files that are not going to be used by CPLAY/CMP, or where their removal will improve SQ. The hub and update device etc are already disable anyway, so they will not cause any problem. But the Processor issue demands thorough investigation. I will skip "intelppm.sys" during installation tomorrow and compare the differences.
I recall some other devices with exclamation mark: Firmware Hub; Microcode Update Device; System Management BIOS Driver; Numeric Data Processor; Programmable interrupt controller
I've had all these devices disabled in DevManager on my main cMP^2 system for at least a couple of years. I've also disabled the DMA Manager and other devices with (I like to think . . . ) sonic benefits. My hunch is that the likes of DMA are handled by the chipset drivers and that the MS devices are inessential. Whatever, the box seems to work fine without them. Have I removed all the relevant drivers?
Good question.
HTH
Dave
System Management BIOS Driver,Numeric Data Processor,Programmable interrupt controller are already disable too. See "Step 1 – disable unnecessary system devices
" here http://www.cicsmemoryplayer.com/index.php?n=CMP.07Optimisations.
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2 Ru, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Liv
I am going to see what happens when I remove processr.sys tonight.
Should have known you were studying the registry!!!
.
Most of us just believed that if cmp/cplay works ...all was well. Don't think this was ever discussed.
Especially interest cMP2 with network.
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2 Ru, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Liv
Don't know where to put this so I'll put it here
..I tried maxmem at differing levels besides maxmem=256 (I have a 256 physical memory stick in use). At maxmem =256 it was ok maybe a little hard so I tried 217 per Steppe. Sounded like I softened everything up a bit in a negative sense. Loose bass, smaller soundspace, and dynamics suffered a bit. So I tried maxmem=227 and everything tightened up a bit. Thats where I am at now and I like it there.
..I tried hpet enabled in bios. At first I liked it but admittedly brighter but more dynamic. Then I tired of it...disabled it which is where I am now.
..Couldn't try the last registry tweaks because I am having trouble with the load hive technique. Here is what I do: I load the cmp ssd into a docking station hooked up to my home desktop. I open regedit, goto Hklm key (all trees closed), goto file click load hive, point to my config file in ssd, click on software, name it theo, click ok. Now I click on hklm showing in registry to edit it. Do my editting then click on theo then unload hive.Is that correct? I once wandered to another key or closed out before unloading and this tequnique no longer brings me predictable results. Sometimes it edits as expected sometimes it edits my desktop sometimes both (desktop and cmp ssd). So how do others do it? Also jolida mentions he cant edit currentcontrolset this way but I never had this issue...just sayin'.
..I am using the g31-e2sl mobo. I usually run memory timings in bios at 3 3 3 5. I ran once at 3 2 2 4 (just recently with all the steppe memory tweaks except the nonpoolpage files)and everything sounded way too slow, fat bass, not as clear a soundspace, less dynamic so I went back to 3 3 3 5 which I like. There is no question this stuff affects the sound but in a detuning kind of way. Almost like detuning the powersupply or putting a fat tube somewhere in the chain.
..I tried again to get the mihaylov technique to work for running regedit on the cmp but to no avail. I even loaded a prior level of operating system and ran mark russenovich process explorer to see what files I really need and it was the same as jolida pointed out several days ago. Is anybody besides mihayalov using this technique? If so I could use your assistance on making this work.
it does not really exist like set 1 or set 2.
So, one really only needs to modify the numbered sets because when you reboot whichever set the machine decides to use will have been modified.
You should not have had a CurrentControlSet within theo. That is very strange and can only be explained by something having gone wrong. Who knows how or why?
I have used the LOAD HIVES method tens of times with no problems other than the one's I caused for myself.
You were not clear - you are only modifying items within the theo heading, right? I feel sure you are but it is a little murky in the wording.
I frankly don't what I did. So are you saying to click on 'theo' (my named hive) and stay within that or click on hklm and stay within that?
otherwise you are changing the registry of the host computer.
Just as you would open any other entry - click the + next to theo and keep track of that vertical line extending below the +!
thanks ....didn't do that.
on your main machine.
If you had gone too far, like I did, you would have had to reinstall WINDOWS.
Yes I agree.
Hi All,
I like Steppe’s idea too delete files so that XP cannot use them any longer too run unnecessary processes. What I don’t like is: we don’t know exactly which files too delete.
So we must randomly delete files too see what works and what not, untill we reach a point where our cMP setup no longer work.
I don’t like this try-and-error approach very much. I feel such approach is like: shooting in the dark, hoping one does hit something.
But thankx too Mihaylov files list, we now know which files Mihaylov’s headless and networked cMP is using too run properly without any problems.
And more pleasantly, I think Mihaylovs file list opens up a way too indentify the files we don’t want XP too run in our cMP setups.
I will name that list: ‘ Steppes shortlist of files’
Here’s the concept that might allow for indentifying these files. For this I used:
1. List of files which my standard, headless and networked cMP setup is using. I created this list with 'Process Expolorer'.
I use a headless and networked cMP setup and used ‘Process Explorer’ too see which files XP uses while playing.
2. The list of files which Mihaylov’s headless and networked cMP setup is using after ‘Steppe’ deletions, as posted by Mihaylov.
List 1 (hfavandepas file list) MINUS list 2 (Mihaylov’s file list) =
list of files hfavandepas cMP set is running, but which files are not strictly needed / which files we don't want XP to run in a cMP setup.
Here is the list of files which my standard cMP setup does run, but which are not run in Mihaylov’s cMP setup (‘Steppes short list of files to be deleted’):
advapi32.dll
AcGenral.dll
activeds.dll
actxprxy.dll
adsldpc.dll
apphelp.dll
arial.ttf
arialbi.ttf
atl.dll
ATMFD.DLL
audstub.sys
batmeter.dll
browselc.dll
browseui.dll
cabinet.dll
credui.dll
cryptui.dll
cscdll.dll
cscui.dll
davclnt.dll
dmio.sys
dmload.sys
drmk.sys
drprov.dll
dxgthk.sys
explorer.exe
Fips.SYS
Fs_Rec.SYS
index.dat
intelppm.sys
JulaWdm.sys
kbdclass.sys
kerberos.dll
ks.sys
linkinfo.dll
lucon.ttf
mlang.dll
mnmdd.SYS
mouclass.sys
msacm32.dll
MSCTF.dll
Msfs.SYS
msgina.dll
msimg32.dll
mssmbios.sys
mstlsapi.dll
msutb.dll
nddeapi.dll
netlogon.dll
Ntfs.sys
ntmarta.dll
Null.SYS
odbc32.dll
odbcint.dll
portcls.sys
powrprof.dll
PROCEXP141.SYS
profmap.dll
psapi.dll
rasadhlp.dll
rdpdr.sys
regapi.dll
rpcss.dll
rsaenh.dll
rtutils.dll
scecli.dll
schannel.dll
sfc_os.dll
sfc.dll
shdocvw.dll
shimeng.dll
srv.sys
sserife.fon
stobject.dll
sxs.dll Fusion 2.5
tahoma.ttf
tahomabd.ttf
termsrv.dll
themeui.dll
update.sys
urlmon.dll
uxtheme.dll
vgasys.fon
w32time.dll
wdigest.dll
webcheck.dll
wininet.dll
winrnr.dll
winscard.dll
wlnotify.dll
wm_hooks.dll
wtsapi32.dll
xpsp2res.dll
In the attached PDF file you can see how I created this ‘Steppes shortlist of files too be deleted’
Download PDF here: http://www.mediafire.com/?74oo51p9ao18v0t
I did not implement this list yet.
I first want too hear any comments and thoughts on this approach too get this ‘Steppes short list of files’.
So any suggestions and thoughts are vey much welcomed
Mark
fully optimized cMP2 PC -> ESI Juli@ -> Van den Hul Optocoupler MkII-> Lavry Black DA10 -> XLR Mogami Gold -> Klein & Hummel O300
unless you find a need for them ...
Hi Rick,
What more is needed? And why?
Please explain.
Mark
fully optimized cMP2 PC -> ESI Juli@ -> Van den Hul Optocoupler MkII-> Lavry Black DA10 -> XLR Mogami Gold -> Klein & Hummel O300
I have posted a few lists.
I posted this on the 12th of November.
I am using a PS/2 mouse which lets you do away with the USB stuff. If you are still using an USB mouse leave in the stuff for that. But I would recommend going with PS/2 - You will most likely have to reinstall WINDOWS to do that but try and see if your machine will recognize it.
This is what I needed before doing away with acpi.sys which you probably still need.
No question after this your machine is a CD player. Won't do anything else. Also, assuming you're working completely within cMP mode. If you do these deletions outside of cMP you will not be able to get back in unless you keep reg.exe. Best to do final deletions with your drive connected to another computer. BUT make sure you exit WINDOWS while in cMP mode so you can get rid of reg.exe. cMP makes a registry change but once in there no need for this any longer.
After getting into this state all changes will HAVE to be made with your drive connected to another computer.
Are you following the FOLDERS deletion also?
My installation is a little over 35mB.
advapi32.dll system32
authz.dll system32
basesvr.dll system32
bootvid.dll system32
c_437.nls system32
c_1252.nls system32
comctl32.dll system32
comdlg32.dll system32
CONFIG.TMP system32
cryptdll.dll system32
csrsrv.dll system32
csrss.exe system32
ctype.nls system32
dnsapi.dll system32
FNTCACHE.DAT system32
framebuf.dll system32
gdi32.dll system32
hal.dll system32
juliaasio.dll system32
kdcom.dll system32
kernel32.dll system32
l_intl.nls system32
locale.nls system32
lsasrv.dll system32
lsass.exe system32
mpr.dll system32
msasn1.dll system32
msprivs.dll system32
msvcp60.dll system32
msvcrt.dll system32
ncobjapi.dll system32
netapi32.dll system32
ntdll.dll system32
ntdsapi.dll system32
ntoskrnl.exe system32
ole32.dll system32
oleaut32.dll system32
oledlg.dll system32
olepro32.dll system32
perfc009.dat system32
rpcrt4.dll system32
samlib.dll system32
samsrv.dll system32
scesrv.dll system32
secur32.dll system32
services.exe system32
setupapi.dll system32
shell32.dll system32
shlwapi.dll system32
shutdown.exe system32
smss.exe system32
sortkey.nls system32
sorttbls.nls system32
svchost.exe system32
sxs.dll system32
umpnpmgr.dll system32
unicode.nls system32
user32.dll system32
userenv.dll system32
vga256.dll system32
vga.dll system32
watchdog.sys system32
win32k.sys system32
winlogon.exe system32
winmm.dll system32
winspool.drv system32
winsrv.dll system32
winsta.dll system32
WLDAP32.DLL system32
ws2_32.dll system32
ws2help.dll system32
acpi.sys driver
atapi.sys driver
classpnp.sys driver
disk.sys driver
dxapi.sys driver
dxg.sys driver
dxgthk.sys driver
ftdisk.sys driver
hidclass.sys driver
hidparse.sys driver
hidusb.sys driver
i8042prt.sys driver
isapnp.sys driver
jula.sys driver
julawdm.sys driver
kbdhid.sys driver
ks.sys driver
ksecdd.sys driver
mouclass.sys driver
mountmgr.sys driver
npfs.sys driver
ntfs.sys driver
partmgr.sys driver
pci.sys driver
pciide.sys driver
pciidex.sys driver
vga.sys driver
videoprt.sys driver
volsnap.sys driver
wmlib.sys driver
Hi Rick,
Thankx for your post.
But I still don't see how your reply relates too my question and post.
* On a headless PC one doesn't use a (ps2- and or usb-) mouse (or monitor)
* as a consequence I don't (have too) use USB
* I don't want to wreck my cMP setup
That is the whole point of my post and suggested approach.
I only want too delete files that are actually going to be used by XP while playing mysic in a (headless and networked) cMP setup.
What good is in deleting files that are not even going to used by XP while playing music anyhow?
I don’t think you red my post.
I don’t want too take the route you are suggesting because it wil UNNESSECARELY wreck my cMP setup. I’m looking for an alternative approach.
I think there is one.
Mark
fully optimized cMP2 PC -> ESI Juli@ -> Van den Hul Optocoupler MkII-> Lavry Black DA10 -> XLR Mogami Gold -> Klein & Hummel O300
Sorry for that!
Just missed the point entirely!
is in our settings of the default user/users during Nlite iso preparation. After the superminimal install Windows/system32/GroupPolicy folder shows EMPTY SUBFOLDERS, thus Group Policy Object Editors sees NO ADMINISTRATOR settings whatsoever. I have prepared two disks with different settings myself from VL version. No Out of the box experience wit vl. No registration. Works and loads perfectly. Windows = 204mb, registry folder - I mean config - is 5,6mb. But still NO F*g AWE!!!
Will try another way.
Serge.
Edits: 12/06/11
in your cPLAY diagnostics?
That is hard to believe!
I know it does not install "correctly" but I have made several installs and have been able to get AWE even though it will not accept ADMINSTRATORS.
You are checking within cPLAY?
No AWEEE!
That is why I started sneaking like a Gollum around and inside a normally installed Windows and how I found this groupobject empty folderrrrrrr!
Serge
Serge
For what is worth, I had the same problem with Junaid's nLite. It would not accept Administrators, and AWE would fail.
I got it to work by picking another ID that was listed. Instead of Administrators I used both Anonymous User & Everyone. It took it, and then AWE was enabled.
Hope this helps.
a wrong iso and do it wrong myself. When I click on advanced I DON'T FIND THERE ANY NAME.
Serge
I know with my "minimal install" I cannot get it to accept ADMINISTRATORS but when I go to cPLAY diagnostics it shows up.
Now I am assuming you are attempting to implement AWE early on.
My install process (done many times and I always get AWE)
as soon as I have a working WINDOWS I:
install chipset drivers
implement AWE
implement MINLOGON
install cMP and cPLAY
start machine in cMP mode and stay there.
do the basic cMP optimizations, which are almost minimal, too.
run Junaids registry batch files all consecutively
uncheck as much as I dare from AUTORUNS while deleting .dll's and .sys's the entries correspond to. This requires some back and forth which I do just to be careful
After tuis the rest of deletions and reg changes are made from the "other" machine.
I delete all but the files I keep in one big deletion.
I have found that the files batch files have IN MY SYSTEM created problems. I am making all deletions while connected to the "other" machine.
I wonder if your problem is WHEN you are attempting to implement AWE? Are you doing it early on?
after installation of windows-lynx-cmp-cplay. After that - restart with a new hope and then start cplay to see it failed.
Serge.
P.S. Read my post below about nonpagedpool size. Can You try it? I am typing from that machine. Just try, not more than 15 minutes then revert.
Serge.
This is very important, I mean feedback.
Hi, i couldn't install awe either, but found a solution (for my machine):
Follow the cics-directions until step 9. Then:
- when you must enter "Object Types" button, don't do that, but do 'advanced' button.
- Find Now button
- In the field below, under Name(RDN) select Everyone
- OK button
- Object Types button
- check Groups > uncheck Built-in security principals & Users > OK button
- again OK button
- apply button
- again OK button
- File > Save
- File > Exit
- Reboot
This worked for me, hope it works for you
Douwe
Will go, pick more blank disks, try anothe nLite iso, reinstall and try. Will report. I want to find a routine that will be at least as easy as what Cics told us to do. It worked fine then, so it shouldn't be more complicated now. The easier it will be, the more people will join us.
Serge
I know you probably have but I thought I would ask anyway.
Now to look at the pagefile post ...
with nLite I did have a few disks that I could not implement AWE which was one of my first frustrations with using it.
NOW, I must say I did not find it unlistenable by any means. I did wonder if when one minimized WINDOWS did this diminish the need for AWE?
I do know that the formula I am using and the one in my post will most assuredly allow AWE. Junaid's formula does allow AWE. Did you make any changes from either of these two versions?
The "version" of nLite that would not allow AWE had more stuff in it than the one I posted about.
I just had another thought: maybe I was getting AWE after all but when clicking OK after typing in ADMINISTRATORS and this was not accpeted maybe I gave up. I cannot remember whether I checked the cPLAY diagnostics since my assumption was IT DID NOT WORK. We have all learned much with this experience. Never be too sure ...
Excuse this stream of WINDOWS minimizing consciousness ...
with all of the images being made and then written to our SSD's I started thinking that this probably created some confusion based upon the stories we hear about how SSDs are different than HDDs.
SO, I thought I would format my SSD partition before replacing the current image. Doesn't take very long - so it is easy to do.
I think I am hearing an additional clarity - maybe I have talked myself into this but I think it sounds much better.
I would welcome slings and arrows of derision for outrageous recommendations if you find there is nothing fortunate going on here.
It is probably correct to say that formatting SSDs would shorten their life. Intel recommends formatting SSD in a couple of scenarios, including before reinstalling an operating system.
http://www.intel.com/support/ssdc/hpssd/sb/CS-032319.htm
But Rick, you are not talking yourself into this. Fragment of data are hidden in slack spaces of unused clusters in unformatted disks. These fragments are read but not identified to be useful data by computers. You can check these fragments by hex viewer or diskexplorer and they are dirty. That's why, without these rubbish in your CMP, you were hearing an additional clarity . But I gather that these are very low level noises and that additional clarity could only be heard in a very good audio system, i.e. a fully Tweaked CMP machine with a low noise power supply. It seems that you have reached thus far though.
One way to implement this is to install CMP using a regular HDD. After all tweaks are done, ghost the image and clone to a formatted SSD as the last move. I've been installing using a HDD and am going to try this first on a CF card disk. Thank you for bringing this out.
That is how I will do my next install and tweaking which involves much back and forth with the images.
Maybe NEWEGG will have a sale on a small capacity SSD and I can copy that final image to the new SSD.
That is such an obvious and GOOD idea!!! But who is surprised by good ideas from you? I have come to expect them.
Got my desoldering stuff yesterday and tonight I start practicing removing video chips from unusable MB's. Can't wait to hear what this will do in combo with minimized WINDOWS together. At the same time I hope to finally install a P24 linear supply I have tinkering with for months (years?).
Thanks!
I bought a 1600W adjustable hot air gun and can now remove chips with this gun and a sharp plier. I just heat up the chip directly for about 15-20 sec, pick up the chip when the tin solder melt. It's hot, need to wear clovers.
Be careful with removal! Remove the audio chip first and then network.... Good luck.
This one is cheap and looks like it has a good nozzle size.
Can't beat the price!
1500W/1000F
http://www.amazon.com/Buffalo-Tools-PS07343-1500-Watt/dp/B004GG0SS8/ref=sr_1_60?s=hi&ie=UTF8&qid=1323332162&sr=1-60
1500W/920F
http://www.amazon.com/American-Tool-Electric-Heat-Accessories/dp/B003ZEJ6S0/ref=sr_1_39?s=hi&ie=UTF8&qid=1323332105&sr=1-39
920F
http://www.amazon.com/All-Power-America-APT2005-Piece/dp/B0018E22US/ref=sr_1_14?s=hi&ie=UTF8&qid=1323331644&sr=1-14
1500W/920F
http://www.amazon.com/Heat-Gun-1500W-Nozzles-700F/dp/B002J9N7M8/ref=sr_1_19?s=hi&ie=UTF8&qid=1323331644&sr=1-19
1300W/1100F
http://www.amazon.com/Black-Decker-C800621-Heat-Plus/dp/B000FV8SC6/ref=pd_sim_sbs_hi_30
1200W/920F
http://www.amazon.com/1200-WATT-Electric-Heat-Gun/dp/B000NS5W48/ref=pd_sim_sbs_hi_17
I prefer small nozzle ones, but I couldn't find them. In the list above I would pick 1000F or 1100F product to ensure rapid heating.
The MAKITA heat gun has an accessory nozzle that is 9.5mm in diameter so I got one of those thinking I need all of the help I can get from the heat gun to not ruin the MB.
Not sure if the nozzle could fit on any heat gun.
There were a few other companies offering nozzles.
THANKS again for the help.
Any recommendations (again) on this hot air gun?
Do you think that any adjustable 1600 watt gun is as good as another?
Thanks,
Mine is, again, a cheap Chinese product @USD30. It has a nozzle smaller than regular ones, not too small though. I think there are 2 nozzle-size and I couldn't find a smaller one. Adjustment is made by turning a cap at the end of the gun. At maximum power, the heater power increases and air flow is slightly reduced. The product claimed to produce 1100 degrees C hot air.
This one is very hot, much hotter than the previous one I used. After experimenting with this gun, I gathered that very hot air produced in a streamlined nozzle will heat up a chip very rapidly. Heating up at close proximity, tin solder securing a chip will melt quickly and I could see it. I just picked up a couple of chips by a forceps. Much easily this way, and very neat. Less works as the caps and R could remain intact.
I must warn that it is very very hot. Wear 2 cotton gloves in the hand which holds the forceps.
The gun is not with me right now. I'll take some photos tomorrow. As to recommendations, I picked up the one because of its 1600W power which is adjustable, 1100C and with smaller nozzle. It turns out to be the right choice. There were a couple of others with similar characteristics and I picked this one for its price and supply of replacement heater coil. Pick one with similar characteristics and you will not go wrong.
Hope to be posting results soon.
Hello! Rick, I think that ssd disks shouldn't be formatted ever. It severely shortens their life. As for the sound, yes it should sound differently, because of the following:
Though we set windows' paging file to 0, it's not a 0 after all. Some volume is created anyway and it partially containd The F* registry and partially what's in the memory - music in our case and chunks and threads' of other processes' remnants, and so on. Now, data from ssds is read faster and requires less processor-memory work, thus different (probably better) sound.
Serge.
Rick! You and some other inmates managed to set the size of the registry to a minimum, but what about the sound. Does it bring the desired change for the better?
Edits: 12/05/11
did not realize formatting would have the same bad effect.
So, when you do a fresh install on a previously used SSD you do not let the machine format the disk? I would think that would result in an extra copy of WINDOWS on the disk and would make it crowded.
Looking around I see this is not absolutely settled. As few writes as our minimized OS initiates I would tend to think we do not have much to worry about.
When I think of all of the installs I have done on my two SSD's and with the re-formatting I allowed WiNDOWS to do ...
I would think we would have to do this MANY times for it to have a bad effect. If the drive does go bad we can hope they become cheaper in the future if replacement becomes neccessary!
I am of the same opinion as Mihaylov, that minimizing WINDOWS has to be a good idea and even if one cannot hear dramatic changes with every change there is a cumulative effect.
If I did not think this was going in a good direction I would not post. If minimizing WINDOWS did not improve the sound we would have dropped this tactic weeks ago.
I can say I am very pleased with the sound I am hearing. Micro-dynamics that I did not know existed. Vibrato clearly delineated. My brain is having to do much less work when listening to music! The low and mid bass seem to have been released from captivity. Not a subtle change there. I must say that much of this has come from your latest registry and boot.ini tweaks but I feel they have been unhindered by the lack of other useless stuff hanging around.
I have done most of Mihaylov's registry slimming and though I cannot point to what they have done (we could chase our tails well into our seventh lives doing that, just trying all of the combinations, like an artist deciding to finish a painting, there is always something else that could be done but you have to tell yourself ENOUGH - that is how I feel about obsessing about various combinations of slimming techniques).
I am more than pleased with what I am hearing.
I would ask (since I do not know) how could superfluous registry entries improve the sound? Just like .dll's that have nothing to do with what we are asking the OS to do in this very specialized application, how could a file for telephony make for better sound? Just one example.
I have found that as I take more stuff away from the registry it does not show as getting any smaller. Last I looked it is sticking at 4.25mB. Was it you or Junaid who pointed this out? I have found it to be true.
1. Rick. Nothing wrong will happen after 5-6 formattings. What gives the sonic result after installing on a newly formatted disk is the paging file being initially defragmented (even though we set it to no paging file, it is there and is at least half of your registry size.)
2. About your Registry Size Limit RSL it is here:
http://www.audioasylum.com/forums/pcaudio/messages/9/99455.html
3. You say: I am of the same opinion as Mihaylov, that minimizing WINDOWS has to be a good idea and even if one cannot hear dramatic changes with every change there is a cumulative effect
Truly, when I posted this, (even the word cumulative is from my post) at the tweaks part 6-7 time, I was ridiculed...
Serge.
on to something. Some will be jealous they did not think of it. I know I was not jealous but wondered why I had not ventured into this. You look at these files in system32 and the balloon (on a non-cMP machine) pops up and tells you what it does. In your head you know that is meaningless for the cMP application but it is easier to think, better leave well enough alone.
Thanks to you we now are looking for the new residence of "well enough" - have not found it yet though we are getting closer.
I do think I know of one who did this and seems to have come round rather strongly in favor of your initiative.
Maybe it was your playful language; your speaking of starting this journey after feeling the effects of alcohol maybe struck some as being unserious. I can assure you without the benefits of alcohol I would have hesitated doing many things that turned out well!!!
I do not think ANYONE is ridiculing you now. And if they are, luckily, they do not speak out on the forum because if they did they would meet massive resistance.
When you consider install s I have formatted the two SSDs tens of times, each. I cannot see or hear any problems. Of course, I would have to try a new one in comparison to really know.
Thanks for the link.
Rick! There was no bitterness in what I posted. It is just the funny side of it. And besides, it's not really me, it's our COMMON effort, no less. It all started with Cics, after all.
Anyway. If you have time and leasure and are willing to risk (not very much) open your registry here:
HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Control\Session Manager\Memory Management
in the window to the right find NonPagedPoolSize. If it doesn't exist, just create it.
Doubleclick and set the value to 4194304 (that's in bytes. See the pick. I have just done it on my working machine I am posting from. (15-20 minutes ago) Everything seems suddenly so agile... It has to do with AWE)
Serge.
Aside from the confusion over hex vs. deci, you gave the number "4194304 (that's in bytes" which equal 4.19MB but then said you were aiming for 40MB. Likewise the larger number in parenthesis in the registry key is 68MB not 682MB, so I am at the least unsure how to proceed whichever number system. Then, of course, there is the question of what this is for or about given that the existing value for me too was 0. Are you still recommending this be tried, and what would I who has 3GB ram total put in my registry entry?
Gentlemen... please be careful to specify numbers like this as either HEX or DECIMAL values, lest we be led to wrong conclusions.
Serge - the number you quote in the body of your post ("4194304") is in hexadecimal format. It's DECIMAL (ie: base-10) equivalent is displayed in your pic to the right (in brackets, "68764420"). That is the number of bytes Windows is using for this parameter.
I mention this because Rick later used a number "halved", and it looks like he has done base-10 division on a base-16 number. The result might "work", but it's certainly not what he presumes it to be.
Just a little "gotcha" we need to avoid in these deep internal tweakings.
Hope this helps...
Grant
That's not a Toy... IT'S A TOOL !!
it more deeply. I wanted to set the nonpaged pool to 40 mb. According to You, I have set it to 687mb now ? Ok, I see. I have on my working machine that is currently working much faster and I am posting right now from it I have 2gb of memory. So, for 1 gb it should be appr 480 and for 512 - 256? No math here in my post, only a common sense, though it doesn't always work with ms programs. What can be recommended or safe numbers to set for an experiment in our cmp for 1 gb and 512 mb, do You think? What is Your opinion?
Serge.
Edits: 12/06/11 12/06/11
Hi Serge
Until I understand a little about how the NonPagedPoolQuota & Size parameters work, I have no recommendations for using them. Checking my main system here, both are set to ZERO (I don't remember setting AWE on it).
I'm not sure exactly what you're trying to achieve with these settings... how it will improve playback.
You know the math: 1024 x ?? = your mb value. My point was to be clear on exactly what we are telling Windows to do via these numbers.
Cheers,
Grant
That's not a Toy... IT'S A TOOL !!
Hi Serge
After much clicking, I came across this statement in an article by a Microsoft Critical Problem Resolution Platforms Team engineer (...with a title like that, who's to argue !):
*NonPaged pool size is not configurable other than the /3GB boot.ini switch which lowers NonPaged Pool’s maximum. 128MB with the /3GB switch, 256MB without
The very in-depth article can be found here -> http://blogs.msdn.com/b/ntdebugging/archive/2006/12/18/understanding-pool-consumption-and-event-id_3a00_--2020-or-2019.aspx
My head still hurts from trying to grasp it all... but these parameters do not get you closer to the Madman's Dream (in my opinion). But I could be wrong - it happened once before ;)
Cheers,
Grant
That's not a Toy... IT'S A TOOL !!
wondering what is the difference between the system using RAM or using the SSD, is there that much difference in speed between them?
Reading the attached article with my breathtakingly limited understanding of how WINDOWS works leads me to think this would not have anything to do with our music files in memory - I get the feeling it is concerned with OS stuff in memory.
I guess if we could find a way to limit how much of that (OS) is kept in memory would that necessarily result in more room for the music files? I am getting the feeling that is not the case.
I know the folly of coming to a conclusion after reading one article but I had spent some time chasing my tail at the MICROSOFT site before. One gets dizzy ...
I used halves since that is all I could think to do!
It seems even after dividing Serge's number by four the tracks have more segments than they did at zero.
It did not make a difference in segments (that is an easy to see difference)when I went from half to a quarter.
I was thinking that since the "full" number would not allow the track to be loaded at all that it might be working in an opposite way from what had been assumed. Then again, I was assuming what the assumption was.
I listened for a couple of hours with the quarter number and it sounded good. I cannot honestly say it sounded any better. I think my first impression was because I had the volume turned up too high!
No question we need to find out just what this does.
I thought Serge was thinking it was a way to either implement AWE or make the AWE "lock" larger. But the fact that my machine which does have AWE working also says "0" leads one to believe that this is not part of AWE,
UNLESS
I have noticed that when connecting my music computer SSD to my other computer than when I load the hive I will have entries on the music SSD's registry that have nothing to do with the "music" registry - it will show the drives in my other machine and things like that so there is not a complete divorce between the two registries.
When I change the registry on the music SSD it will not affect the other machine, so that is not a concern (unless, as I have two times now I lose track of where I am and move from the loaded hive to the "other" machines registry (very stupid)). But this could be one of those situations where the registry is reflecting the host computer's registry and not the registry's hive you have loaded. Excuse the fact that I do not know how to speak the computer vernacular. I hope this makes some sense.
This may be an instance where I am not able to truly see the setting my music computer is using when active.
Rick! I have (it seems) found the way to make awe work, so will risk these settings on cmp machine myself. The working machine works Much Faster! The fact that more segmenting occurs shows, that,
1. Cplay works as described
2. awe, even after all deletions, works
3. There is smth fishy here, that needs to be uncovered (about nonpaged pool size), because it affects the sq.Generally, there is a madman's dream in my head to achieve the following:
1. Make as much (if not all) clocks in the system work in unison as possible
2. Make awe create large pages' size so that locked pages would be few and the cpu would have to make shorter excursions to them and back and that bigger chunks of the song would fit in.
3. Exclude everything possible from the process above.
4. Create a program or a process or just a command line to constantly assign affinity to our cmp-cplay process to a DIFFERENT processor, that's why acpi multicore/multiprocessor (though it affects current stability) installation is needed (remember cics saying that even one core would enable multiple threads? Yeah, but he didn't say of affinity, and many of us tried it manually.)
AND WHEN CICS IS BACK, we'll have to try to persuade him to get back to all the cue loaded into memory, but not just into memory, but in a NonPaged Pool.
Serge.
You must understand, that I am fogwalking as I am not a computer man. I read read read read, try to make my assumptions on the level of my poor ken and try, try, try. All I ask for is more and more feedback. Your feedback is like a thread of Ariadna that must get us out of the labyrinth.
Serge.
Edits: 12/07/11
I have a similar problem with registry tweaking via a host computer. It worked fine for awhile then because of some mishap I made (exitting regedit w/o unloading hive 1st or doing what you did) I now only change my host registry and not the ssd registry with this method. This is why
I keep trying to make the Mihaylov routine work ... but unsucessfully so far.
Edits: 12/07/11
I was pointing out how one has to think about what they are seeing. Sometimes it is (on some entries) showing you what the host machine has installed. Some places it would show both the music computer AND my host computer devices connected.
When I have had a problem it was caused by my paying too little attention and deleting away like a madman!!!!
Have had to reinstall WINDOWS twice. Now to learn to make an image of the other machines OS!!!!
You can reload windows from a prior date from the windows start button on bottom left of xp screen, all programs-> system -> restore or something like that.
than I had before. (2097152)
With your number cPLAY would not load any files. Couldn't "find" my files.
Before each approx. 3 minutes was a segment. Now it is like 2 minutes.
I have halved the number again to see what happens.
So far, it certainly is not doing anything obnoxious.
Will see what 1088576 does and will let you know.
halved the number. And it works.
Haven't had a chance to listen just thought I would tell you.
How big do you think I can make the number with my half-size memory?
I will report on what I think of the sound when I can segue to EXTREME BIOS and listen for awhile.
I can tell you (who knows how influenced I am by your comments) but the music did seem to leap out of the speakers.
More to follow if you are still awake!
so I am not sure this is AWE.
Meant to say that in the first note.
writing at work so I will look as soon as I get there.
Well, after I pour some whiskey in a glass THEN I will take a look and report.
Since Junaid and I are debating the merits of including this in our installations I have been looking around.
This one has something to say about a similar application to ours:
The Advanced Configuration and Power Interface (ACPI) has caused many frustrations for musicians over the last few years, and since there's a switch in the Windows XP setup procedure that lets you install XP without using ACPI at all, this is the first topic on my agenda. It is also possible to remove ACPI once Windows XP is already installed, as I'll explain, but I can't guarantee that this will be as clean.
ACPI is a power-management specification developed by Intel, Microsoft and Toshiba, and support for it is built into every Microsoft operating system since Windows 98. It's designed to let the operating system control the amount of power provided to each hardware device or peripheral attached to the computer, and makes it possible for Windows to turn off your monitor screen, CD-ROM drive, and so on. As you might expect, It's particularly important in laptops, where extending battery life is a priority, and makes it possible for the entire PC to be turned on or off by external devices, so that you can for instance wake up your PC at the touch of a mouse or key: this technology is known as OnNow.
And more concerning IRQ assignments in the linked article.
From what I am reading I cannot figure how this would improve the performance for our purpose.
Hi, tonight i tried two fresh minimal nLite-installs, one acpi, the other non-acpi. Only lightly tweaked (awe and maxmen=256). The non-acpi sounds better, more detail more musical.
For non-acpi i selected 'MPS Multiprocessor' as i run dualcore.
You can select the in nLite, or by pushing F5 during start of installing.
Douwe
When going for for total minimal installation I did not include the MULTIPROCESSOR option in nLite.
I had no idea this had anything to do with acpi other than I noticed I no longer had that driver in the folder.
I can be accused of trying justify this after the fact! I readily admit to guilt!
I have been extremely pleased with the sound I am hearing. From what I read about acpi I do not see from the conceptual point of view (of someone who has no idea how all of these pieces fit together) how this could benefit the qualities we are searching for with out cMP machines.
Junaid has come to a completely different conclusion.
One wonders if different setups could form different synergies (I hate that overused word but cannot think of another to use) with different systems. Junaid found the non-acpi to be bland. I find it to very musical. It could easily be explained by electronics and loudspeakers used. I am using horns with the notoriously tricky to tame TAD drivers (4001 and 2001's above 500 Hz) which could explain my being very pleased with the non-acpi install. This blandness which Junaid hears might be what my system needs.
I have yet to try a minimal setup with MULTIPROCESSOR for lack of time and for the time being my happiness with the current setup. So even though I did plenty of nLite installs with MULTIPROCESSOR before arriving at my current setup I did not try one with only the addition of MULTIPROCESSOE so I am not trying to say I have made a proper comparison.
From what I have read acpi is something intended for laptops for its ability to offer the different power settings. The question remains what is the default setting for a non-acpi machine? I assume HOME/OFFICE or is it MINIMAL POWER MANAGEMENT. Either one would be fine with me since we are doing the slowing down with BIOS settings now and one has to wonder how intrusive to sound quality these settings could be.
I would not want the OS to play around with power in a music computer. I am sure this is advantageous for more typical computer applications but just as with an amplifier you want as steady a flow of power as possible as in a classic class A amplifier. I wonder if these acpi power settings are akin to a positive feedback scheme which never work well in high quality audio.
Time will tell ...
After comparing one-core and dual-core (on my h55m-usb3) in different setups n tweaks, i now have decided to run on two cores. In my setup this gives slightly more stage and music.So when going non-acpi i have to select MPS-multiprocessor (when i select 'standard PC' or MPS-uniprocessor' i get bsod because i have two running cores. When you run one core, my guess is you would'nt select MPS-multiprocessor.
Edit: when one core disabled in a dualcore you should select MPS-uniprocessor.
http://support.microsoft.com/kb/299340
In my setup non acpi is better: more musical.
Edits: 12/08/11
I ask for i486 with C-Step when installing nLite without MULTI PROCESSOR and with BIOS set to one core enabled before beginning the install.
I use the basic cMP recommended BIOS at install with the exception that I have CDROM as my first boot device.
if you run on one core enabled, you should select MPS uniprocessor.
The C-step processor is a very specific and/or old processor.
http://www.northernsounds.com/forum/archive/index.php/t-3917.html
http://support.microsoft.com/kb/299340
Douwe
I guess I had better give uniprocessor a try.
This time using a regular HDD until it is right and then to copy the image to the SSD per Jack Wong.
that both acpi and non-acpi can be installed from the same installation disk, when you press f5 ?
Also, as you write that You implemented awe, it means that your install is ok, did you try
acpi with awe + 256
acpi wo awe +256
non acpi with awe + 256
non acpi no awe + 256?
One more questions: did You do any tweaks after install? Does your mobo bios allow to control vdrop directly?
Serge
yes, both acpi and non-acpi are options during installation. The thing is, the screen where one can choose these options is hidden. Just repeat pressing the F5 button during the first 20-30 seconds of the booting install-disc. Then you et the acpi-selection-screen.
More info: http://support.microsoft.com/kb/299340
I didn't try all the awe+256 variations you mention.
I first tried acpi +awe +256 in dualcore and singlecore intalls. Dualcore was better, so i decided to keep that install.
Then, in dualboot, i installed non-acpi (MPS Multiprocessor) +dualcore +awe +256 and found out this was more musical than acpi.
rickmcinnis also preferres non-acpi, on Jolida's system acpi is better, so i guess it's very mobo-dependent.
I didn't do any other tweaks (exept for display-resolution). Yet.
I do'nt know what is vdrop mobo is ga-h55m-usb3.
Douwe
http://www.audioasylum.com/forums/pcaudio/messages/9/98962.html
It's there. with a pic.
Don't do the /clklvl tweak, described there unless your mobo and cpu are NOT underclocked and undervolted. If they are, you will most probably get a crash like me and Theo after a short period of a very good encrease in sq.
Serge.
P.S. AWE problem settled. Can send via email an ini for nLite iso.
Have done all cmp with all my 22 tweaks exept part 21 - tomorrow. It's deep night here.
I also like Jolida's variant, we have the same mobo.
the feature, that I proposed to do in bios - Enable the vdrop control directly. Enabling it won't not allow the cpu voltage to fluctuate with energy saving routines. (But it is counterIntel specs) So, I am against acpi
Serge.
Sorry to bother you! I just started implementing the Tweaks about 2 weeks ago and am missing a lot of details. Here's my questions:
1. What's Bold Fortune Registry changes? Can't find it.
2. What's the recommended sequence for implementing the Tweaks?
e.g.
Steppe 2
Steppe 3
Steppe 4
Jolida 1
Jolida 2
Steppe 5
Steppe 6
Steppe 7
Steppe 8
Steppe 9
Steppe 10
Steppe 11
Steppe 12
Steppe 13
Steppe 14
Steppe 15
Steppe 16
Bold Fortune Registry changes only
Mihaylov's first tweak posted in September
Mihaylov's My Tweak
Mihaylov's Tweak 2
Screenshot 1
Screenshot 2
3. Where should Steppe 17 - 22 step-in on the above list?
4. Should Xplite be implemented still during Steppe 6?
4. Where to find Screenshot 1(Part 22?) & 2 (Mihaylov's screenshot?)? Or are there more update ones?
Thanks for your help!
The steppes can be run in any order. No hard n fast rules apply there. Xplite was applied between steppe 5 & 6. But I guess u are following the nLite route now, so xplite will be of least importance, as 95% of the stuff is already avoided with the nLite install. If u still want to apply xplite after installing the nLite version, u will have only two video codecs that will need removal from it.
Bold fortune is a different approach. It has it's own batches. If u want, I can send them to u. But I suggest u save it for the last IF needed. Concentrate more on Registry mods, as I now find they are very very effective.
Steppe 17 was a Bios tweak alone. Steppe 18 to 22 were substituted with Screenshots. Both are mostly the same. That section was concentrated on what needs to be LEFT IN. And I did not find it appropriate to make Batch files for them because they anyway had to be removed by connecting the SSD to another computer as a slave, so the drive letter changes depending upon the one which the Host computer assigns. So it was Implemented manually by concentrating on what Remains, more than whats to be removed.
While doing screenshots, u need to be very cautious & slow, as different Xp installations will require different files to work. For eg. If u follow Mihaylov's screenshots, u will remove some fonts which he did, & Ur system will tentatively not boot, as he does not use a monitor. And the font which he needs (866.fon) is not needed by the rest of us. Likewise there are few other files that need to be left in depending upon the MoBo drivers, soundcard used etc etc.
The first time u follow the route, u are most likely to have a BSOD. Until u are too sure of what u are doing, because it would be the final list of files to be removed.
It's highly suggested that u take a back-up after all the batches, & before starting on Screenshots. Mihaylov & Me had posted our screenshots a few weeks back. U will find it in the early November session..
Junaid
Do send me the Bold fortune batches, as may need them finally .
Yes. I was at Part 22 when I deleted some .nls and Windows needed some some files back.
Sure. Will send them to u. For a quick way out, let us know which MoBo & soundcard u r using. I'm sure many would be using the similar combo, so they could list files needed in their set-up..(Screenshot)
Junaid
I recently switched to H67MA-UDH-B3 with i3-2120T, I guess not many of us are using this one. Sound card is Musiland 01USD, I already know what files it needs.
I also have a H55M-UD2H and H61M-S2H for use with Julia@. Thanks!
Language pack used for Xp??
Junaid
XP Pro SP2 English (US) with Chinese Traditional and Simplified languages.
I really don't use a monitor but it does not mean that it doesn't work when I connect it to the cMP2 I. It works!
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2 Ru, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Liv
I'm sure it does :)
I was explaining the difference in the fonts which are different for each system. By the way, did u do nLite?? What's the scene at Ur side?
Junaid
Now I do not see any necessity in nLite for me. If anew to reinstall system that perhaps there is a sense to use nLite. Though it is better to do all manually.
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2 Ru, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Liv
Can you remove the files associated with the graphic? I think SQ will have major benefit if you can disable it.
Edits: 12/04/11
I left only those files without which the system is not boot.
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2 Ru, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Liv
How many have installed a multilanguage version of nLite?
Junaid
Edits: 12/03/11 12/06/11
Junaid can you please explain how you use this? Do you start with a windows os disc to load along with this or is it something else?
Just burn the ISO to a cd. Then install Xp just by using this disc alone. U do not need any other disc. It will do a minimalistic Xp installation just required for our purpose.
I just LOVE the sonics I am getting with this set-up...
Junaid
Edits: 12/04/11
.
Hello, everybody. This is what the mermaid finally told me: keep my memory tight, keep my timings tight, starve me, be strict with me and put this line in your boot.ini:
/nopae /noexecute=alwaysoff /fastdetect /UsePMTimer /timeres=9766 /nodebug /pcilock /onecpu /MAXMEM=256 (if you use /noguiboot, place it at the end of the line). If You use 512 mb of memory, be more carefull with maxmem. The smallest achievable setting is 215-216 mb, provided that you have done all 22 of my tweaks and your memory timings in bios are set to the extreme shortest. Give some time to burn-in.
Serge.
N.B. To Enjoy the mermaid to the fullest, don't forget to ENABLE Precision Event Timer in BIOS!!!
Edits: 12/03/11
Now that I have my registry tweaking process working ok (thanks again Rick) I have settled on maxmem back at 256 and hpet enabled.
Guys,Thanks for the new tweak !
However, after editing the boot.ini, I got serious pop and click sound under cplay when upscan to 192khz, 96khz also the same (a bit better)
I downclock my CPU to 1.2GHz from 3GHz, now i try to give it more power, finally at 2Ghz, upscan to 192khz play without problem.
So this boot.ini tweak will let CPU work harder?
Thanks
Edits: 12/07/11
Hi Serge,
Van Morrison's 'Saint Dominic's Preview' always sound a bit thin and 2 dimensional. With your latest boot.ini tweak, it is thin no more, but now very liquid and 3 dimensional for the first time. Even a bit tube-like, in the best sense of the term.
A great tweak, thanks for a wonderful find.
KC
Serge after reading about riboges tests I still wonder what does maxmem=256 mean when you have a 256 ram? Is there any merit in trying something lower than 256 in my case? Also if maxmem allocates the most memory for cplay should not I get more time between swaps versus baseline (no maxmem setting in boot.ini). Baseline would give me 2 minutes 14 seconds between swaps now I get the same with maxmem=256. Puzzling.
Edits: 12/06/11
Theo, The absolute minimum is 214 mb With less cplay starts, will try to load the file but will finally say it has only so and so (e.g. 160), but needs so and so (it's 163mb, IIRC). By substracting and rounding to the highier value you will reach your absolute minimum. If you set up your bios with cmp002 settings it will ask for 215-216, and IF You are, like Rick, an absolute tweaker, then set your memory timings for extreme cmp003 settings but give 217mb in boot.ini. Read my tweaks 21 with care once again. Try, but do it slowly.
Serge.
P.S. It doesn't matter how much memory you have in your stick, it is how much of your stick you show to the mermaid, if you see what I mean.
One thing you should remember is that I use the G31-E2sL mobo which does not have many of the settings you reference in Tweak 21. So although I can set timings very low for memory and I can adjust cpu speed and voltage low thats about it.
Other settings are less important. The trick is in extremely short memory timings.
Serge.
LOL!! I'm afraid if I show any stick to the mermaid...I'll get arrested.
I'll try 217.
timing adjustments? Can some approximation of this work, like using Maxmen=512 or something a bit higher as an improvement over dual core and 3GB of ram? You probably can't say with any strong basis but does this seem worth trying or obviously unlikely to work?
but I am not certain how I tell if it is really being implemented in my system with none of the advanced bios adjustments. Is this boot.ini setting necessarily effective in any installation with whatever bios?
enough. It tells window to see only amount specified. There is another interesting switch. It does to windows memory almost the same, but not absolutely. It should be even better, but I haven't tried it yet. /burnmemory=mb You have to calculate how much memory to HIDE from windows, Just the opposite of the /maxmem. Try, if you wish and please report. Feedback, even negativeб is precious to us all. Example, if you have 512 mb stick put /burnmemory=256, if 1gb, put /burnmemory=755
Serge.
The minimal amount of memory is 215-216-217 mb. This row is such, because I hope, that You have tried my bios/memory settings from tweaks part 21corrected. The shorter are the memory timings, the more memory (217) you will need. The sound is EXTREMELY good with Extreme settings. Try, be careful and patient. (give real 24 hours burn-in time.)
Also try to put ALL THE BOOT.INI LINE, including /nopae and /onecpu
The difference is substantional. I,ve been playing my music with this bios line for a month.
Edits: 12/05/11 12/05/11
The BURNMEMORY switch works fine. The effect is not obviously different from maxmem but my system is sounding so big and full and balanced at this stage that it is hard to pick up further improvement...maybe with time. I do notice that my cpu usage has gone up from a typical 1-3% to more like 11-13%. As I said I am using a Lenovo laptop which does not provide for the bios/memory adjustments of your step 21 but I am using the full boot.ini line with nopae and onecpu.
I am glad you are so much better with computers than with arithmetic. :) 1000-256 does not come to 755.
You say cpu is working harder. Is it with /maxmemory or /burnmemory?
I reverted to maxmem as suggested. This time the cpu ran in the teens rather than low single digits I can't explain the change. Now maxmem runs just a few percent less cpu usage than burnmemory, and perhaps the latter sounds a bit better since now with our new math I am burning more memory to get a truer 256 remainder (3072-2716).
Please, if you have time and leasure, keep the record of these settings reverting from one to another and try to record the cpu usage. I wonder if both commands allow to bite off a different part of the physical memory every time, and it is important to catch the least possible, if it is possible at all.
Serge
(Certainly, if You have time to do it)
ALSO, Did You try yhe full line with /nopae and/onecpu settings?
What do you think of the sonic change?
Edits: 12/05/11 12/05/11
I have gone back and forth between maxmem and burnmemory three more times. I complicated it a bit by reducing the resulting memory used progressively. Still I never reproduced the very low cpu usage I saw at first. It must have been an anomaly or mistake of some kind. In all cases usage varied mostly in the range of 15-20% with spikes into the 20's. I have now gotten down to using 226MB with maxmem. It is hard to say there was any significant improvement in sq beyond the initial benefit of reducing to maxmem=256. If any it is refinement of ultra-refinement. I repeat that I have been using the full boot.ini line with /nopae and /onecpu all along with this.
It is puzzling to me given the overall goal of reducing cpu stress that this memory limitation improves things though increasing cpu usage--though it is now usage of one core instead of two. Contradicts our premise, doesn't it?
And it is something to think about.
Serge.
It is working harder with burnmemory by a factor of 5: 1-3% for maxmem vs up to 15%. These memory limitations should increase it some, yes? That is one reason I was questioning whether maxmem was really having an effect. Clearly burnmemory is re cpu work.
Decisively, /maxmemory should be used. And, btw. I am already hearing the roar of distant laughter at our common math. 1 Gb is 1024 mb (512+512)
Serge.
Obviously, we both have our minds on a higher plane...and what's a few dozen MBs one way or the other anyway.:) I am heading back to maxmem on the double.
with my own mind. I have prepared everything to rebuild my cmp again and can't make myself begin. As for the math, yeah, it is funny, because another thing was on my mind absolutely. Certainly, 1000 - 256 makes 778 and not 744 anyway.
Serge.
Last time I did not like maxmem=256 but putting the other stuff in (nopae/onecpu) makes it all work very nicely. Thanks for insisting I try the whole line as you wrote it in your post. Nice big sq pop! Not only is timbre right but dynamics and air are also better. Everything sounds very balanced.
Hi Serge
I'm very curious to know what you hear when switching HPET ON/OFF in your BIOS (...no other changes of course, just the HPET).
This question comes from the research I did earlier this year on using the HPET within a Windows XP context. The results/conclusions were posted in the cMP forum, see -> http://www.audioasylum.com/forums/pcaudio/messages/8/88573.html
In essence, XP-SP2 (and beyond) can recognize the presence of HPET on a modern mobo, but XP has no driver for it, and therefore CANNOT benefit from it's use. Not directly anyway.
However, if you are hearing a change in SQ by toggling this component in BIOS (and I certainly trust your reporting on the effect of subtle changes), some other mechanism must be responsible.
Here's a wild guess: perhaps it's related to EMI/RFI pollution on that particular mobo ? Maybe other mobos don't react the same way. This is all speculation on my part, but I just had to make up some kind of shaky answer.
Looking forward to your reply.
Cheers,
Grant
That's not a Toy... IT'S A TOOL !!
In fact the hpet device is in your system and is enabled without drivers like a vga compatible video controller, see the picture. The sound seemed colder? but more precise. (or I talked myself to believe). I want yet to try bigger values. and see. After all it was about a month ago. But I Kept this setting, so there was a change for the better. Grant, try the whole string I posted, in your bios and say if you like it or not. Only, the whole string.
Serge.
Give it a 1-1,5 hours time to settle.
Edits: 12/03/11
The pic is from untweaked russian xp on my work machine it is High precision event timer. If you rightclick it you read the drivers are not installed (Code 28) and below in device usage - Use this device (enable) and in resources you will see the irq etc.
Hi SergeI too get the equivalent Device Manager info for the HPET device... BUT, XP is not able to use it ! The fundamental code of XP was cast in stone long before HPET devices were implemented on motherboards.
If you read the (rather long) post indicated above (from the cMP thread), you will see the quote supporting this notion, and a link to where I found the information.
Regarding BOOT.INI parameters, my file contains all switches shown above except "/onecpu" and "/MAXMEM=256"... it's been that way as far back as spring of this year. I use a 512 mb stick of Hyper-X RAM, and run cMP^2 on the G31M mobo (...ancient, but still effective).
When you get the chance to test HPET In/Out, please lets us know your findings.
Cheers,
Grant
That's not a Toy... IT'S A TOOL !!
Edits: 12/04/11 12/11/11
Did you use the WHOLE line, or only /maxmem? Try the WHOLE line.
Serge.
I'm a contrarian on this one...sounds very similar to the known dll's deletion tweak. Things sound a bit more midrangy almost like a tube stage was added somewhere in the link. Its not unpleasant but disappointing for me. Maybe if you use nopae which sounds very uber solid state to me(tight bass, great dynamics but a bit edgy) it evens things out. I may leave it but I may have add nopae back in.
What if you have 256 mb ram? I'm going to try anyway ... just thought I would ask. If it works does that mean longer times between swaps and fewer (swaps) per track in cplay? Or not?
What sense in used memory limit?
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2 Ru, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Liv
To shorten "walks" between processor and memory and decrease the number of pages.
Serge.
Amazed by what this did. Outstanding tweak.
Fabulous work serge.... Thanks a lot..!!!
Junaid
Hi Inmates,
I went through Jolida's .Bat1-16 and registry tweaks in CMP mode, with Notepad.exe, Device Manager, Regedit.exe and Cplay opened. Notepad.exe and Regedit.exe were retained after file deletion. However, I was not able to go back to XP mode and the Device Manager no longer worked(can't install driver and components anymore). Any idea what files should be retained in order that XP mode and Device Manager?
Restore back reg.exe from Mihaylov's list to system32 & u will be able to get back to Xp mode..
I have sent my new updated batches today to u & all others.
Device manager needs a lot more files to work. We had given up device manager somewhere mid-way while doing the steppes. What do u need it for??
Junaid
Thanks. I have not executed Mihaylov's list yet as I have problem with them earlier. I checked and found Reg.exe in bat.15. I'll redo to see what will happen. I need Device manager as I want to use a master copy to install on different computers(my friends'). If this can't be done, I'll have to do fresh installations.About these tweaks, should we apply them all directly to the trimmed Nlite XP? I ask this question because on reaching bat.16 I can hear changes in SQ, I'm not sure it's improvements or changes and I need to compare carefully.
Highly appreciate what you are doing for the community. If our ultimate goal is to trim down XP to the smallest possible, I think it would be more effective to do it the other way round(if and only if we know what files should be kept and what should be deleted. Instead of deleting huge number of files, create a bat file to save all files that are required for CMP to work good(using another computer), delete the rest from C drive and put what we have saved earlier back in. Just an idea. Don't whether this will work better or not.
And in order to streamline what you guys are doing right now with different versions of XP Pro, a master copy should be used by all. Your Nlite ISO image is good for this purpose but I can't use it because it hasn't got Chinese language in it. I created my own Nlite XP but can't make it to 89mb, mine is 107 mb. Can you upload one without deleting the languages?
Edits: 12/03/11 12/03/11 12/03/11 12/03/11
I think this '...And in order to streamline what you guys are doing right now with different versions of XP Pro, a master copy should be used by all...' is a great idea!!
I think this '...And in order to streamline what you guys are doing right now with different versions of XP Pro, a master copy should be used by all...' is a great idea!!(nb: this is not a reflection on theob, just a convenient place to post this observation)
Gentlemen (and Ladies, should any be reading...),
It's one thing to talk about tinkering with Microsoft's XP product - and the many ways to bend it to our very particular needs - but it strikes me as an Invitation To Trouble for our "community" to openly circulate a copy of the OS.
How long before the Microsoft Police pull up in their virtual squad car, and issue a Cease and Desist order, brandishing words like "bootleg", "pirate" and "lawsuit" ?
There can be no valid objection to sharing the method for creating your own nLited XP install (from your own copy of XP), but passing around a live install disk (or ISO image) goes against that "terms of use" thing we never read.
By using the same "Last Session.ini" parameter file in nLite, we can each generate "the same" install from our individually held sources - conceptually at least. Minor variations may arise due to different language editions and/or service pack updates.
Just an observation. Let the fun continue.
Cheers,
Grant
That's not a Toy... IT'S A TOOL !!
Edits: 12/04/11
Ho gwjaudio,Why would Microsoft care?
As long as you have paid for a license too use XP, its fine.
(so no damage too potential turnover)Microsoft should better not put energy in alienating more customers.
They better put energy in making a better next generation operating software.
Windows 7 is still a little less flimsy Vista version in disguise.
Hope they do a (much!) better job with Windows 8.
As for me, it is far from sure that I will run Windows 8 on my next computer.
Nowadays Linux distro’s (from Ubuntu, Mint, Suse, etc) work instantly flawless right out of the box on the very latest (!!) hardware.
And also Apple is growing every year.
The only thing keeping me from Linux right now is lack of good high quality drivers for pro audio soundcards.Mark
fully optimized cMP2 PC -> ESI Juli@ -> Van den Hul Optocoupler MkII-> Lavry Black DA10 -> XLR Mogami Gold -> Klein & Hummel O300
Edits: 12/04/11
Hi Mark
I agree with all of your observations on Microsoft.
That doesn't change the fact that their business model relies on a license fee for every installed copy of their product (in this case XP). If the MS Auditor were to find the same serial number on 10 different cMP machines, I doubt the response would be "Move along, and Have a Nice Day".
To be openly posting what is in effect "Git yer free Windows XP Pro here..." cannot reflect well on our very well-meaning group effort. Far too easy to twist the facts into something "illegal", and bring dishonour to an exemplary piece of open-source cooperation.
By now, web-bots have no doubt informed their masters of the transgression. It remains to be seen if the imminent sun-setting of XP support pushes us off their priority list.
That's all for this topic. Let's be careful not to attract the wrong kind of attention to the cMp/cPlay Project. Amen.
Cheers,
Grant
That's not a Toy... IT'S A TOOL !!
to suit your own purposes.
Excellent point, Grant.
(sorry I missed you call)
It would be quite simple to change slightly the ISO nLite master file.
Jolida needs to upload his "Last Session.ini" file from nLite XP.
With such file, combined with your own XP master disc, you can make a new ISO file slipstreaming Chinese language, drivers and so on...
U want to integrate Chinese language support in the disc ? I will upload it in while. It's 3.am now. Will do it in the morning. Do u want just the ini file or do u want me to slipstream Chinese into the ISO & upload it again? Either way is fine with me..
Junaid
Not the .ini file, to be useful to me I need Chinese Traditional and Simplified included in your Nlite image. My XP SP2 is of a different built from yours and the size is different after trimming. I suggest the Languages should be left intact so that inmates and select on their own.
Had thought of that idea a few weeks ago. I guess it's gonna end up likewise, as the batches have been made already. If they were not, then a reverse gear would have taken us ahead..!!
Try to ascertain if u like the sound after step 16. As u said, & as I have put it many a times, a change need not always be an improvement. Listen to a lot of familiar recordings to analyse what's happened. Once u install nLite & cics optimise it, take a back-up, before u proceed with deletions or touch the registry. Always helps to get back to where u started. With back-ups & batch files handy, u don't have anything to lose, except some time, but it's worth experimenting to ooze the utmost potential out of Ur system on a software-level.
My recent nLite install has totally changed the performance of my cMp. The best I've heard in a long time, even without the deletions done. I intend to proceed slow & listen a lot more at each stage, to ascertain the improvement each steppe makes. I guess it's high time we finalise what's to be removed & what to be left in..
Steppe (serge) has done a remarkable job throughout, & set a benchmark in the history of cMp. Absolutely stunning work, & we all appreciate his efforts.
What u like & what u don't also depends on Ur entire set-up as well as Ur listening taste. Some like a very laid-back sound, some like it Live with lots of energy. So one tweak may sound quite the opposite to another. As I said, take Ur time & analyse, & take a back up at every stage which has made a pleasant change to u..
Afterall, it's all about what sounds best to You..!!
Junaid
file deletions? I press enter and there is a long list rapidly scrolled, saying smth like bad location or smth, some files however get a longer comment, but still I can't read that fast. Anyway, nothing is deleted, I even copied a bat into system32 and started it there....
Sorry, new thing for me.
Serge.
Do I have to have another (drive d) attached as a must?
I just modified the Batches for u & sent it to ur email. Its the same, except that it does not require a 'D' drive to work.
Just double-click on the Batches & u r good to go...
Junaid
Junaid! You are REALLY helping me and also thanks to NAGUAL19. I read what he said and finishing step12. Will see. I tried step2 or 3 it created a folder on my disk e. there are files inside but while doing the echo is: the system cannot find the file specified or syntax incorrect is it ok? I think it's just because of our minimalistic install. BTH, read my reply post to Mihaylov's registry tweak. ABOUT THE SIZE.
Serge.
While running my batches, it's absolutely normal to see the File not found or the System cannot find the file specified error. It's because I have written the batches in such a way that it will search for the file in multiple folders, & since one file cannot be present in all the folders, it will show up that error. Don't bother about it. The Batches will work as intended & the files will be moved.
Yes I did read your reply to Mihaylov. I too am looking for that info online. Hope to sort it out..!! See if this can be of help:-
http://blogs.technet.com/b/markrussinovich/archive/2009/03/26/3211216.aspx
Junaid
Edits: 12/03/11
mostly not the info, though it is of great importance. I mean the size of the registry and our approach to what we aim at. It's not the size, as long as it is less than 4mb which is a champion's achievement itself. I mean that we ought to strive to cut off everything really unnecessary. I have just tried my own registry tweak about known dlls - good but not overwhelming, though it's hard yet to tell. The further we get during deletions the more evident and true is any registry change. BTW, I think I should have set cmp to start cmp, before proceeding with batches, shouldn't I? I didn't. Run all 16 at once, rebooted and couldn't log on. No minlogon was implemented yet. So many variables. Will start now once again. (thanks to you and nagual19 the batches are ready.)
I think I have found the mistake while preparing nLite install see the picture. This thing shouldn't be deleted, otherwise - the awe problem. Also see that shutdown.exe and taskkill.exe are not removed from installation.
Serge.
Will be posting comments, if you permit and if they are not obsolete.
I have not deleted Group Policy Management Console while making the nLite disc. It's in. Also, I am using the same nLite disc I have uploaded & AWE works absolutely fine. Just follow the instructions & it's done. Even if it does not accept the word Administrator, it's fine. Just click Ok & cancel to close the small windows, & in the Main window click on "File> Save", then again "File> Exit". Reboot, & AWE is successful....
Junaid
Hey Serge,
Looks like u do not have a 'D' drive on ur Hard Disk. If not, dont worry.. Do u want me to edit my Batches for u in such a way that the files are moved in another manner? I sure could do that. Let me know.
And nagual19, u r cool :)
Junaid
Yes, the way Junaid has designed his batchfiles, you'd need a D:\ partition to make them work per default. But of course you can edit them to make it work, just open every *.bat with notepad or your favourite text editor, for example this is Junaid's step2.bat:IF NOT EXIST d:\step2 md d:\step2
MOVE c:\windows\system32\audiosrv.dll d:\step2
...etc...First line creates a new folder 'step2' on D:, second line moves the file audiosrv.dll to the newly created folder. Just find-replace all D:\ strings and edit to point to any directory you like.
When you finally execute the batches, be aware that there will be still some saying "file not found...", because the batches are searching for the respective files in several locations (in this case, audiosrv.dll is searched for in \Windows\system, Windows\system32, and Windows\system32\drivers)
Some kind of brute force attack, but works quite well. Keep in mind, if a file is locked in memory by Windows, the bat also cannot override this (just the same if you'd use just windows' GUI) and the file stays in original location.
PLEASE NOTE: IIRC, only the later (16...?) Junaid batches are dealing with *folders* that are to be removed. So just move folders like WinSxS\Manifests\... manually. Anyway, these batches do speed up the whole process a lot.
Edits: 12/03/11
Will try it again now. Thank You!
Serge.
Now the size of registry (size of the Config folder) is 929 KB.
First I used the portable version of the jv16 PowerTools to clean and compact the registry. Then I cleaned the registry manually using regedit.exe. After all I exported default, sam, security, software and system from registry saved them as hives on my NAS through network. Then I replaced the files default,sam,security,software and system in Config folder by saved hives (under the same names).
To describe the registry removings is difficult because they are very numerous.
Here are some screenshots which help to fathom that it is possible to remove.
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2 Ru, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Liv
See link, password: Serge
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2 Ru, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Liv
Edits: 12/03/11
Wow this is really an accomplishment that you have reached....bravo. So tell us how it sounds. Apparently there is a bright light in Moscow.
Theo, for a long time already the sound in my system is great therefore it is difficult to tell about any appreciable changes as I do much tweaks recently without having time to trace changes carefully.
In this case I probably take the more interest in how much possible to reduce a system size.
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2 Ru, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Liv
Hi, Serge!
1. Did you delete all the branches inside the selected branch, lika ALL HKEY CLASSES ROOT or you show only what remains?
2. Having an extremely small registry size is NOT ITSELF AN OBJECTIVE. Matter is, however small, the registry eats 4mb in paged memory pool. There is a feature like RSL (registry size limit), if RSL or the actual size of the registry is less than 4mb its size is ignored and by default 4mb will be allocated anyway. Now the deletions of entries that invoke unnecessary threads is MORE IMPORTANT, and here I congratulate you.
3. Serge, you seem to be much more apt at computer than me. Please, try to find these two things: assigning memory (in mb) to an application - in our case it is cmp-cplay and MOST IMPORTANT - how to set the size of pages, that awe locks from 4kb to ! 4mb. It IS POSSIBLE in xp, but the info is somehow hushed or vague.
With all my respect, Serge.
http://www.pctools.com/guides/registry/detail/1015/
Junaid
http://msdn.microsoft.com/en-us/library/windows/desktop/aa366527%28v=vs.85%29.aspxand
http://blogs.technet.com/b/askperf/archive/2007/03/07/memory-management-understanding-pool-resources.aspx'The physical pages that can be allocated for an AWE region are limited by the number of physical pages present in the machine, since this memory is never paged'
It appears then, that we (assuming 1x512MB ram) want to maximize the non-paged Memory (which would be regvalue 'NonPagedPoolSize' in bytes) ?
Steppe of course has the idea (sounds promising) to make the AWE memory pages larger (http://www.audioasylum.com/forums/pcaudio/messages/9/99374.html), I'll also try to search for info on this ;)
First link says at last, 'AWE provides page size granularity appropriate to the processor (for example, 4 KB on x86), which is more useful to applications than large pages (for example, 2MB or 4MB on x86).'
So we might try to search from another perspective, that is making Windows believe its using some other 'CPU type' (...) if possible at all, to make it use larger pages. Thinking of, a lesser interval that the ram is accessed would lead to less mobo/psu-disturbances (which leads to ...cMP users know ;)
Or second possibility, Coder Cics joins back in and may shed even more light on the mermaids (already shimmering in 24bit colours I think) tail ;)
Edits: 12/03/11
Throw more light please. Its still dark :)
Junaid
In Moscow it is already daylight ;).
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2 Ru, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Liv
It's daylight in India as well :) but Ur tweak has clouded the light... Let the details rain out of Ur tweak so that it gets brighter again...
Junaid
Made a new install with an nLite disc. Sounds entirely different..
Junaid
Edits: 12/02/11 12/03/11 12/06/11
now? It won't let put in administrators!
Serge
I have installed using that very same nLite disc & enabled AWE.
When u type in Administrators & click Ok, that window will remain after a blink. At that moment two small windows are open. After u click Ok once, press Cancel & close both the small windows, & in the main window, goto 'File' & SAVE, it will assign a name something as Console root 1 in a directory. Save it. Then again on the same main window, goto 'File' & exit.
AWE will now be enabled. Restart the system for it to take effect & confirm it in cPlay diagnostics...
Junaid
for me the trick was to replace "Administrator" with everone and anonymous user. then it would save. I followed the other User Accounts settings in cplay optimizations, rebooted and then awe would enable.
or after minlogon? I have tried before. I typed administrators, Administrators, administrator, Administrator, computer administrator and so on. No way...
That's good. Though I got AWE enabled using Administrator itself...
Either way, as far as it worked :)
Junaid
Go ahead and type in Administrators and click whatever it is you click, can't remember!, yes it will tell you it does not recognize the name. Cancel that screen. Proceed as you would with a more normal install. Be sure to SAVE and EXIT.
I then go to cMP and see if I get the AWE Successful message on diagnostics and I do. Every time.
SO, don't worry about it. Just make sure you get the assurance.
awe failed... reverting to normal or smth like that and AWE IS ESSENTIAL for future tweaks.
This is what I think: As awe is a group object setting the error (advanced tab shows an unrecognized error) is to remove Guest from nLite install. Yes, it removes 1 control set from the registry, probably, but AWE is MORE IMPORTANT.
Serge
It can be done.
Do it again.
Please confirm, since I had already burned the first to CD.
Exactly !!! The 89mb one is the correct image.
I made a fresh install using it, installed drivers for my soundcard & chipset, disabled codecs, applied minlogon, AWE, & installed cMp & cPlay. I haven't even completed the basic Cics optimisations in the Registry nor any other apart from the above mentioned yet..
Have been listening for almost an hour now, & it sounds just amazing at this stage itself. I wonder where the mods & tweaks would take this. The best cMp I have heard so far. I strongly suggest all inmates to do this.
By the way, the size of my config folder without any registry mods itself is 6.8mb...
Junaid
this is what was on the iso
I'm sorry. I just posted another ISO image & edited my post. Please download it & burn it to a disc. I just did an install using it & confirm working !!!
Junaid
It is I who has to apologize for causing so much trouble. Thank You, Junaid!
Serge
Please download the ISO & install Xp using it, install Ur soundcard drivers & apply minlogon. I just made a fresh install using it & am listening to it. Even without the basic Cics optimisations, it's sounding so good. I wonder where this will land up after the optimisations & registry mods & file deletions done. Sounds so promising. Am excited altogether....
Junaid
Just curious if you left more "in" than the three things I kept.
And if you did, WHY?
A very important thing to be left in was the Multi-processor support in the Hardware tab. If not, the installation would be restricted to a standard install with maximum a single Acpi install. I had followed Ur instructions to make my first nLite disc which was successful but hugely different in sonics compared to the one I made today.
I guess now things would move in the right track. The sound is absolutely balanced from the top-most to the bottom registers...
Junaid
I installed with BIOS set for one core.
Do you think this would make a difference?
I so knew that question would come up :)
I can answer that in one line. If the OS is seeing only ONE core in the bios, then HOW DOES IT INSTALL AS AN ACPI MULTIPROCESSOR PC, when the setting are in Automatic?????
As simple as that. Even I have disabled multithreading in my Bios long back, & only one core is enabled. But inspite of that, windows recognises it as a multiprocessor. I guess there is something more to it that just enabling or disabling...
The nLite disc which I have uploaded was made with Multiple processor support enabled and the Computer type I left it to Automatic.
Try installing with that disc & I'm quite sure that even ur's would be installed as an Acpi Multiprocessor, even though only One core is enabled...
Junaid
Edits: 12/02/11
for MULTI PROCESSOR SUPPORT.
Set to one core in BIOS before installation.
Extremely spacious, no lack of mid bass energy, clean, sweet high frequencies ...
When i had made my first nLite disc following ur instructions, i did like it too, & i had even made a post on that. The sound was clean but BLAND ( a bit tasteless). I was expecting it would settle in after deletions. But the thing that got me worried was the high latencies i was getting with that install. Upon checking in the device manager, i found that it was installed as a 'Standard Pc'.
I very well remember that all our previous installations of windows, which had been done in all these months, are installed as an "Acpi Multiprocessor PC". Im sure u & all other inmates have previous back-ups handy. Try restoring one of them & check for urself. My question is, "Why was it installed as a Multi-Processor when only One core was enabled even then??????????????????????????? There should be something to it right...!!!
And i too assure u, that the sonics are way superior in every aspect in this new install, compared to the one i made as per ur recipe...
I strongly suggest u try & hear it for urself..
Junaid
If you did not have but one core enabled your installation is not the same.
Ofcourse I had only one core enabled. And I always did..
Junaid
Was your BIOS setup for only one core enabled during the installation using the nLite disk?
Do you use the standard cMP optimized BIOS setting for installation or do you implement after installation? This makes a big difference in what is installed. I just need to know this before I start the install process again. If you did use the cMP BIOS then I will have to try. If you did not, our installs are not the same.
I know you would have used the cMP BIOS eventually but I am trying to understand how you handled the installation. Your answer could mean many things.
Sorry to be so determined.
I have always used only ONE CORE ever since I started Setting up my cMp. I have never used multiple cores ever. It's been around 2 years almost, when I has set-up my cMp & since then I have not touched the Bios. Only one core is enabled, whether it's a regular Xp install or an nLite...
I started on nLite just last week for the first time. When I made the first disc, I followed Ur instructions step by step. But all I could have, was a Standard & 486 something (not sure) during the installation. And I VERY WELL remember that all my previous installs were Auto-selected by Xp as " ACPI MULTIPROCESSOR PC" even-though-only-one-core-enabled..!!!!!
So while making my next nLite disc, I left the Multiple Processor Support option remain, but chose the Computer type as "Automatic" to see what the installation prefers for my Hardware. And not surprisingly, it again installed as an Acpi Multiprocessor.
I don't know how this could happen, but I'm very sure that it will happen to all of us. Why don't u restore a previous non-nLite image u have to check how Ur computer was detected. Even then u had only one-core enabled. But I'm sure it would have been installed as a Multiprocessor Pc...
Hope this helps..!!!
Junaid
Just had to be sure since I thought many of us would revert to DEFAULT BIOS at installation.
I know with a standard WINDOWS install I would do this since it would stall when I tried using the cMP BIOS.
I remember you posting about having installation difficulty with 1G memory and this was what made me wonder.
I will now have to try with MULTI PROCESSOR in the mix.
One interesting thing was the machine working without ACPI.SYS since it was never there. Before when I tried to delete it (without the minimal install) it would not work. No surprise there.
Does this discussion explain why my setup sounds better with two cores (and HT) enabled? i did the cics- and bios-optimations after install. Should i reinstall wit one core and HT disabled?
tia
Douwe
I know I am quite pleased with what I am hearing and so is Junaid!
I am going to try putting the MULTIPROCESSOR thing back into the nLite install disk and hear what I think.
What is HT? I cannot think of what that is. Dull head this morning.
I tested: two minimal nLite installs: one single-core, the other dual-core. Two cores give little more space an depth. Sounds better in my setup (h55m-usb3-i3-530).
HT is hyperthreading.
Later maybe test acpi vs standard-pc install.
Thanks Junaid, great!
What did you include / exclude?
Is it a startup like RickMcInnis described? Or are already some steppes/BitchBatch involved?
Douwe
It's a minimalistic Xp install disc. Burn the ISO to a CD & install Xp using it. It's already optimised eliminating 70% of the components in a regular Xp installation. After installing Xp using the nLite disc, install Ur chipset & soundcard drivers, & do cics optimisations..
Junaid
First, let me thank Junaid for this generous gift.
Just install it as recommended, and it is one of the better setups I have heard yet still fully functional.
First I applied the CICS optimizations. I played around with various levels of batch file deletions and have not yet concluded which is my favourite. I did the Bold Fortune deletions followed by steppe to 12. At this stage it is very nice having opened up but remained very full and musical. I added Steppe's lost registry tweak and the watchdog and I think it may have been perfect. I went on and applied Steppe to 16 and the Mihaylov tweaks. Still very good but I am not sure if I might prefer 12. I kept images and will toggle to see.
In all and excellent.. and easy!!
I have not yet started file deletions. Have just made cics optimisations & just ran my registry batches followed by lost & last registry, & watchdog tweak. In short, I am just tweaking with the registry to it's fullest possible to analyse how impactful it is vs file deletions. Serge's boot.ini tweak has entirely changed the performance. It's just remarkable.
At this stage, I am very happy with how it's playing. No complaints at all. I intend to move very slow & careful so as to not get into a complication of whether this file is better to keep or delete kind. Once I am certain with the registry section, only then will I start to run batch by batch & decide how to proceed.
It's good to know u liked Ur sonics as well. I guess we are moving in the right direction :)
Junaid
As most of the files have been removed, & Registry Mods still keep coming up, its quite necessary that the mods be done in their Primitive way instead of Loading hives via browsing to the Config folder. One example is while doing Steppe's Watchdog tweak in the Registry. If done from another pc by Loading the Hive, only the Control set 001,002,003, or likewise would be available. But it needs to be done at the 'Current Control Set', which would show up only on the Music computer. So its better to do it from the cMp itself. But here arises another problem of Regedit being deleted already. So all that needs to be done is to restore back the few files that Regedit needs to work, make the changes in the Registry, & if needed, delete those files back again. The files required for Regedit to work are :-C:\WINDOWS\AppPatch\acgenral.dll
C:\WINDOWS\system32\aclui.dll *
C:\WINDOWS\system32\advapi32.dll
C:\WINDOWS\system32\authz.dll
C:\WINDOWS\system32\clb.dll *
C:\WINDOWS\WinSxS\x86_Microsoft.Windows.Common-Controls_6595b64144ccf1df_6.0.2600.2180_x-ww_a84f1ff9\comctl32.dll
C:\WINDOWS\system32\comdlg32.dll
C:\WINDOWS\system32\gdi32.dll
C:\WINDOWS\system32\kernel32.dll
C:\WINDOWS\system32\msacm32.dll
C:\WINDOWS\system32\msvcrt.dll
C:\WINDOWS\system32\ntdll.dll
C:\WINDOWS\system32\ole32.dll
C:\WINDOWS\system32\oleaut32.dll
C:\WINDOWS\regedit.exe *
C:\WINDOWS\system32\rpcrt4.dll
C:\WINDOWS\system32\shell32.dll
C:\WINDOWS\system32\shimeng.dll
C:\WINDOWS\system32\shlwapi.dll
C:\WINDOWS\system32\ulib.dll *
C:\WINDOWS\system32\user32.dll
C:\WINDOWS\system32\userenv.dll
C:\WINDOWS\system32\uxtheme.dll
C:\WINDOWS\system32\version.dll
C:\WINDOWS\system32\winmm.dll
C:\WINDOWS\system32\ctype.nls
C:\WINDOWS\system32\locale.nls
C:\WINDOWS\system32\sortkey.nls
C:\WINDOWS\system32\sorttbls.nls
C:\WINDOWS\system32\unicode.nls
Only a few of the above (marked with a *) had been removed with the steppes, so they need to be temporarily restored. If u plan to delete them back after tweaking the registry, then just write down the names of those u had restored.
Also i am enclosing a few other important sets of files needed for the working of other essential tasks like :-Notepad:-
C:\WINDOWS\WinSxS\x86_Microsoft.Windows.Common-Controls_6595b64144ccf1df_6.0.2600.2180_x-ww_a84f1ff9\comctl32.dll
C:\WINDOWS\system32\winspool.drv
C:\WINDOWS\system32\winmm.dll
C:\WINDOWS\system32\version.dll
C:\WINDOWS\system32\uxtheme.dll
C:\WINDOWS\system32\userenv.dll
C:\WINDOWS\system32\user32.dll
C:\WINDOWS\system32\unicode.nls
C:\WINDOWS\system32\sorttbls.nls
C:\WINDOWS\system32\sortkey.nls
C:\WINDOWS\system32\shlwapi.dll
C:\WINDOWS\system32\shimeng.dll
C:\WINDOWS\system32\shell32.dll
C:\WINDOWS\system32\rpcrt4.dll
C:\WINDOWS\system32\oleaut32.dll
C:\WINDOWS\system32\ole32.dll
C:\WINDOWS\system32\ntdll.dll
C:\WINDOWS\system32\notepad.exe
C:\WINDOWS\system32\msvcrt.dll
C:\WINDOWS\system32\msacm32.dll
C:\WINDOWS\system32\locale.nls
C:\WINDOWS\system32\kernel32.dll
C:\WINDOWS\system32\gdi32.dll
C:\WINDOWS\system32\ctype.nls
C:\WINDOWS\system32\comdlg32.dll
C:\WINDOWS\system32\advapi32.dll
C:\WINDOWS\AppPatch\acgenral.dll
Process Explorer :-D:\ProcessExplorer\procexp.exe
C:\WINDOWS\WinSxS\x86_Microsoft.Windows.Common-Controls_6595b64144ccf1df_6.0.2600.2180_x-ww_a84f1ff9\comctl32.dll
C:\WINDOWS\Temp\Perflib_Perfdata_2bc.dat
C:\WINDOWS\system32\xpsp2res.dll
C:\WINDOWS\system32\wtsapi32.dll
C:\WINDOWS\system32\ws2help.dll
C:\WINDOWS\system32\ws2_32.dll
C:\WINDOWS\system32\wldap32.dll
C:\WINDOWS\system32\wintrust.dll
C:\WINDOWS\system32\winsta.dll
C:\WINDOWS\system32\wininet.dll
C:\WINDOWS\system32\wbem\wbemprox.dll
C:\WINDOWS\system32\wbem\wbemcomn.dll
C:\WINDOWS\system32\version.dll
C:\WINDOWS\system32\uxtheme.dll
C:\WINDOWS\system32\userenv.dll
C:\WINDOWS\system32\user32.dll
C:\WINDOWS\system32\unicode.nls
C:\WINDOWS\system32\sorttbls.nls
C:\WINDOWS\system32\sortkey.nls
C:\WINDOWS\system32\shlwapi.dll
C:\WINDOWS\system32\shgina.dll
C:\WINDOWS\system32\shell32.dll
C:\WINDOWS\system32\shdocvw.dll
C:\WINDOWS\system32\setupapi.dll
C:\WINDOWS\system32\secur32.dll
C:\WINDOWS\system32\samlib.dll
C:\WINDOWS\system32\rpcrt4.dll
C:\WINDOWS\system32\riched32.dll
C:\WINDOWS\system32\riched20.dll
C:\WINDOWS\system32\psapi.dll
C:\WINDOWS\system32\powrprof.dll
C:\WINDOWS\system32\oleaut32.dll
C:\WINDOWS\system32\ole32.dll
C:\WINDOWS\system32\odbcint.dll
C:\WINDOWS\system32\odbc32.dll
C:\WINDOWS\system32\ntshrui.dll
C:\WINDOWS\system32\ntlanman.dll
C:\WINDOWS\system32\ntdll.dll
C:\WINDOWS\system32\netui1.dll
C:\WINDOWS\system32\netui0.dll
C:\WINDOWS\system32\netrap.dll
C:\WINDOWS\system32\netapi32.dll
C:\WINDOWS\system32\msvcrt.dll
C:\WINDOWS\system32\msgina.dll
C:\WINDOWS\system32\msasn1.dll
C:\WINDOWS\system32\mpr.dll
C:\WINDOWS\system32\locale.nls
C:\WINDOWS\system32\linkinfo.dll
C:\WINDOWS\system32\kernel32.dll
C:\WINDOWS\system32\iphlpapi.dll
C:\WINDOWS\system32\imagehlp.dll
C:\WINDOWS\system32\gdi32.dll
C:\WINDOWS\system32\dbghelp.dll
C:\WINDOWS\system32\ctype.nls
C:\WINDOWS\system32\cscui.dll
C:\WINDOWS\system32\cscdll.dll
C:\WINDOWS\system32\cryptui.dll
C:\WINDOWS\system32\crypt32.dll
C:\WINDOWS\system32\credui.dll
C:\WINDOWS\system32\comres.dll
C:\WINDOWS\system32\comdlg32.dll
C:\WINDOWS\system32\clbcatq.dll
C:\WINDOWS\system32\browseui.dll
C:\WINDOWS\system32\atl.dll
C:\WINDOWS\system32\apphelp.dll
C:\WINDOWS\system32\advapi32.dll
C:\WINDOWS\system32\aclui.dll
C:\WINDOWS\Registration\R000000000014.clb
Another very important task, that will be of use throughout our Modification route, is Registry mechanic. I will post a list of files needed for it to work...
Junaid
Edits: 12/02/11
I was able to do Watchdog tweaks for Current Control Set while in another computer using load hive routine. Its interesting that you cant.
The Load Hives option is best used in cases of emergency or during a Virus attack on a Pc, when the system does not boot. In our case, it's optimising the system. And it relies mostly on the Current Control Set. At this moment, when we are mid-way, mixing up things, & expecting more registry tweaks to show up, all we could do is get back Regedit to work temporarily to apply a tweak. Once Serge had FINISHED his registry tweaks altogether, then it makes setting up a full-fledged, hassle-free cMp much defined & easy. At that stage, we could accomplish ALL registry tweaks first, & then proceed with file deletions, followed by a side-by-side clean up using Registry Mechanic.
But at this stage, the only way to implement it properly is to restore Regedit back & do it in the primitive way...
Junaid
1. Is not the current control set drawn from the numbered control sets? That is, if you delete from the numbered ones will not the current set on restart also show these are deleted?
2. Does the system need at minimum more than one control set? That is could we possibly remove all but one numbered one?
CurrentControlSet is simply the set that is being used.
From my experience in my setup this is ControlSet1 - I would think one could delete the other even though it is considered a backup - BUT with the .sav files deleted maybe the OS won't care.
With an image made of a working installation even if something goes wrong in the future (ControlSet1 is corrupted) one could just re-copy the install. There is always the possibility that the OS will simply make another copy and deleting might not get us anywhere.
I will try tonight!
Did you ever get to trying paring down to only one control set?
I have not much of an idea regarding that. But I guess the Current Control Set differs from the other sets...
Junaid
Enough to save only the Current Control Set and Control Set 001 (they are identical in content but requires both). My cMP2 works fine. Although perhaps it's again feature of the Russian Windows XP ;).
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2 Ru, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Liv
Junaid! You were the only one to say you would try deleting entries in Known dlls. Did you do it? what was the sonic impact, if any? Did Serge Mihaylov scare you and other inmates with his funny post?
As for the info about the memory. It was not EXACTLY what I need. Matter is that awe locks pages into memory, but they are only 4kb in size. What I am looking for is the way to set in the registry om any other way this 4kb limit to! 4MB. It will reduce processor-memory interaction TERRIBLY/ And this interaction is one of the sources of playback inconsistency and jitter.
That is all. I will go rebuild my cmp, or it will never happen!!!
I went through DEFAULT and SOFTWARE and deleted everything with no entries or things that corresponded to files deleted.
Went through SYSTEM, same installation on a different SSD deleting anything that looked useless. Both booted up and played just fine.
I had no time to listen to the first version since it was too late. The second version, which I did first (???) and which includes your latest recommendations worked fine, also. Listened to that for a couple of hours and liked what I was hearing.
Will know more tonight. I will copy the modified DEFAULT and SOFTWARE folders into the SSD with the modified SYSTEM folder tonight and hear what results.
From what I can see config should be around 4mB after this.
Yes I was scared by his post :)
Will try it for sure. I have nothing to lose, as I have back-ups already...
Junaid
I have the feeling that CURRENT CONTROL SET is something the system creates for itself within a working installation.
I do not think it really exists in the config file.
My opinion is that if you delete an item from both sets (my setup only has two) it will not be in the CURRENT CONTROL SET since it is NOT there any longer. I have found from my manual registry deletions that the config file does get smaller. Not very quickly and not by large amounts but it does get smaller. I trimmed about 100K from the software folder last night.
Of course, if you are going back and forth from being able to use REGEDIT in the cMP machine you could easily prove this one way or another.
I think if one does the "load hives" method and removes stuff there it will not be in the "expanded" version of the registry that is displayed in a working installation. Where else could this stuff come from?
I look forward to doing a new install with RM to see how different the two registries will be.
2nd cmp2 box has been doing well. I have only done steppes 1st and 2nd steps.
So I boot up and get a bsod. It had a bunch of zeros. Then I reboot and same thing. No time to mess but the 2nd bsod had the following:
0x0000007e (0xc0000005, 0x805cac46, 0xF897666c, 0xF8976368)
Looked on line an no real info.
Here is what fixed it and if anyone can shed some light I would be grateful: I went into the bios and changed the speed setting of the cpu from 1.6 to 990mhz since that was the only thing that had changed. It booted right up and has done so 2wice now.
Never had a bsod that was fixed by underclocking. Was this just a coincidence or does deleting some of the files outlined in steppe's steps cause this???
No one here remembers the bending of our minds
It shouldn't happen. In none of my tweaks and after none of them, a blue screen should appear. I can only think of one thing: if you were removing files not from your cmp hdd directly, but from another windows and DELETED the windows pcm codec? It's a sticky one with *.acm extension that won't let itself be deleted on the early stages. Otherwise, I really can't think of anything. Especially that early. What is your boot.ini "legend" by the way? Have you done /CLKLVL? Remove it.
Serge.
Hi Serge,
I probably wasnt clear. It had nothing to do with your steps. FOr instance I did the first 2 steps and everything worked fine.
The system just worked and was consistently booting at 990mhz. Then one time it didnt and BSOD. Reset to 990 and viola it now boots fine.
WHat do you mean by "legend"?
And what is clklvl. It has been a long time since I did the steps.
Have you tried changing your cpu speed in the bios with your steps in place?
It might not have anything to do with the steps, but all this time and so many auto bios reclocks from 990 to 1.6 and no problems. I do the steps and now it seems that a reset can cause a blue screen.
No one here remembers the bending of our minds
Hello! It's best, imho, not to reset the speed of the cpu. Also, take a look and read carefully, and if you wish, implement, but slowly my tweaks 21(corrected edition) - it's all about bios and memory timings. You can do it anytime. It doesn't depend on my windows tweaks. Gives splendid results but requires a long break-in period. After a 24 hours initial break in time it will usually start sounding very good in about 35-40 minutes (for cmpextreme).
/CLKLVL is a setting in the boot.ini line like /fastdetect. I called this last line with kernel parameters legend here just for fun. Use it, but use it only once, to taste it, as IT will reset your bios FOR SURE, unless you perform it with cpu working at normal speed. Very dangerous but shows what we are missing with so many different clocks in our computers differently ticking. A thing to work on for the future.
Serge.
I have great stability in my system running @ 840 mhz after all the steppes (1-22) and almost all of the registry tweaks. My windows size is something like 45 mb. My home pc windows is like 10 gb so I believe the smaller the better esp for stability. The other day I took out one of my hdds to load some new music and I got a bsod while playing cplay (before the steppes I used to get these once a month or so). So I dont think steppes 1-2 will do anything to increase bsod risk. Sometimes sh** happens.
After making the batch files for various Steppe's, Jolida's & Mihaylov's lists, as well as their Registry Mods, i now wish to make a SINGLE batch file, which will take the cMp from the basic Optimised Cics stage to the Steppe 22, including all the above mentioned lists, Registries, Screenshots, & Optimisations...in ONE-CLICK..
Since inmates have different flavours of cMp (Julia@, Lynx, Networked), i request u to give me a list of files to be RETAINED for each of the above. For eg. I am a Non-Networked Juli@ user, so i may remove files which the Network-user may need to retain. Same goes to users of Lynx cards. As i have completely built my cMp various times, i know which files are needed for a user who uses the cMp with Juli@ & without network. I guess only Two other sections remain..."Lynx, & Network"...
If someone can give me a list of files to be retained for the above Two, i can make an All-in-One Batch that would be Customized to the above Specs...
Junaid
Hello, Junaid, and everybody! Tell me please, if registry mechanics allows to load and optimize a registry hive from a slave drive, I attach with esata.
ABOUT THE LIST of files not to touch, do you need the list of all the remaining files or only those, that I know are necessary for, say, Lynx, that are: drmk.sys, portcls.sys, ks.sys and wdmaud.drv?
Also, about cics' registry tweaks, aren't they supposed to superceed services and awe and minlogon tweaks? If so, then would it be logical to add them to a bitch bat?
Serge?
Edits: 12/01/11
Registry Mechanic is an automated Registry Cleaner. I have tried attaching my SSD to another computer & then Load the hive of my SSD. Then ran Registry Mechanic on that computer to see if it picks up errors on the Loaded Hive, & Yes it did. It repaired a lot of errors, but when I connected the SSD back into the cMp, all I got was a BSOD.
This shows that automated registry cleaning should not be done by loading hives. Instead must be done on the music computer itself. I will post a method shortly that will help in sustaining the program for later as well.
For Ur second question, Yes those are the file I need. I do not want a screenshot of all the remaining files in Ur system, instead just the ones which Lynx would not work without. I hope other inmates using different soundcards may post their list as well.
For Ur third question, I wanted to make the Bitch batch such that it starts work After cics optimisations already done. Whereas u have told me to include that part as well in the bitch? Is that what Ur saying ?? U want the Bitch to work soon after installing Windows with Cics Optimisations integrated??
Junaid
I have seen that nLite allows to initially set the display settings to default refresh, but will it remove dllcache folder, No, I don't think that cics optimisations should be integrated. I think some manual labour is still required. Also, nLite allows to disable services from the beginning, but how to install lynx drivers then and or cmp-cplay software? I also disable these services (rather check that they are disabled with msconfig) msconfig is also used by me to remove fonts, to remove lynx mixer from loading on startup, as I don't need it with cplay and that is one service less. Really, while I was serching for new registry-memory tweaks all of you have gone ahead. I will start rebuilding today from scartch, but will also try to create my nLte iso myself, just for practice. My first experience was a 86 mb iso, but it failed.
Also, using mmc you can disable ip security. I have posted refferences to it, noone reacts.
And there is a very interesting thing to do - disable periodic checking if the network is present!!!
One more thing to consider is the mouse - ps/2 or usb support and keyboard support or removal. On what stage? I am still at a loss. Slow cpu inside my head...
Summing up, I think that the Bitch Bat should still allow for manual tweaking, starting with at least some cics recommendations.
Serge
Edits: 12/01/11
so having nLite do it is no big deal.
I do not mark any of the services on my nLite disk since, as you have found, there are very few of them left.
Easy enough to do after all software and hardware is installed.
I have made many iso images for my nLite install. But I am confused at a particular stage. Though the install is successful & working, my Dpc Latency has shot up. In my previous regular builds, it was 10uS. With nLite installs, it's shot up to 50 to 60. Something needs to be checked. It's quite easy to integrate Ur soundcard drivers in an nLite iso. Select "Drivers" as well in the Main screen, & keep the inf file for Ur Lynx ready, & point to it when the screen for integrating drivers shows up...
Do check Ur latency after installing with nLite. And use the computer type as Acpi Multiprocessor Pc...
Junaid
Good reminder about acpi - I left it to automatic. As for drivers - they are on a cd and then I upgrade them with v2setup32.exe downloaded from their site. Do I need to have the inf from the cd on a hdd?
nLite will integrate Drivers based on the inf file u provide for the device. The exe files will be taken only if their inf is present. The easiest way is to download Ur Lynx drivers & keep them in a folder. The folder will also contain an inf along with the exe files needed by Ur card. When the Driver integrating page pops up, just point to the inf file in the folder & all the necessary files will be slipstreamed automatically...
Junaid
Junaid, is the nLite.iso, that you have created, you know the one I mean, is set for unattended setup. And, if yes, will it format/quickformat the drive?
Serge.
No. It has to be manually done. I have integrated the serial key though. I guess we still need to work on nLite a bit. The latencies cause worry.
By the way, do u have a Skype ID ??
Junaid
Junaid! I am so outdated in all things network, that I only know what skype is. No, I don't have one and no camera, sorry.
Serge.
As i mentioned earlier, i was facing issues with latency boosts when installing from my nLite disc. I had followed Rick's method of creating the nLite install disc. I guess he must be using a single core processor. But as we are using a MultiCore i3 processor, we need to retain the "Multi-Processor" support under 'Components'. I created my nLite install disc again with the above method, & my issue is resolved. If its removed from the disc, u will have standard install & i guess it creates latency issues with the MoBo...
Only if the Multi-Processor support is retained, the Computer-Type can be selected as "Acpi Multiprocessor Pc", which in our case is a must..!!
Attached is a snapshot of the same...
Junaid
so I need to try leaving that in.
I will say that the LUCY tool gave me the same basic latency as I had got before. I do not know how to do the measurement you are speaking of. Missed something (AGAIN).
That must explain why I do not have the acpi.sys in my installation!
There is a lot of difference in sonics if Multi-Processor is not selected. The Pc should be installed as Acpi Multiprocessor Pc. I just finished a new nLite install, & it sounds great out of the box. I'm sure there is a much larger scope for improvement with this one...
Rick, U had not selected Multiple processor support. Please make a note of it. It's very very important ..
Junaid
I think you might want to try installing with optimized BIOS and see if it makes the same difference.
If the MB only sees one core what's the point of multi processor support?
When cics was recommending the hardware set he said that it is enough to enable only one core, it will do all the multithreading (2 simulaneous threads).
Serge
I installed as an ACPI multiprocessor, then reverted bios to
CORES ENABLED 1
Serge.
P.S. there is more to acpi-multiprocessor, but let me finish the cmp. Stuck with awe.
Even if u do an automatic install, it will sense the Pc as a multiprocessor.
Though only one core is enabled. U can check with Ur previous back-ups as well, & all will show the installation as an Acpi Multiprocessor Pc.
That's exactly why I was insisting Rick to retain Multiple Processor support. Else u will only have a standard install which sounds a lot different that this one..
Junaid
One has a choice of ACPI Uniprocessor or the i486 C-Step that I chose since it has long been said that many times the older technology was better for our purposes.
And with the benefit of lower power requirements.
I can assure you the install is anything but bland. I am hearing much greater dynamics on drums and a surprising increase in low bass energy. Energy I had been missing.
I will try the other way but even then I will still install with only one core working and choose ACPI Uniprocessot.
As u please my friend :)
As far as u like it. That's what counts..!!
Junaid
I so knew that question would come up :)
I can answer that in one line. If the OS is seeing only ONE core in the bios, then HOW DOES IT INSTALL AS AN ACPI MULTIPROCESSOR PC, when the setting are in Automatic?????
As simple as that. Even I have disabled multithreading in my Bios long back, & only one core is enabled. But inspite of that, windows recognises it as a multiprocessor. I guess there is something more to it that just enabling or disabling...
The nLite disc which I have uploaded was made with Multiple processor support enabled and the Computer type I left it to Automatic.
Try installing with that disc & I'm quite sure that even ur's would be installed as an Acpi Multiprocessor, even though only One core is enabled...
Junaid
I have no idea where to look.
I am sure it is in the registry somewhere.
I do know that acpi.sys was not installed.
I will give it a try but I am mystified why the installation would override the BIOS when it doesn't for other things.
Try
http://falconfour.wordpress.com/2011/03/19/how-to-switch-windows-xp-hal-_back_-from-standard-pc-to-acpi-the-clean-way/
Where to look, it should be
HKLM\System\CurrentControlSet\Enum\Root\PCI_HAL\000
When you have non-ACPI, it should state 'e_isa_up'. Eventually you are lucky that way, but the site says, that afterwards you have to update the (fundamental) driver of it using MMC/ device manager.
Probably this is some regvalue too.. If you have the time, consider a re-install using ACPI from the start, I think honestly this would be a more clean approach as to not break things :)
Just go to Device Manager & open the first tree names " Computer". There u can see the type of installation been done.
Cross-check this in Ur nLite install with the previous images u had backed up, by restoring any non-nLite image u had earlier. I'm sure all Ur previous back-ups will have used the Acpi Multiprocessor Pc...
Junaid
Standard PC with C-Step i486.
Something to consider:
When I installed nLite my BIOS is set for the use of only one of the cores of the CPU.
As with the 1394 and the other unneeded ports left disabled at install so they don't install or at least are minimized, I wonder if the same thing occurs with this MULTI PROCESSOR thing?
If the installation only sees one core ...
I have not noticed any abberations in the sound that would point to latency problems and as I said LUCY would give me the same basic latency which my memory thinks was 32 as it was in every other install nLite or otherwise.
Could it be important to have this BIOS setting at installation?
Junaid! I was going just to start. can You upload once again?
Serge.
Sure...
Will start now..!! U will receive my email once its uploaded..
Junaid
Junaid! Thank You!
I have uploaded the ISO & emailed u the download link...
Junaid
I alraedy have it. Starting. Will report. Meanwhile, if You have patience and time, Junaid, can you, please, try to find in the Internet the way to set in the registry here:
HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Control\Session Manager\Memory Management
the following parameter EnableLargePages or smth like this. It has to do with awe and is easily done in windows 7 and Vista and windows 2003, it CAN be done in xp. but I don't exactly know the right syntax. IT IS VERY IMPORTANT, I assure You!
Serge.
Microsoft is very "slippery " and "slimy" about this thing.
We can do it just adding 1, But we need the way to establish a size in mb.
Is this it
NonPagedPoolSize REG_DWORD Number of bytes
Default: 0
Specifies the size of nonpaged pool in bytes. When this value is 0, the system uses the default size (based on physical memory). The maximum amount is about 80 percent of physical memory.
PagedPoolSize REG_DWORD 0 to 128 MB
Default: 0x3000000 (32 MB)
Specifies the size of paged pool in bytes. When this value is 0, the system uses the default size (32 MB). See also the entry for RegistrySizeLimit at the beginning of this section.
PagingFiles REG_MULTI_SZ System_Paging_Files
Default: C:\pagefile.sys 27
Specifies page file information set by choosing the System icon in Control Panel.
Optimizing Your Memory Configuration
The default memory-management settings for your system are optimal in most situations and do not need to be changed. However, the system can provide a much larger working set and additional virtual address space for the system cache if some settings are manually tuned. These settings include the following registry entries in the HKEY_LOCAL_MACHINE\System\CurrentControlSet\Control \Session Manager\Memory Management subkey:
LargeSystemCache . A value of 1 maximizes the size of the system cache's working set and, with PagedPoolSize set to 192000000 and SystemPages set to 0, provides a large virtual address space for the cache, provided that Terminal Services and the /3GB switch are not used. In addition, the maximum working set and large virtual address space for the file system cache are unavailable if Driver Verifier, a tool for testing kernel drivers, is enabled with the special pool. For information about Driver Verifier, see the Driver Development Kits link on the Web Resources page at http://windows.microsoft.com/windows2000/reskit/webresources .
PagedPoolSize . Changing this to 0xFFFFFFFF or leaving it at the default of 0 maximizes the size of the virtual address space used for the paged pool. To set aside extended virtual address space for the file system cache, change this value to 192000000, set LargeSystemCache to 1, and set SystemPages to 0.
SystemPages . Changing this to 0xFFFFFFFF provides an additional 464 MB of system page table entries for the virtual address space. Notice that making this change prevents expanding the system cache by 464 MB (limiting it to 512 MB) and limits the paged pool size to 192 MB, provided PagedPoolSize is not set to 0xFFFFFFFF to obtain the maximum virtual address space for the paged pool.
Notice that you have the option of changing the settings for LargeSystemCache in the user interface (using the Server Optimization tab in File and Printer Sharing for Microsoft Networks properties). However, you can only change the settings of SystemPages , PagedPoolSize , and NonPagedPoolSize using a registry editor.
Caution
Do not use a registry editor to edit the registry directly unless you have no alternative. The registry editors bypass the standard safeguards provided by administrative tools. These safeguards prevent you from entering conflicting settings or settings that are likely to degrade performance or damage your system. Editing the registry directly can have serious, unexpected consequences that can prevent the system from starting and require that you reinstall Windows 2000. To configure or customize Windows 2000, use the programs in Control Panel or Microsoft Management Console (MMC) whenever possible.
If you are using Terminal Services or your system is configured to start with the /3GB switch, these features consume the extra virtual address space that you might otherwise allocate to the system cache or the paged pool and that space is unavailable.
The above was taken from Microsoft site see link.
Here u go...http://technet.microsoft.com/en-us/library/cc938581.aspx
But be careful as well:-
http://tweakhound.com/xp/xptweaks/supertweaks11.htm
Junaid
Edits: 12/02/11
Junaid, wow...wow
I had already given full screenshots of my cMP. Maybe this is enough?
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2 Ru, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Liv
Given your screenshots, it would support Juli@ with Network correct?
It would be helpful if u could collect the list of files u had posted which should not be deleted for Network. It were the ones in the earlier steppes, which u had alerted people not to remove for network functionality.
It would be more appropriate if I could get a proper LIST of files from users of various peripherals & functions, so that the Bitch would look more adorable :)
Junaid
May be to add the cic's "Chapter 7.3 - Optimising Windows using Registry Editor (regedit)" into batch file?
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2 Ru, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Liv
It would be appropriate to get a LIST from EMU 1212m users as well.
Thanks
The list would keep going on & on.. It's best if users of various soundcards make a note of what files are being used by their card by looking in it's Properties under Device Manager & post it here. And another user of the same card confirms it. This way, it would leave no room for errors. I'm sure that even after being careful, the batch would initially end up with issues for some, & will need Updation thereafter. So I would need a sure-shot list of the critical files being used, either by the above method, or by Process explorer. I hope inmates read this post & start preparing the list for their requirement & post it ASAP. Once all information is collected & confirmed, I could start work...
Junaid
Thanks Junaid.
I can see that you are following my suggestion...
In your questions could you please include:
- Firewire for protocols (a few people employ Lynx Aurora, Weiss DAC202, Prism Orpheus; possibly Play-mate and/or Riboge can supply you with file names)
- Asus Xonar for sound cards (same as above, Audiodan can supply you with file names).
Thanks for the consideration of us variants. It may be a bit more complicated. There are items to be not deleted for WeissDAC202, for instance, drmk.sys, which is listed in driver details under device manager. But there is also one or more needed for its asio to be recognized by cPlay even beyond the proprietary driver components it takes to get the DAC connected via firewire. I have yet to narrow it down completely so far, having been a bit over-inclusive to make sure it worked. I'll work on it.
Then there are the firewire related items: I retained pretty much any dll, sys or exe with 1394 in it unless clearly specialized for something else, which did the trick, but again I have not made sure I got down to the absolute minimum of these yet.
I will try to come up with complete lists in future.
Then there is a set of items starting with syn_ which are for my trackpad. But my trackpad also requires portcls.dll, mouse.sys and maybe one other I don't recall right now.
Then there is a series of items for my Intel ethernet adapter beginning with epi or epy plus a couple others. I believe most others have Realtek and wonder if items needed for this are being included which I could exclude.
BTW, I find my keyboard keeps working even after removing all keyboard.* files and all kbd* files save kbdclass.dll in System32.
Thanks Riboge, my keyboard is working again! i think you meant to keep the kbdclass.sys file (you described kbdclass.dll)?
Douwe
Right you are! That's what I get for writing from memory.
can further help with his Aurora variations...if he is doing the tweaks.
Well, as I had mentioned in a post a few weeks ago, I have practically NO knowledge in writing Computer programs. A day before I posted my first Batch, I dint even know what it meant by a Batch file. Writing one was a far away chapter... I was way exhausted by manually deleting each dll & running into problems, only to re-do the whole thing again. It was at this point that I started reading about a quicker way of doing the same, & here I am.
I sure could include Ur lists as well, provided I'm given the details of the files to be retained for what u have requested. Its just that I can't make it AS interactive as u have put..!!
I intend to follow a simple procedure of making a COMMON LIST (which I call the Bitch Batch), which will clear all the files Except the ones needed for the requested Peripherals or Functions, & then make another set of Batches which will get rid of what u choose not to keep thereafter...
I guess this procedure is as simple as an Interactive program itself..
Junaid
Hello, Junaid. You seem to read my mind. I have posted something similar to that yesterday, but deleted as too brave to ask.
Serge.
Oh really? Thats cool :)
I have planned to make a Bitch Batch in such a way, that a Single batch file would delete ALL the files, leaving alone the ones needed for Juli@, Lynx, & Network. This will be a One-Click job. Then after this gets done, i will include 4 small Bitches as well, which will read..
1) I Dont use Juli@
2) I Dont use Lynx
3) I Dont use Network.
4) I Dont use USBAfter the One-Click Bitch Batch finishes, users can further use the 4 small Bitches mentioned above, as per their requirement, & get rid of what they dont require..
For eg. in my case, since i am a Non-Networked juli@ user, i would click on the 2) & 3) batches above & get rid of the files not necessary in my set-up. Likewise others using Lynx can choose the 1) & get rid of files needed by juli@, & so on...
This sure would be labour-intensive to make, but i guess its Worth it !!
Junaid
Edits: 12/01/11
Please, as I cannot currently hear the change, folks, make this tweak. It is safe, I have done it, the system works and reboots normally, if only slightly slower.
It should free the memory from crap, because most of these *.dlls cannot be removed and, sitting in this section of the registry, they are all read into memory on start up TOGETHER with all static dlls, that are associated with them.
Go:HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Control\Session Manager\KnownDLLs
in the window on the right remove all entries EXEPT DllDirectory
Will post the pic.
Serge
Edits: 11/30/11
At 1st I did shy away because of Mihaylov's warning but finally tried it. Sonically not sure I like it...it tends to add weight in the lower midrange and on some music it sounds good but on big classical music it can obscure details in that range. Also depth seems to be a bit foreshortened. Highs are still good. What is this supposed to do again ...not load those dll's at boot? I'll listen for a day/go back and listen again to prior level.
Went back to the prior level and yes what I said about the known-dll tweaks is true but in retrospect I do like some aspects of the tweak now that I'm out of it. Either way it is very good. May have to go back and listen again.
cool... will try it & post the feedback...
Junaid
No. This is not cool. This is too cool. About as cool as delete cPlay or cMP because these programs use these files in their work. ;)
Previously I had already given screenshots from Process Explorer for cPlay and cMP.
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2 Ru, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Liv
Edits: 12/01/11 12/01/11
Lol :)
Junaid
of the following tweak. I have done it on my working machine. No crashes. But I don't have a cmp and a soundcard here. Cmp will be rebuilt only by weekend, but the tweak is itching. It's about the hatefull watchdog.sys
http://msdn.microsoft.com/en-us/library/ff553890.aspx
Set BreakPointDelay value to 8640 (rightclick create the new REG_DWORD entry, then doubleclick it)
Set DisableBugCheck value to 1. Everything is working and rebooting.
Be careful.
Serge
2010-01-12 : subject line should read...Serge - but WHY set BreakPointDelay = 95.46 hours ?
Hi Serge
I've been slow to implement all recent tweaks, and last night the Watchdog was in my sights. The msdn page you reference specifies:
Set the value of BreakPointDelay to the watchdog time threshold, in 10-second units. For example, a value of 200 specifies a threshold of 2,000 seconds.
In the case of 0x00008640 (a number expressed in hexidecimal notation), the conversion and math works out to 34,368 x 10-second units, setting the BreakPointDelay to 95.46 hours .
Is there something special about choosing this length of time that I'm missing ?
Perhaps you mean to set the value to 0x000021C0 , which works out to 24 hours ?
In any event, thank you again for blazing the trail of new (...and obscure) adjustments for making cMP sound better and better... and better...
Cheers,
Grant
That's not a Toy... IT'S A TOOL !!
Edits: 01/12/12 01/12/12
Hi Steppe,
What do you think of the 'EaRecovery' registry value in the watchdog registry section ?
Does it also need too be changed from 1 to 0 ?
See: http://msdn.microsoft.com/en-us/library/ff553893.aspx
In Microsoft Windows XP SP1 and later operating systems, GDI uses a watchdog timer to monitor the time that threads spend executing in the display driver. The watchdog defines a time threshold. If a thread spends more time in a display driver than the threshold specifies, the watchdog tries to recover by switching to VGA graphics mode. If the attempt fails, the watchdog generates bug check 0xEA, THREAD_STUCK_IN_DEVICE_DRIVER.
Because timeout recovery code is complex, it might cause incompatibility with display drivers. To resolve the compatibility problems, timeout recovery can be disabled.
To disable timeout recovery, create the following REG_DWORD entry in the registry, and set its value to 0:
HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Control\Watchdog\Display\
EaRecovery
Mark
fully optimized cMP2 PC -> ESI Juli@ -> Van den Hul Optocoupler MkII-> Lavry Black DA10 -> XLR Mogami Gold -> Klein & Hummel O300
Hello, thank You for this info. Probably worth doing. Have You done it? What's the perceptible difference? Do You Like or Dislike the Watchdog tweak in general? Feedback is necessary. Even negative one.
Serge.
Hi Serge,In my system the watchdog registry tweak is the best software tweak after using ASIO or WASAPI.
To give some kind of quantification/impression: it improves sound quality at some 30% – 50% level compared too the positive impact on sq ASIO or WASAPI has. So a real significant one in my system which doesn’t need extensive A <-> B comparisons.Most software tweaks need extensive A <-> B comparisons too indentify their mostly minor and/or very subtle SQ changes/improvements. However, the SQ change of the watchdog registry tweak is very significant. It’s positive impact on SQ can be heard easily and immediately.
It ads to my system a touch of ‘that expensive sound’ one gets when using the top-level range AD-converters from makes like Prism, Lavry, Weiss, Metric Halo, ectSo all in all the second best software tweak for windows XP I’ve come across.
IMHO a must do tweak which should be listed at the cicsmemoryplayer website.Mark
fully optimized cMP2 PC -> ESI Juli@ -> Van den Hul Optocoupler MkII-> Lavry Black DA10 -> XLR Mogami Gold -> Klein & Hummel O300
Edits: 12/05/11
Are you specifically talking about your suggested registry tweak or steppes or both of yours combined?
nt
fully optimized cMP2 PC -> ESI Juli@ -> Van den Hul Optocoupler MkII-> Lavry Black DA10 -> XLR Mogami Gold -> Klein & Hummel O300
Hi Serge,
You struck some gold here!
The SQ improvement is real significant in my setup.
Theo already described the nature of the SQ improvement.
I get the same type of SQ improvement in my setup.
Thank you !
NB Your picture shows REG_SZ values instead of REG_DWORD values
fully optimized cMP2 PC -> ESI Juli@ -> Van den Hul Optocoupler MkII-> Lavry Black DA10 -> XLR Mogami Gold -> Klein & Hummel O300
Mark nice to see you post. I have to share an experience. The other day I took out one of my hdd's to rip some new music (cant do it anymore via thumbdrive after 1-22 steppes) and while ripping (with hdd out)I listened to some cmp^2 music. All of a sudden the same music I just listened to with all my hdds hooked up sounded way better. In an effort to understand this (and capture it) I concluded that the hdd I took out of the pc was hooked up to an external ps with no/minimal ferrites/caps. So when I replaced the hdd I plugged it into my fully capped/ferrite infested internal ps and sure enough the lower mid/upper bass grunge was way better. I tried a retired granite digital on the hdd...not good. I soldered some silmic caps into the granite ... a lot better but not as good as my baseline Antec fully capped and ferrite infested. Once again ferrites/caps on the powerline shows its excellence. Too bad others have not tried this very effective tweaks.
Hi Theo,It is my experience too, that filtering and/or cleaning the power too a cMP and ALL (!) it’s auxiliaries will yield (much) better sound quality.
Not only filtering the power of a switching PSU’s will give better SQ in a cMP setup, but also filtering the output of standard of the shelve linear bench PSU’s, will result in better SQ when using a cMP setup.Even filtering the power too my tinny 16Gb Kingstone SSD gives a little sound quality improvement.
I filter the power too this little 16 Gb Kingstone SSD with a coupled inductor filter (as I showed in earlier posts) and a ferrite coil with 7 windings behind this coupled inductor filter.
Although the SQ improvement is only very little, I find it worth the trouble.I think (!) it was TeddyPardo who stressed the importance of filtering VHF and UHF frequencies in power supplies for analog audio systems. Although these frequencies are not even close too the audible analog frequencies, there negative effects can be clearly heard (according too TeddyPardo. If I remember correctly)
I guess in digital audio filtering VHF and UHF is even more important, because all kind of logic and clocks suffer performance we need for audio quality, when they receive DC power, which is contaminated with VHF and UHF.Unfortunately, ferrites + caps filters can’t filter much lower than 1 mHz. I’m still looking for ways too effectively filter 80 Khz (ATX switching noise) or 300 kHz (pico switching noise).
I looked at using gyrators for this. But understanding/using/implementing gyrator based power line filters, is very complicated for me. As I lack knowledge (not being an EE).Although ‘(much) ferrites + caps’ or ‘ferrite based chokes + caps’, can only effectively filter from roughly 1 mHz and up, there is still a profound positive effect on sound quality when filtering the power too a cMP setup. (also when applied on linear PSU’s !!) . It is indeed a pity that other inmates make so little use of them or experiment so little with them.
Mark
Edit: Instead of TeddyPardo it also could have been Andy Weekes.
I can't remember. But it was one of those PSU guru's. :-)
fully optimized cMP2 PC -> ESI Juli@ -> Van den Hul Optocoupler MkII-> Lavry Black DA10 -> XLR Mogami Gold -> Klein & Hummel O300
Edits: 12/01/11
Hi Mark, i tried the ferrites a while ago, sorry i didn't report. Thanks for your suggestions.
I tried the ferrites on p24 (the flat ones), p4, 12v-line-to-pico, hdd/ssd, and of course on the 220v powercords. When using too much ferrites, i got a slight loss of dynamics. Especially the flat ones on the p24 caused a lot of loss. iirc this is a known issue of ferrites. Now i just have a few small ones on the p4, pico, ssd, monitor-dvi-cable and powercords.
I stopped experimenting when i read about your experiments with the filters. i ordered them, but didn't have time yet to implement (and when i have don it, i regret i didn't do it before; it's an ongoing process...).
Besides that i tried sikorel and Elna-SilmicII caps on the p4 (pico already capped). Didn't work; sounded slightly better without caps.
BTW i can recommend bypassing all caps on mobo like JackWong did: sq boosted!
I hope to build a new cmp during christmas holidays, atom-based, headless. Lots of questions to think about (XPslimming, battery-psu for memory), especially about networking on which i'm a noob.
I'll report on the filters, but that may take some time.
Douwe
Hi Douwe,
Thank you for your feedback.
When reading your reply, first thing that sprang too mind was: ‘it is very strange that Douwe’s computer needs UHF polluted DC too sound good’.
Ferrites can effectively provide high resistance (‘filter’) for UHF from approx 80 mHz too 300 Mhz (when applied correctly and depending on type off ferrite material used).
Have you any idea why this UHF pollution on the DC power supply lines, makes your computer sound better ?
“iirc this is a known issue of ferrites”
Are you really sure about that? I also read this at regular intervals. But I feel this is a misconception from the analogue audio domain which now (unfortunately) also is projected onto the digital audio domain.
Although I did not do/tried it myself, I can easily understand in terms of extra ‘power supply noise suppression’ and extra ‘decoupling’ how bypassing all caps on the mobo will boost SQ.
But I lack the ‘micro’ soldering skills too do this.
I also do believe that powering memory separately will boost SQ.
But is this because memory needs cleaner power which somehow results in better sound quality?
Or is it because in the process Jack is also removing a noise source?
In this case I find it hard too differentiate between ‘cause’, and ‘effect’ (and possibly also ‘symptoms’).
Why do you choose for an atom based setup?
I haven’ compared atom based mobo’s with the latest 1155 chipset mobo’s myself.
But I feel it is definitely worth to try Jackwongs suggestion using a 1155 based mobo with a 1155 socket Celeron dual core. Jackwong reported that this computer setup sounded best in his system.
I think (not sure yet) I will try it myself in a few weeks.
Half a year ago I also changed too using my Gigabyte G41M based cMP setup headless.
This allowed me too get rid of using USB.
While on the same time giving a lot of convenience controlling the cMP setup through my android tablet and smartphone (through using VNC)
Using the VGA output too my Sony KDL-37/W4000 TV has absolutely no impact on SQ.
Also using LAN has absolutely no impact on SQ.
But using USB does have a negative impact on SQ.
This can easily be heard in my setup.
Reducing the polling USB frequency (as found by Ryelands) does give some SQ improvement.
But disabling USB all together in BIOS and XP gives an even greater SQ improvement.
And going headless through using an android tablet is sooo convenient !!
I would not put much effort in XP slimming.
I see no ration behind it.
I consider it as shooting in the dark, hoping you hit something.
And surely you hit something because the results are stunning as reported by other inmates with good equipment and ears.
But one also does totally cripple the cMP machine.
There has too be a more clever way.
I feel a much better approach could be:
1. check with process explorer which files and processes are being used while running cmp + cPlay.
2. Than compare this list it with the list of files Mihaylov’s is running.
Mihaylov’s file list see: http://www.audioasylum.com/forums/pcaudio/messages/9/99505.html
3. Subtract: 1. (process explorer file list) minus 2. (Mihaylov’s file list) = file deletion list.
I will try this somewhere coming weeks.
If you ever manage to find out why your setup needs UHF polluted DC too sound good, let us know.
It’s very intriguing, why it does need it.
Mark
fully optimized cMP2 PC -> ESI Juli@ -> Van den Hul Optocoupler MkII-> Lavry Black DA10 -> XLR Mogami Gold -> Klein & Hummel O300
Hi Mark, thanks for your post
On the ferrites
I don't know why too much ferrite chokes dynamics, i'm not audio engineer. So i just try and listen, keeping what works.
As you know i use sla-batteries and inverters (to 220v). For the analogue parts i use sinus-inverters; for the digital parts (cmp-machine) block-inverters sound much better! Less filtering, more open sound. Maybe this is an explanation, i really don't know.
I'm very curious about your UHF-shunt: what did it bring you and where did you put it?
On the Atom
- cpu frequency matters. In my former and current machines (Asus-socket A and Intel-1156) cpu-frequency makes a lot of difference. This particular atom-board (asrock ad525pv3) can go as low as 900MHz. Besides that has a very detailed bios (unlike most atoms) and tweakable memory-settings.
- This atom board has no unneeded chips and unneeded functions
- i don'n need 24/192; 16/88 is ok. atom can do this.
- the board is very simple (elegant?). i think i can tweak cpu-psu (lowdrops in stead of 78xx). Besides that: only one powerphase, so no/less switching (i suppose) between phases.
- Ryelands has very good experience with atom
- low power, so easier to battery-psu
- last but not least: learning is as important as sq to me. i just want to try and hear. If it is not good, i'll change to 1155. As jackwong is working on GA-H55M-USB3 (my current board) we have some same reference.
I'll report the results when ready.
I'm looking forward to going headless CMP, my current dualboot setup alows me to go headless in Windows 7. Very convenient!!
What about safety: isn't it very vulnerable when connected to the www? or do you have a closed network with dedicated nas and router? (i'm sorry, networking is not my strongest field)
you should try capping the mobo, just try out on an old mobo. I'm not very experienced, it's not that difficult. Just take time and use tape to hold the caps on the mobo before (and while) soldering. This is another argument for atom: only 20-25 caps to bypass.
You are right about usb: i never used it because of negative impact on sq.
about slimming XP: lets just see what it brings. For me its ok if it gets smaller than 1GB: i ordered a 1gb SATA SSD DOM (disk on module). Bot wow, the 36 mb of Rickmcinnis is very intriguing. I like non-rational solutions, but your approach is very interesting. Please report the results!
Oh, about the UHF: i think my setup and psu is so clean that, in order to sound natural, it just begs for a little pollution ;))
Douwe
Douwe
Thankx for your response. I enjoy the communication.
I’m also in computeraudio for learning combined with hobby.
I'm not sure what you mean with: UHF-shunt.
I just use ferrite clamp-on beads (combined with caps to create a 2nd order filter). I don't think that is the same as 'shunting'
Here I lack the knowledge of an EE or any other indebt knowledge of electronics too understand what you extacly mean.
I’m agricultural engineer. (I now can hear you laughing, but in the Netherlands this is a real degree. :-) )
“ Oh, about the UHF: i think my setup and psu is so clean that, in order to sound natural, it just begs for a little pollution ;)) “
Well, no kidding, I already thought of some kind of ‘dither’ effect.
About your atom mobo choice.
I now understand your choice for it.
I also don’t care for re- or upsampling or hig res.
I even don’t play FLAC files.
I convert FLAC files too WAV-files
I use a 2 Tb NAS, so why bother about storage space.
It is also my experience that running at low speeds indeed has some positive effect on SQ
But…. Just for that, you can still go with a Gigabyte MoBo
I run mine at 600 mhz. No kidding. See picture.
So on a GA-H55M-USB3 you should be able too go as low.
Although I have never used one, I agree the Atom boards appear to be nice mobo’s for audio PC or even HTPC’s
About going headless.
I guess I have a standard home LAN setup
WWW <-> modem/router <-> LAN 1 Gbps Swith <-> cMP PC
The routing from my android Tablet too my headless cMP is:
Android Tablet <-> LAN WiFi Acces Point <-> 1 Gbps Swith <-> cMP PC
There is also a firewall in my modem/router (as in any modem/router nowadays)
So I don’t care about security issues and safety on my cMP computer.
On my cMP is not any information that has any value too others.
On my small Kingstone 16 Gb SSD there is only an XP instalment optimized for music.
So I really don’t care about safety issues on my cMP
So if you have a WiFi Acces Point on your home LAN,
Just don’t care about safety and do try an android Tablet or smartphone as a wireless remote.
Mark
fully optimized cMP2 PC -> ESI Juli@ -> Van den Hul Optocoupler MkII-> Lavry Black DA10 -> XLR Mogami Gold -> Klein & Hummel O300
Hi Mark, thanks for reassuring about LAN-safety, i enjoy the conversation too!
I found a picture of your UHF-shunt (or maybe it was HF-shunt) on your picassa-page, which i cannot find now. You didn't post on it, only the picture about shunting HF to earth or so. And of course, as an agricultural engineer, you know everything about earth and grounding and stuff, so i wondered...
600MHz is great! H55M-bios cannot go below 900 MHz; i chose it for the quiet cpu-psu (ic's in stead of rattling 78xx). Downside is it's crowded with unneeded chips: dvi, hdmi, usb-3, ide, even Floppy-slot. Wouldnot recommend it.
I will let you know when i finished atom-set, maybe you can have a listen when you're in amsterdam.
cheers!
Douwe
An agricultural engineer is never laughable. I myself am a trained mechanical engineer trying to become an electrical/computer technician. At any rate you are right learning new stuff is half the fun. When I was working I tended to spend way too much on audio. Now that I am retired I see that very little expenditure (in most cases no expenditure) which can cause way bigger sq changes.
This is with Watchdog.sys deleted as well or Retained? I have already deleted Watchdog.sys from my System32. Would it still be effective?
Junaid
Junaid! You have managed to delete a watchdog.sys with the system working?
I couldn't. It would always go to a black screen on startup and immediate reboot. It must have been with Mihaylov's tweak, but watchdog controls video process on booting. Does it mean that you also managed to delete bootvid.dll?
Serge
I'm so sorry. It was a mistake. I just came home & checked. Watchdog.sys is not deleted. I thought I did. Sorry for the confusion..
Junaid
Junaid! There were so many files, no wonder, you could have forgotten.
I am just thinking about bitch bat. This beast is something to be really very carefull with. What if some more registry tweaks would come up? What if somebody would like to reinstall drivers and so on. The beast should be created, but please, don't be in a hurry and give some time for testing. Let others post their propositions of the logical order of this routine.
I will send you my Lynx files on Sunday, if I may.
Serge
Sure I will. All that has to be done, is to send me the names of files which are present in steppes 1 to 22, Jolida 1& 2, and Mihaylov 1,2 & 3, WITHOUT whom either Juli@, Lynx, Asus & Network would not work..
Junaid
Hi Jolida,
I haven’t deleted the watchdog.sys file
(personally I see no point in (more or less) randomly deleting files in an attempt too improve SQ. I like a more rational approach)
Reading your question on effectiveness, first thing that springs too mind is: your system should let you hear if this tweak has any effect or has no effect.
In my setup the tweak had a very noticeable and profound positive effect on SQ.
The effect on SQ can easily be heard.
So my guess would be: if you can’t hear the effect on SQ, than there is no effect.
Mark
fully optimized cMP2 PC -> ESI Juli@ -> Van den Hul Optocoupler MkII-> Lavry Black DA10 -> XLR Mogami Gold -> Klein & Hummel O300
I haven't tried it yet. Will probably try this evening. My only concern was if watchdog.sys was deleted or not..!!
Junaid
It is necessary to create REG_DWORD entries! The right screenshot:
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2 Ru, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Liv
Edits: 11/30/11
With Ryelands help I was able to finally do this and the sound is fantastic. Yes noise is lower but its more than that: very lively with inner detail revealed..not hard...spacious...and even better bass than prior level. Serge how do you find these gems! Thank you.After listening a bit the soundstage sounds very open and big: wider and deeper. Its almost like I got a bigger power supply somewhere in the chain. Well done again.
Edits: 11/30/11
Ok I went toHKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Control\Watchdog\Display\BreakPointDelay
as per the link and all I have is
HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Control\Watchdog\Display
with nothing that looks like BreakpointDelay within Dispay. So what am I doing wrong here? Do I click on any of the stuff in the right hand pane to get BreakPointDelay to show?
Edits: 11/30/11
I'm in marathon ripping today. Its a brand new world post steppe 22 et al because I have to pull the hdd, put it in another computer via the docking station (what a brilliant tool this has been...thanks ryelands)and so on...but I will try later today. This really sounds promising.
While I'm ripping I'm listening to my other hdd's on the first day with a good clean poweline and I want to tell you the Lost and Last Reg tweaks are really outstanding...dynamics, space, beautiful highs and tight bass...what more could an audiophile want!! Thank you Steppe!
Cant wait for the watchdog--catcher.
BTW for those that still think pc audio is behind traditional cd playback ought to hearing what the mermaid is singing today.
Please, check the sound change
I've tested the tweak by listening to familiar recordings first with the tweak made then with it removed then with it replaced. The change is very noticeable and, in my setup at least, definitely for the better.
It's too early to say that it works without fail on every system and all that but it's certainly sounding good here. My wife, who had no idea what I was changing and cares even less, confirms the result. That's no proof of course but it's encouraging.
Clever, these mermaids.
Thanks.
Dave
Wondered if you had tried.
Hi! Most probably, it won't. I haven't tried, as I can't allow myself to crash the work machine. But at least this change stops the barking and howling for 24 hours.
Serge.
I created the keys and set the values. No problems on startup and plays with seeming lower noise floor as very,very quiet bowing in Vasks quartet at the beginning of 4th movt of String Quartet 3, which is marvelous by the way, is now audible clearly. I have to listen more and at a different time as well to be sure I am not talking myself into this. Surely no harm done so I will let the sleeping dog lie in any case.
I came across http://www.pcregedit.com/ which offers a free registry editor that works on registry files in config folder from outside via startup on cd made from .iso provided. By far the easiest way to prune registry at late stages of our cMP system trimming without having to add back .dlls or export hives.
one wonders if this is really any different than using another computer to load hives from your MUSIC SSD?
If one does not have access to another computer it is the only way.
No question, a great find.
If one could use this same approach to use a registry cleaner now that would be really something.
Yes, not really different in effect, but you can work on all the hives at one startup. Also if you have, as I do, multiple partitions with different stages of cMP stripping on the same ssd then you can work on the registries of each at one time.
I agree that being able to clean and optimize from this platform would be ideal. I have been thinking of calling the maker PC Disk Tools to see what they may know is possible along those lines.
In anticipation of using Junaid's recommended registry cleaner I took a look at the registry (using Junaid's method of loading hives into my "other" computer) and started looking through the entries for anything that had to do with files we have deleted.
Somehow the registry is much simpler this way. I think the registry we see on an active installation has many of the entries duplicated. The whole thing looks much simpler when loading hives into another computer.
I found another place where the same things Serge told us to delete the audio junk (#2 on his last registry post). Many easy to see places that are concerned with .dll's and .sys's that are longer there.
Unfortunately, I deleted things having to do with swenum.sys before I realized I should not have so I did not bother seeing if it would work since I feel confident it won't and it was too late and cold to see just for the fun of it.
Tonight I will try a new install and see if Junaid's recommended cleaner will work and THEN to see how long it will work as files are removed.
Then to go through the registry and see what is left.
If the registry cleaner will not work I am feeling good about doing it manually. Yes, it will take much more time but it is not daunting.
Lots of USB stuff to remove, services, etc.; basically, as Grant has long held that the cMP optimizations can be best done within the registry. WHo knows what all of those alpha-numeric things do - but if they have a sub-heading it is explained or there is more alpha-numeric which I would leave alone.
When u start doing it, follow the instructions in my post. Do it in xp mode, & start Registry Mechanic & minimise it before u run the Batches. The reason I insist on this, is because if u close the program, the files needed for it to work may be deleted in those batches & it would no longer work. By keeping the program open & minimised, the batch would not delete the file, as it would be in use. After running all the batches & arriving at the final stage, restart in cMp mode & delete the few files that were in use by Registry mechanic, & Ur done. I personally feel that the registry cleaning is AS important as deleting the files themselves. Close registry mechanic & restart in cMp mode ONLY when u feel u have deleted all the required files, as this would be Ur only chance to do automated registry cleaning....
Junaid
Thanks for the good advice.
I was tempted to do it within cMP but that is all the reason one needs to be able to keep two windows open. XP it is!
doesn't allow to do anything with it.
I have just downloaded from microsoft website the file:
WindowsXP-KB835935-SP2-ENU.exe. Started nLite and told it to use this file as sp source. It simply unpacked it into a temp folder, congratulated me and went on to components removal. So, no sport here.
Serge
Hello, everybody. I have burned a cd using nLite but it doesn't contain ntldr file and autorun.inf and setup.exe. Installation won't start.
First time I used an XP sp2 pro corporate vl version that doesn't need registration - nlite crashed before finishing its work without a serial number. Next I INSERTED serial number from my own disk. It went smooth, burned a 86 mb installation disk without initial installation files... Strange. What could have been done wrong, folks?
Serge.
.
That's not a Toy... IT'S A TOOL !!
It was designed to be used with the standard XP disk.
The "corporate" disk is different, I would assume, and there was a conflict.
Easy enough to use your standard disk.
I made a folder that contains the XP files from the CD which I then copy to another folder which I use to make the nLite disk.
You can always use that same file to make a smaller nLite installation but not the other way. If something was left out I would re-copy the XP CD files from the folder to the other folder and start again. nLite will remember your settings for desktop and start menu which can be a pain to select after the third time!
Were you able to install the LYNX?
I was not able to eliminate SP2 from my install disks. Have you tried?
Hello! I didn't try to preinstall lynx, its files are on a different (cmp) computer and I did this trial just for experiment. I had only 1 blank cdr disk. I will try today, but, I think, I have to download a separate spak at first.
Serge
Hi Serge,
I have emailed u an ISO image of the nLite Xp installation. Just burn it using any program & u r good to go...
Junaid
Oh, THANKS A LOOOOOOOT!!!
U r most welcome. The file is on it's way. Uploading. U will receive it in less than 20 mins.
I made my first nLite install last night. Took less than 10 mins & then made the cics optimisations. Overall it was done in about an hour. The sonics at this stage itself are way superior to anything I had heard before. I wanted to listen for a couple of hours before starting with the Batch deletions. The sound is way civilised, & controlled. The noise floor has dropped a real lot. The most evident aspect was the front-to-back imaging. Very well defined. The tone is clean with no peaks or sibilances. The only thing I felt that was missing was a little Salt in the dish. I'm sure that will settle in after the deletions...
Junaid
with impatience. Will be reassembling-reinstalling-reeverything these days.
Serge.
So Serge and Jolida,
You are no longer using XPlite in your file removal procedures?
KC
Hello! I really do not know yet. I will be using nLite-prepared installation for the first time. I have seen and tried to remove as many things as I thought possible or sensible, but I suspect that some components may still remain. But I cannot say FOR SURE, I haven't tried it yet. All I know is that file deletions are still to be done, and here the bat files from Jolida are most welcome, just as his reg files are. He even prepared a reg file for the lost&last registry tweaks. Bless him! But bat files cover only my 1-16 tweaks while the tweaks go to part 22 and theye cover almost all of Mihaylov's tweaks as well. Almost nobody seemed to follow or reported to have come only my way. And it IS GOOD, because different approaches give a variety of results, we all can choose from by trial and error. Take Rick and Grant with their registry slimming down and an idea (simple and beautiful) to use ps/2 mouse. It is these things that made me redo the cmp again. I will certainly report. But the smaller the registry-the better is (imho) beyond doubt, even looking from the viewpoint of reduction of threads, severed by file deletions and memory management by windows that we can affect only partially. I am experimenting a lot now with memory timings and boot.ini memory settings now. Some results are very interesting. One was splendid but unstable. Will soon post about the nLitening and my small "finds" that are no secret, certainly, to more advanced computer folks.
Serge.
Edits: 11/30/11
there is a conflict. It will seem to sort of work but it makes a mess.
Give it a try since it is easy to start over but I know you will not like what happens.
If you are using the minimal nLite install I doubt it will work at all. I never tried it again after my first experience with it months ago.
To future minimalists: don't waste your money!
Hi to all!
It is a pity to me of those people who cannot start those or other programs on tweaked cMP2 (for example, regedit or programs for registry). I want to help them. Rebuild the classical version cMP2 on your cMP2 (when all programs still work). Start Mark Russinovich's Process Explorer and, for example, regedit. Look at what files are used by regedit. Add those from them which is not present in tweaked cMP2 to run regedit on tweaked cMP2. That's all.
For work with the registry I use the portable version of jv16.PowerTools starting it on tweaked cMP2 from NAS.
See also for details here http://www.audioasylum.com/forums/pcaudio/messages/9/98066.html and here http://www.audioasylum.com/forums/pcaudio/messages/9/99013.html.
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2 Ru, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Liv
I want to try this on my home pc (which is xp also) do you see any reason it cannot work?
Thanks. It would be of help to all...!!!
Junaid
Multiple ways, numerous approaches, Subjective results !!! A lot has been happening even after the deletion process. As far as the Registry slimming process is concerned, there are a few approaches as well, but i resorted using Registry Mechanic for the same. Its a software i have been using since 3 years now, & it works flawlessly well. The method of using it for our purpose here is quite different though !!!
As mentioned above, there are numerous approaches to this as well, but i would outline the procedure that suited the best to me...Starting with the basic Cics optimized cMp :
1. Install Registry Mechanic (Full version)
2. While in XP MODE, run the Registry Batches first ( all of them)
3. Run Registry Mechanic & then minimise it to the system tray.
4. Start running each Batch file for every step followed by Scanning the Registry. For eg. Run the Batch file for Steppe 2 > maximise Registry Mechanic > Scan the Registry > Select Repair > Minimise Registry Mechanic again to the system tray...
5. Do the above for all the Batch files, but be sure NOT to close Registry Mechanic, just minimise it.
By following the above procedure, i was able to clean the Registry even after completing all the steppes & Mihaylov's lists. (Juli@ users need to kill the Julapan.exe using the taskmanager BEFORE starting the deletions, else the file will reside in the system tray & will not be deleted while running the batch for Mihaylov's list..)
It cleaned nearly 2200 + errors from the Registry. I could never imagine that the sonic impact of doing this would be of so much importance...
I have recently re-built my cMp for using ps2 mouse, & eliminating all usb drivers. Absolutely Worth it !! I am going to modify the ps2 mouse to use an external power supply so that it does not use the power off the MoBo, using a Granite Digital psu & feeding the 5v to the pin 3 & 4 of the ps2 mouse..
Junaid
Edits: 12/20/11
Great Junaid!!
I can confirm that ps/2 is better for sq, i never used usb for this reason.
Douwe
Many good ideas there, Junaid!
Thanks,
=I have recently re-built my cMp for using ps2 mouse, & eliminating all usb drivers. Absolutely Worth it !! I am going to modify the ps2 mouse to use an external power supply so that it does not use the power off the MoBo, using a Granite Digital psu & feeding the 5v to the pin 3 & 4 of the ps2 mouse..=
Great Junaid!
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2 Ru, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Liv
:)
Junaid
Hello, everybody! My mermaid gave me something after a long and hard time together. It was a shiny scale, and on its avers was written:
1. Open your boot.ini and add the following: /CLKLVL
Put it somewhere in the middle of the string (before /nopae and after fastdetect..., e.g.) This will stop processor from answering to IRQ requests (constant) from those onboard devices that are designed by old specifications. There is a number of such. Your soundcard IRQ won't be affected.
2. On the revers of the shiny scale I found this advice:
get into bios-M.I.T intelligent tweaker, open ADVANCED VOLTAGE SETTINGS
and set LOAD LINE CALIBRATION to ENABLED (allows to control vdroop [probably voltage DROP]}directly). It will more or less make cpu work at constant voltage without being reduced or increased.
The effect of both tweaks is immediate and grows within a 1-2 hours burn-in.
Serge.
Edits: 11/26/11
Sorry, folks. The boot.ini setting IS RISKY. It can crash your machine and reset settings in BIOS. Experienced by Theob and myself.
Serge.
The boot.ini setting IS RISKY.
It works fine on the Atom :> )
That last set of reg changes c/w deleting logonui.exe has made as noticeable a difference as any other "XP Slimming" exercise on my systems. Thanks for the research!
BTW, anyone using a headless setup should also be able to delete the Pointer port, Keyboard port, Pointer class, Keyboard class and Parallel Arbiter (spelling?) keys.
Can't say if doing so makes any difference but it makes me feel good.
Dave
=It works fine on the Atom :> ) =
Of course. Atom is not downclocked. ;).
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2 Ru, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Liv
Atom is not downclocked.
You're right. It runs properly right out of the box. I said only that it worked, I didn't say it did any good.
Seriously, I don't see what downclocking has to do with it. As MS notes,/CLKLVL Causes the standard x86 multiprocessor HAL (Halmps.dll) to configure itself for a level-sensitive system clock rather then an edge-triggered clock. Level-sensitive and edge-triggered are terms used to describe hardware interrupt types.and seems relevant to IRQs from legacy (edge-triggered) devices. I'd be a bit surprised to find one on a modern mobo but I could well be wrong. Can anyone explain?
And I said only that it worked due to it's stable mode because it is not downclocked. Obviously the option /CLKLVL leads to some instable mode with downclocking settings. So BIOS settings reset into default at start up as at overclocking.
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2 Ru, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Liv
Ok I am not able to do the load calibration revision (not available on my mobo) I did only the /CLKLVL in the boot file. And while it changed the sonics I'm still not sure I like it. Whenever I shut down my cmp pc and then eventually reboot it reverts bios to 1600 mhz and slower memory timings. So when I rebooted after /CLKLVL it sounded great. When I eventually got back to my 'listening bios levels' which is 840 mhz and faster memory settings it sounded a bit underpowered...soundstage moved closer, bass got sloppier and highs were more midrangey. It sounded a bit voluptuous if I had to give it one word. I will give it a bit more time but if I stay with /CLKLVL I'll probably have to bump up cpu speed. I'm reluctant to go to 1600 mhz always for the risk it may be to cpu heat.
Hello, Theo! I just wanted to post about these dangers. It went OK for 18 hours then crashed. I didn't know it was because of /CLKLVL. From the start it sounded very clean, so i posted it immediately.
Serge.
1.=This will stop processor from answering to IRQ requests (constant) from those onboard devices that are designed by old specifications. =
Why you think so?
I have found this:
"/CLKLVL Causes the standard x86 multiprocessor HAL (Halmps.dll) to configure itself for a level-sensitive system clock rather then an edge-triggered clock. Level-sensitive and edge-triggered are terms used to describe hardware interrupt types." CLKLVL = Clock Level.
2. LOAD LINE CALIBRATION is the feature to automatically maintain constant Vcore in overclocking systems when current consumption of CPU is very high. The current consumption of CPU in cMP2 is relatively low so all automatic features should be disabled because they generate pollutions.
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2 Ru, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Liv
Almost twice the Vcore voltage?
DSIABLED when you write ENABLED.
I assume the photo has nothing to do with the trick other than to show where the trick is performed BUT who, following you, would not know and hence, my confusion!!!!
where to make this change. The pic is from manual pdf. It shows the default values. Sorry, if it is confusing
Serge.
Hello, everybody! This is the lost part of registry tweaks that are best performed during part 13-15, but you can do it anytime BEFORE part 17 and onwards.
1. Go HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Hardware Profiles\0001\System\CurrentControlSet\Enum\PCI\(Your soundcard name may be different)VEN_1621&DEV_0023&SUBSYS_00231621&REV_00\4&6A4EAF8&0&08F0
open this key and find there your sound card, delete DirectSound.
Perform this step in all sets.
2. Go HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Enum\SW and delete every key inside. You will find there our mortal friends.
Note, that the 7th key from above (sysaudio) or some other (the only one) won't let itself be deleted, so enter it and delete inside, Perform in all sets.
3. Go HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Control\GroupOrderList and delete:
Antivirus, Encryption, Content Screener, Copy Protection, Physical Quota Management, Quota Management, Security Enhancer,
System Recovery, NetBios Group, Network, Spooler Group, Streams Drivers, Activity Monitor, Compression, Continuous Backup,
Replication, SCSI CDROM class, SCSI Miniport, Content Screener.
Repeat in all sets.
4. Almost in the same place:
Go HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Control\ServiceGroupOrder
As You open this key, in the window to the right you will see only 2 lines. Doubleclick List,
The dialog that opens will show you almost the same services and groups as in GroupOrderList (see above). Almost the same, but there
is a slight difference. So, Delete:
Antivirus, Encryption, Copy Protection, Security Enhancer, Physical Quota Management, Quota Management, Content Screener, Streams Drivers,
NetBios Group, Spooler Group, Network Provider, Audio Group, SCSI Miniport, SCSI CDROM Class, Compression, Continuous Backup, Replication,
Activity Monitor, Event Log, Local Validation, Scheduler Group, Smart Card Group, Remote Validation.
Perform in all sets.
Serge.
P.S. I had written that tweaks part 15 would be very good, however, nobody confirmed. I think it is because I have forgotten about these registry tweaks in the havoc I had been in.
I AM sorry, folks.
P.P.S. Don't know if these are still relevant after nLite registry swapping.
I was able to do 2 thru 4 using Mihalov's method, having done all steppes and M's list before. I deleted all keys under SW without resistance though each deletion asked for permission. I don't use a soundcard so step 1 is irrelevant for me.
I had also attempted registry cleaning by using Advanced Registry Cleaner on a partition at an earlier step after applying later registry changes only. I then substitured the config files System and Software from this system into the fully savaged system. Then I did these final registry steps. I have not done the xplite thing, however.
Hello! Inmates! No permission was asked for or given. I made these changes during part 13-15 of the tweaks and AFTER the services and legacy services (around tweaks part7) were deleted from the registry. No permission was demanded. The xplite was implemented to its utmost after part 3, then minlogon (in my current setup I didn't need to make a minlogon acrobatics). When making standard cics' tweaks I DISABLED practically all the services, not just stopped them and disabled IP security from administrative tools-local security settings. This weekend I am going to rebuild everything from scratch (to use ps/2 mouse) and will go this way again but with Junaid's batch files. Will report, if anything goes differently. Also, a nontechnical comment: when you start from scratch, make all the cics changes + disable ip security and proceed with my tweaks. make xplite, awe, minlogon. THEN... LISTEN. If it sounds really good after my tweaks part 3-4, it WILL sound good. If it doesn't, you better start again. You may repeat the same steps, but the sonic result will be different. I can't explain why.
Serge.
I think you should try an nLite install(unless you already did and did not like what you got). Seems to me you would have to have a smaller config folder than you could get otherwise. Registry is built at installation. I have no idea how much registry modification is done within XPLite so my assumptions could be completely erroneous.
Had not looked at the Administrative Tools/Local Security Settings before. Looks like lots of things worth exploring there. I guess you could implement AWE from there ...
Don't know if you used AUTORUNS before but I found I could uncheck/delete everything corresponding to a file to be deleted and lots of stuff dealing with SERVICES dependent on SHELL32. This was done within cMP. More stuff eliminated from the registry. There was little left of AUTORUNS when I was finished. Only the entries dealing with stuff I knew was needed. My config folder is now 4.2 mB.
Also went through some of the registry deleting stuff dealing with the services not needed and whose .dll files are deleted anyway. Also, deleted from DEFAULT all file suffixes dealing with audio and video. Got a list from WIKIPEDIA of all file extensions to go by.
Files that were needed (for the machine to start) from an earlier big nLite install (only keyboard and languages unchecked) were not there to begin with with the minimal nLite install leading to a smaller WINDOWS.
Now I want to do it again with this new idea about Administrative Tools/Local Security Settings and hear what happens.
Hi, Rick! I want to make a superminimal nlite install, but I don't know if it will allow Lynx drivers installation, if it will remove all the files and so on. I will use batch files, reg files, but how to clean windows to the extent I am at now,I don't yet know.
Also, how do we do nlite? make a disk and install xp from it? What next?
About the size of the registry. Did diminishing it work positively for you? As far as I understood ALL registry all 16mb(from the start) is always read and is residing in the nonpaged memory, and this is the most precious part of physical memory, that our soundcards' drivers use and, I think, partly cplay/cmp. I tried reducing "visible" memory for windows with /maxmem=*** setting. Interesting that reducing memory timings demands for slightly more like 215-216-217 mb. to ramload in cmp. I am yet to test /burnmemory instead of /maxmemory. The latter setting allows to avoid "holes" in nonpaged memory, that some people report. Anyway, will soon post about the results, including anothes STABLE boot.ini setting. Very good results.
Serge.
Please, for those like me, can you, if enough time, post a more extended NLITEing process? With traps to avoid. What is the size of windows after nlite instal. How functional it remains. What is this talk about registry cleaning?
It is free and simple. And the install goes so quickly finding something that doesn't work is no big deal. About ten minutes start to finish.
You will have to have NETFRAMEWORK on the machine you use to produce the install disk.
There is a place where you can tell it to delete or retain specified files.
I had success deleting the codecs but when I tried to make a big deletion list something or other would go wrong so after goint through eleven disks I gave up. I will try this again but that day it was all I could stand!
If Junaid's registry cleaner works this becomes a moot point anyway
Go download it and play with it. http://www.nliteos.com/
I have made ten's of disks over the last three months to find how little I could get away with.
The minimal recommendation I gave works with JULI@ and PS/2.
If you know which files the LYNX needs just make sure you retain them in the spot where this is allowed (ADVANCED tab) I don't have it in front of me so forgive my minimal help!
You can also load drivers into the install. Though I must add I tried to load JULI@ drivers and this did not work for me. I probably did something wrong.
It is very simple to use just take you time looking at it the first time.
I look forward to hearing what tricks you come up with!
I am planning on making a list of what I did but as soon as I think I have it ready (in my head) you and Junaid come up with bew ideas to incorporate.
As I mentioned before the registry cleaner fclo recommended would not work with my nLite install. I will try Junaid's recommended one when I do the next install which just might be tonight! Can't wait to power that PS/2 mouse with my 5 volts sdd/hdd power supply. Why did that not occur to me?
I could not get this to delete using Junaid's method (music computer SSD oconnected to my regular computer) so I got excited about the MIHAYLOV method which simply did not work.
Desparation ensues.
I noticed the PREMISSION choice and see that permission must be asked for before you delete these files. With each one you must ask and as soon as that is done deletion occurs without complaint.
Now to see if it will work ....
Your ask for '...permission...allow...' worked like a charm for me. I was able to get Steppe's registry, step 2 completed now. Thanks Rick you are right: its a big big impact on transparency.
~could not get this to delete using Junaid's method (music computer SSD oconnected to my regular computer~
The batch files have to be run on the Music computer itself. If they were to be run on a different computer, by connecting the SSD like u tried to do, then u need to edit my Batch files & change the Drive letter of the Root drive to the one which ur host computer has now assigned...
By default, the batch files have the Root Drive letter as C:\
If u connect this SSD to another computer, then it will assign a seperate drive letter to it, (eg K:\)
Then edit the batch file & all u have to do is Use the "Replace All" switch & replace the C:\ with the Drive letter now assigned...
Junaid
I am not aware you had done a batch file for these latest registry mods.
No batch files can be run on my music computer at this point. I had to connect to the other machine, using the LOAD HIVE method.
Would not allow deletion until I right clicked on the entry to be deleted and click PERMISSIONS. Check ALLOW and all was easily done.
Maybe it is because I am making the changes from another computer loading the hive from the MUSIC SSD.
Asking PERMISSION is not a big deal - just right click your mouse and check ALLOW. Simple, but I had to find it by looking at everything since I do not know anything about computers other than what I have learned from cMP. I certainly did not mean to imply this was difficult or scary.
I can say having been "forced" to save this for last the change it made was dramatic. One of the most dramtic changes I have experienced with the process. Not subtle in the least.
My curiousity got to me and I wondered what would happen if one removed SAM and/or security from the config file. Well, the answer is you get to re-do the STEPPE BIOS optimization. No need for anyone else to try this.
or the subfolders below the main entries?
I just deleted the subfolders.
Should I go back and get the other stuff?
.
me too
*************
Sorry, but in what step of 4? If it is step 2 I can't help much, because the comp I am now at, doesn't have a soundcard inside enabled and no drivers installed, there is only asyncmac under sw now on my "virgin" internet/working comp, my cmp hdd is already cleaned. But IIRC there should be entries inside for aec, splitter wdmaud, sysaudio... directsound and so on. DETETE everything INSIDE sw, concerning sound. I mean that when you click on sw it will open and display entries like {eeab7790-c514-11d1-b42b-00805fc1270e} in this examole it is asyncmac, but you should have other entries also like the above example numbers. They can be all deleted, exept sysaudio entry. You should get INSIDE sysaudio entry and also delete inside. Hope this is not too confusing. But I think that there is everything about sound, including asyncmac, but be carefull. I cannot say it from memory. Too much time has passed and too many changes.
Serge.
Edits: 11/28/11 11/28/11
Serge,
I wish I knew computer-speak better. You would probably understand what I was asking if I knew how to ask it!
Under SW, when you first open it there are the basic entries with + signs which when clicked give us the entries with the audio stuff.
I deleted only those entries, the ones that appear after the + is clicked and wondered if I should delete what is left after deleting the items shown when the + is clicked.
I bet if you did the PERMISSIONS thing you could have deleted sysaudio just like you did the others.
It is funny how different each of our installs can behave!
I am assuming I should go back and delete the items with the + sign next to them, also. Of course, now they won't have a + sign next to them ....
Btw even though I could not complete step 2 the sonic impact of these Reg tweaks had a significant clarity, purity, 'improved timbre' effect on my system.
A big thank you.
Serge I could not delete any of the keys in step 2...not just the 7th key down. Anyway you say go inside the keys and delete that way. Can you say a little more about this? How do you exactly do it?
first one made "no sense" in my registry.
Don't know what I was thinking.
On #2, even connected to my other computer (of ocurse, the only way I could make the changes) there was no way to delete the stuff - neither left column or the right.
Any way to trick the machine into letting me do these deletions?
Did you get a sq pop for 4 assuming you could not do 1,2 or 3?
First off, my thanks to Grant (gjwaudio) for his insistence that nLite was the way to go.
My installation has no network and I am using the PS/2 port so I can get rid of the USB stuff. (I delete the USB drivers while connected to another machine at the end of the process)
Make an nLite install disk with the only the following remaining:
DRIVERS
IBM PS/2 TrackPoint (for obvious reasons)
HARDWARE SUPPORT
CPU Intel
Intel PCI/IDE Controller
OPERATING SYSTEM OPTIONS
Certificate Management
Group Policy Mgmt Console
Local Security Settings (this one might be done without. I will try eventually)
Out of Box Experience (if you delete this the machine will not work)
SERVICES
RPC Locator (figure cMP needs this)
Make sure you disable Windows File Protection and Prefetch.
There are other things to do that should be obvious. Look at all of the choices after the main deletion menu. Many are personal taste choices and others disable or eliminate stuff you do not want. Do not rush through here the first time. Think about what you want.
This will result in the fastest WINDOWS install you have ever experienced.
Install the chipset drivers
Implement MINLOGON
Install JULI@ - be sure to go ahead and turn off the undeletable AUDIO CODEC and to check "do not map through this device' and "do not use the mixer".
Implement AWE - when you type in ADMINSITRATORS it will tell you it does not recognize the name. Disregard the warning, cancel out, SAVE and restart.
Install cMP2 software. make sure all is working and check to see if "AWE successful" - if not go through the MMC process again.
Do the optimizations - you will find many of them moot due to this minimal installation.
On my current build I disabled the SMBus Controller in HARDWARE and now it is tricky to restart into STEPPE's BIOS. Next time I will leave this enabled. Anybody know how to RESTART without turning the PS on and off?
Disable the SERVICES. You will notice the list is much smaller and many are already disabled.
Do AUTORUNS. I took a chance (all worked out well) and checked a few of the VIDEO CODECS.
Use the Steppe REGISTRY mods all at once. Junaid's batch files make this quick and easy.
I find it better to delete the files without the batch files with the exception of Junaid's batch file of Mihaylov's last list. After this is done you no longer have access to REG or AUTORUNS. I have had problems with the batch files at times but never with that one. Use one of the lists of files to keep to know what to leave in. You can delete whole swatchs of them pretty quickly between the files you must keep.
I have found it makes no difference in what order the files are deleted. I start at the top of the alphabet and work my way down. I started this on the music machine. You will find the machine will not let you delete some of the files. No big deal, you can get those when you connect the DRIVE to another machine which is the only way to get rid of the last files.
While connected to the cMP machine I did the FOLDERS deletions. I was not able to delete the entire SXS folder but I deleted as much as the machine would let me knowing I could get rid of the FOLDER itself when connected to another machine.
I did restart the machine every forty files or so.
I followed everyone's experience with the FOLDERS, the two Serge's and Junaid's screen shots. Whoever had taken away the most was what I followed.
With this my config FOLDER is a 5.1 mB, my total WINDOWS FOLDER is 36.2 mB.
Been listening to this off and on during the day. Tomorrow will sound better after Serge's Extreme BIOS has had time to settle in.
All in all, this is still quicker, much quicker, than a standard WINDOWS install and just the cMP optimizations.
I do not think it possible to get below this. I am going to try to delete a few files just for the sake of curiosity but I have a feeling I will see only blue screens in trying to go further.
Hi Rick,
As I am about to build my 'real' server soon (I've been using a linear powered netbook since I started on the cMP trip), your process is going to be a big help. It would be nice if we can see if Registry optimization can fit into the process somewhere.
I had tried nLite once when I built a setup for a friend, but did not had the time to evaluate the result properly; but another friend of mine had been using nLite to good result in his system.
KC
before I had done anything other that load the chipset drivers.
nLite must have taken out something it needs.
Already found that going ahead and unchecking as many items in AUTORUNS that correspond with files to be deleted makes a difference.
This, alone, brought my total WINDOWS size to 35mB.
I do not think there is much left to remove but I think there is still something to deleting a few files within nLite; tell it not to load. Mainly stuff like the video and audio codecs and redbook.sys and a few others that do not pop into my head.
I did another build yesterday where I tried the AUTORUNS idea and instead of deleting files in groups I connected the SSD to another computer and did everything at once, all folders and files and the boot.ini file. All started up without incident.
The trick, if you want to use the registry cleaner would be to delete files in groups, becing careful not to delete the files that allow AUTORUNS to work which for me is somewhat mysterious, and then run your registry cleaner. I would do this after the AUTORUNS unchecking. And hope that the registry has not been changed for making changes to the registry!!!
When I was doing my AUTORUNS I unchecked something which allowed the keyboard to work. I need to do another install and pay closer attention. This is what made me do ALL of the file and folder deletions from another machine.
After the next rebuild I will attempt to make a procedure list. I will assume those who do this have some experience with the cMP process. I do not think I would have made a good teacher.
Hi, Rick! At last I am back here on a regular basis and the first thing I want to do, is to thank you for eNlitening! Also great thanks for ps/2 mouse advice. As I am sitting here playing with my mermaid, she is whispering that Nlite is just the toy we'd been wanting for from the tweaks part8. Now, as we are occupied with smth positively different (soon to be posted), you have found the way to slim things down even more. Beginning to rebuild CMP tomorrow (mouse, you know) with nlite. If you will allow a word of criticism..., you should have written, that we don't need to load service pack and hot fixes. Or, am I wrong?
Serge
to the ESSENTIAL MAN of this whole endeavour.
If not for you, Serge, who knows if we would have ever started thicking about this or at best it could have been years down the road. I know it never occurred to me but once you made us aware of your idea I was intrigued and started my computer tinkering with the help of Grant, Mihaylov and Junaid. I have tried to synthesize what I have learned from all of you.
Another thing, the idea of the PS/2 mouse was another of Grant's ideas. I had mentioned this before but think it should be mentioned again.
Thanks to you, especially, and to all, sincerely,
and I did not think it could be deleted! Did not try. I assumed that if SP2 was on your disk you are stuck with it.
If it can, that is better still.
All credit for my interest in nLite goes to Grant. He was the one who brought it to my attention. After a few false starts I would tell him it was not the way and he amiably disagreed vehemently. He is right as he usually is about audio.
I tried to delete more files with nLite and had no success. If WINDOWS would not load then I would run into trouble elsewhere. I do think there is a little more that can be done with this but after ten CD's I gave up.
Another thing I have discovered while doing a fresh install so I could RESTART reliably (SMBus controller) I found that in AUTORUNS - uncheck everything that relies on a file that you know you are going to delete. I did this within cMP and had no trouble.
Could not live without being able to easily get into STEPPE Extreme BIOS.
Hi Rick
Congratulations on Leading the Charge with nLite !! Your success with the process puts the lie to all previous dire warnings against using nLite for a cMP install. It is very powerful - and therefore great care should be exercised when choosing what to cut out, but the payoff is obviously worth the effort.
The next challenge is to work a registry cleaner into the regimen - to learn at which point in the file deletions you stop it executing. Obviously there will be a point beyond which the cleaner won't run, and files deleted after that will remain orphans in the registry.
With the recent report from KCLO, it seems as though Advanced System Care (ASC) is aggressive enough to clean up the mess left by hundreds of files being deleted.
Your next mission - should you decide to accept it - is to add ASC to your recipe, and compare and contrast SQ between versions with a cleaned/not-cleaned registry.
Thank you once again Rick for your persistence in The Quest !
Cheers,
Grant
That's not a Toy... IT'S A TOOL !!
I take it that there is no way to easily do this from an already 1-22 stepped system. I know Mihaylov gave a way to do it from CMP which I was hoping to do but obviously you cant do the nlite stuff that way. I am currently @ 45 mb windows and it would be great to eliminate another 9mb.
Maybe there are a couple of files less than my last list and this is only becasue they were not there to delete.
Since the REGISTRY is created at installation the only way to get a REGISTRY this small is either to do this or to take a chance of doing this without the file deletion part and copy the registry into your current installation.
Who knows how long it would take to whittle down your current registry to this size and even harder to do would be to know what to delete?
I remembered after posting that nLite does allow one to delete specific files. The problem is you have to type them in; won't allow copying for some bizarre reason. So, I guess the REGISTRY can be made smaller still!
with this installation. Took me by surprise.
I have had no problems changing settings within cPLAY with this installation.
I had lost the ability for some reason with the last attempt.
Intrigued by Kclo and Rick's registry posts I took a turn at it with to my ear very good results.
I reloaded an image back when the system was still pretty functional. In my case it was at Steppe7 with Bold Fortune level 9. I installed Kclo's
recommended tool Advanced System Care. I ran all the deep cleaning options. I believe it reported over 5k errors cleaned, and then compressed the registry. I then applied Steppe up to 14. Ran the cleaner and then uninstalled the cleaner. Carried on to apply step16 and Myhalov 1,2&3. Rebooted from another partition and did the folder deletions.
The cmp partition worked without issue.
The SQ in my machine has taken a significant jump. I cannot be fully sure it is all due to the reg optimization as I had made some minor changes to my DAC power supply and it is possible I had not made all the same folder deletions previously. So I'd be interested if anyone else sees this level of change. Basically the imaging is precise in 3d as it never has been before, and the whole thing is very real. It is more or less don't touch it perfect. Definitely taking an image of this one.
It is so nice to see another inmate having good result from Registry optimization. I was trying not to oversell the positive benefits in my previous mails as I was really looking for confirmation for what I was hearing.
To me the sound is now very analog, and there is now a delicate subtlety (sorry, don't know how to put it into words) to the whole sonic spectrum.
KC
Congratulations!
=Basically the imaging is precise in 3d as it never has been before, and the whole thing is very real. It is more or less don't touch it perfect.=
This is very very good.
Excuse me but what is Bold Fortune level 9? All write about it but I don't know nothing :(.
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2 Ru, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Liv
Bold Fortune is a guy who has developed quite a following writing about how to slim down XP. His site is currently here:
http://www.graphixanstuff.com/Forum/index.php?showforum=68
It appears that for a $15 donation, he will supply his batch files.
I put everything he advocates in level 2 into a bat file. Happy to email you a copy.
-
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2 Ru, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Liv
Coming to the realization that I was not going to be able to use the various registry cleaners or AUTORUNS I decided to try something else.
I made a nLite install disk as miminum as I dared that would still allow AWE, JULI@, and all of the stuff that matters. I used another SSD to install this, did the optimizations, installed JULI@ and cMP stuff, chipset drivers. All of the STEPPE registry modifications. Everything that I had done with my main installation. Make sure you leave in cMP mode before making the registry swap. Nothing bad happens, just nothing. I forgot to do this my first attempt and you just get a blank screen.
When done the registry of this version was half the size of the registry on my MUSIC SSD.
I then replaced the registry of my MUSIC SSD with this registry and it workds. My WINDOWS size is now 36 mB.
I did not get a proper chance to audition since I know that without STEPPE Extreme BIOS implemented there is no reason to get in a hurry to assess sound quality.
Tonight I will get more of a chance to see if this is worth it.
Of course, one cannot leave well enough alone so I will try even more minimal nLite installs and see of I can get it smaller, and hopefully sounding better
I thought I made it clear but I guess I did not.
The "registry" build -
approx 1/4 size WINDOWS install disk using nLite (I think this can be made smaller and will try smaller)
There is not much of the cMP optimizations to be done with this install but the ones of importance were done - no paging file - all of the SYSTEM/ADVANCED stuff. I did not turn off services because I knew the corresponding files had been deleted from the cMP Build and I do not think these setting through COMPUTER MANAGEMENT affect the registry.
All STEPPE registry changes were made. No need for deleting files since what I want here is the registry and deleting files does not affect the registry (as far as I can tell).
JULI@ installed - cMP software installed with cMP mode implemented.
SO, other than file deletions this is the same build as my cMP Build which is next to impossible to make wide ranging registry changes.
All of the WINDOWS mass I lost was due to this. I was down to 42.4mB and now it around 36mB. The next time I hook the drive up to the HOST computer I will note the config folder's size.
So other than the deleted files & folders this registry build is basically the same as the cMP Build.
In future I will try smaller and smaller nLite builds that still allow JULI@ to be installed and AWE implemented. From playing with it in the past there is a point where you can't. Wish I had taken notes!
Once you get your registry build done (making sure to shutdown while in cMP mode) you make copy these files to a FOLDER you create while connected to another machine and do the same for the files in the cMP Build machine, just to be safe. Then replace the files in your cMP Build with the files from the registry build.
I had already deleted the registry file that is kept in DOCUMENTS AND SETTINGS since it is just a compilation of parts of the main files.
The only files I moved were the main files - no need to worry about .sav and .log. I deleted them both but the .log files came back.
All I know is it works. Sounded fine when I got a chance to listen to it last night. As I said, until you have listened to the STEPPE's extreme BIOS and let it burn in for a day or so you do not know how good the machine can sound. Tonight I get to hear if I can hear a difference.
Hi Rick,
Did you do the steppe files/ddl deletions before or after the Registry optimization?
It seems to me that you did the Registry optimization before the steppes deletions, and shouldn't it come after to be more effective?
KC
=My WINDOWS size is now 36 mB.=
and what is size of registry (Config folder)?
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2 Ru, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Liv
Did you at least see if it boots/plays cplay?
I did not listen for long since the sound one gets with the extreme bios settings (and that is what I was listening to before I made the change) are so important that a comparison could not be made.
SO, it works just fine. The questions is if it is better and that I will get a chance to ponder this evening.
Maybe I made the post confusing by reminding those interested to be sure the registry you transfer has it start in cMP mode. If it is left in XP mode it will not work since that is part of what cMP does - it changes the registry.
I am only at steppe 2 or 3 but I think something I deleted is needed for Taskkill.
I might not have looked before but now I have the task manager running and I have the lsaa and svchost services set to be killed in cmp.
But when I play music lsaa and svchost still show when playing music. I have it in my head that these used to be disabled by cmp when playing music. IIRC it was a wash in that task kill and cplay showed up and lsaa and svchost disappeared.
Anyhow am I remembering right, and has anyone who has done the more advanced steps looked in the task manager to see if the services are being disabled??
No one here remembers the bending of our minds
d
No one here remembers the bending of our minds
Once you get to the minimum there is nothing to check. They are gone other than the obvious svchost and lsaas.
Lsass & svchost will certainly show up in the Task manager, even if u have set them to be disabled in the cMp settings. A total of 11 processes including Task Manager is what u need to look for when cMp is loaded...
Junaid
Lsass & svchost will certainly show up in the Task manager, even if u have set them to be disabled in the cMp settings.
IIRC, Task Manager itself launches svchost and possibly lsass. If Task Manager is disabled, the two processes are indeed suspended when cPlay is launched. But the only way to check that is to invoke Task Manager which . . .
.
Exactly..
Junaid
Trying to use the Sonicweld Diverter HR (SWD) with cPlay. After uninstalling ASIO4ALL and then installing the SWD driver, cPlay won't run. It says:
"No valid ASIO drivers found!
Player requires at least one ASIO driver to be installed.
If an ASIO driver was installed, its not operational - make sure soundcard is enabled (in hardware device manager)."
The SWD's creator says he can successfully use the Diverter’s ASIO interface with any other player software except cPlay so thinks something is different about cPlay and the way it queries the system for an ASIO driver.
With the SWD being the best USB to SPDIF converter and cPlay the best-sounding Windows software, it would be nice to listen to the combo. Anyone have any advice about the above error message?
Anyone have any advice about the above error message?
I'm pretty sure that cPlay only works with 32-bit ASIO drivers. If your driver is 24-bit only (as the product's web site seems to suggest), it may well not work.
If I'm right, a short-term fix at least is to use Foobar. I don't find it quite as good as cPlay but others do - it's certainly not a night-and-day difference in a cMP^2 setup.
Edit the the line in:-
C:\Program Files\cics Memory Player\cicsMemoryPlayer.pth that starts CUE_PLAYER so it points to foobar.exe.
I've no idea whether it's a big thing to modify the drivers to overcome the anomaly.
HTH
Dave
Thank you for the advice Dave, yes that was a big help. Foobar does indeed work on its own as a player. In my case though, Foobar hangs up when called from cMP; actually, the only player that doesn't freeze when using cMP on my machine -- which is fully tweaked as per the web site (but without the recent series of file deletions) -- is cPlay...
Many thanks again.
that will allow running such a thing?
Has anyone mastered this?
I am assuming explorer, regedit, and reg - but I am sure there is more.
And the question remains - will resoring files be enough or have we passed the point of no return?
Anyone's experience would be appreciated.
I wonder how many are utilizing these?
If you have not you do not know what is possible with cMP.
Serge is not exaggerating when he says once the EXTREME settings are put into place that 24 hours are needed for the full effect. Not just the first time you do this but EVERY time.
One good thing about that is that it keeps you from making changes and encourages you to just enjoy the sound.
When this setting is first heard it is not ugly but it bears no resemblance to what you will hear the next night. It might even improve further after another day.
The spaciousness is simply astounding. Some records I love but have always dismissed as, at best, average recordings reveal themselves to have more to them than I would have ever believed. I figure what sounded like a confused mess turns out to be puzzle that had yet to be solved by my playback gear.
Of course, the good recordings sound good. Less than great recordings can tell you much about your system. I am not speaking of terrible recordings with ridiculous tonal balance but records that sounded too dense for their own good. Mainly pop/rock records with too much Phil Spector influence.
Since I am stuck in SOX (until I replace what is needed to change back to SRC) I have found it to sound good. I had used it before and thought it aggressive. Maybe now is a good time to re-appraise if you are using the upsamplers.
How many are using the extreme BIOS settings?
Rick,Though I am not an EXTREME user (at steppe 4 +jolida 2) I agree with you totally on the effect of a good source on 'terrible' recordings.
When I first got my LP version of George Harrison's "All Things Must Pass" (a Phil Spector recording), in my twenties, I played it a coup[e of times and totally shelved it. No way I could tolerate that mess. Only now with the remastered CD, played in a later version of cPlay can I appreciate the greatness of the Spector production. This is now the only post Beatles record that I play regularly.
KC
Edits: 11/18/11
I must admit I am not much of a fan of his music, not saying I dislike it in any way, just not fanatical. But I always liked him as a human being. He was an interesting mix; spiritual and level-headed all at once.
In the ANTHOLOGY series interviews you cannot help but love the man.
I agree most of the post-Beatles stuff is disappointing. I still listen to PLASTIC ONO BAND and have a soft spot for MEMORY ALMOST FULL, which I think is the most interesting of P. McCartney's solo work. That it came so late is amazing. Pure pop, pure Paul but somehow not as cloying as most of the other stuff. Has some exquisite moments and plenty of embarrassing ones also!
My cMP2 is actually settled with a complete bold fortune + step 1-16 (jolida batch) + jolida 1-2 + mihaylov 2-3 (jolida batch) + some adjustements for my sound card and my system and some manual deletions.
SQ is very good, my windows folder is about 100 mb and i think it's a good result but I've noted that if I keep stopped (no playing music)my system for 1 or 2 minutes, my PC freeze itself and I need to switch off it from the master switch of PC. I don't know which deletion made is responsible of it.
Can someone help me? Thx in advance!
I just noticed this subtopic. I have had this phenomenon on and off at various stages of tweaking cMP and even before the steppes. Since no one ever mentioned it I assumed it was related to my diffent system running on a Lenovo laptop though following all cMP recommendations that can be implemented on it. Presently I have down all the steppes, jolidas and mihaylovs and approached their screenshots as closely as possible given that I have a different network adapter, different 'sound card', that is, a Weiss DAC202 via firewire with its needed drivers and dlls as well as the thinkpad trackpad which has it own required set of these, etc., instead of a mouse. Windows folder=120mb. [It has been quite an ordeal ferreting all this out.] I get this freeze often if cPlay stands idle more than a short while. I also get it sometimes playing a long work: it will keep playing from track to track if I don't I stop it, but if I do it won't start again. If I double click another track while it is playing it almost always happens.The other strange aspect is that it is not itself regular or predictable. Each new state of tweakage in which this happens seems to settle in such that this freezing often becomes less frequent, as if cPlay is 'learning the ropes'.
Edits: 11/22/11
Hello, this should happen with my tweaks (at times) during tweaks part 14-16. Just reboot and start doing smth as soon as you boot into windows. Proceed to part 18 as soon as possible. If this freeze happens after part 20, it is a different story, and is explained (imho) by the fact, that AWE still works, and cplay doesn't have enough time to empty the paged memory block with part 21 bios settings, that severely shorten the memory module timings. (It sometimes happens when you are playing a song and suddenly click on another. The system hangs. The only remedy I know is to set CMP into load into memory and always to use STOP button).
Serge.
THX a lot to you, guys.
In this time I've rebuild from scratch a new cMP2, again, and, with no deletions, THIS PHENOMENON SOMETIMES COMPARE THE SAME!
I'm thinking it was caused by some HW problems ( I use a 2100T with GA H67 MA UD2H B3 mobo)integrated with the serge'explication about RAM.
I've seen too that if I came back, after a while when the system is stopped, and I try to charge another folder on cmp shell, everything works with only a slight wait during the operation.
In this way I don't reboot the system. It's just a little boring...
Daniele
I have from time to time had similar issues. I run using a Toshiba laptop. It did not seem to make a difference what set of deletions I was using. Usually if I moved quickly using a usb mouse it would freeze. Only way out is forced shutdown.
Recently I have disabled the touchpad in the bios. Turning this off, as well as ensureing the wifi radio is powered off makes a substantial improvement in sound. Similar to turning off unused usb ports.
Try going usb mouse without touchpad. I believe I have not seen this freeze since making this change reqardless of file deletion level.
I have done steppe's 1-22 and have no problem as you are describing. What was last level you had where this problem did not occur? Reload if you have an image to verify no freeze. Then proceed with additional steps (already taken previously) until you rediscover freeze. When you get the freeze again it will at least isolate a group of files that are suspect.
Edits: 11/16/11 11/17/11
Check to see if it's not hardware related. I am past steppe 23 with no issues, using the same batch files. I suspect the mouse..
Junaid
Junaid, what is steppe 23? I know steppe 22.
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2 Ru, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Liv
Lol. Sorry, that was a typo. I meant steppe 22 :)
Junaid
THX for your response to my question.
Freezering concerns only cPlay, mouse ( a wifi wireless model) continue to move on the screen butit is inactive. Now I'll retry step by step to searching for the exact step where freezing appears.
Have you made a mix of steppe' steps with mihaylov' or not?
Now I'll try steppe only without bold fortune ( very effective but poor controllable from an informatic inexpert like me).
Daniele
I also do not have this problem. I run Steppe 1-16, Jolida 4.5, Bold Fortune through step 9 and Mihaylov 1,2 & 3. I do no touch winsxs folder as I think it destroys the sound. I use autoruns to delete reg entries with missing files. I think that helps stability.
i run only cmp.
Sound is wonderful.
I do get a lock up as you describe when I shut down. always need manual power off.
I would check your asio drivers. just a hunch
The Win_SxS contents shift the tonal balance. Though the reason for this cannot be technically explained, because some may disagree that it has anything to do with the above. But it sure does. Especially in the Bass & Body of the instruments. As I have mentioned earlier, quite a number of times, it all depends on how a system is set-up, such that one deletion may benefit one, & does quite the opposite to the other. I have always been careful with Just this one folder. It's best to be touched at the End for fine-tuning if necessary....
Junaid
And Yes, I have tried a mix of Steppes with Mihaylov's. In fact my batch files are arranged in alphabetical order in the folder, so when I need to start the process from scratch, I get my 2 lists first, then Mihaylov's 3 lists & then the 16 steppes...
Junaid
Edits: 11/17/11
While doing Bold Fortune, u need to be careful in it, as one batch file will remove the system folder. So watch out for that. I had to edit his batch in that section, so as to retain the above...
Junaid
With all those file, DLL, Registry entry deletions, it suddenly dawned on me that I could do with some Registry cleanup/optimization. After doing just that with CClearner I found the end result quite worthwhile (a denser soundscape is one of the attributes). I wonder if anyone has the same experience? I am at steppe 4 + jolida 2.
KC
Encouraged by the improvement after applying CCleaner's Registry cleaning, I want to try further optimization with optimizers that can dig deeper. And I tried 2:
1. Advanced SystemCare. This utility picked up 1300+ error entries mainly related to ActiveX.
2. Free Registry Cleaner which picked up 360+ errors mainly related to deleted/non-existing Netscape and Microsoft DDLs
The overall effect is improved transparency and detail. Recommended.
A side note: After step 1 above, when I rebooted the system into cMP, a blue screen appeared. I thought I went too far and may be I need to do a partition restore. But it turned out it was not the BSOD, but XP decided to go into Registry defragmentation after a much reduced Registry. After taking a few minutes to defrag the Registry, everything is back to normal.
KC
Tell me please what is the size of the folder Config in the System32 folder in your system now.
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2 Ru, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Liv
Hi Serge,
The size is 48MB. Pls note that I am at steppe 4+ jolida 1 & 2
KC
Hi Rick! Thanks. It's strange because the size of Config folder in my "classic" (before steps) cMP2 OS is 15.5 MB, after Steppe 16 is 15.2 MB, after My Tweak is 15.1 MB.
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2 Ru, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Liv
I don't know. May be because of the different XPs? My cMP is running on a Home Edition. I just checked my other XP system, a Professional, the config folder is at 121MB.
KC
with the music computer SSD attached to a host machine?
Have you tried using these?
Sounds like a great idea, just can't use it on my cMP machine.
Excuse me please but I did not get you Rick.
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2 Ru, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Liv
KC,
Very interesting. Does the utility produce a report of what it found and what it did? If so, can you post that... I'd be very interested in seeing it's impact. I'm especially interested in knowing if it removed the registry entries associated with some of the files removed.
TIA!!!
Greg in Mississippi <---Also at Steppe 4 / Jolida 2
Everything matters!
Hi Greg,In the case of CCleaner, it first does an analysis of your Registry, produce a report (in my case, it had hundreds of entries, and I didn't print it out). Then it would ask if you want to do a backup of the Registry, and then goes ahead to do the work.
No, I did not look deeply into the CCleaner report, (I should have for the sake of science.) With my partition backup and the CCleaner Registry backup, I figured I could always go back if something went wrong or if I don't like the result. Well I do like the result. The sound is now very analog without lost of detail.
Next step I might try a 'deeper' optimizer. CCleaner is considered to be a medium one.
KC
Edits: 11/16/11
I have found the registry stuff in DOCUMENTS AND SETTINGS can be deleted.
In my machine they were in the NETWORK SERVICES folder.
I have also found the .log files come back so it is not worth deleting them.
I find I can run AUTORUNS from a host computer with those still in place.
Would this mean without these files AUTORUNS settings are no longer in effect? My guess is yes.
It will work without them, but ...
AUTORUNS led me to think I was modifying the music computer's registry but the stuff I deleted came out of my regular machine.
Hope WINDOWS repair works.
Excitement can lead to disaster!
and showing us the way for what to do after file deletions and registry changes.
Far more dangerous and no opportunity of going back but this is the next step.
He is the ONE when it comes to these modifications.
After hearing what is possible with file deletions I am ready to give this a try. We are lucky to have him as a guide.
In fact we are a lucky group to have the input from very talented tinkerers. Beyond the realm of amateurs, the tinkerer seeks to improve something just because he HAS to. He is driven by a personal quest for knowledge and insight into how things work and how to make the object of interest better. We are lucky to have these people share their discoveries with us. I know I am not alone in being very grateful.
Yah, it's great to see him back! I wish he'd post more.
In the meantime, while I am no Jack Wong (to paraphrase Lloyd Benson), I've posted the first set of results from my latest Extreme Juli@ mods over in this thread:
Greg in Mississippi
Everything matters!
It is necessary to remove following folders:
disdn
system (with the file winspool.drv)
x86_Microsoft.Windows.GdiPlus_6595b64144ccf1df_1.0.2600.2180_x-ww_522f9f82
x86_policy.1.0.Microsoft.Windows.GdiPlus_6595b64144ccf1df_x-ww_4e8510ac
x86_policy.5.2.Microsoft.Windows.Networking.Dxmrtp_6595b64144ccf1df_x-ww_362e60dd
x86_policy.5.2.Microsoft.Windows.Networking.Rtcdll_6595b64144ccf1df_x-ww_c7b7206f...and following files:
ansi.sys
default.LOG
default.sav
Egosys.ico
logoff.exe
pciidex.sys (only from the folder system32!)
ping6.exe
qprocess.exe
qwinsta.exe
regini.exe
rwinsta.exe
SAM.LOG
SECURITY.LOG
software.LOG
software.sav
system.LOG
system.sav
TempKey.LOG
tscon.exe
tsdiscon.exe
tsshutdn.exe
userdiff
userdiff.LOG
usrlogon.cmd
winspool.exe
x86_Microsoft.Windows.GdiPlus_6595b64144ccf1df_1.0.2600.2180_x-ww_522f9f82.cat
x86_Microsoft.Windows.GdiPlus_6595b64144ccf1df_1.0.2600.2180_x-ww_522f9f82.Manifest
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2 Ru, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Liv
Edits: 11/13/11
With respect to config....I confused config folder within system32 with config.sys which is a file within windows root. Ok I have that clarified now. But I agree with Rick I believe it was Steppe #22 that said to delete all folders within Windows to leave only system, system32 and fonts. Hence Registration was deleted and it works fine for me but Mihaylov says its needed because ... drivers ASIO of Juli are not defined in cPlay without the file R000000000007.clb (which is in Registration).
We may be at the end of the deletion party and I am convinced of the small variations of XP that seem to be the cause of these inconsistencies of what can/cannot be eliminated.
I am convinced of the small variations of XP that seem to be the cause of these inconsistencies
Different XP releases, SP levels, mobo drivers, BIOS firmware, soundcard drivers, configurations (e.g. headless v KVM). On my two Atom mobos, many settings that work fine on one do not work at all on the other.
For example, my Fit-PC does not boot if /nopae is present in boot.ini or /fastdetect is set to always off. There are ten or more LAN-related devices I can turn off in DevMgr on my Intel mobo that I cannot turn off on my Fit-PC although the two boards use the same LAN driver .
I suspect it also depends on how many devices one has disabled in DevMgr (cics' list being by no means exhaustive). For example, he rapped me across the knuckles for advocating switching off DMA but I am convinced my systems sound better for it. They're certainly no worse - my guess is that the chipset drivers handle DMA and that the disabled XP device is a default.
In short, the various sets of instructions, batch files and the like are a welcome guide - many thanks to the various authors - but they are not a panacea. They get one most of the way there and there's nothing wrong with that. The rest of it is subject to test with way too many variables for a prescription.
which can be removed:AUTOEXEC.BAT
CONFIG.SYS
IO.SYS
MSDOS.SYS
(thanks to Dave)
R000000000001.clbIf to convert file system of the disk C from NTFS into FAT32 (by means of Acronis) that folder System Volume Information can be removed without restoration at system reboot. (In this case ntfs.sys in the folder Drivers it will be necessary to replace with fastfat.sys).
- How to remove folders СatRoot and CatRoot2 that they were not rebuilt after system reboot?
- How to reduce the registry size?
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2 Ru, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Liv
Edits: 11/13/11
The file pciidex.sys needs to be removed only from the folder system32.
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2 Ru, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Liv
Rick you said:
'....I tried deleting pcidex.sys with no success.
Forgot, too, that without setup.api JULI@ drivers do not load.
Meant to say vga.sys on my previous post; haven't tried yet to remove...'
Yea I tried pcidex with no luck also. Yes you need setupapi and vga.sys
You deleted the whole config folder? I'm at 47 mb if I delete my config folder which is 16mb I'll be at 31 mb??
Could you use different sample rates when you deleted config?
no way you would every start with that done!
I hope I said I deleted the SXS folder so all of those deletes with the long names will be long gone (no need to delete when they're gone).
You cannot play with cPLAY if you delete the .log and whatever the other one's name is within config. You have to leave the main ones. There is that one section in DOCUMENTS AND SETTINGS, too.
I wonder if all of those most be returned. Maybe only SOFTWARE. I will try tomorrow and see how far one must go.
the registry .log and .sav files are deleted.
I was playing around with the two up-samplers; had been trying SOX and had tried to return to SRC and something went wrong: like and on and off sound and impossible to get out. Had to restart; even required resetting BIOS. All was fine after that but I guess to use another one these files must be returned.
I found that one cannot delete WINSPPOL.EXE - the rest were alredy deleted and with the exception of the problem described - no other problems.
I have deleted SXS entirely - a few weeks ago with no ill effects.
Deleted the folders RYELANDS mentioned. All went fine.
Wonder how many of the empty DOCUMENTS AND SETTINGS folders can be deleted? Most of them are empty. Will try later this week.
I tried deleting pcidex.sys with no success.
Forgot, too, that without setup.api JULI@ drivers do not load.
Meant to say vga.sys on my previous post; haven't tried yet to remove.
Current WINDOWS size: 41.7 mB with the registry files deleted.
- I checked. The files *.log and *sav removed from the folder config do not influence the settings of cPlay - all works fine.
- The remained content of the folder WinSxS is necessary for remote access to cMP2.
- The drivers ASIO of Juli are not defined in cPlay without the file R000000000007.clb therefore do not delete this file!
Rick, what size of the folder config in your OS?
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2 Ru, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Liv
- The drivers ASIO of Juli are not defined in cPlay without the file R000000000007.clb therefore do not delete this file!-
Strange. I have deleted all Registration entries without having issues with cPlay or cMp..
After steppe 22, the sound is Awesome. I don't feel I need to go ahead with anything else. Feels like cMp optimisations are finally complete. The Most critical files influencing the sound along this delete route seemed to be the ones in the Win_SxS folder. Technically the contents of this folder may not have anything to do with the Audio section of a computer, but in Practice, it did...
If one is experiencing a forward or thin sound, it's advisable to experiment with what goes into or out of Win_SxS..
Junaid
And if so, can you still go back and forth between the upsamplers?
I am not sure if you use them at all.
I had not deleted the REGISTRATION file, which does not contain the file that has been mentioned as necessary for cMP, I do not think my machine ever continued the file ending with "7".
I have not tried going back to my last image and seeing if I can change the up-sampler. I do think I had made the change (from SRC to SOX) after that image had been made.
Just curious ...
Junaid, are you ready to remove components from your MB?
but I very well may have.
Thanks for the correction. I will try restoring that first. Tomorrow.
Don't have access to the drive but I believe I remember correctly that without the superfluous registry files I was at 41.7 nB. No more than that.
I figured you had your reasons for retaining SXS due to your network stuff.
Though I thought the other Serge had said you recommended deleting it entirely. I do not remember you saying this directly. I tried it and have had no problems. It is staying out.
I guess your network arrangement allows you to delete WINSPOOL.EXE. My machine would not allow it!.
I do think I am down to the minimum. Other that deleting empty folders and some kind of massive registry deletion (!!!) I think I am as low as I can go.
Does seem there is lots of registry that could be deleted. Much tricker than files deletion.
Wish I could use AUTORUNS. Don't feel like re-installing all of those necessary folders.
That is the only conclusion I can think of too (i.e., different xp systems). Not a problem though I always have my prior snapshot level to go back to.
But no problem I can always go back to a prior snapshot level
Here:
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Live, iPad
Edits: 11/13/11
Many thanks for that. Not that it matters in the scheme of things but IO.SYS, MSDOS.SYS, Config.SYS and Autoexec.bat can safely be deleted from C:\.
Thank you. I'll try.
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Live, iPad
Tried these deletions ...mostly config stuff plus pcidex...and it wouldn't boot. I restored all the files and it still would not boot. Back to a prior level via Snapshot.
Edits: 11/13/11
Theo, I do not know what to say. It is strange. I have the feeling that we have different OS :).
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Live, iPad
As of today. I am going to try losing shutdown.exe, vha.sys, and perfc_009 again.
advapi32.dll system32
authz.dll system32
basesvr.dll system32
bootvid.dll system32
c_437.nls system32
c_1252.nls system32
comctl32.dll system32
comdlg32.dll system32
CONFIG.TMP system32
cryptdll.dll system32
csrsrv.dll system32
csrss.exe system32
ctype.nls system32
dnsapi.dll system32
FNTCACHE.DAT system32
framebuf.dll system32
gdi32.dll system32
hal.dll system32
juliaasio.dll system32
kdcom.dll system32
kernel32.dll system32
l_intl.nls system32
locale.nls system32
lsasrv.dll system32
lsass.exe system32
mpr.dll system32
msasn1.dll system32
msprivs.dll system32
msvcp60.dll system32
msvcrt.dll system32
ncobjapi.dll system32
netapi32.dll system32
ntdll.dll system32
ntdsapi.dll system32
ntoskrnl.exe system32
ole32.dll system32
oleaut32.dll system32
oledlg.dll system32
olepro32.dll system32
perfc009.dat system32
reg.exe system32
rpcrt4.dll system32
samlib.dll system32
samsrv.dll system32
scesrv.dll system32
secur32.dll system32
services.exe system32
setupapi.dll system32
shell32.dll system32
shlwapi.dll system32
shutdown.exe system32
smss.exe system32
sortkey.nls system32
sorttbls.nls system32
svchost.exe system32
sxs.dll system32
umpnpmgr.dll system32
unicode.nls system32
user32.dll system32
userenv.dll system32
vga256.dll system32
vga.dll system32
watchdog.sys system32
win32k.sys system32
winlogon.exe system32
winmm.dll system32
winspool.drv system32
winsrv.dll system32
winsta.dll system32
WLDAP32.DLL system32
ws2_32.dll system32
ws2help.dll system32
acpi.sys driver
atapi.sys driver
classpnp.sys driver
disk.sys driver
dxapi.sys driver
dxg.sys driver
dxgthk.sys driver
ftdisk.sys driver
hidclass.sys driver
hidparse.sys driver
hidusb.sys driver
i8042prt.sys driver
isapnp.sys driver
jula.sys driver
julawdm.sys driver
kbdhid.sys driver
ks.sys driver
ksecdd.sys driver
mouclass.sys driver
mountmgr.sys driver
npfs.sys driver
ntfs.sys driver
partmgr.sys driver
pci.sys driver
pciide.sys driver
pciidex.sys driver
vga.sys driver
videoprt.sys driver
volsnap.sys driver
wmlib.sys driver
I did not use XPLite on my current OS. I bought a copy but there was something about it I did not like. I may have been overly influenced by BOLD FORTUNE. Who knows?
Now I wonder just what it does.
Does it simply delete files or does it also make registry changes? Anyone know?
I found a program that allows one to modify the registry on your music computer HDD from another computer. I will lost a link later. It is at home.
Ready for Serge's latest registry changes!!! I think ...
Inmates, who have passed all Steppes including Mihaylov's lists, may not have access to their Registry to make further changes onto it. To do the same at this stage, we are left with Two options...
The First option is to attach ur Hard Disk to another Windows computer via eSata or USB or whatever, & access the registry by this method:-
If you put the drive from machine (A) into another machine (B) you can use the regedit on machine (B) to access the hive filesSimply run Regedit select the HKLM key and from the "file" menu you should see an option to load hive.
Browse to the desired hive on the hard-drive you connected (ensure that you have access to where the hives are stored, for XP it will be in the "windows\system32\config" directory...
The Second method, is quite the same, but does not need another pc. Instead we create a Bootable Windows Image on a CD & boot to it from the Start-up. The procedure is listed below...
Make sure that your system has about 500MB of free disk space!
Download the latest PE Builder version (self-installing package) and install it. Link:-
http://www.nu2.nu/pebuilder/#download
Start PE Builder (pebuilder.exe). When you start PE Builder for the first time it will ask if you agree with the license agreement.
Now PE Builder will ask to search for windows installation files. If you don't have your windows XP setup/installation files on your system you must insert the original Microsoft Windows XP installation/setup CD at this point.
The files you have at c:\windows are not installation files. They are your already installed files!
Click "yes" to start searching. PE Builder will now search all fixed- and CD-Rom drives for Windows installation files. This will take some time. When more than one valid location is found, a dialog will appear where you can select which location you want to use.
At the main PE Builder dialog, select the "Burn to CD/DVD" option. When you are using an erasable medium, make sure that the "AutoErase RW" option is enabled. The "burn using" option should be set to "StarBurn". Select your CD writer device from the Device list.
Hit the "build" button. PE Builder will now ask you to create a BartPE directory, answer with "yes".
The license for your Microsoft Windows XP product is shown. Read it and agree to it to continue.
PE Builder will now start building BartPE. This will take a few minutes.
You will see a lot of files getting copied and/or decompressed, the ISO image build and the data recorded to your CD/DVD writer.
If the data verify was correct and there where no errors reported you can boot the CD/DVD!1.Insert the BartPE CD into the drive, and boot the system from the CD. Once the file loading phase is over, the Bart PE desktop will be visible, as shown in Figure 1.
2.Type Regedit.exe in the prompt, and press Enter. Select the HKEY_USERS hive
3.From the File menu, choose the Load Hive option. Browse to your Windows installation drive, for example the following location:
C:\Windows\System32\Config\
4.Select the file named SOFTWARE (the file without any extensions), and click Open
5.Type a name for the hive that you've loaded now. (Example: MyXPHive)
6.Now the SOFTWARE hive is loaded, and present under the HKEY_USERS base hive.
7.After entering the correct data, you MUST unload the Hive. To do so, select MyXPHive branch, and then in the File menu, choose Unload Hive. It's important to note that you'll need to select the MyXPHive branch first, before unloading it.Another advantage of creating the Bootable CD, apart from using it to make changes to the Registry, is to make use of the USB functionality which some of us lost in the deletion process. As we would be requiring the Thumb drive to transfer music onto the HDD, it can easily be done by booting to this version of Windows (via the Bootable CD)....
Junaid
Edits: 11/08/11 11/09/11 11/09/11
I have my music computer SDD attached to the "other" computer with the GRANITE USB adaptor, since otherwise my machine will not boot with the drive attached to a SATA port.
I thought I had tried this before and could not get it to "LOAD HIVE". The choice is not offered at all. You can see the shadow, but that is all.
I tried playing with that thing I sent out. It does not allow you to delete. You can change values (I guess) in the right panel but I have no idea what I am doing with that.
I am starting to think that if Serge's new changes are extensive it might be best to stat over. I am already thinking of better ways to get where I am.
Also, are you really able to boot your music computer with gdi32.dll deleted?
Rick,
Im sure u have made some mistake here. The procedure is very simple & i myself have done it. I guess u quite dint understand it properly. Let me explain in a straightforward manner...
With ur SSD attached to the 'other' computer, in any which way (it does'nt matter), it would show up in its "My Computer" as say 'E' drive.
All u have to do is click 'Start'> 'Run'> regedit..
Now u will have the registry editor open. In case u have any tree open, collapse all of them. Now here comes the tricky part...
Just click on HKEY_USERS once so its just selected. NOW go to File menu, & u will have the "Load Hive" option available. Click on it & browse to the Config Folder of ur SSD ( which in this case would be E:\Windows\System32\Config). And from it, select 'software' by double-clicking on it. Now it will ask u to assign a name for this Temporary tree u are creating under the HKEY_USERS. Name it something like 'Rick'.
Now under this Rick tree, u will have access to the Registry of ur SSD. After making any changes whatsoever, remember to UNLOAD the Hive by Just clicking on the 'Rick' & from the File menu, select Unload Hive..
You're Done...!!
Junaid
Big issue: while making registry changes this way one time I did not unload hive after changes and unfortunately I did revise my home pc registry. No problem though everything worked well so no issues but I learned not to make that mistake again. But after latest Steppe registry tweaks I noticed not only did I revise my cmp^2 ssd but I also inadvertently revised my home pc registry. All works ok but I am reluctant to use this method any more period. So it looks like I need another method. Mihaylov's method is probably the way to go. More to come.
THANKS, again!
That thing I sent seems to not offer complete access. And I have not tried to actually change anything with it.
This seems like the way to go.
AS ALWAYS, THANKS for your research and shared intelligence.
My pleasure :)
Junaid
Hi Rick!
You can edit the registry from the shell cMP if temporarily return the four files: regedit.exe, aclui.dll, ulib.dll and clb.dll. You must also edit the line "TOUCH_SCREEN..." in the file cicsMemoryPlayer.pth to the prescribing there path to regedit.exe.
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Live, iPad
Serge I tried restoring these files and amending the CicsMemoryplayer.pth as you indicated and then I just doubleclick on touchscreen and I should get into the registry editor right? But nothing happens I just stay in the cmp screen. Is there something I am doing wrong?
That's all right. Now I editing the registry in manual way. I have returned the files aclui.dll, ulib.dll, clb.dll into System32 folder and regedit.exe into Windows folder. Also I have replaced cicsMemoryPlayer.pth in Program Files/cics Memory Player folder with folowing cicsMemoryPlayer.pth:
RIPPER "c:\windows\system32\taskmgr.exe"
CUE_PLAYER #N "c:\program files\cics Play\cicsPlay.exe" %C
LIBRARY_MANAGER "C:\Program Files\dpclat.exe"
OSK "C:\Program Files\cics Memory Player\ahkosk.exe"
PROCESS_EXPLORER "c:\Logon.bat"
EXPLORER_KILL "c:\windows\system32\taskkill.exe" /F /IM explorer.exe
TOUCH_SCREEN "c:\windows\regedit.exe"
PREP_PLAYER
REFRESH "Эверест" ""
.............................
(I use the only button T.Screen in cMP settings interface so do not pay attention to other lines of this file).
And after pushing the button T.Screen the Registry Edit window is appeared. I do not fathom why it is impossible to you. :(
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2 Ru, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Liv
Serge Mihaylov as I have previously stated this method does not seem to work for me. I replaced the files both you and jolida mentioned and double checked them by reloading a 3 month old version of my os, opened up regedit and ran mark russinovich process explorer to verify. So the only difference could be in the cicsmemory player path file. So I changed that to replicate your ripper 1st line....then tested both. Neither works for me. Maybe taskmgr.exe is deleted too. Anyway did you put regedit.exe in System under Windows or in Windows root directory? Also when I double click T screen it just recycles the cmp screen without ever getting into registry editor.
Edits: 12/07/11
Thanks Serge. You direction has been very clear. I understand what you are saying but just cant get it to work. Thanks again for taking the trouble to post.
Once again. To return the files aclui.dll, ulib.dll, clb.dll into System32 folder (maybe some others else) and regedit.exe into Windows folder. Also replace cicsMemoryPlayer.pth in Program Files/cics Memory Player folder with folowing cicsMemoryPlayer.pth (see link below). Then to run cMP and to open settings and:
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2 Ru, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Liv
Don't know how I missed it.
Thanks,
(You should use a heading that lets it be known what is "inside"!)
I put regedit into system32 but in the pth file I wrote
TOUCH_SCREEN "c:\windows\system32\regedit.exe"
That should work shouldn't it?
Theo, the file regedit.exe don't run from System32 folder or from anywhere apart from Windows folder!
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2 Ru, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Liv
I tried it by putting regedit.exe in windows and revised my cicsMemory.pth as you specified...still doesn't work. Oh well probably another difference in our xp files.
Can you post the precise language to put into the cicsMemoryPlayer.pth for guys like me that dont know how to?
TOUCH_SCREEN "c:\Windows\regedit.exe"
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Live, iPad
Thanks
since I did not use XPLite.
I did on a previous build and had problems which I blamed on it.
Wondering if I may have a cluttered registry AND if XPLite would have made a difference there. Does it do more than just delete files and remove stuff from the registry? I have looked and this is not made clear at their website.
Found a neat thing that allows reg changes from the HDD/SDD without all of that returning of files.
I will post the link to it.
Thanks,
This one can be copied since it is not a screenshot.
My system uses a PS/2 mouse so I was able to get rid of the USB files.
I have added disabling all USB ports in BIOS for the extreme setting. It does make a difference. PS/2 is the way to go.
My folders are as Serge prescribed. I am just over 50mB for WINDOWS.
I would appreciate any suggestion of additional files to delete! I know there are some I need to get rid of such as the eegosys.ico. In time I will. In the "perf" series I suspect there are some here that are unneeded. As the week goes by I will see what I can delete.
Does anyone know of a way to work with the registry with the SDD attached to another machine? That would be a great thing if it is possible. The thought of having to reinsert files for this ... But if must be done!
I would have thought when we setup our cMP with the initial optimizations that we would have eliminated the page file. Did it come back due to our file deletions? Is it worth worrying about?
Hope someone finds this useful.
SYSTEM32
advapi32.dll
at.exe
atrace.dll
authz.dll
autodisk.dll
basesvr.dll
bootvid.dll
c_1251.nls
c_1252nls
c_437.nls
c_1257.nls
cdm.dll
clbcatq.dll
clipsrv.dll
comctl32.dll
comdlg32.dll
CONFIG.TMP
cryptdll.dll
cscdll.dll
csrsrv.dll
csrss.exe
ctype.nls
diskperf.exe
dnsapi.dll
egosys.ico
FNTCACHE.DAT
framebuf.dll
gdi32.dll
h323.tsp
jscript.dll
juliaasio.dll
kdcom.dll
kernel32.dll
l_intl.nls
locale.nls
locator.exe
loghours.dll
lsasrv.dll
lsass.exe
mpr.dll
msasn1.dll
msidle.dll
msisip.dll
msprivs.dll
mstinit.exe
msvbvm50.dll
msvbvm60.dll
msvcp60.dll
msvcrt.dll
ncobjapi.dll
net.exe
netapi32.dll
netrap.dll
nmmkcert.dll
ntbackup.exe
ntdll.dll
ntdsapi.dll
ntoskrnl.exe
occache.dll
ole32.dll
oleaut32.dll
oledlg.dll
olepro32.dll
pci.sep
perfc009.dat
perfci.ini
perfdisk.dll
perffilt.h
perffilt.ini
perfctrs.dll
perfnw.dll
perfos.dll
perfproc.dll
perfwci.h
perfwci.ini
pid.inf
pid.PNF
ping.exe
prodspec.ini
qprocess.exe
qwinsta.exe
reg.exe
regini.exe
rnr20.dll
rpcrt4.dll
rsaenh.dll
rsvpperf.dll
rwinsta.exe
samlib.dll
samsrv.dll
scesrv.dll
secur32.dll
services.exe
setupapi.dll
shell32.dll
shlwapi.dll
shutdown.exe
smss.exe
sortkey.nls
sorttbls.nls
ssdpapi.dll
svchost.exe
sxs.dll
tftp.exe
tlntadmn.exe
tsshutdn.exe
udhisapi.dll
umpnpmgr.dll
unicode.nls
upnp.dll
upnpcont.exe
upnphost.dll
upnpui.dll
user32.dll
userenv.dll
vbicodec.ax
vbscript.dll
vga256.dll
vga.dll
watchdog.sys
wdmaud.drv
win32k.sys
WindowsLogon.manifest
winlogon.exe
winmm.dll
winspool.drv
winspool.exe
winsrv.dll
winsta.dll
wkssvc.dll
WLDAP32.DLL
ws2_32.dll
ws2help.dll
wsock32.dll
xpsp3res.dll
SYSTEM32/DRIVERS
acpi.sys
atapi.sys
classpnp.sys
disk.sys
drmk.sys
dxapi.sys
dxg.sys
dxgthk.sys
ftdisk.sys
hidclass.sys
hidparse.sys
hidusb.sys
i8042prt.sys
jula.sys
julawdm.sys
kbdhid.sys
ks.sys
ksecdd.sys
mouclass.sys
mountmgr.sys
npfs.sys
ntfs.sys
ohci1394.sys
partmgr.sys
pci.sys
pciide.sys
pciidex.sys
portcls.sys
vga.sys
videoprt.sys
volsnap.sys
wdmaud.sys
wmlib.sys
Delete @ your own caution but stuff I dont have in my list are
at.exe
cdm.dll
clbcatq.dll
clipsvr.dll
cscdll.dll
diskperf
jscript.dll
locator.exe
msidle.dll
msisip.dll
mstinit.exe
perfc009.dat
perfci.ini
perfdisk.dll
perffilt.h
perffilt.ini
perfctrs.dll
perfnw.dll
perfos.dll
perfproc.dll
perfwci.h
perfwci.ini
pid.inf
pid.PNF
ping.exe
prodspec.ini
qprocess.exe
qwinsta.exe
regini.exe
rnr20.dll
rsaenh..dll
rsvpperf.dll
rwinsta.exe
ssdpapi.dll (i use an ssd)
tftp.exe
upnp.dll
upnupnpcont.exe
upnphost.dll
upnpui.dll
vbicodec.ax
vbscript.dll
vga256.dll
vga.dll
wdmaud.drv (I use juli@)
windowslogon.manifest
wkssvc.dll
wsock32.dll
xpsp3res.dll
drmk.sys
julawdm.sys
ohci1394.sys
I was double checking mine to see if I had further opportunities so it was no trouble doing the list.
msvbvm50.dll
msvbvm60.dll
net.exe
occache.dll
pci.sep
Had no trouble. Deleted them all except for the JULI@ file. Will try that tomorrow. Same with those c_**** files!
As it as the moment:
System32
advapi32.dll
atrace.dll
authz.dll
autodisk.dll
basesvr.dll
bootvid.dll
c_1251.nls
c_1252nls
c_437.nls
c_1257.nls
comctl32.dll
comdlg32.dll
CONFIG.TMP
cryptdll.dll
cscdll.dll
csrsrv.dll
csrss.exe
ctype.nls
dnsapi.dll
egosys.ico
FNTCACHE.DAT
framebuf.dll
hal.dll
juliaasio.dll
kdcom.dll
kernel32.dll
l_intl.nls
locale.nls
loghours.dll
lsasrv.dll
lsass.exe
mpr.dll
msasn1.dll
msprivs.dll
msvcp60.dll
msvcrt.dll
ncobjapi.dll
netapi32.dll
netrap.dll
nmmkcert.dll
ntbackup.exe
ntdll.dll
ntdsapi.dll
ntoskrnl.exe
ole32.dll
oleaut32.dll
oledlg.dll
olepro32.dll
reg.exe
rpcrt4.dll
samlib.dll
samsrv.dll
scesrv.dll
secur32.dll
services.exe
sfc.exe
shell32.dll
shlwapi.dll
shutdown.exe
smss.exe
sortkey.nls
sorttbls.nls
ssdpapi.dll
svchost.exe
sxs.dll
tlntadmn.exe
tsshutdn.exe
udhisapi.dll
umpnpmgr.dll
unicode.nls
user32.dll
userenv.dll
vga256.dll
vga.dll
watchdog.sys
win32k.sys
winlogon.exe
winmm.dll
winspool.drv
winspool.exe
winsrv.dll
winsta.dll
WLDAP32.DLL
ws2_32.dll
ws2help.dll
System32/DRIVERS
acpi.sys
atapi.sys
classpnp.sys
disk.sys
dxapi.sys
dxg.sys
dxgthk.sys
ftdisk.sys
hidclass.sys
hidparse.sys
hidusb.sys
i8042prt.sys
jula.sys
julawdm.sys
kbdhid.sys
ks.sys
ksecdd.sys
mouclass.sys
mountmgr.sys
npfs.sys
ntfs.sys
partmgr.sys
pci.sys
pciide.sys
pciidex.sys
portcls.sys
vga.sys
videoprt.sys
volsnap.sys
wdmaud.sys
wmlib.sys
Can't wait to get home and delete some more!
I did not use XPLite so I figured there was stuff on my list that TAHT took away.
I assume you have got rid of the USB drivers?
Add to your cmpEXtreme bios disabling the all of the USB. Yes, you have to short the pins to restart but if gives you the screen with the BIOS choices so it not that big a deal to restart.
Bye,
Yes I tried leaving usb in but it hung up on boot so I disabled all usb in bios and it booted fine. I have stuff in system 32 and drivers that you dont. But as I recall juli@ would not work w/o setupapi and prodspec and pschdprf but that was in xp. Maybe its different in cmp mode. You dont use moushid and your ps/2 mouse works? If so maybe I'll try. Also you dont use lanmandrv and I do. There are about 6-8 others but I cant remember their names now. Did you try the nopae boot.ini changes
? That is only one I have not done. Also I deleted all the files I put back in to make cut&paste from a thumb drive work. Does not make sense to leave in that stuff since I cant get back to xp anyway. I guess I'll load new music to the hd directly or by reloading my last good xp image via Snapshot.
After getting used to the sonics I really like it now.
as far as files differences.
I need to take a new SNAPSHOT and that will be a good time to re-check my work.
Yes, I do think cMP ONLY has changed everything. Most for better but certainly a bit of a negative for convenience. If we wanted convenience we would have left all of this alone. But once you hear what this can do it is an easy compromise to make.
I did make the changes to boot.ini. I have not made any comparisons between the "new" one and the old one. I do know I am very pleased with where this is going.
I have found something that allows changing registry entries on the cMP HDD from another computer.
Thanks, again,
Ever since this cMp project has started, has anyone compared it to a CD player??? If so, against which player & where does it stand in comparison...???
Junaid
I keep a CD Player handy for when I seriously break my DAC.
The CDP is a highly modified Sony with a nicely done Tube Lampizator output stage with 6n6p tubes (as popularized by BAT). Lots of power supply mods with tons of OSCON SEPC. Certainly orders of magnitude better than stock.
When I plug in CMP2 on a 1541a based superdac, no comparison. CMP pushed the CDP aside. Way better clarity, sound stage and sense of reality. There is texture in instruments. The thing I am noticing more now is the sense of life in singers. The breathing between notes is just becoming spooky. None of this feels real on the CDP.
"The thing I am noticing more now is the sense of life in singers. The breathing between notes is just becoming spooky. None of this feels real on the CDP."
Agreed... 100%
Junaid
I have a modified Oppo 870 that outputs digital streams from 44 - 88khz sample rate. I bought it 4 years ago to be able to get to my single layer sacd's (ie,. no cd layer) and cmp beats it handily when compared to files converted to 16/44 off the analogue outs of the Oppo. Ccmp sounds better with files via ripped cd layers off my dual layer sacds versus 88khz out to my Buf32s dac (which is fed by cmp as well so good apples to apples comparison).
Anecdotelly via memory, cmp sounds way better than my heavily modded SCD1 which is now sold.
Edits: 11/07/11
I guess the Oppo is a Blu-Ray player...??
Junaid
yes
Thought so. What I meant was comparing the cMp with a dedicated CD player, say in the $2000 range or more...
Junaid
to insert any component easily especially a CD player.
Doesn't matter to me. I have no intention of returning to the CD PLAYER so it is a moot point for me.
I suspect we are approaching a sound quality as good as any CD player of comparable expense. At this level of performance I think the differences are subject to various compromises. Some may prefer another set of distortions. Perfection is not possible so, one way or the other, you learn to live with shortcomings no matter what the primary source may be.
I prefer working with the limitations of any one component by adjusting the system. Who knows the source of anomalies within the system? If it is easier to adjust the tonal balance with your speaker's crossover I will do it there. Who knows which one was absolutely correct before adjusting? It is a puzzle and will remain so.
I run Win 7 which does not have a boot.ini file.
Would anyone be kind enough to tell me how to set these parameters in Win 7 (assuming they yield SQ improvement of course)?
Thanks
Jonathan
Hello, everybody. This is tweaks part 22 and I hope the last part. First, let me thank Serge Mihaylov for finding *.nls files that are safe to remove and for suggesting the basic trick of the last tweak. Also Great thanks to Theo for reg.exe file
It goes Like that:
Restore all files necessary to start cmp from reboot and to shut down windows from cmp. Their list + shutdown.exe and reg.exe are found in Serge Mihaylov’s answer to my plea for help. Set Your cmp to start cmp.
Next reboot your computer from another (untweaked) windows and remove *.nls files EXEPT 1252.nls (Theo, checkit!).
Make sure that cmp windows works and cmp starts and shuts down normally.
Once again reboot into untweaked windows and remove:
From windows root – EVERYTHING, even explorer.exe. Leave only 3 folders: FONTS, SYSTEM and SYSTEM32
Folder Fonts should contain for original US version of windows these files: 8514fix.fon, 8514oem.fon, 8514sys.fon and vgaoem.fon and ttf fonts: arial.ttf, l_10646.ttf, marlett.ttf and micross.ttf. If you have there desktop.ini file, DO NOT TOUCH IT
Folder System should contain only MOUSE.DRV.
Remove as per Serge Mihaylov’s recommendation CatRoot and CatRoot2 folders from System32, folder WinSxS completely. Remove Microsoft\Protect\User folder, Temp folder, pchealth, resources. Leave only 3 folders: Fonts, system and system32. Inside system 32 3 folders should remain: config, drivers (36 files left), DRVSTORE (created by Lynx. Don’t know if you have it).
Remove these files: (Every single one is a jewel and all make great improvement)
Explorer.exe, sethc.exe, userinit.exe, apphelp.dll, browseui.dll, cfgmgr32.dll, crypt32.dll, cryptui.dll, imagehlp.dll, iphlpapi.dll, msports.dll, msv1_0.dll, rpcss.dll, setupapi.dll, shdocvw.dll, srvsvc.dll, uxtheme.dll, vdmdbg.dll, version.dll, wininet.dll, wintrust.dll, vga.drv, acpiec.sys, dmboot.sys, dmio.sys, dmload.sys, enum1394.sys, fs_rec.sys, imapi.sys, mcd.sys, udfs.sys, eula.txt. PLEASE NOTE, that as I use mouse, I have left several hid*.sys usb*.sys and mouhid*.sys files, that Serge recommends to remove.
After that the size of my Windows folder is 49.7 mb. System32 contains 80 files, of them 49 dll, 11 exe, 13 nls files and fntcache.dat, wdmaud.drv, winspool.drv, watchdog.sys, win32k.sys, config.tmp.
After this comes the MOST IMPORTANT PART:
Open your cmp boot.ini file with a notepad and add /nopae and /noguiboot, next remove from windows\system32\ logonui.exe. Reboot and be prepared for surprises!
NOTE 1 Your cmp should be set to start cmp, not xp.
Note 2 Your cmp boot.ini file should read like this:
after windows xp professional
/noexecute=alwaysoff /fastdetect /timeres=9800 /nodebug /pcilock /nopae /noguiboot
Have fun!
Some of you, inmates may still remember the discussion about /nopae key in boot.inni. It was agreed, and I advocated as well, that this option brings harshness to the sound. It was, but not now with the mids so clean. The /nopae option will bring more meat to the highier mids and to cymbals. (IMHO) Certainly, /nopae is optional, while /noguiboot is a must. Don't forget to remove logonui.exe. It in itself is worth a minlogon.
Funny Thing: removing explorer.exe allows to remove SHELL32.dll. Just imagine the possible improvement after removal of this memory residing monster. CMP won’t load without it, alas. It starts, but fails to load… Removing shell32 and winspool.drv would be a great achievement!
That is all, inmates.
Serge.
P.S. Sorry, if it took too long and if it was so troublesome.
P.P.S. I have a 4-step Big and Important registry tweak from part 12-15 period, which I have forgotten to post and a trick with ipsecurity and services that is to be done at the Cics’ part of windows tweaking. Will post it, if anybody is interested. Also, as I have rebuilt my cmp with my tweaks 5 times, I have a certain when-to order of things that can be helpful for those who start from scratch. Tell me if it is interesting.
ONE MORE PLEA FOR HELP: Folks,there had been a discussion concerning hipet and timeres=97?? option in boot.ini. Would experiment with it, but please, post THE EXACT syntax...
To disable both DEP and PAE:
/noexecute=alwaysoff or /noexecute=alwaysoff /nopae
(Windows XP with SP2)
(These parameter combinations are equivalent.)
See here:
http://msdn.microsoft.com/en-us/library/windows/hardware/ff542275(v=vs.85).aspx
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2 Ru, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Liv
Tried nopae. Initially liked it...bass, dynamics better, darker overall sonics. Tired of it eventually and went back. But it is closer versus last time. I can see that nopae may be system dependent. Probably worth a try.
First you write "remove *.nls files EXEPT 1252.nls" but later state after all removals "System32 contains ... 13 nls files" ??
Hello! I wrote remove the *.nls files, mentioned and discovered by Serge Mihaylov. So the only file not to be touched from His List is 1252.
For reference use my screenshots, please.
Also, make sure you checked the list of my screen fonts. They differ from his (Russian) version.
;)
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2 Ru, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Liv
Certainly, inmates, You understand that I took the snapshots from another windows install.
Serge.
all of those USB drivers and portcls.sys.
Then turn off all USB stuff in BIOS.
I hear an improvement.
No need for all of those USB optimizations!
Of course, one must start with a PS/2 mouse to do this.
Credit Grant (gjwaudio) for suggesting the benefits of a PS/2 mouse.
+1 This is the way I do it too.
Hello, everybody. In case there is any doubt or misunderstanding, here are the snapshots from my folders. Also, no, I haven't posted the "lost and last registry tweaks". There are 4 of them and they are significant, but well documented and easy. Will post them shortly.
Serge.
just put whatever is left in there inside SYSTEM32. This is per BOLD FORTUNE.
Also, have you cleared out the stuff in config/systemprofile?
Cleared out the DOCUMENTS and SETTING folder, too. Except for the cMP stuff, of course!
Hello, everybody. Will post screenshots, surely. Now! nogui is necessary to remove logonui.exe. These tweaks should be done from ANOTHER windows.
Serge.
Sorry, gone now. Will be back tomorrow.
= nogui is necessary to remove logonui.exe =
- It's not so. logonui.exe is removed perfectly without nogui in boot.ini. (See My Tweak). Probably because logonui.exe it is not logongui.exe.
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Live, iPad
I was doing great sonically with the 1st part all the way up to
'.....Remove these files(Every single one is a jewel and all make great improvement)
Explorer.exe, sethc.exe...' and it was very good.
After that I did all the deletes and sonics were a little sharper therafter. Not real sharp but subtly sharp. Then I did the noguiboot switch in boot.ini and things got better. Probably still not as good as 1st part above but close.
Finally got to the end almost. There's probably a few straggler files I need to delete and I haven't tried nopae yet but whoa is this good. I finally deleted many of the files I was maintaining for sonic reasons just to try and I now believe they should stay out for sonic reasons. The major impact is a total richness to the sound, super clarity with no edge to the sound. As many of you know I have electrostatic speakers and had to update the panels a few months ago. New panels have more clarity than prior panels but because I cross them over @ 100 hz (which I now know was too low) I kept getting certain frequencies that would subtly zzz or buzz. Today I moved the x-over to a higher value (160-200 hz..not settled yet) and it snapped the highs into focus and solved my zzz problem. Once I settled on a higher volume for my mid bass drivers it filled the lower midbass suckout that resulted from lifting the panel crossover. I have always overlapped the panels and mid bass to get the tone right but not now. I cannot say enough good things about the highs from the CMP^2 all Stepped and Mihayloved out. It is precise, it images, it is smooth it is wonderful ...well done brother Serge's and major subcontractors like Jolida and Ryelands.
I have changed directions many times in set up. 1st it was 192 src, then 192 sox, a brief interlude at 96 sox, then finally back to 44. I used to detest cmp mode because I could not consistently get in, never knew if I was really in, so when Steppe 16 dumped cmp mode I was relieved. Not now I am fully commited to CMP mode. I have relinquished all my usb/thumb drive funtionality so getting new music will have to be via extraction of hd, filling up hd with files/cues and replacing hd back into CMP computer. A big pain in the a** but I will not ever get rid of these sonics. I cannot overstate how good this CMP^2 all Stepped out sounds. Best front end I have ever had.
Now for my question what register changes do we still have to make? Can we still do the autoruns thing for missing files? Did Steppe ever post his registry tricks?
.
No. You can.
Noguiboot works only in OS and does not influence in any way show BIOS.
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Live, iPad
Thanks Serge!
Please!
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Live, iPad
/NOGUIBOOT - Forbids VGA-driver usage at a stage of loading of OS - thus the Windows splash screen will not be displayed and at failures dark blue screens BSOD will not be displayed also. And how it can influence SQ?
/NOPAE - Forbids usage PAE on systems with ram more than 4Гб supporting PAE. Somebody here uses more than 4 GB ram?
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Live, iPad
I have a question about the Registration deletion from My Tweak then the restoring of R000000000007.clb (into folder Registration) My Tweak Corrected. Somewhere I lost this file but can retrieve from another pc. Do you think this would work?
I'm thinking maybe I need to restore this to get back into cmp mode. I never worried about it until Steppe 22 since my cplay worked ok without it.
= Do you think this would work? =
I don't know. R000000000007.clb is used by cPlay. But cPlay works w/o it.
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Live, iPad
Yes cplay works fine w/o it in my system. I guess restoring is worth a try then.
I'm back in the cmp saddle again. But (and remember I still have explorer.exe still in ) the explore button does not work. So there goes my idea of adding new music and/or deleting/adding back files via this method.
Hi (steppe) Serge
Thanks so much for the "endless" research and effort to make the most of software tweaking for cMP. By now there are so many chapters in the story (with LOTS of duplicate files listed) that it may be time to consolidate The Steppe Method into a single, optimized Path to The Mermaid's Tail .
Since you have rebuilt your cMP FIVE TIMES (yikes !), you must have a "big bang" recipe that can be used on the Classic cMP (as prescribed by cics on the website) - one that does all the right things, and avoids the pitfalls that break cMP. Please consider a new, "Complete, All-in-One" post.
And YES... please list and reveal ALL the reg tweaks we need to reach The Mermaid's Tail .
Am I correct thinking you've abandoned the "cPlay without cMP is best SQ" position... and therefore that part of STEPPE_16 should be ignored ?
Thanks again for leading the way.
Cheers,
Grant
That's not a Toy... IT'S A TOOL !!
I run Win 7 which does not have a boot.ini file.
Would anyone be kind enough to tell me how to set these parameters in Win 7 (assuming they yield SQ improvement of course)?
Thanks
Jonathan
You said'...After this comes the MOST IMPORTANT PART:
Open your cmp boot.ini file with a notepad and add /nopae and /noguiboot, next remove from windows\system32\ logonui.exe. Reboot and be prepared for surprises!...'I just cant resist this tease. You should be in advertising.
Now since this last tweak requires one to be in cmp mode I restored some files from the mihaylov tweak posts and your notes above
'...Their list + shutdown.exe and reg.exe...' then rebooted but it still boots in xp (which it should not). So simply stated do you (or does anyone) know what files may be missing from my xp?Also once in cmp mode it should be possible to delete these files from the EXPLORE button in cmp? I cant get this mode either so I conclude I am not really in cmp mode.
Anyway great post I'm just hoping I can get there.
edit: I think it was Rick who recommended reg.exe restoration. but I did recommend the shutdown.exe deletion.
Edits: 11/05/11 11/05/11
Try this .
I have decided to not use XPLite since it did not get along well with my nLite install so I am wondering if the above files are removed by IT since we have not had these on any of the lists of files to remove.
They seem useless for our purposes.
Just curious.
It would seem that XPLite is superfluous especially with screeenshots showing up.
Has anyone compiled a list of files the XPLite is removing?
Thanks,
These are the Screenshots of Folders in my cMp...
Junaid
Edits: 11/04/11
Hi Jolida,
Could you please clarify:
-- does USB still works after all the deletions, and
-- any Registry changes after all the file deletions; and if yes, how to do them.
KC
NO to both. I have deleted all the necessary files for the above to work. I can use a USB mouse though, but not a pen drive. Also Registry access is possible only by following the two methods I posted yesterday. Refer to my post "REGISTRY ACCESS" for info...
How does Ur system sound now??
Junaid
I think you had misunderstood my second question. I mean this:
-- if I had done all the deletion per your screen shots, do I have to do any corresponding Registry changes? and how do I do them?
Followup question to my previous question 1:
-- What files do I need to retain for USB to work (I have a USB dac)
As to my current status, I am happily listening to 4.6 (steppe 4 + jolida 1 & 2). But your screen shots prompted me to may be try tweaking again. If I can do everything in one step, then it is worth the effort.
KC
My screenshot is the result of steppes 2 to 16, along with two of my (Jolida's) lists, three lists of Mihaylov, Mihaylov's two screenshots & BF step 2 to 10...
I did all the above in about half n hour using my batch files, which includes registry as well...
USB is quite tricky. If u wanna play safe, then I suggest u run each Batch, then re-boot to check it's functionality. This way, u could pin down the list which may eat-up Ur USB driver.
The screenshots have no separate registry mods to be made. I hope I have sent u the batch files for all the above including the registry....
Always take a back-up of a stage which suits Ur taste... And always revert to that back up again later, to re-check if the mods done thereafter were worth it..
Junaid
Hi jolida,
No, I hadn't requested for your batch files previously. May be I would this time, in the case that I would go forward with it. I would if I am sure about the USB issue.
My impression is that not every step brings an improvement, at least not so from step 4.6 to steppe 8, which was as far as I went before I abandoned the whole thing.
KC
...an editable text file is much more so.
Thank you Junaid for all the hard work (and critical listening) you've contributed to the Software Tweaking Quest... it's helped many of us save hours of tedious effort.
While the screenshot approach to defining "what's left" has a certain visual familiarity, it is not that easy to use for a checklist. May I suggest an easy way to generate text lists for each of the relevant Windows folders: simply make a DIRectory listing of each folder (from the Command Line window) and save the result to an appropriately named ".txt" file. Zip up the text files, and they become a neat package defining what files you have left in your cMP system.
Open a command window in the folder you want to list and enter this command:
DIR /a > "descriptive file name.txt"
and hit ENTER.
The pipe command (ie: the "> " symbol, followed by a file name) redirects the screen output to your text file. You will get the file written into the same folder as the one you are listing, unless you give it a fully qualified pathname within the quotes.
To get a listing with filenames only (omitting any other information), add the "/b" switch to the command, like this:
DIR /a /b > "C:\Windows\another descriptive name.txt"
For a complete list of switches, type "help DIR" at the command prompt.
I hope this is useful... and Thanks again.
Cheers,
Grant
That's not a Toy... IT'S A TOOL !!
Are you back into cmp mode?
Absolutely. With the files in the screenshot I posted, I'm able to start in both cMp & xp modes. Do u have all the files in the screenshot available in Ur folders???
Junaid
Yes I do I should try again even though I tried a few hours ago.
A new reference point!
THANKS very much,
Let me know if u could...!!
Good Luck..
Junaid
Fantastic, you are a gentleman. This was just what I needed. Thank you.
U r welcome :)
Junaid
Can you gentlemen please post a list of WinSxs files and folders that you leave in the Windows folder? This is the only place where (plus the .nls's) where I have not had much luck eliminating stuff. I did some then tried to do some more and got a no boot. Before I try again I'd like a little direction from you guys. Also jolida I know you left some of this in for sonic reasons. Can you tell if it was all or some and if some which ones did you leave?
How many of them are there really? I see a first tweak and then a tweak#2. However, #2 has, as had been usual, a list of items to remove. But then it had a bunch of screen shots showing the state of his system which resulted from far far more extensive deletions. He makes no comment about the screenshots. They seem to reflect the result of a third tweak pursued by eliminating whatever is not in the screenshots. Is that right? Is this the intention? Why is there a list first,then, if one is working by what is not there in the screenshots? Makes no sense.
Anyway, I did Jolida's batch for tweak #2 and then tried achieving the screenshots. I got close but avoided winsxs further deletions and a very few other items I feared removing. I eliminated all the folders not shown everywhere and all the dlls and drivers not shown. Still I got nowhere near down to 56.3MB for windows. It was more like 570MB. I did't see where there was 500MB left that isn't in the screemshots. Can someone explain this.
Worse, I could not boot up getting a ntfs.sys is missing or corrupt message even after replacing it twice (it had been there all along). Can anyone clarify, explain or assist? (I am back to step 16 plus M1 and M2list only for now and liking it.)
Hey Riboge,
As I had mentioned earlier in a post last week, I started from the Original cMp, implementing the lists using my batch files in the following order:-
Original cMp
Steppe 2
Steppe 3
Steppe 4
Jolida 1
Jolida 2
Steppe 5
Steppe 6 ( omitting xplite )
Steppe 7
Steppe 8
Steppe 9
Steppe 10
Steppe 11
Steppe 12
Steppe 13
Steppe 14
Steppe 15
Steppe 16
Mihaylov's September list
Mihaylov's My Tweak
Mihaylov's Tweak 2
- - took a back up here - -
Screenshot 1
- - took a back up here - -
Screenshot 2
- - took a back up here - -
The size of the 'C' drive came down to about 68 mb here.
So just to check further, I tried Bold Fortunes batches here from it's Part 2 to Part 10.
Now the size of my 'C' drive is close to 58 mb after all the above done...
I took back ups at the above mentioned places because each of them changed the flavour of the sound from Airy to Mellow to Dynamic. So I felt its best to determine which suits best after completing the deletions & analyse them carefully with peace of mind.
It's important to do the Bold Fortunes sets with care. Especially in it's Part 5, where I did not run the Win_SxS batch, & in it's Part 10, where I had to edit his batch there, so as to NOT remove the System Folder. The Part 10 has a batch which removes the System folder, but we need it in our case where it contains winspool.drv & mouse.drv...
Junaid
What are screenshots 1 and 2? I see several screenshots shown and linked to in Mihaylov's tweaks #2 posting. Is this screenshots 2? You seem to take them as an implicit further tweak while others say they just show what resulted from all the steps to that point.There are screenshots posted earlier after M's first tweak about what is shown in process explorer with cMP running. Are you calling these screenshots 1? These are shown to explain adding back the list of files listed above them not as a step of further deletions to make.
Where is it recommended to eliminate Documents and settings altogether and all the folders but 5 in windows?
Can one do the whole thing by simply starting with full cMP and reproducing screenshots 2 all at once by deleting or cutting everything not shown? If not, why not? How can the sequence matter if this is where you end up regardless?
Edits: 11/03/11
Ooh la la. Now u r confusing me :)
Mihaylov... Please post the link to Ur screenshots once again. Both were uploaded by him to Rapidshare. Let alone all steppes, the final list of files PRESENT after all deletions in the system, is what is shown in the screenshot. It makes things a lot easier that way. All u have to do, is take a printout & go to the respective folder, drag the files that are present in his screenshot to one corner in a folder, & highlight & delete ALL others remaining. It's way easier that way because the list of files to be retained is much smaller than the one to be deleted...
Junaid
Edits: 11/03/11
Those are the ones with his second tweak. What is the other set of screenshots, since you list a 1 and a 2? I did the procedure you describe and ended unable to boot with error stating that ntfs.sys was missing or corrupt even though it was present. Can one do this right from the start and skip all the step by steps? Any idea why I got the unbootable error?
Oh yes. Absolutely !! It's way easier isn't it?? No matter on which steppe Ur system presently is in, u can follow this procedure & finish in no time. Just printout the screenshots, drag the contents present in it to one corner of a folder, & highlight-delete all others.
Please note that my system needed a few more files present in his screenshot, to boot or work flawlessly. They were a few more fonts, & sys files that I had to keep to make windows boot, Juli@ to work, & start in cMp mode...
U delete one wrong font, & windows will not boot....
Junaid
Great!
"system needed a few more files present in his screenshot, to boot or work flawlessly. They were a few more fonts, & sys files that I had to keep to make windows boot"
Which ones? I know about vgaoem font, for one.
Yes that one, & a few more. Will let u know once I'm home. Am driving at the moment :)
Junaid
Maybe I did not understand something but the Screenshots 2 shows the result of the package of the following tweaks have been applied to the classically optimized system cMP2: Steppe's 1 ... 16, tweak from here http://www.AudioAsylum.com/forums/pcaudio/messages/9/94530.html, XPLite Tweak, Jolida's 1 ... 2, My Tweak & My Tweak 2.
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Live, iPad
Edits: 11/02/11
Where in that list of tweaks does it say to delete the entire Documents and Settings folder? Where does it say to delete all the folders under Windows except for those 5 in your windows screenshot? Etc, etc? I have once again done all the tweaks you list save for the list you surprisingly bring up from September before any of the Steppes and still windows folder is way over 1GB. The folder system32 in your screenshot has dozens and dozens fewer elements than result from the combined tweaks listed. I must really have missed something and remain no less confused.
= Where in that list of tweaks does it say to delete the entire Documents and Settings folder? =
- Here: http://www.audioasylum.com/forums/pcaudio/messages/9/96990.html
(quote:
It is necessary to remove following files and folders.
From the directory C and subfolders:
3com_dmi
1025
1028
1031
1033
1037
1041
1042
1054
2052
3076
Common Files
dhcp
Documents and Settings
DRVSTORE
export
IME
inetsrv
ras
Restore)
= Where does it say to delete all the folders under Windows except for those 5 in your windows screenshot? =
- here: http://www.audioasylum.com/forums/pcaudio/messages/9/96990.html
(quote:
...From the folder Windows:
...
addins
AppPatch
Config
Connection Wizard
Debug
explorer
help
ime
Installer
java
msapps
pchealth
PIF
Registration
srchasst
Temp)
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Live, iPad
I was using the Jolida batch file for your tweak which does not deal with the folder deletions you listed. Now it makes sense to me at last.
I still wonder whether one can do it all from the start at once now that we know the end point. When I tried after 16 steps and the batches for your tweaks I couldn't boot up but got ntfs.sys missing when it wasn't.
Try to execute My Tweaks manually and gradually deleting folders and files (by groups) to define that file or a folder after which removal system it ceases to be loaded (if the such occurs to be).
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Live, iPad
I eliminated windows folders not shown> okay. Then, I eliminated folders from system32> okay. Then I eliminated the drivers not shown in drivers folder> no boot saying isapnp.sys was missing. I replaced that then no boot saying ntfs.sys missing. I am stuck here trying to identify which driver(s) must be replaced--no obvious choices. Suggestions?Btw, I never had LtcycfgWMp.sys or mv2.sys or rtenicxp.sys
At this point windows is 315MB
Edits: 11/03/11
= Then I eliminated the drivers not shown in drivers folder> no boot saying isapnp.sys was missing. I replaced that... =
- Here the screenshot of the Driver folder in my OS (from package "Screenshots 2"):
As it is possible to see the file isapnp.sys is present (is shown!). Tell me please why you moved it and where?
The file ntfs.sys is shown too.= Btw, I never had LtcycfgWMp.sys or mv2.sys or rtenicxp.sys =
- Because you do not use the network configuration of cMP2. LtcycfgWMp.sys is driver for PCI Latency Tool, mv2.sys is driver for UVNC server and rtenicxp.sys is Realtek network driver.
Show the screenshot of your Drivers folder after elimination because I cannot understand in any way that you delete and that you leave.
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Live, iPad
Edits: 11/03/11
Sorry if I wasn't clear. I removed isapnp.sys by mistake then returned it. ntfs.sys was present all along despite the error message. When I returned all the drivers it worked again, so I need to identify which driver removal of those not shown caused the problem. It is not one you listed to stay.
I do in fact use the network configuration of cMP but not the latency tool or a Realtek network driver. Mine is Intel on my Lenovo laptop.
Later I did the removal of files from system32 and again couldn't complete booting getting an "error c0000135 unable to locate component, application cannot find winsrv" despite winsrv.dll being still present. I returned first the .exe files> no improvement. Then I returned the .sys and .nls files> no error but boot stalled after showing the usual windowsxp screen with progress bar. Only when I returned the .dlls did full boot get restored. So, in toto, I need to find a driver in the driver folder and a .dll in the system32 folder that I have to keep in addition to what your screenshots show...apparently.
In my system the contents of the winsxs folder are entirely different from what you show, that is, none of the ones you have in your screenshot are in my winsxs at any point.
Well. Now as you have understood you have to find the necessary dll and sys. This is what I was doing when I was searching for files that can be removed. The easiest way to do it by successive bisection: return first half of all deleted files dll and check the system boot. If the system does not boot then remove this half and return the second half. If your system boots then again remove half of the recovered files and check the system boot, etc.
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Live, iPad
Edits: 11/03/11
I must confirm the riboge issue: me too can see about 250MB checking Windows folder.I'm not worried because sound is gorgeous but my curiosity is the same of riboge.
I've performed step 1-16 (plus XPlite) +jolida 1-2 + Mihaylov 1-2 ( with winsxs and .nls delections but fonts untouched.
Can you post number of files in windows, system32 and drivers along with size in mb?
One more thing are you sure you're getting properties of Windows and not C drive? My c drive is still over 1 gb.If you are do a properties read on Windows and all its sub folders then a read on system32 and its drivers tell us how many files and how much mb?
Unfortunately right now I have overriden that with a copy of just before pursuing the screenshots. I do recall that I did properties on windows folder to get this number and on system32. The latter was around 275MB. I kept a copy so I can look thru it later again.
But why do steppe 17-20 then M1 and M2 if you can at step 16 just work from the screenshots by eliminating all not shown? Also, Mihaylov presented the screenshots as part of M2 (?) well before steppe came out with 17-21. That's what is puzzling, the sequence of offering them and how it should relate to the sequence of performing deletions and whether one is intended to do both the latter steppes and M tweaks or whether they are alternative methods. Please note, I am working on a Lenovo laptop not a custom cMP rig. The bios changes are not possible for me.
Also, I am hoping Jolida will do batch files for these last steppe steps.
I personally do not use the screen shots because I cant d/l or want to pay to d/l. I think the screen shots are just a double check on the deletions. But anyway I did Steppes 1-20 plus My Tweaks 1 and 2 and I get down to about 80+ mb. I left all the .nls files and fonts because I had so much trouble with them initially. Are you sure its 570 and not 57 mb?
I must share a surprising improvement in sound quality. I run USB from CMP machine to the DAC. Quite simply, I built an improved USB cable. Impact was as big as going from Step4 to 16. Litterally hearing stuff for the first time. Three features seemed contribute.
1. replaced standard wire with 2 runs of CATV twisted pair
2. made it very short 6"
3. shielded each of the 2 pairs with aluminum foil. This had big impact.
With this in place I attacked a short piece of I2S cable in the same manner. Positive impact but not on the same scale.
Clarity, sound stage, balance all significantly improved. Got inspired by the LessLoss site. They claim cable length and shielding very important on all digital signal lines before the DAC chip to reduce HF induced jitter. I think they are right.
Thanks for sharing your experience.
I'm not surprised. I've found very few situations where a shorter connection wasn't better... most times markedly. And digitally-encoded music signals seems to be even moreso in this regard.
There's a lot of blather about USB cables on the PC Audio Asylum. I posted this response to one thread... the experience related there was very similar to yours:
Http://www.audioasylum.com/forums/pcaudio/messages/9/96004.html
Greg in Mississippi
Everything matters!
Thanks for posting ling to these pics Greg. I had forgotten about Peter's post. It was infact his work that inspired me initially to build my own usb cable. I was running something about a foot long. The initial change this week was to shield the power pair separately from the signal pair with ordinary kitchen tin foil. Shocking improvement. Next I took it down to < 6". Modest improvement. I think as a result the deal is to reduce HF induced jitter. The lessloss guys are building stacks of chips direct to chips to eliminate cable all together. much as Peter does with his clock/2706/1543/cap stack on his latest dac experiments. These pics reminded me that I should eliminate the connector at the dac end and just leave one to connect to the PC.
Another change that brought crazy improvement was eliminating even the 3 or 4 mm of leads on caps by going to surface mount polymer decoupling caps directly on the legs of the 1541 dac chip per ecdesign. I think every mm counts in digitally induced jitter on signal path.
Antenna length is only about an inch in the 2.4 ghz wifi band. We all have 5.8ghz cordless phones around our rigs. Even shorter antenna. All this stuff is injecting substantial jitter into our highly exposed cmp systems. I have now proven that even wrapping the digital signal wires in a little foil make a difference if you cannot eliminate them all together.
I2S is better than other signal transports because it carries separate clock signals. Still shield it from the antenna factor.
At this stage, where there seems nothing more left to delete or change, & windows sits around 50mb, arises another question. I have been using a cordless mouse throughout this project, & now it seems like it is left alone without siblings. The system does not accept any other mouse, be it USB (corded or cordless) as well as ps2. I guess the HID service had been stopped before services were deleted along the route. What if this mouse stops working at a later date? There seems no chance at that stage but to re-build with the default settings & re-do all the optimisations. Has anybody any experience about this???
Junaid
to settle on which mouse early on.
In my experience there is nothing you can do, especially now.
Consider starting over with the PS2.
To really go for broke once all is settled and BIOS optimized you can delete USB in BIOS!
Rick, I'm unable to understand the last line u wrote..
Can u explain what it meant..?
Junaid
all of the USB entries.
No polling worries after that!
I have tried it (at Grant's suggestion (cgwaudio)) and though subtle I thought it was worth the trouble.
Of course, it could make for pain in the event of a power failure! But without pain how can one call themselves an audiophile? As someone remarked at the beginning of the CD era how audiophiles would find a way to make playing CD's as inconvenient as playing LPs (properly) and I think we can say we have achieved this goal/prophesy. Exceeded it, actually!
Have you tried Serge's BIOS regimen?
Hi all
my experience is the same of jolida but I can add an interestin "USB memory" effect. After step 7, i cannot remember at which stage, we lost USB use. If you, like me, use the batch steps keeping connect one HDD or two, and the mouse/keyboard wifi thumb, they will continue to work until the end of deletions without any problem but only and solely in that precise USB port where they are connected. So you'll can decide of disconnect and reconnect the device but only in his old position.
Interesting and useful, I think
But are u able to use a different mouse other than the one u have been using all this way???? My concern is not with the port, but with the mouse itself. Will we be able to get a new mouse to work, just in case the present one gets damaged???
Junaid
No, I' don't think, only an identical model of mouse.
Otherwise you must redo everything from the start.
Bless Acronis ( or other backup and restore system) and your batch files!
I'll try, I get some different type of wifi mouse. It could be easy to try if they works or not
Bought an exact model of the mouse I'm using to test it. No luck. Doesn't work. I presume we need to enable the HID service again to recognise it once. To do that, Services.msc needs to work, which again demands restoring files that support it..
Junaid
Edits: 11/01/11 11/01/11
thought my MB had died, went so far as to order another one (now I have a backup) and just for "fun" I put in the other precious stick of 512 mB I possess and everything worked fine.
Had not done anything unusual. Just would not get to the BIOS screen.
Any ideas? Don't want it to happen again.
Any ideas? Don't want it to happen again.
Set your BIOS back to the default settings and try again. If, under these circs, Stick 1 runs and Stick 2 doesn't, fair bet that Stick 2 is dud. If OTOH Stick 2 runs, you've maybe been "under-clocking" more than that particular device warms to.
when this happened I could not get to BIOS to change anything. Just sat there with those lights blinking on the board - they would stay on about 15 seconds and then go off, come back on for another 15 seconds ad infinitum.
Using only one. Luckily I had another.
Tonight I was listening to music and it just stopped. Tried to restart, same thing as with the bad stick. Turned off all power, took it out and re-inserted and all was fine. Makes me think the "bad: one might not be bad.
Thanks, as always,
Wasn't doing anything extreme in BIOS - standard recommended settings.
Tried STEPPE's BIOS settings later on and had no problem getting that to work. A mystery ...
Because there are problems with functioning of usb ports on tweaking cMP2 why not to use external eSATA HDD?
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Live, iPad
Edits: 10/29/11
Why not use external eSATA HDD?
A cMP2 box I built for a friend uses an eSATA HDD with an OS and a data partition. It works just fine. My good idea was that it would be easy to replace the stock HDD PSU with something better, my bad one that I would have to make it. As yet, I haven't - but the eSATA interface is (as you'd expect) no problem.
Another idea at the back of my mind is to try booting and pulling data from a USB HDD connected via one of those ADuM4160 USB isolator chips. Slow speed and all that but not outrageously expensive and plenty fast enough for audio and, with slimmed-down OS installs, sensible boot times.
Hopefully.
Dave
So does this device allow me to go from usb (say from my docking station) to a sata input on my hd? If so that would fix my problems as long as I could cut and paste(which I think I now know how to do)?
does this device allow me to go from usb . . . to a sata input on my hd?
No. I'm not clear why you might want to but you can connect a docking station c/w eSATA interface (most newer devices have one) either to an eSATA interface on the motherboard or (slightly inferior but most unlikely to matter in this context) a SATA> eSATA adapter on the back panel:
On that system I mentioned, I connected an external boot/data drive with its own PSU via one of these adapters. Functionally, it's an internal drive but it's not connected to the mobo PSU and is open to improvement. Some day.
You said '...No. I'm not clear why you might want to ...'
I'm just trying to get new music into my hdd's. So unless I figure out how to reactivate thumb drives and to cut & paste from them I don't have a way unless I swap a a sata hdd to my home computer and then swap it back into cmp. Maybe one of your above mentioned adapters will work. I'm only looking for a way to easily get new music data into cmp.
So I assume my Thermaltake docking station is a type of esata drive. So now that I can copy or cut & paste I could load my hdd's that way. Will try in the am.
Thanks Dave and Serge.
Did not work...oh well back to restoring more files.
I have bought a Fit-PC2 Ultra. I have (licensed ) copies of both Win 7 & Win XP SP2.
The Fit-PC2 is to be used exclusively for playing music with CMP2 to a USB DAC.
I want to modify Windows to minimise its footprint and will be following the most helpful file removals on this forum.
I wondered if anyone had done back to back comparisons of the two OS similarly cut down using CMP2 and whether there is a preference for either for best SQ?
Many thanks
Jonathan
I have bought a Fit-PC2 Ultra.
Wise choice! I've been using the rather slower 1GHz, single-core Fit-PC2 for over two years now. Though it can be a bit sluggish at browsing, it never falters when playing and I've yet to find anything better sound-quality wise. Note that the Atom cannot upsample to 192KHz using cPlay.
I can't say whether Win 7 beats XP using stock drivers but suspect that, if your DAC has bespoke ones or you use the likes of the AQVOX, the difference is negligible.
One point you may want to bear in mind is that the user base (and thus the number of people who can offer support on this forum) for cMP/XP setups is much wider than for cMP/Win7 ones.
HTH
Dave
Thanks Dave. Sensible advice. I think I will stick with SP2 XP.
I am using a M2Tech Evo based DAC which does have proprietary Windows drivers. Will that work OK with CPlay?
Jonathan
Will [the M2Tech Evo] work with cPlayIf its driver is ASIO-compatible, almost certainly Yes. If for some reason it doesn't work with cPlay, you can always use a player such as Foobar. Not quite as good as cPlay but closer IMHO than some suggest.
Dave
Edits: 10/29/11 10/29/11 10/29/11
= I am using a M2Tech Evo based DAC which does have proprietary Windows drivers. Will that work OK with CPlay? =
- No. The drivers M2Tech Evo do not support ASIO. cPlay does not define M2Tech Evo with drivers ASIO4ALL also.
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Live, iPad
Damnit!
Would the new ASIO2WASAPI driver from levm also not work?
I did not try driver ASIO2WASAPI. I heard for the first time about it from you now. Unfortunately I already gave M2Tech Evo therefore check up its operation with this driver I can not.
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Live, iPad
Serge - I am pleased to say that the ASIO2WASAPI driver works very well with CPlay on my M2Tech Evo USB interface.
Many thanks levm.
Jonathan
Thank You for the good news.
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Live, iPad
For those who have lost thumb drive functionality after deletions, look up for the following files in the folders where u moved them to, & restore them back ( this can also be done temporarily till u load music from the thumb drive & the files can then be deleted again if u feel they effect the Sonics )...
Anything that starts with 'hid'
Anything starting with 'usb'
And the file which was deleted from c:\windows\inf\INFCACHE.1
After completing almost all the file deletions, I am still able to use my thumb drive exactly like how it was with a clean install of xp. Please note that I had also not deleted any file or entry that had to do with 'smartcard'. I have no idea if it has anything to do with USB, but I preferred not to take chances , & reserve it's removal in the End if needed...
Junaid
There must be different versions of xp. I finally got my way back to cut & paste and having cmp recognize thumb drives.Requirements for cut and paste : rpcns4, rpcss, rsfsaps,rshx32,rsmps all dll's. It could be this subset is smaller I just haven't narrowed it further.
Requirements for recognizing thumb drives: wshatm.dll, wshbth.dll, fs_rec.sys, mcd.sys.
I left the inf folder deleted and all works fine.
Go figure maybe the files can be further narrowed but I'm done after looking for a solution the old fashioned way ... hard work(restoring files 5-7 at a time to see if it fixes the problem).
The last few days demonstrated what functions I need and it isn't keyboard. It mouse, cut & paste and recognizing thumb drives.
Edit: changed file name for thumb drive functionality correct file is fs_rec not imapi....
Edits: 10/31/11 10/31/11
A friend of mine is Still unable to use pen drive even after restoring those 4 files u specified. Are they cumulatively related to the files u have chosen for 'cut n paste' to work, or are they independent of each other?? Have u restored the 4 files after restoring the files for cut-n-paste??
Junaid
I did the cut and paste 1st then the thumb drive.
Thumb drive Still does not work even after restoring all files..
Junaid
Your restorations didn't work either?
No. He still gets the "Disk not formatted" error..!!
Junaid
One thing I did do cut/paste files then the wshatm and wshbth then lastly the 2 files Fs_rec and mcd as a set. Rebooting after each set of deletions.
Dont know if it makes a differnce but there it is.
What level of other deletions has he done (e.g., Steppe 1-20, My Tweak 1...)?
Yes he is done with all deletions. He could cut 'n' paste but thumb drive doesn't function. It shows up in My Computer but Not formatted error as usual..!!
Junaid
Did he reboot?
When I restored these files I just put them into Windows not back into a folder called inf. Should I try that instead? Is that how you did it?
Re-create the c:\ windows\inf folder & put back the INFCACHE.1 as well as everything to do with USB..
Junaid
still does not work. Since I lost this function in My Tweak page 5 I'm going to try to restore those files. Think there was only 11 or so.
Hope u get it to work asap...
How's the sound at this stage ??
Junaid
I restored all the filoes (17 of them) from page 5 of My Tweak and I can now cut & paste ok but cmp does not recognize thumb drive. Still get a 'Do you want to format the disk' error message.
even though I restored the files mentioned...sonics are still very good.
Ok I narrowed down the cut & paste function to a small group of dll's: rpcn4,rpcss, rsfaps, rshx32. When I put these files back in cmp boots faster and cut & paste is restored. The process I used was restore 5-7 files reboot see if I fix the problem and it worked well for the cut & paste function. However when I used this process to find why I could not get thumb drives or my esata docking station to work I ran into weird results.I restored 8 drivers from my temp folder beginning with wshatm through drmk and I got a no boot. Put acpiec through drmk.sys back into the temp folder still no boot. So I restored wshatm, wshbth and so since this was where I was before it should boot ....but still no boot ...very weird. All I can figure is that is that some interaction between rpcn4... and wshatm...is going on or the order of file deletes is important or I goofed up somewhere. . Anyway I restored my last good os disc file for all the tweaks through My tweak 2 and I'm back in music
but no cut & paste and no thumb drive functionality but it sounds great.
Edits: 10/30/11 10/30/11 10/30/11 10/30/11
And the file which was deleted from c:\windows\inf\INFCACHE.1
Question do you remember either the name of the file in inf\INFCACHE.1 or in which post it was described (ie My tweak or My Tweak2...?
Edits: 10/29/11
I'm not too sure which steppe it was, but I vaguely remember it being mentioned in the last set of steppes. We were told to delete the entire inf folder, which actually had the motor to make the thumb drive run. If u have Moved the files elsewhere & not deleted them permanently, then it's quite easy to get hold of it... I think it should be steppe 8.
Also reading this page might be useful:-http://www.uwe-sieber.de/usbtrouble_e.html
and
http://forums.cnet.com/7723-6142_102-276502.html
Junaid
Edits: 10/29/11 10/29/11
The inf stuff is in the original My Tweak stuff right near the front. So anyway I put back inf cache stuff but also found some smartcard stuff in inf & I put that back but I still get the disk not formatted response. I did find some usb stuff in the inf folder so I'll try restoring that. Did you restore these as well?Remember to try cut & paste also.
Unfortunately even with the usb files from inf folder restored I still get disc not formatted error message.
edit: I restored hid***.inf files still no luck
Edits: 10/29/11 10/29/11 10/29/11 10/29/11 10/29/11
Annoyed by the results after steppe 16, & reverting back to steppe 5, & finally back to where it all started (original cMp), I was all set to discover the best combination out of all that we had accumulated so far. And then Mihaylov posted his tweak. Since I had implemented all steppes, I thought why not give it a try, afterall nothing to lose since I had a back up handy. As posted earlier, I found it to be a very good step heading in the right direction. Inspired with the result I began stepping ahead with the Original cMp in the following order ( continuosly without a break or audition in between, by using the Batch files)...
Steppe 2
Steppe 3
Steppe 4
Jolida 1
Jolida 2
Steppe 5
Steppe 6 ( I skipped xplite)
Steppe 7
Steppe 8
Steppe 9
Steppe 10
Steppe 11
Steppe 12
Steppe 13
Steppe 14
Steppe 15
Steppe 16
Bold Fortune Registry changes only
Mihaylov's first tweak posted in September
Mihaylov's My Tweak
Mihaylov's Tweak 2
Screenshot 1
Screenshot 2.
There were a lot of changes to be made in many of the above posted lists, to make boot-ups go smooth, get Juli@ to work, get cicsremote to work , & restart in cMp mode. After making all the corrections, everything seemed to work perfectly well.
End Result :- unbelievable Sonics, excellent liquidity & dynamic range, very good blend & spacing between instruments, pleasantly laid-back & non-fatiguing.
Only after reaching this stage, I could sense what a cMp is capable of doing, & where it stands in comparison to awfully expensive transports....
Ofcourse there are short-comings as well, but there is still a lot of room for improvement at a hardware level, which tentatively will unleash what's been veiled still..!!
Worth it.. Absolutely !!!!
Junaid
You inspired me to go all the way.
I followed your path to steppe7. Then did the complete Bold Fortune. Really like the sound at that stage. Took a system image after ever step so comparison is easy.
Then went up to steppe10. Took a defrag while I still could. Then applied 11 thru 16 (just the first safe part). Started & restarted CMP and then applied the 3 Mihaylov lists file deletions from exporer within cmp. I forgot to do the folder deletions (will try that in future) Also from explorer entered Autoruns and deleted each and ever registry entry that had file not found.
Then reboot back into cmp.
Music very good. Have not had time for serious assessment, but after 30 mins it is clear it is in another league.
Thanks for your batch files and the inspiration.
Cool. Good to hear that :)
Do the screenshot 1 as well. It changes the flavour of the sound. I liked it though...
Junaid
Congrats on your final end result...lots of hard work there I'm sure. Can you explain a little about screenshots I neverdid d/l them. What are they? What do they do, how do they sound?
The screenshots which Mihaylov posted. Shows the remaining files/folders after deletions. Helped a lot in the process BUT must be done carefully. Especially with fonts. My cMp needed more fonts than on his screenshot. Also Juli@ refused to play without it's needed sys files. Rest can be replicated though..!!! I still have some more of it to be done. But I will first take the back up & then proceed. Autoruns too is way risky unless one is 100% sure what's to be removed. Else, a Blue Screen..!!!
Junaid
Which makes me wonder if, when it was working, it would REPORT but not be able to AFFECT?
It is possible.
I was deleting stuff like mad!
Absolutely. One wrong thing, & it takes u to the bsod. Unless u are very sure what to remove from it, stay away..!! It's best done after taking a back up after all file deletions so that u know which entry in it is causing the blue screen, to pin it down in a safe zone..
Junaid
thanks
Any way to find the last source code?
Is this an open source project?
Who has cicsPlay 2.0b ~ 2.0b7 version?
I need cicsPlay for Pentium3 version.
I work a old Pentium3(sse1) notebook for music-center...
If who have cicsPlay for Pentium3 version, Please send me to mail...
Thank for your kindness..^^
I GET cplay 2.0b1 to 2.0b7 for PENTIUM3 OR SSE1 from Dave BruceI uploaded these files my blog
-> http://blog.naver.com/examanida/100141805852
if you need these files, come my blog and get these ^^
thank for Dave Bruce's kindness
Edits: 10/28/11
Thanks!
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Live, iPad
I want to offer the My Tweak 2 in addition to My Tweak. My Tweak 2 include the files and folders from Steppe's Tweaks parts 17-20 missing in My Tweak and applicable for my network configuration and also more files.
It is necessary to remove following files and folders:
x86_policy.5.1.Microsoft.Windows.SystemCompatible_6595b64144ccf1df_x-ww_a0111510
x86_policy.7.0.Microsoft.Windows.CPlusPlusRuntime_6595b64144ccf1df_x-ww_a317e4b3
$winnt$.inf
12520437.cpx
12520850.cpx
activeds.tlb
alg.exe
amcompat.tlb
AUTOEXEC.NT
bios1.rom
bios4.rom
bootstat.dat
cdplayer.exe.manifest
chkntfs.exe
command.com
CONFIG.NT
CONFIG.TMP
dcomcnfg.exe
desktop.ini
dssec.dat
dwintl.dll
ega.cpi
emptyregdb.dat
esentprf.hxx
graphics.pro
h323log.txt
himem.sys
ipxroute.exe
JulaPAN.exe
key01.sys
krnl386.exe
login.cmd
logonui.exe.manifest
mqoa.tlb
mqoa10.tlb
mqoa20.tlb
mqperf.ini
mqprfsym.h
mqtgsvc.exe
mscdexnt.exe
msdatsrc.tlb
msdtc.exe
msdtcprf.h
msdtcprf.ini
msg.exe
ncpa.cpl.manifest
net1.exe
netsh.exe
nscompat.tlb
ntimage.gif
ntio.sys
ntio404.sys
ntio411.sys
ntio412.sys
ntio804.sys
nwc.cpl.manifest
oembios.sig
pagefileconfig.vbs
perfc019.dat
perfd019.dat
perfh019.dat
perfi019.dat
pschdcnt.h
qappsrv.exe
rasctrnm.h
rasdial.exe
remotesp.tsp
reset.exe
rsaci.rat
rsvpcnts.h
sapi.cpl.manifest
secupd.sig
sfc_os.dll.orig
shadow.exe
stdole2.tlb
stdole32.tlb
tcpmon.ini
termcap
tskill.exe
tslabels.h
tslabels.ini
unimdm.tsp
v7vga.rom
wdl.trm
WindowsLogon.manifest
wpa.dbl
x86_Microsoft.Windows.SystemCompatible_6595b64144ccf1df_5.1.0.0_x-ww_fc342b0b.cat
x86_Microsoft.Windows.SystemCompatible_6595b64144ccf1df_5.1.0.0_x-ww_fc342b0b.Manifest
x86_Microsoft.Windows.SystemCompatible_6595b64144ccf1df_5.1.2600.2000_x-ww_bcc9a281.cat
x86_Microsoft.Windows.SystemCompatible_6595b64144ccf1df_5.1.2600.2000_x-ww_bcc9a281.Manifest
See all screenshots below.
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Live, iPad
There are a few exceptions. I left in usb stuff thinking I could load music via my thumb drive which I lost anyway for some other reason. Also left in some files per Jolida for sonic reasons...doesn't make sense but it works. Left in all WinSxs & I think taskman and a couple of others. Don't know how to load music other than taking out hdd's from cmp and loading them on my home pc but I am not going back.Hats off to Serge and Serge...these tweaks make magnificient music...beautiful mermaid indeed.
Edits: 10/27/11
Hi Mihaylov,
I would like to bring to your notice, that when following Ur screenshots, one has to be a bit careful in exactly replicating it. Mine needed more fonts than u have in Ur screenshot, & deleting them caused windows not to boot. Also I'm surprised how u r able to get Juli@ to work without ks.sys, drmk.sys & portclys.sys, as they are not in Ur Drivers folder. My Juli@ wouldn't work on deleting any one of these files...
Junaid
My Juli (I use only digital part) works fine w/o ks.sys, drmk.sys & portclys.sys.
The list of used drivers at work of cMP and cPlay shown in the screenshot here http://www.AudioAsylum.com/forums/pcaudio/messages/9/97337.html (this screenshot was obtained in the system before removing any files).The files you specify is not listed. They are not used for playing audio so they are not needed. I do not understand the causes of your problems.
About fonts. To keep only those files in the Fonts folder are shown in the corresponding screenshot here http://www.AudioAsylum.com/forums/pcaudio/messages/9/96990.html. But replace the Russian files 8514fixr, 8514oemr, 8514sysr by the English files (if you use the English version of OS) 8514fixe, 8514oeme, 8514syse. I have already written about this.
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Live, iPad
Well well. That's quite strange. I too am using only the digital part of my Juli@, but removing any one of those files doesn't get it to work. Also, I did exactly the same with regards to fonts (replaced the 'r' with the ones with 'e') but my system refused to boot without vgaoem, & also did not allow me to delete few other font files as well.
Really strange though :)
Junaid
I do not know what to say. It is a pity :(.
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Live, iPad
I have found this to be the way to go after you lose your start button or RUN entirely.
Just double click on any entry and it takes you to REGISTRY EDITOR. Of course, depending on where you click will take you to THAT part of the registry but easy enough to find your way.
Of course, without the keyboard you can only delete entries. Easy enough to put the keyboard back if it is needed.
AUTORUNS is really a great tool. Thanks to MIHAYLOV for reminding us of this.
I have not deleted much of 13 and beyond, though I did go ahead and delete the files of MIHAYLOV #2 (except for the ones he said to not delete and the two drivers the PS2 port needs).
Did some folder re-arranging with the BOLD FORTUNE batch files.
Tried to run AUTORUNS and it cannot find it or something to that effect.
Anyone know which files AUTORUNS needs?
Thanks,
Hi Rick,
I checked with proces explorer.
In my setup autoruns.exe uses these files:
Look from this line: "Process: autoruns.exe Pid: 1444" and beyond.
Process PID CPU Private Bytes Working Set Description Company Name
System Idle Process 0 97.01 0 K 16 K
System 4 1.49 0 K 208 K
Interrupts n/a < 0.01 0 K 0 K Hardware Interrupts and DPCs
smss.exe 404 164 K 364 K Windows NT Session Manager Microsoft Corporation
csrss.exe 452 0.75 1.428 K 2.784 K Client Server Runtime Process Microsoft Corporation
winlogon.exe 476 5.680 K 1.660 K Windows NT Logon Application Microsoft Corporation
services.exe 520 1.604 K 3.308 K Services and Controller app Microsoft Corporation
svchost.exe 720 1.424 K 3.596 K Generic Host Process for Win32 Services Microsoft Corporation
svchost.exe 816 5.376 K 6.860 K Generic Host Process for Win32 Services Microsoft Corporation
winvnc4.exe 852 836 K 2.708 K VNC Server Free Edition for Win32 RealVNC Ltd.
svchost.exe 1048 2.556 K 3.788 K Generic Host Process for Win32 Services Microsoft Corporation
lsass.exe 532 1.836 K 748 K LSA Shell (Export Version) Microsoft Corporation
explorer.exe 1076 5.652 K 9.540 K Windows Explorer Microsoft Corporation
procexp.exe 1388 6.720 K 8.220 K Sysinternals Process Explorer Sysinternals - www.sysinternals.com
autoruns.exe 1444 0.75 12.188 K 12.024 K Autostart program viewer Sysinternals - www.sysinternals.com
Process: autoruns.exe Pid: 1444
Name Description Company Name Version
advapi32.dll Advanced Windows 32 Base API Microsoft Corporation 5.1.2600.2180
apphelp.dll Application Compatibility Client Library Microsoft Corporation 5.1.2600.2180
autoruns.exe Autostart program viewer Sysinternals - www.sysinternals.com 9.35.0.0
clbcatq.dll Microsoft Corporation 2001.12.4414.258
comctl32.dll User Experience Controls Library Microsoft Corporation 6.0.2900.2180
comdlg32.dll Common Dialogs DLL Microsoft Corporation 6.0.2900.2180
comres.dll Microsoft Corporation 2001.12.4414.258
crypt32.dll Crypto API32 Microsoft Corporation 5.131.2600.2180
cryptnet.dll Crypto Network Related API Microsoft Corporation 5.131.2600.2180
ctype.nls
dnsapi.dll DNS Client API DLL Microsoft Corporation 5.1.2600.2180
gdi32.dll GDI Client DLL Microsoft Corporation 5.1.2600.2180
imagehlp.dll Windows NT Image Helper Microsoft Corporation 5.1.2600.2180
kernel32.dll Windows NT BASE API Client DLL Microsoft Corporation 5.1.2600.2180
locale.nls
mpr.dll Multiple Provider Router DLL Microsoft Corporation 5.1.2600.2180
msasn1.dll ASN.1 Runtime APIs Microsoft Corporation 5.1.2600.2180
mstask.dll Task Scheduler interface DLL Microsoft Corporation 5.1.2600.2180
msvcrt.dll Windows NT CRT DLL Microsoft Corporation 7.0.2600.2180
netapi32.dll Net Win32 API DLL Microsoft Corporation 5.1.2600.2180
ntdll.dll NT Layer DLL Microsoft Corporation 5.1.2600.2180
ntdsapi.dll NT5DS Microsoft Corporation 5.1.2600.2180
ole32.dll Microsoft OLE for Windows Microsoft Corporation 5.1.2600.2180
oleaut32.dll Microsoft Corporation 5.1.2600.2180
R000000000007.clb
rpcrt4.dll Remote Procedure Call Runtime Microsoft Corporation 5.1.2600.2180
rsaenh.dll Microsoft Enhanced Cryptographic Provider Microsoft Corporation 5.1.2600.2161
secur32.dll Security Support Provider Interface Microsoft Corporation 5.1.2600.2180
sensapi.dll SENS Connectivity API DLL Microsoft Corporation 5.1.2600.2180
setupapi.dll Windows Setup API Microsoft Corporation 5.1.2600.2180
shell32.dll Windows Shell Common Dll Microsoft Corporation 6.0.2900.2180
shlwapi.dll Shell Light-weight Utility Library Microsoft Corporation 6.0.2900.2180
sortkey.nls
sorttbls.nls
unicode.nls
user32.dll Windows XP USER API Client DLL Microsoft Corporation 5.1.2600.2180
userenv.dll Userenv Microsoft Corporation 5.1.2600.2180
uxtheme.dll Microsoft UxTheme Library Microsoft Corporation 6.0.2900.2180
version.dll Version Checking and File Installation Libraries Microsoft Corporation 5.1.2600.2180
winhttp.dll Windows HTTP Services Microsoft Corporation 5.1.2600.2180
wintrust.dll Microsoft Trust Verification APIs Microsoft Corporation 5.131.2600.2180
wldap32.dll Win32 LDAP API DLL Microsoft Corporation 5.1.2600.2180
ws2_32.dll Windows Socket 2.0 32-Bit DLL Microsoft Corporation 5.1.2600.2180
ws2help.dll Windows Socket 2.0 Helper for Windows NT Microsoft Corporation 5.1.2600.2180
xpsp2res.dll Service Pack 2 Messages Microsoft Corporation 5.1.2600.2180
Mark
fully optimized cMP2 PC -> ESI Juli@ -> Van den Hul Optocoupler MkII-> Lavry Black DA10 -> XLR Mogami Gold -> Klein & Hummel O300
I wrote a note of thanks this morning and for some reason I see it is not there.
PROCESS EXPLORER is a tool I need to use AGAIN!
Funny thing is I am pretty sure all of those files are still in my OS. Will have to check when I get home.
Thanks, again, for the information and help,
Hi Rick,
I can only agree: autoruns and process explorer should have be used more. Also during the development of all the ‘Steppe tweaks’.
I still have a hard time in understanding how sound quality can be improved through deleting files that are not in use and/or are not going to be used.
Using autoruns and process explorer I found that the ‘Steppe tweaks’ indeed do delete some files that are used when running a cmp setup.. Such as: ntmarta.dll, shiming.dll, batmeter.dll, davclnt.dll, webcheck, imagenhlp, kerberos.dll, urlmon.dll, ect. (I roughly checked until Steppe Tweaks Part 7).
But most files on those deletion list aren’t used or in use at all when playing music. So how can this improve sound quality?
I even discovered that deleting files can rise latency significantly. When I delete igfxpph.dll in my cMp setup latency doubles ! This can’t be a desirable thing and is certainly not good for SQ.
My strategy would be too use process explorer too create a list of files which are used in a running cMP setup. And than experiment which of these files can be deleted.
I already found some (see above).
But I prefer too wait and see how the Steppe Tweaks process will develop on the asylum.
I like too concentrate first on experiments with power filters on the 5 volt DC and 3,3 Volt DC power lines that power my Jyli@ digital part separately.
Mark
fully optimized cMP2 PC -> ESI Juli@ -> Van den Hul Optocoupler MkII-> Lavry Black DA10 -> XLR Mogami Gold -> Klein & Hummel O300
I am making some comments ... not trying to argue.
You said: '...I still have a hard time in understanding how sound quality can be improved through deleting files that are not in use and/or are not going to be used...'
I have a difficult time understanding too. But I remember Cics commenting that in the work of developing one of the cplay versions he discovered that changing the order in which the coding was compiled had an effect on sonics. So leaving a file in that is not used could also change how coding is compiled.
You said '...I even discovered that deleting files can rise latency significantly. When I delete igfxpph.dll in my cMp setup latency doubles ! This can't be a desirable thing and is certainly not good for SQ...'
This is a very good point and is maybe another reason why putting files back in may be recommended.
You said '...Using autoruns and process explorer I found that the ‘Steppe tweaks' indeed do delete some files that are used when running a cmp setup.. Such as: ntmarta.dll, shiming.dll, batmeter.dll, davclnt.dll, webcheck, imagenhlp, kerberos.dll, urlmon.dll, ect...'
Isn't it interesting that none of these files (except for imagehlp)are in the Mihaylov post concerning what files are needed for cmp and cplay.
These facts are certainly strange and should convince anyone to have an image backup program (like Snapshot) when doing the tweaks. You just never know what is and what is not required to boot and to run cmp/cplay...in your machine.
I can also tell you that having a windows footprint of only 80+ mb (which is where I am now)is definitely a good thing imo and in my experience for sonics.
Hi Theo,
Why should there be an argue in a negative way?
I think Steppe's idea/insight is very clever.
“Kernel streaming itself is clogged with attached and unreported services that are like leeches on a mermaid's ass or like fleas on a golden fleece. These services are from Direct X, Sound services and kernel audio "helpers". Also there are two DRM descramblers, that I don't know how to remove yet. By removing these files , some of which are of greater, some of lesser importance, we leave our KS beauty almost totally naked and busy with sound reproduction exclusively”.
It is not only an clever idea, it also works !
The sound quality improvement is heard by everyone.
So we all start removing 'these files'
But unfortunately removing 'these files' is done rather blindy.
We don't know exactly which files too remove.
Thus: files are being removed untill the cMP setup no longer works.
But why do we do that?
What thought or ration is behind it?
May be: “Less is beter'?
May be there is a way too identify 'these files'.
So that we no longer brick our cMP setup in an attempt too remove 'these files'.
I do not know anything about software or programming, ect.
But my feeling tells me: there has to be a better way than just blindy removing files in an attempt too hit 'these files'.
So I'm looking for a way too implement Steppe idea's without crippling or breaking my cMP setup.
In my reply too Rick I was trying too excite/stimulate/tease other inmates too explore other less intrusive ways too implement Steppe's idea's
I'm not a native English speaker but the dictionary tells me that 'argue' also can mean:
'to discuss, giving one's reasoning'. So let's argue. May be there's a clever inmate who knows a clever way too identify 'these files'.
Mark
fully optimized cMP2 PC -> ESI Juli@ -> Van den Hul Optocoupler MkII-> Lavry Black DA10 -> XLR Mogami Gold -> Klein & Hummel O300
n
Doing my "editing" at another computer so I do not do anything but play music and AUTORUNS at the music computer.
So you load new music onto hdd via another computer?
since I have never connected the music computer to the internet AND after finding how nice it is to have the titles of the songs filled in I have been using my regular computer for that so I can use the database.
I leave the panel off so I have easy access to the sata cable and a power connection. This also has come in handy for file deletions and making the SNAPSHOTS. Finally got a copy and find it works perfectly. Thanks for the recommendation (If I have not already thanked you).
Thanks my cmp is not internet connected either but for new music I run EAC on my home computer then save the data on a thumb drive which I then plug into my cmp and cut & paste onto a cmp hdd. But no can do now. Now we both know why other smart guys (Ryelands, Mihaylov) use a network.
Edits: 10/27/11
I am not convinced of the need for a network.
The idea of a minimal computer (not the same as minimal operating system) is intriguing but I like the upsampler within cPLAY so I am sticking with what I have.
The old story of there being more than one way to skin a cat, or any other animal for that matter.
While following the deletion route, I strongly advise inmates NOT to move anything out of the Win_SxS folder. After a long analysis, I find that it's files have a direct impact on the Sonics. If anybody needs to mingle with it, do it in the End. Keep it as the last tweak. Also I suggest not to move it's files to any separate folder, but rather delete them into the Recycle Bin, so that they can be restored back to their exact locations if needed (as their filenames are too long, & one could easile be mistaken where to restore what)..
Junaid
.
With those files deleted, the bass definition & body of instruments decline. But the Seperation increases. It's a matter of careful mix-n-match, best left for the final fine-tuning stage..
Junaid
Inmates! To finish with windows tweaking at last, I suggest to (finally) revert to start cmp not start xp mode of cmp. This way several very dirty and important files can be removed at last. I know that in the frenzy of removing I have removed some file that is necessary for start cmp option to work. Someone here even killed his cmp because of this file. So my plea is WHAT FILEs are necessary to start cmp from boot up? Also, Does CMP allow to RESTART, not shutdown computer.
Please, help me find the fileS necessary. Is it shutdown.exe? Any other file?
Serge
*
I had this same problem. Could not remember exactly but I remember someone mentioning this so I re-inserted reg.exe and I can start in cMP!
Here these files
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Live, iPad
My memory indicates you need at least winspool.drv plus shutdown.exe. Maybe more but I dont recall any others.
Hello, inmates! The tweak is splendid, as it helps to finish windows' tweaking. I would be carefull with 1252.nls file. Don't remove it for english version of OS, also screen fonts - *.fon files are slightly different for the original version of Windows. Also, Serge proposes to delete wdmaud.drv. Serge! Is it drv or SYS file? How does Juli@ work without it. Lynx users, DON'T delet portculis.sys.
I haven't done all the tweak, as I am almost at the same level of windows castration, but what I DID works fine, so Hats off to Mihaylov!
The difference will be posted soon as tweaks part 22.
I use a PS2 mouse at the suggestion of Grant (the great Canadian) who opines that this way you can disable all USB devices in BIOS.
As all of these mods require some sacrifice, one does have to reset BIOS to get USB services back but how often will that happen?
I have been doing this and find it better.
I deleted this file last night with the last group of MIHAYLOV's drivers suggestions and lost the mouse. As soon as it happened I thought that "port" part of the name was the clue and sure enough that is what it does.
Anyone have any idea which drivers the PS2 port needs?
Thanks,
Here is what I posted yesterday. I think it was msssm something something...check the link
I thought I read them all.
Thanks very much AGAIN!
I know you have tried to retain USB, but since the thumb drives won't work consider disabling it ALL in BIOS.
If you still dont get your mouse back I remember a Mihalov post recommending to keep hid*** files in. I still have these in so they may be necessary too. Hid***are human interface device drivers or files.
Edits: 10/27/11
= Also, Serge proposes to delete wdmaud.drv. Serge! Is it drv or SYS file? =
- It's drv from folder system32. wdmaud.sys is removed for a long time already.
= How does Juli@ work without it. =
- Fine! Neither cPlay, nor cMP do not use these files. See screenshots here http://www.audioasylum.com/forums/pcaudio/messages/9/97337.html
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Live, iPad
Hello, everybody. This is tweaks part 21 and this is the first illogical and strange thing that I promised. Initially it was intended as a tail-end to part 20 but with the possibilities it had opened, I thought I should post it before, especially as it is large and significant enough to be an independent tweaks part. Also I feel that some inmates are becoming restless and possibly bored with part after part of just removing files without a “BIG” advance. Really, inmates, you are wrong because removing those files you are cleaning the ground for future tweaks (17-20). Yes, I know that from part 12 it may sound sometimes slightly grainy. Also I know that some instruments are shifting places – this is because processor and memory are getting less and less things to do and there is less and less electrical phase shift (you may laugh at this explanation, but remember that Cics implemented the phase shift button in the player. He is a far-sighted person and he knew what he was doing), but again I want to reassure that all grain will settle down with the soundscape becoming very clean and smooth.
Now the tweaks part 21. It’s all about BIOS. Not very dangerous, but be attentive and prepare a small screwdriver or the white jumper to shorten the cmos pins if the things go suddenly wrong.
They won’t IF YOU ARE CAREFULL!
Switch off your CMP machine if it is working. Reboot and enter the BIOS setup. If you have GAG31ms2l or, like me, have GA-H55M-UD2H mobo you have a wonderful capability to save bios profiles and load them without any trouble. F11 and F12 buttons serve to do it. So you have entered bios from the state of cmos that you usually listen to your music at. It works rock solid, I presume, so SAVE it now pressing button F11 an name it CMP 001. Exit Save utility and enter PC HEALTH STATUS, find there CPU SMART FAN CONTROL. Enable it and enter CPU smart fan mode. Set the mode to VOLTAGE. Get back to CPU SMART FAN CONTROL and disable it. Below, blued out should remain voltage. Exit PC Health Status.
Next Go to MOBO Intelligent tweaking M.I.T. (on the top). Enter and enter Advanced Frequency Settings (second from the top). Find third from the top Advanced CPU Core settings. Enter it and change all AUTO values to DISABLED
The values and parameters to DISABLE are:
Enhanced halt (C1E)
C3/C6 State Support
Thermal Monitor
EIST Function
Bi-directional PROCHOT.
This will look like core 1, Multithreading - enabled, all other 5 settings DISABLED, not Auto. Mark for yourself C3/C6 state support and CPU Enhanced halt. You will possibly play with them later. Now exit Advanced CPU Core settings, exit Advanced Frequency settings and go to Advanced Voltage settings. My cpu (540) has an in-built graphics core. Hope yours has also. Enter Graphics Core menu and set voltage to 0,650. Press F10, Save, exit Bios, the system will start to reboot, enter bios again set graphics core voltage to 0,625. Save, exit bios and after that let the system completely reboot, but don’t listen to the music. Tell windows to shutdown-restart and enter bios setup again. Enter MOBO Intelligent tweaking M.I.T. - Advanced Voltage settings. With graphics core voltage reduced to 0,625, enter CPU Vcore end set the voltage to 0,68125. Save and exit and enter bios again and so on, set vcore voltage to 0,66875. Save, exit (I hope your system starts, mine does.) enter bios and press F11. Save current settings as cmp super (or cmp 002), exit save utility and back to M.I.T.. Set Graphics Core voltage (don’t mix things) to 0,612, Save and exit, enter bios, set Vcore voltage to 0,65625. Don't change the order of lowering down the voltages! Don’t attempt to go lower with voltages of both Vcore and Graphics Core. DURING ALL THESE TWEAKS YOU SHOULD NOT ALLOW THE SYSTEM TO SHUTDOWN!!! Save the bios values as CMP extreme or cmp003 and now go to M.I.T. Advanced Memory Settings. Set DRAM Timing Selectable (SPD) to Expert. And now go to Channel A Timings. Here we tweak timings of our memory as follows:
This is the table of timings and how I changed them (a dash instead of a value means that I didn’t change the value in this time. Dash between values serves to separate restarts (about 8 steps) but not complete reboots) Save by pressing F11 to cmp extreme 2 (cmp003) often and restart but enter bios and go on.
I have a recommended DDR3 Kingston Value Ram in 1GB module. If You have a DDR2 (for previous MS 2l mobo and e7200 processor), your timings are probably tighter.
Especially, if you are lucky to have a LL or even ULL HyperX memory!!! So, use your common sense. Smaller values in timings can be reduced by half and bigger values 3 or even 4 times. But Be carefull.
TRAS 11 - 10 – 8- - - 7
TRRD 2
TWTP 14 – 12 – 10 – 9 – 8 – 7
TRFC 34 – 30 – 29 – 25 – 20 - 20 – 17 – 14
Static TRD - - - - - - - 5
TFAW - - - -7
TRC- - 11 - - - 8
TWR - - 9 - -6 - - 3
You Don’t need to follow this table exactly, but I recommend to follow it more or less.
This is not all. After You have achieved final values, saved as cmp extreme 2 (cmp003), you should know, that the system won't reboot from the “cold start” with these values. So reboot and listen to the music. Let the system burn in for 24 hours. Before attempting to shut down ALWAYS RESTART and ENTER BIOS SETUP. Switching the machine on should start with rock solid settings and only after 5 minutes of playback reboot to cmp extreme and after 15 minutes more to extreme 2. Otherwise the system will fail to start. Press F12 and load CMP001 (your rock solid values). Reboot and then shutdown completely. I Know it is a lot of trouble, but it is worth it. Technically it is like a soft on and soft off with big power amplifiers.
Serge.
P.S. It is best to do this tweak after part 20, but you can do it anytime. The first result will be the soundstage too close and even crowded and in your face, but it will settle to a bliss with boundless depth. The ease and fluency, liquidity and natural tone of reproduction is astonishing. Give the system time to burn in.
Also, You start the system with rock solid bios settings, proceed to cmp002, and after 5-7 minutes restart with cmp003 (use F11 and then F10 buttons). You Switch off in the opposite manner. From cmp003 (extreem) restart from windows, enter bios, load cmp001, boot to windows and ONLY after that shut down.
You are right. It sounds rough at the start.
Things were starting to get better when I stopped listening after an hour. Tomorrow night will be the test!
The procedure works just like you said it would.
Have never had any success getting much below 0.75 volts in the past. This is the way to do it.
Serge, you are quite a clever fellow!!!
Hello, everybody. This is Tweaks part 20, and this is what I propose to do:
Remove these folders:
Crypto
from
C:\Documents and Settings\All Users\Application Data\Microsoft
C:\Documents and Settings\All Users.WINDOWS\Application Data\Microsoft
C:\Documents and Settings\YOUR LOGON NAME NERE\Application Data\Microsoft
and from anywhere in Documents and settings. Go Real Deep!
Manifests from C:\WINDOWS\WinSxS
NetworkService from Documents And Settings. Go deep. Not everything can be removed, get inside the folders and exterminate everything inside.
Remove these files: vga.dll desktop.ini 12520437.cpx 12520850.cpx dssec.dat emptyregdb.dat wpa.dbl mouse.drv VGA.DRV termcap ntimage.gif
esentprf.hxx logonui.exe.manifest ncpa.cpl.manifest nwc.cpl.manifest sapi.cpl.manifest WindowsLogon.manifest certmgr.msc command.com
rsaci.rat AUTOEXEC.NT CONFIG.NT sfc_os.dll.orig graphics.pro bios1.rom bios4.rom v7vga.rom $winnt$.inf oembios.sig instcat.sql himem.sys
ntio404.sys ntio411.sys pagefile.sys activeds.tlb msdatsrc.tlb wdl.trm pagefileconfig.vbs login.cmd
Now our kernel beauty mermaid is really shaved and wearing only her underthings. Part 21 remains to be corrected and part 22 (hat off to Mihaylov!) is to conclude this process.
Serge.
Serge,
You machine must be re-installing files!
Hello! Sorry to present such a dirty set of tweaks. They were done during my influenza and my cat dying and one more cmp crash and a hurried rebuild, that is why some files from parts 5, 11 and, probably, 12 got listed in parts 17-20. The last (I hope) part will be 22 and is soon to be posted thanks to Serge Mihaylov.
Serge.
certainly not seriously giving you a hard time! Far from it.
If I had written all of that stuff out there would be more missing than duplications.
You know you are due nothing but THANK YOUs for getting us all to think more about how our computers work.
The death of two dear dogs were very hard and few things were more painful than that. Only my Mother's death has followed me for longer and with greater sadness. As the story goes time is the healer AND it is hard to beat getting another cat to bring life back to normal. I do not think anything honors the memory of a loved pet more.
Nothing but good wishes for you.
Hi Serge. Doesn't mouse.drv eliminate mouse function? Also what functions do you you still use? For example mouse, I know you dont use keyboard, anything else?
Hello, everybody. This is Tweaks part 19, and this is what I propose to do:
Remove these files (some of them have names that can be found elswhere, like desktop.ini, I refer to the files residing in Windows, Windows\System and Windows\system32 + drivers folders):
imsins.BAK setup.bmp winnt256.bmp winnt.bmp control.ini desktop.ini system.ini win.ini bootstat.dat pschdcnt.h rasctrnm.h rsvpcnts.h WindowsShell.Manifest REGLOCS.OLD
setup.inf _default (With MS DOS logo) 0.log bitssetup.log cmsetacl.log comsetup.log DtcInstall.log FaxSetup.log iis6.log imsins.log MedCtrOC.log msgsocm.log msmqinst.log
netfxocm.log ntdtcsetup.log ocgen.log ocmsn.log OEWABLog.txt regopt.log SchedLgU.Txt sessmgr.setup.log setupact.log setupapi.log setuperr.log setuplog.txt Sti_Trace.log
tabletoc.log tsoc.log wiadebug.log wiaservc.log WindowsUpdate.log wmsetup.log XPlite.log SET3.tmp SET4.tmp SET8.tmp explorer (this is Not Explorer.exe, this is Windows Explorer Command with a lens)
After parts 19-20 you should have only 1 file left - real explorer.exe in Windows Root. Only MOUSE.DRV in Windows\System., So proceed to part 20, please.
Serge.
The file setupapi.log will form anew by the system (by the file setupapi.dll) after removal. That the size of this file did not increase with each loading of system it is necessary to delete all its content in a notepad, and then to specify "only for reading" in properties of this file.
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Live, iPad
Hello, everybody. This is Tweaks part 18, and this is what I propose to do:
Remove these files: regedit.exe (2 of them?) rshx32.dll spoolss.dll vwipxspx.dll SYSTEM.DRV (from System) also this file, find it with search - d42cc0c3858a58db2db37658219e6400_be532a57-712d-4ba9-81e8-8ab896330805
ntio404.sys ntio411.sys ntio412.sys ntio804.sys ntio.sys
Now THE FOLLOWING FILES! They come from Windows\System32\CatRoot, Serge Mihaylov suggested to remove both CatRoot and CatRoot2 folders completely It's a bold and good measure!. You should probably rather do it his way.
I only post the names of these security catalog files to follow the time order of things. They cannot (if I remember correctly) be removed from your windows. You need to enter tweaked windows and remove these files from other windows install. Say from another HDD.
FP4.CAT HPCRDP.CAT IASNT4.CAT IMS.CAT MAPIMIG.CAT MSMSGS.CAT msn7.cat msn9.cat MSTSWEB.CAT MW770.CAT netfx.cat NT5.CAT NT5IIS.CAT NT5INF.CAT OEMBIOS.CAT SP2.CAT tabletpc.cat wmerrenu.cat. Also, while working from another
Windows. remove the file ega.cpi
Next remove from the Folder Policies:
x86_policy.1.0.Microsoft.Windows.GdiPlus_6595b64144ccf1df_x-ww_4e8510ac
x86_policy.5.1.Microsoft.Windows.SystemCompatible_6595b64144ccf1df_x-ww_a0111510
x86_policy.6.0.Microsoft.Windows.Common-Controls_6595b64144ccf1df_x-ww_5ddad775
x86_policy.7.0.Microsoft.Windows.CPlusPlusRuntime_6595b64144ccf1df_x-ww_a317e4b3
Serge.
Hello, everybody. This is Tweaks part 17, and this is what I propose to do:
Remove these files: mscdexnt.exe msconfig.exe chkntfs.exe notepad.exe (there are 2 of them!) taskkill.exe (CMP will kill processes anyway) TASKMAN.EXE (both of them) user.exe asycfilt.dll COMMDLG.DLL dbgeng.dll devenum.dll dmdskres.dll fdeploy.dll hid.dll localui.dll msimg32.dll msratelc.dll nwapi32.dll
olecli32.dll olesvr32.dll olethk32.dll rasapi32.dll rassapi.dll rpcns4.dll rsvpsp.dll security.dll SHELL.DLL untfs.dll wshatm.dll wshbth.dll desk.cpl main.cpl keyboard.drv system.drv secupd.sig ansi.sys drmkaud.sys fltMgr.sys i8042prt.sys key01.sys keyboard.sys mnmdd.sys mssmbios.sys ndistapi.sys secdrv.sys
tunmp.sys stdole2.tlb stdole32.tlb
PLEASE NOTE, that from now on You are without keyboard. So, for the sake ov convenience, mark the keyboard files to be deleted later and go on to part 18.
BUT DELETE the keyboard drivers ABSOLUTELY, funny, but they are very dirty.
Serge.
I tried yesterday leaving in what I thought would be necessary to have a monitor and mouse but I must not have any idea how to either.
Just get a blue screen - initially WINDOWS would tell me what it was missing but eventually just blue.
The question is: does one have to come to this incrementally like lowering the CPU voltage or can one just do it all at once?
Any ideas of what needs to be added to Mihaylov's set-up for monitor and mouse purposes and what can be eliminated due to "no network"?
Trying to work backwards but any such help would be appreciated.
Waiting for STEPPE/Serge to show us what his 20mB XP contains!
Thanks,
Hi,
does anyone know how to get the latest source code of cPlay? There is nothing on Sourceforge, and I sent an email to cics but get no reply.
1. The monitor works fine in my setup.
2. To works mouse It is nessesary to return all that responsible for usb as minimum - all this I have removed a long time ago.
3. = Waiting for STEPPE/Serge to show us what his 20mB XP contains! =
This cannot be because only registry (the folder Config in system32) is 15 MB.
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Live, iPad
I notice the following previous "deleted" files in your system32 - are these files involved with networking?
desktop.ini
FNTCACHE.DAT
imm32.dll
ipsec.sys
msgpc.dll
mswsock.dll
mup.sys
net.exe
ent1.exe
betbt.dll
netmsg.cll
netplwz.dll
netsh.dll
netshell.dll
netui1.dll
netui0.dll
ntlanman.dll
ntlanui.dll
psched.dll
riched20.dll
riched32.dll
wshtcpip.dll
Can these also be removed if one is not using a network?
Are there files that need to be added for a "regular" installation?
Thanks,
Rick McInnis
Hi Rick!
Yes, you can remove these files if you do not use network (but you should leave the files desktop.ini and FNTCACHE.DAT - they are not related to the network). You can also remove the file tdi.sys.
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Live, iPad
If you print out the original My Tweak posting I have gotten as far as half way through Page 4...except I deleted none of the following explorer, notepad, regedit, mouse, browseui, all of the .nls files, crypt32, desk, deskadp, hidserv and I'm happy with the sonics. I thought it was best of best.
FAR more extensive!
I never did d/l those what is it? Just say a little more about it.
I went down the same path. My machine was a Steppe16 and unable to run CMP.
explorer.exe was the missing ingredient after this tweek. I put back explorer and then one by one several dlls but ultimately just ended up with teh blue screen and a cursor. I suspect we were intended to run this on a machine locked into cmp, in which case explorer and its supporting cast was not required.
This prompted a complete rebuild. At least now I have multiple partitions so I can take an image before I go down this path again. Now back to cmp, it is surprisingly good, but I believe 4.5 and 16 were better. Until I am satisfied I know how to quickly do a restore, I am leaving it alone.
I feel Mihaylov's list serves as a saviour to the already complete steppe 16. Just run it in cMp mode, using explore, & u r good to go. For users of Juli@, beware of what u are deleting if not using my Batch file. The drmk.sys needs to remain for Juli@ to work...
Junaid
Hi Jolida! Are you running Steppe 16 plus Mihaylov My Tweak in cmp mode? Or is it Basic cmp^2 plus My Tweak?
Edits: 10/27/11
Well I have tasted all the flavours of the ice-cream. I must admit that Mihaylov's list does justice to the deletion route. If only could the bottom-end improve, it would be flawless..
Junaid
Did you put back all of the files you feel help the bass?
Not yet. I've yet to experiment with them. But it needs serious attention. Seems like the range below 200Hz is kinda rolled off to a considerable level. Since the deletion route is at it's twilight, there are not many things left to play around with, except to try un-deleting some that have their effect on the bass. I hope to sort it out some-way or the other...
Junaid
Why is everything on a stand-still??? Nobody seems to post anything or share their experiences with their present set-ups or step-ups...
I'm sure there is activity going on back-stage as usual. Or has everybody decided to be content with whatever has been achieved so far ??
Junaid
Hi Junaid
thanks to your job I'm working around various steps.
Until now I can tell you that in my integrate system ( Ga H67 MA Ud2h B3, i3 2100T, Kingston 1333 1GB, OCZ 64GB, PicoPsu 160 xt feeded by Belleson based linear PSU (like P4 and soundcard), linear PSU non belleson 7812/7805 based to feeds dirty devices, Asus Xonar Essence STX deep modded analogic unbalanced output) I've found that step 7 (XPlite delections) with SOUND and msacm32 spared is really astounding and better of each previous one and, obviously, of the native cMP2.
Now I'm trying to save a OS image for each significant step, using various little SSD (8GB) to test faster differences.
It's an hard and long job
I'll post my follow up
Daniele
Oh that nice. Just be careful & slow at making A/B comparisons. It should in fact be an A/B/A comparison given a little time...
Junaid
I corrected the list of my tweak in my message here http://www.audioasylum.com/forums/pcaudio/messages/9/96990.html
I returned following files:
c_1252.nls
clbcatq.dll
comres.dll
crypt32.dll
imagehlp.dll
JulaWDM.sys
R000000000007.clb (into folder Registration)
wintrust.dll
because they are used by cMP and cPlay (you can see it in the program ProcessExplorer and Everest launched on "classic" optimized cMP2).
cMP:
cPlay:
The list of running system drivers (from Everest):
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Live, iPad
Ok I am done with My Tweak except for the .nls files and the fonts. Does anyone know what we need and what we can delete?
I've gotten about 2/3's thru Mihaylov My Tweak corrected and the sonics are scarey good but I can no longer cut and paste which precludes me from adding new music files from thumb drive to hd or to swap spool.drv in and out so I can run cpuid. So the files that I deleted that caused the problem are somewhere between mprui through rsmps (page 5 of the printed My Tweak). Any ideas which files are the culprits?
Edits: 10/27/11
Suspect files causing no cut/paste are msimg32, psports, msratelc, ntkmlpa, nwapi32, rasapi32, rassapi, rpcns4, rpcss, rsfaps, rshx32, rsmps.
Looking these up on paretolgic gives no clue. Anybody have an idea?
In the meantime I'm enjoying best sonics I have ever had.
It was i804prt.sys because I use ps/2 port not usb. I'm back in business. Although I have to say I left mssmbios in also.
Edits: 10/26/11
I may have to update music to hdd's on another computer because this really sounds good. I has the air and delicacy but with great bass and it is holographic as all get out.
Ok did another page of My Tweak stuff dont know if sonics are better but certainly no worse. Although music at same volume settings sounds louder as a result I set it lower and noise seems lower. One thing if others do it my way (one deletion at a time) when booting it takes about a minute longer from xp screen to desktop. Still cant cut and paste but I can delete so all deletions here were done with delete button not cut and paste into a temp file on c drive.
One more page of stuff to go...so far you rock Mihaylov!
Edits: 10/25/11 10/27/11
;)
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Live, iPad
I did the cat root deletions and loved it but after all the drivers I somehow lost mouse cursor. So what do I need in the last driver list to maintain mouse point & click function I did leave all the stuff like hid*** you recommended but still lost mouse.
Edits: 10/26/11
And the usb port to which the mouse connected works?
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Live, iPad
Dont use usb mouse but did find the culprit ...i8042prt...I use ps/2 for mouse and monitor.
OK. The solution is found. Thanks to Rick.
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Live, iPad
Ummm maybe mssmbios I'll try it
...and it is a sonic wonder. Very nice. Windows now is 88mb not bad and a ways away from 50mb but I'm getting there.
A big way to go Mahaylov.
An example of stuff I deleted but cplay works fine is R000000000007.clb (into folder Registration) and comres.dll. Dont know where I put the Registration folder but I restored comres but observed no noticeable sonic performance change in cplay.This seams very contaradictory.
Also I'm sure you delted this stuff too but cmp/cplay worked for you.
Edits: 10/24/11
Simply I distinguish concepts: saves workability and works as it is conceived by the author cics (it's about cPlay).
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Live, iPad
Thanks again for the additional info. What I find interesting is that some of your stuff in your new list I either don't have anywhere (could be support for other software you have & I dont) or I do have but its in my temp list and not active because I put it there. Weird.
Edits: 10/22/11
Ted,
Did u do Manage to do his list & listen???
Junaid
No never could get bat file to work.
for those interested.
Dave Davenport assures me this is the case.
He says the "isolation" is affected at the RECEIVE board.
After researching this subject last week, here's what I wrote here (http://www.audioasylum.com/forums/pcaudio/messages/9/96941.html):
"RE: You think one is better off with SPDIF? - GStew 20:02:11 10/14/11 (0)
In Reply to: RE: You think one is better off with SPDIF? posted by rickmcinnis@dogwoodfabrics.com on October 14, 2011 at 07:55:45
Rick,
-SNIP-
Then on galvanic isolation with the K&K I2S connection, while the connection circuits are NOT isolated, the RAKK DAC's I2S input IS. This is from pages 6&7 of the manual... and it has some additional grounding/isolation information that might be useful, but that I need to digest.
-SNIP-
And that your setup already has galvanic isolation on the I2S lines MAY mean that you are already getting better sound quality using the Juli@ than the rest of us with lesser DACs."
This is what they say in the current (3.1) version of the manual:
"For the signal interface, we have chosen to provide a galvanically isolated I2S interface (both signal and ground) on the RAK dac Mark III. This prevents external ground current, potentially containing digital noise, from entering the RAKK dac ground system.
The I2S Receive boards do not have a galvanically isolated power supply interface. Rather, the preferred solution is to provide a totally separate DC power supply. This means a power supply derived from a separate power transformer or a separate (non-shared) winding on a power transformer. The RAKK dac Mark III version 2 Low Voltage Power System provides separate power supplies for the RAKK dac and the I2S Interface Adapter."
They provide somewhat different (and still useful) information in the earlier version of the manual, 2.0. Both are available for download here:
http://www.raleighaudio.com/i2s_interface.htm
This also agrees with the info page 3 of the RAKK DAC manual, available for download here:
http://www.raleighaudio.com/rakk_dac_mark_iii.htm
It may be semantics, but all of this strongly suggests that the galvanic isolation is in the RAKK DAC, not on the I2S interface. This is also strongly suggested by the interchange between Soundchekk & KevinC in this thread:
http://www.audioasylum.com/cgi/t.mpl?f=kandk&m=8609
I'm not saying all of this to be argumentative, but to point out that according to this information, you ONLY have galvanic isolation using the K&K I2S interface (either from the Juli@ or another source) when using the RAKK DAC. The information STRONGLY suggests you don't have that with other DACs.
Now, without having heard the RAKK DAC, but from the reviews and the information posted about it, I think it is a very fine one and a great choice. So this is not a knock on it... just a caution to people like Theob who might want to get the K&K I2S interface thinking it provides THEM with galvanic isolation for their DAC.
Greg in Mississippi
Everything matters!
Dave said this in his note:
"Galvanic isolation needs to be done only once and we do it on the Receive side: the Send side is not isolated."
Not having read the manual in a long while I assumed he meant the receive board not the DAC board - both being on the receive end but an essential difference for those using another DAC.
Thanks for making this clear.
I do wish Dave's DAC did not upsample. I do know it sounds very good to me but I have never sampled another DAC in my system since that is basically impossible. I am committed to Dave's stuff.
Yah, that's a bit of a bummer... but I'm beginning to think that galvanic isolation for I2S is just a matter of finding the right interface chip and putting it in. The chip on the EXAU21 sure looks that way (although the setup on the RAKK DAC looks more complex).
On the upsampling, it's too bad you don't have a choice. I still upsample to 192 & have compared with each major upgrade of my SW/HW and keep coming back, so I don't think it's necessarily a bad thing. Being able to try upsampling in the PC versus NOS would be nice. But that would be a significant change to the architecture... I understand why it's not there.
I do think it is one of the better choices available for a DIY DAC in that it includes both ASRC and galvanic isolation. While I think that you can do better than using these techniques, I don't know of anyway to do that without a very integrated, dedicated, and comprehensive implementation. The ASRC and galvanic isolation make it a much more universal device.
I do think that power regulator upgrades might make a difference, similar to adding Twisted Pear Tridents or Paul Hynes regs to a TP Buffalo DAC. But it looks to be a very involved, complicated, and expensive job... it looks to have 10 (or more?) on-board regulators! Good for sound, bad if you want to try to get better sound!
Again, I think it is a good choice. More expensive (especially the output stage) than some of the alternatives (like the TP Buffalo), but it looks like they did their homework and did a LOT of the right stuff to begin with. Like I said, I suspect upgrades are going to be difficult because it looks like it's so good to begin with. As best I can tell, it should be the equal of many DACs 2-4x it's cost as a fairly easy-to-assemble kit.
Of course, I don't have one and haven't heard one, so take that all with a grain!
Greg in Mississippi
Everything matters!
but as I read the article it seems to me that GI has nothing to do with the DAC but how the power supplies are implemented. If the receive board is powered separately from the DAC and the grounding instructions are followed ...
From what I have read I am not getting the benefits of GI - my version of the board has the dreaded DC to DC converter, I hope I can remove it - I was using my same power supply for the DAC for the receive card (a large HYNES supply - same as the one used by the ALTERMANN DAC folks would use instead of a battery.
Have to get another 5 volts BELLESON on the way and hope for a better presentation with the benefits of GI. We'll see ...
When I receive Dave's reply I will post that.
Thanks for bringing this to my attention!
LOL.
Sounds like you may have additional SQ improvements available when you separate the supply. I just assumed (stupid of me, I know, but I'm so good at doing it!) that you had the receive card separately powered.
Looking at the manuals for both the I2S interface and the DAC, to me they aren't totally clear and consistent on how that works, but the comments in the manuals plus the post exchange between Dave & Soundchekk made it clear to me.
And then I cheated... I looked at the picture of the RAKK DAC board and could follow the I2S and SPDIF lines coming into the board and see them go through the GI devices (optocouplers or something similar).
On the Belleson, PartsConnexion has them on sale through the end of the month. They are on back-order, but expect a shipment soon... I've already got an order in.
Later!
Greg in Mississippi
Everything matters!
so I didn't!
I installed it and never considered that anything had changed. How many times to I have to learn THAT lesson?
Getting the converter off the board will make a difference also.
I am looking forward to an improvement (so what's new?).
Just hope there is one!
Still waiting to hear from Dave.
Bye,
had to look back at correspondence to remember. Used Dave's old DAC supply for the receive board which means it is still using that converter.
Guess I will still make the change to 5 volts and get rid of that converter but I have been enjoying the benefit of GI.
YIKES!
So it has a send and receive board & how much $?
http://www.kandkaudio.com/digitalaudio.html
Look close to the bottom of the page.
Are you powering your JULI@ with five or three point three volts?
Hopefully five since that would make installation easy.
You already know which pins are which.
I like the fact I can see a longer cable and get the computer away from the amplification. System is absolutely silent other than the rush from my line stage tubes.
I put 3.3v onto the juli@ but still use 5v from the mobo. I'm thinking of doing 5v LiFeo4 (2 in series regulated down to 5v).
Dear Forum,
Does anyone have experience with asynchronous USB DACs using NO ASIO-driver software ?
Certain DACs nowadays offer asynchonous protocol without drivers (Zodiac, Arcam, etc.) but nothing is mentioned about true ASIO.....
Does cPlay recognize these DAC as native ASIO devices ?
Please tell us about your experience.....
kind regards
cMP2 Computer: i3-540 Core cPlay039 >FIR filter w/ convolver> Lynx Aurora 8 FireWire /192kHz throughout. 2x AcousticReality Ref. 202 & 2x AcousticReality Ref. 601´s ICEpower. Magnepan MG3.3R beechwood frames & custom stands. Miller chokes
Mihaylov,
Since your's was the only list I had not tried, I thought since I made it's Batch file, why not give it a try. I do not know if it's mandatory to do Ur list Only after doing All steppes, or if it's independent in it's effect. Anyways, I gave it a try & this is what I felt....
The Mids had good clarity & also instrument separation is a bit more. The vocals cleaned up. But the Bass goes down a bit losing it's muscle & blend in other instruments, making them sound less-full-bodied than without the deletion. The mid-bass is a wee bit lean as well. The balance tilts slightly pushing down the overall bottom-end BUT not drastically. The other aspects have an overall improvement. All in all, I find only a very less downside, & more of an improvement. If only the bottom-end improves a bit, this will be an out & out success. Worth implementing this Batch. Fantastic list..!!
When I ran the batch file, windows refused to boot missing on one more nls file than the ones mentioned by u & Theob. . I will make the correction in the Batch & re-post it. Also, as expected there was no mouse, keyboard & USB. So I restored their offending files as well. After doing all this, Juli@ was missing files as well, so there are three sections which need updating in Ur list. I will make the changes & run it again to check the functioning of all the above, & send it to all...
Thanks for Ur efforts...P.S :- the above improvements are evident with the Base as Original cMp & cPlay b38. The list was implemented straight from this stage with only the FILES deleted (I did not delete folders yet nor did I touch the Win_SxS folder or any of it's contents, as it has given me trouble earlier with windows functionality) & without involving any of the Steppes & My (Jolida's) list... I have no idea how it would sound with b39 or with any of the omitted lists included..!!
For Juli@ users, DO NOT attempt to remove files used by Juli@, else it sounds terrible with uncontrolled highs & rough edges in the lower midrange. The files that Juli@ needs to give a balanced sound are :-
In system32 folder :-
1. wdmaud.drv
2. jula.sys
3. JulaAsio.dll
4. JulaPan.exe
5. JulaWdm.sys
6. ks.sys
7. ksproxy.ax
8. ksuser.dllIn System32/Drivers folder :-
1. drmk.sys
2. jula.sys
3. JulaWdm.sys
4. ks.sys
5. portcls.sys
6. stream.sys
Junaid
Edits: 10/20/11 10/20/11 10/20/11 10/20/11
I believe that My Tweak should applied after all the tweaks (Steppe, Jolida and me) as final which means the suitability of the system only for audio playback using cMP/cPlay after it (possible view of interfaces of cMP/cPlay I had already given).
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Live, iPad
Edits: 10/20/11
Junaid is right - would it be possible for either/both of you to show us which files you are keeping?
This would be helpful to those of us taking different approaches to end up in the same place.
Also, Mihaylov was right to bring us back to AUTORUNS. There is much one can do here.
After another rebuild one can see that if one's intentions are to go with file deletions most of the original cMP optimizations are moot. Most can be done with the file deletions without dealing with SERVICES and DEVICE MANAGER and the one's that remain you wonder if they still are effective after the file deletions (in total).
"which files you are keeping?" -
I already two times gave link to screenshots of the system. I give once again.
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Live, iPad
Edits: 10/20/11 10/21/11
I found it a minute ago tucked away in an answer to Ted.
Thanks.
It is worth posting three times!!!
Thanks, again,
:)
Junaid
As mentioned in my posts, I would not be following the deletion route anymore. But since Mihaylov has posted his set of tweaks, I thought I would make a Batch file for his list as well, so that inmates could benefit from it. I have managed to make the Batch for his list today. Anyone interested may email me....
Junaid
Based on Serge's plan, I was hoping to finish up the "path to deletion" and Drink The Cool-Aid along with theob - yes, I was willing to steppe out of the comfort zone of cMP (and use of the keyboard), and sample the Gold at The End of The Rainbow (ie: cPlay-only in a stripped OS).
Aw... shucks - no, keyboard, no cMP, No Nuthin'... except a message on a blank screen stating:
"cicsMemoryPlayer.exe - Unable to Locate Com... This application has failed to start because WINSPOOL.DRV was not found." Hmmmm... not good.
My only thought is the OS is stuck in cMP mode, and now that critical files are gone, it's Death to All Music (btw: I NEVER use XP mode, since the SDD is serviced by connecting it to another computer, so I haven't tried it during the Daze of Deletion).
At this level (STEPPE_16, Pt1), I can't get back into XP mode, and therefore the Windows desktop is unavailable to me (from where presumably I must launch cPlay for the Full-MONTY experience). I suppose somewhere along the STEPPEs we lose the ability to switch between XP & cMP modes... does anybody know which files are responsible for this behaviour ?
What settings do I need to make before clobbering the OS with Pt2 of STEPPE_16, so that I have "something" to play with (ie: cPlay) ?
Thank you to everyone who's participated in this cMP2 Resurgence, your contributions make the world a better (sounding) place !
Cheers
Grant
That's not a Toy... IT'S A TOOL !!
Try restoring explorer.exe, winspool.drv(to both system32 & system folders) & olepro32.dll. I hope u have'nt deleted shutdown.exe as well.
Since u are unable to boot from ur hard-disk, i suggest u hook it up to another computer as a slave & restore the above mentioned files by navigating to the folders. Also deleting some of the contents in the Win_SxS folder are responsible for malfunctioning of Xp. Im surprised that even after reading so many posts, u have not taken a back-up of ur partition before u went ahead with the deletions...
Junaid
Edits: 10/18/11
Have been listening to the Original cMp version since the past two days. Can't do anything but just laugh at myself. This plays like how cMp actually has to play. This was why we built our cMp for. Each n eveything sounds so "Correct". Sounds miles ahead of any step or all steps put together. Just a wee bit of fine-tuning is all thats required. The settings in cPlay are more important & critical than any file deletion. Of all the files i had deleted in the past weeks, i feel that most influential of them is winspool.drv. But cMp would not start without it. And the equation is incomplete without cMp. If only Cics could release a new version of cPlay that will not use this file, then I would certainly delete that one file..
Have been totally lost in a different track since one month with these deletions. Feels like I'm listening to Music after such a long time. Cics really knew what he did.... I feel ( I FEEL ) its more wiser to wait for Cics new release than doing all this...
I'm so so glad that atleast i realised & woke up before losing the faith in"Hi-End"....
Junaid
Edits: 10/17/11
Ok this is probably unfounded since I have just completed steppe 2 or 3, but I wouldnt be surprised if the "original" unaltered recipe is the best.
Here is why. Remember that in cplay developments there would always be a fundamentally new release and invariably it would be super clear and detailed but invariably many would go back to the "warmer version". And then a new version would warm things up?
I think cics got to the point where he went too far in the warm direction (IIRC that was 38) and then 39 balanced the thin detailed sound with the warm sound. And we were all happy best of everything.
FOr those who are getting that detailed "unmusical" sound, could you try v38 and see what happens?
Said another way I think cics probably figured out how to change the sound and ended with the most balanced player he could based on our feedback. Now we might have changed that balance with all the files we deleted and other things.
I think this shows us all the hard work he did to get things where he did. If you think about the things that go into his recipe it is amazing the effort it would take to figure out all that and listen to the results.
Now the skeptic in me would like to say something. Perhaps this whole thing has just been a funny exercise in the placebo effect??? Certainly if you read some of the posts where people heard or didnt hear things and then got some feedback and suddenly heard things it does make one wonder.
And how many of the recent epiphanies with steppes mods are just the fact that by the time you get a step finished it is 2am and everyone knows things sound better then anyhow??? :)
And I wouldnt hold my breath for any new release from cics.
Skeptic off.
No one here remembers the bending of our minds
Yes. That's exactly what I feel, has happened. I am not saying that the steppes had all downsides. I guess most people have not clearly understood my last two posts (U turn, & theory vs Practicals ). I certainly agree that steppe 5 has a very open & clean sound. In fact I had again n again suggested people to move back to steppe 5 & analyse (in my posts last week). If u read my posts carefully, u will understand what I'm trying to say...
Steppe 5 was the best of "The Steppe Regimen"... It was the most balanced since the deletion process commenced. BUT... when one goes back to the unaltered cics recipe, u will immediately realise that there is SO MUCH that has been lost till steppe 5 as well. It's not just the Air & Openness that we have set up the cMp for. I too loved steppe 5 a lot TILL i woke up to the Original Optimised cMp.
WE FIRST HAVE TO GET THE BASICS RIGHT !!!!!!
Most inmates have never bothered to even play with the cPlay settings elaborately. There is so much in them. Everything including the buffer, phase, SoX, VHQ, HQ, tuning with alias checked/un-checked, all the above settings done to 95.0 or 96.0 & so on, Intermediate 10,15,20,minimum,maximum.... Without playing with all these combinations, how can u get the Basics right for YOUR system. Every combination of these settings change the way the Vocals blend with the Midbass, with the Highs transforming into sibilance, with the definition of instruments vs their spacing, the presence of instruments vs their imaging etc etc. With just the above cPlay parameters, there are nearly a hundred or more combinations. And every combination effects the overall nature of the Tone & it's presentation. As I have been doing this since the past 2-3 days, I am surprised how I had lost track & deviated from all this. There is so so much that is lost, yes LOST, not just thinned down or changed. The sound totally lost it's FULLNESS, BLEND, IMPACT, DELICACY, & most of all MUSICALITY...!!
And one more thing I really find funny & totally impractical in this forum. The cMp is a player, NOT an active equaliser or Room Corrector with which u can cure all your other faults. Instead, people here are trying to use just this ONE piece of their equipment & expecting everything else to benefit from it. This is totally wrong. For eg. No Room in the world has a totally flat frequency response. Every single room has Different Modal resonances, reverberation times, Peaks & Nulls. These are acoustical problems, & should be treated Acoustically. Just because one has a standing wave at 400Hz in his room, & a thin sound in some Steppe with less Mid-bass may compensate for that peak, he treats it like a saviour. Acoustical issues need to be treated acoustically & electronic issues must be treated likewise. If we could get EVERYTHING out of cMp, why would we ever upgrade components, speakers, interconnects, power cords, Acoustics or isolations. This is the basic mistake of most people in this forum. They are using this one cMp dart to hit on every issue they may have elsewhere...
Get the basics correct, then move on with other things....
Junaid
Edits: 10/17/11
Dawnrazor, I have had that same train of thought at several points in this group pairing down party. Yet at other points, I feel pretty sure I hear something better for real. It is complicated by all the factors that cause us to like several different renditions of the music under different condtions internal and external(e.g.,group enthusiasm) at one time or another--the mood, what was heard last, the hour, expectation, hobby-horses, what we had for dinner, etc.
Jolida, I for one got all along what you say about step 5 and then your U turn. Taste and personal preference play an inescapable role in all this. Also, you are just as subject to subjectivity, bias, group influence, etc. The passion of your conversion back to optimized cMP makes me wonder. So does your suggesting that many/most of us have not explored cMP and its settings over the years we have followed it intensely thru many versions, many discussions of different settings, etc.
Hey Riboge,
I have had the same experience where I hear an improvement and am all excited. A few days later things seem a bit normal. Kind of a roller coaster at times, but that could just be the tunes I listen too.
Also this is on my headphone rig which is not the best so I am not making any conclusions based on this.
I am totally disappointed though in that my main rig with the icore 3 stuff isnt as good as my headphone rig in terms of upsampling. I cant get over 176k on the new hardware without stuttering but the old one does 192k easily. This makes me wonder if maybe the old hardware isnt better??
Now I think I have to take Ryelands advice and make images of each step. Damn you Ryelands :)
No one here remembers the bending of our minds
You said: '...A few days later things seem a bit normal. Kind of a roller coaster at times, but that could just be the tunes I listen too...'
Welcome to the world of tweaking. Even after the 'best tweak' I sort of subsume it after a day or so and I'm then ready for the next big thing.
Riboge,
I had not invented the stage till Step 5 nor was it me who invented cMp. Cics has put in a lot of effort to get things where they stand now, with his b39 version. It's upto inmates to understand why I initially suggested step 5 & what's actually the reason for me to revert back to the unaltered cics version. There should be some reason for it right? I have not posted it just like that. I have analysed & auditioned all steppes till step 16 & felt that after step 5, things don't stand a chance. This was when I MYSELF reverted to step 5. At that stage, I too was back to step 5. I had absolutely no idea that I would re-check the unaltered version ever. But something made me do it. And here I am. I still like steppe 5 a lot, but when compared with the Original cMp version, it too misses on quite a few things. What I have written In my posts is just MY OPINION. I am not compelling inmates to follow my route. It's their system, which they have worked very hard to optimise. I posted because I felt it's important to share our opinion. And I DID ADMIT that I too got carried away. Dint I mention that in my earlier posts??? Read them again !!! There should be some reason to sit down & write loads of data to make all the batch files & then swaying away from them. Who would do that without a reason. Yes it sure may be subjective, but I repeat "THAT'S MY OPINION".
And I guess u misunderstood my last line in the previous post. I said "get the basics right & then proceed". It did not imply to the cPlay settings. It relates to other issues one may have apart from their cMp, like Room mode problems, interconnect mismatches, & other faults elsewhere in the whole set-up & not to look for their rectification just by compensating for the same through the cMp. U sure must have played around with cPlay settings, but others might not. I'm just throwing light at the area of fine-tuning within cPlay itself, rather than looking for solutions elsewhere.
I hope that answers your question..
Junaid
Edits: 10/18/11
Jolida, what I wrote was a caution not an attack. I had read all you wrote before carefully and also with great appreciation. You have done enormous work to the benefit of all. I took you seriously enough to reinstall unaltered cMP and cMP up thru 5 in order to compare to thru16 to try to hear for myself what you were talking about. I was ready to be reborn to plain optimized cMP as you seem to have been, but I found thru5 still sounded better to me. I don't (yet?) perceive what you find missing. I am now comparing thru4+Jolida1,2 to thru5 at kclo's recommendation.
You have shown much to us. It may be possible that we can show something to you...or bring you back to something...or, without there being any disparagement of you.
Hi riboge and jolida,
I am now back to 4.5 (Steppe 4 + jolida's first batch).
As I come down from the full stepp 5, to steppe 5.5, to 4.6, and now 4.5, its like turning the tilt knob (remember the Quad amp?): the high is reduced and the bass filled up. What attracted us in the file deletion path was that the high frequencies seemed to be more refined as more files are removed. By focusing on the highs we were not aware that the balance was being tilted, and artificiality crept in.
With 4.5, Paul Mccartney's bass in Sgt. Peppers now has the right weight. How do I know? UK pressing vinyl is my benchmark.
KC
Hey kc,
What u mentioned above, is it any different from what I've been saying all this while?? This is what I have been trying to convey ever since I got back to the Original cMp. The only difference is that u are taking a reversed-walk & I took what I would say a reversed-jump...
Now as u are moving incrementally in the opposite direction (which is a good thing), u are getting a sense of what had been compromised. This 'tilt' which is apparent in the music, is a cumulative effect of hundreds of file deletions. So to rectify, or to say "balance" the equation, one needs to pin down the offending files which should-not-have-been-deleted. But as the number of deletions are a lot till step 5, it made more sense, TO ME, to start from the Original cMp, which sounds closer to that "balance" I am looking for.
I agree, & have always agreed, that step 5 sounds very clean & strain-free, but the Scale & Weight that's in the Original, is lost somewhere in the route..
Also, I suggest u do try one more thing at this point. Staying at step 4.5 (where u are now), uninstall the cPlay b39 u have, reboot the computer, install b38, & give it a listen.
I have been listening to five cPlay versions since two days now, starting from b35 to b39. It may seem un-necessary for others to validate what I am doing, by trying all this. I do not intend to post my opinion about this b35 to b39 session, as I feel, or as riboge said, may bias others as well. So try & decide for urself. If not all the five, try b38 atleast...
Junaid
Hi Jolida,
I am reversing stepwise because of 2 reasons:
1. steppe 4.5 is my earliest backup, so I can't jump any further backwards; and
2. I was very happy with each tweak until after 4.5, or 4.6; somewhere around there is the optimum for my setup, I think, and I want to capture that.
Will try out b38 like you suggested.
KC
That's wonderful. U are close to the Optimum for Ur set up, which is a real bold statement to make. Im glad u feel that way. I wonder what the others are upto. No news about the outcome of Steppe 21 or Mihaylov's list??? Some are at step 5, u are at step 4.5, I'm at where I started (original cMp with b38).
Also I would like to mention one more thing. When I put up my post for the Batch files, I received hundreds of emails even from people who do not post on this forum. This proves that there are a lot of them who are just playing safe & being Spectators of this whole Deletion roulette. I strongly suggest, they too post their opinions so that this discussion would be more elaborate & beneficial...
I agree with Ur other post as well. I'm doing exactly the same since I reverted to the Original cMp (looking for a good Transport or Cd player to assess where we stand now). I hope to get hold of one soon, to double-check the worth of all the pains we have taken in this project...
I'm sure cMp will have it's own Class. Atleast that's what I hope happens :)
Junaid
Hi Jilida,
Just tried b38 with steppe 4.6 (with most codec removed or disabled).
Comparing to b39, there is a sort of glare to the high which could make it more compatible with a native cPlay. with 4.6, because of the cleaner high end, b38's high frequencies might seem to be exaggerated. Moreover, the bass is too controlled such that some of the bloom of natural music is gone.
To my system b39 is more natural and is a better choice.
KC
That's good. It's all about synergy. Though I find quite the opposite here. The Highs are very smooth & the Bass needs control :)
Junaid
"With 4.5, Paul Mccartney's bass in Sgt. Peppers now has the right weight. How do I know? UK pressing vinyl is my benchmark." -
Tell me please you were listening the own digital copy of YOUR vinyl at cMP2? If not then you need to do such copy. Then the comparison will be correct.
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Live, iPad
Hi Mihaylov,
If I do that, then I would be comparing 2 digital transfers (assuming playing both on the same cPlay/cMP platform): one is the CD remastered from the original master tape, and the other is my own digital conversion from the LP.
What we are trying to do here is to test/confirm the validity of the cPlay/cMP platform. To do that I think we need another trusted platform as a benchmark. To me, they are:
1. My vinyl setup with a good pressing as source, Another digital source (a good CD player?) might do as well, but a good CD setup is WAY too expansive. Moreover, a vinyl platform gives a very good contrast as it has a different set of good/bad points comparing to a digital source. And my other benchmark is;
2. Life. Play some acoustic instrument recording recorded in a natural space and ask oneself, is the human voice , etc, more natural in this cPlay configuration comparing to the previous cPlay configuration? Good candidates for comparison are: human voice because we are all very sensitive to it, piano, brass, and strings. I also found that cPlay tweak # 5.5 onward creates an artificial space that sounds quite attractive, but when played a live recording with real space information, the natural sound stage of the recording would collapse.
KC
Well. I agree that a proper comparative testing cMP2 is not easy. I just would like to say that in my opinion any CD remastering from old master tapes doesn't sound as good as the good old vinyl Edition.
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Live, iPad
At the moment, I am at the plain Original cMp version, trying hard to locate sections which need refinement. The two main areas I find needing attention is as I mentioned earlier, the Upper Bass & the Upper midrange. Rest everything sounds perfect to me. I agree it does not sound as Liquid as step 5, but the offending parameters for it are the ones I mentioned above. Everything else sounds way superior, Correct & Full sized & scaled than step 5, atleast according to me. I have images of All the steppes till step 16. So it's easy for me to revert to any step anytime. And I did so. Back & Forth with step 5 & Unaltered cMp, I find the difference is not subtle, but a lot. If I am to listen to step 5 & revert back to the Original cMp in less than an hour, then I too may get puzzled. Because initially, the Original cMp sounds dry & hard, as our ears will be used to the Open & Liquid sound of Step 5. But when u remain with the Original & listen to it for atleast two days, anybody can sense what has been compromised & to what degree.
As I have posted, the Original cMp too needs tweaking. But slightly & carefully, so as to not let go of it's main qualities. This tweaking may be in terms of settings, file by file deletions, hardware optimisations like capacitor choices (I use Elna Silmic 2 which had a warmer sound than the Oscons), or a newer version of cPlay. I don't intend to do deletions in Full Steps which will take ages to ascertain what contributed to what. Though the file by file deletion is as labour-intensive as the former, it can make me pin down the effect of a single file vs the Cumulative effect of many files in a particular step.
By the way, my cPlay settings as of last evening are as under:-
Sampling rate: 44.1
Buffer : Small
VHQ with alias at 96.0
Intermediate 20
AWE engaged...
When playing any track, & adjusting each parameter shows a very profound effect in it's tonal variation. For eg. When I change the Buffer to Medium along with Intermediate Minimum, the Bass definition changes. So once we, or should I say "I" arrive at the best setting that gives the cleanest Blend I'm looking for, I can then move to the next line of tweaking which may be any of the ones mentioned above.
I do not want to follow the route of starting from step 5 & going backwards. Instead, I move with the Original cMp & go forward incrementally, because I feel it's more closer to the Sound I'm looking for...
Junaid
Edits: 10/18/11
You realize don't you that VHQ with alias at 96.0 Intermediate 20 and all that doesn't make any difference when you use 44Khz sr and listen to 44khz sr files. They only apply when upsampling.
It has never made sense to me, but these settings have quite an effect in my system, even though I rip basic CD's to 44.1, and play at 44.1. I use VHQ into a NOS DAC. If I have a harsh szz at instead of a sssss in female singers, I adjust down to linear and the szzzz turns to ssss. I also cannot check the alias box without losing dynamics. All this and the software is not suppposed to be engaged. go figure
Well to me it does most certainly. May be because all my music is 24/96 either from HD-Tracks or ripped via LP in 24/192. None of the Albums in my cPlay library are in the Native CD bitrate...
Junaid
ok fair enough but why would you listen to 24/96 or 24/192 files downsampled to 44khz sr?
I knew that question would come up :)
As simple as it is... My Wadia Dac is not an Upsampler. It accepts only 44.1 my friend...
Junaid
ok good answer. :)
Now don't ask me why am I keeping 24/192 files having a 44.1 Dac :)
Junaid
In the process of getting back to musicality, I had restored and compared the followings:
1. steppe 5 complete
2. steppe 5 partial (up to step j)
3. steppe/jolida 4.6 (steppe 4 + 2 batches of files from Jolida), and
4. steppe/jolida 4.5 (steppe 4 + 1st batch of files from Jolida)
Item number 3 (steppe/jolida 4.6) is the best: bass, definition, sound stage--just play a live recording and you'll see-- and dynamics.
If you are now on steppe 5 (riboge?) may be a tiny step back?
KC
step 5 to what you call step 4.6 for an hour or so: I find not a lot of difference although with some music I hear a bit fuller bass with 4.6. On the other hand, with some things 5 has more presence and realness. This may be related to what others call "air" or "liquidness" though those are terms I can't easily relate to. Sometimes I get the impression there might be a volume effect, i.e., that the music is a bit louder with 5 though I change nothing switching between them but them. I'll keep comparing.
Jolida is making it clear, if it wasn't already, that so many factors make a difference and that different combinations of cPlay settings, step version or individual item changes, etc, can 'synergize' to reverse or alter these differences for any one system--to say nothing of different systems with different rooms and equipment, etc--that it is like chasing your tail trying to pin down what is best generally or in any linear incremental way.
For instance, I am using network versions of all. That makes a considerable difference in step 5 in terms of reducing the number of .sys .dll and registry entries eliminated. Wouldn't one expect that network 5 and non-network 5 would sound different? Which were you using in finding 4.6 best?
In my system (no network, USB-> battery powered USB isolator-> Devilsound dac) 4.6 and 5.5 (steppe 5 up to step j) have the same audio characteristics--analog like and realistic human voice, full bass etc. though the former has a bit more.
But the difference between 5.5 and the full steppe 5 treatment is huge. Yes, steppe 5 is clean, sharp, and the sonic image well defined with a spacious sound stage, but it does not sound natural, esp. the human voice, and the bass is too lean. I come to this conclusion by playing Van Morrison's "It's Too Late to Sop Now" though I guess any good live recording will do. VM,s record has mainly acoustic instruments (voice, horns, strings) and a lot of space information. With the complete steppe 5, the natural acoustic space just collapsed and the horn and human voice very unnatural and artificial.
KC
I'll work on it and let you know what results for me.
Influenced by your most recent comments and facilitated by all your work I have set up three partitions on my ssd: 1. Basic, original cMP 2. cMP after the fifth step and 3. cMp after the 16th step.
After my first round of listening, I find 2. to be superior to the others. It has scale and dynamism along with spatial accuracy and openness better than 1. though still without a sense of sacrificing tonal weight as 3. does. It sounds the most accurate and clear as well.
I will see how this seems over time and repeated comparisons.
Hi Jumaid
I've covered your same road, I've rebuilt cMP2 in all my three systems and I'm completely in agrement with you.
I've only a problem with a really good sensation I've felt when I've tried step 4. I remember a beautiful sense of spatial reconstrution, with open and airy sound. I can't remeber a lean sound in these phase, but now I want to try again, with and without registry change, with and without Bold Fortune step 2. I think that cMP2 is really the best sounding system but it seems to me that we can improve something. Or almost I'll try to improve.
Now, thanks to your job, I can try without effort. Maybe it could be a waste of time but I think there is a little margin to improve SQ.
Daniele
Where is ASIO4ALL? I don't have any sound card and this message says "If you don't have an ASIO compatible soundcard, use ASIO4ALL."
----- Dan
"Where the hell's all the f***ing kangaroos?", Thelonious Monk in Australia, circa 1965
Got to to do an internet search for it then download it.
I want to offer the tweak in addition to the Steppe's tweaks.
My Tweak for COMPLETE/FINISHED system in which nothing should change. The idea is to make the fullest possible delete all that is not required for audio playback.
The size of the folder Windows became equal 58.3 MB. The sound is detailed and smooth. The mode Start cMP is used only. I don't use the mouse and the keyboards. I don't use any usb devices. I use english and russian fonts.
I shall explain that I remove files and folders from the system drive of cMP2 using another operating system from the external hard drive. Thus you can always get access to the file system of the cMP2.It is necessary to remove following files and folders.
From the directory C and subfolders:
3com_dmi
1025
1028
1031
1033
1037
1041
1042
1054
2052
3076
Common Files
dhcp
Documents and Settings
DRVSTORE
export
IME
inetsrv
ras
RestoreFrom the folder Windows:
0.log
cmsetacl.log
comsetup.log
control.ini
desktop.ini
DtcInstall.log
EventSystem.log
explorer.exe
FaxSetup.log
GSetup.exe
GSetup.ini
iis6.log
imsins.BAK
imsins.log
NOTEPAD.EXE
MedCtrOC.log
msgsocm.log
msmqinst.log
netfxocm.log
ntbtlog.txt
ntdtcsetup.log
ocgen.log
ocmsn.log
OEWABLog.txt
regedit.exe
REGLOCS.OLD
regopt.log
SchedLgU.Txt
sessmgr.setup.log
SET3.tmp
SET4.tmp
SET8.tmp
SET25.tmp
setupact.log
setuperr.log
setuplog.txt
Sti_Trace.log
system.ini
tabletoc.log
tsoc.log
wiadebug.log
wiaservc.log
win.ini
WindowsShell.Manifest
WindowsUpdate.log
winhelp.exe
winnt.bmp
winnt256.bmp
wmprfRUS.prx
wmsetup.logaddins
AppPatch
Config
Connection Wizard
Debug
explorer
help
ime
Installer
java
msapps
pchealth
PIF
Registration
srchasst
TempFrom the folder system:
COMMDLG.DLL
IsUninst.exe
MCISEQ.DRV
MMSYSTEM.DLL
MOUSE.DRV
setup.inf
SYSTEM.DRV
VGA.DRV
WFWNET.DRVFrom the folder system32 (be careful with the files *.nls):
acledit.dll
activeds.dll
adsldp.dll
adsmsext.dll
adsnds.dll
adsnw.dll
alrsvc.dll
appmgmts.dll
appmgr.dll
bitsprx2.dll
bitsprx3.dll
browser.dll
browseui.dll
c_037.nls
c_437.nls (don't remove for English Windows)
c_500.nls
c_737.nls
c_775.nls
c_850.nls
c_852.nls
c_855.nls
c_857.nls
c_860.nls
c_861.nls
c_863.nls
c_865.nls
c_869.nls
c_874.nls
c_875.nls
c_932.nls
c_936.nls
c_949.nls
c_950.nls
c_1026.nls
c_1250.nls
c_1253.nls
c_1254.nls
c_1255.nls
c_1256.nls
c_1257.nls (don't remove for English Windows)
c_1258.nls
c_10000.nls
c_10006.nls
c_10007.nls
c_10010.nls
c_10017.nls
c_10029.nls
c_10079.nls
c_10081.nls
c_10082.nls
c_20127.nls
c_20261.nls
c_20866.nls
c_20905.nls
c_21866.nls
c_28591.nls
c_28592.nls
c_28593.nls
C_28594.NLS
C_28595.NLS
C_28597.NLS
c_28598.nls
c_28599.nls
c_28603.nls
c_28605.nls
catsrv.dll
catsrvps.dll
catsrvut.dll
cdmodem.dll
certmgr.msc
cfgbkend.dll
cfgmgr32.dll
clbcatex.dll
cmprops.dll
colbact.dll
comaddin.dll
comrepl.dll
comsnap.dll
comsvcs.dll
comuid.dll
credui.dll
cryptui.dll
CSVer.dll
davclnt.dll
desk.cpl
deskadp.dll
dgsetup.dll
difxapi.dll
dmboot.sys
dsquery.dll
dsprop.dll
ersvc.dll
es.dll
eula.txt
eventvwr.msc
fdeploy.dll
fs_rec.sys
hid.dll
hidserv.dll
iasacct.dll
iasads.dll
iashlpr.dll
iasnap.dll
iaspolcy.dll
iasrad.dll
iasrecst.dll
iassam.dll
iassdo.dll
iassvcs.dll
icaapi.dll
instcat.sql
ipmontr.dll
ipnathlp.dll
iprtprio.dll
ipxmontr.dll
ipxpromn.dll
ipxrtmgr.dll
ipxsap.dll
ipxwan.dll
jsru.dll
licwmi.dll
lmhsvc.dll
localui.dll
logonui.exe
main.cpl
mfc40loc.dll
mfc42loc.dll
mmfutil.dll
mprdim.dll
mprmsg.dll
mprui.dll
mqad.dll
mqcertui.dll
mqdscli.dll
mqgentr.dll
mqise.dll
mqlogmgr.dll
mqoa.dll
mqperf.dll
mqqm.dll
mqrt.dll
mqrtdep.dll
mqsnap.dll
mqtrig.dll
mqupgrd.dll
mouse.drv
mqbkup.exe
mqsvc.exe
msdtclog.dll
msdtcprx.dll
msdtctm.dll
msdtcuiu.dll
msgsvc.dll
msi.dll
msiexec.exe
msimg32.dll
msltus40.dll
msports.dll
msratelc.dll
mstlsapi.dll
mstsc.exe
mstscax.dll
mtxclu.dll
mtxdm.dll
mtxex.dll
mtxlegih.dll
mtxoci.dll
newdev.dll
notepad.exe
ntkrnlpa.exe
nwapi32.dll
olecli32.dll
olesvr32.dll
olethk32.dll
perfts.dll
printui.dll
psnppagn.dll
qmgr.dll
qmgrprxy.dll
rasadhlp.dll
rasapi32.dll
raschap.dll
rasdlg.dll
rasmxs.dll
rasphone.exe
rasppp.dll
rassapi.dll
rastapi.dll
rastls.dll
rdchost.dll
rdpcfgex.dll
rdpclip.exe
rdpwsx.dll
rdsaddin.exe
rdshost.exe
reg.exe
remotepg.dll
rpcns4.dll
rpcss.dll
rsfsaps.dll
rshx32.dll
rsmps.dll
rtm.dll
RtNicProp32.dll
RTNUninst32.dll
runas.exe
sclgntfy.dll
scoru.dll
scrrnru.dll
seclogon.dll
security.dll
sens.dll
sensapi.dll
senscfg.dll
servdeps.dll
sethc.exe
setup.bmp
sfmapi.dll
shdocvw.dll
spmsg.dll
spoolss.dll
spxcoins.dll
srclient.dll
srrstr.dll
srsvc.dll
srvsvc.dll
stclient.dll
sti.dll
sti_ci.dll
strmfilt.dll
system.drv
taskkill.exe
tcpmib.dll
tcpmonui.dll
termmgr.dll
termsrv.dll
trkwks.dll
tsappcmp.dll
tscfgwmi.dll
tscupgrd.exe
tsddd.dll
txflog.dll
umdmxfrm.dll
unimdmat.dll
uniplat.dll
untfs.dll
url.dll
user.exe
userinit.exe
utildll.dll
uxtheme.dll
vbsru.dll
vdmdbg.dll
vdmredir.dll
vga.dll
vga.drv
vssapi.dll
vwipxspx.dll
w32tm.exe
w3ssl.dll
webclnt.dll
wdmaud.drv
wiadefui.dll
wiadss.dll
wiascr.dll
wiaservc.dll
wiashext.dll
wininet.dll
winlogon.exe.bak
wmimgmt.msc
wshatm.dll
wshbth.dll
wshru.dll
xolehlp.dllCatRoot
CatRoot2
Com
ias
Microsoft
MsDtc
ShellExt
spool
wins
xircom
From the folder drivers:
acpiec.sys
AppleCharger.sys
dmboot.sys
dmio.sys
dmload.sys
drmk.sys
drmkaud.sys
enum1394.sys
fs_rec.sys
HECI.sys
hidusb.sys
hidclass.sys
hidparse.sys
i8042prt.sys
imapi.sys
keyboard.sys
mcd.sys
mnmdd.sys
mouclass.sys
mouhid.sys
mssmbios.sys
ndisuio.sys
ndistapi.sys
portcls.sys
rdpcdd.sys
rdpwd.sys
secdrv.sys
sr.sys
tcpip6.sys
tdpipe.sys
tdtcp.sys
tunmp.sys
udfs.sysFrom the folder WinSxS:
x86_Microsoft.Tools.VisualCPlusPlus.Runtime-Libraries.Resources_6595b64144ccf1df_6.0.0.0_ru-RU_b372f0f
x86_Microsoft.Windows.Common-Controls_6595b64144ccf1df_6.0.0.0_x-ww_1382d70a
x86_Microsoft.Windows.Networking.Dxmrtp_6595b64144ccf1df_5.2.2.3_x-ww_468466a7
x86_Microsoft.Tools.VisualCPlusPlus.Runtime-Libraries.Resources_6595b64144ccf1df_6.0.0.0_ru-RU_b372f0f9.cat
x86_Microsoft.Tools.VisualCPlusPlus.Runtime-Libraries.Resources_6595b64144ccf1df_6.0.0.0_ru-RU_b372f0f9.Manifest
x86_Microsoft.Windows.Common-Controls_6595b64144ccf1df_6.0.0.0_x-ww_1382d70a.cat
x86_Microsoft.Windows.Common-Controls_6595b64144ccf1df_6.0.0.0_x-ww_1382d70a.Manifest
x86_Microsoft.Windows.Networking.Dxmrtp_6595b64144ccf1df_5.2.2.3_x-ww_468466a7.cat
x86_Microsoft.Windows.Networking.Dxmrtp_6595b64144ccf1df_5.2.2.3_x-ww_468466a7.Manifest
From the folder Fonts It is necessary to remove all files of that are specified on a screenshot (instead of Russian fonts with letter "r" probably it is necessary to leave yours):
Launch the program Autoruns and delete records in the registry for missing files ("Files not found").
Now the system folder Windows is looked so:
See all screenshots below.
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Live, iPad
Edits: 10/17/11 10/17/11 10/17/11 10/18/11 10/21/11 10/21/11 10/21/11 11/03/11
The Everest/AIDA (http://www.lavalys.com/) is very useful test program which can be launched from the cMP shell
(see here http://www.audioasylum.com/forums/pcaudio/messages/9/97080.html)
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Live, iPad
Serge! It's great! I couldn't figure out how and what nls files to remove. I have now only explorer.exe in windows root. I wish only that you post the procedure in a more detailed way. I know this is my sin, also. In System-drivers I have only mouse.drv. Will post my tweaks by weekend. Please, compare them with your own and let's think how to remove crypt*.dll files and security files.
Serge.
Hi Serge.
1. "I wish only that you post the procedure in a more detailed way." - I do not know what else particulars are necessary. All that I wrote it is possible to repeat safely in system similar to mine, that is with the same hard configuration and OS Russian WinXP SP2 Pro. (in OS with other language it is necessary to define correctly in practical way the files *.nls and the files from folder Fonts which need to be removed and which to leave). Eeasier to be guided by a necessary minimum of files and folders left in my OS (see my message http://www.audioasylum.com/forums/pcaudio/messages/9/97075.html) because earlier I removed still files which other users could leave in the system (these files were not in lists of tweaks and I removed them at own discretion).
2. The differences between my and your final tweaks will be defined by that I use cMP as shell and you use explorer in this kind. And still I do not use peripheral devices and I use LAN.
The files crypt32.dll, cryptui.dll, secdrv.sys, security.dll are removed w/o problems in my setup.
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Live, iPad
Edits: 10/18/11
I think I will take another run @ your 'My Tweak' stuff but only after Steppe publishes his new stuff so that they will truly be the last tweak. Appreciate your efforts.
Here are screenshots:
The only test program Everest which can be launched if needed is left. Certainly it is possible to leave any program but it is necessary to edit the file cicsMemoryPlayer.pth in notepad.(archive password: Serge)
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Live, iPad
ok I'm done would not boot w/o crypt32.dll. Serge did your cmp boot w/o this ...oh maybe because you boot into cmp it works for you.
Yes, Theo. My cMP2 runs fine w/o crypt32.dll. But the file crypt.dll should remain in system32.
Here are the content of the system folders of my system and screenshot of Autoruns (see link below; archive password: Serge).
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Live, iPad
Edits: 10/17/11 10/21/11
I did leave crypt.dll in system32. Its a mystery then why it works for you but not me.
I bombed out on another c_437.nls
I'm going to reload and skip these nls files.
Don't delete browseui.dll otherwise cant use explorer to get into C or other drives to further edit or to run cplay.
Once again. My Tweak for COMPLETE/FINISHED system in which nothing should change. The idea is to make the fullest possible delete all that is not required for audio playback. (anyone not will require access to the drive C in the Linn Klimax:)).
I shall explain that I remove files and folders from the system drive of cMP2 using another operating system from the external hard drive. Thus you can always get access to the file system of the cMP2.
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Live, iPad
Edits: 10/17/11
Just bombed out on c_1257.nls. So dont delete this one. Does anyone know how to safe start and get into these files?bombed out on c_1252.nls also ...don't know if my original post was based on my hand written notes (where often a 7 looks like a 2 or vice versa) or if both are needed.
Edits: 10/17/11
Sorry but I forgot to notify that the file с_1251.nls should be left for Russian Windows (as shown on a screenshot), but for English Windows it is necessary to leave instead of it the file с_1257.nls.(*.nls are Microsoft National Language Support Files)If the system is failed to get access to its file system I boot from the external hdd connected to esata port - it is very convenient.
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Live, iPad
Edits: 10/16/11
no problem I just had to reload my snapshot image of Steppe16.
OK.
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Live, iPad
Edits: 10/16/11
I see you are deleting WINLOGON.
Are you doing the AUTORUNS stuff and then returning to MINLOGON before deleting WINLOGON?
Why not?
I installed the program Autoruns upon the button T.Screen in cMP settings and launched it in the mode Start cMP.
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Live, iPad
Edits: 10/16/11
I thought I remembered cics saying something about it not working there. Who knows where I think I read this!
Glad to know it is not true,
After Greg's bringing up getting the controller out of the way - is this how you do it?
That is quite a list. It reminds me of a complete BOLD FORTUNE deletion with some extras just for good measure.
So you obviously still have your NETWORK capability?
What have you done with registry changes? You mentioned more AUTORUNS - have you kept to STEPPE's registry regimen?
Can you still access RUN and regedit with this configuration?
Bye,
The file JulaWDM.sys isn't used during playback in my system. (What drivers are used in the operating system currently can be viewed using the Everest/AIDA)."So you obviously still have your NETWORK capability?" - Yes!
"Can you still access RUN and regedit with this configuration?" - No.Full functionality is maintained in my setup without any concessions. I just deleted everything that did not participate in the work of my network player. ;)
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Live, iPad
Edits: 10/16/11
1. I cannot make out for sure what you are recommending with "From the folder Fonts It is necessary to remove all files of that are specified on a screenshot (instead of Russian fonts with letter "r" probably it is necessary to leave yours): " Do you mean: remove fonts in this screenshot except for ones with an "r"? English ones like the ones with "r" except for the "r"? or are these the ones you say to leave? Is it only the ones with "r" at the end or also marlett and micross?
2. My Weiss driver for DAC202 seems to degrade over time and need reinstall periodically. I gather that would not be possible after this tweak amd thus make this unworkable for me (save replacing the entire partition when that is the case). Or am I wrong?
3. You say it sounds "detailed and smooth" but don't say you feel it is a significant improvement or how much of one you find it to be. Comment?
1. You should leave the files in the screenshot. The files 8514fixr.fon, 8514oemr.fon and 8514sysr.fon should be replaced by the corresponding English, for example.
2. My Tweak be used in a finished system. After my tweek you won't be able to change/reinstall anything in the system but also how you nothing can change in operating systems used in the industrial network players type Linn, Olive, etc.
3. I find it hard to clearly speak of the degree of improvement because the sound before was good. It seemed to me that some of the sharpness/pseudodetailness has disappeared and the sound became very analog but very detailed. In any case I would like to minimize the size of the system because I think this is going to benefit.
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Live, iPad
Serge do you know which files/folders may be associated with keyboard, mouse, usb and monitor? Some may be easy to identify (for example the phrase 'mouse', 'keyboard' or 'vga' in the name). But others are not.
Hi Theo!
Perhaps the files mouse.drv, mouhid.sys and mouclass.sys are associated with the mouse, and the file keyboard.sys isassociated with a keyboard. The monitor work after my tweek now. Perhaps usb ports stopped working even after tweaks Steppe (I haven't tested it, I have turned off it in BIOS and system). Perhaps the files hidusb.sys,hidclass.sys, hidparse.sys are associated with usb.
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Live, iPad
Thank you Serge.
Ever since I got to step 5, it reminded me as to how deceptive changes could get. I was lucky to have taken a back up at this stage, so that it's state could be exactly revertible. Going all the way to steppe 16 was most definitely, undeniably, an enthusiastic experience, even though it was an ordeal to almost all of us to have reached till there. With innumerable file deletions, settings & registry mods, the sonic balance did change to a very great extent, later to realise that it was more of a tilt than an overall improvement. At a certain stage (steppe 5.5, if I'm not wrong), the emphasis was more towards bringing the 'Air' out of the music. Each step thereafter brought about more n more transparency, more ambience & as I would say 'moisture' in the presentation. Agreed..!!
But.. !! As each steppe was done, & heard, we perceived a change, which had a sense of 'something new'. And it was said that each steppe be given some time for the so-called 'burn-in', which in another aspect made us Get Used to that new sound, thereby over-writing the sonic registrations which we previously had in our ears about our familiar music. This happens a lot to any of us. Atleast it has happened to me, inspite of trying to be careful.
After xplite, things started to change directionally. It was, as mentioned above, given Most emphasis on Air & Transparency. This Open sound was initially appealing, but started to show the downside it brought as steppes progressed, getting very Analytical by steppe 16.
I had always liked Steppe 5, amongst all the steppes starting from steppe 2 to steppe 16. So I decided to swap over to steppe 5 to cross-check if I was right. And Yes, it was ( atleast according to me).. Listening to steppe 5 for two days, & trying with the Bold Fortune route definitely gave a Plus. It was atleast not as bad as what xplite did. The sound was warmer, yet open. And the best combo was steppe 5 with The part 2 of BF. With Part 3, the bass was different.
All along this way, for about one full month, ever since I has started off with deletions, I was listening to a certain set of Music Albums. It was yesterday, when i felt bored listening to the same music all along, that I wanted to pop in another album which had been quite a while since I heard.
Very surprisingly, I felt many sections Missing in the presentation. This was alarming. It was at this point that I felt something somewhere has been lost, altogether.
And there came this time, when I restored our Original Optimised cMP with b38... The one we had set-up as per cics recommendations. And what more could I tell myself other than "what have I done..!!!
The qualities of the original optimised cMp had their own class. The IMPACT in the bass, the SCALE of the piano, the ATTACK in the drums, the BLEND in the vocals, the DELICACY in the treble, has totally been compromised along this route of Software-level mods. The FULLNESS in the mids is totally gone past the initial Steppes. The only downside I could sense in this stage was a little boost in the Upper midrange & a little Upper bass. If just these two aspects get rectified, the whole spectrum will balance itself & the Air which is now veiled by the above two things, would automatically be released.
I mean how could we lose that Bone & Muscle just to get some ambience & transparency. The foundation itself is wronged here.
I initially did not want to post this, because it might directly or indirectly offend the hard-work Serge has put into all this. Two thumbs up for him. Hats off to his dedication in this project. It's actually not him who is to be blamed. He did it for himself & posted to make others benefit out of it, which is a good thing. But somewhere it went out of track... May happen to any of us..
I don't expect anyone reading this post to agree with me. I know that Most of u will certainly disagree with my findings here. But whatever I have mentioned above, are my own personal observations.
So summing it up all, I intend to spend some time with the original Optimised cMp, & figure out a line of progressive implementations, that will give the best of everything.
This post has become quite long. That's all for now...
Thanks for reading...
Junaid
Edits: 10/15/11 10/19/11
I agree with much of what you said.
When I got my regular, from WINDOWS disk, cMP optimized (what do you call it? What word should go here?) I was surprised by how good it sounded. The only thing that did stick out was that the high frequencies were now perceived to be grainy, very grainy in comparison to STEPPE 10. I never got to listen to anything past that.
I do not think you are being contrarian at all. There is always a compromise and as you put it this configuration (is that the word?) suits you better.
In my experience I perceived a greater impact and dynamics with #5 than with the baseline.
One can wonder are we doing things that are confusing the machine. I think of the AUTORUNS aspect which I assume is just a shortcut to registry changes. Of course, I do not know the details of these changes but one wonders if we are to do these file deletions and these registry changes, both STEPPE and BOLD FORTUNE, that we should not use AUTORUNS?
By the way, Mihaylov's tweak looks a lot like a full BOLD FORTUNE deletion regime. I went through number nine and lost my START button (again), RUN, and NOTEPAD. All of these can be lived without but one would like to know what it is that makes them go away so you could plan how to proceed with deletions and registry edits.
I continue to think there is much good in this idea. My mind finds it plausible that calming the veritable Tower of Babel that is WINDOWS in full has to make for a better device for playing music. This discussion is only a month old; there is much to be learned. The learning can be frustrating and is getting in the way of listening to music in my case.
I finally took the advice of doing images and made an image of the optimized basic cMP. I tried to use this image yesterday after a BOLD FORTUNE overreach and it was corrupted. Any ideas of what I did to do that?
Your comments are needed to keep this discussion going in the right direction. I must admit I was taking the easy way out assuming the more stuff you removed the better the sound would be. Rarely is anything that simple and your asking us to consider there is more to the process than simple subtraction.
Nonetheless, STEPPE deserves praise for getting us thinking about how to make this better. The forum has been exciting again. Something good is gong to come out of this. Just think how much we have learned about .dlls lately.
I think the BOLD FORTUNE approach is superior to XPLite. We just have to learn to tweak it for our purposes.
Bye,
I am not saying that the Original optimised cMp is the ultimate. What I'm trying to say is that, whatever Software-level changes we make, it's mandatory to revert back to the previous step to double-check if the upgrade is worth it. It's very easy to get carried away when we get to hear new details these changes make. An A/B comparison, I feel, must always be done after giving some time with the new sound. By this, I don't mean Burn-in. For eg. if we are happy with steppe 5 & then we delete some files, & if the new sound reveals more details & Air or whatever, we MUST go back to steppe 5 back again after a few days to be sure it's an Improvement. Else we would just keep going on n on leaving a lot behind. This was evident when I came back to the Original optimised cMp. The sound of the Original Optmised cMp may sound coarse or dry initially, because we are so used to the Ambience of the new steppes. But I stayed back & wanted to keep listening. And it made me feel, that there are so many aspects in the Original, which have been blown away with the mods we have made. The sound is fuller, more Blended, & Plays in the right Scale. Piano sounded more Realistic in terms of size & Scale. Also the IMPACT is lost after the initial steppes which I'm now hearing in the Original Optimised cMp. The sound of Barry White is so effective, which lost most of it's Body in the deletion process. The only thing I feel that needs to be rectified is the Upper Bass & the Upper Midrange. Because of the dominance of these two sections, it sounds coarse or grainy. And I'm sure, quite sure, that it does not need hundreds of file deletions & registry changes just to get those two things right. This is where our focus comes to play. We need to move very slowly & be extremely critical in our analysis. Just because we have all the Batches ready, does not mean that we go at light-speed implementing them. This will only lead to confusion & losing the sonic registrations we have in our ears.
For this reason, I have decided to choose only 3 tracks for this whole process. I am repeatedly listening to them so as to grasp the reproduction of each note being played.
Now at this point, where I have decided to start-over from the Original version, I intend to proceed very carefully, and go back & forth till I'm very certain, that the new sound is absolutely superior in every aspect to the previous one. I don't know if I would follow the order of the Steppes which I did before. In fact, I don't even intend to do each steppe fully. I know this may seem like a long procedure, but I feel it's high-time we get back on track. The qualities of steppe 5 are much superior to the Original version, but the performance of the Original version has so much more than Steppe 5.. The bottom line is that So far, every improvement made has been born with a compromise. And it's this compromise that I want to eliminate...
Junaid
The night before reading your post, I was listening to Van Morrison's "Too Late To Stop Now" (Remastered) and was very unhappy with the sound, no bass, too thin, and the human voice too artificial. Another bad remastered disc, thought I.
Then the next morning I saw your post and since I was never totally happy with any tweaks after 4.5 (steppe 4 + your first batch) I decided to go back to it. However, I discovered that I had deleted that partition image for lack of space! The latest version that I had was steppe 5.5 (steppe # 5 up to step (j)) so I restored that. Comparing steppe # 5 complete and partial (up to step j) is very educational. Yes, the partial brings the bass back, so is the bloom of a natural soundstage, and the human voice is human again. While the complete #5 has a very 'clean' sound, an expanded soundstage (quite intoxicating), tight but thin bass, and an edginess to the human voice.
Yes, the transition to artificiality begins with steppe 5.5.
KC
Hello, everybody. This is tweaks part 21 and this is the first illogical and strange thing that I promised. Initially it was intended as a tail-end to part 20 but with the possibilities it had opened, I thought I should post it before, especially as it is large and significant enough to be an independent tweaks part. Also I feel that some inmates are becoming restless and possibly bored with part after part of just removing files without a “BIG” advance. Really, inmates, you are wrong because removing those files you are cleaning the ground for future tweaks (17-20). Yes, I know that from part 12 it may sound sometimes slightly grainy. Also I know that some instruments are shifting places – this is because processor and memory are getting less and less things to do and there is less and less electrical phase shift (you may laugh at this explanation, but remember that Cics implemented the phase shift button in the player. He is a far-sighted person and he knew what he was doing), but again I want to reassure that all grain will settle down with the soundscape becoming very clean and smooth.
Now the tweaks part 21. It’s all about BIOS. Not very dangerous, but be attentive and prepare a small screwdriver or the white jumper to shorten the cmos pins if the things go suddenly wrong.
They won’t IF YOU ARE CAREFULL!
Switch off your CMP machine if it is working. Reboot and enter the BIOS setup. If you have GAG31ms2l or, like me, have GA-H55M-UD2H mobo you have a wonderful capability to save bios profiles and load them without any trouble. F11 and F12 buttons serve to do it. So you have entered bios from the state of cmos that you usually listen to your music at. It works rock solid, I presume, so SAVE it now pressing button F11 an name it CMP 001. Exit Save utility and enter PC HEALTH STATUS, find there CPU SMART FAN CONTROL. Enable it and enter CPU smart fan mode. Set the mode to VOLTAGE. Get back to CPU SMART FAN CONTROL and disable it. Below, blued out should remain voltage. Exit PC Health Status.
Next Go to MOBO Intelligent tweaking M.I.T. (on the top). Enter and enter Advanced Frequency Settings (second from the top). Find third from the top Advanced CPU Core settings. Enter it and change all AUTO values to DISABLED. This will look like core 1, Multithreading - enabled, all other 5 settings DISABLED, not Auto. Mark for yourself C3/C6 state support and CPU Enhanced halt. You will possibly play with them later. Now exit Advanced CPU Core settings, exit Advanced Frequency settings and go to Advanced Voltage settings. My cpu (540) has an in-built graphics core. Hope yours has also. Enter Graphics Core menu and set voltage to 0,650. Press F10, Save, exit Bios, the system will start to reboot, enter bios again set graphics core voltage to 0,625. Save, exit bios and after that let the system completely reboot, but don’t listen to the music. Tell windows to shutdown-restart and enter bios setup again. Enter MOBO Intelligent tweaking M.I.T. - Advanced Voltage settings. With graphics core voltage reduced to 0,625, enter CPU Vcore end set the voltage to 0,68125. Save and exit and enter bios again and so on, set vcore voltage to 0,66875. Save, exit (I hope your system starts, mine does.) enter bios and press F11. Save current settings as cmp super (or cmp 002), exit save utility and back to M.I.T.. Set Graphics Core voltage (don’t mix things) to 0,612, Save and exit, enter bios, set Vcore voltage to 0,65625. Don't change the order of lowering down the voltages! Don’t attempt to go lower with voltages of both Vcore and Graphics Core. DURING ALL THESE TWEAKS YOU SHOULD NOT ALLOW THE SYSTEM TO SHUTDOWN!!! Save the bios values as CMP extreme or cmp003 and now go to M.I.T. Advanced Memory Settings. Set DRAM Timing Selectable (SPD) to Expert. And now go to Channel A Timings. Here we tweak timings of our memory as follows:
This is the table of timings and how I changed them (a dash instead of a value means that I didn’t change the value in this time. Dash between values serves to separate restarts (about 8 steps) but not complete reboots) Save by pressing F11 to cmp extreme 2 (cmp003) often and restart but enter bios and go on.
TRAS 11 - 10 – 8- - - 7
TRRD 2
TWTP 14 – 12 – 10 – 9 – 8 – 7
TRFC 34 – 30 – 29 – 25 – 20 - 20 – 17 – 14
Static TRD - - - - - - - 5
TFAW - - - -7
TRC- - 11 - - - 8
TWR - - 9 - -6 - - 3
You Don’t need to follow this table exactly, but I recommend to follow it more or less.
This is not all. After You have achieved final values, saved as cmp extreme 2 (cmp003), you should know, that the system wont reboot from the “cold start” with these values. So reboot and listen to the music. Let the system burn in for 2 hours. Before attempting to shut down ALWAYS RESTART and ENTER BIOS SETUP. Switching the machine on should start with rock solid settings and only after 5 minutes of playback reboot to cmp extreme and after 15 minutes more to extreme 2. Otherwise the system will fail to start. Press F12 and load CMP001 (your rock solid values). Reboot and then shutdown completely. I Know it is a lot of trouble, but it is worth it. Technically it is like a soft on and soft off with big power amplifiers.
Serge.
P.S. The promised parts 17-20 are soon to be posted. I simply want to finish with Windows in part 20, but not yet sure how.
Steppe,
Appreciate the effort and all the hard work. But if you are going to take us down this path, could you at least throw us a bone and include a screen shot or 2?
I THINK I get what you are saying about the memory timings, but that chart is a bit confusing. And it might help those of us who have different bios settings as each mobo is slightly different. For instance your "all other 5 settings DISABLED, not Auto" might be confusing. What if we have 4 or 6 in our bios. It would be nice to see exactly what those are called so we could account for off recipe hardware.
Please take this as constructive criticism and not any kind of detraction from your incredible work.
No one here remembers the bending of our minds
Aren't you saying in effect ...just lower voltages as much as possible while still maintaining stability? For those of us with e7200 intel cpu/gigabyte mobo's ... most of are there allready.
I have been happy with JULI@ but I have not heard anything else.
I like the fact I can access the I2S signals and have been using K&K AUDIO's version of the PSAudio system to take the signal from card to DAC.
I have soldered caps to the SMD pads on JULI@ and they are starting to fatigue after having to re-solder a few times.
Is there a compelling reason to replace this?
I like the fact that JULI@ is inexpensive enough to play around with. I scrapped the contact material from the "pins" so I can power it externaly but wonder if I would have the same bravado to do this to a six hundred dollars device.
So, I know I am asking the near impossible question - would a LYNX card be better than an optimized JULI@ setup. If anyone with a LYNX responds I would be curious to know if you have made any mods to the precious thing!
STEPPE using this card has got me thikning. Along with the fact that the LYNX needs fewer files to work compared to JULI@. One wonders if "we" (JULI@ users) are hearing what he is hearing?
Any comments are appreciated.
Rick,
I thought more about your question of whether the Lynx card would be an upgrade to your use of a Juli@ digital section as an interface to your DAC. After doing some research (including reading Mark's & Serge's comments below), here's some thoughts...
1. If you were just going to use the analog out of one of the cards or the other, one of the Lynx cards would almost surely be better.
2. There doesn't appear to be any easy way to get I2S out of the Lynx card, so you'd be using SPDIF out of it. I believe that a well-implemented I2S is better than a SPDIF connection, so you are ahead using the Juli@ with I2S if it's done right... and I think your K&K Juli@-> I2S & the galvanic isolation in the K&K DAC make it a good I2S implementation. So I think you're already ahead.
3. As you noted, the Juli@ is pretty easy to modify and at <$150, not so expensive that you worry excessively about breaking it. So you can do additional things to improve its effectiveness that you are less likely to do to one of the Lynx cards because of their cost (and as best I can tell, higher level of complexity).
4. As Mark noted below, a Juli@ with some basic power mods and a galvanically-isolated SPDIF TOSLINK connection equaled or beat the Lynx cards. You already have equivalent mods on your Juli@ AND a galvanically-isolated I2S connection. So (without being able to easily compare) your modified Juli@/DAC setup should be equaling or beating a stock Lynx/DAC setup.
All of that is to say, you're probably not going to do better with one of the Lynx cards for your setup.
I hope this makes sense!
Greg in Mississippi
Everything matters!
Hi Rick,
I owned and used the Lynx L22, the Lynx AES16 and the Juli@ digital part for at least over a year and was able too compare them in a cMP setup exactly (!) as per Cics recommendations except for:
- I’m using a LAN connection to my Buffalo NAS where I store my music files.
- the cMP setup was powered ‘hybride style’: Linear PSU -> P4 & Linear PSU -> Pico
All these three cards I only used as a digital interface to my Lavry Black DA10 DAC.
As per Cics recommendations I started with the juli@ digital part with an RCA S/Pdif connection to the Lavry DAC. This sounded okay. But not really better than an RCA S/Pdif connection from a EMU USB 0404 to my Lavry DAC or better than the RCA S/Pdif connection from a EMU PCI 1212 or better than the RCA S/Pdif from a RME9632.
So I looked for interfaces which use an XLR AES/EBU connection as this is theoretically a better connection method being balanced and also transformer decoupled by specification.
So I ended up with the Lynx L22 and the Lynx AES16 connecting via XLR AES/EBU too the Lavry.
Both Lynx cards sound the same too my ears. I could not detect any sound quality differences between the Lynx cards when used as a digital interface.
Connected through XLR AES/EBU both Lynx cards sounded better than the Juli@/EMU/RME connected through RCA S/Pdif. And yes I did carefully check if I was using an 75Ohm video coax cable with 75ohms Neutric connectors for the RCA S/Pdif connection.
Lynx drivers only ad some 2 to 3 latency in the Thesycon DPC Latency on a constant, non-fluctuating level. The juli@ drivers however ad 6 to 10 extra latency, which levels are also fluctuating much more.
Although both Lynx cards through XLR AES/EBU sounded better than there RCA S/Pdif competitors, the differences where cleary audible, but where not shocking. A linear PSU on the P4 has a bigger impact.
Having friends in professional music and audio production, I find myself lucky too being able too hear my favourite music on top notch DA converters and monitor speakers a few times a year. From this I knew that I still wasn’t getting maximum SQ possible from my cMP setup. Of course at every occasion I asked studio technicians about what too change in my setup to get better SQ, but unfortunately I got all kind of different advices. The only consistent advice I got was: what ever you do, it's not your Klein & Hummel O300’s.
So I could not think of any thing else left, than too follow Cics advice and use a good quality Toslink cable between the Juli@ digital part and the Lavry DAC. This was the only advice from Cics which I had not implemented yet. I already had tried a toslink connection with the plastic toslink cable that came with my RME 9632 card, but this was disappointing. So I dismissed a toslink connection.
So I decided too buy a Van den Hull Opto-coupler toslink cable. And whow !!! This was a real jump in SQ. I now finally could hear what Cics ment. After a month of intensive listening I decided too sell my Lynx L22 and AES16.
After a few months I did juli@ sound card modification number 1: Providing cleaner power (picture @10pin header). It indeed toke away some glare in the HF regions.
And only recently I applied modification number 2: desoldering the U1 voltage regulator and feed the Juli@ digital part with a better separate 3.3 V Linear Technolgy voltage regulator with extra input and output filtering of this regulator.
As also reported by others: unsoldering the U1 voltage regulator and feeding the Juli@ digital part with better power, gave a major jump in sound quality. It almost equals powering the cMP setup with a hybride power supply. So a real major step up in SQ.
So in short:
1. in my setup the juli@ digital part is significantly better than a Lynx digital interface, but only when used via a high quality Toslink
2. unsoldering the U1 regulator and feeding the Juli@ digital part with better/cleaner 3.3 volt improves SC significantly.
Hope this was informative.
Mark
fully optimized cMP2 PC -> ESI Juli@ -> Van den Hul Optocoupler MkII-> Lavry Black DA10 -> XLR Mogami Gold -> Klein & Hummel O300
Hey Mark. I use 5 LiFePo4's in parallel for juli@ 3.3v input. When I was trying to rid my system of metallic noise I briefly tried cut pci fingers and 5 volts from a 12 v lead acid battery regulated down to 5 volts. This scheme did not rid me of metallics but wow did it sound great and I mean great. However for other reasons I trashed the sound card and abandoned external 5 v feed to (a new) Juli@ because I solved the metallic issue. Now I am thinking of going back after the external 5 v feed. Greg has reminded me of how good it was.
Hi Theo,
Thank you for your response.
Just to check if I understand your post correctly.
You first powered the juli@ digital part with 3.3 Volt through battery’s, but the 5 Volt was still coming from the pci socket (power situation 1).
Later on you cut the 5 Volt PCI fingers and also powered the 5 Volt from battery’s. So both 3.3 V and 5 V are provided through battery’s (power situation 2)
Now about the sound quality.
Are you saying that changing from power situation 1 too power situation 2 again gave a (big?) sound quality improvement?
Mark
fully optimized cMP2 PC -> ESI Juli@ -> Van den Hul Optocoupler MkII-> Lavry Black DA10 -> XLR Mogami Gold -> Klein & Hummel O300
Yes a big sq pop again.
Hi Theo.
Looks like I have too build myself a second DIY voltage regulator.
This time a 5 Volt one.
Building a 3,3 Volt Voltage regulator was my first electronic DIY project ever.
And… it turned out too be surprisingly easy.
I deliberatly build a prototype with only the LT regulator and 2 caps on the PCB.
Like this:
Van on filtering, de-coupling, demping, ect
This construction with P4 connectors allows for easy swithing in and out all kinds of filter types, before and after the DIY regulator.
Van on filtering, de-coupling, demping, ect
And guess what?
In my setup there is no sound quality difference between feeding the 3.3 V DIY regulator with 5 Volt DC coming from a Pico or 5 Volt DC coming from a linear PSU.
Provided (!) ….. that you filter the input too the regulator board.
Ripple rejection of the LT regulator is arround 75 dB at 1000 Hz but only 15 dB at 300.000 Hz.
(the Pico swithing frequency = 300.000 Hz)
But if you filter this 300.000 Hz swithing frequency before the LT regulator,
than there is no sound quality difference between 5 Volt used from the Pico or 5 Volt from a linear PSU.
I (mis-)use a simple Schurter 5500.200 netfilter for this.
Specs see data-sheet: http://datasheet.octopart.com/5500.2014-Schurter-datasheet-2547.pdf
The Schurter netfilter has 2 coupled ferrite based 40 mH inductors inside and filters from 100.000 Hz too 10.000.000 Hz with 50 dB.
When you ad extra (Oscon) caps on the outside, the 2nd order effect (filterslope) can be bettered/altered/tailord
I haven’t used any batterys and thus I don’t know this for sure, but from this experiment (no diffrence between a linear PSU and a switching PSU when the input is proparly filtered) I think there is no need too use a battery as a source for digital components. The noise level only needs to be on a low level. It doesn’t necessarily need to be approaching almost zero.
Since you already constructed DIY regulators before, give it a try and feed your DIY regulator with filtered input from a pico PSU or a linear PSU and compair that too a battery source.
Mark
fully optimized cMP2 PC -> ESI Juli@ -> Van den Hul Optocoupler MkII-> Lavry Black DA10 -> XLR Mogami Gold -> Klein & Hummel O300
Hi Mark!
Here is the standard schematic of the stabilized power supply unit.
The explanatory notes to schematics:
1. TV1 – the power transformer
2. VD – the Schottky rectifiers
3. C1 and C2 – the filtering capacitors, C3 and C4 – output capacitors
4. С2 and С4 - the qualitative film capacitors with small capacity (for example Wima) shunting electrolytic capacitors for filtering HF noise (the film capacitor capacity is around 1/100 part of the electrolytic capacitor capacity). С2 and С4 are possible not to install in case of use of electrolytic capacitors Black Gate, Elna Cerafine, Elna Silmic, Panasonic FC and similar.
5. R1 – the load resistor providing the minimum load current (5-10 mA) in the psu’s output.
6. U1 – the regulator.
In your case:
the filtering capacitor capacity (C1) should be not less than 4700 µF and C3 should be not less than 470 µF approximately. Oscon does not approach here.
R1=500-1000 Ohm
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Live, iPad
Is R1 mandatory in the design?? I have made my Linear for the Juli@ long back following this schematic, but without R1. Also the Filter caps I have used are 20000uf Elna Silmic II & output caps are 1000uf Elna Silmic II. The regulator I have used is a Dexa UWB series 3.3v in the linear psu itself & de-soldered the onboard reg on the juli@, & powered it through it's pin 10 & pin 8 straight with this 3.3v linear psu I made...
Junaid
= Is R1 mandatory in the design?? =
No. R1 is necessary for maintenance of stabilisation state of ps when loading is disconnected. For example if you want to measure correctly an output voltage of ps without loading connection.
= Also the Filter caps I have used are 20000uf Elna Silmic II =
You are cool :).
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Live, iPad
:)
Junaid
Hi Serge,Thank you for the schematics.
But it is not at all my intention too build or construct a separate high quality DIY linear PSU for several reasons.1.
I would need too buy a whole lot of extra other materials like: an extra transformer, rectifier diodes, housing/casing, fuse/holder, on/of swith, cables/wires, sockets/plugs, ect etc. I also would have too spend a considerable amount of time too for assembling and constructing a DIY linear PSU.2.
I don’t think it is always necesarry too strive for lowest possible power supply noise.
I can fully understand the need for the lowest possible power supply noise in analog audio.
But this is NOT ALWAYS necessary for digital circuits.
For example see this article.
http://www.analog.com/static/imported-files/rarely_asked_questions/RAQ_tamingADC.pdf
The article clearly demonstrates that lowest possible power supply noise is a ‘very nice thing too have’. But striving for lowest possible noise levels, is not always necessary.
It can also be a waist of time, materials, money, ect.
The article outlines a situation where striving for lowest possible noise levels is not necessary at all. The article explains that in given situation, power supply noise levels only have too stay below a certain level. Only TOO MUCH noise wil degrade performance. Striving for lowest possible power supply noise levels is not necesarry at all (given the situation in the article)Now back too the Juli@ digital part.
The logic on the Juli@ card will produce noise itself. And the Juli@ card will also receive noise through the PCI-bus from the MoBo.
Considering these noise levels: is it really necessary too construct an external DIY linear PSU with lowest possible power supply noise? Or is it enough/sufficient too construct an external power supply for the Juli@ digital part, that keeps power supply noise levels below a certain level?However not exactly knowing these noise levels, it is a safe approach too construct an external DIY linear PSU which has lowest possible noise levels. One never can go wrong with this approach.
But, ….. if one doesn’t want too build a DIY PSU or one is not capable of building one with lowest possible noise level, than may be also good results can be achieved when using a LT regulator as a post regulator after a switching PSU or after an off the shelve linear bench PSU.
The filtering effect of the LT regulator, together with pre- and post filtering before and after the regulator, might reduce power supply noise levels sufficiently.I guess for many inmates building there own high quality, low noise linear PSU, will be a ‘bridge too far’. However they may be capable of construction a simple (post) voltage regulator with pre and post filtering. Constructing such a (post) regulator, with pre and post filtering is rather easy.
(Also there is no dangerous 230 AC high voltage is involved.)Although when using these simple DIY (post) regulators, power supply noise levels are not lowest possible, the effect on sound quality is nevertheless quite remarkable (!!).
Hoping you are not the only one who is reading my posts, I post my flimsy constructed filters and voltages regulators too inspire other inmates, who cannot build a high quality linear DIY PSU, but who might be able too construct such a simple (post) voltage regulator.
It is defenately worth the effort and costs for materials.Mark
fully optimized cMP2 PC -> ESI Juli@ -> Van den Hul Optocoupler MkII-> Lavry Black DA10 -> XLR Mogami Gold -> Klein & Hummel O300
Edits: 10/21/11
Hi Mark!
I recommend to you to make a separate linear supply for Juli and to install RCA (or BNC) connector how is stated in the document (see the link below). Earlier I used just the same as your hybrid power supply and when I did that I wrote above SQ was refined by qualitative leap! I explain it to that the subsystem of Juli's power supply is bad and cannot cope to the right degree with an impulse noise from PicoPSU - first of all Juli's clock suffers from it. In this sense Lynx is better. Unsoldering the U1 voltage regulator is not necessary as it is an additional step of stabilization.
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Live, iPad
Edits: 10/18/11
Hi Serge,Thankx for your suggestion on how too further improve the sound quality from the digital part of the Juli@.
But I do not exactly understand what you mean and what next step you are suggesting.
I already did unsolder the U1 regulator.
So right now my Juli@ digital part looks like this.
Van on filtering, de-coupling, demping, ect I power it through a newly constructed voltage regulator based on a higher quality LT IC voltage regulator
(Too further improve DC quality I also extra filter the input too and output from the regulator bord. These filters are not shown on this picture)
Van on filtering, de-coupling, demping, ect My first question is:
- If I cut the 8 PCI traces on my Juli@ digital part as shown on the above picture, will my Juli@ digital part still function?
(it than will only receive 3,3 Volt DC from the external regulator)Or (second question)
- does my Julia digital part as shown in above picture (with 3,3 voltage from external power source) still need the extra 5 Volt DC when those 8 PCI traces are cut?
So: 5 Volt DC (from external power source) + 3,3 Volt DC (from external power source)?Mark
fully optimized cMP2 PC -> ESI Juli@ -> Van den Hul Optocoupler MkII-> Lavry Black DA10 -> XLR Mogami Gold -> Klein & Hummel O300
Edits: 10/19/11
Hi Mark!
I am not sure that Juli receives a power supply from the motherboard only through regulator U1. Probably that some elements on board Juli receive a power supply in the bus wire +5V from the motherboard directly or through other regulators (not U1). Therefore it is necessary to supply them from the external power supply cutting 8 traces on a board Juli.
Most easier to cut these 8 traces and will check up to work Juli. If Juli will not work it will be necessary to connect additional external power supply +5В instead of the power supply from the motherboard.
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Live, iPad
Mark and Serge,
First of all, I can answer the question about whether the Juli@ digital section needs 5v to operate. While I haven't tested it, I determined that 3 of the devices, the HC125AG CMOS Buffer/Line Driver (that appears to enable the switching of the SPDIF input and output signals), the PLT133/T10 TOSLINK output, and the AV73-01L are powered by the 5v. Without the 5v from either the motherboard or supplied by a separate supply, you definitely won't get a SPDIF output from either the RCA or TOSLINK. Actually, you probably won't get any function at all. Even though I cannot find any info on the AV73-01L chip, it sits between the Envy VT1721 (PCI interface) & the XILINX (local controller) and as such, likely plays a very key role in the Juli@'s operation.
See my long, rambling post here for more info: http://www.audioasylum.com/cgi/t.mpl?f=pcaudio&m=96880
Second, I want to thank you for your impressions on the sonic benefits of various modifications to the Juli@. While I've done a bit with the card too, almost all of my use has been with I2S-connected DACs and not using the SPDIF connection.
Mark, I especially was interested your comments on the benefits of using the Van den Hul Opto-Coupler TOSLINK. I do expect benefits with properly-implemented galvanic isolation from the Juli@ card (and I expect Van den Hul to properly implement it!). But I need to do more investigation on how to best implement it for the I2S connection.
Two asides... one, I have a bunch of caps coming to experiment with additional local bypassing/filtering on the power supply leads to the 'key' devices on the Juli@, the Envy VT1721 PCI interface & the XILINX local controller. Given they have to be very noisy chips, I expect there to be significant performance improvement potential here. Its a pity that one can't easily provide separate power to these chips, but the power runs are in one of the internal layers of the multi-layer circuit board of the Juli@.
Still, I hope this will provide a good improvement above just providing seperate power for the 5v & 3.3v on the Juli@ digital. I'll publish my results in the subthread I listed above.
Then, recently I saw a number of interesting discussions on DIYAudio dealing with wringing the best performance out of the ESS DACs. Many of the ideas and techniques discussed will have application on other DACs. These threads included:
http://www.diyaudio.com/forums/digital-line-level/117238-ess-sabre-reference-dac-8-channel-4.html
http://www.diyaudio.com/forums/digital-line-level/196474-ess9018-try-new-try-more.html
http://www.diyaudio.com/forums/digital-line-level/147817-ackodac-based-es9018-8.html
http://www.diyaudio.com/forums/digital-source/190033-upgrading-modding-new-oppos-bdp-93-bdp-95-a-2.html
http://www.diyaudio.com/forums/digital-line-level/183374-exau2i-multi-channel-asynchronous-usb-i2s-interface-17.html
http://www.diyaudio.com/forums/digital-source/188902-xmos-based-asynchronous-usb-i2s-interface.html
http://www.diyaudio.com/forums/analog-line-level/196049-dual-i-v-stage-implementation-buffalo-ii-iii.html
http://www.diyaudio.com/forums/digital-line-level/196552-buffalo-iii-dac-tube-output-stage.html
http://www.diyaudio.com/forums/twisted-pear/164294-buffalo-ii-transformers-5.html
http://www.diyaudio.com/forums/feastrex/197223-premium-finemet-transformers.html
Note that some of these threads are pretty long... the most relevent info is in the last several pages of posts. Also great to see Jonathan Carr, the designer of the Lyra Connoisseur electronics participating in these threads.
Again, to both of you, thanks for the very good info!
Greg in Mississippi
P.S. Serge, an extra thanks to you for your previous publishing of the details of upgrading the clocks on a Juli@. I have a dual clock coming from http://www.fidelityaudio.co.uk/clocks.html and hope to get a good upgrade with it.
Everything matters!
Hi Greg,
Thank you very much for joining this thread and providing the answer too my question !
Knowing the answer now, I’m still wondering if it will be worth the trouble of cutting the 8 fingers on the Juli@ that take 5 volt from the PCI slot. How much extra sound quality improvement will there be if I start providing 5 Volt high quality DC power too the Juli@ digital part from an external source?
Right now my Juli@ digital part gets:
- 5 Volt ‘dirty’ DC power from the PCI socket
- 3,3 Volt DC clean power from external source
Providing clean 3,3 Volt DC from an external source gave a very big rise in sound quality !
But will there be again such a big improvement in sound quality when I also switch too 5 volt clean power?
I would be very curious too know if there are any inmates who have compared the following two power configurations on their Juli@ digital part:
* Power configuration 1:
- 5 Volt ‘dirty’ DC power from the PCI socket
- 3,3 Volt high quality DC power from an external source
* Power configuration 2:
- 5 Volt high quality DC power from an external source (with the 8 pci fingers cut)
- 3,3 Volt high quality DC power from an external source
Until now I haven’t red any postings from inmates who did some experiments on this.
Probably most inmates changed over both voltages (5 V and 3,3 V) at once too an external source.
So the contribution of clean power too sound quality from each voltage separately, is still unknown.
I think I will wait for a while and see if there will be any response from inmates who might have done this experiment.
Mark
fully optimized cMP2 PC -> ESI Juli@ -> Van den Hul Optocoupler MkII-> Lavry Black DA10 -> XLR Mogami Gold -> Klein & Hummel O300
I don't think anybody else has as much knowledge & expertise in all sections of cMP hardware, like Gstew. He knows his work, & we can close our eyes & follow his mods, no questions asked. His work is a masterpiece...
Junaid
Rick,I can't answer your question of whether the Lynx would be better than an optimized Juli@. But I've been looking at the Juli@ again over the last couple of weeks with an eye to what is right about it and what can be improved. And I'm looking at what the next level of upgrades and modifications might look like.
For the the Juli@'s positives attributes, we have the following:
1. It looks to be designed pretty well, with a multi-layer circuit board and even stock, a fairly sophisticated power regulation and bypassing scheme. An example of this is how it is designed to work and sound good even though it's intended to be powered by the computer power supply via the computer motherboard power rails. That it works as well as it does in this environment means the designers did SOMETHING right!
2. It does provide a lot of flexability in being able to output I2S and SPDIF, along with being usable via it's analog outputs (in a pinch).
3. It can be modified fairly easily for better performance.
4. It is sufficiently inexpensive that one doesn't feel too bad taking the risk of killing the card while modifying.
5. And as a bonus, you have the K&K I2S interface which allows you to bring the I2S out of the computer and connect to a remote DAC. I'm not sure how well it compares to a short cable ( <3") and a close DAC (an option not available for most), but I have a lot of respect for the K&K guys (and the PS Audio ones too) and know that they wouldn't put it out if it was bad.
But the Juli@ does have the following drawbacks:1. It does a lot of things not needed if you are using it only for an computer-to-DAC interface, especially if you are using the I2S bus. This means you have components that you don't need that are potentially degrading the signal quality and is true both for the hardware and the driver software. It DOES help that you can remove the analog card and it's devices.... too bad you can't do the same for the driver processing associated with the analog output.
2. PCI looks to be end-of-life, as you found with the latest motherboards either not having a PCI slot or using a PCI-e to PCI bridge (which HAS to be bad for SQ).
I think that the Steppe & Jolida Windows Slimmings do some positive things to reduce the extraneous processing in the Juli@ (or any sound card) processing stream. I SERIOUSLY DOUBT that you'd be able to run the Juli@ Control Panel after you've gotten very far down the Windows slim-down path.
I also think that the mods that I and others have posted for the Juli@ digital section go a significant way in reducing the impact of the excess complexity AND also the impact of it being a sound card powered by the computer's power supply via the computer motherboard rails.
What I'm working towards right now are a further set of mods that can take the Juli@ a few notches better, building on the ones I've generally done in the past and the clock mods shown by Mihaylov.
I'll be starting a sub-thread over in the cMP thread about this soon.
But finally Rick, after saying all of this, if you are looking for the best available alternative today to getting an I2S stream out of the computer, look at an EXAU21. Reviews have been very positive (including a couple comparing it to somewhat-modified Juli@) and it costs $430 CDN, significantly less than the Lynx. AND being a USB interface, it won't have the PCI end-of-life problem. AND since it is designed to ONLY produce an I2S datastream coming out of the computer, it starts with a lot less extraneous complexity that can impact the SQ. And it will need less work than either a Juli@ or a Lynx to serve it's best in that role.
And I almost forgot to mention that the EXAU21 provides galvanic isolation between the computer and the DAC (and I don't believe the K&K I2S interface does this... NOTE that I was not correct on this, while the K&K interface does not provide isolation, the K&K RAKK DAC does have isolation at its I2S inputs. This is important as this is the DAC Rick uses).
Also, since it mounts at the end of a USB cable, it provides another way to bring the I2S to a remote DAC.
These are two more important benefits of the device.
The positive comments about this new interface are what have prompted me to take a closer look at how to get the best out of the Juli@.
This is probably more like $2.37 instead of my $0.02.
Greg in Mississippi
Everything matters!
Edits: 10/13/11 10/13/11 10/14/11 10/14/11
Thanks for the help.
I have not compared the two though it would be easy enough.
As soon as I get something working I find myself more interested in listening than playing further.
Have you tried using JULI@ without CONTROLLER installed? That would be useful since it is totally useless when one is using the LUCY Latency thing.
Knowing how Dave has incorporated galvanic isolation with the SPDIF input I would be surprised if the HDMI approach does not offer this also, but I have no idea. I will ask him.
I started out with USB and found JULI@ a revelation so I am hesitant to go back to USB. If someone I trust says it is great, THEN that's another story; I am not going to give it a try until then.
My JULI@ sounds fine for the moment. I worry about its future. The price cannot be beat. I see them now for $125.00. I think I paid more for my first one.
Rick,Sorry, I think I gave two wrong impressions.
First, I do think I2S is the better way to connect to a Juli@ as long as you do it right... either a short connection or using K&K's setup as you are. SPDIF has to go through additional devices and processing to get out of the Juli@ & of course, goes through a conversion back to I2S at the input of most DACs.
I do agree with you that Theob's I2S setup was likely not optimal.
Second, I was only assuming that the Juli@ Control Panel would not run after one got far enough into the Steppe Windows Slimming-down... and Theob corrected me in that he has gotten through Steppe 16 and still could access the Juli@ Control Panel.
Then on galvanic isolation with the K&K I2S connection, while the connection circuits are NOT isolated, the RAKK DAC's I2S input IS. This is from pages 6&7 of the manual... and it has some additional grounding/isolation information that might be useful, but that I need to digest. I do intend to try isolation on my I2S lines at some point, but I plan to work other areas first including doing additional filtering on the motherboard (Ala JackWong) to reduce the need for that isolation.
And that your setup already has galvanic isolation on the I2S lines MAY mean that you are already getting better sound quality using the Juli@ than the rest of us with lesser DACs.
Finally, while I instinctively like PCI over USB as an interface, I believe that any number of recent USB interfaces, especially with some modifications, will work well. And as at least two people have commented, they found the EXAU21 a significant improvement over their modified Juli@ cards. I still think that the Juli@ can be raised to that level. But I can't discount their impressions.
(And for a very informative set of comments about the relative benefits of different interfaces for getting music out of a computer, read John Swenson's comments in this thread: http://www.audioasylum.com/cgi/t.mpl?f=pcaudio&m=90268. If it doesn't make your head hurt, you are either much smarter than all the rest of us like John appears to be or you just didn't read it closely enough!)
Later!
Greg in Mississippi
Everything matters!
Edits: 10/14/11 10/14/11
Happy you are back into music...what Steppe are you at? I had been using I2S (for about a year) but went back to spdif which I prefer. However I'm sure my I2S implementation was not the best.
I kept screwing around with "cleverness" and then took Junaid's advice and did it the old fashioned way.
Listening to STEPPE 5 last night to get a baseline and tonight I will go further.
Yes, I do think the implementation of I2S you were using was dicey.
I am intrigued by "losing" CONTROLLER. I guess I could give it a try but if Greg has already done the experiment ... I am a little shy of experimentation at the moment!
I wonder if SOUND requires setupapi.dll? If not, and doing without CONTROLLER works, then it would be good to be able to delete setupapi.dll.
I am assuming you had to put that one back, too, to get your system to work?
Let me know.
Bye,
You said: '...I am intrigued by "losing" CONTROLLER. I guess I could give it a try but if Greg has already done the experiment ... I am a little shy of experimentation at the moment!...'
What Controller are you talking about?
You said: '... wonder if SOUND requires setupapi.dll? If not, and doing without CONTROLLER works, then it would be good to be able to delete setupapi.dll.
I am assuming you had to put that one back, too, to get your system to work?...'
I have put setupapi back in yes. I have left Sound and wdmaud.drv out. But wmdaud.sys is in. I believe it sounds better with this combination ims.
So u made a fresh installation of xp, made all optimisations as per cics website, & implemented the batch files getting to steppe 5 ?? I hope u went in this order :-
1. Optimised cMP
2. Steppe2
3. Steppe 3
4. Steppe 4
5. Jolida1
6. Jolida 2
7. Steppe 5...
Is that right ??
Junaid
do you think it would make any difference?
I finally made an image of the cMP Optimized WINDOWS! I will make another of this STEPPE 5 version tonight before going further.
First I deleted the folders you have in Step 1!
Seems to me as long as deletions and registry changes are close to each other it would not make much difference.
Consider xmiprovi which you deleted early on and STEPPE delete almost at the end of his published recommendations AND looking at the files deleted with Serge's and your recommendations compared to BOLD FORTUNE who recommends deleting many of these same files all at once.
Just curious why you think the order, at the beginning, has importance. Towards the end I would tend to agree there is some merit. Just an instinctual assumption. I have no real idea one way or the other!
How far have you proceeded with BF? Have you deleted further files after STEPPE 5?
THANKS!
Well well... I'm in a moment of shock at the moment. Did something that made me realise how deceiving a human notion can get. If I post it here, it would raise a debate, because it's way beyond our expectation. But it's an eye-opener & a wake up call that displays a message that reads " My friend, u r a human, & u r bound to get carried away"...
I know what I've written above would puzzle u as well as others who read this post. But it's about time we get to reality.
In the meantime, I suggest u stick on to step 5, as u already have. I will keep u posted !!
Junaid
I hope I did not say something offensive.
I cannot tell if you are having fun or being serious OR not really responding to my post at all O R have come across something close to "mind-blowing".
Talk about cryptic!
I will try to patiently wait for the explanation.
In the meantime I will make an image of FIVE and play with BF tonight.
Do tell.
Great analyses of the juli@. You bring up a real good point about pci as we go forward. I had been toying with the exau21 idea as a substitute for juli@ based on the great customer reviews on DIYaudio. You make a strong case for the exau21 vs another pci soundcard. One minor point though is I have gone pretty much Steppe 1-16 and I can still access the juli@ control panel fyi. What is the K&K I2S interface ?
Thanks Theob.
The K&K I2S interface is a couple of small circuit cards that transmits and receives the I2S signal across an HDMI cable. K&K Audio makes it (they also make the DAC that Rick McI uses). You can get more info in these threads:
http://www.audioasylum.com/cgi/t.mpl?f=kandk&m=8778
and
http://www.audioasylum.com/cgi/t.mpl?f=kandk&m=8609
and
http://www.audioasylum.com/cgi/t.mpl?f=kandk&m=8522
BTW, I forgot to mention the galvanic isolation between the computer and the DAC provided by the EXAU21 and also it's inherent ability (using the USB cable) to bring the I2S to a remote DAC... two other important benefits. I'll update my post above with these.
Later!
Greg in Mississippi
Everything matters!
Thanks I also saw your other post.
Edits: 10/14/11
Finally the job is done. I managed to finish the Registry as well. So now we have ALL the Tools ready to Review their impact on this project.
For those who want to start-over from scratch (the original cMP installation ), i suggest its about time u get handy with back-up tools like Acronis or Snapshot, so that if at a particular stage, u like the performance of a certain combination, its mandatory to Back it up, & then move ahead to the other sections. This makes less room for errors, since u can restore back to that exact state where things sounded best so far..!!
I have tried my best to edit entries in all these batches, so that USB thumb drive functionality is maintained even after part 16..
Move slow, & observe the changes. Just because one has all batches ready, usually is in a haste to jump to the next stage in minutes. Please try to avoid that, & give some time, NOT FOR BURN-IN, but to analyze what has changed, or 'WHETHER IT HAS CHANGED, OR IMPROVED..."
Once we are able to observe That Change from an Improvement, we cant go wrong..
Junaid
Hi Jumaid
you are a great man!
A lot of GRAZIE for your fantastic job, you are un example for all forumers.
Your batch files works properly and FAST! Now it will be easy to find best solution for everyone, in term of best step to stop.
I've deleted step 12, very disappointing, and I'm come back to step 7.
I've found that keeping msacm32.dll this improve SQ a lot and allows an excellent tonal balance. Very pleasant sound.
Thx again
Daniele
PS if possible send to me the regedit batch, please....
Since i have finished creating All the Batch Files for all Steppe's, Networking, & Jolida lists, the only thing left was moving on to the Registry. Trust Me...!! This is AS painful as doing the Batches. I am through with them, say just 20% of the work left. SO MUCH data involved in it as well. I have made all the Registry changes in One single file, which will execute them all at once. This way, the user need not intervene in the process. Gets done in a matter of say 3 seconds. But took 8 hours for me to write them :)
At this moment, im even unable to type. So exhausted. I think i sure need to take an X-ray of my neck. Im sure i have Batch modified its shape along with these files...
Junaid
the only thing we can offer is gratitude ... unless you can think of something more we offer in return!
Hope your neck starts feeling better soon.
Thanks for your gifts.
Those who have gone past step 15, try restoring taskman.exe to the root drive & also everything back to the Win_sxs folder...
Also when going with the Bold Fortune route, be careful in it's Part 5, so as to not implement the win_sxs folder deletion & with taskman.exe.. I tried & it sucks out the ambience & body of the instruments !!
Junaid
Taskman.exe appear again in BF Part 10:Listed Alphabetically.bat file so watch out for it there.
There is a REGEDT32.EXE in BF Batch Files\PART 10 Individual system32 Files Batch Files\Individual PART 10 Batch Files\Various Windows Applications.bat.
System would always boot to XP and not into cMP even when the option Start cMP is chosen if this file is deleted.
:)
Junaid
Junaid
Must agree. Restoring the WinSxS folders brings texture to instruments. I find it more emotionally involving. My toe is taping.
While admittedly skeptical about these recommendations (I tried restoring other files like wdmaud before and did not like) I tried them and yes they are directionally what you say. Adds body and and an aliveness that is not there without this restoration. I tried because I believe your dedication to sonics and although I did love Steppes 8-16 carte blanche its now better with these file restorations. Thank you Jolida!
Theob,
We have no idea how many files, mindlessly removed, have contributed to what... Seems like we still have a long way to go, unless we go back ten steps & do the same but in increments. I mean not each steppe fully, but by slicing each step into sub-divisions, so as to maintain our focus on the sonical differences it makes...
Thats exactly why i jumped back to step 5, because its superiority over the other steps was apparent, when i proceeded thereafter. I am following a different route from that point & trust me, Its Worth it..!!!
Junaid
I just want us all to be content with the outcome of our cMP. It's upto all of us, as to how we put these batches to use with various combinations & achieve something which we will all cherish, for a long time to come...
Junaid
A huge thank you from me as well. Your work makes it possible for us to experiment.
I ran your batch files for step 2 to 7 including the steps new steps after 4. I had missed quite a few files in my manual application. This brought back the air and texture that had been exchanged for detail in steps 9-16. I look forward to trying these next suggestions for fine tuning.
I have created an entire set of Batches from Steppe 1 to 16, for Network users of cMP. Thought it might be useful for those using Network functionality to have a seperate set for themselves.... I have already emailed it to users who had requested me for my previous Batches...Edit:- How stupid of me. I made batches for all the steppes but forgot to create the ones for the two lists i had posted (Jolida1 & Jolida2) myself. Please excuse me for that. I just made batches for the two & added them to the archive. Please find attached sets of all the steppes including Jolida1 & Jolida2 in ur email...
Note that my two lists were implemented after Steppe 4 & before Steppe 5...
How forgetful can one be... :)
Junaid
Edits: 10/12/11
Hello, everybody. I am here today only for a short while. I have lost my cat and I am sorry that she didn't have an opportunity to listen to Diana Kall after tweaks part 19 and 20. She loved her voice, btw. I will post these tweaks shortly, but to do it, I have to switch in another hdd. No Notepad anymore. I'd like to point out that I initially suggested to change dvmt memory for 256 mb AND then to 128. I guess it works only for 1 gb memory. Also, if you would think the sound a little strained, don't be alarmed. After part 18-19 you will need less than 20 mb for windows. My windows folder is 79,5 mb and the sound is smooth like a calm after the storm. Even Alice Cooper is not a nightmare anymore. The step from 16 to 20 is really significant.
TO LYNX USERS and maybe for everybody. There is a dll called mfc20 or mfc 40. You won't miss it it's the only one with a name like this. IT IS necessary for changing Lynx mixer settings, but not for the sound, so don't delete it permanently. Also, after part 12-14 don't exactly know when, Lynx output will change to +4 dbv from -10 dbv (fit for consumer electronics) To reset it you need the abovementioned dll, that can be removed once you checked your driver.
Strange and illogical would be the things I propose in parts 17-20, but they seem to work. What You think is slightly too much now soundwise is a slight strain of the last effort to come.
Serge.
Edits: 10/11/11
Sorry to hear about your kitty. Take care. Remember all the life/joy she brought you. If you post her picture maybe we will all remember her too. I will listen to Diane Krall today in her memory.
After numerous attempts to get a good install disk of nLite over the weekend I thought I would give warning that if you want to go with the full STEPPE regimen you will likely find disappointment, as I did.
For all of the attractiveness of eliminating stuff at install, along with the magic of a fifteen minute install and the elimination of many optimization steps, it just will not allow all that is required.
If you go truly minimal, no AWE. If you go with a disk that allows AWE then you run into problems further down the line.
Gave up and just did it the old fashioned way.
Found that one MUST do XPLite when Serge said to because after ocmanage.dll and WBEM folder is deleted it won't work anymore. No big deal you just put them back in. It lets you know what is missing. It does not tell you one of the files it wants is in the WBEM folder. Good to have another machine around.
I did most of my XPLite after the cMP optimizations as an easy way to get WFP disabled and figured it would eliminate a few RED changes even though you still have to look for them just to be sure. I went too far with file and folder deleting before finishing up and found what I said above.
What a way to spend a beautiful weekend!
I went all the way to steppe 16 & was truly disappointed with the end result. So I backed out to steppe 5, which as I mentioned earlier, was the most Musical & Balanced overall. Xplite has caused a sure downside thereafter. No step after xplite can get that Balance corrected. Steppe 9 just makes it better of the worst.
I am now following the Xp slimming method by Bold Fortune after steppe 5, & it's in an entire different league altogether. The sound is of a different flavour. I can easily hear an IMPROVEMENT, not just a CHANGE, which the steps 5.5 to 16 did. It's Worth to the core!!! I have just applied the Part 1 of the Bold Fortune's route, & it sounds so seductive & Musical. Wish I had done this earlier.
Xplite.... A big NO !!!
Junaid
> > > "I have just applied the Part 1 of the Bold Fortune's route,"
Did you just apply this part only:-Slimming Down Windows XP PART 1: Temp And Junk Files?
If you had created the batch files for this, would appreciate a copy from you.
Best Regards
My apologies. It was a typo. I meant to say I implemented just the Part 2 of Bold Fortune's xp slimming process. It's " Part2- Disabled services batch files".
This was what I did after restoring my cMp to steppe 5. One can easily sense how progressive the transformation gets from here. It does not alter the balance as xplite does.
I do not intend to move ahead with steppes 6 to 16 till I get done with the slimming process.... And I feel thereafter as well !!!
Junaid
Did you run the system32 Files Disabled Services.bat, and remove in one click, all 435 system32 files, for every Windows Service , or you only remove these Services individually, keeping some of them?
Regards
The Part I implemented consists of the following:-
1) disabling printers & faxes
2) terminating individual services (about 30 of them)
3) removing network connections
4) removing scanners & camera
5) removing schedule cast applet
4) removing 435 system32 files
These alone made a tremendous difference. The above mentioned is just one section of the slimming procedure. There are totally Nine sections in the slimming process.
Junaid
Thanks.
I have Part 2 to 10
2) and 4) are actually the same.
2) allows you a choice of what to delete.
4) is the option of deleting all the services in one go.
Did you delete everything or you kept some of the services?
Regards
I Removed ALL the services in Part 2. In short, i did not omit anything from that Part. This was done after reverting back to steppe 5. I wanted to listen for a few days before i move on to Part 2. As it is, it sounds very promising....
Junaid
Since I am in the midst of trying to find the best foundation for my system I did an install yesterday with a mildly de-contented nLite install disk. It makes for a quick install and has WFP disabled.
I first installed chipset drivers, cMP, cPLAY, JULI@ and a few of the important cMP optimizations.
I went through and did all of the BF deletions. I did pick and choose serivces (and would appreciate knowing which ones you deleted).
Well, I lost CONTROL PANEL and the START button. If I hit the WINDOWS button the menu would appear but with most of the (few) items I had left on the menu still available. Luckily I could access WINDOWS EXPLORER and was able to engage cMP from there. And it still worked and played all night. My computer is not connected to the system so I have no idea what it sounds like. Wanted to get something stable before doing that because access to the computer is cramped when it is installed.
I know this doesn't tell us much but it is interesting.
Also, I am wondering if getting the best out of Serge's regimen is to use the LYNX card? Wish it was not so expensive. The fact that JULI@ seems to need more .dll's than the LYNX is intriguing.
First of all I am unable to understand why do u need to install via nLite ?? Why can't u just do a clean install of xp & disable services according to the cMP website, take a back up & name it step 1. Thereafter u can proceed with whatever u want to. Now that batches 1 to 16 are ready, it's gonna take absolutely no time to work through till part 16 if u wish to.
As for my side, I'm doing a very very critical analysis of various combinations. And I'm very very happy with their outcome. One thing I understand now... There are no hard & fast rules !!! Something that may seem illogical ends up sounding way beyond u expect. Such is what's happening here :-)
Junaid
I am unable to understand why do u need to install via nLite ??
Me neither. As I see it, the problem with programs like nLite and XPLite (although the latter is excellent for disabling WFP) is that they delete we know not what using rules we don't know to configure a system (cMP2) which the programs' authors know nothing about and probably wouldn't understand if they did.
Add to that the fact that, while the process of "slimming" XP is proving sonically very fruitful, no-one taking it forward, not even steppe who started it all by disabling a few codecs back in July and has done very fine work since, is quite sure why it's effective.
On top of that, users have different motherboards and soundcards. I, e.g., have two cMP2 systems, one running on a single-core Atom board, the other on a dual-core Atom one (both controlled via LAN). The boards use exactly the same drivers but a set of batch files that worked fine on one system gave me a "no boot" scenario on the other. Not sure why yet. Mihaylov's list of "network-safe" deletions was not safe on either even if I needed to reinstate only a couple of files.
I don't see any time soon a "royal road" to making this work though Jolida's current and impending batch files seem to offer a robust way of approaching things. It looks like being a slow, systematic process for almost all of us. With respect, neither nLite nor XPLite offers that.
Whatever, it's been fun has it not?
I too had used xplite initially, just to disable wfp. This was when I had started to set up my cMp, following instructions on the cics website. That's it. It did the job very well. But this time, when the same was used in the slimming down process, it proved highly disappointing.
Swapping over to steppe 5, & adopting the Bold Fortune route, was an eye opener. I'm glad things dint work out even after steppe 16, else cMp
Would have carried on with it's journey with a scar on it's face (xplite). No matter what we try to balance it with, it can only be partially masked like a make-up, but could never be healed. With the procedure I have followed now (involving three major steps), I feel I am almost done. Just a bit of fine tuning left. I do not know how the BF route is technically different from xplite. And I don't care to know as well. I'm just glad I'm enlightened at the right time, before things would have gone haywire...
Junaid
Like I said I used the nLite disk just to bypass WFP and to see what would happen with doing all of BF's deletions.
And it loads so quickly.
SO tonight I put on my hair shirt and wait for WINDOWS to load and see what happens. (And grit my teeth while I slog through the WFP disabling)
By the way, I have not received the gift of your batch files containing 1 to 16. When you get the chance I would be most appreciative!
Thanks for your advice.
Oh is it?? I'm sorry. I dint know. Actually I get so many emails everyday for the batches that I don't remember to whom I sent or not. Check Ur email now...
Junaid
steppe 6 is very detail, but I found the sound thin, so I immediately went onto steppe 7 which restored some of the fullness but not much. And the bloom is gone. Reading your post I am now back to steppe 5 and relax.
I am very interested in your experience with BOLD-FORTUNE. Please keep posting.
I was intrigued by this after RYELANDS made mention of it.
I sent the fellow some money and got the batch files so my new install will revolve around Serge's recommendations and BOLD FORTUNE's. Certainly cheaper than XPLite and his "stuff" has explanations of what it does.
I find many of the same files are deleted in both regimens.
I did not have a chance to listen to the XPLit modified installation long before it died but I know I felt I was missing some of the atmosphere I had heard before.
Thanks again for bringing this BACK to my, and our, attention.
This is an unending journey and that keeps it interesting.
I don't understand why I get such a wonderful sound and you guys don't.
Theob,
Im sure u do..!! As i said earlier, MAY BE its subjective. When one attains a stage where he feels that everything sounds Just the way he wants, then he sure has achieved his goal. But even after the huge ordeal of modifications, if one feels that "Something Somewhere is Just NOT Right", then it defeats his purpose. I presume we are of the latter category. Im glad your set-up sounds Good.
You said "I don't understand why I get such a wonderful sound and you guys don't". Its a big statement. You are lucky. I wish i could say that as well. But unfortunately not..
As i once mentioned in one of my posts, NO two systems, even though identical, can sound identical in different surroundings. Much is Boosted or Bucked by our Cabling, Speaker placement, Room Acoustics, Isolation Platforms, Room size, Listening tastes, etc etc...
Likewise, implementing the same set of steppes, may or may not yield a similar benefit. Its a sure possibility that it can sound as a downside as well. Which in our case, it has !!!!
Junaid
I hope I have got the name right. I know it had advanced in it and you engage it by clicking on the leftmost button where the "X" is at the top left of the window.
Quite a few additional components appear and I wonder if Serge used these and for that matter has anyone deleted these?
I have managed to create ALL the Batch files that were remaining. So Now i have the entire set of Batch Files (steppe 1 to steppe 16) ready. Was working on them Non-Stop since midnight, & finally finished a while ago. Now the only files that need to be made are the Reg files for the registry entries. I hope to get them done as well, soon...
At this stage, i feel my back is in two pieces & am unable to move my neck as well. But i intend to take a break ONLY after i finish the job i have taken up....
Junaid
Edits: 10/10/11
When i had decided to go with the Batch file method, i had no wild idea what it actually is, or how its written, as i had no knowledge in writing any software or programs. This was my first time. Since i had made up my mind to follow this route, i struggled for nearly 3 full days, READING n reading web pages of tutorials as to how one could go about making them. With numerous errors in doing the same, i somehow managed to create a batch file for step 9. Upon its success, i thereby went ahead in writing the other parts. The batch files i had made did not include Registry modifications, as i had no knowledge in intergrating them. So i had to READ about it again & now i think i can finally go ahead with writing them up for all the steppes, as i managed to gather quite a bit of information regarding Batch Commands & their syntax.
I had to RE-DO the entire set of Batches i had made, to make them more accurate. I now have an UPDATED set of Batch files from steps 8 to 16. The previous Batches that i had made (and sent to inmates who had requested me through email), also work the same, but the Updated sets have more accuracy in their search paths, so now there is absolutely no room for errors. It takes a real long time in writing the data into the files, so i have managed to do Part 8 to Part 16 since this morning, without taking a break. I intend to make the rest of the Batches (1 to 7)as well in the coming week.
Those interested in using the Updated sets, may email me..
Regards,
Junaid...
Jolida,
I want to join the chorus of 'THANK YOU's for doing this.
I'd started down the road of making batch files for the Steppe Mods starting at Steppe 5, but have not had the time to make much progress on them yet. So I'm very happy to hear of yours (and sent a request for them via Asylum email).
Without seeing yours, here are some thoughts on what I was working towards with mine:
1. Knowing that a particular file might show up in multiple places inside of the Windows directory structure, I was planning to put in multiple 'removes' for each file, one for each of the directories where I'd found these DLL's & other files.
2. I had scoped out in my head a process that MIGHT work to make creating the batch files easier using MS Word and Excel, but hadn't worked out the details. If you are taking a break before finishing the rest, let me know. If so, I'll go play with that and see if I can make the process of creating them fairly easy.
3. I had thought about doing batch files for registry entries too, but decided that it might not be as good as going in and searching/exporting/deleting. This is because I suspect that the registry entries will vary a bit computer-to-computer depending on the initial Windows install disk (SP1 vs SP2 vs SP3 at least), how the person installed the op system & did the standard optimizations, what other components they use (Juli@ vs other sound cards, PCI sound card vs USB vs firewire, etc.).
I SUSPECT that a set of registry deletes for one person's computer at best might not get everything on the next person's and at worse might brick it. Does anyone have experience with this?
I've got a bunch more thoughts on the Steppe Mods and hopefully I can get to a post that's been rolling around in my head for a couple of weeks. But in short, these mods have been a revelation in my system's performance (and I'm only up through #4 so far!) AND I think there are some implications for those who want the ultimate best sound out of their systems that were not clear before or even were not possible before.
Later!
Greg in Mississippi
Everything matters!
The toughest part was getting to add the various paths to the filename to be deleted. DOS is pretty cranky. It would not accept an extra Space as a character as well. Thats why my efforts of modifying the batches via Excel were wasted. Most of the commands Like 'PATH' & 'IF NOT ELSE', are very critical with their Syntax. I wasted a real lot of time trying to use Excel for this... Try using TextPad. Its way better. U need to download it..
SO i went ahead & manually added an ENTIRE set of files to EVERY single path which may have housed those files.. A pain really. But it worked well...
I intend to finish Steps 1 to 7 in the next 3 to 4 days max, God willing..
I take this opportunity to congratulate & thank u for all the work u have done in improvizing the cMP project. I have copied all my Linear PSU designs from ur posts. Worked very well...
Regards,
Junaid..
Great idea about the steps 1- 9! Also thought about this but as always the time are missing out on all. Propose those files that you and I have proposed to remove as separate steps too, for example, in the form of steps 3.5 and 4.5. It would be cool to take account of all the files to delete proposed by the other users.
Thanks again for your work.
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Live, iPad
Edits: 10/08/11
Yes. That's exactly how I was able to maintain USB thumb drive functionality even after part 16... Also, xplite changes things quite a bit. I felt that the performance suffered a downside after the xplite stage, then re-gained a better balance at step 9 ( just a bit better, but still step 5 was the winner), & again kept on varying erratically all the way till part 16. Now I'm back to step 5 again & carefully observing where we missed out on what. Batch files makes this process a piece of cake. But I want to spend some time auditioning each step & pinning down ONLY those files that are eligible to be removed, I mean only those which are CPU INTENSIVE..
This way, one can maintain a perfect blend of Sonics along with ease of use. I haven't started on the remaining batches yet, as I'm exhausted with writing batch processes yesterday. I intend to start over tomorrow. As I'm writing this, I'm listening to step 5 again. I am quite certain that I will be skipping the Xplite step altogether. My initial impressions were right. Let me know what Ur findings are....Junaid...
Edits: 10/09/11 10/09/11
I went up to #7 and didn't quite like it so I am back at #5. I am now at #5 post XPLite.
I thought I had lost the natural attack, decay, sustain, and release of each notes in the music.
When time permits I would like to try from #8 to #16 and see how that fare.
Thanks for all works you had done and let's know your findings.
Best Regards
U wrote "I went up to #7 and didn't quite like it so I am back at #5. I am now at #5 post XPLite."
Xplite comes after steppe 5. U mean u have gone one step before xplite right???
Im glad I have taken a backup of every single step before proceeding with another. So now I'm able to comfortably run steppe 5 which I liked. Post steppe 5 (after xplite ), I found the spectrum tilting one side or the other with every new set of deletions ( my assumption ). It never sounded good in one band without a downside at the other. The highs opened up at the cost of midbass dropping a notch. The bass became full at the cost of the upper mids getting laid-back more than they should. So I intend to start over from steppe 5 & proceed carefully. May be our findings are subjective as well. But this reminds me of a statement I heard once..."We build our systems to suit our tastes, not to satisfy the critics"..
Junaid
Yes, I just completed XPLite removal of components.
You could be right in your assumptions. I thought some steps favoured a certain frequencies over others too.
More listening required to ascertain these. Hope to find the time to do so.
Best Regards
I finished my new installation (all but the keyboard and cMP affected files deleted) last night.
I had started with a minimalist nLite install disk and then used XPLite, also. As an aside I found XPLite had to run numerous times to get the remaining stuff out of the way. I guess it gets confused bu stuff that is already missing.
When I finished I checked the size of my WINDOWS folder and in PROPERTIES it says 187 MB. I am curious what others who have got "here" get to see if there is any difference between using nLite or not.
One thing I can say is that many files are not there to delete with nLite. All the way up to #16.
This time I can get AWE to work!
I highly recommend using nLite just for the bloat it gets rid of from the beginning and the fact it disables WFP without having to do the hex editor
exercise. I prefer the WINDOWS CLASSIC simplicity instead of the XP look for this application. Another choice you get with nLite.
nLite is free. Give it a try.
Now to get to listen to it tonight.
After a full monty 16 except for any file with usb, fat in it or files needed for cplay I have Windows size of 103mb.
As a testimony to Steppe the temp file into which I placed many of my deleted files (not all but most) is 227mb in size.
Edits: 10/06/11
Makes me wonder if XPLite deleted anything from my nLite install?
Thanks for looking.
Can't imagine getting it much smaller than THAT!!!
I have implemented till steppe 8 & began to realise what pain it is to Do or Re-Do an entire step of deletions. Sometimes more than a hundred in each step. So it came to my mind, 'Why not make a Batch file of a particular step'. This would make things a real lot easier if I had to re-do that step , & most of all, to help others in doing the same in a fraction of a second.. YES ! A fraction of a second to delete an entire set.
For those who do not know what a Batch file is, or what it does, here is a simple explanation. A batch file is a single file that is made with commands to direct Windows to implement a particular thing, either by user-intervention or on it's own. In our case, it's for making the entire deletion process turn into a "one-click" automated process, which will delete ALL the specified files in that particular step, in a fraction of a second..
Since I have followed the primitive method of searching & deleting every single file (as we all have done so far) till part 8, I have started creating Batch files from steppe 9 onwards.. Anyone who has not implemented part 9 or later, just need to click on that batch file of that particular step & all the files specified in that step will be deleted with the blink of an eye. Couldn't get easier right...????
I have made the processes in such a way, that it does not permanently delete the files from the computer, just in case we want to restore them or some of them for other reasons. Instead it creates a folder on the 'D' drive for that particular step & dumps all the files into the respective folder, so that we can figure out which file came from which step..
Inmates who are interested in doing steps 9 to 16 using this method can email me, & I will send you the batch file for the step you have requested....
Edits: 10/06/11
Hi Jolida,
Any chance to get these fantastic batch files ?
Alexandre
After operating the steps through the Batch files, I'm beginning to wonder how it would be to make the Batch files for the previous steps as well. It will give us all a very fair idea of evaluating the changes from step 1 to step 16 in a disorderly fashion. I mean for us who have a back up of the original cMp installation, it a very good chance to do a re-check of what actually changed & to what degree all the way till part 16. And if all the batches are ready, one could do from step 1 to step 16 in less than a minute if needed. Also, it would give us a chance to skip one step over the other or eliminate it as a whole to ascertain it's impact on the sound. Hoping to get the other Batches ready soon..Junaid..
Edits: 10/07/11
Great idea and very generous ...only problem I can see is if certain installations require some files not be deleted. For example I needed to maintain some files for asio to initialize, some which my installation said were in use so I was not permitted to delete, some files required for usb thumb drive functionality. Others who use networks also have a whole slew of files to maintain. It might therefore vary what you can delete vs others. Anyway its a risk to be aware of.
I did that for the batch files already available. Just eliminate any line that removes something that should stay in your case.
Theob,
It's not just that. There is another thing that made me want to re-try all the previous versions of cMP. When I did step 5, I was very convinced with the outcome of it. The moment I started doing from step 5.5 onwards, something ir the other does not seem right. As I'm writing this, I'm at a stage where I have already finished doing all the parts till 16. And even at this stage, things aren't too good either. After step 5, the only stage I found impressive was when I did part 9, followed directly with part 16, eliminating all other intermediate parts. This sounded very promising. I must say it sounded way better than all parts accomplished. It's a good thing to go back ten steps & review it again...
I guess what you may find unacceptable vs anybody else is for sure highly subjective. But what about a complete 16 (all steps) did you find unacceptable? What did you like about 1-9 plus 16?
Yes. Absolutely !! To me, 1 to 9 plus 16 was way superior sonically then all the way from 1 to 16. That's exactly why I started to make batch files so that I & others may benefit from that experiment of re-evaluating what actually contributed to what...
What specifically is superior...highs,mids, lows, dynamics? want to know whether I should restore 10-15. If its just tonal balance I find that difficult to chase.
Sonically superior in terms of dynamics, involvement & presentation..
;)
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Live, iPad
Using the batch files was a breeze and a big time saver.
Inmates who requested for the batch files, please be advised that I will be sending the files in a single folder in .Rar format, as yahoo or gmail do not allow sending attachments like batch files without ripping them into one folder. The files will be in the attached folder called batch files.rar. U need to extract the files from that folder using a freeware utility called WinRar, downloadable in less than a minute from this link
http://www.rarlab.com/download.htm
most of us have it from the early days of cMP.
Thanks for the tedium this required.
Good idea.
Hello, I wonder if anyone can help and offer feedback about the poor latency performance I'm seeing with my recent build?
This machine uses the 35W i3 2100t CPU. The motherboard is the GA-H61M-S2H (H61 R). This was the only board I could find with a PCI slot, 1155 socket, Ultra Durable 2 and it's one of the least populated micro-ATX boards I've seen. Also, the machine uses Win 7 32-bit, SSD for the OS, external powered SATA drive for the music library and the SOTA PCI USB card.
After optimizing all of the voltages and other settings in the BIOS, I'm impressed by the voltage readings (DRAM running at 1.18 and CPU measuring 6.2 watts). However, the DPC Latency test readings suck with everything at idle it hovers between 5-7 and with music files playing in Foobar or cPlay it goes down as low as 5, but often spikes up to 40. My other machine, which used a full ATX board from Supermicro and a dual-core CPU was pretty consistent, reading about 7 with music playing.
Can anyone offer a suggestion about what might be the problem? Maybe I'll just have to follow Greg's build step by step, but I was hoping to at least get the same performance from the 35 watt i3, for a scaled down linear PS.
Here's a list of ideas I've had about what could be standing in the way of the latency performance:
* The MB uses a PCI to PCIe bridge.
* The i3 2100T has some quirks that inhibit low-latency performance. The clock ratio can only be reduced to 16, which means the CPU freq can't go below 1.6ghz.
* The S2H MB doesn't have good latency performance. This board does not have a graphics chip, it uses the CPU graphics engine.
* I could not disable HW Acceleration in the OS. I received an error about the registry, but I'm not sure HW Acceleration is even running.
Your help is appreciated.
Thanks for the feedback Rick. I looked everywhere for feedback on the i3 2100t and remember finding almost nothing, but a post that you were using it in a machine. I should have emailed you about it first, but as is too often the case, I don't ask around enough before leaping.
I ordered the UD2H board and a new chip. What kills me is I had that board a few weeks ago and RMA'd it, because I didn't realize it didn't have the 1155 socket. I'll probably sell the chip on ebay. With this post, the next person has no excuse to make the same mistake.
Thanks for the feedback Rick. I looked everywhere for feedback on the i3 2100t and remember finding almost nothing, but a post that you were using it in a machine. I should have emailed you about it first, but as is too often the case, I don't ask around enough before leaping.
I ordered the UD2H board and a new chip. What kills me is I had that board a few weeks ago and RMA'd it, because I didn't realize it didn't have the 1155 socket. I'll probably sell the chip on ebay. With this post, the next person has no excuse to make the same mistake.
Did not realize that the CPU would be a problem, also, though.
There is not an appropriate MB out there for this socket. I should have asked Mihaylov before I wasted my money on re-stocking charges.
When you think about it the only thing the low wattage CPU does is make life easier on the regulator for the P4 connector. Not really a big deal, anyway.
The big trick with the linear supply for the MB is the 5 volts rail and that is not insurmountable.
One wonders if the Gigabyte GA-H55M-UD2H is the best we will ever have for use with PCI soundcards.
What is "Greg's" build? Do you mean power supply stuff?
Thanks for re-assuring me that my experience with the "new" processor was not a fluke.
Where does one sell a unneeded CPU? I need to get rid of mine, also.
After some troubles I've finished step 12.
Now I can listen to a great sound, it seems me a little more transparent than the step 7 but I can't decide if just a bit leaner too.
The more astonishing feature is the silence and the born of the sounds from a black background, and this one seems to me even more than the reality. Good soundstage too,very precise, large and deep.
It's a very very good result IMHO, tonal balance is near perfect and now I can enjoy myself a lot. Is it better than step 7? Maybe yes, but I need a lot of music now to decide.
Now the problems in my system:
- XP is blow out, you can't manage nothing nor setting anything else.
- No pictures or other images can be showed on the screen
- USB shows a strange beahviour. Devices that are connected from the older steps ( i.e. microsoft keyboard and mouse cableless) are recognized and are usable but no one of new devices USB connected can works. So no external HDD, thumbs, others keyboard or mouse.
- no more possible save an image of OS by Acronis: it doesn't works
These two last points could be troublesome or dangerous if you get a crash of system. It's too hard restart from step 1 (obviously I've lost all backups of my various OS I've done with Acronis. Last was step 9, after it backup become impossible)!
I'm searching for a solution for USB function but I think it's too difficult for me.
I don't know if I've desire of carry on in this direction, PC become too difficult to manage and every problem could be fatal for the listening.
Step 16 is so superior to get the risk?
Any help is wellcome
Daniele
Acronis has Nothing to do with what services work in Ur windows. I guess u are using the software version of acronis which u have installed as a program into windows. That's not the correct procedure though. The actual & most optimal way of using Acronis is by creating a bootable flash drive which can also store Ur back ups on the same. Whether windows recognises Ur flash drive or not has nothing to do with it, because u will tentatively boot via the flash drive when u switch on Ur computer ( before windows starts to load), by either pressing F12 during the initial screen & selecting "USB HDD" as the option. It then boots via the flash drive which u had made bootable, & thus the Acronis menu loads with options to Back up or Restore. This is how it's actually supposed to be done to make a true image of a partition. And as to Ur query regarding USB flash drive functionality in windows, I guess some of us will sort that out soon, once we are done with optimising cMP with all the deletions.
I had problems with usb drives after file deletions as well. For me, I got it back be enabling some of the usb devices I had disabled as part of the cmp optimization. It is a little hard to explain, but usb had been working on the 3 ports I left enabled, and then usb stopped recognizing thumb drives or usb HDD. Enable some of them and back came support for usb drives.
Any ideas?
Can't re-install from RECYCLE BIN because I can't see it.
SAFE MODE has the same problem.
Any idea of what file I deleted by mistake?
With 1100 deletions to get to the end I hope I do not have to start over!
I can use another computer to re-install.
Thanks,
Probably you have removed "explorer.exe"?
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Live, iPad
Edits: 10/04/11
my SSD for OS and software is formatted in FAT32.
Could this have anything to do with this?
Never a problem before.
I did this after reading that SSD's liked FAT better than NTFS.
I noticed a few files with FAT in the name.
I copied the deleted files back into the music computer's WINDOWS from my regular machine and this did not make a difference which makes me think I deleted something else other than what was specified.
So returning the files with FAT in the name did not bring it back.
I know, I know, I should have made an image ...
Thanks for your help,
I started doing from part 9 to the end of part 12 in one go, but without rebooting in between. After such a long set of deletions, wanted to audition the changes & got a Blue Screen. The windows starts & remains with a blank screen with just the mouse cursor for ten seconds & shuts down with a blue screen error. Please help...
Edits: 10/04/11
I've got same experience after part 10. I've made some mistakes because restarting from part seven with calm and attention all now is working but.....
USB!!!!!
I cannot use USB devices: no mouse, no keyboard,no thumbs, no external HDD!!
It's a big problem for me and I can't find any tutorial or news about deleted files I must restore to repair this function. I'm not an expert!! System now recognize only devices attached from step 7, but thumbs go away the same. It's impossible to use a new mouse Keyboard wifi combo.
I can't go on toward next steps if I can't use mouse and keyboard and I can't listening music too!
Can someone help me? thanks a lot in advance
Daniele
and did not have a problem.
I do get the "autochk missing" screen for a few seconds and then it would nexy tell me I am missing some other file which after I click OK a couple of times (for cMP) or four times for XP all seems fine. It complains about a file removed back in 7 or 8 - sorry I cannot remember which one.
When I tried to do 9 through 16 I did get the blue screen. I had made the deletions through the GRANITE USB adaptor with my regular computer. I then reinstalled all of the files and went back to 8, all was fine.
Then I did the 8 through 12 the same way with no problems.
Have you tried SAFE MODE yet?
Last night I did 13 at the music computer, all was fine. Then I did 14 and the problem I am complaining about happened. I tried connecting the SDD to the regular computer. I can see the RECYCLE BIN folder but when I try to open it it momentarily shows the files then it goes blank and worst of all does not give me the restore option.
I know I am not giving you any help whatsoever but my thinking is that something I have done might shed some light for your problem.
Best of luck (to us both).
The masochistic part of me wonders if it might be for the best to start over again, again!
I did step 9 to 12 by hooking up my ssd to a laptop via esata cable & powering the ssd via granite digital power supply. Since it was all done on another computer, I dint find it necessary to restart. After all steps were completed, I hooked up the ssd back to the music computer. Windows boots. All that shows up on the screen is a mouse cursor & nothing else, not even taskbar. Then it shuts does after ten seconds getting a Blue screen "fatal error 000021". This is so frustrating...
Edits: 10/04/11
AND the cursor - but it would sit there like that for the rest of my life if I let it!
I amp hoping my case is a conflict with my FAT formatted drive. I put back the files that (I presume) are associated with that (As if I would KNOW). One never learns anything without making mistakes. I am hoping for a few brilliant mistakes!
Tonight I will see ...
I would replace the files and start over. Worth a try.
The Steppe's Tweaks parts 15, 16 are possible to apply completely in a network configuration of cMP2 BUT:
do not delete following files:
reg.exe (otherwise the mode "Start cMP" will be inaccessible!)
cmd.exe
netui0.dll
winspool.drv, olepro32.dll (otherwise cMP it will not be launched)
The network configuration of my cMP2 is based on cic's "Headless cMP ² using a Fit-PC2" namely:
- Static IP addresses avoid DHCP's overhead and work well with UltraVNC. Recommended.
- The remote control utility is... the best of those tested ... UltraVNC.
- cPlay can be launched using cMP's library facility as normal except that it is necessary first to log on to the network drive (NAS) by means of a command "Net Use M: \\Server\Folder/User:Server\UserName password:" started by click the ‘ RIP ' button in the cMP window.
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Live, iPad
The Steppe's Tweaks parts 13, 14 are possible to apply completely in a network configuration of cMP2. (but I did not find files tapi32.dll, ver.dll , wzcsvc.dll in step 12 and asr_ldm.exe, netapi.dll, xmlprovi.dll in step 14. The files prodspec.ini and pschdprf.ini in step 14 can be removed - cPlay works fine).The network configuration of my cMP2 is based on cic's "Headless cMP ² using a Fit-PC2" namely:
- Static IP addresses avoid DHCP's overhead and work well with UltraVNC. Recommended.
- The remote control utility is... the best of those tested ... UltraVNC.
- cPlay can be launched using cMP's library facility as normal except that it is necessary first to log on to the network drive (NAS) by means of a command "Net Use M: \\Server\Folder/User:Server\UserName password:" started by click the ‘ RIP ' button in the cMP window.
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Live, iPad
Edits: 10/03/11
The files prodspec.ini and pschdprf.ini in step 14 can be removed - cPlay works fine).
Weird maybe I'll try again.
I too was surprised after your warning. But it so at me.
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Live, iPad
The Steppe's Tweaks part 12 is possible to apply completely in a network configuration of cMP2 BUT:
do not delete following files:
imm32.dll
setupapi.dll (used by cPlay - Thanks to theob!)
The network configuration of my cMP2 is based on cic's "Headless cMP ² using a Fit-PC2" namely:
- Static IP addresses avoid DHCP's overhead and work well with UltraVNC. Recommended.
- The remote control utility is... the best of those tested ... UltraVNC.
- cPlay can be launched using cMP's library facility as normal except that it is necessary first to log on to the network drive (NAS) by means of a command "Net Use M: \\Server\Folder/User:Server\UserName password:" started by click the ‘ RIP ' button in the cMP window.
Serge.
http://cmp2-mihaylov.narod.ru/
If cPLAY needed it Serge would not have been able to make it work.
He is using the LYNX card and undoubtedly if does not need the file.
:)
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Live, iPad
I eventually got Steps 10 and 11 to work without losing networking functionality. Below is a list of the files I left in place (Intel Atom mobo, AQVOX USB driver).
It might well be possible to remove some in due course (I left some in as a precaution) but deleting others definitely caused networking on my cMP2 boxes to fail. Whatever, I'd start by leaving them in place.
Step 10 pt 1: dciman32.dll, deskadp.dll, deskmon.dll, deskperf.dll (repeatable failure if these are deleted).
Left in for now: ippromon.dll, iprop.dll, ipsecsnp.dll, ipsecsvc.dll, ipsmsnap.dll, kbdusx.dll, netid.dll, netui1.dll, netui2.dll, ntlanman.dll, ntlanui2.dll, ntmarta.dll, ntmsevt.dll, ntmsmgr.dll, ntmssvc.dll, ntsdexts.dll, ntshrui.dll, nwcfg.dll, rasmontr.dll, usbmon.dll.
Step 10 pt 2, ndis.sys, ndisuio.sys, ndproxy.sys, netbt.sys, oprghdlr.sys (ACPI Operation Registration Driver, repeatable failure if deleted), processr.sys, rasl2tp.sys, raspptp.sys, raspti.sys, rawwan.sys, rndismp.sys, sdbus.sys (repeatable failure if deleted), tdi.sys (network driver, repeatable failure if deleted).
Step 11, mf.sys, msgpc.sys, mup.sys, ndiswan.sys (repeatable failure if these four are deleted).
How does this compare with anyone else's experience? The resulting step up in SQ is apparent - well worth doing. Thanks, Serge.
Dave
The Steppe's Tweaks part 11 is possible to apply completely in a network configuration of cMP2 BUT:
do not delete following files:x86_Microsoft.Windows.Networking.Dxmrtp_6595b64144ccf1df_5.2.2.3_x-ww_468466a7.cat
x86_Microsoft.Windows.Networking.Dxmrtp_6595b64144ccf1df_5.2.2.3_x-ww_468466a7.Manifestcomres.dll (used by cPlay)
mswsock.dllmsgpc.sys
mup.sys
psched.sysThe network configuration of my cMP2 is based on cic's "Headless cMP ² using a Fit-PC2" namely:
- Static IP addresses avoid DHCP's overhead and work well with UltraVNC. Recommended.
- The remote control utility is... the best of those tested ... UltraVNC.
- cPlay can be launched using cMP's library facility as normal except that it is necessary first to log on to the network drive (NAS) by means of a command "Net Use M: \\Server\Folder/User:Server\UserName password:" started by click the ‘ RIP ' button in the cMP window.
Serge.
http://cmp2-mihaylov.narod.ru/
Edits: 10/02/11
The Steppe's Tweaks part 10 is possible to apply completely in a network configuration of cMP2 BUT:
do not delete following folders and files:x86_Microsoft.Windows.GdiPlus_6595b64144ccf1df_1.0.2600.2180_x-ww_522f9f82
x86_Microsoft.Windows.Networking.Dxmrtp_6595b64144ccf1df_5.2.2.3_x-ww_468466a7credui.dll (used for PCI Latency Tool)
netui1.dll
newdev.dll (used for PCI Latency Tool)
ntlanman.dll
riched20.dll
riched32.dll
wshtcpip.dllndis.sys
ndisuio.sys
netbt.sys
tdi.sysThe network configuration of my cMP2 is based on cic's "Headless cMP ² using a Fit-PC2" namely:
- Static IP addresses avoid DHCP's overhead and work well with UltraVNC. Recommended.
- The remote control utility is... the best of those tested ... UltraVNC.
- cPlay can be launched using cMP's library facility as normal except that it is necessary first to log on to the network drive (NAS) by means of a command "Net Use M: \\Server\Folder/User:Server\UserName password:" started by click the ‘ RIP ' button in the cMP window.
Serge.
http://cmp2-mihaylov.narod.ru/
Edits: 10/02/11
The Steppe's Tweaks parts 8, 9 are possible to apply completely in a network configuration of cMP2.
The network configuration of my cMP2 is based on cic's "Headless cMP ² using a Fit-PC2" namely:
- Static IP addresses avoid DHCP's overhead and work well with UltraVNC. Recommended.
- The remote control utility is... the best of those tested ... UltraVNC.
- cPlay can be launched using cMP's library facility as normal except that it is necessary first to log on to the network drive (NAS) by means of a command "Net Use M: \\Server\Folder/User:Server\UserName password:" started by click the ‘ RIP ' button in the cMP window.
Serge.
http://cmp2-mihaylov.narod.ru/
Please, both you and Ryelands, keep this "network version" going as far as possible in relation to Steppe's tweaks. It's very useful.
Thanks.
Steppe's Tweaks parts 8, 9 are possible to apply completely in a network configuration of cMP2.
Agreed - and very good they sound too. Currently working on steppe 10 in which some files definitely have to be retained.
Dave
Hello, everybody.
I didn't realize that you would be so quick at implementing tweaks part 7, so this time I decided to post several parts at once.
I'd like to point out, that already part 8 is setting a cardinal question about network capability of your cmp machine. Inmates, you really have to decide if network is an absolute must for you, because from part 8-9 there will be no network. Also, many other functions will be deleted.
The subdivision is purely out of convenience. I'd like to point out, that I have an external hdd in a esata-usb combo box with external power supply (thermaltake silver river 2) with tweaked Windows on it as well as all my music files. I rip on a different computer and transfer now files plugging the hdd into ripping computer's usb. I don't use flac, by the way.
As a good measure of caution I'd recommend to have an ability to install (in case of a blue-screen crash) your main hdd as a slave or secondary master in another computer to return a file you have removed.
This is what I propose to do in part 9:
I'd rather perform everything at once and then reboot, because I had been removing files in a disorderly fashion and, as experience shows, some files bring blue screen crash at one time but go off easily after some other file was removed.
Remove folders: GroupPolicy from windows-system32 and security from Windows root directory
remove file ver.dll from windows-system
other files are mostly from windows-system 32 or windows-system32-drivers.
List of files to remove:
asr_fmt.exe asr_pfu.exe atmadm.exe attrib.exe auditusr.exe autochk.exe (after you remove this file windows will reboot with a funny warning that it cannot find it) autoconv.exe autofmt.exe autolfn.exe cacls.exe cdosys.dll chkdsk.exe cipher.exe comp.exe compact.exe conime.exe convert.exe ctfmon.exe dbghelp.dll ddeshare.exe dfrgfat.exe diantz.exe
dmremote.exe driverquery.exe dumprep.exe dwwin.exe esentutl.exe eudcedit.exe eventcreate.exe eventvwr.exe expand.exe extrac32.exe fc.exe fltMc.exe fontview.exe fsquirt.exe fsutil.exe gpresult.exe gpupdate.exe help.exe hhctrl.ocx hostname.exe iexpress.exe
kbddv.dll kbdkaz.dll kbduk.dll kbdukx.dll kbdusl.dll kbdusr.dll lnkstub.exe lodctr.exe logman.exe LZEXPAND.DLL migpwd.exe netsetup.exe nslookup.exe ntsd.exe pifmgr.dll pjlmon.dll polstore.dll powrprof.dll prflbmsg.dll proquota.exe psapi.dll psbase.dll
pschdprf.dll pstorec.dll pstorsvc.dll query.dll recover.exe redir.exe regsvr32.exe relog.exe routemon.exe rsopprov.exe rsvp.exe savedump.exe sc.exe sdbinst.exe secedit.exe shrpubw.exe sigverif.exe smlogcfg.dll spiisupd.exe swprv.dll sysocmgr.exe systray.exe
unlodctr.exe unlodctr.exe verifier.exe vssadmin.exe vssvc.exe winscard.dll wowdeb.exe wowexec.exe wscntfy.exe
Serge.
P.S.I want to warn you that at the end of part 9 or 10 you may get a sound that you didn't expect. It may become very detailed but harsh and brittle and not very pleasant. This is not to discourage you. It will all settle to a "relative" bliss in parts 13-14.
In fact the subdivision is only for convenience. It is all one long tweak. I stop from part 12 only when I have achieved some significant (though intermediate it may be) result.
If cPlay plays for more than a few minutes I can no longer close it and revert to cMP. I have to restart the system. It never did this before this step. Anyone know what I might restore to cure this? (shutdown.exe has already been restored.)
New one on me. What I do is restore files deleted a few at a time until your function is restored. I recommend to do as ryelands suggests: delete, test, then back up your os disk.
Let me be the first to congratulate Serge on an awesome job, and to thank you for breathing incredible new life into this system.
I went on a marathon and went did 8 thru 12 without stopping to listen. It seemed from the writeup that 12 is a natural plateau. The steps were well laid out. With the exception that many files were marked in use and would not delete, there were no issues. My system is stable and fully functional.
Now for the sound. At 12 the sound is VERY good. In my case a startling improvement. PRAT, attack, balance top to bottom. Relaxed and fluid but very detailed. Very rich but REAL. I spnt about 90 minutes picking off 1 or 2 cuts out of 20 cds. In every case I wanted to stop and listen to the entire cd, but was curious to go on. Normally many of these I have not listened to for months as they were in some way disappointing. At step 12 all but 1 was a winner and I did not want to leave it. I am listening to the music, and not looking for faults in the system.
I am pausing at 12, perhaps forever. It is that good.
Serge you are the man!. Take care of your cat and your cold. Again thankyou.
Hi Serge,
It's quite impressive to see the amount of dedication & hard work u have put in creating all the above. Part 16!!! Phew! Looks like we have a Long way to go. Talking about the Long way, one thing comes to my mind. Why not make things easier & lead us to a short-cut....!!!!!!
By saying that, I mean instead of knowing what files to delete, I feel (since it's gone upto part 16 as of now) that u could guide us with WHAT FILES TO KEEP!! The situation has become such that the system32 & system folder is getting smaller & smaller with every step. So instead of walking all the way through to part 16 & taking the long route, it would be great if u could post a list of files that remain in your system32 & system folder post-part-16. I'm sure the files remaining will be much lesser than the ones to be deleted till we reach part 16. For eg. From steppe 8 to steppe 16, if we need to delete say 300 dll files, & post that only 50-60 remain, then we could drag the needed ones to one side in their respective folder & highlight & delete the remaining ones. Just a thought. U know better!!!
I originally posted a technical question but forget it I will figure it out. Take care of yourself and your kitty. Thanks for all the hard work.
Hello, everybody. This is part 8 of the tweaks, that I propose to do.
1. Find and open with notepad file admfiles.ini in system32-group policy-adm. Change inside this file values from 1 to 0 in inetres.adm, wmplayer.adm, conf.adm and wuau.adm. Save, reboot, remove to your temp folder files: inetres.adm and wmplayer.adm if the latter is still there.
2.Remove folders: windows-system32-NPP and from 1033- file dwintl.dll, remove file winhlp32.exe, folder windows-setup and clear windows-temp from everything. From windows-system32-config-system profile-templates remove everything, Windows-system32-drivers-remove folder ETC. Reboot.
3.Remove these folders (mostly from windows-system32 or windows): Repair, Inf, CSC, Web, debug (but leave file passwd.log), ehome, mui, peernet, pss, resources, provisioning, wbem. In folder pchealth leave only msconfig.exe (go deep inside). from windows-system32-config-systemprofile remove everything IE and Media Player, also remove Remote assistance, narrator, magnify and 0userkeybrd. Reboot.
3.Remove these files: mciole16.dll mmdriver.inf msjetoledb40.dll oleprn.dll setup.exe (windowsNT setup executable), hscsp_w3.cab and pchshell.dll. Don't reboot, but go on and remove these files: SPGRMR.dll all 22 files starting with wdma_ *.inf (that is if you haven't removed the inf folder yet) + wdmaudio.inf and wdmjoy.inf (check your soundcard, but it should work anyway). next remove files: amdk6.sys amdk7.sys arp1394.sys asyncmac.sys atmarpc.sys
atmepvc.sys atmlane.sys atmuni.sys cdfs.sys cdrom.sys fdc.sys flpydisk.sys http.sys irenum.sys modem.sys mqac.sys msfs.sys netbios.sys nic1394.sys nikedrv.sys nmnt.sys p3.sys parport.sys pcmcia.sys ptilink.sys rasacd.sys rio8drv.sys riodrv.sys rootmdm.sys scsiport.sys serenum.sys serial.sys sfloppy.sys smclib.sys tape.sys usbcamd2.sys usbcamd.sys usbintel.sys wanarp.sys.
Reboot. Listen to some music. I hope you will see the difference was worth the trouble.
Now some words about the ongoing tweaks, and how long will it take, and about my humble system.
The tweaks go as long as part 16 with part 10 and 15 being the longest and other parts very short. Part 16 is only 15 files long. I am positively running out of files to remove, and every new find is a treasure, indeed. It should be said, I think, that an opinion that it is enough to remove only dlls and/or sys and exe files that "directly" concern the cmp-cplay process, is slightly incorrect.
Going deeper and deeper into the maze of Windows one starts to see, that though basically very simple, even primitive, it is bound hand and foot with dlls that are patches on patches on patches of some ill-designed services that are unnecessary for OUR purpose. Most are triggered by logon process only to grab the resources and spread triggering threads to other services and so on. Worst of these
files are APIs, and most won't get off easily or at all. Most resources are spent to service security, cryptography and networking. Now I have to warn you that I would retain network ONLY if it is an absolute MUST, otherwise you cannot get too far from here. Also I'd like to attract your attention to two interesting files that are still in your system" winspool.drv and xmlprovi.dll. The can be
removed even now, but wait till you get to part 14-15. Winspool won't allow you to start cmp, but it is SO DIRTY that without it cplay sounds better alone. And xmlprovi.dll will allow for interesting tweaking of sounstage, expanding or flattening it.
Now about my setup. it is the simplest possible: Gigabyte ga g55m-ud 2h mobo 540 processor in zalman hd 160xp+ case with passive cooling from scythe (a flat one to fit into the case), 1 gb of kingston KVR ddr3 memory, zalman heatpipe psu, lynx l22 soundcard. The only hdd is seagate constellation Es with capacity of 2 tb. It was chosen for its durability. It resides in an external esata-usb combo from thermaltake.
Name is silver river 2 smth. Has external psu, Very dirty, so I put it on a different powerline. Everything is fanless, including psu. All tweaks by Cics done. voltage is 0,7.
The soundcard is feeding analog out to $50> Microlab b73 active multimedia speakers, set for near-field. As they don't have xlr inputs, I made xlr-rca adaptors with Oyaide Tunami cable and good connectors. Adaptors cost 3 times more than the speakers.
Speakers are 5,5 feet apart, have 6 feet spare space behind them, early reflections are coped with. I sit 6 feet from the speakers that have a 30 degree studio toe-in. That is all. No network, no ripping, no cdrom, no nothing else.
Somebody said there is a "typical" clutter inside inmates' cmps. No clutter wharsoever. The innards are well placed, only 2 feeding cables from psu. The case is empty. I transfer ripped music by attaching my hdd as a Usb device to the ripping machine and just copy files.
Serge.
Hello, everybody. This is the longest and dullest part. It is as long as all the remaining parts taken together.
1.Remove these folders:
Dr Watson
InstallTemp
and from Windows-WinSxS and deeper inside it (don't touch there anything else!) remove:
x86_Microsoft.Tools.VisualCPlusPlus.Runtime-Libraries_6595b64144ccf1df_6.0.0.0_x-ww_ff9986d7
x86_Microsoft.Windows.CPlusPlusRuntime_6595b64144ccf1df_7.0.0.0_x-ww_2726e76a
x86_Microsoft.Windows.CPlusPlusRuntime_6595b64144ccf1df_7.0.2600.2180_x-ww_b2505ed9
x86_Microsoft.Windows.GdiPlus_6595b64144ccf1df_1.0.0.0_x-ww_8d353f13
x86_Microsoft.Windows.GdiPlus_6595b64144ccf1df_1.0.2600.2180_x-ww_522f9f82
x86_Microsoft.Windows.Networking.Dxmrtp_6595b64144ccf1df_5.2.2.3_x-ww_468466a7
x86_Microsoft.Windows.Networking.RtcRes_6595b64144ccf1df_5.2.2.3_en_16a24bc0
2. Remove these files in no particular order, but I started with dlls:
advpack.dll bidispl.dll capesnpn.dll certcli.dll certmgr.dll comcat.dll corpol.dll credui.dll crtdll.dll cryptsvc.dll dciman32.dll ddeml.dll deskadp.dll deskmon.dll deskperf.dll (this one will start to yell)
dfrgres.dll dfrgsnap.dll dfrgui.dll dgnet.dll digest.dll diskcopy.dll dnsrslvr.dll docprop.dll dpcdll.dll dssec.dll dssenh.dll duser.dll efsadu.dll esent97.dll esent.dll esentprf.dll eventcls.dll expsrv.dll
extmgr.dll exts.dll feclient.dll filemgmt.dll fldrclnr.dll fltlib.dll fmifs.dll fontsub.dll fsusd.dll ftsrch.dll gpedit.dll gpkcsp.dll gpkrsrc.dll gptext.dll hlink.dll icmp.dll ieencode.dll iernonce.dll ifmon.dll ifsutil.dll
igmpagnt.dll iissuba.dll ils.dll inetmib1.dll inseng.dll iologmsg.dll ippromon.dll iprop.dll ipsecsnp.dll ipsecsvc.dll ipsmsnap.dll itircl.dll itss.dll jet500.dll kbdusx.dll kd1394.dll kerberos.dll keymgr.dll
licdll.dll linkinfo.dll loadperf.dll localspl.dll lz32.dll lzexpand.dll mcastmib.dll mchgrcoi.dll mciole32.dll mdhcp.dll mf3216.dll mfc40.dll mfc40u.dll mfc42u.dll mfcsubs.dll mmcbase.dll mmcndmgr.dll
mmcshext.dll moricons.dll mprapi.dll mqsec.dll mqutil.dll msaatext.dll msapsspc.dll msaudite.dll mscat32.dll MSCTF.dll MSCTFP.dll msdadiag.dll msdart.dll msdmo.dll msexch40.dll msident.dll msidntld.dll
msihnd.dll MSIMTF.dll msisip.dll msjet40.dll msjint40.dll msjter40.dll msjtes40.dll msltus40.dll mspbde40.dll msrd2x40.dll msrd3x40.dll msrepl40.dll mssip32.dll msswch.dll mstext40.dll msutb.dll msvcp50.dll
mswdat10.dll mswstr10.dll msxbde40.dll msxml2.dll msxml2r.dll msxml3.dll msxml3r.dll msxml.dll msxmlr.dll mydocs.dll nddenb32.dll netid.dll netui1.dll netui2.dll newdev.dll npptools.dll ntdsbcli.dll ntlanman.dll
ntlanui2.dll ntmarta.dll ntmsevt.dll ntmsmgr.dll ntmssvc.dll ntsdexts.dll ntshrui.dll nwcfg.dll nwevent.dll nwprovau.dll nwwks.dll oakley.dll paqsp.dll pautoenr.dll pdh.dll pid.dll pidgen.dll plustab.dll
pnrpnsp.dll profmap.dll qosname.dll rasmontr.dll rdpdd.dll rend.dll riched20.dll riched32.dll scecli.dll schannel.dll scriptpw.dll sdhcinst.dll sdpblb.dll sfc.dll sfc_os.dll sfcfiles.dll sigtab.dll sisbkup.dll
skdll.dll softpub.dll themeui.dll typelib.dll urlmon.dll usbmon.dll usp10.dll vbajet32.dll verifier.dll vga64k.dll vss_ps.dll wdigest.dll webvw.dll wifeman.dll win32spl.dll winbrand.dll winfax.dll winntbbu.dll
winshfhc.dll winsock.dll wsecedit.dll wshisn.dll wshnetbs.dll WshRm.dll wshtcpip.dll xpob2res.dll xpsp1res.dll xpsp2res.dll zipfldr.dll
3. Remove these files: clock.avi esentprf.ini rasctrs.ini rsvp.ini mlang.dat bridge.sys cbidf2k.sys crusoe.sys diskdump.sys fips.sys fsvga.sys mrxdav.sys ndis.sys ndisuio.sys ndproxy.sys netbt.sys oprghdlr.sys
processr.sys rasl2tp.sys raspptp.sys raspti.sys rawwan.sys rndismp.sys sdbus.sys sffdisk.sys sffp_sd.sys swenum.sys tdi.sys update.sys usb8023.sys
4. Now remove these files: blastcln.exe defrag.exe dfrgntfs.exe mrinfo.exe nw16.exe osuninst.exe rsm.exe rsmui.exe rundll32.exe skeys.exe sort.exe subst.exe syskey.exe systeminfo.exe vwipxspx.exe winmsd.exe
5. Last registry tweaks: go HKLM-Software-Microsoft and delete DRM, then go Hkey_users (all users) software-microsoft and delete Clock
Reboot. Listen to some music.
Serge.
P.S. I truly hope, you have NO adobe products installed on your cmp machine!!!! Otherwise, first uninstall them all and clean the registry from them completely.
Hello, everybody. This is part 11, and this is what I propose to do:
Find and remove folder DRM
Find in WinSxS folder and remove following folders:
x86_policy.5.2.Microsoft.Windows.Networking.Dxmrtp_6595b64144ccf1df_x-ww_362e60dd
x86_policy.5.2.Microsoft.Windows.Networking.Rtcdll_6595b64144ccf1df_x-ww_c7b7206f
and files:
x86_Microsoft.Windows.Networking.Dxmrtp_6595b64144ccf1df_5.2.2.3_x-ww_468466a7.cat
x86_Microsoft.Windows.Networking.Dxmrtp_6595b64144ccf1df_5.2.2.3_x-ww_468466a7.Manifest
x86_Microsoft.Windows.Networking.RtcDll_6595b64144ccf1df_5.2.2.3_x-ww_d6bd8b95.cat
x86_Microsoft.Windows.Networking.RtcDll_6595b64144ccf1df_5.2.2.3_x-ww_d6bd8b95.Manifest
x86_Microsoft.Windows.Networking.RtcRes_6595b64144ccf1df_5.2.2.3_en_16a24bc0.cat
x86_Microsoft.Windows.Networking.RtcRes_6595b64144ccf1df_5.2.2.3_en_16a24bc0.Manifest
Find and remove from system, system32, system32-drivers following files:
append.exe debug.exe doskey.exe edlin.exe exe2bin.exe fastopen.exe find.exe findstr.exe fixmapi.exe forcedos.exe label.exe mem.exe msswchx.exe nddeapir.exe nlsfunc.exe ntvdm.exe progman.exe replace.exe runonce.exe setver.exe sfc.exe share.exe shmgrate.exe smbinst.exe sysedit.exe winver.exe
Now remove these files: admparse.dll adptif.dll adsldpc.dll adsnt.dll asycfilt.dll atl.dll compstui.dll comres.dll console.dll cscui.dll dbgeng.dll devenum.dll devmgr.dll dfsshlex.dll dmdlgs.dll dmdskmgr.dll dmdskres.dll dmintf.dll dsprpres.dll encapi.dll eventlog.dll fontext.dll hccoin.dll
hotplug.dll isrdbg32.dll localsec.dll mcdsrv32.dll msafd.dll msls31.dll msobjs.dll mspatcha.dll mswsock.dll ntvdmd.dll odbc32.dll odbcbcp.dll odbcint.dll ole2.dll ole2disp.dll ole2nls.dll rnr20.dll routetab.dll rsvpmsg.dll rsvpsp.dll setupdll.dll svcpack.dll syssetup.dll toolhelp.dll
twext.dll usbui.dll win87em.dll winrnr.dll msdfmap.ini ODBCINST.INI vb.ini vbaddin.ini country.sys intelppm.sys mf.sys msgpc.sys mup.sys ndiswan.sys null.sys psched.sys RMCast.sys ws2ifsl.sys stdole.tlb
Reboot. Listen to some music.
serge.
Something is wrong with the forum interface or my connection
Hello, everybody. This is part 12, and this is what I propose to do:
Find and remove:
dmutil.dll dskquota.dll dskquoui.dll imm32.dll mapi32.dll mapistub.dll mfc42.dll mgmtapi.dll mslbui.dll nddeapi.dll ntlsapi.dll nwapi16.dll oleacc.dll oleaccrc.dll olecli.dll olecnv32.dll olesvr.dll rtipxmib.dll rtutils.dll sendcmsg.dll serialui.dll setupapi.dll shdoclc.dll
Reboot. Listen to some music. This is the turning point. If you have reached this far, you will probably hear the whisper of prudence to stop, for it is sooo goood now! But will you?
Warning! Going on will severely cut the windows' functionality. Also, the sound will become very harsh for some time.
Serge.
For juli@ users and possibly others my cmp/cplay would not work (could not find asio drivers)without setupapi.dll. I put it back in and all was ok.Btw: sonics are outstanding although I prefer 11 but it is indeed brighter.
Edits: 10/02/11
Hello, everybody. This is part 13, and this is what I propose to do:
Find and remove:
aaaamon.dll aclui.dll shgina.dll shimeng.dll shimgvw.dll shscrap.dll shsvcs.dll slayerxp.dll
snmpapi.dll snmpsnap.dll stobject.dll tapi32.dll ufat.dll ulib.dll umandlg.dll ver.dll winnls.dll
wlnotify.dll wmi.dll wmiscmgr.dll wow32.dll wsnmp32.dll wtsapi32.dll wzcsapi.dll wzcsvc.dll
xactsrv.dll xcopy.exe
Reboot and listen to some music.
Serge.
P.S. You might have noticed that after you have just removed several files, windows is rebooting slowly at first. This is very good. It means, that big changes in sq are going to happen. Though the resolution increases, it may sound sharp or brittle or sybilant for some time.
Check your mouse, keyboard, search function, notepad and explorer. Everything should work.
Hello, everybody. This is part 14, and this is what I propose to do:
Find and remove:
dmview.ocx msscript.ocx asr_ldm.exe bootcfg.exe control.exe dllhost.exe dllhst3g.exe dmadmin.exe mountvol.exe pentnt.exe rsmsink.exe sprestrt.exe actxprxy.dll autodisc.dll browselc.dll
clb.dll clusapi.dll cnvfat.dll commdlg.dll compobj.dll dmocx.dll inetres.dll miglibnt.dll msvcrt20.dll netapi.dll OLECLI.DLL (from windows system, I think) xmlprovi.dll (yes, the file I told you about),
mediactr.cat NTPRINT.CAT prodspec.ini pschdprf.ini comm.drv AppEvent.Evt SecEvent.Evt SysEvent.Evt MSCTFIME.IME ntdos404.sys ntdos411.sys ntdos412.sys ntdos804.sys ntdos.sys
Please note, that in this series of files some are from windows root, some from windows-system, system32 and system32-config folders.
Reboot and listen to some music.
Serge.
Like before I needed to restore prodspec.ini and pschdprf.ini to get my asio drivers to inialize for cmp/cplay to work.Sonics (ims) are very good ... not sure its better than 11 but for me they are close. If 14 is not first in sonics it's pretty close so far.
Edits: 10/02/11
AND it is such a big file!
But, that is the way ti goes -
I wonder if someone could come up with a way to do without it?
Interesting that LYNX does not need it. Maybe time to consider trying one of those?
You are no longer able to use CONTROL PANEL are you? Not even with COMMAND LINE on mine.
.
Hello, everybody. This is part 15, and this is what I propose to do:
Find and remove:
bootok.exe bootvrfy.exe dosx.exe grpconv.exe mmc.exe reg.exe regedt32.exe tasklist.exe taskman.exe taskmgr.exe (these files, or their absence won't affect the startup of cmp - read below *) dmconfig.dll dmserver.dll dsuiext.dll fde.dll inetcplc.dll input.dll
msgina.dll msvcrt40.dll mycomput.dll ntmsapi.dll ntmsdba.dll osuninst.dll regapi.dll resutils.dll rsaenh.dll ureg.dll Dcache.bin mib.bin oembios.bin appwiz.cpl bthprops.cpl firewall.cpl hdwwiz.cpl inetcpl.cpl intl.cpl ncpa.cpl nusrmgr.cpl nwc.cpl powercfg.cpl
sysdm.cpl timedate.cpl ega.cpi FNTCACHE.DAT oembios.dat secupd.dat compmgmt.msc devmgmt.msc dfrg.msc diskmgmt.msc fsmgmt.msc gpedit.msc lusrmgr.msc ntmsmgr.msc ntmsoprq.msc rsop.msc secpol.msc services.msc winoldap.mod chcp.com diskcomp.com diskcopy.com
edit.com format.com graftabl.com graphics.com kb16.com loadfix.com mode.com more.com tree.com win.com ieuinit.inf
Please note, that in this series of files some are from windows root, some from windows-system, system32 and system32-config folders.
Reboot and listen to some music.
* CMP doesn't react to the absence of taskmanager and loads perfectly. Also, I didn't remove taskill.exe. When I last saw task manager on startup, it showed:
Commit charge (before loading the library titles, but with cicsremote aborted - 20,444 mb, with remote - 22,496, no remote and 700+ titles loaded, OK pressed to close the diagnostics window - 25,628)
Serge.
P.S. Inmates, I know that some of you might have lost your usb flash drive capability. Please, find its dll. I am positively exhausted to return and "search" for it, besides, I don't need it, while other folks may.
Whew!! Finally reached all of 15...very very nice sonically. Talk about Stairway to Heaven this is 'Steppes to Heaven' afaic. I still have ability to invoke Explore from Cmp screen to load all my files into cmp (I use a folder to only hold a few dozen files in Cmp). Also along this journey I did not delete any exe or dll files with 'usb' or 'fat' phrases in them and I still have ability to transfer files from thumbdrive to cmp hd. I probably could delete some of these but for now I'll settle with what I have. I'm still mulling over how much of 16 I want to do...think I want keyboard functionality. I guess when I will need more hd's I'll reload an earlier version to do file management. I know I know I should have done a network.Thank you Serge, this a masterpiece. I cannot express how appreciative I am for all your hard work. If going from Steppe 7 to 15 was grueling (for a non-computer geek like me)I can not even imagine the work you put into this. Well done!!!
Edits: 10/03/11
Hello, everybody.
This is at last part 16.
You see, that it is extremely short.
Also, with all the tweaks from part 11-16 you will be bound to service sometimes your hdd from within another Windows computer or hdd with a copy of windows with all its functions intact.
I almost cannot find any more significant files that can be removed without a blue screen crash.
I will post here winspool.drv and some other files to be removed aside from the main bulk, but you should know that removing them you will have no keyboard, and no cmp.
However, I find that this extreme setup sounds better than cmp. Such is the reality (in my ears).
1.Safe to remove (cmp working): cmd.exe diskpart.exe fastfat.sys gdi.exe netui0.dll objsel.dll openfiles.exe parvdm.sys shell.dll shellstyle.dll shfolder.dll storage.dll storprop.dll vmmreg32.dll w32topl.dll
2. Unsafe to remove (no more cmp shell, no keyboard, no notepad to edit cue files):
winspool.drv (I don't know if it has anything in common with wispool.exe, but WHY would Cics need it, is a mystery)
KEYBOARD.DRV (from windows system)
kbdus.dll (last keyboard layout from system32)
kbdclass.sys (from system32-drivers)
olepro32.dll (from system 32) also necessary for cmp but not for cplay.
A very dirty file as all files from ole sh*tpack
OLESVR.DLL - another ole wonderkid.
Without these files cplay works fine. Moreover, it sounds standalone better than from within cmp but with dirty files in the system.
So it is a "controversy" about convenience versus better sound. Inmates, it is for you to decide, what is more important.
Also, there was a question if AWE still works. I guess it does, for cics warned that, if awe were ticked in the cplay dialogue
without the corresponding tweak, the player would exit. Mine doesn't.
Serge.
Went through steppe 16 (except for key board stuff and olepro32.dll and winspool.drv) and my initial thoughts were 'maybe a step too far' but I gave it time to see if I would get used to it but I couldn't. In anticipation of deleting olepro32 I wanted to see how much degredation I would get by 'not going through cmp' and running cplay naked and guess what ... I flatout preferred it. It is so pure in tone now that it is a whole new level ims. Cant wait to delete olepro32.dll now.
Edits: 10/03/11
Ok more feedback on sonics. I cant really say how cplay within or not within CMP mode may sound because I either have a fouled up cmp program file or I still don't implement it right but....I have gone the full monty Steppe 16 ...all keyboard stuff out along with olepro32 out, winspool out from Windows and system 32 and I'm drinking the coolaid! The Windows folder is down to 103 mb. What a difference these last few file deletions make. Seems as if music gets louder, more dynamic, less bright, tighter bass and bigger soundstage all around @ full 16. Interestingly if I leave winspool in Windows folder but not system 32 it changes the sonics. I don't want to overhype but it is real good.
This is crazy. Took out olepro32 and it was more transparent but after some time (a day or so) I concluded it was a bit too much. Put it back in and the slight hardness went away. Then I thought what the heck lets try taking winspool out along with olepro32 and voila no hardness and bass was back big time. So only thing left to try is the keyboard stuff and only God (and Steppe) know what that will sound like. I now believe there is a pattern here and that is to follow Steppe's order of deletions.
that you are so close to the pot of gold (at the end of the rainbow).
Though, I (still) do not think it makes much, if any, difference, in the order of deletion. I know with my nLite install there were lots of files "pre-deleted" for steps 1 through 16 (though I have not done it yet successfully). I have looked through and a few are not there even for 16.
I am very frustrated by my mis-step.
I would think it easy to go back and forth between keyboard and no keyboard. Just put those in a folder and re-install as needed. That is my plan. I did prefer using my keyboard to the mouse, though, so it will be a hardship. One wonders: can one do without the mouse and get the same effect or are the keyboard drivers "dirtier" than the mouse drivers?
(I would have no idea which files to delete for the mouse)
Hope to hearing what 16 can do in a few days.
Extra Urgent!!! Something is wrong on my side of internet, thus these crazy postings, inmates. Go to your bios setup, press control+F1 to load ehanced menues, get to advanced chipset features inside it - in my mobo only oe item concerning vga or video, sorry,inmates, print it by memory in terrible internet conditions. This settings allows for max or 256 or 128 mb. Set at first to 256 mb. reboot. If cmp works set it to 128 mb.
Hello, everybody.
Here is some info that may or may not interest you, still, I think it should be posted.
The very first time I have done some tweaking of cmp setup was in the times of cplay 1-14 or 1-15.
Some of you may remember, there was a very soffft version followed by first truly great versions 1-17 and 1-19. I had a cmp based on the first prescribed mobo - biostar with sb audigy 2 soundcard and asio4all.
I noticed how cics advocated reducing windows footprint, and it reminded me of freeing memory in DOS times to launch some ega game on my first 286 machine.
So, what the hell, I said to myself, why not try it here. I tried and stopped from msconfig Windows Time. It immediately reduced commit charge in task manager from 36,5-37 mb to 35 and the sq seemed slightly better. I reported it, but no one seemed to notice.
At approximately the same time I asked Cics, if he was going to do anything about "windows' beloved infant terrible" - OLE, but he did not respond. So, be it. I was happy with the way things had been developing with a fresh
cplay every weekend. Still it was nagging me, that so much difference became evident version after version, as if somebody was playing with a d*ned equalizzzer! Though I was sure Cics wouldn't do such thing. Why, then,
the difference is so evident? After I turned to using onboard Azahlea sound device, it seemed so good after soundblaster, that I stopped worrying at all for some more time. And then it happened - Cics stopped giving us
new versions, and here we all were, sitting on our chairs, doing nothing. It was about that time that I have returned home after spending practically 4 years in Moscow. I sold my business, there was nothing pressing for me to do here, but to build another cmp with all new prescribed components.
I did it, but even splendid Lynx soundcard did little to suffuse my disappointment, that there were no new versions. The rest about belgian monastic beer and first set of tweaks, you all know.
It wasn't all blind tweaking, especially from part 5 and onward, as I began to see how things work within the window frame, but, honestly, to a great extent, it was trial and error. With dozens and dozens reboots and crashes. And now we are here.
I know this set of tweaks isn't an elegant thing, moreover, I know how time consuming and dull it is, but the result is worth it, I hope. At least to me.
Some additional info:
Most important listening was done at night. I have a dedicated custom line (made by agreement with servicing company) from the common feeder on the ground floor, no other apartment is hoggin' it. I live on the fifth floor, so there is appr. 65 feet from the feeder to my apartment.
Inside the apartment there are three different lines. Good thick copper wiring from my Accuphase days. The room has splendid proportions, though small.
Hdd, cmp machine and speakers all sit on different lines (2 lines) and all on different wall plugs. Kids from apartment below me have a hatefull wi-fi router, that explains partly my night vigils - their dad makes them switch everything off. No other neibors above or to the side, so we may say that power is more or less clean.
I used for reference the following cd-rips from my library
David Byrn - Uh-Oh (a very well recorded pre-loudness war nonremaster edition with layer upon layer of natural instruments, clever electronics and heaps of percussion)
Diana Krall - Quiet Nights (do I have to comment on the level of sq and amount of tiniest details?)
Porcupine Tree - Deadwing and Fear Of A Blank Planet - charged and smart heavy progressive with very complex sound canvas and both albums terribly compressed. (original editions)
Fish - 13th Star - a Very strong album from former Marillion frontman, overproduced and compressed as sh*t, almost undecodable till the very end. Really, Mr. Dick had it severely overdone.
Robert Plant and The Strange Sensations - Mighty Rearranger (European original edition). Also a very strong album with very complex and varying sounscape. Tracks 2 and 7 still slightly grainy.
Bob Dylan - Slow Train Coming - who would have thought there is such an amount of splendid musicianship! A very difficult album, though seemingly an absolute simpleton and no-problem.
Alice Cooper - Welcome To My Nightmare - a real nightmare remaster probably because of some specific coinsidence of recording (mastering) scheme and a set of instruments. It is grainy.
Adriano celentano - Azzurro (1969) Wait, till mandolin sounds like a mandolin
Jethro Tull J-Tull Dot Com - not exactly difficult soundwise, but I like and know this album well.
Van Der Graaf Generator - Trisector. Beautiful sound, wonderful music, complex at times, but a real pearl. A treat for every serious listener.
That is all, inmates. I am exhausted, got a terrible cold and a high temperature, lost (burned) my flash drive in a public internet cafe, my 14-year old beloved kitty is terribly ill these days (she almost can't walk and feed anymore), I hate to listen to the sound of files. I want to make a break and listen to some sad songs.
Serge.
P.S. All said is within the IMHO frame, I am not away from the forum, I just want a break.
P.P.S. How about some beer and then to remove msconfig.exe?
You said ...'And then it happened - Cics stopped giving us
new versions, and here we all were, sitting on our chairs, doing nothing'...After the Steppe journey I don't think anyone will ever accuse you of sitting on your chair doing nothing. But I find it ironic that you love Dianne Krall and Bob Dylan while I love Arvo Part's symphony #3. Go figure!
Serge,
It is not true that having the button checked will make cMP crash if you have not implemented AWE. Makes no difference on my machine and I KNOW I do not have AWE implemented.
ONLY way to tell is from the DIAGNOSTICS. If it says AWE was used then it was. If it says nothing: no AWE.
Have got to #12 and can't wait to go further this evening.
Tried to do them all at once and it did not work out. Have to re-install all of those files and start over. Working fine now.
Thanks, again,
Hi steppe,
Thanks for all the hard works. Though I am still at steppe 5 (don't want to do the heavy lifting until I find a safe way to mirror my cMP partition in a dual boot system), I can see that there is a great deal to look forward to.
It is great to know that you are a music lover. But then everyone who is so deep in the tweaking game got to be one.
KC
and you are probably not anxious to give your address.
This adventure has been like a good novel. Unexpected twists and turns - cMP actually turning out to not be the ultimate set-up. Did you figure out how to remove what it was you mentioned around #5? Not that it matters; like I said like a good novel.
You must acknowledge our other Serge's recommendation for the setting in ADVANCED CHIPSET FEATURES. It is how he has his BIOS set-up on his cMP website.
I cannot wait to get to work on this. So I will get to work!
THANKS VERY MUCH,
What is the difference in ADVANCED CHIPSET FEATURES per Mihaylov?
I had been using it for awhile, though my 512mB memory will not work with this suggested settings, which has nothing to do with the first part of my sentence. That works very well.
using something like this adapter:
http://www.virtuavia.eu/shop/pci-32-bits-1-x-pci-express-adapter-.html
Mihaylov, what do you think? And any others!
The ASUS board was a complete disappointment. BIOS adjustments were not available as promised.
I know I should give up on this 1155 socket set-up but I bought the CPU and wonder what to do with it.
I would not advise it and another. I would suggest P8Q67-M DO. Look review P8Q67-M DO (link below). The BIOS settings are fine.
Serge.
http://cmp2-mihaylov.narod.ru/
the two people who have used the board say nothing nice about the board; they say it is "cheap".
I think I best stay with what I have.
I am committed to JULI@ being easily used for i2s output and therefore PCI.
Nothing has been suggested as a gook replacement. Not ready to buy the LYNX!!! Too expensive
Thanks for the input, it is greatly appreciated.
I'm sorry for you Rick but I think you've done some strange choices on mobo model.
My excellent GA H67MA UD2H B3 was discontinuated to promote newer Z68 MX ( it seems to me the same mobo with a different chipset!)and this was the mobo I was thinking you purchase.
I'm shure this one nust be a good choice for a cMP but......PCI must be forget, Intel doesn't support it in the future ( and in the present too).
Adaptor PCI-E> PCI not seems to me a good idea: rumor rumor rumor!
Dear community
Just a little warning : When configuring cMP, I use XPlite to disable Windows File Protection (WFP). This works fine, but don´t reinstall XPlite after minlogon to further reduce components.
Even with WFP disabled, I had two Blue Screen crashes when attempting modifying any components.
Prune off components BEFORE minlogon !
BTW.: DLLs remain in the system even when XPlite has disabled lots of stuff.....
kind regards
cMP2 Computer: i3-540 Core cPlay039 >FIR filter w/ convolver> Lynx Aurora 8 FireWire /192kHz throughout. 2x AcousticReality Ref. 202 & 2x AcousticReality Ref. 601´s ICEpower. Magnepan MG3.3R beechwood frames & custom stands. Miller chokes
yet on a new installation I was working from memory and kept wondering why I was having trouble with XPLite.
As soon as I went back to WINLOGON all was fine.
I think deleting as much stuff with nLite is better if only I could figure out what to leave in to still allow AWE!
if one is determined to use the 35 watt processor.
Has voltage adjusting and the ability to disable Spread Spectrum.
I discovered this after buying the h67 version which has all of the voltage adjusting stuff but not the ability to disable SS. Oh, well ...
GIGABYTE is offering nothing like this and it is cheap! And small.
The chipset Intel H61 has not native support for PCI bus. Therefore it is not suitable for Juli.
You should not rely on cheap system boards. At least all capacitors must be solid-state on the good system board.
Serge.
http://cmp2-mihaylov.narod.ru/
How about the H67? Does it have the same problem?
That is the one I purchased. I was disappointed that one could not disable SPREAD SPECTRUM with the H67 version, as if I have any idea what that means and how it affects the board's "audio" behaviour!
Could it be that the 1155 boards are not suitable for cMP use?
I don't quite understand about the board's expense. I do nto remember paying much more for the recommended GUGABYTE boards. This one was cheaper than the GIGABYTE x68 board I had tried out with disappointing results.
Other than your previous recommendation, which I feel quite sure is about the same, BIOS wise, as the one I tried, is there anything else out there that looks interesting to you?
"How about the H67? Does it have the same problem?" - Yes. This chipset has not the native support for PCI also. Also as well as P67, Z68, H61. Only Q67 and B65 has the native support for PCI but it's bios settings do not correspond to requirements of cMP2.
"Could it be that the 1155 boards are not suitable for cMP use?" - For example GA-H67M-UD2H-B3 is quite suitable for cMP use but only with PCIe sound card.
High cost of the system board (not less than 100 $ approximately) confirms the general high level of fabrication and components that influences on SQ.
Serge.
http://cmp2-mihaylov.narod.ru/
I should have been more clear with my question:
is there an 1155 board suitable for cMP with JULI@?
I guess there is not!
Thanks, again.
NO for Gigabyte's system board with "classical" (advanced for cMP2) bios settings. It is possible to try Asus P8Q67-M DO (it's bios settings is quite advanced; see manual from asus website).
Serge.
http://cmp2-mihaylov.narod.ru/
Edits: 09/28/11
might be a good choice.
Can't find one for sale.
In Moscow is on sale :)
Serge.
http://cmp2-mihaylov.narod.ru/
Edits: 09/28/11
Looks like it has good BIOS and the PCI native support and good build quality.
Maybe will we get it in a few months.
Does this look like a good board to you?
Where does it fall in the range of prices?
Yes, it's not bad board. Although at a cost of $150 all capacitors could be solid state (as in a Gigabyte boards for the same or even lower price).
Serge.
http://cmp2-mihaylov.narod.ru/
Edits: 09/28/11
not to rush you, but ...
ready for the next step. (could not resist)
It was not good.
The one I got, because you could adjust cpu voltage, GA-Z68XP-UD3P.
I did not realize when I ordered that its only video output was HDMI. This started off working OK until I tried to install the MB drivers. These could not be installed without ALSO installing .NET Framework which I did not want but without the MB drivers it would not see JULI@ in the PA slot.
It would not allow WINDOWS to be loaded from an nLite disk.
None of the BIOS setting would allow what we can get with the 1156 board recommended by cics.
All in all, very disappointing.
Now I fear that this is how it is for all of the x68 equipped boards.
On its way back to NEWEGG. And the search continues.
Imho we need to use Gigabyte GA-Z68MX-UD2H-B3. It's bios is fully compliant with cMP2.
Serge.
http://cmp2-mihaylov.narod.ru/
it was fun while it lasted.
I got a feeling it is the same as what I had. You can only lower the voltage to 0.85. CPU Clock Ratio - nowhere close to the recommended for the previous board. Anything that could be adjusted could not be brought down to what we have used before.
It looks like it has the same under-running abilities but it does not.
Unless you have one! And that would change everything. If you have no used one I do think it is the same BIOS as what I had.
Committed to my JULI@ set-up, can't make a change even if it was "good".
THANKS for the advice!
Hi all users of cMP2 network player!
You as well as me obviously should connect manually the network disk with the help of bat file. That the network disk was connected automatically at startup of cMP it is necessary to edit in a notepad the files cicsMemoryPlayer.pth and Logon.bat as follows.1. Edit the “PROCESS_EXPLORER” line in C:\Program Files\cics Memory player\cics memory player.pth to read:
PROCESS_EXPLORER "c:\Logon.bat"2. Edit the file Logon.bat that it looked so:
SetLocal EnableExtensions EnableDelayedExpansion
:LogStora
NET USE M: /delete
NET USE M: \\Server\Folder /User:Server\UserName password: /persistent:no
If Not "!ErrorLevel!"=="0" GoTo :LogStora
Exit(I use log in without the password therefore the fourth line in my case looks so:
NET USE M: \\MyBookLive\Public /persistent:no)Now it is necessary to select option "Starup -> Process Explorer" in settings of cMP and voilà :).
Serge.
http://cmp2-mihaylov.narod.ru/
Edits: 09/24/11
". . . select option "Starup -> Process Explorer" in settings of cMP and voilà"
And "voilà!" it is, too. A neat idea; works a treat. Many thanks.
D
You are the greatest! I can't wait to try this. I have tried using a logon.bat file with a net use command in it but clearly didn't know all that had to be in it until now.
Hi cMP friends,
could we please make a map of sound cards used with new tweaks.
It would be interesting, for example, to establish how much effect new tweaks have on asyncronous interfaces
Every forumer just need to say something like:
Juli@/spdif/Weis DAC202
firewire/Lynx Aurora
usb/M2Tech Young
....
Juli@---spdif---> Buf32s dac
firewire(via expresscard in Lenovo laptop) to Weiss DAC202
Juli@ => Spdif => Accuphase DC-61
USB/NOS tda1541a tube output DAC
I had big trouble getting workstation service to start after finally getting all the restoration of files and registry keys right a la Mihaylov's very helpful posts. it turns out one has to add back rasman.dll to get it to do so IF you have the bigger problem I came upon:
I had moved from one home/computer setup to another and only hours into this networking mess discovered that because the first router uses 192.168.0.1-200 and the one I moved to uses 198.168.1.1-200 I could not connect and workstation doesn't get turned on automatically. I first had to figure out how to do ipconfig to see where I was at re static ip. It turns out to do this you have to add back additionally:
dbtcp.dll
netman.dll
netcfgx.dll
I learned my static ip was stuck at 192.168.0.198 while I needed to get it to one with a 1 rather than 0 in the 3rd position to be in the router's range. I couldn't figure out how to do this without going back to services and starting the several needed to get to network connections to set the static ip the usual way. I tried brute force which was to go thru the registry and change every single entry I could find with "find" that had a 0 and change it to a 1. Alas, upon restart ipconfig showed no change in the static ip. I don't know how this is possible, but...
I searched and found a command line method of setting the static ip. The command is lengthily:
"netsh interface ip set address name="Local Area Connection" static 192.168.1.198 255.255.255.0 192.168.1.1 1"
this worked and I was able to connect to the router which was set to the static assignment of 192.168.1.198 as well....BUT ONLY UNTIL I RESTARTED. After restart it reverted to what it had been despite a registry chock full of the new address I was trying to maintain. Maybe one of you computer experts knows how to permanently change the static ip from the command line in this situation. This command is supposed to and probably would in a normal system.
I finally had to do last resort and changed the router over to new 0 addresses for itself and thus all else on the network. Now it connects in the usual way, namely, giving the net use command at each startup. I will try /persistent:yes after a bit to see if that gets it to 'stick' after restarts.
Hi all,Are there any inmates who checked with Process Explorer (Sysinternals, Mark Russinovich) which DLL’s, EXE and other files are actually in use when a standard cMP setup (Juli@ digital / cPlay) is playing music?
I first tried un registering DLL files with this command: regsvr32 /u C:\Windows\System32\dllname.dll.
This way I discovered that approximately half of the dll files on the list are not in use and the other half is in use (DLL which are in use, can not be unregistered at that moment).Knowing what dll’s and other files are in use when playing music before and after Steppe’s SQ Tweaks, might shorten the deletion list dramatically.
So I was wondering if there already are any inmates who checked with Process Explorer which files are actually being used and which files are not being used when playing music on a cMP setup.
I’m especially interested in shortening the deletion list, as I already lost track of what files must be deleted at part 3. I see same dll’s re-appearing on the deletion list.
Mark
NB.
Thank you Steppe and others !! Although the communication through the asylum on which files have too be deleted (and which files not) is somewhat confusing for me, I must say: Thank you Steppe ! (and others). So far I only progressed too part 3 but the SQ improvement is already amazing.
fully optimized cMP2 PC -> ESI Juli@ -> Van den Hul Optocoupler MkII-> Lavry Black DA10 -> XLR Mogami Gold -> Klein & Hummel O300
Edits: 09/23/11 09/23/11
It is all successfully works in my configuration of cMP2.
The network configuration of my cMP2 is based on cic's "Headless cMP ² using a Fit-PC2" namely:
- Static IP addresses avoid DHCP’s overhead and work well with UltraVNC. Recommended.
- The remote control utility is... the best of those tested ... UltraVNC.
- cPlay can be launched using cMP’s library facility as normal except that it is necessary first to log on to the network drive (NAS) by means of a command "Net Use M: \\Server\Folder/User:Server\UserName password:" started by click the ‘ RIP ’ button in the cMP window.
But all components and full optimisation of OS and BIOS corresponds to the "classical" configuration of cMP2 with the involved network possibilities.
Serge.
http://cmp2-mihaylov.narod.ru/
I read that you are already at step 11...
How many more steps do you think are coming (or necessary)?
If you Crtl-Alt-Del, how much memory is used by the system at the moment?
This would be a nice indication to have each step you make. Other users at stage 5,6 or 7 can give it too. Thanks
Thanks for all your efforts
The Steppe's Tweaks part 7 is possible to apply completely in a network configuration of cMP2 BUT:
do not delete following files:shutdown.exe
netmsg.dll
netplwiz.dll
netshell.dll
ntlanui.dll
tcpmon.iniThe network configuration of my cMP2 is based on cic's "Headless cMP ² using a Fit-PC2" namely:
- Static IP addresses avoid DHCP’s overhead and work well with UltraVNC. Recommended.
- The remote control utility is... the best of those tested ... UltraVNC.
- cPlay can be launched using cMP’s library facility as normal except that it is necessary first to log on to the network drive (NAS) by means of a command "Net Use M: \\Server\Folder/User:Server\UserName password:" started by click the ‘ RIP ’ button in the cMP window.
Serge.
http://cmp2-mihaylov.narod.ru/
Edits: 09/22/11
before getting into any of these steps...or, at least, I did not find them when looking to delete them. Are you sure these are needed now?
"I was networking without netplwiz.dll and ntlanui.dll"
I'm sure you were but "networking" covers a multitude of sins. Which XP drivers and which .dll files are needed and which aren't in any given situation surely depends on how the network is being used. Pulling data from a server onto a cMP^2 box's local drive using local KVM is very different to a headless box remotely controlled by a VNC service even if both are "networking". In any case and as you'd expect, different VNC utilities seem to rely on different native files. Some users run the DHCP, others use fixed IP addresses. And so on.
The same applies to soundcard/DAC drivers. Some use the USB, some the PCI, some the Firewire bus. The orthodoxly-pure, the very-sainted-indeed St Juli@ jostles for position with RME and ASUS soundcards, with different bespoke ASIO drivers, with the multi-device AQVOX/Ploytek USB driver, with high-end asynch drivers, with ASIO4ALL and who knows what else.
And, while you're at it, there's folk using VST utilities, room-correction software, digital crossovers and other stuff that it scares me even to think about.
Steppe, Jolida, Mihaylov and others have generously reported on the "resource depletions" that worked in their setup. If they sometimes get a bit carried away, well, if I'd done what they've done, I'd get a bit carried away too.
It's probably fair to say that the cMP^2 "community" generally agrees that the various procedures lead to improved (and sometimes to dramatically improved) SQ.
I'd argue that there is no royal road to realising these improvements - there are, unavoidably, just too many variables. cMP^2 "slimmers" need systematically to test the various recommendations for themselves. It's time consuming, esp for folk unfamiliar with Registry editing and it can be frustrating. But it's gratifying when it works.
Bit like many hobbies, really.
I get what you say, but more importantly: I was wrong. I wasn't networking without those. I misread my notes made in process of following the sacred stages of the cross-out. Yes, we are all on individual odysseys of finding our way to blissful slimness (to screw up the metaphors entirely), but amazingly enough steppe and mihaylov and jolida have guided us with remarkable success over all.
But that is what I have been doing: connecting via net use command, and I did not find these dll's. I will check with another copy of the system on a different disk.
If these files are missing then the network drive (NAS) will not connect using the Net Use.
Serge.
http://cmp2-mihaylov.narod.ru/
My mistake. I misread my notes. I had removed these. Sorry.
The Steppe's Tweaks part 6 is possible to apply completely in a network configuration of cMP2 BUT:
do not delete following files:net1.exe
net.exe
netsh.exe
ping6.exeThe network configuration of my cMP2 is based on cic's "Headless cMP ² using a Fit-PC2" namely:
- Static IP addresses avoid DHCP’s overhead and work well with UltraVNC. Recommended.
- The remote control utility is... the best of those tested ... UltraVNC.
- cPlay can be launched using cMP’s library facility as normal except that it is necessary first to log on to the network drive (NAS) by means of a command "Net Use M: \\Server\Folder/User:Server\UserName password:" started by click the ‘ RIP ’ button in the cMP window.
Serge.
http://cmp2-mihaylov.narod.ru/
Edits: 09/22/11
The Steppe's Tweaks part 5 is possible to apply completely in a network configuration of cMP2 BUT:
do not delete following files and appropriate records in the registry:
nddeapi.dll
netrap.dll
afd.sys
ipsec.sys
rdbss.sys
tcpip.sys
tcpip6.sys
termdd.sysThe network configuration of my cMP2 is based on cic's "Headless cMP ² using a Fit-PC2" namely:
- Static IP addresses avoid DHCP’s overhead and work well with UltraVNC. Recommended.
- The remote control utility is... the best of those tested ... UltraVNC.
- cPlay can be launched using cMP’s library facility as normal except that it is necessary first to log on to the network drive (NAS) by means of a command "Net Use M: \\Server\Folder/User:Server\UserName password:" started by click the ‘ RIP ’ button in the cMP window.
Serge.
http://cmp2-mihaylov.narod.ru/
Edits: 09/20/11 09/22/11
I went ahead before this to try on my own preserving networking and did fairly well but did delete some on this list and associated registry keys. How do you put them back. I have another install of cMP pre steps 5-7, so can I copy keys from them...and how, if so?
Using .reg file that can be done using a template for minlogon (file name and registry keys must be of course different):
Using Notepad, create a file called minlogon.reg containing the following text:
Windows Registry Editor Version 5.00
[HKEY_LOCAL_MACHINE\SOFTWARE\Microsoft\Windows NT\CurrentVersion\Winlogon]
"Config"=dword:00000017
Double-click on minlogon.reg to make the requisite Registry change.
Serge.
http://cmp2-mihaylov.narod.ru/
Thanks so much, Mihaylov. I found on my own the way wlowes recommends. theob, I put the .reg exported files on the cMP desktop and double-clicked them in explorer. Doing so from the thumb drive would also probably work as well.
Mihaylov, I restored everything as best I could to what you said should stay. When I try to use the net use command I am told the workstation service is not started. When I give the net start workstation command I get "Workstation service terminated with service-specific error 2250". I have googled for solutions but find none that work. One message at microsoft said to use the net config workstation commmand to see what's what but this just says workstation is not started. I will review what I did to see if I missed something (once again), but meanwhile do you know what might be missing or wrong?
I put a bunch back by going to another XP machine,
open regedit
go and hilite the key you want
file-> export to a filename on a jump drive
then move drive to the cmp machine
each reg file you double click will restore the key you saved. prevents typos
This is an excellent solution...just some clarification. You said:
'...go and hilite the key you want...'
ok clear to me
'...file-> export to a filename on a jump drive...'
I assume the 'file-> export' command is available from regedit. Correct? Is filename pre-existing? By jump drive do you mean thumb drive?
'...then move drive to the cmp machine
each reg file you double click will restore the key you saved. prevents typos...'
So while still on thumb drive double click?
Thanks again
Basically yes to all. To export, in regedit it is File-export.
It will export everything below the spot you have highlighted. It will prompt you for the file name. you need to get that file to the target machine. Again in regedit File-Import. it will ask for the file name. You will be prompted for "do you really want to update the registry?'
click yes and it is done.
I restored all the Steppe 7 registry changes using this approach as I tried to debug my shutdown problem. Take a back up of the registry before starting for safety. I believe you get the whole thing by highlighting the very first entry.
thanks
The good way!
Serge.
http://cmp2-mihaylov.narod.ru/
Edits: 09/22/11
I discovered from my weekend adventure with nLite that even though I left GROUP POLICY blah blah (can't remember the rest) IN I could not engage AWE.
Makes me wonder if the same thing occurs with XPLite since Serve recommended taking out more stuff that I did with nLite.
Those that have tried it might check their diagnostics and see.
So far no problems, other than that, with the nLite installation.
I've not started any file deleting, yet. It will be interesting to see if any of the recommended deletions were deleted by nLite. Even better if many of the registry changes are made.
I did listen to the new install. So far, just the standard cMP recommendations and removal of audio codecs. The sound is IN THE ROOM, all of the spaciousness has gone and the high frequencies are COARSE in comparison to what I heard before ...
So tonight, I will try to get as many deletions of files I can before I fall asleep. Reading that article Ryelands linked gave me the impression that I should be able to delete all of the files in one go and then to the registry. At that point I will get the software that allows you to of back to what worked if you do something wrong (like I did before, did not think I did, but I must have)
Bye,
I had just installed a nLite version, my first, for a friend of mine. Same experience as yours, that is no AWE, and I was wondering what I had done wrong. My experience with nLite, one that is in non audio terms because I did not compare it with my main setup:
-- it speeds up the computer response a great deal comparing to a standard XP installation, and
-- a lot of the steppe # 4 registry deletions are already done by nlite, say 90%
wondering if the same thing happens for them?
The nLite is very interesting.
I had intended to do some file deleting last night but went to see SENNA, the documentary about the life of Ayrton Senna which was very entertaining but ultimately sad.
Tonight I will get to work and am glad to hear I will have less deleting to do.
Thanks,
Finally had the time to go ahead with Part 5 today. This time, I wanted to do this Part in three stages. So that I could analyse which section is doing what, & also if any complications show up in the working of windows, it would be easy to undo them.
In the first stage, I accomplished from the start of Part 5 till the section "i". Analysed this change for more than two hours & was very impressed. Very open, full-bodied & images had a good Size to them. A sure improvement.
Went ahead to doing stage two, which involved continuing from "i" till the end of Part 5, BUT without deleting any dll or exe or sys files. Just the Registry deletions. This stage was totally different. The Music became totally laid-back & distant, definitely cleaner, but lacked the bite & involvement that was apparent in the first stage. No downside though. Just a different presentation. Listened to it for another two hours & went to stage three.
This stage involved doing the left-over deletions, all of which were deleting dll, exe, & sys files (except the wdmaud.drv & the wdmaud in the registry). This stage proved to be an eye opener. As if the ENTIRE music spectrum had cleaned up. Absolutely controlled right from the bottom to the top-end. Not a hint of glare or grunge in the lower-upper mids. Bass seemed a weeeee bit less, but very very controlled. Noise floor dropped down unbelievably low. Had to turn up the volume more & more & more, finally to realise that Music is now playing like Music. The sense of Electronic-sound had begun to diminish. Very analogue like. Very pleasing & there is a Grip on the entire frequency range. A slight feeling of the highs going down a notch (little less definition of instruments), but im sure it will get way better till Part 7 is done. An out & out success. Excellent job Serge..
A milestone in the cMp project !!!
Edits: 09/18/11 09/18/11
I too had finally done steppe # 5. I had also repeated the experiment with wdmaud.drv.
steppe # 5 after all is a loooong process so much so that I had forgotten about wdmaud and had go ahead and deleted the related register entries. Too late, thought I, might as well go ahead and remove the driver which after all those deletions would probably do not have the same significance as was back then. I was wrong.
After firing up steppe # 5 and listened for a full day, though the sound was very clean, somehow it seemed light weight. Then I thought of wdmaud.drv. Too late to restore the registry entries, but I can restore the driver and see if that works. It does!
To my great relief, the weight is back, bass guitar again sounded right, instead of in the next room. Thanks steppe for the great and meticulous work, and Jolida for the finding about wdmaud.drv
I'm sure there are many people who have done the same. I was cautious at every step about anything to do with wdm*** deletions, be it files or registry. Also, I take back-ups after every step I do. Let me know in which step u are, & I shall email u a back-up of the registry at that particular step so that u can restore it...
Hi kclo
are you performed some step of " cancellations files fever"?
I think it could be possible you've cancelled something wrong for your system. It seems to me the only valid explication for your strange fau
Daniele
Why i can't run cPlay under Win 7 x64 with native ASIO driver of my card? cPlay working only when i use ASIO4ALL.
Hi
which sound card driver are you using?
Note that cplay needs 32bit ASIO only.
If you are using foobar too you can open foobar' setting, open the window for setting ASIO and select 32 bit from here but it needs your driver get this option, obviously!
Daniele
My apologies! I think it was something wrong with the OS. But today i came home and starting to search, and finaly i found it. It was selected a 24 bit in ASIO driver. I select 32 and it's run perfect!!! This programm great! I love it! Thank you for this great player and many thanks for your help! God bless you!
Hello, everybody. This is tweaks part 7 and it is very simple, even dull. But it is really worth it, I think. Before you start, while you still can, remove all video codecs from your system. I have forgotten to mention it before, however, you will delete these files anyway.
This is what I propose to do in part 7:
Delete or remove to your temp folder these files:
getmac.exe ipconfig.exe ipsec6.exe ipv6.exe lights.exe makecab.exe mpnotify.exe mshta.exe nwscript.exe packager.exe powercfg.exe rsnotify.exe shutdown.exe spnpinst.exe wextract.exe winhlp32.exe wpabaln.exe 6to4svc.dll
6to4svc.dll atkctrs.dll atmfd.dll atmlib.dll atmpvcno.dll batmeter.dll batt.dll bthci.dll bthserv.dll btpanui.dll cabinet.dll cabview.dll camocx.dll cards.dll ccfgnt.dll cmsetACL.dll cnbjmon.dll cscdll.dll ctl3d32.dll
ctl3dv2.dll ddraw.dll dgrpsetu.dll dgsetup.dll dhcpcsvc.dll dhcpmon.dll dhcpsapi.dll docprop2.dll dsauth.dll els.dll EqnClass.Dll faultrep.dll fwcfg.dll hhsetup.dll hnetcfg.dll hnetmon.dll hnetwiz.dll httpapi.dll iepeers.dll
imeshare.dll inetcomm.dll inetpp.dll inetppui.dll ipv6mon.dll ipxrip.dll irclass.dll jgdw400.dll jgpl400.dll jgsd400.dll jgsh400.dll KBDAL.DLL kbdcr.dll kbdaze.dll kbdazel.dll kbdbe.dll kbdbene.dll kbdblr.dll kbdbr.dll
kbdbu.dll kbdca.dll kbdcan.dll kbdcz1.dll kbdcz2.dll kbdcz.dll kbdda.dll kbdes.dll kbdest.dll kbdfc.dll kbdfi1.dll kbdfi.dll kbdfo.dll kbdfr.dll kbdgae.dll kbdgkl.dll kbdgr1.dll kbdgr.dll kbdhe220.dll kbdhe319.dll kbdhe.dll
kbdhela2.dll kbdhela3.dll kbdhept.dll kbdhu1.dll kbdhu.dll kbdic.dll kbdinbe1.dll kbdinben.dll kbdinmal.dll kbdir.dll kbdit142.dll kbdit.dll kbdkyr.dll kbdla.dll kbdlt1.dll kbdlt.dll kbdlv1.dll kbdlv.dll kbdmac.dll kbdmaori.dll
kbdmlt47.dll kbdmlt48.dll kbdmon.dll kbdne.dll kbdnec.dll kbdno1.dll kbdno.dll kbdpl1.dll kbdpl.dll kbdpo.dll kbdro.dll kbdru1.dll kbdru.dll kbdsf.dll kbdsg.dll kbdsl1.dll kbdsl.dll kbdsmsfi.dll kbdsmsno.dll kbdsp.dll kbdsw.dll
kbdtat.dll kbdtuf.dll kbdtuq.dll kbdur.dll kbduzb.dll kbdycc.dll kbdycc.dll kbdycl.dll lpk.dll lmrt.dll lprhelp.dll lprmonui.dll mdminst.dll mdwmdmsp.dll mlang.dll mll_mtf.dll mll_qic.dll modemui.dll modex.dll mscms.dll
msexcl40.dll msftedit.dll msimsg.dll msw3prt.dll ncxpnt.dll netcfgx.dll netevent.dll neth.dll netlogon.dll netman.dll netmsg.dll netplwiz.dll netplwiz.dll netshell.dll ntlanui.dll ntprint.dll ocmanage.dll p2p.dll p2pgasvc.dll
p2pgraph.dll p2pnetsh.dll p2psvc.dll panmap.dll photowiz.dll pmspl.dll winhttp.dll winipsec.dll wowfax.dll wowfaxui.dll wship6.dll xmlprov.dll tcpmon.ini irprops.cpl joy.cpl lanman.drv netware.drv edit.hlp net.hlp winhelp.hlp
html.iec webfldrs.msi eventvwr.msc logon.scr pcl.sep pscript.sep sysprint.sep sysprtj.sep ipconf.tsp ndptsp.tsp eventquery.vbs pubprn.vbs prncnfg.vbs prndrvr.vbs prnjobs.vbs prnmngr.vbs prnport.vbs prnqctl.vbs
Remove FOLDERS:
twain_32 and x86_Microsoft.Windows.Networking.RtcDll_6595b64144ccf1df_5.2.2.3_x-ww_d6bd8b95
Several simple registry tweaks
go: HKLM-System-current control set-control-Device classes and delete inside of the following entries, but not the entries themselves:
dff220f3, bf963d80, ad809coo, a7c7a5b1, 9ea331fa, 3e227e76, ffbb6e3f.
Now try to recall the services you have deleted in part 5 (print for convenience)
Now delete them in 2 other places, only here some of them are absent and others have word legacy in front.
Example: Beep from part 5 here is LEGACY_Beep
1 place: HKLM-System-Controlset001 (all sets) - Enum-root
2nd place: HKLM-System-ControlSet001-Hardware profiles-current-system-current control set- Enum-root
Remove Legacy services here too.
Reboot. Listen to some music.
Serge.
Finally got 7 done, and Serge is right. It is worth it. Scarey good SQ.
from the folder System32 otherwise you cannot change the start mode of cMP2 with "Start XP" to "Start cMP"!
Serge.
http://cmp2-mihaylov.narod.ru/
Edits: 09/21/11
THANK YOU!
I was going nuts. I thought I was headed for a rebuild. All works now. :)
And I cannot thank you enough for bringing this to our attention!!!!!!
This must be what caused my problem.
I can proceed with less concern about THAT happening again.
THANKS,
That's all right. :)
Serge.
http://cmp2-mihaylov.narod.ru/
Edits: 09/21/11
I thought I had compiles all of Serge's file deletions and I do not see it on there.
Did I miss it?
Thanks for the warning but wondering what brought this up.
"Did I miss it?"
Yes. I spotted the suggestion and skipped deleting the file but, unlike Mihaylov, didn't post on it. Should have done . . .
D
In fact I'm surprised how good these Steppes (0 through 7)have gone wrt implementation. Now a question: I deleted the subject file but never had a problem, should I put it back?
I wonder if there might have been something of a cumulative effect with my deleting this file after XPLite?
I do not think you have done this yet? Am I assuming incorrectly?
Very interesting, isn't it?
I think I would put it back in. Even Serge has thanked Mihaylov for the notice which makes me think he thinks it might be "a bridge too far".
I'm through with xplite/steppe 6 and 7 and all works well
Please let me know.
yes
Maybe I need to do re-install with the standard disk and then XPLite.
I wish I knew how important AWE is to a completely STEPPED machine.
Thanks!
I finished step 7 and have the same result as Rick. CMP always comes back to the desktop even tho I set it to start cmp on reboot. So I really can't get down into real cmp without explorer. I always have the desktop in the background.
I was confused about how to do this reg tweek..
"HKLM-System-current control set-control-Device classes and delete inside of the following entries, but not the entries themselves:
dff220f3, bf963d80, ad809coo, a7c7a5b1, 9ea331fa, 3e227e76, ffbb6e3f."
I went to dff220f3 and did not know if I should delete the keys below it in the left pane or the stuff dwords and values in the right side. Serge, Rick and Theob, which route did you take. I deleted the children keys on the left side.
I can still go back and redo this reg step because I took a reg export before starting.
The sound is very nice. Very clear and very rich. good imaging, but less air around the instruments. I suspect that is because I can't get into pure cmp. I like it in some ways but in some ways I preferred 6 for the air and micro detail.
thanks for your help
I am sure you saw his post.
I hope you can get in there through EXPLORER and return that file.
Let us know.
Rhanks Rick
really helped mebefore I messed it up .. whew!
I did as you did. It works fine. who knows if doing it on the right would be better?
'... dff220f3 and did not know if I should delete the keys below it in the left pane or the stuff dwords and values in the right side. Serge, Rick and Theob, which route did you take. I deleted the children keys on the left side...'I deleted stuff on right (misread you 1st time through) as I figured I can always delete stuff on left below dff220f3 if that is correct approach.
Edits: 09/21/11
Serge you said: '...Remove Legacy services here too...' at the end of your Steppe 7. Does that mean delete everything with the word Legacy in front of it?
Hello! NO! ONLY Legacy_service that corresponds to services from part 5.
Serge. Also big thanks to Mihaylov for finding the problem with shutdown.exe
Serge
Thanks
Theob
looks like you got thru step 7. Congrats. What was the answer here? Delete everything with word legacy in it?
Did all but that, still waiting for an answer.
Funny thing happened part way thru Steppe 7...I got Control Panel back and was able to get into Device mgr but not services. Wonder if I'll lose it again.
Done with 7 (except for last registry tweak) Serge brings it all back to a perfect blend of pristine/clear highs to full airy bass...just wonderful. Thank you Serge. Guess what I can still get into device Mgr (but not services). I can still stop Crypto with the batch file Cics posted for those into the new mobo hw. I think it was 10-18 months ago. Do a search for it.
For anybody who did the last registry deletions: did you delete all the individual items within the keys or just the dwords on the right?
I snapshot my os disc just before last 2 or 3 steps in 7 then plunged ahead and deleted all the interior stuff in the registry keys per Serge. Worked like a charm. The sonics improved as well. Last bit of making absolutely pure highs and a sound which is fully integrated and nothing dominates from highs to lows was the result. Great sense of fine imaging on classical music was never better ims. Serge saved the best for last. For anybody mulling over whether to go with 7 or not...I say go.
And I still have device mgr but no services access.
> > "And I still have device mgr but no services access."
Admistrative tools> services requires IE5.5 or above to run. You had deleted Internet Explorer in Step 5, item 2 e. I suppose.
I restored registry back-up to pre-2 e. and got back access to services.
Too bad I can't go beyond Step 5 as system insist that I activate my Windows when I get back to winlogon. But clicking on "Activate Now" yields nothing.
Maybe time to do a fresh Windows installation.
I have never connected my machine to the internet and therefore had not activated WINDOWS either.
MINLOGON does not care about activation as you well know.
When I did XPLite (before everything fell apart) I was shocked when I returned to WINLOGON that it wanted to be activated. I thought MINLOGON stopped the clock. I guess it just slows it down!
Tried SAFE MODE and IT WORKED!!!
Give it a try.
after I kept getting an error about windows being "unable to exactly check my license for this computer error code:0x80090019" when I tried to winlogon. I ran xplite in safe mode and let it do its thing which appeared to go fine, then did step 6. All is fine at this point. I will report back after I do step 7 about whether or not I have the problem rickmcinnis had at that point.
and I doubt you will.
No one can get these things absolutely right and with the amount of work Serge has put into this project I would NEVER give him a hard time.
What is GOOD is that someone like Mihaylov is there as a backup.
As you can tell I have nothing but the warmest of feelings and regard for Mihaylov for his warning and about a thousand other things.
Strangely, XPLiite is always requesting for Windows CDRom to be inserted during components removal.
Anybody experience this?
Right you are about Mihaylov. He posted about what not to delete to preserve networking as well--too late to stop me deleting some of them but now I know what to go back and try to fix. Without him, I'd be completely done for.
Thanks for the tip.
I did activate Windows via the telephone when I did my first installation.
Anyway it said that got 3 more days to activate Windows. Let's see if I can start Windows tomorrow.
Did you do the XPLite removal of components in Safe Mode?
I did end up losing the installation entirely after a few days.
It worked fine up until I deleted the files from the "seventh directive".
The the next morning I wanted to do the registry deletions and I could not get back to XP.
I have not considered whether that (using SAFE MODE) had anything to do with it until now. Unfortunately, I have no idea.
When you do the XPLite trick see if you still have AWE by checking the cMP diagnostics.
Bye,
I guess only problem one might have with 'no services' is disc management to add new hard drives. With device mgr one can uninstall stuff not needed on the IRQ and as I recall one can add new soundcard/soundcard drivers. So until those of us run out of hdd space for new music I think we're ok.
To add sound card drivers one would need to Start Event Log, Cryptographic Services, Windows Audio, and WMI.
I think you can do it w/o restarting these services. Cics outlined a way to do it from within minlogon and device mgr....a while back. Unfortunately I don't know a way to do the same for adding a hdd.
Edits: 09/20/11
I have had a few things I was not able to do which I will mention in a later post.
An interesting anomaly occurred which I thought I would tell: when asking cMP to OPEN with cMP after I said said yes to all of the questions and hit the "X" and it asked me to restart which I said YES to the machine would restart but not into cMP, back to DESKTOP. I found after the request and it restarted I had to use the START button and request restart then the machine would restart into cMP.
Needless to say this did cause a moment or two of panic.
Thought I would mention this in case anyone else has the same problem.
Thought I would finish up the last big registry change so I ask cMP to start with XP - it did not want to turn itself off but this was the case last might.
Comes back to cMP. Tried about five more times with the same result.
Then I had the bright idea of moving the cMP files out of the PROGRAMS folder thinking this would force an XP opening.
Well, you have guessed that this did not work and now I have nothing. I tried SAFE MODE hoping for a miracle but ...
I guess what I went through last night was part of this.
What's strange is that it sounded good last night even in what was obviously not the way it is supposed to be. For a few hours I had a dedicated cMP machine!
I assume there is nowhere to go with this but to start over. It would be really good if XPLite took care of HALF of these but I guess that is not the case.
If one does start over is it still a good idea to install XPLite in the order it was given or could there be some use in implementing that right before going into MINLOGON?
Sure wish there was a way to do batch deletions!
At least I will get more yard work done today, and tomorrow.
Hello!. Don't do xplite trick before you complete part 5. It is out of experience. The sound will be very edgy.
Serge.
P.S. I boot up into windows xp, then start cmp, everything works like a charm, and I am almost over with part 11 tweaks.
Don't worry, I am not mad or disappointed at all!
Did an nLite install (if you did not see my other ramblings) and figure I got rid of most of the stuff to begin with.
I wonder if there can be a conflict with XPLite?
I tried doing it to begin with without being careful to not delete stuff that was not there - I think it confused the issue.
Before I try it again I will make sure to only delete what I left in.
Last night I deleted all files up to number six and was surprised how good it did sound (all codecs disabled or deleted and cMP optimizations done) and though it was not where it was before I was thinking for the squeamish it sounded pretty good!
Have you noticed whether you still have AWE after all of the changes?
Bye,
Happens to me all the time that's why I have Snapshot.
but as I admitted earlier I am mindlessly decisive.
I am trying an nLite install disk.
Will be interesting to see what happens.
You said: '...An interesting anomaly occurred which I thought I would tell: when asking cMP to OPEN with cMP after I said said yes to all of the questions and hit the "X" and it asked me to restart which I said YES to the machine would restart but not into cMP, back to DESKTOP...'
Just so I understand you clicked on the Cmp logo to start...it started...you clicked on cmp mode...it asked the 3 questions... you answered yes to all and then clicked on the x button. Is that right?
and it would bring me back to XP.
When I did START/RESTART, instead of trying for the fourth time doing it the old, standard way it opened in cMP mode. Thankfully!
Strange, isn't it?
I sent you a note asking for help on accessing SERVICES through the COMMAND LINE and you did not respond! I could use the help.
Bye,
Rick
Have you figured out what is happening here? I have the same deal. Following step7 regedits no matter what I do, when I exit CMP, it exits back to windows desktop.
Therefore I cannot get CMP to start in real cmp mode. I wonder if I misunderstood one of the registry deletes.
I can still back out step 7 piece by piece as I took exports before each phase.
That may help me pin point where it changed.
I'd just like to fix it and be ready for the next phase.
Edits: 09/20/11
Check your email now.
Hello, everybody. First, let me thank you for your interest. Next, some explanations. I have only one hdd and that is external sata in a esata-usb conbo box. I don't use this computer for anything but music, next, I strive only for the best sound even at the expence of functionality. Next, If your system sounds good, why reinstall drivers?
Next, I believe, I have posted all the warnings about network abilities and so on. If you are going with these tweaks even further, you will settle only with regedit, notepad, picture viewer and copy-paste and search function. AND NOTHING MORE, Folks. The sound will be absolutely devastating.
Serge.
This is the end state I seek. I use a PC dedicated to be a music transport into a USB DAC. I have applied every CMP optimization and would like to end with nothing running but the bare essential required to take bits from my hybrid SSD/HDD and get it jitter free into the decoder chip. My only short cut is I use a laptop with roughly the specs of the first CMP machine. I like the idea of a selfcontained keyboard, LCD and processor. Also like the fact that sith only the cost of a new HDD, a recycled notebook became a world class transport.
Sometimes I think we have it backwards even starting with XP.. It might make sense to start with a minimalist OS and add enough to run a player. Again thanks to CICS, this is essentially free, if you use a cast off notebook.
Right now I am finally taking an image backup so I can press on without the fear of losing it all.
Hi steppe
I think that your giant work is basically valid.
I've adopted it until the step 4 but I've noted a sound lighter and more open after this step.
Post of Jolida about utility of reinstall wdmaud.drv file, and SOUND too (for Juli@ users but it works well in my setting too, ASUS ESSENCE STX based) cured sound of my system but tell me also that not all cancellations must be useful in my setting. The true risk is of not understand who does what. I hate this!
Now I think it could be important we can know your Pc setting and, more important, your musical setting. It's too frustrating spend a lot of time for an upgrading and discover a downgrade instead! I'm puzzled, now, i'm seeing too files and function blown out withouth a clear explication on the function (on sound) of each one. Now I'll wait some feedback about sound quality, and some news about your settings too
Daniele
Serge, describe please the configuration of yours audio system: components of cMP2 (system board, CPU, sound card, RAM, PSU, system storage) if you use cMP2, whether the full optimization of cMP2 is made as in cics's website, DAC, amp (pre-amp), speakers.
It would be desirable to know level of audio system on which become and listened tweaks.
The cic's system was the highest level.
http://cmp2-mihaylov.narod.ru/
Hello, everybody. It is very good that our community started to be suddenly so active. It's what I was hoping for, but, folks, really, let me post all the tweaks (# 9 is ready), or we will produce a mix, that we won't be able to find the way out from. This is what I propose to do in part 6.
ALL FOLLOWING TWEAKS SHOULD BE DONE AFTER “MINLOGON INTERMISSION” and XPLITE tweaking.
See earlier posts.
First Go to the folder Documents and settings. Remove from everywhere Internet Explorer, Windows Media, Media player and or Windows media player. GO DEEPб GO TO THE VERY BOTTOM. Don't Miss a single instance. Do It In ALL subfolders. Pe patient, please, and thorough. This folder is very deep.
Registry tweaks. Go HKLM-System – Control Set 001 (all sets)-Control-Print (delete it)
Dangerous but necessary : HKLM-System-control set 001 (all sets) –control-device classes -2eb07ea0-7e.., open it and find ##?#Root#System/ open it find inside Acoustic echo canceller, wavetable synthesis and so on - 4 items, delete them.
Now go HKLM-System-Control Set 001 (all sets)-control-class-find E0CBF06C-CD8B (it will be Bluetooth radios) delete it, next delete 68dd1fc5-810f Infrared devices. Reboot. The system works! Now delete in the same place IEE1394, IEE1284.4 printers, IEE1394&SCSI printers, 1284.4 devices, Ports & LPT, Printers, SCSI & Raid controllers, Floppy (yes!), Smartcard readers, IEEE1394 Bus host controllers, Imaging devices, tape drives, non-plug and play drivers, IEE1394 IP Network Enumerator. The name of the thing you are deleting is in the right window, so be wary, but just read. Reboot.
Go HKEY-USERS-Default (all users)-AppEvents-Schemes (delete it), Delete also here Printers
Go HKEY-Users_default (all users) – Software-Microsoft and delete Messenger Service and Microsoft Agent
Delete FILES: confmsp.dll, tapi.dll, wfwnet.drv, eventtriggers.exe, net1.exe, net.exe, netsh.exe, ping6.exe, print.exe, proxycfg.exe, rtcshare.exe, stimon.exe, wiaacmgr.exe, wpnpinst.exe, mpeg2data.ax.
Reboot 3 times (rebooting, IMHO IS the process of system burn in as I begin to think)
Go HKLM –Software-Microsoft- delete smart card
HKEY_LOCAL_MACHINE\SOFTWARE\Classes\Media Type\{e436eb83-524f-11ce-9f53-0020af0ba770} delete Extensions.
HK-Users-default (all users) –AppEvents-Event Labels –delete all
HKey-Users-Default (all users) –Software—Intel –delete it (it's a codec)
HKLM-Syste,-Current Control set (all sets)- Services-DRMK and Beep – delete both
HK-USERS-Default (all users)-Software-Microsoft-windows-curr.version. delete telephony and themes
Remove or delete following files:
dxmrtp.dll, winstrm.dll, streamci.dll, mspclock, wiavideo.dll, mciavi32.dll, strmfilt.dll, msacm32.dll (at last), msacm.mll, msvcirt.dll, rasser.dll, sbe.dll, shmedia.dll, spnike.dll, sprio600.dll, sprio800.dll, tcpmon.dll, traffic.dll, vcdex.dll, vjoy.dll, wiausd.dll, viewchannels.exe (application, VERY IMPORTANT), wmploc.dll, mmsys.cpl, wiasf.ax, msrle32.dll, kmddsp.tsp, hidphone.tsp.
Remove folder x86_Microsoft.Windows.Networking.RtcRes_6595b64144ccf1df_5.2.2.3_en_16a24bc0
This is the end of part 6, and you may say that parts 5 + minlogon acrobatics + part 6 are The Rubikon.
Parts 7, 8 and 9 will seem dull and too easy. But all these tweaks make for the great sound that Cics was leading to.
Also you will see, how great is the math behind cplay.
Serge.
P.S. Folks! Somewhere in these battles I lost the capability to use flash card. If you know to what class it belongs, tell me. I don't suffer much, as I use external esata drive, but I don't want anyone else to suffer.
I am a bit late compared to the others here, but implemented everything up to part 6, where Serge you wrote: "viewchannels.exe (application, VERY IMPORTANT)" - however I do not find this .exe, not on my cmp drive and not on my normal PC. Only thing I find is something like an Explorer command called "viewfiles", but this is not an application!
Thanks to Serge and all the other posters who make this possible. I could only hear the first 2 steps on my chain because I had to send in my amp for repair today. But already those first steps sounded like nothing that had come out of my speakers before. Hard to believe...
Cheers
Bernie
Serge
Step six brings it all home. This is nothing short of amazing. You have given me a new system, and a whole new library of music to experience.
For those following, I found most of the files called for deletion were not there. I went through them all and maybe found one or two to delete. No worries. Not sure if they were never installed or if XPlite got them.
Several items I could not find or found in different places or found as a variation to the above:
..Under HKLM\software\classes\Mediatype( e436eb83...) Extensions was found in parellel to above key not within. Did others find it this way? Ok to delete?
..Under HKLM\system\Current control\Services ...I found DRMKAUD not DRMK. Is DRMKAUD ok to delete?
..Under HKLM-System-Control Set 001 (all sets)-control-class-find E0CBF06C-CD8B (it will be Bluetooth radios) delete it, next delete 68dd1fc5-810f Infrared devices. Reboot. The system works! Now delete in the same place IEE1394, IEE1284.4 printers, IEE1394&SCSI printers, 1284.4 devices, Ports & LPT, Printers, SCSI & Raid controllers, Floppy (yes!), Smartcard readers, IEEE1394 Bus host controllers, Imaging devices, tape drives, non-plug and play drivers, IEE1394 IP Network Enumerator.
I did not find any of this.
Yes kill extensions
Yes to DRMKAUD
In the last paragraph, look again. I found it all and deleted it. Can't recall the trick to it but you will find it.
Walter
Thanks plus I found other stuff too. Think there was a typo in the key.
Haven't started yet but looking forward to it.
I would like you to comment on Jolida's opinion about wdmaud.dll and JULI@.
I did restore it and I do believe he is right.
If we want to keep this file should we NOT delete the REGISTRY entry? I would think we would not.
I am sure we will all be asking you lots of questions. The post is very dense with information and some of us (me!) might find ourselves a little confused. I will know better when I get into it.
Certainly you have been immersed in this project the last few weeks. I am hoping, and I hope not too selfishly, that you might be more available to help the less capable of us to find our way to the end.
This is a magnificent project and you are quite a fine fellow to share it with us. I feel sure cics is quite proud of the interest you have shown in his endeavor. AND we would love to hear from the man one of these days!!!
THANKS very much,
Hello, everybody. This is an easy thing after tweaks part 5 but, unfortunately, it leads to an OBLIGATORY thing.
YOU SHOULD DO IT ONLY AFTER TWEAKS PART 5 !!!!!
OBLIGATORY THING is that to go on you have to PURCHASE or OBTAIN XPLITE (FULL VERSION!!!) a portable full version will do nicely. I HAVE ABSOLUTELY NO CONNECTION TO XPLite AUTHORS or VENDORS!!! But I’d like to thank inmates for showing the way.
1. Recollect everything you did to perform minlogon tweak. Find your former real winlogon.exe file. It must be now winlogon.exe.bak. It should be in windows-system32, probably. Check its size (appr. 450-490 kb). Find Your minlogon.reg file
2. Check that windows file protection is DISABLED!!!.
Rename your current winlogon.exe (former minlogon.exe, size appr. 17-23 kb) back to minlogon.exe.
3. Rename winlogon.exe.bak (the original winlogon.exe) to winlogon.exe.
4. Reboot (You should see a user logon dialog.)
5. Install xplite (full version) and run it.
Folks! I type it from memory, I am not at home and cannot get inside xplite, but what I write is easy, you won't get lost.
6. Inside xplite click on the left (windows accessories and optional components). XPlite will check and say you've got smth. like 134 of 169)
Check to uninstall EVERYTHING. Only Notepad, and windows CORE system fonts should remain. Be thorough, get inside every item and, rather, untick everything manually. XPlite is a polite program. It will ask you if you REALLY want to remove system restore, Internet explorer and other funny things. Say YES. After that say ok,or go, or next, or whatever. Xplite will think a little and start working showing you 2 gears turning. Let it work and BE PATIENT. It takes more time than to smoke a cigarette. After everything is completed, say ok, reboot.
7. Now open xplite once again and check that wfp is disabled, go back to windows-system32 and rename winlogon.exe to winlogon.exe.bak, rename minlogon.exe to winlogon.exe, double click on minlogon.reg, apply it to the registry and reboot. If All went well, You will enter a “minloggedon” state again.
In OTHER WORDS, You should revert from minlogon to winlogon perform xplite manipulation and back to minlogon!!!
If You attempt to do xplite tweak in minlogon state, you will get a blue screen crash!
Reboot, listen to some music.
Serge.
P.S. Tweaks part 6 and 7 are ready, but I do not post them now, just to give you some time to assimilate with the new sound and because there is a “hydra” secret in registry that I haven’t solved yet. I personally recreated all cmp tweaks from windows installation to the end of part 6 in 18 hours, but 2 days. And also, folks, if you are reluctant to do all the tweaks with registry yourself, wait till part 6 and just ask somebody to help, as I myself did the first time. It is worth it.
Hi, I applied the Minlogon/ XPlite tweaks and while Windows and cPlay seem to work flawlessly after the changes, J. River Media Jukebox does not run anymore.
This is a pity, because I just like to listen to lots of different files, also ape and mp3 - and Media Jukebox sounds quite good and musical, and is my first choice for overviewing the music library and for casual listening. I also find cplay - as great as it sounds - a bit frustrating to operate, not very versatile, a universal program like Jukebox is just so much easier to use... But on opening it needs to read a index.html file, but this does not work any longer after the XPLite tweaks (no program associated with html files.
So, is there another recommended Music Playback Program as second Music player next to cplay? Otherwise I may just go back to Step 5 and try to change only the things that leave Media Jukebox functional...
Thank you
Bernie
Hi, I applied the Minlogon/ XPlite tweaks and while Windows and cPlay seem to work flawlessly after the changes, J. River Media Jukebox does not run anymore.
This is a pity, because I just like to listen to lots of different files, also ape and mp3 - and Media Jukebox sounds quite good and musical, and is my first choice for overviewing the music library and for casual listening. I also find cplay - as great as it sounds - a bit frustrating to operate, not very versatile, a universal program like Jukebox is just so much easier to use... But on opening it needs to read a index.html file, but this does not work any longer after the XPLite tweaks (no program associated with html files.
So, is there another recommended Music Playback Program as second Music player next to cplay? Otherwise I may just go back to Step 5 and try to change only the things that leave Media Jukebox functional...
Thank you
Bernie
.. simply return the file "nddeapi.dll" which you may removed in step 5 (if indeed correctly understood that Steppe's file "nddeap" refers to the file "nddeapi.dll" and removed it and as I did at first ;)).
Serge.
http://cmp2-mihaylov.narod.ru/
Hi Guys,
I need some help, I tryed to revert minlogon after step 5 but after reboot I got blue screen ( fatal error ) , any ideas ?
Thanks.
Alexandre
Hi Alexandre!
Try to return the file "nddeapi.dll" which you may removed in step 5 (if indeed correctly understood that Steppe's file "nddeap" refers to the file "nddeapi.dll" and removed it and as I did at first ;)).
Serge.
http://cmp2-mihaylov.narod.ru/
Hi Mihaylov,
Thanks a lot, it works finally.
There are two things I can think of for you to try before starting over:
1. Have you tried to start in safe mode? I don't know if you can, but...
2. If you can connect your cMP drive to another computer or start from another system partition
Either way: then you can change min- and win- back so you go back to the regular winlogon and try to figure out and fix what went wrong
I hope you copied an image of your Operating System prior to the change. If you did reload it. Otherwise you have a rebuild in front of you. The usual error is not unhiding the file extensions for minlogon and winlogon.
Hi Theob,
In fact I did, I'm using snapshot, but I'm stuck in part 5, to be honest it's fantastic, these guys are doing a great job, but I'm curious how could improve doing the other steps. I'm not sure if I get your comment about hiding extension, is it a trick ? because my folder is not hiding any extension, really sorry about my misunderstanding english is not my natural language.
Regards,
Alexandre.
If you are showing all extensions then just make sure you have no typo's when doing the minlogon/winlogon swap. Good thing you have snapshot. :).
Hi Theob,
I couldn´t agree more.
;)
Finished the XPlite mod. Big step forward. In my system, things are getting really good. I imagine the noise floor took a drop. Still more space around instruments. Close miked female singers are very natural. The breath between notes and during notes very real. You can sense a human with emotions. Quite seductive. Strings and piano took on a new level of realism. Time to listen to the entire collection over again.
First on to 6
After installing this I think I am getting a glimpse of what will be realized after all of the deletions are made.
Spaciousness! An almost incredible increase in space around instruments and voices. On pop records it is amazing how long you can hear the artificial ambiance as it decays, one can hear the separation of the voice (in the recordings I was listening to) and the effect. I am having trouble describing this artifact of multichannel recording but I use it as an illustration of the resolution.
Started doing some of the NUMBER SIX registry deletions and had some mysteries which I need to re-check before "complaining".
I am seeing the banks of the Rubicon with this latest change. I can't wait to reach the other side. (Do not fear, fellow Romans, I have not brought an army)
Thanks, Serge. This is getting really REALLY good!
Don't know what happened but I reinstalled winlogon, renamed minlogon rebooted and pc booted/wanted to install juli@ but couldn't (didn't like drivers I had) wanted to reinstall usb drivers but couldn't find them. So I am stopped at steppe 5.
Hello, everybody. In this period I have rebuilt my CMP 3 times, 2 times deliberately leading to a crash and each time applying my own tweaks in a slightly different sequence just to see the "right" way. This is what I propose to do, and it is safe.
Remove to your temp folder, or whatever folder you have chosen, these folders from Windows or Program files or from elswhere: Help (windows help, surely), Internet Explorer, Macromedia, Messenger, Movie Maker, msagent, Netmeetings, Online Services, Outlook Express, Speech, Speech Engine, Video Lan, Windows Media, Windows Media Player and reboot two times. (this is in case you haven't done it yet).
Here goes the tweaks part 5 itself
1. Remove the files: iprtrmgr.dll, rasctrs.dll, mprddm.dll, rasrad.dll, prtmgr.dll, mprddm.dll. Reboot.
2. Registry tweaks
a. Go to regedit, open your registry: H.Key-Classes-Root. Delete wmp.audiocd, wmptheme, wmpskinmngr.1, avifile, audiocd, audiovbscript, audiovbscript.1, wav, wma, wmv, wax, everythimg mp, mpg, mpeg2,3,4, flac and so on (all audio-video multimedia suffixes you know), delete Internet Explorer application and its double - .1, Windows media.
b. Go to H.KEY- Current User-Software-Policies- delete Netscape (repeat with all users)
c. Go to HKLM/System/Control Set001/Services (the same with control set 002 and 003) and delete sysaudio, splitter, cdaudio, swmidi, redbook.
d. Go to HKLM-Software-Active Setup-Installed Components. (Don't be scared) The 4th (cdd79...) from below should be Windows Movie Maker (check in the field to the right) Delete it. The 3rd from above (22d6...) - media player. Delete it.
e. Go to HKLM-Software-Microsoft-delete audiocompr.manager, Internet explorer, IE4 and IE setup
f. Go to Hkey-Users- (you may start with default user but do it within every user listed) s1-5-20 - software-Internet explorer delete it.
g. go to HKLM-System- (all control sets)- services-and delete MSPQM, MSKSSRV, MSPCLOCK
h. HKLM-System-Control set (all sets) ENUM-ROOT-MEDIA delete media
i. HKLM-System-Control set (all sets) -control-Media properties and delete inside everything that doesn't bear the name of your soundcard. The same about Media resources
j. HKLM-System-control set (all sets) -Services and delete: aec, Alerter, asyncMac, audstub, wdmaud, AFD, Browser, DHCP, Dnscache, HTTP, IP6FW (and its .sys file from drivers folder), IPFilterDriver and (ipfltdrv.sys file from drivers folder. All other services with corresponding files will be enumerated below in paragraph j. as files. Remove or delete them from registry AND from folders), IpInIp.sys, kmixer, kssrv, ipnat.sys, ipsec.sys, swmidi, modem, mnmsrvc.exe, splitter, netdde, netddedsdm, netlogon, netman, nla, nic1394.sys, nwlnkflt.sys (+nwlnkipx.sys, nwlnknb.sys, nwlnkspx.sys, nwrdr sys), nwlnkfwd,
perfnet.dll, policy agent, RasAcd, RasAuto, RasMan, rdbss.sys, rdpdr.sys, rdpnp (+drprov.dll), raspppoe.sys, rdsessmgr (+sessmgr.exe), Remote Access, Rpclocator, Shared Access, Spooler (spoolsv.exe !VERY IMPORTANT), Srv.sys, stisvc, ssddpsrv (don't know if it has to do with ssd disks, so be wary), smlogsvc.exe, sysmonlog (+smlogsvc.exe), tapisrv, tcpip.sys (+ tcpip6.sys,), TermDD.sys, termservice (+tcmsetup.exe), Themes, TlntSvr.exe (+telnet.exe, tlntadmin.exe), W32time, wanarp.sys, wdmaud, wmdmPmSn, wuaserv, WZCSVC, xmlprov.
REMOVE THE CORRESPONDING FILES!!!, Perform this castration on all control sets!!!
k.Go to HKLM-System- Control set (all sets)-Control-Class (find 6th from above) {4D36E965-E325-11CE-BFC1-08002BE10318} open, find cdrom, remove digital audio, NEXT IS VERY IMPORTANT
HKEY_LOCAL_MACHINE\SYSTEM\ControlSet (all sets) \Control\Class\{4D36E96C-E325-11CE-BFC1-08002BE10318} open and delete everything (you may probably retain drm and kx audio driver 0005, but I deleted everything, though drm and kx driver I deleted later)
As You Delete these services from control set 001 it wil be more or less automatically deleted from current control set, but CHECK ALL SETS. When you will see screaming dialogues, don't be scared.
This is a list of files to be deleted mostly from system32 and system32\drivers
exe files: locator, mnmsrvc, netdde, scardsrv, sessmgr, smlogsvc, spoolsv, tcmsetup, tcpsvcs, telnet, tlntadmin, tlntsvr.
dll files: drprov, iprtrmgr, mprddm, ncxpnt, nddeap, nddenb32, netapi (be careful), netcfgx, netevent, neth, netid, netman, netrap, nlhtml, nmevtmsg, nwwks perfnet rasctrs rasrad
sys files: afd, http, ip6fw, ipfltdrv, ipinip, ipnat, ipsec, modem, rio1394, nwlnkflt, nwlnkfwd, nwlnkipx, nwlnknb, nwlnkspx, nwrdr, raspppoe, rdbss, rdpdr, srv, tcpip6, tcpip, termdd and a file wave.inf (search for it)
NOW REBOOT. Listen to some music! (Let your cmp burn in for an hour)
Serge.
P.S. Please, try to do it with a system backup or smth... just in case. I don't expect ANY troubles.
P.P.S. This is the hardest part of the tweaks. Starting from part 7 and up to 9 it will be piece of cake.
there were a few things I could not find.
Let all of the wmdaud stuff in.
Does seem to offer a clearer view into the music but not enough time before or after to make a real assessment.
One thing for sure it did not go backwards.
Truly a pain to do - probably took almost as much time as the standard optimizations.
Downloaded a copy of XPLite and will make those changes tomorrow.
Ditto here.
On my system it was a small but worthwhile improvement. It is all cummulative and heading in the same direction. More air and separation around instruments. In this case the bass sprung to life with more texture and definition. I look forward to next steps but first I have to figure out how to take an image of my system, as I don't want to go through this again.
Thanks Serge
Finally got thru Steppe 5 too (minlogon/winlogon/xplite....). I agree bass comes back.Another observation regarding post xplite. Prior level sounded very deep as if soundstage went way back but you were still outside the soundstage. Like there was a narrow door that was open to the soundstage but you were still outside. Now (post xplite) it sounds as if you entered the narrow door and the soundstage is deep but you are within it and all is very clean/clear not bright just clear.
Edits: 09/17/11
Does Control Panel still work for you?
*
Usually when I boot up I go to Control Panel and stop Cryptographic service and I uninstall Numeric Data Prcessor and unused Usb ports to get them off the IRQ. Also when I install a new hdd I enable one of the other Stopped Services (forget which one) to do routine hw management. I have come up with a workaround for Crypto and have discovered how to access control panel from the Command Line but it still does not allow me to get to device mgr. I do not notice any ill effects from inability to get into Control Panel (because it sounds excellent!!) and I could live with it but would rather not.Also I do not know how to change sound cards or load new drivers without control panel.
Edits: 09/15/11
Theob & others...
If you can't get to the control panel anymore, here's how to get to some of the functions directly...
First, set cMP to open the Task Manager at Startup. In the Task Manager, select File, Run, then Browse to one of the following files in windows\system32\ for some useful Control Panel options:
Services.msc - Services
appwiz.cpl - Add/Remove Programs
desk.cpl - Desktop
sysdm.cpl - System Properties (for the Hardware Manager)
main.cpl - Mouse
I've only gotten through Steppe 4 & Jolida 1 & 2 so far, but these still work for me.
Greg in Mississippi
Everything matters!
Great cant wait to try!Unfortunately did not work...I get same result. I can access what looks like device mgr for example but when I click on it I get an error message.
Edits: 09/16/11
*
Look I think what you have done here is brilliant not because you are a computer expert but because you are not. I don't know how I would ever accomplish what you have other than thru brute trial/error. If that is how you did it I salute you even more because that is a whole lot of work. And to the extent that you are sharing it here I thank you. Let me be clear I am not going back from Steppe 5 it is that good...pure highs, depth, transients and dynamics to die for. If I have to reload a prior image of windows to load new hw I will do it. And if I can figure out why reloading Winlogon did not work for me I will be implementing Steppe 6, 7...
I am just looking for a way to avoid reloading windows.
I do not recommend to owners of network configurations of cMP2 do this tweak. Only lose time and will restore the system. Too many of the removed files and registry entries are related to network capabilities of the system and the Steppe not even warn about it!
http://cmp2-mihaylov.narod.ru/
these mods have been offered as something for the purist.
Network capabilities, from what I have read in the FORUM, are mainly for the sake of convenience and CERTAINLY no one would do this if that was of importance to them.
Only for those wanting to take their computer transport to the highest level. As with all gear that I am aware of one gets to choose between sound and convenience with each option taking a toll against the other one.
I am not interested in any convenience and only want the best sound I can get out of our humble approach.
Thanks to Serge for his efforts.
My apologies for not paying better attention.
I do wonder, though, if with the removal of the networking capability could there be such an improvement, without using a network, that one could come out ahead?
with the removal of the networking capability could there be such an improvement
It needs to be tested but my guess would be that a good network-connected but headless setup would come out on top.
The orthodox setup includes a keyboard, a mouse (possibly a wireless one) and a monitor (perhaps with a touchscreen), all of them noisier than school children at a boy-band concert. Many builders use separate switching power supplies for the CPU, for the motherboard and for disk drives and adopt a bewildering variety of measures to tame them. Data are stored on locally-connected (notoriously noisy) spinning disks. The whole lot is typically squeezed together in a case beneath a rat's nest of cables. What a mess. (If you don't believe me, look at almost all pictures of inmates' systems - they're a veritable plague of "leeches on a mermaid's ass".
Compared to that lot, the likes of a Fit-PC2 (or indeed almost any mini-ITX mobo where the total power draw is about the same as most soundcards on their own) with a good linear PSU sited remotely, an SSD for the OS (also with a remote linear PSU) and a LAN to provide control and data seems to me as simple as it gets.
Hmmmmmm. A Keyboard, a mouse, a monitor, switching PSUs, at least one spinning disk, a soundcard with enough PSU add-ons to rival a small aircraft - or a mostly-inactive networking stack. Tough call.
Mihaylov: I do not recommend to owners of network configurations of cMP2 do this tweak
I've done most of the steppe 5 deletions and am reasonably convinced that there's an improvement to be had but I have not tried deleting anything that might be to do with networking and probably won't.
"Сonvenience" and "best sound" not is mandatory contradict each other.
Agree completely.
Dave
I haven't had time to attempt step 5 but am also daunted re impairing network function. Could you say which items you do not change or delete in order to preserve this?
There you go again smoking that cigarette (in a holder of course) talking with a David Niven accent, :)
Hi Dave!
Please describe your version of step 5 for the network configuration of cMP2 a la Fit-PC2.
Thanks.
Serge.
http://cmp2-mihaylov.narod.ru/
Edits: 09/15/11 09/15/11
Please describe your version of step 5 for the network configuration of cMP2 a la Fit-PC2.
Sorry for the delay - I decided to check that I'd done everything on the lists before answering your question. (I hadn't but I have now.) I don't have a "version" as such though I do have a procedure. In case it's helpful, here it is.
1. I have already worked at length with Bold-Fortune's XP Slimming Guide (see link). Much of what steppe recommends I have, for different reasons, already done and, in many cases, have gone further. (Details later!) That said, his suggestions have a marked effect on SQ and are well worth doing.
2. I would advise anyone against making changes like this unless they have an imagefile backup utility such as Snapshot or Acronis. It's a good idea also to have a docking station on your desktop box as it makes for quick recovery. Non-recoverable errors are pretty much inevitable.
I would also urge inmates NOT to use XPLite for anything other than disabling WFP. See link for a good explanation of why XPLite is best avoided.
But you know all this.
3. Copy the AA post into a text file and format it so it's easy to read (e.g. one step per line). See what steps are likely to involve networking. Clues include, obviously, the likes of files with net, tcp and ip in the name.
4. Perform the various steps in cMP Mode - it makes testing much quicker. Create shortcuts to regedit.exe and cmd.exe and place them somewhere easy to find. Launch Explorer from cMP's GUI and navigate to the shortcuts.
5. Work your way through the lists line by line. Skip any steps that you suspect may affect networking.
As you're accessing the cMP^2 box via VNC on an Internet-enabled desktop, it's easy look up any files you're not certain about. Much of the info is provided by companies trying to sell "security" software and isn't much use but you can usually get a rough idea sooner or later.
If the named file is present, decide whether to keep or delete it; make a note in your text file of what you delete. XP denied me access to some files - I didn't try to over-ride it.
If you disabled as many devices as possible prior to or during the XP install, many of the listed files may well not be present. I could only find about a quarter of them.
6. Test the system every few steps. This is easy if you stay in cMP Mode - a quick reboot is all you need.
7. Make regular backups of the OS partition - much better than the oft-recommended procedure of copying files you delete in the hope you can restore them in the event of fatal error. You'll forget where they came from and, in any case, copying doesn't protect you against Regedit errors.
The more often you make backups, the less time it takes to pin down the source of a problem.
8. I suppose, what with this and that, steppes 5 and 6 took me about two hours. I had no fatal errors (though I've had plenty before).
HTH
Dave
So Dave is not going thru Xplite (or an equivalent) a pre-requisite to Steppe 6 & beyond? That's how I read Serge. Did you do something similar to Xplite? Was not Autoruns something similar? When you went through Slimming XP did you use his 'routine' to do the changes or did you just plow through all the individual changes?
I am not anxious to buy a program for $40 to do one task and maybe never use it again.
I normally adhere to that procedure which you described. But it demands a lot of time. I thought that may be you have the LIST of changes in the registry and file system which it is necessary to make for the step 5 on full optimized cMP2 with network. But all the same thanks!
Serge.
http://cmp2-mihaylov.narod.ru/
Edits: 09/16/11
"Network capabilities, from what I have read in the FORUM, are mainly for the sake of convenience." -You are mistaken.
"I am not interested in any convenience and only want the best sound I can get out of our humble approach." -
"Сonvenience" and "best sound" not is mandatory contradict each other.
http://cmp2-mihaylov.narod.ru/
Edits: 09/15/11
I suppose the usual caveat applies to the tweakers.
But one can always try this freeware for registries backup.
http://erunt.en.softonic.com/images
Just a few clicks of the mouse to save and restore those deleted registries.
Provided, always, the mouse is still alive after the tweaks.
"I suppose the usual caveat applies to the tweakers." -
Similar claims to cics could be never!
For example look here http://www.cicsmemoryplayer.com/index.php?n=CMP.08AutoRuns (steps 4 and 5) and here http://www.cicsmemoryplayer.com/index.php?n=CMP.07Optimisations (step 6).
http://cmp2-mihaylov.narod.ru/
since I use SSD for my opertaing system I wanted to check on this.
Fist thing O fond is that there is an extra letter in Serge's report.
The next thing is, from what I can see, it has nothing to do with SSD. All I read talks about USB and plug and play ...
I wish you would have made it clearer on what is what with the registry and .dll/.ysy.etc paired deletions. Do all of them have correspondents?
If not it would be good for the squeamish to know which has which.
Maybe there is no confusion when one gets going. I am looking at my office computer to get a feel for what's what. It just seems there could be some confusions with .exe and .dll files if both exist. Do we delete them both if this is the case?
Thanks and hope I do not sound ungrateful - just a bit nervous. I know you recommend an image of WINDOWS but the programs I have looked at seem like they are always ON which seems to me to be a problem in themselves. Is this not the case and if so, whose software are you using for this?
Thanks, again,
Hello, thank you for the interest. Dlls and exe programs are different. It often happens that svchost is starting with a -key that starts some service and THREADS to other services, represented by other dlls. also there is an utility thst is called run dll as a program. I want to find and EXTERMINATE IT. If I suggest to remove (read delete) a dll it doesn't mean to delete an exe or sys file with the same name. it may come later. I am not a computer specialist, so I may be all wrong. but what I suggest is derived from practice and only after I make sure that the system is stable. Take for e.g. file winspool.exe. It has impact on sq, but though cplay would work without it, cmp wouldn't. So I sort the files, I propose to delete. also, if you have a sp3, PLEASE, make a differences list, that other folks will know that there are some more files to delete.
Serge
My suggestion was to make it EVEN cleared for us who are much less savvy that you just which suffix to remove.
As if you have not given us enough, I know, it does sound ungrateful, but I figure if you did not want this to be used by others you would not ahve gone to all of the trouble to write it down. Please accept my criticism in that light.
Thanks, again,
Hello. Criticism accepted. From tweaks part 7 I try to be absolutely precise with filenames. Also, if network defenders will read this, real orthodox setup was no network and no keyboard and even no mouse (why then write a program called cicsremote and use a touchscreen.) I don't understand why we need network. To grab music from torrents? I don't do it.
Serge
"I don't understand why we need network." -
1. To provide the FULL GALVANIC ISOLATION of cMP2 from hdd.(I explain that it is important first of all for a SQ).
- It is irrelevant certainly only if you use ssd and do not use hdd (internal, usb hdd, esata hdd, etc.) generally. Big ssd (more 500 GB) too expensive for me.
- And certainly it is critical if you made the full optimization of cMP2 with the full linear psu and your audio system is good enough.
2. Usage of network capabilities allows to delete the mouse, the keyboard, the built-in monitor, to disconnect all usb ports as already correctly noted Dave.(I explain that it is important first of all for a SQ).
Serge, I recommend to you to study the content of website CICS attentively.
Serge.
http://cmp2-mihaylov.narod.ru/
Looked at the article, which I of course remember, but dismissed it at the time because I am not using an USB dac and likely never will.
I am confused though about the elimination of the monitor, mouse, and keyboard. I assume these are "turned off" within the machine and not simply left unplugged?
It would be interesting to be able to compare both approaches. Never will happen at my house, sorry to say (no help here with the answer).
One wonders, though, with all of the assumptions we have had to temper about digital audio if a network cable really does bring about the galvanic isolation that is desired? Assuming a fully functioning computer as your SEND unit it is not beyond conception that the pollution from that machine could ride the signal you are sending to the minimal machine.
I continue to think both approaches are valid and I suspect very good results can be achieved with both. Not that they will do all of the same things well but that each will have its own strengths.
Could be a form of personality test as to which one one pursues? I wonder what is says about those of us willing to squint for hours at a monitor deleting files?
All I know is that something quite good became apparent with the cumulative effect of the previous deletions and XPLite. Yes, it is a little frustrating having a computer that does not do all of the things it did before (and that was not much!) but as with the graduation from a changer to a proper turntable, these things must be done.
Bye.
real orthodox setup was no network and no keyboard and even no mouse (why then write a program called cicsremote and use a touchscreen.)
Because some users find a touchscreen convenient. I don't - nor was I ever convinced that the sonic crud it must cause made it worthwhile. For that and other reasons, I eventually rejected the "orthodox" setup and looked for a different solution. Others have done much the same.
I don't understand why we need network. To grab music from torrents? I don't do it.
For the benefits of networked (headless) setups, see here:
http://www.audioasylum.com/forums/pcaudio/messages/9/95222.html
and here (though it's a bit dated):
http://www.cicsmemoryplayer.com/index.php?n=CMP.Headless
Nothing to do with downloads! Other reasons? Monitors are ugly, greedy for space and, esp with a touchscreen, expensive to buy and run.
++++
Theob sees me as a smoothie, David Niven smart ass.
So how do I picture him? Hmmmmmmm Thinks . . . How about as one of those hicks a'jumpin' an a'cursin fit to bust - while a'blastin' away at yet another Goddamn motherboard that's just gone up in smoke??
Only jokin', Pardner.
Ok got through Steppe 5 and it sounds good ...a little light (Jolida might not like) but I cant get Services from control panel (to turn off Crypto) and I cant get into Device Mananger to get some stuff off the irq etc. But how would I change drivers for juli@ or load a new sound card or add a new hard drive? In my case I could reload an earlier image of windows but this might be a major issue for many.The sonics (if they get no better than post Steppe 5) are not enough to justify giving up the above functionality (device mgr, ability to stop Crypto, add new disks etc) from within minlogon. In other words I will try Steppe 6 and 7 but I am skeptical now.
Edits: 09/14/11
I went thru Steppe 5 and I missed some stuff which I picked up the 2nd time thru. For Jolida I also listened to Steppe 5 with/without Sound.drv. I like it way better with Sound.drv deleted. It sounds really good. I am still concerned about Crypto and getting into Control Panel to stop it and also to eliminate stuff from IRQ (like Numeric processor and unused usb hubs). I am currently doing Steppe 6 (I Snapshot my image as a backup) but am doing it slowly (I hate minlogon).
If I had to live with Steppe 5 I would. I prefer it now to pre Steppe 5. If anyone knows how to do the above stuff without going thru Control Panel pls let me know.
Theob,
I haven't implemented steppe 5 or 6 yet. Will probably try it soon. And with regards to my liking or disliking of the music, which may get, as u say 'lighter', is truly subjective. Its not a preference of a particular flavour of sound that I have. What I look for, is a Balance in the entire frequency range. As mentioned earlier, if the Highs sound open or airy after a deletion, but at the cost of the lower range going down a notch, then it's a compromise, or as I put it, 'a downside'. It's something like u pulling Ur speakers a few inches more off the Walls, than they are at the moment, which would somewhat result in the same kind of a tonal change. An improvement, on the other hand, is clearly audible, only with familiar recordings which we are well versed with. Making the presentation Lighter, than it should actually be, is like an equalisation, which is not a very good thing in High-end audio.
As I have mentioned already, I haven't tried steppe 5 or 6. I shall post my opinion when I'm done with them soon :-)
As for the sound.drv file, I still prefer it in. Must be an Acoustical difference between Ur room & mine, which likes either. My room measures almost flat using REW software, using 16 ASC tubetraps, 8 Panels by Realtraps & a couple of RPG skyline diffusors on the front wall & ceiling. Did u like the file in more after steppe 5 ??
I meant no disrespect or even a value judgement on your taste or your system. I was just giving my subjective impressions to hopefully help others (especially you) who otherwise might wonder what Steppe 5 might do for their systems. I do prefer Sound.drv out post Steppe 5.
I am however perplexed as to why Winlogon acts like I have missing drivers for juli@ and usb ports and wont use drivers I quite clearly have in in my Windows System32folder. When I reloaded my image file (which includes minlogon enabled) my system works fine.
I am also concerned that post Steppe 5 I cannot use Control Panel.
Did you ever get 1g ram to work?
No I could not. Last night was even worse. Since my attempt of changing to a 1gb ram, the Juli@ console is acting pretty funny. When I start the computer, the "esi" icon loads in the system tray, but when I take the mouse cursor on it, it disappears. The Juli@ is active in the device manager, shows up in the cMp as well, but does not output any signal. Now when I revert back to the original 2gb ram, it behaves the same, even if I restore a working Image back-up. Will try again this evening, else I'm afraid I will have to start from scratch all-over again :-(
Let me know.
Thanks,
Its german program called snapshot. Slight cost something like $40 or so.
Edits: 09/14/11
Something I should do.
Thanks, again,
Hey Guys,
It would be great to know if anyone has any experience with USB/FireWire DACs in regards to the DLL & Driver modifications.
Which are needed in such setups ?
kind regards
cMP2 Computer: i3-540 Core cPlay039 >FIR filter w/ convolver> Lynx Aurora 8 FireWire /192kHz throughout. 2x AcousticReality Ref. 202 & 2x AcousticReality Ref. 601´s ICEpower. Magnepan MG3.3R beechwood frames & custom stands. Miller chokes
any experience with USB/FireWire DACs in regards to the DLL & Driver modifications?These things are so device specific that you really need to experiment for yourself. I and another inmate who both use the AQVOX USB driver independently and successfully deleted:
usb8023.sys, usbintel.sys, usbcamd2.sys, usbcamd.sys, usbccgp.sys, usbintel.sys, ubstor.sys, usbehci.sys and kmixer.sys.
The overall effect in both cases was a slight "refinement" of the sound - AFAIK we've both left the "tweak" in place.
I couldn't delete usbuhci.sys and usbhub.sys (failed to boot) and didn't try the few others that seemed USB-related. I didn't hear any change on deleting the kmixer driver and can't say which of the rest actually changed the sound (as opposed to doing nothing).
I've had no disastrous side effects but I don't use USB ports for any other purpose than driving a DAC on the two machines I tried it on. As ever, YMMV.
Edits: 09/13/11
No one has said much about it for awhile.
Considering getting the GIGABYTE GA-H61M-S2H LGA 1155 Intel H61 and the above processor would like to hear of others recent experience.
Thanks,
Hi Rick
I've build my last release of my former cMP2 with a combo of i3 2100T+ GA H67MA UD2H-B3.
This is a fantastic and stable foundation for our system, clearly better than my previous H55+i3 530 combo.
Power consuption of new set is above 30% less than the previous one, and quality of regulators on board is also better .
But.....now my H67 is not more available, after very few months, and his substitute is the Z68MX UD2H-B3, maybe better in the BIOS setting but I'vn't any news about this in a cMP2.
The great advantage of a H67 ( and of Z68 more) is his compliance with next generation of CPU, the Ivy Bridge series, a new 22nm technology, that could spend fewer current of 2100T.
But H67+ 2100T is very effective and stable, great sound!
The Z68 seems like a mixed blessing for cMP.
Please tell me why you think it would be good for cMP machines.
Thanks for your help,
Hi Rick
Z68 could be better (could be, maybe)formerly for three reasons, IMHO:
- it could mounts new 22nm Ivy Bridge. They can consume less than i3 2100T, lowering currents circulating on the entire mobo
- the regulators on mobo are always switching ( like everyones!) but are really better builded, a lot than the previous models of mobo ( are the same of H67). In my H67 mobo ( now discontinued, a true pity) current for picoPSU drops from 1,7A of H55 to 1,2A of H67.
- Z68 get a BIOS totally manageable, more than my H67. This could be an advantage to lowering tensions and downclocking CPU.
But someone must try it, I can attest that H67+ i3 2100T is largely better than my previous system but I've scheduled a new mobo CPU combo to april-may of next year, when I'll can buy new Ivy Bridge CPU, if review will demonstrate real improvement.
Ciao
Daniele
Whenever I read all of the "it is faster" stuff I start to worry which was my main concern for our application.
I will hope for a flexible BIOS and let you know what I find.
Looking at the manual does not show anything different from their typical BIOS but the manuals never show what happens with control/F1 which is what really matters.
THANKS for your help!
I'm so curious about it.....
But I want wait for new setup when will be available new CPU too.
I think you'llhave some surprise using new setting, it's largely better than H55/i3 530 combo in my experience.
I was doing some digging, trying to understand how deleting DLLs can have such a profound impact on SQ, and came across a registry change that forces DLLs to be unloaded. I added it to my CMP XP SP3 machine. I don't know if it is my imagination, but I think its quite positive. Give it a try and see if you concur. Here is the tweak:
When programs run in Windows XP, they frequently use (Dynamic-Link Libraries), which contain shared programming instructions that different applications use in order to run. DLLs are stored in RAM whenever the system summons them; when you quit a program, XP is supposed to release the DLL from memory. But sometimes DLLs can get stuck and do not get released.
You can force Windows XP to release DLLs by adding a Registry key. Run the Registry Editor and go to My Computer-> HKEY_LOCAL_MACHINE-> SOFTWARE-> Microsoft-> Windows-> CurrentVersion-> Explorer. Create a new subkey called AlwaysUnloadDLL and set the default value to 1. Restart Windows to activate the change or reload registry this way.
I set the default value to 1.
here is the reference http://www.dq.winsila.com/windows/windows-xp-tips/speed-up-your-pc-force-windows-xp-to-release-dlls.html
I have now toggled it on "1" and off "0". There is no question on my system that this makes a very noticable change. Sound stage came forward and a little more 3d. Main change is more air and less congestion.
I was out of town for a few days. Listened briefly to 3 well known cuts. Booted back to explorer, into regedit made the change and rebooted to cmp and listend to same 3 cuts. Immediate and not subtle degradation with setting at "0".
I am thinking the difference is I am running USB out to a DAC. Perhaps this has more impact then with the more conventional @Julie configuration.
I have had no instability issues and I am leaving this one in.
I think it is especially important since I cannot implement AWE.
Thanks for your research and suggestion.
One wonders if this will be less useful once we have discovered all of the .dll files that can be eliminated?
For the time being one would have to think this is a boon.
Theob has accurately pointed out that there is lots of debate on this one. Many posts say that it was used for pre win2000 systems and is no longer used by XP. It is applicable to Explorer so logically it may do nothing for CMP. Others say it is intended only for programmers to help them in debugging, implying it is still there but not publicized. Could be my imagination, but I believe it had an impact, so I'd be interested in your listening impressions. It is easy to do and easy to reverse. Take a registry export to be extra safe.
No troubles booting up.
There have been so many improvements realized lately with the .dll and driver deletions (along with files ...) that I hesitate to say that I heard something.
If I do it is subtle (duh).
I think this is part of the cumulative gains made with this new initiative. It is likely an important building block. I still wonder, depending on how many of these .dll files we find we can delete if this will become less important? Going in circles, I know.
My machine is minimalist so I can only speak for machines in similar states of tune.
Adding AlwaysUnloadDLL set to 1 somehow caused cPlay to freeze if left unused for awhile. Deleting the key and also reinstalling cPlay did not fix this so I had to reinstall the whole system from backup.
The problem appeared in the reinstalled system so it must have been latent...about to happen even aside from this key being added. ksuser.dll is needed for Weiss driver installation and, it seems, proper functioning sooner or later.
I was going to write the same thing as Rick that any change is subtle but seemingly no harm. But now I have noticed that if I leave cPlay running but idling for an extended period, when I go to close it or try to play a track it freezes and a hard system restart is needed. This has happened two or three times now but never before. Sometimes in the past if left it idle I would have to reload cPlay to reset the asio driver but never a freeze. I am going to remove the AlwaysUploadDLL key. Please note I use a Lenovo laptop not the full cMP2 build, and I use a Weiss dac not the julia card.
is it Weiss DAC Asio compatible ?
Of course. Weiss has his own asio driver for it, and I have been using it for years now.
Fantastic!!! Just curious , are you using Firewire conection ?
Yes, firewire, which the Weiss DAC202 sounds best with, as did the DAC2 before it.
Thanks for the feed back. I'm out on business for a week. I'll try removing and reinstalling when I return to see if it is the same on mine. I have all the CMP mods, but it is a laptop as opposed to the target configuration.
so have not encountered an idling problem.
We'll see what else might develop.
Little confused here. Your link does not work on my pc. Also googling AlwaysUnloadDLL indicates in many cases: does not work or creates other problems.
Interesting.. I really just tried it and it seemed like a positive step in micro detail and balance. However, at this point one could argue its just hearing what you want to hear. I had no expectations of success, so I was pleasantly surprised.
I figure, we want to get down to only what is required to run this one app and the sound card. Everything else is overhead that seems to degrade performance.
When you say it did not work, did it cause a problem, or just not produce any change at all? Should be easy enough to back out. Just delete the key.
No just the link you provided did not work. When I did my own search I came up with '...does not work...' and '...may cause boot problems...' as a result of the search. I did not try.
As we all probably know, The digital data is fairly modified in the audio-chain, right from its source, till it finally reaches our ear. This Audio chain, as far as the cMP is concerned, routes its way in the following order.
Flac=> Sata cable=> Motherboard=> Processor=> de-compresses to Wav=> Soundcard=> Digital cable=> Dac=> Interconnects=> Amplifier=> Speaker cables=> Speakers=> ROOM=> perceived by the ear !!!
No matter which brand of Signal cables or Power cords or Speakers being used, each have their own Colour !! Every piece of electronics in this Audio-chain induces this Colour on the music, as it passes through it. A cable, which sounds perfect in one system, may quite certainly sound dull or bright or bass-heavy in another system. Another MOST important aspect of this audio-chain, often neglected, is the Room itself. It Influences nearly 50% or more of what we hear.
Cutting the long-story short, we TUNE our systems to adapt to our listening tastes. Though the digital data may be 'Bit-Perfect', it still has to pass through these Peaks & Dips. Like a piece of Electronic, a tweak may as well sound as an Improvement, or as a down-side. Its entirely based upon this Equation, which i call 1+1-2=0.
An interconnect may boost certain frequencies ( add 1), a power cord or speaker cable or the relationship of the Speaker> Room> Listener may dip certain frequencies (minus 2). So at each step, we willingly or un-willingly tailor this line-up to arrive as closely as possible to a Balance (Zero).This is not only evident in low to mid-priced set ups. Even the most expensive set-ups i have auditioned have sounded nothing more than 'Just Ok' or 'Not Bad' to me.
Likewise, all these file deletions serve as tweaks that we are doing to an already set-up project by Cics, with an intention of improving it further if possible. But while implementing these, or for that matter any tweak, what we need to have is Focus. As i already mentioned in a previous posting, " A change need not always be an Improvement". If we have those so-called "Registrations" in our mind, as to how it sounded before making those changes, then there are very less chances of making mistakes. To accomplish that, we either need to audition recordings which we have heard hundreds of times, & are so well versed with them, that we can distinguish between that Improvement or Down-side in that Change thats taken place. I understand that its not easy. But if we could kinda stick to this policy as much as we can, then we probably will lead in the Right direction.. Only by trying to follow it as much as i could, i was able to spot out at the 'sound' & 'wdmaud.drv' files.
Just thought would share my experience with all. Sorry for the long post :-)
Have been listening for more than an hour after doing my 3rd set of deletions...
OK... I admit it.
I am hoping you post today, so I can have the fun of playing with it late tonight!
I think that was half the fun of being on this wonderful ride. Every now and again, CICS would surprise us with a new release. It was like receiving a new $15k CD in the mail from an anonymous benefactor.
I thought it was not possible to improve .39 when Serge came along and took it up a level. Now I hear of new OS tweaks and I scoff that it cannot get better.. Until it does.
Walter
Ok don't hold us in suspense. What are deletions?
:-) Will post it in a while. Have to write it down. One of the cleanest Bass qualities is apparent after these deletions. Very deep, as well as controlled... Still analysing critically. Will upload once I feel satisfied with all aspects !! Shortly..
Hi all
I've read with great interest the posts about juli@ drivers and now I can confirm that SOUND (C:\Windows\system) and wdmaud.drv (in C:\Windows\system32\drivers) must rest in place with Asus Xonar Essence too.
wdmaud.drv is an important driver for this soundcard, that works the same without it but produce the strange sound I've listened, incoherent and not balanced. When you put again this driver in his folder and reboot, your musical world changes and sound come back to a good balance, with the right midbass.
Adding SOUND.sys also, you can enhance this effect and you can listen to a solid, physical bass and midbass,i.e. piano gains depth like real,
But all music is enlighten by a transparency and a precision when you add the last Jolida' modding, now all is allright.
Simply wonderful.
I'm very happy now and I was thinking about the effect of single file on sound. Maybe, with a great job, we could identify the effect of single file removed on sound quality, to do a file (fine) tuning of our sound.
theob had mentioned something about being unsure of the propriety of tuning with various file deletions and even though I can understand his purist vision behind this comment; when one is removing veils can there be veils that actually enhance the transparency of a total system (for a lame try at a poetic explanation). But in the real world one has to take what they can get.
There is no perfect resistor or capacitor but the most dedicated hobbyists and designers search for the combination that allows the most pleasing sound.
When dealing with an operating system one (now we can see) has to look at these components as we have traditionally looked at resistors, caps, etc. in an electronic circuit.
Which makes me think of the tunings our revered cics made with the various iterations of cPLAY.
I would like to think this new found energy the file deletions have brought to this project might inspire cics to take another look at cPLAY. As good as it is and as much we are enjoying using it one has to wonder if at the end of this XP slimming project that it might bring opportunities for a re-tuning of cPLAY? Compromise being the name of the audio game and with less compromises imposed bu XP (my assumption) there could be good reason to look at a new set of compromises within cPLAY.
I hope cics is paying attention to this because it is bringing us closer to realizing the art of the computer transport.
I second that.
With the latest edition of file removal (and re-installation), cPlay's performance is on another level. I, for one, is very interested and would appreciate what cics-who had single handedly started us on this wonderful journey-would think about this latest development of his creation.
Hi All
With so much activity and dialogue concerning the SQ improvement wrought by deleting Windows system files (and registry trimming), it may be time to gather the collected wisdom onto a webpage.
Does anyone have the power to modify/update the cicsmemoryplayer.com website ?
A full master listing of all "candidates for deletion" would be very helpful. Since no two XP installations can be expected to have identical system file contents, this would be more of a "suggested list" than a 100% "must remove" requirement.
As some early adopters have noted, the presence or absence of certain files may have a "tone control" effect, and this information should be part of the list.
Should the files be divided into groups or phases... reflecting the way they were first posted ? Are there sound technical reasons to apply the changes in stages - or can they be executed in one go for maximum SQ boost. I have no idea, and defer to the authors to make that call.
Thoughts ?
Cheers,
Grant
That's not a Toy... IT'S A TOOL !!
Well... As for ur 1st question, we must leave that to Cics. The cMP website has been hosted by him, & most of all, this project was initiated by him alone, so its upto him to add or modify its contents. And as for the question "When???", i feel its best to be done ONLY after all possible/needed deletions have been made. By saying that i mean, that it should be done at a so-called 'Saturation-Point', after which deleting any additional file will have no effect on the SQ. And/or when ALL the 'Not-really-needed' files are removed, thus leaving back only those files which are critical for the operation of Windows...
Was auditioning the new deletions I just made. I'm SO beginning to wonder how much potential is hidden under the hood of this so called Marvellous cMP project. Try & hear it for yourself :-)
wowexec
Winfax.dll
Winhttp.dll
Wpnpinst
Wupdmgr
Cdrom
FlpydiskFinally go to the Drivercache folder & delete the i386 folder. Yes the entire folder.
Have been listening for an hour after making these changes & am surprised where is it finally going to land up :-)
Truly Awesome !!!
Note:- please restart in cMP mode to apply the changes..
Edits: 09/08/11
Yes, the latest deletion list has a lot of transparency and air. But is it possible that there's too much of a good thing?
After moving the whole list, I find my attention kept on focusing on the high frequencies, while there is a thinness in the mid and bass. So finally I had to move back the winhttp.ddl file, and now there is much better balance in my system.
I wonder if this is unique for my system. If like me, you find that the high is drawing too much attention, try moving back winhttp.ddl and tell me what you think.
KC
Hi Jolida,
I am sorry about my previous contrary opinion. I was wrong! This is a great list, More transparency, wider soundstage, and more details.
I am not totally sure why I was not convinced previously. I can think of two excuses: firstly I did not give the new deletion enough burn-in time; and secondly I think that the difference in sound of the new tweak is so significance that a shift of focus was needed.
This time around, I took my time and did the deletion in stages, making sure that I like what I hear before moving on, or move back and forth if I was not sure of the benefit of a certain removal (e.g. i386). It was also a great learning exercise, how the removal of a single ddl would make such improvement is quite unbelievable.
Anyway, thanks again Jolida for the great effort. cPlay is now playing better than ever. I think I learn the lesson to wait before voicing my opinion.
Thats absolutely fine. While assessing or could I say 'evaluating' the changes made to the sound via deletions, it's better to give some time, because certain factors play a role in the Process, like our day's mood, quality of the AC power at that time etc. It's good that u te-evaluated them to see if it suits your liking :-)
Hi Jolida,
I have a contrary opinion this time. I move the files, listened for 2 hours and have to move them back.
Yes, the latest version might have more transparency, but certainly too polite for my taste, the excitement is gone. The magical immediacy of the previous version (what I called Tweak #4.5) is gone. Move the files back and I am now back to Joni Mitchell's live performance. The difference is not subtle.
I am mainly a rock and roll listener, and to me immediacy and excitement is important. I guess a chamber music listener (thoughI listen to lots of Bach too) might have a different opinion.
I had a similar reaction. As stated by many, we amay be in the relm of fine tuning. I had also made a minor tweek to a power supply, so I let the system run overnight to settle in. I must admit it was very good in the morning. Could be we need to let these things run in for 8 hours before doing the A/B listening test with drivers in/out.
It does take something like 8 hours of playing. I had to do it in steps to ease it in and it took me several days. Now I had removed the whole list and the sound is great.
Hi kclo,
If u feel that the changes made by the new deletions are not appealing to ur tastes, then its best to stick on to the previous stage. At the end of the day, its all about Enjoying music :-)
With this latest round there was an increase in transparency and resolution that is almost hard to describe.
Maybe there is a cumulative effect?
I have not re-installed the two "JUL@" files yet. I do not find any hollowness in the region described but many times (most times) it is easy to delude oneself so I am going to try it.
Being a rather mindlessly decisive type I had deleted those files completely so I had to copy them from my other computer. I put them on a thumb drive and discovered that the music computer no longer recognizes it. Even though it shows up in DEVICE MANAGER/HARDWARE it will not show up on MY COMPUTER. I will have to put it on a HDD to give it a try. Not that that is a problem. Just thought it was interesting.
Also, as I had mentioned before after these deletions you will find some of the cics recommended optimizations have disappeared. Most notably was the SYSTEM/ADVANCED/PERFORMANCE setting. I had to reset to BEST PERFORMANCE. One should take a quick run through to see what your machine might have done with these settings.
little bug: mine doesn't shut down because freezes itself closing cMP (I must turn off by PC switch )but restarts with no problems at all.
I don't know which file needs to a regular shut down, I've deleted too much but sound is fantastic. I'll can switch off throughout my life :-)
Audiodan,
I have implemented all deletions prescribed, but my cMP shuts down with absolutely no problem, like it used to. Since when are u facing this issue?
It happens after steppe's step 4. It was too many files and registry change to record. Moreover pc restarts perfectly so I cannot know with precision at which file or goup of files I could attribute this fact.
It's just bit boring but it doesn't matter
Oh ok.. When I implemented my cMP last year, I had disabled the restart option in the device manager. So I can only Shut down instead of a reboot...
I too lost ability to use a thumb drive somewhere on this journey.
If anyone figures out how to get USB to recognize a thumb drive or usb hdd, I would love to hear it. Not sure at what point it failed to be found, but I need to get this back to add music.
thanks
Do you go back to xp mode to add music via thumb drive? Also make sure at least usb 2.0 is enabled in bios/control panel.
Thanks, I'll check. The thing is it used to work. Now I put the same device in the same slot and it never gets discovered. I have to think one of these drivers was used by the external drive, or the plug and play subsystem. Just not sure how to go about finding which one.
In case someone else goes down this rat hole..
I fixed it by enabling all the usb devices. Now the usb hdd is recognized in the same usb ports that previously saw my mouse or dac but not the hdd. thank heavens I dodged a rebuild.
I too found that the system performance setting had changed.
My settings remained un-changed though !!
Thanks for heads up on this (ie, best perf).Edit: but mine did not reset.
Edits: 09/09/11
My feedback: more transparency, louder, more of the same the last jolida batch brought us. When I listened yesterday it was solid homerun. This am my system showed more of the grunge I usually get until the system warms up. But it was plainly/painfully noticeable. In 15 minutes it was back to yesterday's performance level... an out and out success!
Exactly !! More transparent. Most of all I liked the Delicacy in the Midrange & Instrument-definition. And again, Noise-floor dropped further. Am able to hear how long the piano notes are sustained in certain recordings. Overall, more clean & Musical. I'm surprised to see how quick the cMP shuts down after this deletion..
Quick question does deletion/removal of cdrom preclude you from reloading windows from a disk? That is ok for us that use image restorers like snapshot or ghost but for those that don't this could be an issue.
http://m.computing.net/answers/windows-xp/where-do-i-put-cdromsys/81265.html
We would be needing a CD-Rom drive only when we need to re-install a corrupt Windows installation (i.e. when it does not boot). So having the cdrom.sys file in the system is of no use when we are unable to boot to Windows itself. Either way, a corrupt Windows installation has to be repaired before Windows starts, by booting from the Windows set-up CD & selecting to repair a Windows partition. And as for us, who have made an image of the cMP partition with Drive-backup solutions ( i use Acronis), we as well need to start the Recovery procedure by booting from a CD of that Recovery program. The CD rom just needs to be enabled in the BIOS for both the situations to work.
As for the cdrom.sys file in Windows, its only for using a cd-rom to install something while windows is already running ( which ofcourse we would not)...
Can't wait to try you are amazing!
Hey Guys,
When trying to move certain system drivers and the option "extract" or "compress" comes up : what to do then ?
greetz
cMP2 Computer: i3-540 Core cPlay039 >FIR filter w/ convolver> Lynx Aurora 8 FireWire /192kHz throughout. 2x AcousticReality Ref. 202 & 2x AcousticReality Ref. 601´s ICEpower. Magnepan MG3.3R beechwood frames & custom stands. Miller chokes
These are files that are stored inside of '.cab' files, which are similar to '.zip' files. These are packaged for installations and are not executable in this form. My opinion is that you don't need to bother with these files... they aren't running and aren't accessable to Windows without an running an install program.
Greg in Mississippi
Everything matters!
I dont quite get ur point..??? Those options show up when ur trying to delete which file(s)??? Deleting .dll or .drv files are as simple as deleting any other file for that matter. Are u trying to delete them to the Recycle Bin or do u wanna move them to a Temporary location? Either way, the options u are getting will show up only when u right-click on a file & have something like WinZip or WinRar installed.. A deletion is straightforward. Just Right-click & delete. Or am i missing something from what ur trying to ask????
Jolida,If you use Windows' Search to find files, it will show you the ones inside of '.cab' files too. These are the ones that will give you the option of "extract" or "compress" when you try to move them.
And as I mentioned above, they can be ignored.
Greg in Mississippi
Everything matters!
Edits: 09/08/11
Hi serge. I made deletions as per what u suggested in Part 4 but was somehow not as pleased as I thought I would be. It dint sound balanced. Especially in the bass & mids. Waited for it to burn in for almost 48 hours now. But still not happy. So I went ahead & tried something wild. Made a list of files which I thought could effect the performance in addition to what we already have deleted as per Ur suggestions. So I deleted a few files & was stunned at what happened. It's like the whole music spectrum has cleaned up. Sounds like as if the noise-floor came down a couple notches. Am able to listen at a much more louder volume than I mostly do. Sounds very controlled as well. These are the files I have deleted. A few are from the System folder as well. But most are from System32.
fp40.ext
fxsocm.dll
iis.dll
imsinsnt.dll
msgrocm.dll
msmqocm.dll
netoc.dll
ocmsn.dll
tabletoc.dll
zoneoc.dll
Avicap.dll
Avifile.dll
MCIAVI
MCIWAVE
MSVIDEO.dll
TAPI.dll
winscard.dll
wmp.dll
wmp.ocx
wmpasf.dll
wmpcd.dll
wmpcore.dll
wmpdxm.dll
wmploc.dll
wmpshell.dll
wmpui.dll
wmspdmod.dll
wowfax.dll
wowfaxui.dll
I see a file NTOC.dll, but not netoc.dll. Perhaps I missed it. Is it possible this is a typo?
No it's not a typo. The file is netoc.dll
The file u specified is a non-system file. Must be belonging to some software u have installed on Ur system
Edits: 09/09/11
I have been noticing this since about a week. The Bottom end, & for that matter the whole Range does not sound FULL without the wdmaud.drv file in it's respective folder. I agree that Juli@ does work without the file as well, but the bottom line is that the end-result sounds incomplete without it. Without the wdmaud.drv file, I feel that, if not full, atleast half an octave sounds missing in the lower registers ( somewhere between the bass/midbass/lower-mids section). The male vocals sound leaner & the impact in Bass guitars lacks the Bite which kinda makes the sound, as we could say "light-weighted". Whether Juli@ needs the file to work or not, it sure is needed to make the presentation Complete.
Last night when I was having my listening session, I did one more thing. Serge had suggested that we delete the file named "SOUND" in his 1st list. I put it back in it's respective folder, & heard a wake-up call again. When we make deletions, we need to be absolutely focussed & sure as to how it sounded before & what it's sounding after the deletion. A CHANGE need not always be an improvement (No Offence to Serge). That's what has happened by removing those two files. I suggest u all try it as well & post a feedback.
I cannot comment on the improvements of users who use other soundcards apart from Juli@. It's a sure possibility that they too might equally benefit by putting back the wdmaud.drv & SOUND files in their original folders. But as far as Juli@ is concerned, it's a MUST !!!
Edits: 09/08/11
No, I am not a Juli@ user. I use ASIO4ALL and the Devilsound dac through an USB isolator with an external battery power supply.And as I was listening to Tweak #4.5 (that's what I call Jolida's first list) I thought to myself, why don't I try Jolida's other suggestion posted a week ago, putting wdmaud.drv back. After all no harm could possibly be done because I can always remove it if I don't like it.
To put a story short, putting wdmaud.drv back does make a big difference. The sound is now no longer thin; not only is it fuller and more natural, the soundstage is now more extended to the back, so that it is now square instead of a thin rectangle.
Then Jolida make another suggestion, put SOUND back. After putting back the sound.drv (lower case in my system) back to System32 and dllcache. I immediately played Van Morrison's 'His Band & The Street Choir'. I always thought that record is the worst recorded of his 3 classics which include "Astral Weeks', and 'Moondance'. This applies to both the LPs (Japanese pressing) and the wave files ripped from Japanese remastered CD. But by putting back the sound.drv file, the sound of that record is no longer a bit recess, but is as immediate as the sound on 'Moondance', one of the most immediate recordings ever! Imagine my surprise! Everything is now in my room!
A revelation indeed. Everyone MUST give this a try.
P.S. I understand that Jolida had created a second list. I think I will sit this one out, at least for a couple of days. The sound coming out of the steppe tweaks 1-4 + Jolida's 4.5 + 2 DRV files back has created such fantastic sound I think I will just happily enjoy that for the moment.
Edits: 09/09/11
Hi kclo,
I understand your situation completely. It has happened to me many a times in this hobby, that when ur system plays flaw-less after trying some tweaks or upgrades, one does not want to mingle with the Already perfect Equation that has been created :-) & wants to enjoy by re-discovering his precious recordings which now sound "ENJOYABLE"...
Sound goes back into system32 drivers subfolder, correct? I tried but did not reboot and I did not like it. Did you reboot after replacing Sound?
The file that I deleted earlier was the "SOUND" in capital letters. It was in the Recycle Bin which I just restored. I think it does go to the drivers folder if I'm not wrong. Yes a reboot is a must. Reboot in cMP mode. Restore the wdmaud.drv file as well...
Rebooted and its better but still not sure I prefer. There is more (apparent) transparency with Sound deleted and a deeper soundstage but with Sound back in there is a more traditional sound...a certain brightness that disappears (which is appealing) and soundstage moves more up front (which ims is not appealing). I'll listen a few days and see/hear which I prefer. There is a theoretical aspect of this that bothers me. I get that deleting stuff not needed reduces pc overhead. Putting stuff back in to eq the system strikes me as wrong. I have a digital crossover that allows me to eq stuff so I prefer to do it that way. However with Sound deleted I did try to eq with my crossover and it was not as effective as 'putting Sound back in'. But I am pragmatic about this so I'll live with this a few days and make a final judgment then.
Thanks again for all the suggestions. BTW I take it that you prefer Sound back in (I have all juli@ drivers still in Sys32) but what drove you to try this? Was it a dissatisfaction and did you try to put other drivers back in as well?
A week ago, I had made deletions based on a screenshot by Mihaylov. I deleted all the files listed on it which CHANGED the sound though, but I wasn't sure if I liked all of it. Felt as if some in-between stuff was missing. Though it became more Open & light, I felt that passages having drum-strikes & lower frequencies lacked Attack. Before making those deletions, I had noted that the wdmaud.drv was on the list of the drivers which Juli@ was using (in the Control Panel > Sounds > Juli@ > Properties > Driver. So curiously wanted to cross-check if it's deletion was worth it. And to my surprise, it brought back that missing section which I felt had to be restored. I immediately notified Mihaylov about it & he said he would look into it.
Last evening, I was guiding a cousin of mine who lives close-by with the procedure of deleting the files step by step. He first deleted the 16 files of Serge's first list ( which included SOUND) & upon completion we noted that though it became better in the high frequencies, something was giving a feeling of Hollowness in the lower mid-band. Ran it continuosly expecting things to settle, but no luck. Thereafter we started restoring each file at a time, & deleting it again to re-check, we came to a point where restoring the Sound file filled that Hollow section & sounded more correct.
He is using Juli@ as well. I understand that deleting un-necessary files reduces pc overheads that tax the motherboard thereby polluting the supply. But what I am most careful about, is not to use any tweak in the form of a "tone control" or as u say an EQ, which may reveal more airiness but which is tentatively the result of pulling down other frequencies, thus creating a feeling of something New..
I certainly agree that we all have our personal listening tastes, which may appeal to me & not to others. But it showed a positive change to me, so I thought I would encourage other inmates who use Juli@ to also try the same.
Thanks for Ur feedback :-)
Edits: 09/09/11
Fair enough believe me I am not criticizing...just providing the feedback you requested. Make no mistake the files you suggested to delete are perhaps one of the biggest tweaks I have made to CMP...keep up the good work.
Thanks :-) Ur feedback is important to me, as well as to all who are reading this forum. Audio Asylum, by itself is a place where we share our views about each other's findings, so that other inmates can also benefit from them. I truly appreciate the efforts taken by Serge & Mihaylov, who have let us in this direction, of improving the cMP, based on a Software-Level modification. Two thumbs up for them..!!!
Hi all
also this new chapter of "deletion files war" or better "russian roulette" was done.
I agree on transparency, clearly best, others veils are dropped (a lot!) between my loudspeakers and my ears and from this background silence new particulars shows up. I could name this like Silent Music, music born from deep silence? Or better from silence and vacuum, it's a completely new experience for me, very exciting.
I need other hours of listening because I've had a first impression of a leaner sound, more open but with tonal balance slightly on highs. No cold, no harsh only more open so I need more time to find proper new settings, but results are outstanding.
It could needs a software burn-in (another strange phenomenon, but true)?
Impressive!I can't believe that deletion of few files could causes so big change in SQ.
Thanks a lot for your job, jolida
My pleasure :)
By the way Which soundcard are u using ?
I use an Asus Xonar Essence STX with his analog output. I'm waiting for next steppe' steps for make a complete update on all files deleted.
It seems to me that also your "list of deletion" needs a little burn-in time.
It's strange but after few hours sound is clearly better balanced, no more "lean" sensation. It sounds great, really
THX again Jolida
Ryelands points out it takes time to appreciate subtleties. I add that in taking time one accommodates to the 'new' sound so it seems to become more balanced...i.e., more familiar. Or, OTOH, one gets over the thrill of the new and finds it thin or flat after a while, which also has been reported. I believe there is no doubting that these deletions have improved the sound, each phase of deletions, but I am a bit skeptical of the reports of clear changes due to deleting one element or returning one element though, of course, it as possible as any of these changes--all are hard to explain.
Thanks, none the less, to Jolida.
. . . your "list of deletions" needs a little burn-in time
If you mean that it takes even experienced listeners time to appreciate the effect of subtle changes in a configuration, I agree with you.
Simply detecting a change (which experienced listeners can do very quickly) and properly evaluating it are different things. Listeners are inevitably "anchored" to the old base-line.
If OTOH it is implied that the way data are processed changes for the better over time (all else being equal) simply as a result of minor software changes then I cannot agree.
If people find that it takes time fully to appreciate a given change, might I suggest that we avoid the term "burn-in" as it has a reasonably well-understood meaning associated with chemical and other changes in audio components as functions of age and temperature.
Something like "give this a little time" perhaps?
Dave
Really nice to hear that. Thanks for your appreciation...
My favorite time to listen to my system is 5:00 am - 7:00am. This is when my powerline is the least distorted. I just listened and wow these deletions take my system to a level I have not heard before. Can't say enough good things about the effectiveness of your recommendations.
I now hear what you say about mids and mid bass but for me the beauty of these recommendations is in the highs. All that has been said in this thread are indeed what any keen eared audiophile can hear. Do check what files you need for your soundcard and yes please do try these deletions.
It's because the Highs are free from the grunge of the Bass which was dominant on the whole spectrum till this had been done. The deletions have balanced the tonal equation to a very great extent. The experience is more "Out-of-the-box". If the tonality is not balanced, it sounds Analytical. I guess now it's become more "Musical". I'm glad you liked it :)
Yes exactly... I perceived a very strong change as well. Was apparent immediately upon doing it. I made the deletions based upon critical listening after deleting each file. Most important was getting rid of the dll files starting with the 'wmp'. I specifically wanted to delete them, as it did not make a Very strong change by just cutting the Windows Media folder from the Program Files. I suppose those files were still messing up things in the background.
Wlowes :- I think ur soundcard uses the 'wincard.dll' file which i specified. As i am using Juli@, it worked for me, but since u r using different soundcard, it must have had a conflict. I suggest u put it back in the system32 folder IF u have not permanently deleted it. If not let me know & i shall send it to u by email...
Theob :- Yes the edginess has vanished. I too was noticing the vocals to be a little strained especially in the lower-mids. If u have noticed, the mid-bass has also cleaned up, thus cleaning up the Vocal ranges as well.
Riboge :- You got that right !! the Imaging now looks spot-on, because the noise-floor has gone down making the spacing between intruments to EXIST. Prior to this, it was'nt completely deliniated. Now that the background is more Black, we are able to localize the presence of instruments more clearly without straining.
This is very inspiring. I had made many Hardware-level upgrades to my cMP like making Linear supplies & Ferrites & Caps & what not. But im surprised that these Software-Level changes can lead to an equal or if i could say a better improvement than the former. And best of all, its FREE :)
I don't want to raise an unwarrented alarm, but for me this did not work out. I can boot, cplay starts, and appears to play, but no sound. I am using aqvox asio driver. Perhaps something else is wrong. I'll debug tomorrow.
After deleting files, some exe failed to start due to wincard.dll missing.
otherwise system normal, but no sound. running xp home sp3.
My appologies, this was truely a false alarm. It seems a signal wire deep inside my DAC broke as I deleted the files.
I have now got it working again, and have made these deletions and could not be happier with the results. I'll save listening impressions for when I have had more time, but no question the noise floor has dropped away leaving air, and details not before heard.
Thank you again, and sorry to those to whom I have caused hesitation.
Walter
Hey Guys,
We also had problems with a cMP running a RME Soundcard. All deletions worked (uptil Part 3), except : ks.sys & ksproxy.ax.
Just beware.
kind regards
cMP2 Computer: i3-540 Core cPlay039 >FIR filter w/ convolver> Lynx Aurora 8 FireWire /192kHz throughout. 2x AcousticReality Ref. 202 & 2x AcousticReality Ref. 601´s ICEpower. Magnepan MG3.3R beechwood frames & custom stands. Miller chokes
Hi wlowes
has you checked the drivers used by your aquox before starting this "russian roulette", on "sound" of your control panel?
I think this is mandatory for everyone before any alteration of files inside system 32 folder, maybe you've deleted some of these ones.
I'm sorry for you, but these "delection work" is the main upgrade until now,IMHO, but is "PC sensitive" so results can diverge from one to other, formerly for the different soundboards used.
Daniele
Everything does sound YET better.
I think you are on the right track about noise. Not just a noise floor but noise that was there and not perceived until it was gone.
Noise between instruments, notes, etc.
Thanks,
Ok Jolida magnificient! The air around instruments is stupendous. How can it keep getting better. What strategy did you use to find these files? Was it just random or was there some method to the madness because you chose the best ones to delete/move. This might be best of the best based on a cursory listen to. I have a file Bach Concerti for 3 harpsichords by the Cologne Chamber Orchestra which was always on the border of brightnes/edgyness. Now the instruments just hang into space dripping of liquidity!!
Well done!
I deleted your list save the few I didn't find. It seems to bring a further improvement along the lines of the last two lists. Greater clarity and dimensionality with more exact localizations. All ranges sound full and, as you say, clean. I am listening to a marvelous recording of Porgy & Bess, the full musical on EMI CD with Rattle and the London Philharmonic. I can close my eyes and be in orchestra center so for real I wonder where my tickets are.
which is not saying much (I am no one to make such a judgment!) but they seem to be more of the same ilk as Serge's deletions. I gave a cursory look-up with GOOGLE and saw nothing scary about any of the files.
As I mentioned before, three of the files were on Serge's list.
I will try them tonight and give my opinion tomorrow.
I only saw b39 as the lastest version.
This Tweaking should have been posted on the cMP Forum, and not been called "a new version of cPlay" !
cMP2 Computer: i3-540 Core cPlay039 >FIR filter w/ convolver> Lynx Aurora 8 FireWire /192kHz throughout. 2x AcousticReality Ref. 202 & 2x AcousticReality Ref. 601´s ICEpower. Magnepan MG3.3R beechwood frames & custom stands. Miller chokes
"...not been called "a new version of cPlay" !" -
I completely agree with it. These tweaks have no relation to cPlay.
http://cmp2-mihaylov.narod.ru/
Edits: 09/03/11
Well I think it has been posted in the correct forum. In the cMP forum we discuss hardware-level modifications. In cPlay forum we are discussing software-level modifications. It makes things a lot easier & organised...
There is no such official version of b4xx. 2 b4 onwards are in reference to the tweaks by inmate steppe.
Was brought to my attention by Dr. Bill Gaw's column.
He has used cPLAY in the past but his system is so complicated I guess cPLAY does not work for him (video and all of those extra channels).
He writes about this in his column http://enjoythemusic.com/magazine/viewpoint/0911/aa_chapter_142.htm. He is one of the best in our hobby and his system is unique in every sense of the word. Nothing like it in my experience!
Maybe this will be good for those with VISTA and 7; it is not for XP.
Seems like a small program in size - much smaller than cPLAY.
Personally, I cannot imagine it being as good especially considering all of the added work done by others on the FORUM.
How about that name? Not very original!
"He writes about this in his column http://enjoythemusic.com/magazine/viewpoint/0911/aa_chapter_142.htm." -
Josef Piri wrote that "Like Marcin, I also used Foobar initially..." and no words that Marcin was cMP2's fun earlier.
http://www.audioasylum.com/forums/pcaudio/messages/7/70465.html
http://www.audioasylum.com/forums/pcaudio/messages/7/70512.html
Generally the authors of JPlay like to compare own player to other players but ignore cPlay as though did not hear about the such. It is strange.
http://cmp2-mihaylov.narod.ru/
> > > Generally the authors of JPlay like to compare own player to other
players but ignore cPlay as though did not hear about the such. It is
strange.
JPlay is a commercial product.Certainly they would not to bring to the readers attention a competition, which is better and free.
Hello, everybody. Here is the fourth set of tweaks. It is very long and the changes in SQ are profound. Some of the proposed things may sound scary but they are not. The system is totally stable after 72 hours with a dozen reboots, file copying and so on. It plays very much better. As usually it is assumed, that you have no network, removed Windows Media Player folder and don’t watch videos on cmp computer, no modems and the stuff. The tweaks were NOT tested on Asio4all, so be wary.
1. Files to remove (there may be repetitions from part 3) from System, System32, Drivers, Dllcache folders:
Amstream.dll, cewmdm.dll, splitter.sys, swmidi.sys, sysaudio.sys, cinemst2.sys, mspqm.sys, tosdvd.sys, tsbvcap.sys, vdmindvd.sys, gm.sys, cpqdap01.sys, sonydcam.sys, msvfw32.dll, msvidc32.dll, msvidctl.dll, msvideo.dll, msyuv.dll, wmplayer.adm, browsewm.dll, cdplayer.exe.manifest, dsound.vxd, wmiprop.dll, wmnetmgr.dll, ksproxy.ax, iac25_32.ax, wdmaud.drv, dxdiag.exe, dvdupgd.exe, dpvsetup.exe, logagent.exe, amstream.dll, datime.dll, dmcompos.dll, dvsetup.exe, encdec.dll, l3codeca.acm, msg711.acm, msg723.acm, msgsm32.acm, sl_anet.acm, tssoft32.acm, drmclien.dll, dimap.dll, dimput8.dll, dinput.dll, dplaysvr.exe, dpnsvr.exe, daxctle.ocx, msdxm.ocx, actmovie.exe, dvdplay.exe, avifil32.dll, avifile.dll, avicap32.dll, avicap.dll, winhelp32.exe, winhelp.exe, hh.exe, wmsysPr9.prx, aec.sys and beep.sys (these 2 are from system32\drivers), msvideo.dll, mciavi.drv, mmtask.tsk, timer.drv. As usually, many are doubled, if any file is wanted by your soundcard, just put it back.
2. Folders to remove from Program Files and Windows completely:
Program files\common files\speech engine
Program files\common files\Microsoft Shared\ speech
Also folders from Program files : Videolan, Movie Maker, NetMeeting, Internet Explorer, Outlook Express, Messenger, Online Services. Yes! Yes!!!
From Windows\Media, Msagent, Help
From Windows/system32/ Macromed (all told 13 folders. Some will start yelling, don’t pay attention. If You are sorry to part with them for good, remove them to your temp folder.) Reboot. Play music.
3. Now comes the “scary” part, but the most important. If you are following the tweaks from part 1, you probably remember me mentioning registry items deleted. Here it comes again but first come words of reassurance. I did it together with my former sysadmin, and everything is well documented. Also, yes it IS dangerous to change registry values but NOT DANGEROUS to delete some items completely.
Here They are (for the beginning):
Start-Run-Regedit-press enter.
To The left of the window you will see the structure (just like a normal but deep catalog tree).
First go
Classes-Root, find wav and JUST DELETE IT by right click on your mouse
Next delete everything BDATuner (about !48 entries), next DirectSound (22 entries), next MediaPlayer and MediaPlayer1 (4 entries), WindowsMovieMaker, Mmedia, DirectPlay, Direct Sound, mplayer, wma, Microsoft Direct Music, Microsoft Direct Sound (some have many entries).
Second go Hkey-Users and open Default-Software-Microsoft. Find and delete Internet Explorer, Media Player and Windows Media, repeat it with S-1-5-18, S-1-5-19 and so on (I don’ know how many users you have)
Some entries may yell at you that they cannot be opened. That’s because you have already deleted their contents from default user. Reboot about 3 times when editing Registry. Play music. Have fun. Tell your impressions, please.
Serge.
P.S. The tweaks in the registry are only a beginning of what will come. I suggest you to get used to the idea of editing this sacred cow lightly. I have already done a lot more with registry, but that will come next time. Let your system burn in for 15-20 min.
P.P.S. I have already prepared tweaks part 5 and was in the middle of part 6 when Windows finally crashed. I don't know if I would be able to restore the registry or would have to start EVERYTHING from scratch, but I may be absent for some time. This part 4 and part 5 that I will present to you later, are perfectly safe. Just wait, please.
Edits: 09/02/11
What's interesting is these drivers could previously NOT be disabled. But after deleting, Device Manager allows you to disable them. For '2.0b41' under "Sound, video and game controllers" you get:
All apart from your installed soundcard can now be disabled.
I still have to do '2.0b42' and '2.0b43'. Are there more devices that appear with ! after additional file deletions?
PS - It's been a while. Nice work Serge!
.
So nice to see a post of yours. Welcome back. Serge, Mihaylov, and Jolida have been doing a great job of keeping CMP^2 moving.
Hi serge. I made deletions as per what u suggested in Part 4 but was somehow not as pleased as I thought I would be. It dint sound balanced. Especially in the bass & mids. Waited for it to burn in for almost 48 hours now. But still not happy. So I went ahead & tried something wild. Made a list of files which I thought could effect the performance in addition to what we already have deleted as per Ur suggestions. So I deleted a few files & was stunned at what happened. It's like the whole music spectrum has cleaned up. Sounds like as if the noise-floor came down a couple notches. Am able to listen at a much more louder volume than I mostly do. Sounds very controlled as well. These are the files I have deleted. A few are from the System folder as well. But most are from System32.
fp40.ext
fxsocm.dll
iis.dll
imsinsnt.dll
msgrocm.dll
msmqocm.dll
netoc.dll
ocmsn.dll
tabletoc.dll
zoneoc.dll
Avicap.dll
Avifile.dll
MCIAVI
MCIWAVE
MSVIDEO.dll
TAPI.dll
winscard.dll
wmp.dll
wmp.ocx
wmpasf.dll
wmpcd.dll
wmpcore.dll
wmpdxm.dll
wmploc.dll
wmpshell.dll
wmpui.dll
wmspdmod.dll
wowfax.dll
wowfaxui.dll
Avicap.dll
Avifile.dll
MCIAVI
were on Serge's list. Just thought I would make mention of this.
I have, to this point, been pleased with Serge's recommendations as to balance. One wonders if they are happy now would we not be happy after making these additional deletions?
So no problems with these?
Might have to give it a try.
It is amazing how much smaller my WINDOWS folder is now.
Thanks for YOUR intrepidity, also!
I have been meaning to ask this for a while. The registry AFAIU is just a catelogue of settings, paths, etc. While the effects of many of the cMP2 modifications are hard to specifically explain, these in particular seem inexplicable. If you remove keys for needed items you get errors and/or malfunction. If you remove keys for unused,unneeded or deleted items, these entries are not used even if present, so how does removing them change anything--except perhaps for freeing up minute amounts of memory for those that are loaded at startup despite going uncalled on?
I wonder if anyone--Serge or Mihalov?--has actually tested the effect of removing only registry items in the recent 'versions'?
Deleted files from step 4 & let it burn for 24 hours now. The highs have dropped down considerably. Sounds dry. Is this normal ?? It was better when played soon After step 4. But now it's starting to get dry in the upper mids & treble. Does it need more burning in??
I went back to see if the cMOP "standard" setting had changed after finding out I could no longer change my DISPLAY settings and sure enough you cannot.
Everything seems to have reverted to standard WINDOWS settings AND you cannot change them because nothing happens.
The questions posed are:
1. Are these settings no longer applicable once these files are deleted?
2. Have the previous setting actually remained we just can't "see" them any longer?
3. Can one bypass this part of setting up a cMP machine when deleting the files or should one do it anyway because it just seems that they have reverted?
Or is this unique to my machine?
Bye,
Interesting questions, rick.
I was worried about this but results tell me that my two systems are better than previous. Another question rises, now: what happens if we set only minlogon,kernel, cics'registry settings and memory page locks then steppe's steps only?
It could be interesting to try.
Daniele
nothing relpaces minlogon, autoruns, and other registry changes.
I meant the CONTROL PANEL settings.
But, the question remains: do those settings remain if made before the latest deletions or are they made redundant or useless?
Could save some time on a new system. Not that the time required to make the deletions is minimal and with more to come ... Can't wait for the next round, don't get me wrong!
Serge, you have made this adventure an adventure again.
Now if you would just clarify what we do with wma.
Bye,
Do we delete .wma or wmafile?
Thanks,
Rick WMA is in the top list of lower case stuff in this key. I was looking in the bottom one and didn't find it 1st time through.
.
Now that I deleted it I don't know. But if its in the top alphabatized list then yes.
My problem was that he made it clear we deleted .wav but when it came to wma that is all he said. No period in the prefix.
There is also a file called wmafile with no period later on down the list.
This led to my confusion.
Did not want to delete something THERE without being sure. If you deleted and not having problems (??!!!) I will take your word for it and delete it.
Easy enough to replace deleted files but I have no idea how to replace deleted registry entires so I thought I would wait for an answer.
Thanks,
Dear Inmates,
Stepped into these modifications and like to confirm : GREAT !
I deleted the first 15 DLLs and the improvement was apparent right away :
More air and ease combined with more focus and deeper soundstage.
Step 2 turned the mid-bass markedly leaner and made the overall presentation turn brighter.
Following another long bunch of DLLs & Drivers the sound resurrected its natural balance, but improvement was less marked.
-but what a revelation !!
Has anyone a technical explanation to this ?
I´m AMAZED !
Thanks a bunch, guys.
cMP2 Computer: i3-540 Core cPlay039 >FIR filter w/ convolver> Lynx Aurora 8 FireWire /192kHz throughout. 2x AcousticReality Ref. 202 & 2x AcousticReality Ref. 601´s ICEpower. Magnepan MG3.3R beechwood frames & custom stands. Miller chokes
This evening I've finish to climb the mountain and the four step also was done.
The sound quality is now miles away from the previous, no doubt, no problems. All perfect
Everythings works, a little bug shutting down the PC, it cannot, so I must turn off the switch, no matters.
A question to steppe; I've found in the registry a lot of keys Direct sound with extensions. I must cancel everything because extensions are no critic, or only the former? Also Microsoft Direct Music and Direct sound get the same problem. Can we have some more informations about?
Great job, I can't believe it's so good.
BRAVO!
Daniele
Oh shit after listening for an hour or so I get a very loud staticky sound that if left unchecked will probably take out my panel amplifier. Hope nobody else has this. At 1st it sounded like Cryptograph Service kicked in and it did so I set it back to Stop. Didn't work. So I rebooted 3 times still there. I am going to have to retreat to a prior windows image. Sorry Steppe.
I reloaded a Stewppe pre-tweak # 4 and got same result. Then I took out one of my ferrite/cap infested p24 cables and now I can't even get the thing to try to boot. So I think I have a hw problem somewhere.
,,
So I ran into a problem where I could not boot and after a day of trying to figure it out I discovered my agi psu (essentially a dirty power unit running all except p24) was faulty. I run dual switching psu's ala Cics recommendation a while back. So the slow boot, and the anomalies I have been suffering the last few months was probably caused by this AGI psu going bad and finally failing. So I'm running my antec for all power feeds and it works well but just does not sound as good as dual psu's. This is for the benefit of those wondering '...what did Theob do this time...' or is Steppe #4 safe. I'm sure it is safe although I have no guarantees about Theob.
Anyway I need another psu (no more AGI's). But all my local stores are closed (Labor Day in the States) so I have some time to think about it. Should I get a pico to run p24 or another Antec or similar ranked standard ATX brand?
Shure! A picoPSU 160 XT feeded by a little linear PSU Belleson Superregulator based. It's a bit tricky, just a bit, but improvement are greater than bother.
Your mobo must get a power consuption lower than 2A however.
I live happy with it. Great device!
Daniele
Ok finally got through the whole list...magnificient! When I went through the register at first I missed Direct Sound and while sonics were ok but they were not as good as everyone reported. Now that I deleted Direct Sound the highs are way better. I agree with all other comments on sonics but for me highs are now perfect.
Edits: 09/03/11
I had just done part 1 and 2. Amazing improvement on the get go. Unlike tweak # 3, sound is excellent without even burn-in. Soundstage depth is now very plausible for the first time. And what clarity.
Just could not imagine the improvement that is still possible after tweak #3. And there are more to come. Can you believe that guy steppe?
Just deleting my pet hate IE and Outlook is enough to make me delirious with joy.
Fellow Inmates,
It seems to me that an awful lot of these deletions could also be performed by XPlite.
Maybe also at a time earlier in the entire setup/optimization of cMP....
-what are the thoughts on this ?
Kind regards
cMP2 Computer: i3-540 Core cPlay039 >FIR filter w/ convolver> Lynx Aurora 8 FireWire /192kHz throughout. 2x AcousticReality Ref. 202 & 2x AcousticReality Ref. 601´s ICEpower. Magnepan MG3.3R beechwood frames & custom stands. Miller chokes
Hi all
I've finished the third step in both my two cMP2 systems. It works but ks.sys is a part of Xonar Essence ASIO driver, so I've reinstall it.
I've noted that each is slightly different from the other,although both are born from the same installation CD. They get same chipset mobo ( H67) but different models ( GA H67MA UDH2-B3 and GA H67N USB3-B3) same CPU and some difference in HW.
Now, before step 4, I wish to know if PC retains residual functionality or It will be used for music only or could install a new program or using a new usb device i.e.
It seems to me a very aggressive setting, I haven't fear but change registry key is not a trivial operation and could run to a crash of the entire system. About disk image backup which programs are better?
THX
Daniele
"About disk image backup which programs are better?" -
Acronis! http://www.acronis.com/homecomputing/products/trueimage/
http://cmp2-mihaylov.narod.ru/
I have been using Acronis True Image since i started cMP about a year ago. It works best for me as well. A snapshot(back-up file) of an entire cMP partition comes to less than 750MB..
About disk image backup which programs are better?
Everyone has their favourite; mine is Snapshot (see link). One useful feature is that you can set an imagefile to look like a CD by clicking on it in Explorer. That allows you to extract individual files. (Other products might by now do the same but they didn't when I bought Snapshot over five years ago.)
With a good utility such as this and a docking station (so you can restore on a different computer and avoid all that "how do I get the thing to boot?" hassle), you can go from cMP^2-is-dead-in-the-water to back up and running in under ten minutes.
Obviously, you need either two drives (or a partitioned drive) or a network as you can't restore the only disk/partition on a computer from that disk or partition.
They're pretty much essential tools if you're trimming an OS and are also useful for general maintenance.
I am pleased with the sound. As stated below there is more depth. I find a more stable center image. General tonality is improved.
I thought I would mention things that have occurred for me. My machine is with the latest recommended components. I am using JULI@ digital only. I think I have incorporated all of the suggestions made in the past. The machine is only used for transport duty and has never been connected to the internet.
1. Contrary to theob's experience I find booting up even faster and this is with the SSD. Maybe a second faster to the blue screen.
2. One no longer has the ability to change the color of the desktop screen. It is now default blue. When you go to PROPERTIES you are no longer given any choices. I noticed when going to MY COMPUTER and C drive when the screen came up it was with the standard screen - no files showing. You had to click the request to show all files. Going to the VIEW/FILES gave no option. This must be the default screen. I am thinking that tomorrow I will take a run through the standard cPLAY settings and see if anything has changed.
3. No problems with keyboard (my main input) or mouse.
4. I made all of the "deletes" in one go without any problem.
5. I made all of the registry changes at one time, also. Booted right up after each set of changes were made.
6. Concerning registry changes I could not locate DIRECT SOUND. I deleted everything that had DirectPlay, MicrosoftDirectMusic and MicrosoftDirectSound in the beginning of the title. I was confused by wma - is this .wma or wmafile?
7. For what it's worth these are the files that did not exist in my machine, keep in mind the files deleted were both Serges #4 group and the files suggested by Mihaylov:
amstream.dll
cewmdm.dll
cinemst.sys
cpqdap01.sys
dvsetup.exe
gm.sys
iac25_32.ax
mciavi.drv
mfo42.dll
winhelp32.exe
wmiprop.dll
wmnetmgr.dll
The following files only showed un-deletable CACHE files when searching:
splitter.sys
swmidi.sys
sysaudio.sys
tosdvd.sys
tsvbcap.sys
vdmindvd.sys
wdmaud.drv
8. A week or so ago I deleted the DirectX file with no complications.
Hope this is comforting to those that have had similar experiences. If anyone recognizes something I have done wrong PLEASE tell me.
Same experience on many fronts. Faster boot as well. No issue doing all deletes then one reboot then all registry changes then reboot and play. Also confused by WMA.
OMG who would have guessed this much improvement. Thank you
I was impressed by the impact of steps 1, 2, & 3, but not sure I liked the outcome. More detail for sure, but a lot of trebble on my system.
This tweak #4 just brought it all home. I now have deeper bass than ever. The entire spectrum is detailed clear and natural. But so much more bass filling in everywhere. Lots more texture. Sound stage a little smaller? perhaps and for sure moved back deep into the room. But very natural and relaxed.
It did need to settle in as if there were new caps and wire in the system. Why is that? I do have a lot of fairly new caps in my dac and lightspeed. Do they need to adjust to the new level of detail and the beautiful textured sound?
Sorry you have to rebuild.
Again.. Bravo!
I completed all of part 4 deletions except the 4 items possibly related to wired mouse and monitor. OMG! How can there still be such improvement!
I did have trouble again with the Weiss dac202 driver. It wouldn't run complaining of the absence of MSVFW32.dll. I replaced it but a driver reinstall was still necessary.
There was no problem with networking--as Mihaylov posted to reassure me. Meanwhile I found it out on my own, too. Strangely, the changes fixed a problem I had posted about before about larger files that require segmentation not playing via network beyond 30 seconds or so. Now, so far, they do.
Hats off to Serge and Mihaylov. It is incredible that there might be further improvements still, but I eagerly await them anyway.
1st report. I'm only part way through...did not do the registry yet. But 1st thing I noticed is a very slow boot. But it does boot up. Sonically its very smooth, less edge, good solid bass very very nice. Today's weather is very hot 96 degrees f. But my power line is good at 120v. Yesterday it was hot but very difficult to listento: edgy (powerline noise from all the a/c units running I surmise). Today is hotter and very very nice way better than yesterday at same time. So far so good. Well done Serge.
I noticed there were some duplications on both lists and figured it would not hurt to compile them in alphabetical order.
Serge, if there is a reason you had your files in an optimum order for deletion please let it be known.
Thanks to both of you for your gifts. (both physical and mental)
I find it easier to copy these lists into a work processor so I can make the typeface larger!
I hope I got these right since I had to type in Mihaylov's.
Combo list of Mihaylov and Steppe's recommendations
actmovie.exe,
aec.sys
amstream.dll,
avicap.dll,
avicap32.dll,
avifil32.dll,
avifile.dll,
beep.sys (these 2 are from system32\drivers),
browsewm.dll,
cdplayer.exe.manifest, my office machine cannot find this one
cewmdm.dll,
cinemst2.sys,
cpqdap01.sys,
d3d8.dll
d3d8thk.dll
d3d9.dll
d3dim.dll
d3dim700.dll
d3dpmesh.dll
d3dramp.dll
d3drm.dll
d3dxof.dll
danim.dll
dataclen.dll
datime.dll,
daxctle.ocx,
ddraw.dll
ddrawex.dll
diactfrm.dll
dimap.dll
dinput.dll
dinput8.dll,
dmcompos.dll
dplaysvr.exe
dpnsvr.exe
dpvsetup.exe
drmclien.dll,
dsound.vxd,
dvdplay.exe
dvdupgrd.exe
dvsetup.exe,
dx7vb.dll
dx8vb.dll
dxdiag.exe
dxdiagn.dll
dxtmsft.dll
dxtrans.dll
encapi.dll
encdec.dll,
gcdef.dll
gm.sys,
hh.exe,
iac25_32.ax,
ir41_32.ax
ir41_qc.dll
ir41_qcx.dll
ir50_32.dll
ir50_qc.dll
ir50_qcx.dll
ivfsrc.ax
joy.cpl
ks.sys
ksproxy.ax,
l3codeca.acm,
logagent.exe,
mciavi.drv,
mciqtz32.dll
mfo40.dll
mfo42.dll
mmtask.tsk,
mpeg2data.ax
mpg2splt.ax
msdmo.dll
msdxm.ocx,
msg711.acm,
msg723.acm,
msgsm32.acm,
MSPCLOCK.sys
mspqm.sys,
msvfw32.dll,
msvidc32.dll,
msvidctl.dll
msvidctl.dll,
msvideo.dll,
mswebdvd.dll
msyuv.dll
pid.dll
qasf.dll
qcap.dll
qdv.dll
qdvd.dll
qedit.dll
qedwipes.dll
quartz.dll
sl_anet.acm,
sonydcam.sys,
splitter.sys,
swenum.sys
swmidi.sys,
sysaudio.sys,
timer.drv.
tosdvd.sys,
tsbvcap.sys,
tssoft32.acm,
vbisurf.ax
vdmindvd.sys,
wdmaud.drv,
winhelp.exe,
winhelp32.exe,
wmiprop.dll,
wmnetmgr.dll,
wmplayer.adm,
wmsysPr9.prx,
wsock32.dll
wstdecod.dll
I can't delete wsock32.dll. Can anybody tell me what's wrong?
TIA
You say "it is assumed, that you have no network" but what if you do? Are you able to point out which items you list not to delete in that case? Also you say to remove items from dllcache, so how does one delete items from a .cab file?I want to mention that I came to the same method for doing the tweaks that theob describes and agree it is a good way to proceed.
Most of all I want to chime in with thanks, Serge, for your hard work and helpful results.
Edits: 09/02/11
Don't worry! I made all as Steppe wrote and network works :).
http://cmp2-mihaylov.narod.ru/
You are amazing. Thank you for all your work. Can't wait to get into this part 4. Here is a process I have ginned up for myself. It may be useful to others. It may be obvious but it helps me to take my time getting through the list.
1) Create a backup image of your system...just in case.
2) Print a copy of Serge's list
3) Find out which files your soundcard uses, cross them out on the printed list
4) Get into a search program. I use the default Search command in Windows. Ryelands uses Command line searches.
5) Search for each and every file & either delete or move to a temp file per Serge's recommendation.
6) Have at it.
Thanks again Serge for your efforts!
I will try more and will try to be more cautious. I am now going to try to revive my registry or will have to rebuild the cmp from scratch, a heavy duty but I know now what to avoid in the future. The tweaks I presented here are SAFE and those, that will come with part 5 are ALSO SAFE.
I don't post them now to give folks some time for their machines and systems to settle.
Serge.
Hello, everybody. First let me thank Mr. Mihaylov for pointing out that hdaudio.* files come from Realtek driver pack, so you may possibly not have these files on your system. Also thank you all for encouragement. Folks with asio4all! Take care, I didn’t test these tweaks, but I don’t expect trouble. Just reboot and check more often. ALSO! DO EVERYTHING - the result is very good. When checking after rebooting you don’t have to start cmp every time. If cplay works, it is OK - just click on any *.cue file.
Here is another, and much longer, list of files to remove. As usual, it is assumed, that you have already removed previous 16 or so files, had disabled WFP, disabled or removed all the codeks. For those folks who retained Microsoft PCM converter, I strongly suggest to remove it . It is one of the dirtiest things. Also take a note, that I don’t use my cmp for ripping, don’t have a cd drive in it and have no Internet or any network connection. I transfer ripped music with a flash drive.
A WORD OF WARNING: NEVER TOUCH ksecdd.sys. The system will not boot until you put it back.
Now, if somebody starts experimenting with removing other files or programs, let others know of dangerous files and results. Somebody, please, take trouble to make a list of never-to-touch files!
The files to remove come in a more or less alphabetical order. Most have doubles in dllcache.
The list #2 (most are *dll files, unless specified otherwise). Most are from System32 and Dllcache, doubles are in Dllcache and sometimes in Drivers folders. Some files give immediate and strong effect, removing others GIVE cumulative effect. Note that some files have same names but different extensions (suffixes? Like *.dll and *.sys, or *.drv)
dmloader.dll, dmmime.dll, dmscript.dll, dmsynth.dll, dmusic.dll, dplay.dll, dplayx.dll, dpmodemx.dll, dmstyle.dll, dpnaddr.dll, dpnet.dll, dpnhpast.dll, dpnhupnp.dll, dpnlobby.dll, dpnmodem.dll, dpwsock.dll, dpserial.dll, dpvacm.dll, dpvoice.dll, dpvox.dll, dpwsock.dll, dpwsockx.dll, dsdmo.dll, dsdmoprp.dll, dsound3d.dll, dsound.dll, dxmasf.dll, h323cc.dll, h323msp.dll, hdaprop.dll, HdAudRes.dll, (these 2 are probably also from Realtek), jgaw400.dll, jgmd400.dll, laprxy.dll, ksuser.dll, mciwave.dll, mciseq.dll, mcicda.dll, midimap.dll, mll_hp.dll, mmdrv.dll, mmulse.dll, mspmsnsv.dll, mspmsp.dll,mssap.dll, msscp.dll, mstime.dll, rdpsnd.dll, sbeio.dll, serwvdrv.dll, strmdll.dll, usrvoica.dll, usrvpa.dll, wmvcore.dll, wmstream.dll, wavemsp.dll, wmamod.dll, wmadmoe.dll,wmasf.dll, wmdmlog.dll, wmdmps.dll, wmerrenu.dll, wmerror.dll, wmidx.dll.
AFTER REMOVING THE ABOVE FILES to TMP or any other folder, REMOVE THE FOLLOWING
BE CAREFULL AND REBOOT AFTER 5-8 files, listen to the changes, but don’t jump at conclusions and DON’T STOP (as usually, some files have doubles)
These come mostly from Drivers folder and some have a very strong effect:
amstream.dll, cewmdm.dll, kmixer.sys, redbook.sys, cdaudio.sys, swmidi.sys, sysaudio.sys, stream.sys, wdmaud.sys, splitter.sys, cinemst2.sys, tosdvd.sys, tsbvcap.sys, vdmindvd.sys, wmiprop.dll, wmnetmgr.dll, ksproxy.ax, iac25_32.ax, wdmaud.drv
AFTER PERFORMING the abovementioned tweaks DO THE SIMPLIEST THING : Cut the Windows Media Player folder (It starts to howl, but don’t be scared!) from Windows folder and paste in TMP folder or whatever folder You are using. Reboot. Have fun! Please tell us of Your experience.
Serge.
P.S. Folks! I don’t insist on the order of things, but that is how I've done it. No trouble whatsoever - the system is stable and playing for 3 days brilliantly, with air and ease we have never had in cmp before. Somebody mentioned, that the mids came slightly over the board, they will settle. The backyard will be swept almost totally clean. The same about the attic. Overall sound quality will be up several steps. Funny, but it needs a breaking time of appr. 15-20 minutes. How it can be with software is beyond me.
Edits: 08/23/11
The only file from the above mentioned lists which is used by Juli@ seems to be the wdmaud.drv. This was the only file (according to Me), that needed to be left untouched for Juli@ to work. Since all other files have been deleted, including 'ksuser.dll', & Juli@ continues to work without reporting an error, does it effect the sound in any way??
Has anyone succeeded in removing the Microsoft PCM converter? Im not even able to locate it in the 1st place....
Hi all!
I deleted files which are used (can be used?) by Juli's driver: stream.sys, wdmaud.drv, ksuser.dll, ksproxy.ax.
All while works stably. I use only digital part of Juli and motherboard Gigabyte GA-H55M-UD2H with Intel Core i3-530.
The files stream.sys, ksuser.dll, ksproxy.ax are the part of DirectX file package (http://en.wikipedia.org/wiki/DirectX) and they are not necessary for playback with Juli in ASIO mode as however and remaining files DirectX(imho) which I deleted too (except the file dxapi.sys without which system does not start and the file wsock32.dll without which Everest does not start). But I use cMP2 as streaming player w/o mouse, monitor and keyboards.
The file wdmaud.drv probably is used for audio Windows and it is not required for sound output with Juli in ASIO mode too.
Here the list of deleted files from folder system32:
The files ks.sys, MSPCLOCK.sys, MSPQM.sys, swenum.sys are in the folder Drivers on screenshot (also removed).
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Live, iPad
Edits: 10/14/11
I forgot to mention that after deleting the said files in Ur screenshot I am not able to use the wheel & right-click on the cordless mouse. Before I did the mod, the wheel scrolled down highlighting the next items on the list & clicking the wheel activated it. Now I can only use the left-click alone to point at songs & play by double-clicking them. Which file needs to be replaced to sort this out???
I deleted the above mentioned files in the screenshot excluding the direct x input files which u suggested deleting in the second queue. Have been playing for almost two hours now. The music spectrum seems tilted towards the upper mids & highs. Entire presentation seems very Forward. Though detail has increased, there is no grip in the upper midrange which makes vocals less controlled. The bass & lower mids have gone down two notches. In short it sounds equalised. Does it need to be run-in longer???Edit :- I put back the wdmaud.drv file back in the Drivers folder just to check again. It has cured 90% of the flaws I mentioned above. I guess we have to be extremely focussed while making changes to the system. That file has a definite purpose of being used by Julia, though it runs even without it. The files you have deleted have a profound effect on the performance. Except for the wdmaud.drv file, which if deleted, reverses their effect. Thanks :)
Edits: 09/02/11
Hi Jolida,
First of all, thanks for your list, there is now better left/right balance, more precise image, and a yet tighter bass. And as I was listening to Tweak #4.5 (that's what I call yours) I though to myself, why don't I try your other suggestion, put wdmaud.drv back, after all no harm can be done and I can always remove it if I don't like it.
To put a story short, I think you are right again. Putting it back does make a big difference. The sound is now no longer thin; not only is it fuller and more natural, the soundstage is now more extended to the back, so that it is now square instead of a thin rectangle.
I suggest others to try it to see how you feel.
In fact I was JUST about to post a new topic regarding the same. I have been noticing this since about a week. The bottom end & for that matter the whole Range does not sound FULL without the wdmaud.drv file in it's respective folder. I agree that Juli@ does work without the file as well, but the bottom line is that the end-result sounds incomplete without it. Without the wdmaud.drv file, I feel that, if not full, atleast half an octave sounds missing in the lower registers ( somewhere between the bass/midbass/lower-mids section). The male vocals sound leaner & the impact in bass guitars lacks the Bite which kinda makes the sound, as we could say "light-weighted". Whether Juli@ needs the file to work or not, it sure is needed to make the presentation Complete.
Last night when I was having my listening session, I did one more thing. Serge had suggested that we delete the file named "SOUND" in his 1st list. I put it back in it's respective folder, & heard a wake-up call again. When we make deletions, we need to be absolutely focussed & sure as to how it sounded before & what it's sounding after the deletion. A CHANGE need not always be an improvement. That's what has happened by removing those two files. I suggest u try it as well & post a feedback.
I cannot comment on the improvements of users who use other soundcards apart from Juli@. It's a sure possibility that they too might equally benefit by putting back the wdmaud.drv & SOUND files in their original folders. But as far as Juli@ is concerned, it's a MUST !!!
Hi Jolida,
Thanks for another excellent suggestion! Do note that I am not a Juli@ user. I use ASIO4ALL and the Devilsound dac through an USB isolator with an external battery power supply.
After putting back the sound.drv (lower case in my system) back to System32 and dllcache. I immediately played Van Morrison's 'His Band & The Street Choir'. I always thought that record is the worst recorded of his 3 classics which include "Astral Weeks', and 'Moondance'. This applies to both the LPs (Japanese pressing) and the wave files ripped from Japanese remastered CD. But by putting back the sound.drv file, the sound of that record is no longer a bit recess, but is as immediate as the sound on 'Moondance', one of the most immediate recordings ever! Image my surprise! Everything is now in my room!
A revelation indeed. Everyone MUST give this a try.
Hi Jolida,
Though I'm not a Juli@ user (yet, I now use ASIO4ALL), I had the exact same experience as yours when I did Tweak #3, i.e., tilled high, no bass, recess vocal. It was only after several hours playing in a period of 2 days that this thinness of sound was finally gone and everything became glorious.
Yes it does. But without that file, the music does not seem to be balanced at all (atleast with Julia)...
Perhaps. I did not note it.
http://cmp2-mihaylov.narod.ru/
After deleting the files in Ur screenshot ( except the wdmaud.drv which I mentioned) & running in for a day, it sounds OUTSTANDING. I still wonder what role does that one file play in maintaining tonal balance. Excellent job. Thanks :)
It would be great to have your feedback on the same...
Quote :- "The files stream.sys, ksuser.dll, ksproxy.ax are the part of DirectX file package (http://en.wikipedia.org/wiki/DirectX) and they are not necessary for playback with Juli in ASIO mode as however and remaining files DirectX(imho) which I deleted too (except the file dxapi.sys without which system does not start and the file wsock.dll without which Everest does not start). But I use cMP2 as streaming player w/o mouse, monitor and keyboards."
If i delete ALL the files which you have mentioned (which ofcourse i would love to), & use a wired usb mouse, will it be alright?? I do not use the Zalman case so i have to use a Monitor. Does the list of deleted files in ur post effect the working of the mouse & monitor? If not, i would rightaway go ahead & delete the whole lot...
I have Zalman case, but I find it less and less convenient, so I also use a wheel usb mose and a monitor.
Steppe.
Mouse is corded or cordless ?? Have u tried deleting all files which Mihaylov suggested? He suggests not deleting some files if using a Mouse & a Monitor. I take this opportunity to thank u for sharing Ur ideas on taking the cMp further ahead at software level....
I already answered you here: http://www.audioasylum.com/forums/pcaudio/messages/9/94547.html
http://cmp2-mihaylov.narod.ru/
Are these new files you have found to be sq opportunities if deleted? Are they recommended for those of us with the old E7200 based system?
Perhaps. I am not sure. At least the sound does not worsen ;).
"Are they recommended for those of us with the old E7200 based system?" - I think that systemboard and cpu does not matter here.
http://cmp2-mihaylov.narod.ru/
Excellent. I will try them as well. By the way I use a cordless mouse for my cMP. Will deleting all the above mentioned still get the mouse to work ??
Don't remove the files DirectInput: dinput8.dll, diactfrm.dll, dimap.dll, dinput.dll or try to remove them in second queue.
http://cmp2-mihaylov.narod.ru/
Is this a warning for all or just jolida?
Most likely it is reinsurance. Files DirectInput probably concern only game pointing devices (joystick etc.)
http://cmp2-mihaylov.narod.ru/
when I was using a wireless mouse I had problems.
A sloppy mis-click could result in having to re-boot.
Since doing away with it I have NEVER had to re-boot unless I wanted to.
I wonder if one is better off without the wireless mouse?
I have found I much prefer having the keyboard with an USB extension cord.
Exactly !! U r correct. I am facing such issues with the cordless mouse as well. I dint know it's been an issue with other users as well. So u mean instead we use a corded USB mouse with an extension cable right?? With the cordless mouse, a sloppy mis-click always results in an error reporting "Asio driver not installed" or initialised or something like that, & the only option left would be to re-boot using the restart button on the cMP...
Instead of using the mouse I went to keeping the keyboard close by.
I keep the mouse at the computer so the keyboard is what get used.
Like I said no re-boots since getting rid of the remote mouse.
I did not get any messages when I had a problem,the machine would just freeze and the hourglass would rotate endlessly. I would re-boot and all was fine. Though I did have a few times when I had to clear the BIOS and reset that.
I had made mention of this before but no one else seemed to be having the same problem.
I guess I have about eighteen feet of usb cable between keyboard and machine. I am using a stout linear supply for usb power.
Personally, I much prefer the keyboard. Much easier to switth back and forth the "phase"/polarity to find the best setting
Linear psu for the USB?? but the port will just be used with a mouse/keyboard or an external drive to transfer files right ? How does a linear benefit for USB then??
hdd drives.
The extra power, I thought, might be important with the long cable run.
Also, this was part of a general move to eliminate all switching supplies from the machine.
I should have made that clear.
that I was not having any problems.
I deleted the DIRECTX folder with no problems.
My machine is only using cMP and cPLAY, nothing else is asked of it.
I spoke not about the folder DirectX but about the files in root of the folder system32 and in the subfolder drivers ;).
http://cmp2-mihaylov.narod.ru/
Edits: 09/01/11
Being basically an incompetent computer dilettante I greatly appreciate what the LEADERS have shown us (and continue to show us)
Now to get to that list! (thought I had got off easy with the folder deletion. Oh, well ...)
THANKS!
Do you only use juli@ as a digital out device? Are the other juli@ files only used by juli@ analogue?
Edits: 09/01/11
Yes I use the Juli@'s optical out only, not the analogue. I have removed the analogue board...
But you are using the new i3 platform, Like Rick maybe the new hw doesn't care about these files.
Do u think the i3 has got anything to do with what files juli@ needs??? In what way are they connected ??? i3 is just the Processor. I can understand if u point out at the Motherboard though...
By i3 platform I meant all associated hw. But I still dont understand why even a new mobo would allow juli@ to work w/o some of these files. But I was asking to see if I could delete some more files. Since I am using the old hw I'll probably not try.
I see that U r reluctant to try :) I guess u have nothing to lose though. It's not a system file. Delete it & see. If Juli@ does not work, put it back from where u removed it. I mean cut the file from the directory & paste it on Ur desktop, so that u can put it back if circumstances need u to ....
Last time I deleted all files it was a disaster. Had to short my battery terminals to boot then reload windows via a file image with Snapshot. Not my favorite stuff to do.
Did u manage to do the file deletions as suggested by Serge? What effect did u perceive in terms of tonality?? I recently made linear supplies for Julia & P4 using Elna Silmic 2 caps. The burn-in has crossed 100 hours & still the sound varies drastically each day
Yes I posted my reactions earlier in this thread.
Hi all
first a lot of thanks to Steppe and Mihaylov for the great job done.
I've tried and results seems to me very good at the first listening.
I've checked Essence's driver and none is on the Mihaylov's list.
So I've removed all files of this clear list(.dll,.drv,.sys) and the whole system works well, but I see a lot of boring alerts about the absence of msacm32.drv.
Window alert is opened but OS works well, just a click and alert disappares so I cannot decide if reinstall this file or not. I'll try to test SQ with or without it.
Mihaylov's list is perfect for Asus Xonar Essence soundcard too, this is important and safe.
Why Steppe's list forsees a greater number of files to remove?
Hi!
Look here, please http://www.audioasylum.com/forums/pcaudio/messages/9/94239.html.
The full list of files to remove, proposed by Steppe and ordered me, there on my site in Optimization> Optimizing sound cards> Step 9 (use the button "English" to translate my website from Russian to English).
http://cmp2-mihaylov.narod.ru/
I removed all the listed dll & sys files from the system32 & drivers folder. Restarted the computer & was impressed at the improvement. You have mentioned removing the Microsoft PCM converter. How do I do that?? But i feel that the wdmaud.drv & SOUND files need to remain untouched, to maintain tonal balance...
Also u had mentioned deleting certain entries from the Hkey local machine & Hkey current users in the Registry. I would love to implement that as well. If u could be a little more specefic of which of the keys to be removed from the list. Thanks. And once again, we all appreciate your contribution towards this. Truly awesome....
Edits: 09/08/11
The corrected list of files for removing (system cMP2 Advanced with Juli):Part I.(from folder system32)
dmloader.dll,
dmime.dll,
dmscript.dll,
dmsynth.dll,
dmusic.dll,
dplay.dll,
dplayx.dll,
dpmodemx.dll,
dmstyle.dll,
dpnaddr.dll,
dpnet.dll,
dpnhpast.dll,
dpnhupnp.dll,
dpnlobby.dll,
dpnmodem.dll,
dpwsock.dll,
dpserial.dll,
dpvacm.dll,
dpvoice.dll,
dpvvox.dll,
dpwsockx.dll,
dsdmo.dll,
dsdmoprp.dll,
dsound3d.dll,
dsound.dll,
dxmasf.dll,
h323cc.dll,
h323msp.dll,
hdaprop.dll, HdAudRes.dll (if you installed drivers for Realtek),
jgaw400.dll,
jgmd400.dll,
laprxy.dll,
mciavi.drv
mciseq.drv
mciavi32.dll
mciwave.dll,
mciseq.dll,
mcicda.dll,
midimap.dll,
mll_hp.dll,
mmdrv.dll,
mmutilse.dll,
mspmsnsv.dll,
mspmsp.dll,
mssap.dll,
msscp.dll,
mstime.dll,
rdpsnd.dll,
sbeio.dll,
serwvdrv.dll,
strmdll.dll,
usrvoica.dll,
usrvpa.dll,
wmvcore.dll,
wmstream.dll,
wavemsp.dll,
wmadmod.dll,
wmadmoe.dll,
wmasf.dll,
wmdmlog.dll,
wmdmps.dll,
wmerrenu.dll (or wmerrRUS.dll for russian Windows XP),
wmerror.dll,
wmidx.dllPart II (*.dll and *.ax from folder system32; *.sys from subfolder Drivers)
amstream.dll,
cewmdm.dll,
iac25_32.ax
wmiprop.dll,
wmnetmgr.dll,
cdaudio.sys,
cinemst2.sys,
kmixer.sys,
redbook.sys,
splitter.sys,
swmidi.sys,
sysaudio.sys,
tosdvd.sys,
tsbvcap.sys,
vdmindvd.sys,
wdmaud.sys(The file h323cc.dll is not found. Microsoft NetMeeting uses the file h323cc.dll to handle network data transactions between clients in a network.)
http://cmp2-mihaylov.narod.ru/
Edits: 08/25/11 08/25/11 08/25/11 08/25/11 08/25/11 08/26/11
for posting them. I was going to do it on tuesday-wednesday, after enough time passed to make sure the system is totally stable. There are more very aggressive tweaks also, but more on that later. Especially shrewd of you was to find splitter - the file that breaks wavs into chunks to feed to kernel streaming. Don't know if it was dictated by the times when computers had less memory on board. also attached to it is a datime file which is attempting to attach a clock of its own to any media file.
Please, if you have time take a look at windows time service. What else can be removed.
Serge.
I removed all of the files per Serge's instructions.
I use JULI@ and have had no problems with these files removed.
Maybe I should try to replace them?
I have left off the REALTEK files
Steppe's List Mihaylov's List
amstream.dll, amstream.dll,
audiosrv.dll,
audstub.sys,
avwav.dll,
cdaudio.sys, cdaudio.sys,
cewmdm.dll, cewmdm.dll,
cinemst2.sys, cinemst2.sys,
dmband.dll
dmloader.dll dmloader.dll,
dmmime.dll dmime.dll, Serge had an extra "m" on this one
dmscript.dll, dmscript.dll,
dmstyle.dll, dmstyle.dll,
dmsynth.dll, dmsynth.dll,
dmusic.dll, dmusic.dll,
DMusic.sys,
dplay.dll, dplay.dll,
dplayx.dll, dplayx.dll,
dpmodemx.dll, dpmodemx.dll,
dpnaddr.dll dpnaddr.dll,
dpnet.dll, dpnet.dll,
dpnhpast.dll, dpnhpast.dll,
dpnhupnp.dll, dpnhupnp.dll,
dpnlobby.dll, dpnlobby.dll,
dpnmodem.dll, dpnmodem.dll,
dpserial.dll, dpserial.dll,
dpvacm.dll, dpvacm.dll,
dpvoice.dll, dpvoice.dll,
dpvox.dll, dpvvox.dll,
dpwsock.dll, dpwsock.dll,
dpwsockx.dll dpwsockx.dll,
dsdmo.dll, dsdmo.dll,
dsdmoprp.dll, dsdmoprp.dll,
dsound.dll, dsound.dll,
dsound3d.dll, dsound3d.dll,
dswave.dll,
dxmasf.dll, dxmasf.dll,
h323cc.dll h323cc.dll,
h323msp.dll, h323msp.dll,
hdaprop.dll, hdaprop.dll,
HdAudRes.dll, HdAudRes.dll,
iac25_32.ax, iac25_32.ax
jgaw400.dll, jgaw400.dll,
jgmd400.dll, jgmd400.dll,
kmixer.sys, kmixer.sys,
ksproxy.ax,
ksuser.dll,
laprxy.dll, laprxy.dll,
maadp32.acm,
mcicda.dll, mcicda.dll,
mciseq.dll, mciseq.dll,
mciwave.dll, mciwave.dll,
mciwave.drv
midimap.dll, midimap.dll,
mll_hp.dll, mll_hp.dll,
mmdrv.dll, mmdrv.dll,
MMSYSTEM.DLL,
mmulse.dll, mmutilse.dll,
msacm32.drv,
msadp32.acm,
msaud32.acm,
MSKSSRV.sys,
mspmsnsv.dll, mspmsnsv.dll,
mspmsp.dll, mspmsp.dll,
mssap.dll, mssap.dll,
msscp.dll, msscp.dll,
mstime.dll, mstime.dll,
rdpsnd.dll, rdpsnd.dll,
redbook.sys, redbook.sys,
sbeio.dll, sbeio.dll,
serwvdrv.dll, serwvdrv.dll,
SOUND.DRV ,
splitter.sys splitter.sys,
stream.sys,
strmdll.dll, strmdll.dll,
swmidi.sys, swmidi.sys,
sysaudio.sys, sysaudio.sys,
tosdvd.sys, tosdvd.sys,
tsbvcap.sys, tsbvcap.sys,
usrvoica.dll usrvoica.dll,
usrvpa.dll usrvpa.dll,
vdmindvd.sys vdmindvd.sys,
wavemsp.dll wavemsp.dll,
wdmaud.drv wdmaud.sys
wdmaud.sys wdmaud.sys
wmadmoe.dll wmadmod.dll,
wmamod.dll wmadmoe.dll,
wmasf.dll wmasf.dll,
wmdmlog.dll wmdmlog.dll,
wmdmps.dll wmdmps.dll,
wmerrenu.dll wmerrenu.dll (or wmerrRUS.dll for russian Windows XP),
wmerror.dll wmerror.dll,
wmidx.dll wmidx.dll
wmiprop.dll wmiprop.dll,
wmnetmgr.dll wmnetmgr.dll,
wmstream.dll wmstream.dll,
wmvcore.dll wmvcore.dll,
Any explanations to be given for the differences? I hope my confusion is understandable.
Mihaylov had recommended not removing ksuser.dll, stream.sys, ksproxy.ax and wdmaud.drv yet this last one remains on his list.
It seems it is impossible to separate the lists the way I intended so forgive the confusion from the list!
Thanks,
My list applies only to New very important tweaks part 3. I made a correction in the name of some files. Excludes the files used by driver Juli: ksuser.dll, stream.sys, ksproxy.ax, wdmaud.drv.
http://cmp2-mihaylov.narod.ru/
Edits: 08/26/11
and did not check from which list the "missing" files were from.
stream.sys was on Serge's "third" list but is not on your list. Is this to be added to the list of files JULI@ needs?
As I mentioned, I am using JULI@ and deleted all of the files. All seems to be well. Should I replace the files you recommended not be deleted?
I have figured that if it works it is OK in the digital world and if the files in question were needed my system would not work at all.
But I know nothing, really, about this stuff and would welcome advice as to whether I should replace.
THANKS,
I initially deleted all the files and it worked for a brief time. Then I ran into trouble. Then I put the juli@ files back in and all was well. In this thread steppe gave a 'how to' find what files your soundcard uses.
which is what has me puzzled.
I guess I could try putting them back and see if it makes any difference.
So far, completely stable and sounds quite good to me. Running continuously since the recommendations were posted.
One would think if JULI@ needed the files in question it would not work at all?
that is a puzzle??
I did all the removals except items in \windows\drivers\drivers.cab. I don't have a \system32\dllcache folder. I don't know how to remove individual items from a .cab leaving the rest intact. I believe if I leave wfp off that doesn't matter. Someone tell me if I have that wrong.
Issues: I got an error window starting cMP saying it couldn't initialize something because msacm32.dll was missing. Closing that cMP came up anyway. I found, though, that I could not use notepad.exe 9to edit cuesheets) due to this being missing. I put it back and all was fine, except...
For those with Weiss DACs: my installation bluescreened soon after starting until I restarted with the firewire cable to the dac unplugged and then reinstalled the Weiss driver. Obviously one or more items removed are essential for these dacs. Its probably among the ones starting with midi.. I haven't traced it out yet, but reinstalling makes that unnecessary.
You don't have to touch files from the cabs. They are not extracted, thereforem- harmless.
Serge.
Thanks, What I don't understand is how wfp if on would restore all the items I removed if they had only been present once in system32 or for only a few there plus a cab but nowhere else. I have no dllcache or other store of them to restore from that I can identify.
How did you get wfp to work?
I finally got XPlite to work after yet another try a couple days later. I can't explain why it worked then when it wouldn't on several tries before. Fortunately, I didn't have to attempt that long method in the link posted.
Let me recommend to all an album that sounds just great with the new improvements. Schorry and Carroll's Percussion in Hifi 24/192 version from highdefinitiontapetransfers.com. Not great music but fun and wow sound.
Folks! There may be typos, though I try me best will check and recheck next time. PCM audio converter, (it seems) is removed only after wfp removal. Next time I will post another series of files to remove and my system history.
serge.
P.S. Next list is almost ready. OK, I know I'm walking on thin ice blindfolded, I am scary of next steps each time I reboot, but I can't stop
Is there something else that need to be done?
I can turn it off but I cannot remove it.
Please clarify, when you have the time.
Thanks again. Other than that all is going well.
(There are no conflicts with SRC with all files removed as specified in case you were wondering)
I think. (don't use the codec option is what I clicked upon)
Serge.
no problem on the typos...cant wait for your new list
Checked a moment ago and saw it was still there and will not allow removal.
How does one do this?
Checked as many old posts as my patience would allow and saw nothing.
I doubt I am the only one wondering about this!
Thanks,
If you look at this in control panel and click on properties you will see a 'do not use this program' or something similar. I think you are ok
I had done that.
Thanks for the re-assurance!
Do you have any idea just what is happening from these deletions?
Is this dismantling the controversial kmixer?
Just wondered if there was a motive other than eliminating useless files to these instructions; did you start with a plan to eliminate something in particular?
Thanks, again,
There are several folks who gather here that when I see/read their new posts it gets my audiophile pulse jumping. Included in the 1st tier of impact posters are: Cics, Dave (ryelands), Mark (hfavandepas), Gstew, Jackwong and I'm sure others I can't think of right now. Serge you have now entered this 1st tier. What a wonderful result this is. The air the smoothness is just great. Way to go!!!Few other comments:
..Several files listed in your original post differed by one letter in my Sytem32 folder. They are: dpwsock.dll and dpvox.dll. Are these typos?
..Also since my experience yesterday I have become very sensitive to stuff my juli@ uses like wdmaud.drv but is wdmaud.sys required for the driver to work? I left wdmaud.sys in tact but if anybody knows I would appreciate an answer.
...Several files were not found: dmmine.dll, hdaprop, hdaudres.dll, mmulse.dll. I'm sure you are right about the Realtek comments.
Edits: 08/24/11 08/24/11 09/01/11
Hi Serge,
Thanks for the detail work.
I had taken 2 steps to execute the process.
1. Remove everything in your first paragraph, except ksuser.dll, which had to be reinstall after cPlay complained with an error message (I am an ASIO4ALL user).
2. Everything else.
After step 1, the SQ is improved further along the line of your previous post. The gravity of the mid-range is most noticeable.
After step 2, there is a shift in balance to the top. Yes, there is a general clean up of everything, but the lower-mid is lighter. I had to turn up the bass woofer just so lightly. Will listen more.
However, I have one problem: just how do you remove the Microsoft PCM Converter? The usually route of the Control Panel> Audio Device> Hardware> Codec doesn't allow me to remove it. So how do you do it?
Thanks again for the great effort.
KC
I agree with your sonics assessment but did you eliminate Windows Media Player? When I deleted this it brought back much of the lower mids/bass.
Yes, but not the 20 minutes that steppe indicated, but something like a few hours. May be loud music is also needed, mine happened during Beatles 'White Album', played LOUD.
Before burn-in, no bass, little dynamic, and a recess sound stage. Everything changes after burn-in.
I am now back to 'Tiny' buffer.
KC
Hi theob,
Yes, I had eliminated Windows Media Player (move the Windows Media Player folder from Programs directory to TMP).
Since yesterday, I had switched the cPlay buffer from 'Tiny' to 'Small'. That seems to have restored the mid-range, top-bass gravity.
Will listen more.
KC
> > > Cut the Windows Media Player folder (It starts to howl, but don't be
scared!) from Windows folder..
Isn't the WMP folder in the Program folder?
Hi Serge!
Maybe you still tell us what is configuration of your music computer - the motherboard, processor, sound card, and have You done all the BIOS settings and system optimizations according to cMP2?
The files ksuser.dll, stream.sys, ksproxy.ax, wdmaud.drv that you offer to remove are used by the driver of Juli. So why remove them?
Serge.
http://cmp2-mihaylov.narod.ru/
Thank you Mihaylov that was obviously my problem because I have a juli@ soundcard. Its not Steppe's fault, its mine for not double checking that step from his 1st note.
Edits: 08/23/11 10/27/11
Using the search function with my office computer for a few of these "items to eliminate", on this machine (which likely has little to do with what I will see with the cMP machine) I see lots of versions of these files. Should one eliminate all variants of these files or only those in WINDOWS or only those in WINDOWS\system32?
Thanks, again,
no problem with the music computer
Serge,
You are the keeper of the cMP flame for the moment! We can only hope you come up with more.
Wish I had the time and ability to do what you have done. Since I don't I can only thank you sincerely for your efforts. Based on previous posts I have no doubts this, too, will become part of what I call the cMP canon. All of your recommendations I feel sure will be incorporated into cics's recomendations when he gets a chance to amend his site.
Thanks very much for moving this project forward,
Rick McInnis
ok Serge very exciting changes but I ran into some difficulties. I got through the 1st list rebooted and listened a very nice sq pop...more air smoother sounds a little louder. I got through the 2nd list (not counting windows media player) and again great sound. Then my system became unstable and gave me a terrible sound. So I shut it off rebooted and it would not boot so I jumpered the terminals next to battery and rebooted but when I was back in I found a whole bunch more stuff on my irq list (which I always goto to uninstall stuff from the irq like unused usb ports). Anyway I dont have the time to get back in now and I don't know if I have to raise the clock spd or vcore to get more stable (currently @ 133 host clock control and .944 vcore). So great sound but I have to work through a few things 1st. Oh also some files I could not find, some were off by one letter or so...maybe a typo.
When I reloaded a bios it apparently reset everything to pre optimization levels. To not take chanches I used snapshot to reload Windows too (thank you Dave). Anyway don't know what happened. I'll now try to figure out what I did wrong.
Hello, everybody! Here is a new set of simple actions that’ll bring our cmp-cplay sound to another level.
IT IS ASUMED, that you have already removed all the audio codecs and Windows file protection, and your Windows Shows Hidden Files!!!
1. Create a TMP folder on disk c (if you don’t have one yet), or any other, that you won’t forget about.
2. Inside this folder create subfolder System32
3. Inside System 32 create 2 folders:
Dllcache and Drivers (not case sensitive)
4. Locate (use Search) the following 16 files in your real (not tmp) Windows/system32 folder AND IN SUBFOLDERS (real Dllcache and Drivers, probably in some other windows/ folder)
NOTE: Some files have doubles. NOTE 2: triples and quadruples from *.cab files are unimportant, only real files are.
THE LIST OF 16 FILES:
audiosrv.dll, audstub.sys, avwav.dll, DMusic.sys, dswave.dll, Hdaudbus.sys, Hdaudio.sys, imaadp32.acm, mciwave.drv, MMSYSTEM.DLL, msacm32.drv, msadp32.acm, msaud32.acm, MSKSSRV.sys, SOUND.DRV , dmband.dll
5. Remove (Cut) these files from their native folders to your temp folder and into corresponding subfolders (ALL THE ABOVE STEPS are only a precaution in case you won’t like the tweak and would like to put the files back)
6. Run Settings> Control Panel>
7. Find your soundcard, and check what files for ASIO it uses. (my Lynx l22 uses 2 Microsoft files and 2 Lynx wdm files). Make sure that your soundcard does not use any of the files from the “remove” list. If It does, return it to its place.
8. Reboot, listen, please write back.
NOW SOME EXPLANATION
Up till now we’ve been cleaning Windows to let the pipline Cplay> ASIO> Kernel streaming run unimpeded. Cics made a great job. However, Kernel streaming itself is clogged with attached and unreported services that are like leeches on a mermaid’s ass or like fleas on a golden fleece. These services are from Direct X, Sound services and kernel audio “helpers”. Also there are two DRM descramblers, that I don’t know how to remove yet. By removing these files, some of which are of greater, some of lesser importance, we leave our KS beauty almost totally naked and busy with sound reproduction exclusively. Make these changes, I am sure, You won’t regret. There are some more relatively unimportant files to remove, but I came almost to the limit of my poor speakers’ resolution. So please, folks with really hi res speakers, report. I especially would appreciate feedback from folks with studio monitors.
The fun goes on, it seems.
Serge.
P.S. Tested with 16-44 wavs, not tested with asio4all
You preface with "IT IS ASUMED, that you have already removed all the audio codecs and Windows file protection" but I am unaware of a way to do the latter in sp3. You have many responses but no one asks about this so maybe I am missing something widely known. How do you disable or remove windows file protection when using XP Pro sp3?
maybe I am missing something widely known . . .
And the rest of us could well have it wrong but I suspect that Serge was just being circumspect (which, in this milieu, is welcome).
Whatever, if you have successfully implemented the minlogon change, you have (SP2 or SP3) "removed" WFP. If you haven't, do as you're told at the link - it's the same palaver with either SP and well worth the slight effort in both cases.
Part 1 (removing audio codecs in DevManager) is also trivial to do but IME equally worthwhile.
XPlite and the manual procedure that it automates are only for sp2 not 3. I used the method via safe mode that does not need wfp disabled to install minlogon.
So I am still looking for a way to follow Steppe's 2 with sp3. I have already dont step 1.
XPlite and the manual procedure that it automates are only for sp2 not 3.
Sorry, I wasn't aware that that was the case. But I don't see why you can't just locate and move/delete the files regardless using Explorer and/or the DOS prompt. After all, that's how we get rid of winlogon.exe in the first place . . .
If WFP is off (regardless of how you do it) and the backup files are removed from the dll cache, I don't see your problem. Sorry if I'm being dumb.
Dave
Doing minlogon setup one substitutes one winlogon file for another--so wfp is 'fooled'. Here one removes files without substituting. I did not think that in this case files would be removed from the caches. This doesn't make a lot of sense to me that it would work this way because if wfp reformulates what files to maintain at each startup how does it know to replace missing files? That is, I would expect it to replace what I removed rather than eliminate what I removed from caches. Do you KNOW otherwise from experience?
I would expect it to replace what I removed rather than eliminate what I removed from caches.
Yes, of course. That's why you have to delete any backup file from the cache as well as the working one from \system32.
Hence my earlier suggestion that you use the /s switch with the DIR command - you see where every copy of the file is located. As both \drivers and \dllcache are sub-folders of \system32, it's easy.
Do you KNOW otherwise from experience?
Well, yes, I think so (he said, blinking in the glare of the bright lamp shining in his eyes). OTOH, I can't see the roof falling in if you just give it a try. It's mostly how I got my experience, such as it is.
As ever, be sure to have to hand an up-to-date backup of the OS partition before you do anything like this.
D
Oh, I took your recommendation to mean that wfp would remove the files from the caches if I removed them from \system32 while it was off. I will try to remove them from all places in safe mode and restart and see what happens. I still wonder why wfp wouldn't call foul in that case causing error or startup failure: does it really only know which files to maintain by what is in the cache and not from some standard list it references?And the bright light in your eyes is to help you listen objectively, i.e., double-blind.
Edits: 08/22/11
I still wonder why wfp wouldn't call foul . . .
So do I. BTW, I see that litepc.com claims that XPLite "Supports . . . all XP service packs including SP3!"
does it really only know which files to maintain by what is in the cache
No idea. Why not ensure you have (and can use) an OS backup and just give it a go?
Best
Dave
at least on my cMP2 installation.
I have removed all the files Steppe listed except the duplicates in .cab files which I don't know how to cut from the .cab. The system only allows extraction of these which I believe leaves the driver in the cab and just allows a copy to be made. What I don't know yet is whether wfp can only replace these or also ones that were only in \system32 but not in a cache (at least that showed on a search).
I do believe I hear an improvement right off, but I need more listening as I first put on a great 24/192 album that sounds super no matter what.
my machine is SP2, not SP3 with WFP removed.
I do not remember hearing that WFP could not be removed from SP3 XP. I could have missed it.
I have gone back to check and the files are still "gone".
Not responding to RYELANDS, thought it best to put my comment in the next logical spot.
on my machine
Hdaudbus.sys
Hdaudio.sys
What am I doing wrong?
SEARCH finds the files on the computer I am using to write this note.
Could it have been removed earlier somehow? Some previous tweak?
Thanks,
I can't find Hdaudio.sys neither.
However I discovered, after numerous reboots and by the process of elimination, Hdaudbus.sys is the only file out of steppe's list that I need. Without it, there was no sound. I think this might have to do with the fact that I am an ASIO4ALL user.
Thanks again steppe for a most significant tweak. There is so much improvement in the middle range that the human voice is now much more realistic.
KC
I cannot find them either.
I understand that this driver High Definition Audio. If you have not installed the drivers for integrated sound card then you will not find these files.
http://cmp2-mihaylov.narod.ru/
d
No one here remembers the bending of our minds
Apart from the two HD files, mine couldn't find avwav.dll either.
Mine was in the system32/dllcache folder.
I wonder what it is for.
No one here remembers the bending of our minds
I wonder what it is for.The dll cache is the place from which WFP takes a replacement file in the event of one of its brood being corrupted. With WFP disabled, it has no role. I've no idea why you don't have the folder but, as it can safely be deleted once WFP is off, I wouldn't toss and turn through a sleepless night worrying about it.
+++
Edit: You do have the folder. I miss-read your note - sorry. Now go back to sleep.
+++No two Windoze installations are ever going to have the same file library. If this or that (or even these or those) files are present on Machine A but not Machine B, again, don't worry about it. At least three quarters of the files on Serge's list are not on my machine as I've been doing a similar purge though not as thorough and for a different purpose.
Whatever, he has passed on a list of files that he was able to remove and get better sound as a result. If you want to do the same, remember that what works for him might not work for you and that, while it might be safe to delete them, their absence may well make no difference to your setup. On the other hand, you won't miss them.
The mantra is "Go carefully and backup regularly". It's not a bad idea to make notes of what you delete as you go along.
Edits: 08/23/11
Hey Rye,
Wow, an inmate who actually reads and re-reads posts they are responding to. Dont make a habit of that or you will not be welcomed here :).
Before I go back to sleep, I remember on my last cmp2 build that I was using the directions off the cicsmemoryplayer.com site and there was a step missing that I remembered doing.
In the Autoruns section there is no mention of deleting the kmixer, dmusic, etc. Though on the directions that you gave me years ago, there is a whole section on that.
Was this an oversite when the site was built, or a change to the "recipe"?
I did the steps and cant see why they wouldnt be a good thing to do. Perhaps some inmates are missing out?
No one here remembers the bending of our minds
Was this an oversite when the site was built, or a change to the "recipe"?
Good point. I suspect it was an error, probably mine way back when I edited the copy a couple of years ago. What we can do to correct it now, I don't know as it's out of my hands.
Do you still have a copy of those "directions"? If so, any chance you could send them to me?
Yikes!!
I looked on the current pc and did not find any such doc. Then I looked through e-mail and besides remembering some of the great ones I found every attachment you sent but that one.
1st off any ideas what it would be called??
I do remember a time when my computer died and I rushed my cmp2 rig into service. There was a lapse I think where some e-mails might not have made it from one computer to the other. The wife has taken time off to be with me this week so give me some time to figure it all out and see if I can find it.
I find the doc I have better than the printouts from the site.
Anyhow I'll just e-mail you when I have a solid answer but you know how long those religious services last...
Oh, and one of those e-mails talked about having higher latency on the fit pc, but better sound, concluding that the dpc latency checker is an indicator of OS slimming but not necessarily quality. I wish I had read that last week. My headphone cmp2 has been plagued with high latency 700 or higher no matter what I do. It does sound pretty good though. I think it is a bad mobo so i ordered a new one. Headphones blow so if I can not spend money on it that would be a good thing...
No one here remembers the bending of our minds
Hi steppe,
Thanks again for another great tweak.
I am also 16-44, but use ASIO4ALL and Windows XP Home Edition.
I removed:
from Windows\System32: msadp32.acm, msaud32.acm, and sound.drv (yes, lower case)
from Windows\System32\dllcache: msacm32.drv, and dmband.dll, sound.drv (yes, again)
and from Windows\System32\Drivers: MSKSSRV.sys
The improvement is systemic and therefore difficult to describe. But the sound is more precise and I am able to listen into the music more. The high is less shrill. Detail more apparent.
Thanks again.
KC
throughout the spectrum in my system.
I do not think I am hearing much different in the high frequencies.
Very worthwhile improvement. Subtle initially but the longer you listen the more you appreciate what it does.
Thanks very much, Serge,
Hi SergeI'm working through your new hit list of files to disable, and I discovered MSACM.DLL & MSACM32.DLL in my system32 folder... but no instruction from you to deal with them.
Is this something you tried already, or another two candidates for exclusion ?
BTW, thank you so much for starting us down this road of improvement !
Cheers,
GrantLATER ON: haw-haw... answered my own question. In my system, Windows complains on bootup without MSACM32.DLL (the other one seems of no consequence, so I left it disabled)
EARLY LISTENING: I agree with other posters here - nice uplift to SQ in my room too. Excellent results from your research Serge ! Can there be any more left to uncover ???
That's not a Toy... IT'S A TOOL !!
Edits: 08/17/11 08/17/11
Didn't come to it yet
I discovered MSACM.DLL & MSACM32.DLL in my system32 folder . . .As Serge noted, "Also there are two DRM descramblers, that I don't know how to remove yet".
If he means (as I'm sure he does) MSACM.DLL & MSACM32.DLL, neither do I. If you do delete them, Windows will almost certainly fail on boot - the Internet is awash with scary warnings about this. (Ever curious, I renamed msacm.dll and couldn't access my headless box on restart. I switched it back and all was OK.)
Just for the record, I've now deleted the listed files on two cMP^2 boxes and, yes, the SQ improvement is very marked in both cases. This is a VERY worthwhile "tweak". Good work, Serge!
Edit : On my cMP^2 systems, the "Search" function doesn't work because something or other is missing. If you get the same, open a Command Prompt window in \system32 and type Dir audiosrv.dll /s to find all instances of audiosrv.dll (/s searches sub-folders) then delete them. Repeat this for each file - very quick.
Edits: 08/18/11
even on my office computer where nothing has been removed from WINDOWS, SEARCH will only find the basic files where COMMAND PROMPT finds the "other" files.
Whether this makes any difference sonically I cannot say since I just discovered this new tweak today. Can't wait to get home and give it a try.
Serge/STEPPE is the man of the season on the FORUM this summer of 2011.
Thanks for letting us know the results of your inquisitive mind!!!
The fun goes on, it seems.
Indeed it does. I haven't done exactly as you suggest but what I have done is most of the general slimming down of XP described in detail at the link below. I was doing this with an eye to fitting the OS onto (and booting from) a 1GB SD using a SD/SATA adapter with an Intel Atom mobo. What I wasn't prepared for was the dramatic difference in SQ that followed.
To what extent I have more or less inadvertently done the same as you I can't really say though I'll take a look shortly and report back. Nor can I say whether doing the exercise would make any difference on another system as I've only tried it on the one so far.
One thing I can say is that it's time consuming to do, especially Part 2 (Disabled Services) which needs also to be done with care. I've skipped most of Part 2 for the moment following repeated lock outs though I will go back to it. I'd urge that you don't even think of trying it if you don't have access to and are familiar with a good "imagefile" backup utility and that you at least think twice about trying it if you don't have a docking station.
I was planning eventually to report what I'd found but I've been sidetracked by other issues. In short, however, on the basis of my experience, I wouldn't be at all surprised if you're on to something. Thanks!
Best
Dave
Dave,
Very interesting link. I'm warming to the idea of taking Windows down another notch or two... Serge showed the first Steppe (PI), this looks like a good 'nuther hundred steps or so.
I take it from your comments you made the changes manually. In the interest of saving time (a commodity I don't have a lot to spare right now), I'm thinking the easiest way to do this is to:
1. "Donate" for his batch files.
2. Go through the batch files with the guide & comment out the deletes that look like they'll stop my cMP from running.
3. Apply them on a test HDD. Then see if it runs still.
4. If no-runnie, back to #2 and repeat as needed. If it runs, copy the system to the SDD & give it a listen.
I didn't see any cautions about using his batch files after a quick 'Google'. Any thoughts?
Greg in Mississippi
P.S. Serge, thanks for doing all this research and sharing your findings. After getting into the cMP groove a few years back, I've gained the opinion that the more un-needed processing (software or hardware) you cut out of the system, the better it will sound. As a matter of course, I removed the codecs when setting up my cMP when doing "Step 3 – disable unused hardware" & "Step 4 – disable unused audio functions" under the Soundcard setup section. But I never listened to see if that made a difference. Now I know it does! You, Mark, and Dave are doing great things to extend our cMP setup knowledge!
Everything matters!
I'm warming to the idea of taking Windows down another notch or two.
Me too. Many thanks for your interest.
The first point I'd make is (a) that what I call "steppe 2" (Serge's codec deletions) makes a significant difference to SQ - it jumps at you - and (b) that "a leech on a mermaid's ass" just has to be the Metaphor of the Year!
The second is that I'm not suggesting that "slimming" Win XP makes for anything like such a marked a difference. I'm wondering if it's not my imagination. It needs more testing. I reported a SQ improvement the first time I tried it - and there certainly was one in my system - but I'm not now absolutely sure it was "the slimming wot done it". This is partly because I can't readily think of a reason why it should and explains why I'm hoping others will try it.
"Donate" for his batch files.
I've been working my way through the ten "Chapters" of the web site and writing my own batch files to do the various deletions. I certainly don't grudge the chap his $15 - if this works out we should IMHO pay him both verbal and more useful tribute - but it soon became clear to me that the scenario he was designing for and the cMP setup are very different and that what he generally recommends and what would work for us are not the same. There is, e.g., a fair bit that he urges his readers to keep that I've deleted without problems and vice versa .
I've had some interesting side effects. As you know, I use a headless setup and for a long while had an odd bug whereby some (though not all) flac-format data (though not wav) would, if running with AWE enabled while pulling data over a LAN, cause certain albums to freeze on launch. It seems (though it might not seem so in the morning) that this irritating bug has been cured because of the files I've deleted. Time will tell but what pleases me is that the system is definitely more responsive than it was. I have no idea why.
Another odd effect is that I now think that playing with AWE enabled is less pleasant to listen to than playing with it off. It's certainly different and I'm not sure which I prefer.
Apply them on a test HDD. Then see if it runs still.
Already done! I've worked my through the text on the web site chapter by chapter to create files that do nearly all the proposed deletions while omitting those that gut feel suggested I shouldn't, at least on a first pass. It was teeeeeedious.
If no-runnie, back to #2 and repeat as needed. If it runs, copy the system to the SDD & give it a listen.
I tested both boot and play at regular intervals as I assembled the batch files. As it goes, this was easy to do with a headless setup as I'd write the file on a desktop then run it on the target without having to boot the latter into XP Mode. I'd just launch Explorer from the cMP GUI, locate and run the batch file, re-boot (to be sure it would still boot) and test play. I reckon I've now done about 90 per cent of the chap's suggestions and have yet to have a lockout. So far, all-runnie. It won't last.
So, if you can bear with me for a day or so while I complete and test the files, I'll happily pass them on to anyone who asks for them c/w some rough and ready notes, screenshots, the usual disclaimers and so on.
As of now, I've taken the OS partition down from something just over 1 GB to under 400 MB (making it feasible to boot from a 1GB SD c/w SATA adapter), fixed my flac bug and possibly (though not definitely) improved cMP^2's SQ.
+++
In short, it's a case of "what are you doing but lure people into time consuming and sometimes useless software ''so called ops' ' w/o thinking and trial?" Shame on you, DR.
Ok Dave please do share which 'file deletes' you did successfully. I have read the link several times and conclude I cannot do any of these w/o some assistance.
please do share which 'file deletes' you did successfully.
I did pretty much the whole caboodle except for Chapter 2 which covers the files that perform XP Services you don't use. I hit problems with that bit and have yet to go back to it. The rest was fine.
As I say, I was aiming only to reduce the installed size of the OS - the SQ change was unexpected and not IMHO as marked as that obtained from "steppe 2". But still worth doing.
Just sit with the instructions in front of you and delete the stuff it tells you to using a bit of common sense as to what to skip.
It's not exciting - but it's not difficult either. Just be sure to proceed slowly and make regular tested backups (so you don't have to go back to Square One when you mess up).
HTH
Its great to see you continously tweaking Windows. I have one question before I try this. You said '... Find your soundcard, and check what files for ASIO it uses. (my Lynx l22 uses 2 Microsoft files and 2 Lynx wdm files). Make sure that your soundcard does not use any of the files from the "remove" list. If It does, return it to its place...'
How exactly do you do that?
Hello.Theo!
Run> Control Panel> Sounds and Audiodevices> hardware> (find your - is it Juli@,-then go > drivers...)I recollect it from memory sitting in front of a russian version of Windows Vista, so can't say exactly right now, but I am sure you will find it once you open your souncard's dialod. I am sure, by the way that no real asio soundcard uses these files.
Try, Theo. It was a real earopener. The mids are delicious. You are one of those few of us who had done all cics's tweaks, so I'm sure You will hear the difference.
Serge.
Ok I had found it before your post and removed all the files I could find. So far yes more transparent. Sounds like more volume. I will post long term impressions after I listen for awhile
More listening impressions after 1 hour of listening. More highs & air and all but I found it bit overdone so I backed off on my panel crossover a bit. Once this adjustment was made I was ok with it and yes more air, more mids/upper mids definition. It is definitely more transparent in the positive sense of the word. More to come.
Ok I finally found the 3 files I missed 1st time through. It sounds better with remaining files deleted. More transparency in mids/highs. Purer sound is the best way to describe it. Definitely a keeper.
Dave I don't know how to respond to your post. I went through the link you provided but there is lot there I don't understand, Will study some more.
Hi all. Can you help me?
Where can I find CPlay sources?
Can I remote control without wireless mouse, through sockets or other?
Thanks.
Hi, everybody! I have some "error correction" concerning the tweaks that I proposed. First. I noted that high frequencies, though free at last came with some excessive sheen, This effect passes as soon as You (and I) put the external esata hdd on a different power line. Second To achieve the best results with the aforementioned tweaks, You don't need to edit the registry.
Now some reports. I recently managed to compare cmp with 2 nice cd players.
First was an Ayre (about $3500 in Russia), another - Gryphon Mikado (12 000 euro). In both cases cmp (with Lynx sound card) was connected with balanced cable to Gryphon integrated using Lynx's analog outs - to Tannoy and PMC speakers.
Ayre cd vs cmp was a very simple and easy case. Ayre was eaten raw without looking back. Nothing, - details, resolution, tonal truth, pace... was on Ayre's side.
With Gryphon Mikado it wasn't that easy. You cannot accuse both Lynx and Gryphon of being inattentive to analog outs, and this is where the competition really came hot. Whenever cmp with Lynx was surprising with more and more details, Gryphon would render them easily though tonally darker. Where cmp was delivering a wider stage, its opponent would deliver more depth. In short, they came "nose to nose" as we say in Russia. I cannot say that any one of them was a clear winner. The general agreement with a lot of comments was that Gryphon was more delicate and subtle, more dark. Cmp was a trifle more in-your-face, more raw and closer (with rock) to life. The resolution, displayed by both players, was very good to excellent but not extreme or "absolute". Next would come reports with Accuphase dp 67 and (if I manage to settle it with the owner) CEC zero smth transport paired with Audio Note Dac.
Wait,if interested.
Serge.
P.S. I read that many people here haven't examined the basic info concerning the Build of CMP.If You did, many questions wouldn't surface at all. At the same time, I'm glad that more people join the project.
I've follow your advice about Audio codec (not registry) but in my setting I didn't notice anything different in SQ.
It seems strange to me that your cMP2 could sound how you describe, but I use always to upsampling at 24/192. With this setting I find a huge difference in SQ.
I'm curious too about your cMP2 setting & system.
Daniele
I shal post exact info of my cmp next time (I am at my friend's working computer now) As I have smth much more important to post now. Read highier
Serge.
I second Mihaylov.
Were you comparing 16/44 files, one from CD and one from HD (cMP)?
Or were you comparing one from CD and one upsampled to 88 or 176 kHz?
Hi! We compared cd to a 16-44 wav file, ripped with eac from this very cd. No upsampling for me.
Hi Serge!
Could you give me the exact configuration of cmp2 and your audio system?
http://cmp2-mihaylov.narod.ru/
Hi! My system or the system cmp was tested in?
Both.
http://cmp2-mihaylov.narod.ru/
is it possible to use more than 1 GB of RAM ...?
I have 2 modules of 2 GB that would slip in , seems like more is better
thanks
Yes you can, but 2G unit needs more electric power consuption ( aka more rumors=more jitter), so SQ could be less than using a 1G unit.
Everything is important in a cMP2 system if you want reach the best result
Daniele
Why don't you try and let's know the result?
But I had tried as single 1Gb and a 512Mb, and the latter is sonically superior.
Hello, everybody!
As I almost despaired to see 2.b 40 I decided to explore what else can be done to improve the sound. I now own Lynx l22 and I noticed that its wdm driver is only 1 of several. Renaming them one by one proved that they are all necessary and that asio is not independent of windows kernel streaming. So I thought I may as well try to do smth about it. But can I, really? I read again AOBCT and found that nothing had been told about different (about 8) sound codecs that are on the system. I removed all of them, including ms pcm converter, some fraunhoffer, truespeech and others. I was heavily under belgian monastic beer and decided to go on. Next thing, I went run-regedit-HKLM and Attention, very important! not only HKLM current machine but also current user and REMOVED FROM THE REGISTRY:
all adobe entries
almost all microsoft entries in SOFTWARE
a lot of other entries from UNKNOWN.
Another entry was from NETSCAPE. But I left it, though it was very tempting to remove it. Please mind that I did it in local machine AND current user.
I rebooted my machine. The sound was so astonishingly different that first I thought I spoiled everything. 5-10 minutes later I realized I have made a change that was much STRONGER than Minlogon. In fact stronger than anything. The highs got so airy and detailed (in fact uleashed at last completely), the scene so deep, high and multilayered, I hardly find words now. Bass extravagantly defined. I call it luck of a drunk. So I suggest that folks with deeper knowledge than mine would experiment and have their say. BUT BE CAREFUL. After my blind tweaking I don't see any hardware in control panel. No devices at all. Those who are brave and wise enough, are welcome to experiment, but please report. I am listening to cmp-cplay for several days and see that a lot is yet to be done with cplay soundwise if my tweaks were directed correctly. Do these tweaks and You will see what I mean.
If You have done this (removing codecs) and I merely missed smth., I apologize for bothering you all.
Also a question for help: What is the exact syntax of bootlog file with HPET/97?? on. Does it work in XP?
Serge.
P.S. Glad to write to You all after such a loooooong pause.
Hello, everybody! I more or less managed to reconstruct what I did.
First I DELETED all codecs, including Microsoft PCM converter. IT DOES MOST TWEAKING
As for the registry, I still remember it vague, but I recollect now that two entries were from antivirus software, one unknown (and very suspicious), one from anti malware bytes smth (hidden worm possibly) and several microsoft entries. I didn't touch netscape entry, though it was tempting.
MOST IMPORTANT I reconstructed the tweak with codecs on my wife's machine
1 I installed Asio-for-all on her comp to drive onboard azahlia soundcard.
2. I installed cPlay 2-39 sse2 on her machine
3 I removed joystics, mixers and all the rest
4 I set ALL CODECS to DO NOT USE. Everything worked fine. The effect is very strong.
5 I started swithing on MS codecs one by one untill the sound came back to initial quality. TWO are most important to remove - Windows media audio AND PCM converter (on both my machines it went off easily).
6 I think registry tweaking is relatively unimportant now.
Now some speculations and reflections. I noticed that HF are a little bit TOO PROMINENT, so I suppose that Cics was "voicing" the player subconsciously to overcome codecs' bottleneck. Now, when there is no bottleneck, the highs are a little bit too loud and somewhat grainy, maybe too "eager". Also I think that our beloved Secretly Rabid Code is probably using pcm converter from Microsoft or touches windows media audio somehow.
I urgently suggest that folks with studio monitors would try codecs tweak and report on the difference.
Also I suspect on the edge of my hearing an increase in hf hiss at higher volume. (Please mind that I use now very poor speakers).
It is getting too long, but I will tell You, that I have compared untweaked And tweaked cmp-cplay with two cd players : one a $3500 Ayre player, another - 12000 euro Gryphon Mikado. The report will surprise You, but leave till my next post, where I will report also of our player compared to my former Accuphase setup.
Happy tweaking.
Serge.
Hi Serge,
Off board Theo and I already communicated about the SQ changes.
See here after.
I only disabled the sound codecs.
I did not delete them.
Reading other posts in this thread,
I think in general, everyone is hearing more or less the same type of SQ changes.
Thank you for this tweak
Mark
> > > > > Hi Theo,
> > > > > I'm right at it.
> > > > > First impressions: indeed a nice SQ pop.
> > > > > There are more micro details and also a somewhat better stereo image.
> > > > >
> > > > > But also some HF brightness added.
> > > > > (may be I just have too get used too the new sound)
> > > > >
> > > > > I think I like it.
> > > > > Especially because the combination of ferrites + caps 'softens' the
> > > > > extra HF
> > > > > brightness
> > > > >
> > > > > Mark
fully optimized cMP2 PC -> ESI Juli@ -> Van den Hul Optocoupler MkII-> Lavry Black DA10 -> XLR Mogami Gold -> Klein & Hummel O300
Hello All:
Started using cPlay and I really love the sound. Is there any helper programs to build cue sheets with the direct paths built in? I know I can edit the text file, but has anybody come up with a better way? Thanks,
Kevin
Yes I tried the Alan Jordan download. Installs fine and brings up all of my music tracks. I specify the root dir to build a cue file in, select a few tracks and yes...it builds the cue file. However to does not work with cPlay, the error I get is that FLAC file can't be found (because its looking for the file in the root dir, not the sub dir). Its possible I don't have it configured correctly...but their don't seem like a lot of options to play with.
Kevin
nt
fully optimized cMP2 PC -> ESI Juli@ -> Van den Hul Optocoupler MkII-> Lavry Black DA10 -> XLR Mogami Gold -> Klein & Hummel O300
I more or less managed to reconstruct what I did.
and I've managed to reproduce your results . . . while stone-cold sober, even.
I deleted all the Audio codecs except for the Windows PCM converter which XP wouldn't allow me to remove. However, I could disable it so I did.
I repeated this for Legacy Audio Drivers, Legacy Video Capture Devices and Media Control Devices. I also "removed" the Video codecs but they reappeared on re-boot so I deleted the .dll files instead. (Whether these matter in the slightest, I have no idea but removing them did no harm.)
I've now listened to the change using familiar material over two lengthy sessions and during casual listening and am firmly convinced that the change is well worth making. (I don't find the treble excessive using what I'd term mid-range kit though I'm not saying you're wrong either.)
Many thanks!
Dave
Hi Ryelands,The deleted dll-file most likely will re-appear through the actions of the windows file protection system. When a vital OS file is deleted, the Windows file protection systems normally will detect this and will try too re-install a copy from the dllcache-directory. See :\windows\system32\dllcache
So don’t forget too delete these file from the dllcache too. But probably than the windows file protection system will ask you for the CD with the original Windows XP installation files.
Mark
NB
This Windows file protection system can nicely be misused when applying the minlogon.
Rename the minlogon.exe too winlogon.exe.
Replace the original winlogon.exe in the dllcache with the renamed ‘fake winlogon.exe’-file into the dllcache.
Delete the winlogon.exe (that XP actually uses) in the :\windows\system32 directory.
The windows file protection system will detect this and will now look in the dllcache for a backup copy.
The windows file protection system will now install the ‘fake winlogon.exe’ for you from the dllcache.
Just as you want this too happen.
:-)
(don't forget too apply the needed registry edit. The windows file protection system will not do this)
fully optimized cMP2 PC -> ESI Juli@ -> Van den Hul Optocoupler MkII-> Lavry Black DA10 -> XLR Mogami Gold -> Klein & Hummel O300
Edits: 07/08/11
just curious.
nt
fully optimized cMP2 PC -> ESI Juli@ -> Van den Hul Optocoupler MkII-> Lavry Black DA10 -> XLR Mogami Gold -> Klein & Hummel O300
Hi, WFP can be completely disabled. Read carefull about minlogon tweak. There is a try-out progran that does it
.serge
see the minlog section
No one here remembers the bending of our minds
Hi Dawnrazor,Thankx for the link on how too disable the WFP feature manually.
But I think using ‘step 2’ in the minlogon implementation is much work.
Why going through the all that trouble of disabling the WFP feature if one can let it do the job for you.
Simply copy the ‘fake winlogon.exe’ in too the dllcache directory, etc.Mark
fully optimized cMP2 PC -> ESI Juli@ -> Van den Hul Optocoupler MkII-> Lavry Black DA10 -> XLR Mogami Gold -> Klein & Hummel O300
Edits: 07/09/11
Well I THINK that windows is smarter than that. If you dont disable WFP then windows will just overwrite the fake file.
Though it is easy to tell. If you load differently after making the switch and you continue to load differently and faster then perhaps you are right. Though if it is so why would someone as savvy as Cics do extra steps, and why wouldnt someone have brought it up before. Like segers who found the simple method, etc.
Anyhow if Dawnrazor can do it, everyone should be able to cope.
No one here remembers the bending of our minds
Hi Dawnrazor,
It’s indeed very easy too tell, in various ways, if the switch to minlogon was successful or not. There is really no need for such patronizing post.
I would have liked you first studied how the WFP feature actually works. Especially from what directory the WFP retrieves the backup copies.
Mark
fully optimized cMP2 PC -> ESI Juli@ -> Van den Hul Optocoupler MkII-> Lavry Black DA10 -> XLR Mogami Gold -> Klein & Hummel O300
Hi Mark,
Sorry. I was shocked at your reply. I never intended to be patronizing. Didnt think I was being patronizing.
Sorry it came off that way.
No one here remembers the bending of our minds
Hi Dawnrazor,
I overreacted.
I should have lifted the thread back too the content-level and should have stuck too the WFP topic.
I instead I continued the thread on the level about persons.
My sincere apologies for that.
Mark
fully optimized cMP2 PC -> ESI Juli@ -> Van den Hul Optocoupler MkII-> Lavry Black DA10 -> XLR Mogami Gold -> Klein & Hummel O300
Hi Mark,
No worries. I have zero issues with you personally or any other way for that matter.
In fact I learn a bunch from you.
Again sorry for the way I wrote that post.
No one here remembers the bending of our minds
Please!
One does wonder how much effect it would have but still there are few things more satisfying than deleting stuff out of XP.
At first I thought maybe I should not do the video stuff (thinking graphics in my head) but all the codec deletions you recommend (audio & video)are good ones and I think they do a small sq improvement. Certainly not as much as the audio codecs but worthwhile indeed.
Edits: 07/07/11 07/07/11
but I do think I am hearing a general improvement.
There is better cohesion of the image. I hear better center fill. Yes, the high frequencies do seem to have less grain. But, I could be kidding myself.
I figure if I think it is better then it is better.
I would really like to try the registry tricks. One is nervous that when you go too far you have to start over again.
Dear Steppe, would you list what you think are good guidelines as to how you approached WHAT to disable? Not asking you to spell it all out.
Also, how many items do you think you disabled?
Thanks for you intrepid nature and your report.
Just when you think the thread is down for the count something like this comes up. No question the onus is on us to keep this vital. I wish I had something to offer other than encouragement.
Thanks, again,
Rick McInnis
Rick I personally wont go into the registry w/o a specfific list of stuff to delete. Too much risk for a won't boot issue for me. If you do it please post exactly what you do so other 'registry unfriendly' folks like me can follow.
Only did some of it (wanted to ease into it) so I checked do not use the audio codecs (instead of removing) and Holy Cow!! Wow! Nice tweak let me buy you a beer Serge!
Thanks for trying, Theo! It is settling now and the initial grain in HF is coming away. Psycho trick?
Serge.
Many thanks for your report.
In Device Manager > Sound, video & game controllers, I removed every video and audio device except for the Microsoft PCM Converter (which can't be removed) and, of course, the driver for the DAC. I agree - there does seem to be a worthwhile improvement.
As soon as I get some time, I'll reverse the changes once or twice to try to be more certain. (I'll also take a look at your suggested Reg Edits.)
It is settling now and the initial grain in HF is coming away. Psycho trick?
Most likely but, if you'd like to wager on it . . . :> )
Dave
Did you compare 'Do not Use' with 'Remove' audio codecs? I did Do Not Use and I get the best highs I have ever had. Dynamics are greatly improved too. Yes it starts out a bit grainly but quickly warms up to bliss.
Though I had not go back and forth between the 2 options, 'Do Not Use' vs 'Remove', (I first tried 'Do Not Use' listened for a couple of hours and then 'Remove'); I much prefer 'Remove'.
I see ten of them on the machine I am using, not my cMP machine. Haven't looked there yet.
Goto CONTROL PANEL/SOUNDS AND AUDIO DEVICES/HARDWARE/select AUDIO CODECS/PROPERTIES/select the codec you want to delete/select DO NOT USE THIS AUDIO CODEC.
One wonders if there would be advantages to removing even more of this stuff?
VIDEO CODECS?
MEDIA CONTROL DEVICES?
One would worry if you could no longer make changes to HARDWAARE in CONTROL PANEL.
Is "current machine" LOCAL MACHINE?
Very interesting.
Have you made a list of which items you have "removed" from the registry?
Is there a way to do this so that one does not have to modify EVERY SINGLE THING? Could take longer than "tuning" XP!!!
Now to try out no audio codecs!!!
Thanks,
Hello! I think I removed media control devices as well, because YES, I don't see any hardware in control panel. Slightly ovedid, I guess.
Next, local machine And CURRENT USER. Read my answer to Theo's questions. I am ready to send my registry for inspection, only I don't know which file(s) contain it. I am not boasting, I ask for help from computer specialists who would wish to follow not my steps exactly, but this registry-editing direction in a sensible way so that others may repeat it safely.
I removed all the codecs. As I understand, while they are on the system, they are permanently attached to kernel streaming drivers (2 of ks*.sys files there, it seems) and produce a lot of havoc.
Thanks for Your interest.
Serge.
Great to hear from you indeed. Just a few questions because I would like to retrace your steps.'...about different (about 8) sound codecs that are on the system.I removed all of them, including ms pcm converter, some fraunhoffer, truespeech and others...'
Where did you do this? Was it under device Mgr? Did you remove all these codecs or just Microsoft codecs?
'...I went run-regedit-HKLM and Attention, very important! not only HKLM current machine but also current user and REMOVED FROM THE REGISTRY:
all adobe entries almost all microsoft entries in SOFTWARE a lot of other entries from UNKNOWN...'So you goto Regedit HKLM and remove ms software entries plus others. Did you set their values to 0? If not how did you remove...just delete? Also you said you removed almost all microsoft entries. Which ones did you not remove? Which specific entries did you delete from UNKNOWN?
Welcome back Serge! Sometimes alcohol can be inspiring!
Edits: 07/04/11 07/04/11
Hi, Theo! Codecs are Control panel-sound and audio devices-there You will find Audiocodecs. Under Properties of these audiocodecs You will find them enumerated (about 10-12) I removed them all as is suggested by the dialog.
Registry was a blind shot.but I recall I left untouched security and those I wasn't sure about. Really, Theo I relied completely on common sense, that is why I ask more knowledgeable guys to retrace my steps. Yes,I just DELETED them. Unknown entries were well.. unknown. Also I remember I deleted entries from alwill software, avg antivirus, anti malware bytes smth. but it may not be relevant to Your machine. Be very careful, because after my drunk escapade I don't see any devices listed under hardware in control panel.
Wish that computer specialists among us join this qwest.
Serge.
ok, what am I missing here? I downloaded cPlay, but it sets on the desktop and remains passive when I click on it. I am trying to see how it works on a new AMD Ontario Laptop. I get this on the read me "cicsPlay installation successful (no settings yet)." But Windows 7 says it can't find it. It's there "C:\Program Files (x86)\cics Play" 1.91 MB. I tried SSE4 and SSSE3.
I just know this is something real simple.
I just know this is something real simple.According to Wiki, Ontario chips support SSE, SSE2, SSE3, SSSE3, SSE4a, NX bit, AMD64, PowerNow! and AMD-V (even) so I doubt that supported instruction sets are your problem. (Some fancy upsampling features work on Intel chips only IIRC).
However, you say, "It's there 'C:\Program Files (x86)\cics Play' ".
If you're launching cPlay from a short-cut, is its path correct? It might be that it's pointing to C:\Program Files\cics Play\cicsPlay.exe [i.e. no ( x86)] as that's the default location for Win XP.
Have you tried navigating manually either in Explorer or via the Run command to cicsPlay.exe and clicking on it?
In the \cics Play folder, there's also a .BAT file that launches cPlay. I don't know if it's used in routine play (I think it's used by cMP) but try double-clicking on it to see if it works. If not, try opening it in Notepad and editing "Progra~1" to read "C:\Program Files (x86)" and having another go.
Note that I know SFA about Win 7 (I'm sort of OK on XP) but it's possibly worth your checking the above.
HTH
Dave
Edits: 07/04/11
Try SSE2.
Rather odd as this is a new chipset that supposedly supports "everything". I wonder why the older SSE2 works and newer ones don't? Is this something cics could address and is there really any need to actually have the newer instruction set?
Thanks to all for your help.
SSE3 and SSE4 instructions by AMD are slightly different from Intel's, that's why.
and see if it opens the file.
If so you stopped at a good place and thanks for all the effort. Cplay is no small achievement and continues to provide a remarkable front end (the best) for my system and perhaps many other AA inmate systems. It was a tremendously exciting ride for me personally.
If Cplay 40 is to come then thanks for the continous effort.
Just wondering if anyone has noticed anything like this, or if they haven't could tell me some areas that could possibly be the cause of this...
Recently I changed the hardware of my full cmp2+cplay setup. I have noticed since this change that on some recordings there appears to be some distortion on the treble. This only occurs with 'loud' recordings, for example on Amy Winehouse's 'back to black' on the louder vocal parts.
Initally the problem was pronounced (and annoying) but I let the system burn in over about 50 hours and it seems much less pronounced, to the point where now sometimes I'm questioning if it's the recording, the rip or my imagination.
Overall the hardware changed the sound signature markedly with the sound being more open, there being more detail and more seperation between instruments.
My system is the recommended i3 system with 1GB of RAM. I use an Asus Xonar Essence for the digital output rather than the juli however. I'll detail the changes. I went from:
-Magnum 500W power supply
-1tb WD green in an external enclosure plugged in through sata with an e-sata to sata converter cable.
To:
-Antec signature 850
-Patriot 60gb SSD as OS drive
-2tb WD green storage drive connected in the same way as above.
I also used nLite to cut down the version of windows (XP SP2) I used and disabled the USB ports I wasn't using, I did not do this before.
Any help at all would be great.
Thinking out loud... could this be due to underclocking the CPU? I have it at .650v I believe.
Hi Apokalypse
have you tried to use the analog output of Essence? Which kind of driver have you in use? Upsampling? I can't believe that it's a problem of undervolting, this could be cause of dropout not distortion.
I think it could be the digital output ( connection-cable- DAC- other?)
Try analog output and listen if distortion persist
Daniele
If the input freq and output freq are the same, can I disable the digital filters in the CPlay.
I understand that if the input and output are the same freq (say both 44.1), the Resampler IS NOT used.
However, Resampler and Digital Filter are two different things, one can bypass the Resampler but still use a digital filter.
So can the digital filters (not the Resampler) in the CPlay be bypassed/disabled completely?
Thanks,
DM
thanks.
Dear DM9999,
In your specific quest about what cPlay does to an equal : in => out samplingrate, only cics can give you an exact answer, but I kind of remember that he has claimed several times, that straight in, is straight out....
There is no (cPlay) filter involved in this equation !!
what leads you to this assumption ?
kind regards
cMP2 Computer: i3-540 Core cPlay039 >FIR filter w/ convolver> Lynx Aurora 8 FireWire /192kHz throughout. 2x AcousticReality Ref. 202 & 2x AcousticReality Ref. 601´s ICEpower. Magnepan MG3.3R beechwood frames & custom stands. Miller chokes
> what leads you to this assumption ?
because the sound is slightly different when I change the filters ... or I am just imagining things?
Using cMP^2 with the net use command to play from a library on an external usb drive attached to my other networked computer, it usually functions well and sounds great, perhaps better than with the same drive attached directly. But any time it encounters a track large enough to be segmented as shown in parenthesis after the title it plays for about 23 seconds and then is interrupted by an asio driver reset request. This is so for all samplerates, buffer settings and regardless of whether resampling is involved or not. This never happens when the drive is directly attached. It would appear that the problem might be with reloading RAM with the next segment, but of course I don't know.
Anyone else encounter this? Or can you report it doesn't happen in your system so I know to look further for a local issue.
I had cMP^2 on one drive, a hdd, that worked on segmented files and one, on a ssd, that didn't. Now the hdd version doesn't let them play thru either. So now it seems it is a corruption of the network function or a router change of some sort that develops over time. I am baffled. Any thoughts and/or suggestions? (Dlink DIR 655 router, xp pro fully modded on Lenovo t500 laptop)
I tried setting up the networking on a hdd cMP^2 setup I had as a backup. It worked properly right off with no aborting of segmented long tracks. So I cloned this partition onto the ssd, replacing what was there that was having the problem, and now this clone of what works on the hdd also aborts segmented tracks. Clearly the problem has to do with the ssd. It is a Kingston 16GB ssdnow 100 like a number of us got on sale, I believe, from Egghead earlier this year. I used Paragon migration software made for migration to ssds. I know little about ssds, so wonder if someone can tell me why this might be and what might be done about it.
The consolation is that I now have an hdd-based cMP^2 setup that can play from a network drive and sounds quite good. It's just that the ssd should be better if I could get it to behave.
wonder if someone can tell me why this might be . . .
I too had issues with one of those "bargain" Kingston SSDs though not as you describe - I was getting drop outs with data residing on the drive. I solved it by streaming over a LAN and having the OS only on the SSD. (Both configurations are fine with an HDD.)
what might be done about it
I bought the bottom-of-the-range 8GB Kingston SSD simply to test the idea and plan some day to buy a more up-to-date one. Hopefully that will fix the problem.
Whatever, the Kingston device made the sound so much better than the old 2.5" HDD (no idea why) that I devised a work-round pending funds.
The rest, as they say, is history. Hopefully the Kingston SSD will be the same before too long.
Dave
Can you report it doesn't happen in your system?
It doesn't happen in my system. Honest.
I see that your music (i.e. shared) drive is connected to your server via USB. That's going to be on the s-l-o-o-o-w side. Can you try connecting it via a SATA port, even if only temporarily for test purposes?
I've no idea why but cPlay doesn't seem to cope well if there's a delay in receiving data for whatever reason. Of course, with a local SATA drive, the situation doesn't arise.
HTH
Dave
Any further thoughts about what the problem might be?
Any further thoughts ?
In cPlay's settings dialogue, disable Autoplay (makes things easier to control; double-click to start play) and AWE. Repeat your test.
Even if it works, it isn't a cure but it would be a step forward.
D
.
In cMP have you suspended both svchost and lsass? Try to suspend svchost only.
No change with only svc host suspended.
Strangely, I have come across one 24/96 album whose 3 tracks though in 2 segments each play right through and don't interrupt after 23 seconds or so. All others with 2 or more segments tried so far do interrupt at about that point.
Thanks for the quick response. I don't believe I have a way to connect this external drive via SATA, but I guess I could try playing some long files from the server computers primary internal hdd to see how that goes.
Hi,
One thing that I don't like with the cmp2 is that it doesn't allow for audio streaming from a NAS. I can understand it SQ wise but not convenience wise.
Therefore I wonder IF letting the network stuff active to copy/paste the intended-to-be-listened-to audio files from NAS to SSD (a partitioned solid-state HD with system and programs on a small partition and a 'music tank' on a different partition)could be an alternative as far as SQ?
Hi Golo,
I wouldn’t worry so much about using your LAN socket as a source. Why do you feel a LAN socket is such an inferior source compared too a HDD?
I know that the communis opinio amongst audiophiles is that using the network function degrades sound quality. But I never read very detailed explanations as too why this is. I even dare too steer up some discussion about this subject because the communis opinio doctorum about this is not so negative at all. In Pro Audio LAN has no bad reputation at all.
Why should retrieving an X amount of bits from a LAN socket, be more troublesome for sound quality than retrieving that same X amount of bits from an HDD? Or why would it be inferior compared too retrieving that same amount of bits through E-SATA, USB or Firewire ?
Either way the transportation of that X amount of bits will cost CPU activity, will use I/O chips and logic, will produce HF noise by the electronics involved, will use current and cause voltage sacks, ect, ect.
So why would taking that X amount of bits from a LAN socket be so inferior too taking them of a HDD? In my opinion a HDD is about THE WORST source too get your music bits from. A HDD is one VERY BIG (!) source of conducted EMI pollution, it is current hungry (voltages sacks), it vibrates, it produces heat, ect. As a source I would choose a LAN socket over a HDD any day.
Transportation of the bits along the MoBo.
I’m also not very much convinced that transporting the bits from the LAN socket via the PCIe bus too the ICH7 intel chipset, is so much worse for Sound Quality than transporting those same bits from a HDD through the SATA controller too the ICH7 Intel chipset. Same question for transporting those bits through E-sata, USB or Firewire too the ICH7 chipset.
All these methods use I/O chips and logic, use PC resources, produce and conduct noise, use energy, etc, etc. So I would really like too see/read a detailed article which explains too me how and why one concept or method is superior over the other with sound quality in mind.
From all 4 concepts/methods too connect a HDD too a mobo (E-sata, USB, Firewire, LAN), I think the LAN connection might even have the best EMI-rejection coming from the HDD.
So if you already have a SSD in your cMP and you must connect a HDD too your MoBo for extra storage, I would seriously consider a LAN connection. HDD’s produce A LOT of EMI (do some google-ing on: EMI + Harddisk) and connection through LAN might have best EMI rejection.
Also one way or the other, those bits have too be transported along the MoBo from a source too the ICH7 chipset. As I haven’t seen any good articles about which concept/methode is best for sound quality when transporting the bits along the mobo too the ICH7 chipset, I also trust my ears.
When assigning a fixed IP-address, cMP can make use of LAN with only 1 extra service running (workstation). Transportation of the bits along MoBo from the LAN-socket too the ICH7 chipset probably doesn’t generate more noise than other transportation methods do (USB, Firewire, e-SATA, ect), because when I use LAN I don’t notice any degradation in sound quality.
Mark
fully optimized cMP2 PC -> ESI Juli@ -> Van den Hul Optocoupler MkII-> Lavry Black DA10 -> XLR Mogami Gold -> Klein & Hummel O300
Hi hfavandepas,
Thanks for answering.
[Why do you feel a LAN socket is such an inferior source compared too a HDD?] Because CICS told it. Although my question was referring to a SSD not to a HDD.
Anyway my knowledge in computer things is so poor (I don't even know what a ICH7 chipset is!) that I can't make an opinion on my own. That's why I'm asking the experts! And I'm very happy of your answer that's saying that by using a LAN connection it shouldn't degrade SQ.
Therefore is there a 'good' way of doing this LAN connection or is it just inserting an Ethernet card in the computer without much of fiddling around?
I've read that by sending RAW wav files makes the player to work less (comparing to sending wav or FLAC files) therefore to increase SQ. Is it true?
I've got plenty of other questions to ask but I don't want to overwhelm you with them from now.
Let be assured that I've greatly appreciated your answer as, as we say in French, 'you call a cat "a cat"' ie you've got your feet well down to the ground.
It was a pleasure to read you,
Cheers,
Golo
Hi Golo,Well Cics is right. Any extra activity in your cMP setup that is not needed too process your digital audio stream will most likely affect sound quality because these extra activity’s use hardware, use PC and processor resources, involved hardware produces noise, causes voltage sacks, ect. So when LAN is not needed: don’t use it and disable hard- and software.
But I red your question like this:
* What too do when my SSD is too small too store (a copy of) all my music files?Your question probably arises from having not enough storage capacity on your SSD. Otherwise you just could simply copy all your music files too the SSD and your problem was solved. But probably your SSD has not enough storage capacity, so you are considering other options.
Best option of course would be a bigger SSD or an extra SSD, but that is not what you are considering. You are considering the option of connecting a HDD through LAN.
But you are worried about the negative affect this LAN connection will have on the sound quality of your cMP setup. You are right. Using LAN will impact sound quality negatively very slightly.
But every other option, will also impact audio quality negatively.If adding more storage capacity through extra SSD capacity is not an option, I read your question like this:
* which option for extra HDD storage capacity will least impact sound quality of my cMP setup?You can connect extra HDD storage capacity through:
1. a second SATA channel on your mobo (internal sata connection port)
2. eSATA (external sata connection port)
3. USB
4. Firewire
5. LAN.
All these 5 methods will impact sound quality of your cMP negatively.
But which of the above 5 options will least impact sound quality of your cMP setup?Option 1 and 2 are out of the question since you already know how big the negative impact on sound is when you connect a HDD through SATA in your cMP setup.
Leaves options: 3, 4 and 5.The whole point of my previous post was that I don’t see why a HDD connected trough LAN should be such a bad choice for sound quality. Method 3 and 4 also use I/O chips and logic which generate noise and pollution, also use processing resources, alsoo draw current which causes voltage sacks, ect, ect. I never read any detailed articles that paint a bigger picture on why and how a HDD connected through USB or Firewire is superior for sound quality than when connecting it through LAN.
So just trust your ears when choosing between option 3, 4 or 5. So why not try it and listen for yourself? I’m with Ryelands on this. It's a perfectly valid alternative. So just try it and see if it meets your sound quality standards.
* Some tips when you use a LAN connection too connect your HDD.
One ingredient in the cMP-recipe is lowering speeds of chips and logic so too decrease the noise these circuits generate, too make them less current hungry, too reduce voltage sacks ect etc. Same applies too the LAN chip on your MoBo.
1. So I use my Realtek PCIe GBE family controller with these setting (in ‘Device Manager’, view: ‘Devices by connection). In ‘properties’ under the ‘advanced’ tab:
- I put it at fixed (low) 10Mb/s speed (so not 100 or 1000Mb/s)
- I disable all automatic energy savings options
- I disable all automatic Wake On ‘etc, ect’ facility’s2. The PCIe channel in my MoBo has 2 root ports. Only one root port is used by the Realtek LAN controller. So I disable the PCIe root port that is not used by the Realtek LAN controller.
(BTW: I also disable channel 1 on the Serial ATA storage controller. I use my SSD on channel 0 so Serial ATA channel 1 can be disabled both in the BIOS as well as in the device manager)3. Cics describes very well which devices NOT TO DISABLE in the device manager when using LAN. So I will point you too cicsmemoryplayer website for that.
4. However Cics is not so clear on his website what services are minimal needed when using LAN.
A cMP setup can run with only 2 services enabled: ‘Plug & Play’ + Remote Procedure Call (RPC). For LAN you only need one extra service: ‘Workstation’.
You don’t need too use the ‘DHCP client’-service if you assign a static IP address by hand yourself.
(you only need too assign this address once)N.B. This is not mentioned as a detailed ‘how to’. I just wrote it too roughly guide you. After reading this you now know all area’s in a cMP PC that need attention. Exact details as where too click and look and also how too configure your IP settings between your PC and NAS, you will have too study yourself. But also that is rather easy not very hard too accomplish.
Succes
Mark
fully optimized cMP2 PC -> ESI Juli@ -> Van den Hul Optocoupler MkII-> Lavry Black DA10 -> XLR Mogami Gold -> Klein & Hummel O300
Edits: 06/14/11
Hi Mark!
You wrote "I put it at fixed (low) 10Mb/s speed (so not 100 or 1000Mb/s)". Are you sure that such speed reduction give improving in SQ. After all thus loading of a content from NAS into cMP2' memory through a network during playback will be very slow.
http://cmp2-mihaylov.narod.ru/
Hi Mihaylov,No I’m not sure. Actually I don’t hear any SQ difference. Not for the better, nor for the worst.
I just lower the speed because in general lowering speed will reduce the amount of noise that is generated/emitted by the hardware involved. And probably also the involved hardware will draw less current or will have less need for current in short periodic bursts (avoiding voltage sacks).
However the above is just speculation/theory.
In practice I can see that cPlay not starts playing before (an amount of) the music file is loaded in memory. While the first music file is being played, at some time the second music file is loaded from the NAS through LAN. Whether this is done slow or fast, makes no sound quality difference when playing the first music file. At least, it makes no sound quality difference that my setup is able too let me hear.
So too be on the save side I put it on slow. Being on slow (10 Mb/s)it takes a fraction more time before cPlay starts playing the first music file. But I can live with that.
* But now for something completely different.
Your message comes at a crucial time. Right know I’m experimenting a lot with ferrites too filter the amount of very HF on the power supply lines too the P4 and too the P24. When used enough(!) ferrites in combination with some decoupling caps too shunt HF frequencies, this results in a very nice SQ improvement.
I use the recommended Peaktech/Velleman Linear PSU too power the P4 with 12 DC. Placing ferrites on the 12 V DC lines improved sound quality noticeably. Not much, but just a little. With some extra decoupling caps it again gave a little improvement. I wasn’t expecting this. However it is just a regular, mass produced linear bench PSU. May be it is still somewhat noisy.
I wondered if you ever tried placing (enough!!) ferrites & some decoupling caps on the DC lines of your custom build, high quality PSU?
I’m very curious if ferrites also improve sound quality in you r setup when (enough!) ferrites are used on such a high quality, custom build linear PSU.Have you ever tried ?
Mark
fully optimized cMP2 PC -> ESI Juli@ -> Van den Hul Optocoupler MkII-> Lavry Black DA10 -> XLR Mogami Gold -> Klein & Hummel O300
Edits: 06/26/11
Hi hfavandepas,
I'm happy that you get back as obviously I misunderstood you.
I can understand that "Any extra activity in your cMP setup that is not needed too process your digital audio stream will most likely affect sound quality because these extra activity’s use hardware, use PC and processor resources, involved hardware produces noise, causes voltage sacks, ect. So when LAN is not needed: don’t use it and disable hard- and software."
That's why my suggestion was to use the LAN connection just to copy what I want to listen to (let's say a 40mn or 90mn album)from the NAS through LAN to the SSD. I was thinking of using a partitioned SSD with programs on one side and data on the other side. Therefore what I am listening to is on the SSD and doesn't involve LAN connection in real time. Does the fact that the LAN connection is used just to copy audio files on the SSD data partition BEFORE the playing time influences the SQ?
Of course if I want to listen to a playlist of hours of music for partying or background music purposes I'd go directly from NAS to cmp ie the upmost SQ is not needed.
I'm also puzzled that nobody in the cmp2 blog is referencing to DRC (Digital Room Correction). Does it mean that they are ignorant of it or that they consider it as a 'minor' improvement well behind the source improvement?
I personnaly won't consider any digital improvement path without a DRC correction. Once you've experienced it there's no way to going back imho.
Hi Golo,When done right Digital Room Correction can sound very, very good. Same accounts for digital correction of speakers. An other audioasylum inmate invited me over too his place too audition his Klein & Hummel O 500 C - Digital Active Main Monitor. These active monitors have digital correction build in (not for compensating room characteristics) but for correcting and compensating characteristics of the applied speaker drive units, speaker cabinet, build in amplifiers, etc. The end result is truly impressive.
He stressed that I should explore the possibilities of DRC because it would greatly enhance my listening experience. He was convinced that DRC would better my listening experience more than buying a new top line DAC from makes like Weiss, Prism or Lavry (I'm planning such a purchase.So when I’m done battling conducted EMI noise too lowest possible levels in my cMP setup, may be next project shuld be Digital Room Correction. There are some professional Digital Correction systems(The Klein & Hummel Pro C 28 - Digital FIR Controller would fit my O300 speakers nicely) and also some ready made out-of-the box solutions for consumer electronics.
What DRC system do you use?
Mark
fully optimized cMP2 PC -> ESI Juli@ -> Van den Hul Optocoupler MkII-> Lavry Black DA10 -> XLR Mogami Gold -> Klein & Hummel O300
Edits: 06/15/11
I'm not dedicated yet to one system although I'm sure to use one: the good thing about cmp is that it allows you for one plug-in which, for me, will be a convolver in which I can load my FIR filters.
Of course it will use more CPU, current etc... But I tend to personally think that DRC makes bigger improvement than a change of output capacitors or ...
I've experienced with ACXO which has the advantage of being very effective and simple but is limited to 16-bit, with Inguz on my Squeezebox Classic which set-up is more complex, but ultimately has the same limitations of ACXO, with pure DRC by Denis Sbragion which set-up is too complex for me and finally I'm now experimenting with Audiolense, which you have to pay for, but the free demo version gives me already fantastic 'in location cues' ie a fabulous 'beyond the speakers depth' with my own recordings of symphony orchestras and choirs. It is less obvious with compressed commercial recordings.
Anyway I'm convinced and I will use DRC. The SQ advantages of DRC are far more important to me than the so-called digital artifacts that some detractors are hearing (?).
Hey golo & hfavandepas
...in all humbleness :
I´ve gone from digital IIR crossover technique (Allocator) to FIR filtering with my cMP2 maschine for some while now, and the result is nothing short of amazing and highly recommendable !
There are two important considerations :
1. -with FIR filtering you may need a slightly higher processor speed and associated higher voltage level in your system.
2. - FIR filters do also need to be trimmed correctly to your environment for highest fidelity. A number of the cheaper software offerings for this are not up to high-end standards, because they are too automated.
I suggest using Acourate for state-of-the-art implementation of FIR, simply because it can be tailored better than others.
I´ve posted a number of views on how it works on the digital asylum, so if anyone wants to consider FIR filtering within cMP2 have a look or send me an e-mail.
kind regards
cMP2 Computer: i3-540 Core cPlay039 >FIR filter w/ convolver> Lynx Aurora 8 FireWire /192kHz throughout. 2x AcousticReality Ref. 202 & 2x AcousticReality Ref. 601´s ICEpower. Magnepan MG3.3R beechwood frames & custom stands. Miller chokes
Thanks play-mate,
To be honest, although I'm sure I want to go the DRC way, I don't have enough knowledge to understand it properly and therefore to be able to use Acourate. What is good for you "because it can be tailored better than others" might not be useful to me as I doubt I know enough on how to tailor it for my needs.
One thing I've noticed when using DRC is the improvement difference it makes wether you're playing a commercial recording versus a original one. I've noticed that on my own recordings, especially those with a huge soundstage ie a symphony orchestra, the improvement made by DRC is huge (specifically regarding depth)compared to a commercial pop recording.
Have you noticed the same?
I also wonder if you could do me a favor and direct me to the right threads (so many are not identified by what they are talking about) regarding the following topics:
- I'd like to know if it's possible to implement cmp2 with 2 different configs: a 'purist' one following cics rules as well as a 'lite', more user friendly one for background music or for my family to use it where use of a touchscreen or screen+mouse and connection to a NAS, operational CD/DVD drive to use direct CD playback etc is possible?
- is there a comparison, SQ wise, between usb+usb dac, usb/spdif converter+dac and soundcard/spdif+dac?
- as the aim of cmp2 is to use the less possible load and energy would it be good if it played raw wave files instead of wave or flac?
golo
Can you give me a link please to your post in Digital Asylum? Couldn't find it.
Thanks.
hey kclo,
I did a summary on my FIR filter project here :
http://db.AudioAsylum.com/mhtml/m.html?forum=mug&n=171469&highlight=FIR+play-mate&r=&search_url=%2Fcgi%2Fsearch.mpl%3Fforum%3Dcables%26searchtext%3Dq-10
-and you will find a lot more when searching "FIR" autored by play-mate....
kind regards
cMP2 Computer: i3-540 Core cPlay039 >FIR filter w/ convolver> Lynx Aurora 8 FireWire /192kHz throughout. 2x AcousticReality Ref. 202 & 2x AcousticReality Ref. 601´s ICEpower. Magnepan MG3.3R beechwood frames & custom stands. Miller chokes
Thanks Play-mate,
If I were a planar speaker user (I once was, decades ago), I would definitely follow your example. But I am now a quite happy Mirage OM-6 user (the optimized cMP system really gives new life to them), I am more interested in the effect of FIR filters to correct room effects, rather than digital cross-over and multi-amp.
But your post is really interesting.
Hey kclo,
-thanks for you encouraging words...
To implement FIR into a traditional setup with regular passive Xo´s is equally possible, and even easier.
Acourate generates then the sweep over the entire frequency range and the result is equally stunning.
-also in a passive crossover does Acourate align a perfect time-, impulse and frequency behavior.
Where as in an active system 6 filters (in a 3-way speaker) are generated, in a passive stereo system Acourate generates two.
These filters are just implemented into the cPlay DSP via convolver. That´s it !
kind regards
cMP2 Computer: i3-540 Core cPlay039 >FIR filter w/ convolver> Lynx Aurora 8 FireWire /192kHz throughout. 2x AcousticReality Ref. 202 & 2x AcousticReality Ref. 601´s ICEpower. Magnepan MG3.3R beechwood frames & custom stands. Miller chokes
I wonder IF letting the network stuff active . . . could be an alternative as far as SQ?It's a perfectly valid alternative - I've been using it successfully for over two years now.
Incidentally, it allows you to run your cMP^2 box headless if you wish. I and others have compared that to a conventional setup (local KVM but no LAN) and found no SQ difference.
For some time, I ran by copying music data one album at a time to the local HDD (much as you suggest) but have been experimenting recently with streaming from a remote "Server". When I first tried this a couple of years ago, it wasn't a great success but, though I don't know why, it's proving more reliable this time. (Another inmate also gave it a try but hit a configuration issue.)
The SQ is also much improved though some or all of that may be down to other changes such as using an SSD to hold the OS. I don't know why an SSD makes for such a marked SQ change compared to a conventional HDD (there is no disk activity during playback) but it does even though I use a good PSU for the HDD whatever the type. What matters is that it is certainly no worse.
Play-mate is right - an eSATA drive works fine. Another system I built boots from one partition on a 1.5 TB eSATA drive (2TB drives were still pretty expensive when I did the upgrade). It hold its music data on the other and also runs headless.
See link for some more information. HTH
Dave
Edits: 06/12/11
Many thanks Ryelands
Many thanks Ryelands
You're very welcome. BTW, I've just remembered that hfavandepas had a post on the same topic a few days ago. I can't find it right now but it's worth a search for as it included some detail. He seems to like the arrangement though I don't know if he runs his box "headless".
Thanks Ryelands,
I'll try to contact hfavandepas
Cheers,
Golo
Hey Golo,
There is in principal a difference in a steaming Client as opposed to the SATA driven approach that cMP is laid out for. The keywords are (as you correctly point out) the network services that are pruned out of the cMP....
-but why the fuss ?
If you absolutely want a remote harddisk eSATA could be the answer.
I´ve now configured two maschines, where I use a small SSD for the operating system + programs, and a traditional 1TB HHD for the music storage.
That works flawless and saves you the expensive SSD tech for storage.
Since cMP loads the files to the RAM anyway, there is no compromise in SQ (granted that the drives have separate power supplies).
kind regards
cMP2 Computer: i3-540 Core cPlay039 >FIR filter w/ convolver> Lynx Aurora 8 FireWire /192kHz throughout. 2x AcousticReality Ref. 202 & 2x AcousticReality Ref. 601´s ICEpower. Magnepan MG3.3R beechwood frames & custom stands. Miller chokes
I also use this hardware architecture.
For the OS drive I use a patriot inferno 60gb and for storage I use a WD green 2TB hdd inside a vantec nexstar enclosure. The HDD sits inside the case while the power adaptor for the vantec sits outside the case.
Thanks for your input play-mate.
Although I didn't fully understood it:
"The keywords are (as you correctly point out) the network services that are pruned out of the cMP" does it mean that even when the player is not 'playing' network services are off, therefore network services have to be forgotten? Does it also mean that remote control from another computer/notebook/iPad/iPhone is not possible?
"I´ve now configured two maschines, where I use a small SSD for the operating system + programs, and a traditional 1TB HHD for the music storage." how do you make your HHD silent?
Hey golo
The main objective of cMP2 is to reduce as much activity on the mainboard as possible, because it all contributes to increased power demand and causes potential jitter.
Network capability requires quite a lot of this "activity", also during memory playback, although most of it is operating in the background.
....that´s the simple reason to prefer a SATA drive instead of NAS.
My HDD is dead silent, but vibrates a little.
Therefore is it de-coupled from the case.
I consider the noise so extremely minimal that it´s not an issue.
kind regards
cMP2 Computer: i3-540 Core cPlay039 >FIR filter w/ convolver> Lynx Aurora 8 FireWire /192kHz throughout. 2x AcousticReality Ref. 202 & 2x AcousticReality Ref. 601´s ICEpower. Magnepan MG3.3R beechwood frames & custom stands. Miller chokes
Thanks play-mate
Dear Forum,
I´ve discovered that Eric de Lupo´s Secret Rabbit Code upsampler has released a newer version than the one installed in cPlay.
In cPlay we find the 0.1.5. version, while as his newest is 0.1.7.
-has anyone tried this new version, and how to go about changing versions within cPlay.
Kind regards
cMP2 Computer: i3-540 Core cPlay039 >FIR filter w/ convolver> Lynx Aurora 8 FireWire /192kHz throughout. 2x AcousticReality Ref. 202 & 2x AcousticReality Ref. 601´s ICEpower. Magnepan MG3.3R beechwood frames & custom stands. Miller chokes
Hey Leif,
Looking at the release notes that version is not new at all...February 2009.
Also I am no expert but the release doesnt really seem to be new, but fixes some bugs on 64 bit machines and some compiling issues.
I am hoping that cics would have seen these and probably found it unnecessary to change.
But I could be wrong. I was probably hounding him with some silly install questions at the time :)
No one here remembers the bending of our minds
for those who want to search for the new version
Hi ,
I've read that it was no good, SQ wise, to use ethernet to get the audio in (in my case from a Netgear Duo NAS). Would an ethernet-to-usb adapter at the end of the cabling ready to go onto the mobo usb port help?
thanks
I've read that it was no good, SQ wise, to use ethernet to get the audio in
It's something of an urban myth, particularly if you use a headless setup, i.e. control the audio box remotely and swap LAN overhead for KVM mush.
What you propose might well introduce as many problems as it solved as USB has a high overhead on its own account. There's some pertinent material on the cMP^2 website - see link.
Thanks Ryelands for your prompt answer.
I'm not a computer expert at all therefore I might have wrong conceptions of how things work. I had the impression that going from mobo-usb-i2s-dac was more 'direct' than to add a SPDIF carrier as in mobo-soundcard-spdif-dac. Also it appears to me that by using stuff like the exaU2I USB to I2S Interface (http://www.exadevices.com/exaU2I/Overview.aspx) with a ES9018 DAC was worthwile compared to using the spdif route.
Thanks for the link. I didn't know of that design - it's not what you'd call cheap . . . Whatever, I thought your question was about moving data from your NAS, not about the DAC interface.
If you haven't visited it before, an excellent place to start is at the link below.
Dave
Hi again Dave,
Another question I've got (please direct me to the right post if it's already been discussed) is that, in the context of the fit-pc2 being the server and another computer being the controller, is it possible to switch from cmp2 to xp back and forworth easily?
The thing it that I'd like to have ONE computer only to serve as the audio player.
For all that is 'already saved' audio files cmp2 will do. But I also like to be able to switch back to XP in order to run DC7 (http://www.diamondcut.com/store/index.php?main_page=product_info&cPath=1&products_id=9) which allows me to listen to vinyl (or any other analog sources for the matter) 'live'.
Thanks
is it possible to switch from cmp2 to xp back and forth easily?
Obviously, a headless setup doesn’t allow access to the boot.ini menu. However, though I’ve yet fully to test the idea, one could easily write a couple of batch files to make the boot.ini file default to the chosen OS install before restarting.
how do you control [ the Fit-PC ] ?
In my case, I select the data from a computer in an adjacent room and use an analogue volume control. But that’s my choice. A system I built for a friend has the cMP^2 box in a roof void, pulls its data from a local drive and is controlled by a Netbook-like device of some sort (I forget now but I think it’s an elderly Dell laptop). Any VNC-capable device is adequate for this including, so help me, an iPhone.
Possible issues with the Fit-PC2 might include:
* If your DC7 application is processor intensive, it might struggle on an Atom processor. You’d need to test it.
* CPlay uses the SSE4 instruction set when upsampling to 24/192 but the Atom processor does not support it (SSSE3 only).
* The 1.1GHz Atom struggles with 24/96 upsampling though the 1.6GHz model should be fine.
[is it possible to switch from cmp2 to xp back and forth easily?
Obviously, a headless setup doesn’t allow access to the boot.ini menu. However, though I’ve yet fully to test the idea, one could easily write a couple of batch files to make the boot.ini file default to the chosen OS install before restarting]
I have to confess that I'm too ignorant to deal with 'writing batch files'
[issues with the Fit-PC2]
It seems that I'd better follow the traditional route.
But, if I get rid of the 'head', can't I swap the USB overhead with the LAN overhead? and have a 'headless' traditional cmp2 computer with the SS4 required power?
if I get rid of the 'head', can't I swap the USB overhead with the LAN overhead
In a headless setup, you swap KVM (Keyboard, Video & Mouse) overhead for LAN overhead. Of course, the K & M bit might be via USB but it's monitors that are really noisy.
and have a 'headless' traditional cmp2 computer with the [SSE4] required power?
Yes, no problem. But it was the Fit-PC2 you asked about.
HTH
Yes but I didn't realize of the shortcomings of the Fit-PC2.
I'm sorry if I waste your time. It's just that I'd like to play with the grown-up but the more I learn from them the more I'm aware of my ignorance. It's at the same time gratifying (to learn) and frustrating (to become aware of my ignorance) and embarrassing (to rely on the generosity of people like you).
You're absolutely right Dave. I was thinking of accessing the NAS through LAN.
The headless fit-PC is a very interesting proposition especially in my case.
One thing I'm not understanding well is how do you control it? Do you really need another computer (which sounds to me impractical for selecting audio files and adjusting the volume from your sweet spot) or is it possible to use a notebook or any other light and small device?
Thanks
Recently ripped CD's with EAC. The cue sheets show the correct track names, but all tracks (except the last one) are listed as the same length. All tracks play the same track??????
Dear Forum,
Haven´t had much success on the digital forum, so I´ll try it here :
Can anyone provide a link or give an explanation of what ASIO actually does ?
how does it work and what does it do to the clock dependency ?
I got into a hardcore technical discussion on re-clocking and ASIO where I need some hard facts.
any ideas ?
cMP2 Computer (XP minlogon) i3-540 Core cPlay039 >Allocator> Lynx Aurora 8 FireWire /192kHz throughout. 2x AcousticReality Ref. 202 & 2x AcousticReality Ref. 601´s ICEpower. Magnepan MG3.3R beechwood frames & custom stands. Miller chokes
PM, search from a different perspective (RECORDING) and you may find more info.ASIO's key benefit for PLAYBACK is that it bypasses typical computer/software layers. Thus, it can help to get cleaner sound.
For PC/Windows audiophiles, this has been a tremendous help, particularly before WASAPI. Until the advent of WASAPI, there was only ASIO or kernel streaming. I am still partial to ASIO (mainly for its variable buffers to affect timings). However, I have heard well-tuned systems with WASAPI that don't leave me wanting.
AFAIK, the most important reason for ASIO to be created was in the sync of input/output timing, particularly during recordings. In dealing with this challenge, features & optimizations were made that benefit those like us, who do mostly playback. So, for core research look into the recording side of ASIO. (You are going to see a lot of Mac users, I suspect.)
Edits: 03/29/11
Can anyone provide a link or give an explanation of what ASIO actually does ?
You could start at:
http://en.wikipedia.org/wiki/Audio_Stream_Input/Output
though there's plenty more. As the link notes, "ASIO bypasses the normal audio path from the user application through layers of intermediary Windows operating system software so that the application connects directly to the sound card hardware".
what does it do to the clock dependency ?
I'm not sure what you mean by "clock dependency".
Hi Dave
thanks for your reply, but I´m afraid my curiosity goes a bit beyond wikipedia.
I got into this discussion about re-clocking a s/pdif signal with an apogee bigben or other suitable re-clocker.
my theory is that the dramatic effect on this re-clocking scenario is the lack of a true ASIO connection to the D/A converter, so I´m getting a little insecure on what ASIO actually does.
in a normal chain of digital transfer from computer to DAC via s/pdif there is surely a level of jitter and it would be correct to assume that looping the signal through a re-clocking device like the bigben will increase quality.
however, my theory is that a DAC with an ASIO driver will be a superior connection mode provided that the clock of the DAC has an equal precision.
-or said in other words : the reason for the improvement of a re-clocking via bigben is due to the fact that there is no true ASIO connection in the chain.
therefore I´ve tried to goggle the technical clue to ASIO but I´ve found next to nothing on this.
I´m very curious.
kind regards
cMP2 Computer (XP minlogon) i3-540 Core cPlay039 >Allocator> Lynx Aurora 8 FireWire /192kHz throughout. 2x AcousticReality Ref. 202 & 2x AcousticReality Ref. 601´s ICEpower. Magnepan MG3.3R beechwood frames & custom stands. Miller chokes
I think there is a slight confusion.
ASIO is a driver protocol and one can have a "true" connection through S/PDIF, AES/EBU, USB or Firewire.
The quality of the clock depends on the implementation at DAC level.
The so called "improvement" one can have with an external word clock is more often due to "chages" in the sound rather diminished jitter.
Yes, you are mixing two sort of unrelated subjects. Next, usage of external clock is not necessarily going to make an improvement regardless of the brand. External clocks are very helpful in the studio environment where many different units are used, and they all need to be synchronized. In the typical home high end set up, where you have transport on one side and DAC on other side, the best result is achieved by having clock closest to the DAC chip. In this situation no external clock will improve jitter performance, quite opposite. The other issue here is master and slave role. In ideal world transport will have option to be switched to slave and DAC needs to be master keeping transport in complete sync with itself.
I believe I have seen your disscussion on DIYaudio. Regarding S/PDIF other formats are much better and if possible one should always opt to use them instead S/PDIF. AES/EBU and even better I2S if possible (short distance connections only apply here)
Is it possible to have console automatically load a saved project file when Cplay starts?
At this point I have to manually load my saved project file every time I change albums in cplay.
Rh,
I dont know for sure as I dont use console, but I recall when cics talked about it, that the free version wouldnt save settings and you had to buy it to save settings.
No one here remembers the bending of our minds
I have purchased a license for the product. all of the setting are saved in a *.cps file. Each time I change albums I have to load the saved settings. I was thinking the program would do that automatically.
OK,
that does seem like a reasonable thing for the program to do.
AFAIK, Presto is the guru on things like console, so try getting in touch with him and see what he says.
No one here remembers the bending of our minds
Try to run Console project standalone, before cPlay, and leave it running. Use cPlay with no VST filter.
Console will work as a VST host and pass the signal received from cPlay.
Edits: 03/23/11
Thanks I will try to right a script for the rip command that starts console.
Hello, Play-mate. I recently read about your system and noticed you do not mention any soundcard, but use firewire version of Aurora dac. I am currently planning to rebuild my very old cmp machine and was going to add lynx 16 xlr-to aurora into chain. However, if it is possible to do without a soundcard with a firewire dac, it would save a lot of money and will make the chain shorter. Please confirm if I am right or wrong and possibly, some how-tos
Serge
Edits: 03/23/11
Hi steppe,
I´m not quite understanding what you´re saying here :
"....going to add lynx 16 xlr-to aurora into chain"
The Aurora comes standard with a AES/EBU in/output via D-Sub 25 cables, right ?
FireWire capability is enabled with the Slot-in FW card which is optional extra, but once installed you just plug-in the FW cable into a port of your computer.
It´s important that your motherboard´s IEE1394 (FireWire) port is operated by a Texas Instruments chip set (TSB43AB23), but most Gigabyte boards are.
-otherwise you can get a cheap PCIe card to do this.
I shall not withhold my relative disappointment with FW over the Aurora. Albeit that the Aurora delivers the most amazing sound quality, the FW has a general 96kHz resolution limit on the newest drivers. On the previous 15 version driver it can handle 192kHz with extended buffer length, but my thoughts are indeed to "upgrade" with an AES16e at some time.
Hope this clears some of your thoughts.
Kind regards
cMP2 Computer (XP minlogon) i3-540 Core cPlay039 >Allocator> Lynx Aurora 8 FireWire /192kHz throughout. 2x AcousticReality Ref. 202 & 2x AcousticReality Ref. 601´s ICEpower. Magnepan MG3.3R beechwood frames & custom stands. Miller chokes
Could anyone please tell me how to save Console setting, with all plug-ins and chains, in order for cPlay to run it automatically, without needing to set it up every time.
Thanks.
Hey Bib,
If you have the trial version that is the reason it wont save. If you have the full version it should save nicely, if not, contact Presto. He is the one who I know has a bunch of experience with console.
No one here remembers the bending of our minds
Running SSD minlogin CMP2. :) T.
No worries!
Awesome!
I still have issues. Dont get why it has been so hard this time.
Do you have the PCI simple communications controller in the device manager?
I do and it wont disable. Well it does but windows always finds it and throws an install party.
Now I cant find the drivers to install the damn thing.
And typical of this build I chased my tail for 20 minutes with AWE until I noticed that I hadnt checked the box in cplay!
No one here remembers the bending of our minds
Hi,
I am now using convolverVST within cplay to correct the defaults of my living room, i noticed that a file named wisdom.fftw is created in all my flac folder as soon as cplay is shutdown, this breaks the last modified sorting which allow me to find the last album i ripped. Would there be a way to prevent this and force the VST to store its settings within cplay folder ?
Thanks in advance
Damien
Hey Damien,
I have previously used convolver and had not the issue you´re describing. strange ?
-but, I DO have a question for you :
as mentioned I´ve had convolver running on cPlay´s VST, but now, after some while, I cannot figure out where and how to call up the convolver configuration.
cPlay just says it can´t find the filter/conf. files...
would be great if you could get back !
kind regards
cMP2 Computer (XP minlogon) i3-540 Core cPlay039 >Allocator> Lynx Aurora 8 FireWire /192kHz throughout. 2x AcousticReality Ref. 202 & 2x AcousticReality Ref. 601´s ICEpower. Magnepan MG3.3R beechwood frames & custom stands. Miller chokes
Hi,
I can deal with it, so it is fine, i am currently testing other convolver VST software and will report when i fine one good one.
About your question:
Launch cplay, open a cue, click on green VST icon, convolver VST configuration window appear, click on litte button close to Filter/config and browse to your impulse response wave file. Tada you are done.
Damien
hey Damien,
yep, that did it !!
-I knew it was a tiny bit I was missing.
Thanks a bunch !
cMP2 Computer (XP minlogon) i3-540 Core cPlay039 >Allocator> Lynx Aurora 8 FireWire /192kHz throughout. 2x AcousticReality Ref. 202 & 2x AcousticReality Ref. 601´s ICEpower. Magnepan MG3.3R beechwood frames & custom stands. Miller chokes
I believe Damien was the one who once suggested loading cPlay not just with high priority but also for a single core.
Works very well for me in cPlay (It also helps Foobar, btw). cMP users may have been doing this from the get go but, if so, I had missed it. Probably for having stayed at just the cPlay level.
Hi,
yes it was me ;) Glad it work fine with you
I am working on a launcher for cplay that will allow tweaking process priority, core affinity and also allow you to lower to 0.5ms the global Windows Clock Resolution (which runs around 15ms by default). When it is done i will post it there.
Damien
Wicked! Looking forward to it, Damien. Tks.
I am wondering if there are others using this combination and whether they encounter this (as described to Daniel Weiss):
"When cPlay is called by cMP to play a cuesheet, if the samplerate set in the driver differs from the setting in the cPlay configuration file, the dac’s setting is changed as cPlay starts but then it freezes showing “medea”, a step toward readying play. cPlay must be killed via task manager and then restarted from cMP. It appears cPlay does not get a confirm of the driver’s readiness at the correct rate fast enough. Also, when the dac is turned on connected to the booted computer it does not get set to the last rate used. It always starts at 48000 actual even though the rate listed in the driver panel as setting is different."
Any ideas about cause or solution? Weiss brought this to the attention of those writing newest driver but the new driver does just the same. I have asked cics in the past. He blames the dac/driver. It seems to me it is a mismatch between how the driver confirms asio driver readiness and cPlay's reception of same, e.g., time allowed to receive before failing. There is no problem when the rates match to begin with.
Hello,
My version of cPlay : 2.0b7 (because I have a SSE1 CPU).
My version of cMP : 1.2 Final.
I have : Windows XP SP3, a sound card E-MU 0404 USB.
My problem :
When I use cPlay alone, everything works, but when I use it with cMP, the music is normal, but I have crackling in the sound when the player changes from one track to another, from one song to another, from one .WAV file to another, I have small crackles in the sound at this precise moment.
What can I change to prevent these cracks ?
My options for cPlay : 44.1KHz, Buffer auto, Resampler 121db, AWE No, cMP No.
My options for cMP : RAM Load Yes, Suspend svchost No - Isass No, Optimize RealTime.
Thank you.
you could try using a larger buffer. the juli@ card has a buffer configuration panel (latency options 46 64 128 ect) choose a larger number and in cplay choose large not tiny or small. or you could set it to auto and see what happens.
What can I change to prevent these cracks ?
You could try running MS update # KB918005 as recommended as a cure for a similar issue on an E-MU 0202. Not even the manufacturer was quite sure why it worked but work it did.
Dave
Hello,
I can not install this update, because there is already a newer version included in my service pack installed (SP3).
I tried other settings with cMP, Optimize : Critical and Player.
But there are always cracks when the player changes from one song to another.
Hello,
I followed the advice of a person that I contacted at the address : cics@cicsmemoryplayer.com, and I tried the following settings :
1. cPlay : check cMP.
2. cMP: RAM Load to No, Optimize to Critical.
But unfortunately, it did not resolve my problem of crackles in the sound in every change of song.
I think that there is a precise reason in these crackles, because I have no crackles when I use cPlay alone.
Would you have other advice for me ? I would like to use also cMP, because I find the sound better (except these crackles !).
Thank you.
Hello,
I solved my problem of cracks at each song change with cMP.
The problem was the Buffer size of the ASIO driver of my E-MU 0404 USB. By
default it was set to 10 ms, I changed it to 20 ms to have no more crackles.
The problem did not really come of cMP, or cPlay, but from the configuration
of the Buffer size of the ASIO driver of the E-MU 0404 USB.
Thank you very much for your help.
NB : Do you go to put the version 2.0b7 (last SSE1 version) of cPlay in
download on your site ? It can not be found elsewhere on the internet, that would be a good idea.
Hello,
I add a precision to my answer. To have no more crackles at all, I had to increase the Buffer size of the ASIO driver of my E-MU 0404 USB.
To have no more crackles at all, I set the Buffer size to 80 ms.
Thank you.
For Windows 7, Multimedia Class Scheduler Service is available. If ASIO driver (or ASIO4ALL) and cPlay are better managed as real time process/thread, would it sound better than XP?
Vista and onwards has seen good improvements over XP's standard audio performance but they also suffer from more complexity, i.e. DRM.
Optimised XP with minlogon offers "least line of resistance". With cMP, we get excellent thread setup (through very few threads being active and affinities) & prioritisation (ASIO thread gets a dedicated CPU core which is better then RealTime). A further refinement is using a single HT CPU core - this removes IRQ overhead thereby removing the need for using IRQ dispatching tools. This is achieved with the new 32nm recommended hw.
NT's kernel as used in XP is very good. If my memory serves me correctly, it was developed by DAC/VAX's VMS architect - at the time this was a highly regarded OS. XP's problem is its excessive bloat - a characteristic of MS adding lots of clutter. Fortunately we are able to remove this...
Hello,
My processor does not support SSE2, but just SSE1.
Where can I find an old version of cPlay (and cMP if possible) for my processor ?
Thank you.
Where can I find an old version of cPlay
I'm not sure if it's the right one though I think it is. Whatever, I hold cPlay version 2_0b12 from November 2008 (all later versions being SSE2 or higher). PM if you want a copy.
(and cMP if possible) for my processor ?
AFAIK, any cMP version is OK.
Hello,
Yes, I want it.
I sent you a private message.
Thank you.
Hello,
Hotmail has blocked the file ! I can not open it !
Could you resend me the file on my personnal address (I have sent to you an e-mail).
Thank you.
Hello,
I received the file, thank you, but this version is not compatible with my PC (cPlay version 2_0b12).
It is a SSE2 version and cPlay closes automatically.
I think the last version compatible SSE1 is : cicsPlay 2.0b9
Because here : http://www.cicsmemoryplayer.com/index.php?n=CPlay.Changelog
It is indicated for the next version (cicsPlay 2.0b10) :
cicsPlay 2.0b10
---------------
November 2008
1. DSP refinement: 128 bit DSP processing
All critical areas changed to use advanced 128 bit processing
capabilities (i.e. vectorisation & streaming) of SSE2, SSE3...
processors. SRC upsampling is excluded (which only benefits from
compiler optimisations).
CPU load reduces and RAM I/O efficiency improves.
2. UI Refinements: Phase icon indicates 0 or 180 (previous '^'
indicator is removed). Settings provides diagnostics information
(available RAM, volume and phase).
Do you have version of cicsPlay 2.0b9 or earlier ?
Thank you.
Hello,
I make a correction, the latest SSE1 version of cPlay is : cicsPlay 2.0b7
Read here :
cicsPlay 2.0b7
--------------
November 2008
1. Minimum processor supported is SSE (Pentium 3). This replaces
the normal cPlay version and will work on all processors
expect deep legacy models (prior to Pentium 3)
2. DSP refinements: more efficient DSP & ASIO processing
The next version only supports SSE2 processors as noted here :
cicsPlay 2.0b8
--------------
November 2008
1. Minimum processor supported is SSE2 (Pentium 4) which replaces
the normal cPlay version (SSE). Processors not supporting SSE2
are not suitable for high quality audio
2. DSP refinement
3. Compiler optimisation done
So, the latest version compatible with SSE1 processors is : cicsPlay 2.0b7
Hello Ryelands,
Thank you very much for the file : cicsPlay 2.0b7
This is the correct version. It works with my SSE1 processor.
I will keep this file carefully, before changing my PC.
Thank you soon.
Hello,
i have a SSE1 CPU, too.
Could anybody of you please send me the old version of cPlay?
Okay, so what am I missing here? Thought that VST plug-ins were processed after re-sampling (and maybe Volume), but when I play a .wav file with SoX engaged for 44.1 ==> 88.2 and hook Voxengo's spectrum viewer, here's what I see . . .
44.1 material has no information above 22kHz. That's why output levels after 22kHz are very low (~-170db). cPlay is performing exactly as per specs/measurements.
Notice the red vertical line - that indicates the max output frequency which is at 44kHz. If you change the output in cPlay to 96k you'll see the red line shift to 48k.
DUH... of course! Sorry to waste your time with neural disarticulations.
BTW: any news on the RAM allocations mystery?
Cheers
Talking about scrolling titles and/or having a popup window in cMP possibly showing the rate as well as full title and cover, I realized the suggestion to add the rate to the end of titles wouldn't work for me who already has well over 200 cuesheets for HiRes albums. I don't see redoing 200 cuesheets one by one. I have thought of a compromise method that works even without scrolling or a popup.The idea is to place a "!" before each such album title, in this case, albums with samplerate of 48k,96k or 192k. For these one would want cPlay set to 192k, while for all others 176.4k. Or, excluding 192k, to use 96k vs. 88.2k. This will show which they are in the cMP windows.
To do this, one must segregate those cuesheets in a subfolder in order to add the "!". I found a free text editing program that can do this for all cuesheets in the subfolder at once: TextCrawler (http://www.digitalvolcano.co.uk/content/textcrawler/tcdownload)
Use the regular expression \nTITLE " for find and for replacement \nTITLE "! . You can substitute another character for !, if you prefer.
Of course, if you want to be more specific about rates you can group each alternative in a separate subfolder and use a series of symbols to denote which specifically, e.g., 1Title for 44.1k, 2Title for 48k ...etc.
Edits: 02/01/11
dear forum,
have we ever discussed the technique of upsampling BEFORE playback ?
cPlay uses a lot of effort upsampling in real-time, so with ever cheaper harddrives the logic of upsampling at the time of ripping seems very alluring, doesn´t it ?
anyone doing this ?
-and which software does enable this ?
kind regards
cMP2 Computer (XP minlogon) Intel E7400 cPlay039>Allocator>Lynx Two B. /192kHz throughout. 2x AcousticReality Ref. 202 & 2x AcousticReality Ref. 601´s ICEpower. Magnepan MG3.3R beechwood frames & custom stands. Miller chokes on Ribbons.
Hi Leif,
What you ask makes sense.
Foobar can do this with the convert feature and you can add dsp to the conversion.
R8brain, adobe, audition (I think) come to mind and isotope rx too, and Sox as well.
I asked cics to include the option of using cplay and its SRC for this very purpose but he never did, though he did add VST (thanks).
I have often thought of doing this but cplay seems to work great at realtime (except for my recent build with the new software).
No one here remembers the bending of our minds
When cics finally released Sox in cplay many of us tried offline upsampling but found that realtime upsampling within cplay actually sounded better.The release of sox caused many of us to become familiar with sox to the extent that we could do offline upsampling.
Edits: 02/07/11
Hey Theob,
Thanks for the reminder. I never dug sox that much, but I never really worked at it. Anyhow I would love to try a file made using src and compare to realtime upsampling.
In theory offline should do better I would think.
No one here remembers the bending of our minds
You have a good point. Back when I compared realtime up sampling (src) versus offline (sox) it may not have been an apples to apples comparison. I agree with you that off line should be better.
Since I sort of opened this subject and I admit I probably did it wrong the 1st time I thought well why not try again but this time apples to apples. Since I now use sox uniquely in cplay it really made sense to try sox upsampled off line to the same parameters.
I listen to sox with Intermediate phase 15 and 96 BW with no aliasing so when I used sox to convert from the base 16/44 file I chose the same parameters...and wow it took about 10 minutes of real time cpu going full time (49% on task mgr) to complete the conversion and the new file is 2.4 gigs (old file is 570 megs). So how does it sound..wow again. Bigger soundstage more tranparency. Its not so good I'll kill my 1st born for it but it is definitely better (maybe my 2nd born). And it is significantly better.
Is it practical probably not but probably worth it for my 4 or 5 favorite files. And it makes sense.
Hey Theob,
Thanks for the update. I'll have to try sox again and repeat your experiments.
Disk space is cheap but if one wants an ssd well best to do it all realtime.
And perhaps there is a compromise or perhaps i dont know how upsampling works. Would it be a benefit to do halve of it offline. Instead of a 192 file, perhaps an 88 or 96, and do the rest realtime?
Just a thought.
No one here remembers the bending of our minds
Don't know the answers try it and tell us. I tell you though if the reason that sox offline upsampling is better is because there is less current and power draw on the cpu (& less rfi on the mobo) then maybe its one more argument for an ssd / nas.
Edits: 02/08/11
Hi Theob,
Thanks for the homework :)...dont hold your breath on that one, I have to get my system RUNNING first!
I find I am torn on the nas thing. In my case it will probably be easier and cheaper to just get a large ssd. I only need about 160g or so having only 120g or so of music at the moment and buying MAYBE a few disks a year.
Once I factor in the price of the nas and the falling prices of ssd, i can probably get a big one an just call it a day.
Now if offline upsampling is indeed better, well the nas makes more sense.
maybe i am following blindly but I trust cics on the networking thing and would like to avoid it all if I can. Though it seems like other inmates are doing the nas with no audible consequences so who knows.
No one here remembers the bending of our minds
I'm not sure I get this... seems like it is saying that there is available over twice the memory required, yet it fails !?!?!?
hey Igo51,
this seems to me that you have NOT made the required AWE settings in the windows registry....
before you apply the AWE check-box in cMP you need to follow the instructions described in the manual :
Perform these steps exactly (applicable to XP SP2 Pro):
Start Menu > Run > enter "mmc" > OK button
File > Add/Remove Snap-in > Add button
Select "Group Policy Object Editor" > Add button
Finish button
Close button
OK button
Local Computer Policy > Computer Configuration > Windows Settings > Security Settings > Local Policies > select "User Rights Assignment"
Double-click "Lock pages in memory"
Add User or Group button
Object Types... button
check Groups > uncheck Built-in security principals & Users > OK button
Enter "Administrators" under "Enter the object names to select" > OK button
Apply button > OK button
File > Save
File > Exit
Reboot
Lock privilege is now enabled. If this is not done, cPlay will fail on AWE and revert to standard RAM allocation. This is reported in Diagnostics only, i.e. "AWE allocation successful" message is given.
before this is applied there will be error msg.
hope this will help !
kind regards
cMP2 Computer (XP minlogon) Intel E7400 cPlay039>Allocator>Lynx Two B. /192kHz throughout. 2x AcousticReality Ref. 202 & 2x AcousticReality Ref. 601´s ICEpower. Magnepan MG3.3R beechwood frames & custom stands. Miller chokes on Ribbons.
Hello again,
Oddly, I had read that in the doc, and had performed those steps - including the reboot. Sorry to report that the error persists even after verification that locking is enabled.
I might draw your attention to the status line following the AWE failure notice in my original post... it is the apparent discrepancy between the available and required memory that does not make sense to me.
Cheers
This happened to me once and it was the fact that too many services were open and were thus using memory which no longer became available for memory playback. Make sure you have just 2 services open.
Have you got "ram load" set to NO in cmp settings,
I think this was required for awe to work.
sorry i cant be more positive,its been a while since i set mine up.
worth a check,good luck.
Thanks for your suggestions. To clarify, the issue is not that AWE is inoperative, to the contrary it works fine for the majority of CUE lists. But for some reason with this CUE list (have reviewed, nothing unusual inside) the memory math seems to go haywire - or the reporting is inaccurate.
Oh, and I am not running cMP.
Cheers
cPlay only checks for valid allocation of RAM from Windows. Insufficient RAM is one cause of many. There was a fix done due to a kernel bug a while ago. However I assume you're on the latest cPlay release in which case this is unexpected.
Try setting cPlay to not use AWE and let me know.
Sure thing.... removed AWE, restarted CPlay (yes, latest issue)
"WINDOWS memory allocation failed!
Physical (3326MB), Available (1717MB), Required (734MB)"
Still no joy in musicland :(
Cheers
Looks like you have 2x2GB RAM sticks installed.
Very interesting, I think this is another kernel bug (esp when only this particular cue sheet causes it). This exact error occurred when using AWE, however standard RAM allocation worked fine. In your case, both AWE and standard allocation failed.
Only way for me to look at this is to recreate the problem on my PC. This would need your cue sheet and its media files.
Yup, 2x2GB is correct!
...you've got mail.
Cheers... and TNX
removed AWE, restarted CPlay (yes, latest issue)
If, as you suggest, it happens on just the one album, have you tried re-ripping the CD (and making a new cue sheet)?
If you can verify the data using accuraterip, so much the better. I've encountered maverick files once or twice but got them to run fine with a re-rip. No, I've no idea why . . .
Dave
Hi Dave,
Thanks for your notes. All of the WAV files play fine via Foobar2k and VLC, so I'm comfortable with the EAC rips. The CUE file was re-generated (and redirected to a different folder) by CPlayListEditor - same results.
Cheers,
P.S. - would also vote in favor of tweak-able playlist fonts !
All of the WAV files play fine via Foobar2k and VLC, so I'm comfortable with the EAC rips.
I'm sure they do but, though you might be comfortable with the rip, cPlay clearly isn't. I've cured issues re-ripping CDs whose files played fine with other software.
You might try launching cPlay, setting it to read wav files ("Files of type" > wav) and testing the files one by one. Even if there's only one faulty file in the folder, the cue-sheet will fail.
If you pin it down to one or two files only, try things like shortening the name and/or removing any unusual characters and re-testing. You'll obviously need to re-create the cue sheet.
If that fails, try re-ripping.
D
Thanks again for your thoughts. Have played every .wav file via CPlay as suggested and they all play just fine individually. Conclusion; no issues with the ripping.
Have also rebuilt the CUE file with the tools on this site and that made no difference in the results.
Do believe this is being reviewed by the cics team -- thanks folks!
Cheers
Maby someone already asked this question. I would like to know has anyone tried to use Linux with jack drivers???
Thanks.
hey ademaro,it´s unlikely that cPlay and cMP will ever appear as a linux version, as the whole idea is to configure windows to what linux already is : a slim, realtime-able operating system.
the fully configured cMP2 can however take advantage of the sophisticated developments of drivers for mainstream hardware.
in the elevated spheres of high-end audio, the hardware drivers do have a significant share to the quality of a soundcard or an interface or the performance of a DAC, and very often the development on these are compromized under the linux platform.top-class drivers do cost a lot of painstaking research and development and are hard to justify for the design departments.
-and since linux is such a marginal platform that´s where corners are cut.
-and where mid-priced DAC´s often are outperformed by top-class ones, even if they use the same converter chip !anyone who is intrigued by connecting different applications and programs in the world of serious high-fidelity be warned :
virtual programs like JACK, console, VINE or LADSPA do offer good performance when need be, but they can only be a second rate solution.
they induce jitter, overheads, latency and less-tranparent audio !...all in contrary of the purist, audiophile idea of cPlay/cMP.
NB : a minlogon reduced windows XP is just 1,8Gb of size !kind regards
cMP2 Computer (XP minlogon) Intel E7400 cPlay039>Allocator>Lynx Two B. /192kHz throughout. 2x AcousticReality Ref. 202 & 2x AcousticReality Ref. 601´s ICEpower. Magnepan MG3.3R beechwood frames & custom stands. Miller chokes on Ribbons.
Edits: 01/28/11
Cics we haven't heard any feedback on your Sonus Faber. Please let us know your experience with them...at least I can live vicariously through you.
Also is Cplay 39 the end of the line or is there a 40 in our future? This may be the longest time between updates that I can remember. Not complaining I could live with 39 forever.
Still burning-in my Strads with about 150 hours done. There's more to come from these beauties.... maybe 300 hours would be ideal. So far am very satisfied with its performance with a few unexpected but positive surprises.
B39 presents a big challenge as I cannot find fault with it for now. Improving on its performance will be very difficult. Until I get my Strads playing at a consistent level I don't think its possible to develop b40. That means there won't be a release for a while.
A potential area for improvement is Intel's 2nd generation 32nm technology. It looks very good for cMP². I want to test this with a Lynx PCIe card.
Thanks and don't hurry on 40 I can live with 39 a long time. I'm sure those Strads are already singing.
If you are using ASIO4ALL, try out the AQVOX ASIO driver. It is quite an amazing jump in sound quality, resolution and dynamics. I had been concerned that the TI PCM2706 based USB controller in my DAC was holding me back based on the lack of a true ASIO driver. Replacing ASIO4ALL with this driver was like going from a non CMP machine to fully optimized. I am now getting everything I can out of my 1541a DAC.
One more update. AQVOX advocates using great USB cables & of course provides a product.
I had previously been amazed at the difference in SQ with USB cables, and had been running a fairly good cable that I had shortened to about 8 inches.
I figured that if solid copper makes a better interconnect than stranded wire, why not for the USB cable. So as an experiment i cut the ends off my cable and used them to terminate a DIY cable with 2 CAT5 twisted pairs. One pair for data (green and white) and the other for the (red/Black) power/ground.
Holly molly, what a jump in SQ. Took 10 minutes and cost nothing other than a sacrificial USB cable.
dear forum,
with prices coming down and availability increasing it may be time to figure out how to use a tablet PC as remote control on cMP2.
has anyone done this and can we know how to implement this ...... ?
I think this could be a worthwhile issue to clear up.
happy new year everyone !
cMP2 Computer (XP minlogon) Intel E7400 cPlay039>Allocator>Lynx Two B. /192kHz throughout. 2x AcousticReality Ref. 202 & 2x AcousticReality Ref. 601´s ICEpower. Magnepan MG3.3R beechwood frames & custom stands. Miller chokes on Ribbons.
has anyone done this and can we know how to implement this ...... ?
Not exactly the same but perhaps close enough to be worth your taking a look . . . feel free to contact me.
can you support ape format? i have to convert all my apes to flac one by one.
You can batch convert your .ape file to .wav or .flac using a utility like SoX. For best results, I recommend using .wav (see FLAC vs WAV ).
When batch converting, make sure you have plenty of disk space - test the process with a few files then do rest. If you use SoX, be sure to use the .wavpcm format (then rename your files back to .wav).
Hi all
I can't resolve the problem of long title, specially with classic music.
I don't know how I could manage the title or better to scroll with a lateral movement both in cplay and in cMP windows.
It could be possible add an horizontal slider to the window so to permit this? It's really frustrating don't know what you are listening to!!
I will appreciate a lot any help.
Daniele
Hello, everybody! Of course, scrolling or resizability of the window would be convenient, but You have apparently forgotten the initial soft-hardware philosophy of the project. The size of the window is defined by the native resolution of 1024-768 touchscreen display built in the zalman 160xt+ case. Also, if adding scrollbar can possibly affect sound quality, I strongly doubt that Cics would add it. I wouldn't.
Serge
Additionally, to Mihaylov, I ask Your permission to write an e-mail to You concerning power supply unit. Especially if You are in Moscow, I'd like to have a "live" talk on the phone, if possible.
I cannot agree with ideological positions about a supposed "pure cmp2".
Just for example, I don't use a Zalman 160 xt but a 501, I don't need a touchscreen because my transport is so far from my chair that it would be only a meaningless waste of money. And I'm sure I'm not alone.
What we are asking is only a simple and safe way to resolve a little but annoying problem, not to do the October's Revolution!
So why an interesting, valid and open project must be stopped in his developement by closing speech?
I can't understand but I'm surprised too......
Hi Serge! I invite you to a forum of russian audiophiles.
http://cmp2-mihaylov.narod.ru/
Here are some ways fuller album titles could be displayed:
1. cMP window scrollbar, as mentioned. A shift to right would show the end of all titles.
2. Scrolling single selected line is limited in showing only one title at a time so little help in selection from all on screen.
3. Option for smaller font, though there is a limit to how much could possibly be shown. Set in .pth file or settings.
4. Option for window size adjustment. Why should the purist limitation for the Zalman case or the like be the only option? Sizability could be selectable in the .pth file or settings.
5. Hover window option: When the arrow is held over (or right clicked on) a given line it could be made to open a window displaying the full album title, or the window could show all titles completely.
None of these methods, it seems to me, would change anything about how cPlay or other player plays a cuesheet or file once selected. In fact, any of the same methods could be used in cPlay as well to display full track titles prior to selection so that these would not be active when it comes to playing the track.
"When the arrow is held over (or right clicked on) a given line it could be made to open a window displaying the full album title, or the window could show all titles completely."Perhaps it could be something like this:
in cMP right-click (a free command in cicsRemote) and a window pops up with cover (a "folder" thumbnail saved in the same directory of Cue file) and three lines of text available. The 1st line is for Artist only, 2nd and eventually 3rd line for Title.
Right-click again and window closes returning to normal cMP listing.
If the album is high res, for example, I always keep the resolution with the title as it helps with cPlay usage/selection - because of "screen restrictions" I can hardly ever read it.This "window popping up option" could be used exactly in the same way for cPlay in order to be able to read a particular long title song.
Edits: 12/18/10
Bibo! This is very good idea!
http://cmp2-mihaylov.narod.ru/
Hey B,
My 2 cents on the subject is that such a feature if it could be added with no SQ issues would be a welcomed addition.
THough my question to that ideas proponents is how could one actually see the cover art??
I had winamp set to show coverart and you could scroll through all the are and select an album...but when I did the graphic optimizations for a cmp2 build the benefit of the art was really destroyed and all the cd cases kind of looked alike.
CAn you help me better understand the need, as I dont find I have it.
In the picts posted wouldnt cplay list "yellow-touch" in the top left hand corner of the player?? So the title "Taklan...." just needs to be displayed in cplay?? Isnt there enough room for that??
And thinking aloud couldnt some of the issues of searching with classical and the need for orchestra, conductor, composer sorting be solved by additional cue sheets?? Just create some with altered genre and artist to give a different way of searching. Alan Jordans playlist creator could be used for this I think, with maybe some manual editing.
No one here remembers the bending of our minds
DR,
The main issue here is that cMP often hasn't got enough space to show a long ALBUM Title; similarly, cPlay often hasn't got enough space to show a long SONG Title.
The need for Album cover is NOT to have a pretty picture, but to help selection. For example, for artists with large discographies someone may have a problem remembering a particular album just by the title as in cMP.
I personally do not see a point in having a "perfect system" playing an album just as an attempt to find the right selection! ;-)
I completely agree with Bibo again.
http://cmp2-mihaylov.narod.ru/
From my point of view, the cover is the least important item needed to appear in a popup window. Most important in cMP would be the full album title and the samplerate. Especially with classical there are often several works in one album/cd and thus in its title. As it is now, one either abbreviates radically as I have often done or divides one cd into more than one cuesheet. This is a hassle, and the latter makes less sense when the works are short and there are several. A popup window with the full unabbreviated title would be a big help.
For cPlay I think the best thing would be to have a window open with right click that lists all the tags of the clicked file much like a properties window in foobar. That could thus show performers, conductor, composer, samplerate, comments et al.
Additional cuesheets makes little sense to me and surely is not convenient, though perhaps I don't get what you had in mind. I have, though, used composer in place of genre for classical music. I don't see going beyond that with the cuesheets.
Yes, indeed. That would be a great enhancement!
You have me wondering what I would do about album covers since I keep all my cue files in one desktop folder with absolute addresses for the music files referenced. I think it would be better to program it to find each cover.jpg in the folder with the music files regardless of where the cue file is, if that is doable.
The title line could add the sample rate at the end even if not included with the title tag. I've been looking for a way to know this before playing a file so I can change between 176.4 and 192 setting for cPlay (via the .ini file) and dac (via driver control panel) to suit the selection.
That is why I keep an album's FLACs, Cue and Cover in the same folder :-)
Having sampling rate in a pop-up window it would be difficult (I suppose) because cMP looks at cue files only.
Excactly for the reason you mention - "change between 176.4 and 192 setting for cPlay" - for high res only I write it in the title. That's one more reason for reading the WHOLE album title.
The trouble for most of us is that we already have a huge number of titles and cuesheets formed without extras like samplerate. If not, we could, for instance, put the samplerate in the cuesheet as a REM COMMENT which cMP could read and put after the title in this popup window.
cMP should be able to read the location of the file right-clicked on from the cuesheet and find the file's samplerate tag and embedded cover and/or the cover.jpg in the same location wherever that is, no?
I agree with Serge.
I second this request.
If it is any easier - cics - perhaps only the selected line can scroll sideways automatically.
Perhaps cics has a hidden present for us for Christmas, like last year :-D
I have complained of this several times and suggested various ways to deal with it over time but to no avail. Cics is indifferent to this apparently, and no one else has come forward to contribute to an improved gui. Obviously, the window could be bigger/adjustable in size or the font smaller as a first improvement, which doesn't seem to me to present much of a problem to program or in terms of compromising performance. It is a blatant and unnecessary shortcoming, so I remain baffled that nothing has been done about it.
You are right, Daniele. I think that horizontal scroll bar is necessary!
http://cmp2-mihaylov.narod.ru/
Planned to create a module or plugin for cplay?
can we use Core i3-540 instead of Core i3-530 ? any difference in quality of sound ?
Only difference is an extra 130MHz CPU freq increase. This is a non-issue as we underclock to 900MHz.
Last night after fiddling around with the balance control the system "froze", though it would continue playing.
I restarted with the usual return to DEFAULT BIOS after these incidents.
I attached a trusty alligator clip to the BIOS pins, went to the store, came back, removed the clip, turned on the power supplies and pushed the start button and nothing happens at all.
I removed the BIOS battery overnight and this morning after replacing and turning on supplies, same result. I think the thing is dead.
Looking at NEWEGG I see a new board that might be even better for the purposes of cMP, the GA-H55M-UD2H-LGA1156. Has the same BIOS setting and less of what we do not need - only two memory slots and no IDE connector.
I am going to order one with the hopes that it will be more stable.
If anyone has experience with this board please let me hear what that has been.
Bye,
Rick McInnis
though I still had ordered the new MB so it will be interesting to see if it makes any difference.
I have the GA-H55M-UD2H and I've had it appear dead, only to start up again after a day or so when I gave it another go. It may have something to do with the extreme undervolting we do - living on the edge.
If not the low voltage is always a problem at startup.
Sorry, I don't remember if it was plugged and/or powered on during these times. I was swapping a high end power cord between the Gigabyte and an Atom box at the time.
After some days with 2.39 I switched back to 2.38 and it sounds a lot better to me.
My MB is intel D510MO, 1G corsair ram, Lynx22 sound card, WD 2.5" 640GB HD, PSU Silverstone fanless 400W, Windows vista32I use only one cpu (disabled HT in BIOS and cpu2 in windows)
(it sounds much better with only one cpu. My digital output is 44.1 so I don't need a lot of proccessing power)I set pci latency in bios to 192 (it makes big difference)
default buffer size in cplay and asio (when I lower the buffer size, audio quality gets worse)
windows32priorityseperation set it to 16 (it means 36 for the foreground and 6 for the background)
I have tested every possible value here. My idea is that the real music player is the cplay program and it should get as much priority as possibleI have also installed a power switch, so I can activate or deactivate the dvd-rom when needed.
and all the other optimizations described by cics..
Thanks cics once more, you have made hi-end computer audio playback possible and at a reasonable cost.
Edits: 12/01/10 12/01/10
Hi everyone,
For those who use cplay without cmp here is a little trick to improve the sound quality by a big factor. Idea is simple, launch cplay with realtime priority. To achieve this i created a link to cicsplay.exe and modify it in the following way:
Target must be like this (adapt it to your windows installation):
C:\Windows\System32\cmd.exe /c start "runrealtime" /realtime cicsPlay.exe
Start in must be configured like this (adapt it to your installation folder)
"C:\Program Files\cics Play\"
Then launch cplay using this link, select a cue file and enjoy.
Warning for this to work when double clicking on a cue file within the explorer you need to modify the default application for cue file from cicsplay.exe to a smilar trick, i will post later on when doing it properly.
Enjoy this simple trick which improve a lot the sound stage and sound quality.
Damien
Hi,
one more stuff to try is to lock cplay to first processor, here is the command i use, i modified cicsPlay.bat in consequence:
C:\Windows\System32\cmd.exe /c start "runrealtime" /realtime /affinity 1 cicsPlay.exe
Happy testing/music
PS: cics should modify cicsPlay.bat to add the /affinity switch (if someone could try under a windows xp as i am using windows seven).
For cue file opening, here is the trick (caution do it carefully in order to not break something).
1) Launch regedit
2) Navigate to HKEY_CLASSES_ROOT\.cue, mine has a "(Default)" key with a value "cPlay" (WARNING if your is different you need to adapt the next points)
3) Navigate to HKEY_CLASSES_ROOT\cPlay\shell\open\command
4) Modify "(Default)" key to launch cicsPlay.bat instead of cicsPlay.exe
5) Here you go
Warning: you will loose cplay icon being displayed on cue files in explorer but it will work.
For those cPlay users with XP Home and not the recommended XP Pro:
XP Home does not include the 'group policy editor' for adding "lock pages in memory" to enable AWE for cPlay. This is a worthwhile feature.
The useful utility below allows setting of all user account features in XP Home - including locking pages in memory and is somewhat easier to use than the 'mmc' route. The utility is on a limited use basis.
This route of enabling AWE has worked for my XP Home installation.
Usual caveats (no warranties etc.) apply with regard to 3rd party software.
Thank you monaco, it works!
The sound is now smoother and there is more space. Worthwhile improvement indeed.
Dear Monaco
Could you describe in simple steps how to impliment the lock pages in memory for AWE in C PLAY using the programme you recommend for XP home
thanks
JON L1
Hi Jon,
After installation, the program starts with a page of all users in the system:
1. Double click on the user under which you play cMP, A page of the user's general profile comes out.
2. Click on the 'Privilege' tab, a page of privileges will show, with red buttons on the right side indicating that they are 'disabled'
3. Locate the 'Lock memory page' privilege, right click it and set to 'Enable'
That's it. Enjoy.
Hi Jon / kclo,
please do not forget to check that AWE is working in the cPlay diagnostics - as per the cMP2 web site instructions.
Glad this works for others too.
Monaco
Dear Monaco
Have followed your instructions with the acc Manager tool and set lock pages in Mem for user. but still says AWE failed and default to stan alloc. Have you any ideas why it doesn't work.
Thanks
Jon L1
Hi Jon L1,
Sorry for the delay in my response - having fun weather here in the UK!
To make my AWE load successful I enabled 'lock pages in memory' for both user and administrator accounts - it is also worth checking that the administrator and user groups have also got this setting.
If you install the 'minlogon' after installing Local Account Manager you will have to re-do the settings.
I have doubled checked both cPlay on its own and via CMP and both have AWE successful in my installation.
hope this helps
Monaco
Dear Monaco
Thanks for the advice, after locking pages in memory in all accounts it works. Sounds smoother (less distortion)and Low Freq are fleshed out and more dynamic overall. Brilliant.
Bye the way using a proper asio driver on trial, to drive my usb
Monica dac, so much better than ASIO4ALL !!!
Thanks again
Jon L1
What is the good usb audio driver you are using? I'm very interested. Want to improve ASIO4ALL + windows native driver. Thanks.
Dear ob2I found a trial version WUA V2.8.45 download for XP at www.usb-audio.com it has an annoying beep every 30 secs or so, but is way better than asio4all in every area, better bandwidth, more dynamic etc. purchase cost around 50 euros or so, but i think it could be a keeper.
Regards
Jon L1
Edits: 12/10/10
I am an XP Home user, and it worked for me.
It was explained in the manual in the Advanced section.
http://www.cicsmemoryplayer.com/index.php?n=CMP.ApdxBAdvanced
Hello, Cics and everybody! Been away for a long time. No computer and no Internet close at hand. Yes, it happens.
THIS IS the question:
Ever since the PLAYER had become sufficiently resolute, and I mean since 1-17 (approximately), I have a strange feeling that the start of the First track is much more grainy than the rest of the set. It seems, like memory after having been loaded with a portion of wav for the first time, comes to a certain initial turmoil. Does it have to do with AWE? Is it only my impression on start-up, though I am sure that later the sound settles and gets much cleaner. Does anybody have a close experience or impression?
Also: is it possible to equip cPlay with a possibility of a 5-7 seconds' delay before it starts to play in order for us to get to our listening place. (I don't use remote.exe).
I would dearly welcome a certain change, also in the UI, namely, to separate balance and volume control as far as possible. If it DOES NOT compromise the sound. Concerning the SQ of 2-39, I'd say that it is a step forward, certainly, very airy and resolute, that I vote for. I am still with my old Biostar setup and use ssse3.
If most of us do not use VST, is it possible to revert to the practice of ssse3b9 at least for more important betas?
With gratitude and respect, Serge.
P.S. Hope to become more active soon and with a completely different setup.
hey serge
dear cics
I must admit I do also have a slight "flickering" and uncontroled playback at the first second of every track and have often thought of suggesting such optional "mute" function as a possible feature for cPlay....
I do however use a allocator VST so this may very well be a part of the reason.
reducing the buffer times either in cPlay nor the lynx ASIO settings does not solve the problem. shorter buffer seems to increase the distortion, but reduces the time the problem occurs.
sometimes the "flickering" in sound quality also happens when the RAM loads in the middle of a track when cPlay indicates the need of RAM loading twice.
kind regards
cMP2 PC audio system playing WAV via cPlay 2.0b39 w/ SRC > thuneau allocator digital crossover > Lynx Two B soundcard. 192kHz upsampling. XLR balanced throughout. IIR filtering.
Zalman HD135, under-volted gigabyte EG45M UD2H. under-clocked Intel E7400 Duo Core, BeQuiet dark power with cap´ed P4 & Mushkin XP 8500 DDR2 RAM. all optimizations according to cics -minlogon included.
is that if I restart the first track, there is no flicker.
In my system it seems to have something to do with a new track selected.
If I hit the roller button and ask for a re-start with Mihaylov's mouse program all is well.
I do seem to have a problem that I did not have with pre-Mihaylov mouse control, my system will seemingly get very confused and revert to default BIOS more than I would like. I do not doubt this is due to fumbling fingers, maybe I still have not learned the new protocols with proficiency?
When this happens I clear the BIOS and turn the machine back on. I get a screen asking if I want the PREVIOUS GOOD SETTING, I assent and all is well. The optimized BIOS is restored.
Oy ve, there are worse things ...
Welcome back Serge!
If I was to express my desire for cplay direction I would ask for more parameter settings within SRC or Sox. For example what would SRC sound like with aliasing? I have become so impressed with Sox with aliasing that just having more things to try might also reveal more hidden truths in the music.
Has anyone tried these?
http://www.bso.org/bso/shop/productCategories.jsp?id=bcat13360032
I have never tries a "hi-rez" download so I am nervous and ignorant.
Will cPLAY play a WMA file and if not what is the best way to convert?
I guess I will have to come up with a cue file? What is the best way to do this?
Their one year subscription of fifty dollars and unlimited downloads seems a great way to gain some important music especially someone like me whose selection of orchestral music is lacking.
Any help will be GREATLY appreciated.
Thanks,
there is not much "depth" to the recording.
At times it can sound like the orchestra has been turned around and the strings are in the back which is strange but does not ruin it.
But the sound is non-fatiguing - I do not hear any unfortunate artifacts.
I thought there was more available but still one could probably get good value for the fifty dollars subscription. A good way to become more familiar with orchestral music.
Thanks, again, to Grant for telling me how to make this work.
Bye,
Rick McInnis
Hi Rick
...what is the best way to convert?
Same advice as given here -> http://www.audioasylum.com/forums/pcaudio/messages/6/67878.html...
Use xrecode II to transcode lossless WMA to WAV or FLAC format. It can also create CUE files, and manipulate metadata for tagging your FLAC files.
There is a new nag system beginning with version 1.0.0.152... if you intend to use the unregistered version, stick with v1.0.0.151.
Hope this helps.
Cheers,
Grant
That's not a Toy... IT'S A TOOL !!
Installed the recommended converter.
Waiting for the Levine MAHLER 6th to download and then to see if I can make it work.
AND THEN, to see if it sounds good.
I am confused about the lossless comment. I figured a 24 bit 88 mHz file would be lossless by its very nature. I readily admit I have much to learn about this.
Thanks to you both for your help and comments.
Bye,
Rick McInnis
Attractive offer, if they clarify; "High-definition" is a very relative term but "lossless" is not.
Nowhere do they mention "lossless", WMA lossless can be converted to a format that cPlay can take. (Non-lossless WMA can also be converted but it ain't worth the trouble.)
After responding to Grant's post I went looking to see if there was something said about the quality of the BSO files and found this in something called MACWORLD - not being a MAC guy, other than my last name, I have no idea as to the veracity of what I am quoting:
Classical orchestras see downloads as a way of distributing recordings that they might not otherwise sell. Fewer orchestras are being recorded these days by major labels, as the cost of recording far outweighs the profits made from limited sales. The Philadelphia Orchestra, for example, sells a wide range of recordings as does the Boston Symphony Orchestra, with the latter selling lossless, high-resolution and surround-sound files in addition to MP3s.
Hope it is true. I would guess the file size will give it away? It is taking quite a long while to load.
Thanks for the comment,
Rick McInnis
Just because I always recheck my assumptions (challenge them if you will ala Ayn Rand) I decided to try sox (at Intermediate 15 phase, vhq 95) and I really liked it. It seems that at least to me it improves transients of flutes, strings, woodwinds and hall space. Upon longer listening I noticed that lower mids to bass were a bit less defined than src 145. I'm not sure where I'll end up but it's interesting. BTW I use tiny buffer.
Any other sox fans?
Dear Theo,
I read this when you first posted it.
Being a creature of habit and being happy with what I was hearing it took me two weeks for me to actually try it.
I have been listening to it for the last few days and I do think I like it better, too. The great thing is if you change your mind you can always go back to the other!
Then there is the aspect where we all grow tired of our systems and something new makes it all interesting again.
I agree I am hearing more details but not in way one assumes the compromises that are made when one chooses detail. Of course, I think much of the ability to enjoy increased detail is due to the excellent effects of the "ground things".
Thanks for the notice. I am glad I took your advice.
Bye,
Rick McInnis
Glad you like it. I've tried going back on and off during past few weeks and I still prefer sox w_aliasing by a big margin.
Thanks to cics for providing this option.
Exactly the reasons that keep me coming back to SOX most of the time. It's good to know that I am not just "hearing things". No cMP, just cPlay.
The very low-end is less differentiated here because of the limitations of my Magnepan MMGs (modded). They sharply become mute under 40hz (but amply strong up from there). I run them with no sub-woofer most of the time; so, I loose the very bottom for the sake of the great textures thereafter. In my environment, SOX at minimum phase fixes the mid-bass definition and still strikes a good overall balance. At 15 I do get a little extra punch.
Have you tried aliasing? What a big surprise when I did. Transients becoming stunning in that there is less smear more air around flutes, triangles and such. I have one Gaite Pariesien file by Arthur Fiedler that always bordered on being too hot but with 'aliasing' on a particular tanbourine type instrument I can now tell there are multiple cymbal like devices all around the instrument wheras before it sounds like one big percussion instrument all smeared together.
Anyway I originally 'summarily dismissed' this feature after a brief test. But after a few hours of listening now with the box checked I have discovered new treasures in my music. May be worth a try for you.
I'm at Intermediate 15, bw 98 and alias checked.
You do remind me that I have to give it and extended listen. I tried aliasing very briefly several versions back. I liked the extra smoothness but feared the buttery sound was more of an artifact that sooner or later would show its ugly side. So, I never left it on for serious eval.
This was early in the year. Now with version b39, one evening I switched aliasing on (SOX at tiny, minimum, 97) for a very "contrasty" Tango piece. Unlike my first time, the smoothness did not sound contrived. Since I was evaluating something else, I made a mental note for the future; gotta give it a good listen.
Tks for exploring and sharing!
After several days now of listening to sox with the aliasing box checked I now prefer this to src. I think you are right that a valid description is buttery smooth but not as an artifact but as in more fidelity. I listened to some of my favorite pieces and toggled between src and sox and you can easily sense (at least in my system) that src pulls the soundstage closer and brightens the timbre but loses detail versus sox. Also all transients are quicker with sox (maybe due to less ringing with aliasing). My wife even declared w/o coaching from me that sox was more distinct and natural than src.Caution to anybody who may try this: do it on a night when everything sounds good (good powerline/mood whatever are your controlling factors) then give it some time. Listen to stuff that has a lot of hall ambiance. While giving it time may initially be a burden I believe you will be rewarded.
I wonder and seem to think 'aliasing' is achieved by bypassing the filter all together. Anybody have an opinion on this?
Edits: 11/17/10
Hi Ted (and other SoX-ers...)
Thank you for prompting me to go back and re-evaluate SRC vs SoX. A "quick listen" last night turned into 3 albums, the wee hours, and a string of "I don't believe it" moments (I much preferred the SoX-sound).
But that was at the end of a long day - and who knows how it might change with a good rest and fresh set of ears ?
Well, today I corralled my teenage daughter after school for a quick listen to our usual reference piece... and she chose the SoX version without reservation or hesitation. Free of audio review adjectives and similes, her reaction was "you can hear more instruments, and they are better separated from each other". (this using Phase-Intermediate:15/ bandwidth-95.0/ alias-checked)
The downside to SoX is - for the obsessive - there's a whole world of parameter tinkering available. The "right" settings are a moving target - varying from record to record - but I sure appreciate being handed the tool.
Is there a consensus for "the sweet spot" using SoX ?
Cheers,
Grant
That's not a Toy... IT'S A TOOL !!
Though I have come to like SRC in the latest version, as always in the past I have found SOX more to my liking. And even more so now.
SOX is more revealing here. SRC more "pleasing" and softer at the expense of detail. SOX can be more crude-sounding as it seems to be more truthful of the state of the system. Luckily, like THEOB mentioned, the aliasing has now improved to the point where it is a useful function. I found, to my surprise, that I can leave it on all the time and not feel cheated out of something, as I do with SRC after a while. It may not be SRC's fault...this is not an optimized system....just cPlay on Win 7 Pro.
SOX settings: small/AWE/phase Minimum/BW 97, with planar speakers.
At times, I suspect that SRC already uses aliasing by default. If so, this may be the reason for no switch.
Isn't it amazing and it was right there all along just waiting to be chosen by us. I'm at the same settings as you. Before I checked the aliasing box I tried min phase and it was too bass shy so I settled at Intermediate 15. I tried 98 and 95 bw and settled on 95 because of theoretical reasons I found elsewhere. But who knows where the sweet spot is.
Ok more on settings. If you try 99 bw it sounds too hot, if you try 90 its not bad but maybe not as open as 95. So at least for me its Intermediate 15 95 bw and definitely aliasing checked.
So theo - how does it compare to no upsampling at all?
I still believe that no upsampling might be most accurate. However upsampling does add airiness, depth and the sense of stage presence in music particularly in classical. Maybe not accurate but certainly enjoyable in the audiophile sense. But sox with aliasing gives more dynamic freedom, more airiness, more front to back (vs src and no up-sampling) more a sense of pure analogue than anything digital I have heard .
You really have to try. Most important thing that sox with aliasing does, is give more of the ictus or that distinctiveness of live music which is clearly related to the better impulse response with less ringing etc etc.
in which case I will have a listen this evening
Hi,
This is a request for cics but also a plea for any other user who perhaps could develop it himself.
Would it be possible to develop a cPlay interface similar to this one?
Incidentally, this is from a Swedish manufacturer - Bladelius. It does not need to be like this one - this is just as a reference.
Perhaps it could be optional, sort of "skin" for the present interface.
It could show all the usual controls and info, so remote controls ans shortcuts stay the same.
The cover to be displayed, perhaps under the name "folder" like in J River Media Center, could be kept in each music subfolder together with WAV/FLAC files.
Furthermore, the added graphical CPU calculation would be only slight if colours are set to 16bit.
I realize that it is not an "audiophile" tweak and it just makes cPlay more attractive, but I am sure it would help many people's fruition.
What do other inmates thinks of this?
I think that you offer it is not so necessary. In my opinion it would be more useful to make a choice of albums in cMP by covers. The visual choice often happens is very useful, especially if the album is unfamiliar to you or forgotten by you (on a cover often there is artist with a musical instrument etc.).
Also I think that it would be more convenient the sort preference "Artist" to remove from the cMP panel and to make its part of the sort preference "Genre". Then the list inside "Genre" would be more compact, especially when it is a lot of albums of one artist (that happens very often).
For example the consecutive choice of the album:
Edits: 11/09/10
do not forget that cPlay is used also by non-cMP users. That's why a "facelift" would be helpfull.
Neverthelss, I agree with your suggestion.
I think though that the possibility to select an album by cover could be implemented for Artist AND Genre. For example, I may have 20 albums for the same artist.
Perhaps there could be 4 covers for each screen and, if there are more, one scrolls down.
to have holding a right-click on a given album in both the genre> albums window and the artist window open a window showing the track title list for that album gotten from the cuesheet. This would not affect the gui otherwise or cpu usage during playback at all and would, I'm supposing, add only a little code to cMP.
This window could possibly also show the album cover for those who keep cuesheets in the album's folder with the files. It would leave out those like me who use a universal cuesheet folder and absolute references to files (without further programming steps).
Nice idea...I see the utility in this.
that we feel our experience diminished by an absence of rich graphical display to satisfy the eye when entering the sonic world of Musical imagery.Upon what is the perceived need for such ocular gratification based?
Honestly I don't even bother adding tag info to my rips anymore..
My cPlay displays just track number and duration.While I did have this odd sense of irritation the sweet tooth of my eyes was being neglected I just concluded for myself such Eye Candy was merely distracting and served not to enrich listening.
I can turn to any number of resources for visual stimulation at other times and eventually the sense of lack and deprivation fades.
Edits: 11/10/10
...if we cannot see beyond our nose.
This must be clear - we are on the same boat otherwise we would not choose and like cMP2 (BTW, I do have a full system).
Everybody here is entitled to his own opinion, but in my view you are exagerating in your reaction if you think that I am looking for "rich graphical display to satisfy the eye". Possibly, I was not able to explain myself properly.
IMHO, Mihaylov's suggestion was acceptable because I can see users like me who have trouble discerning an album among many by the same artists. I must admit, when I see listed the entire Van Morrison's discography or Nina Simone's or John Martin's, I do have a problem recalling a particular album just by the title. Am I looking for "Eye Candy"?! I don't think so.
Sometimes I need to play a certain album or song in order to recall a particular tune I am looking for. This is how my personal memory works either by listening to a tune or by visually recalling an image associated with it. Are you a person who, if sollicited, would mention an album title or describe the cover or recall a music passage? I am much more inclined towards 2nd and 3rd options.
Therefore, for what I just mentioned above, I can see a practical application for covers in cMP and slightly less practical but tied more to music "fruition" as a whole in cPlay. :-)
Cheers
Hey Bibo01,
I think I get what you are saying even though I dont quite have the same problem. I guess it is a small collection, as while I have like 20 cure albums, I still know which is which mostly.
Anyhow you could maybe do the old school thing. See I have a case of discs with the cover art...you know one of those cd cases you can put a whole collections in.
It is by the listening chair, and if I have to see some cover art or look at the song titles (hey all my meta data is pretty much non-existent and everything says "track 1" while the newly ripped titles get names).
I know that is not what you wanted but it might be an option since I doubt cics will do this.
I dont know about your rig but with the video optimizations the cover art on say Winamp is pretty difficult to read at all.
It is funny because I refused to do the cmp2 project because I wanted all the art and lyrics and thought I COULD get it with sound quality. I only built a cmp2 box to debunk it and happened to have an extra pc to try the concepts. So I have shifted dramatically in my views on the pretty pictures aspect.
No one here remembers the bending of our minds
Thanks for your suggestion.
It wouldn't really work for me. What about material I d/l?!
I am not looking for pretty pictures. My cMP2 is a full system.
If i need to listen to MP3s or watch movies, I have a 2nd partion for another cMP-like system.
I am sorry to hear that in your view cics will not implement such "features", but I understand.
This triggers off the reason why I started this theread in the 1st place.
In reality I am quite happy for him to concentrate on the audio quality of cMP2. However, if we truely have this project at heart, I suggest trying to involve other people in developing features/skins/usability for it. Aljordan's Cue Creator comes to mind, but also the latest Mihaylov's remote ontrol.
I am NOT trying to snatch this project, which we must remember is "open source", from cics' hands. On the contrary, I would like it to develop further, more quickly and give its designer a chance to receive some help!
For example, is there any particular reason cPlay can only be 2ch? There is so much good music coming out in high quality multi-ch format which it would be simple to rip and playback. Are we missing out on something?
I am sure cics has many ideas on cMP2 development, but we are not aware of. Most of all, being on his own, he has to prioritize his choices. Is it that bad trying to enlarge the users' base still focusing on audio quality?!
Personally, I cannot write software but I can help develop its users' base, which I do by the way in the country I live in, similar to what Playmate perhaps does in Germany.
I can offer feedback, but cPlay listening sensations is a very tiny kind of feddback!
you the whole title content. Because the text is of a size the information runs off the screen and you can't see which of any given artist of who's material you have several titles loaded you are selecting. Yes I agree that can be problematic requiring you to actually load several in order to get the one you want. Having the means of identifying the full title of each selection would be a desirable feature..I suppose seeing the image you posted i leaped to the conclusion you wanted a fancy GUI.. so sorry..lol
Edits: 11/10/10 11/10/10
nt
fully AOB optimized cMP2 PC -> Lynx AES16 -> XLR AES/EBU -> Lavry Black DA10 -> XLR Mogami Gold -> Klein & Hummel O300
,
I made cicsRemote.exe.
Hi Mihaylov,
would you please contact me on "bibo01 at libero.it" (with no spaces).
I have even tried to register on your Russian board, but it got awfully complicated in Russian (English version or translation does not work during registration).
I would like to get the source code of your cicsremote.exe version.
As I am in the process of developing something for cMP (along cover and extra lines) perhaps I could integrate your program and discuss with you.
Cheers
"I am in the process of developing something for cMP (along cover and extra lines)"
I am happy to see that you've taken up my suggestion, developed it and now will try to implement it. I was afraid it would just fall to the wayside. I am not a programmer, but if there is any way I can help, like testing for instance, let me know. I hope you will find a way to include the samplerate, or simply whether it is a multiple of 44.1 or 48, for those who have not put this in the album title.
I am not a programmer myself, but I am going to PAY some young chap to do it :)
Diplaying the sample rate in cMP, I think it's difficult as cMP gets its info from CUE files.
I suppose that your HighRes albums are not that many yet. I suggest you to do what I did - inserting for example "(24-96)" at the end of the title in the CUE file itself. It won't be such a long job.
cMP will be able to display it then.
If the title is too long and it does not fit in the window, then hopefully my "upgrade" will be the answer.
I have dozens and dozens of hires albums downloaded. But regardless, cMP should be capable of displaying the samplerate somehow to be complete. It really, ultimately, should be able to change cPlay's .ini file to suit this rate so cPlay starts up with the correct/desired rate.
All that is needed is for the program to look at the first file listed in the cuesheet at the address given for its rate and then display it along with the album title. This couldn't be too difficult for your young chap programmer. I'd be willing to chip in for the added cost of this, if that would help.
I hope you will find a way to include the samplerate . . .
I'm not getting involved in this debate :> ) save to note that cPlay already displays both the sample rate of the source and the program's current setting. Look at the bar at the top of the display:
Agreed, it's not ideal as you need to start playing to find out how to set up for playing - but the rates are shown.
Happy New Year to all inmates.
What debate? He says he's adding to cMP not cPlay, anyway. You need to know when choosing a listing in cMP what samplerate it has to set cPlay and in my case with Weiss DAC202 its driver to that rate. Otherwise you make a choice and open cPlay and try playing a track to find out the rate then, if it hasn't as usual frozen due to the wrong rate being set, set the right rate, close cPlay and choose again in cMP after setting the dac driver.
What debate?And a Happy New Year to you too.
He says he's adding to cMP not cPlay, anyway.
Yes, yes, I know that - I'm suggesting that there might just be an easy interim fix that allows you to select cPlay settings using current cMP facilities.
Take a sneeky peeky at cicsPlay.ini and at cMP's dot ini file as well . . .
Otherwise you make a choice and open cPlay and try playing a track to find out the rate then, if it hasn't as usual frozen due to the wrong rate being set, set the right rate . . .
You should be able to avoid the "freezing up" by disabling "Autoplay". Select the album to be played and cPlay will display the play list and report both the source and the target sample rates. If they're OK, hit Enter to start play.
If not, click the settings icon, select the desired target rate and restart cPlay. Hit Enter as soon as the playlist re-appears to commence play.
(If perchance your driver stores settings in a dot ini file, you might also be able to adjust them as well from within cMP.)
HTH
Dave
Edits: 12/31/10
Thanks for trying to help, but I know all that. When I know the rate in advance and need to change I change the setting in the .ini file and in the control panel for the dac's driver. The dac202 with foobar and other players is capable of changing its rate in response to a different rate set in the player but not with cPlay, at least not in cMP mode. If I change the rate in the .ini file then as so as soon as cPlay loads the asio driver fails to initialize and produces an error or a freeze. I never use autoplay anyway. I have asked Daniel Weiss to see if the driver can be modified to perform correctly in this circumstance but this has not been done.
Thanks all for your attention!
Bibo01, to be registered it is not necessary. Do how it is specified on pictures.
Somehow I am always redirected to the registration page - I do NOT access the page you posted.
I think it is due to the fact that I was a RapidPro user myself.
I now managed to D/L your file by using JDownloader.
Thanks again.
Edits: 11/05/10
Thank you! T.
No worries!
The link (RapidShare) asks for a slightly unpleasant free registration. However, even logged in, how do I then D/L the file?
Simply going to your link again, it brings me back to registration page.
Mihaylov, has your Rapid Pro account expired?
Any help please?BTW, THANKS!
I believe that people adding features to cPlay would be the right direction for its development. For example, cics could concentrate on the core software development whilst others could develop extra features for cPlay interface, sort of plug-ins, for its customization.
What do other forumers think of this?
Cics, is this feasable?
Edits: 11/05/10
.
Now the balance control will REALLY be useful.
You are a fine man Mihaylov!
THANKS!
Rick McInnis
&
&
dear forum,I´m having a little trouble to get any signal running with convolver.
the convolver is (-or should be) setup correctly, but maybe the cplay architecture cannot handle this process.any suggestions of what may be wrong ?
(more images in msg. above)
Edits: 10/31/10
Convolver looks like a powerful plugin that allows for DRC and x-over processing.
- cPlay setup:
- Convolver setup:
I also copied Convolver's other .dll files (libfftw3f-3.dll & libsndfile-1.dll) into cPlay's installation folder.
cPlay upsamples 44.1 to 96 using SRC@145. In my test I didn't use a proper IR (just a small 1kHz 96k test tone). Of course this produces terrible noise (52db peak gain)! When I hit vst bypass I get normal music playback. What I did find is convolver's setup requires a cPlay restart (otherwise settings are not affected).
You have a far more complex setup doing x-over and are using the config file approach. I would suggest starting with a simple 2in 2out setup and build from there.
thank you for your reply.
meanwhile I have sorted out how to get convolver running whithin cPlay. the problem was in the configuration file...(of course).
I´m aware that this rather advanced setup WILL induce some noise and some sort of added jitter & polution, but I do have the hope to balance this with the gained control of the loudspeakers physical limitations and the mishaps of crossovers.
it has been a long wish to create a digital setup that allows room correction and crossover duties in one go, and one that does works in ASIO mode with high samplerate at the same time.
i.e. without any routing programs like console or jack.
a poor man´s DEQX so to speak.
while not quite there yet, I´m closer than ever.
kind regards
.
Hi Theo, Play-mate,
Installing the Convolver VST plugin into cPlay two weeks ago was pretty easy. It is indeed just as easy a Cics describes.
So now I’m ready for the next step and I will have too 'tip my toe into the water’: practicing how to make a correct recording of the impulse response of my listening room.
The Simple Automated IR Measuring Tool may have ‘Simple’ in it’s name, but until so far I even haven’t the faintest clue how to work with it.
If there are no other programs or apps, I will have too study the SAIM Tool better. But if any inmates have any recommendations for other (more easy) programs (Win) or apps (MAC), please do post them.
Mark
fully AOB optimized cMP2 PC -> Lynx AES16 -> XLR AES/EBU -> Lavry Black DA10 -> XLR Mogami Gold -> Klein & Hummel O300
dear theob - and dear forum,
this roomcorrection stuff is indeed rather complicated, but done right (and that´s the real challenge) I predict a significant lift in audio fidelity.
most avaliable ready-made solutions do compromize on either the inherent digital -, or on the measurement side.
my experience so far is that, for actively driven multi channel setups the audiolense software is indeed tailored for that, and quite easy to operate.
on the other hand, my own software (acourate) which does not cater directly for such active solutions, enables a more precise tailoring (and requires more work) to achieve the desired result.
no roomcorrection is better than the parameters it is setup to correct !
-that´s the essential hurdle.
most "grown up" audiophiles have the (justified) view that less "complication" usually result in better performance, but with increased processor capabillity I believe the time is ripe to explore this technology further, and not scrape it off as an "effect".
we all know, that the real limitations in our playback systems belie in the transformation of electrical signals into acoustical information and the physical errors that this process incur.
I believe that proper roomcorrection than reduce these errors significantly.
I will of course keep you posted on my proceedings !
kind regards
Refer to ReadMe.txt for important details & specifications.
Change Log (2.0b39 release):
- Fix for balance control when used repeatedly
- Provide keyboard support for balance control: keys 'k' & 'l' changes balance
- Reverse change 5 done in 2.0b37: ASIO interface for general hardware assumed. This should provide interesting comparisons with 2.0b38.
cPlay documentation can be found at the new cMP² website .
Please REMOVE previous versions before installing cPlay 2.0b39. Normal (SSE2), SSE3, SSSE3 and SSE4 versions are available. Running cPlay on a CPU not supporting the required instruction set will cause cPlay to exit immediately. For best results, uncheck "Timer" setting and test different "Buffer" sizes.
As from cPlay 2.0b33, the installation folder is NOT deleted if one or more plugins were used. All plugin settings are stored in files named with the corresponding plugin name and '.vst' as extension. Subsequent cPlay releases will attempt to use these settings. That is, cPlay releases can be upgraded without loosing plugin settings. On installation of a new release and plugins were previously used, the installer prompts to confirm reuse of the existing folder. Answer 'Yes'.
Well, I'm a bit late to the party, I found and installed 2.0b39 - my previous version being b31 (!), and the change was indeed immediately audible to me. I don't have such elaborate words as some, I rather try to say it in a simple way.
When my chain and the sound get better and better - and this also goes for cPlay versions, the difference is often a "less", not a "more". Sound stage, top end, low end, in the end the perfect stereo chain is when all of this gradually dissappears and when we will have the perfect cmp drive, stereo chain and cplay version, then all there remains is a bunch of people making music in front of you. No "big sound" anymore, just aliveness and music.
In this way I happily join the crowd and say thank you very much to cics for this exceptional software.
:-)
regards
Bernhard
In the midst of changing my speaker crossover when this "arrived" ...
I thoroughly enjoyed #38 so I think there was an acclimation period required for me to appreciate #39.
My experience agrees with what I have heard from Theo and others. I hope I am not being hyperbolic when I suggest that #39 helped me to hear where my crossover was going wrong or maybe it all just became obvious to me as a coincidence?
I would not go back.
Thanks, as always,
Rick McInnis
I will be the first to say there are some glitches with the H55m set up/i3/1GHX. Listening to B&W/Very Audiophile New Recordings, L22/300B/SHL5, simply superlative. My regards CIC, T.
No worries!
Edits: 11/02/10
Dear Cics,
After breaking -in B39 ,in my set up, showing not just improvement
but complitly different level of SQ. It is more like live presentation
3D as never before. My cMP still not broking in ,I rebild a sistem -new MoBo , CPU, Juli@,Buffalo,Tube Buffer all in one box, PSU separate.
Cics, I see fundamental changes in B39 , look as you find something,
tell me if I am wrong, but my opinion - it will be turning point in whole project.
Thank you very much .
Gene.
I'm very happy with b39's performance and have the same experience as yours, i.e. live 3D presentation. Strings, vocals, bass etc.. are presented effortlessly and I can't help but want to up the volume.
The fundamental change was achieved in b38. In b39, I wanted revert back to the generic ASIO interface (i.e. reverse b37 ASIO change). Intel provides limited low-level information on how instructions are implemented across its processor architectures (Nehalem, Netburst, etc.) - there are often big differences here for complex instructions. Hence some trail and error was needed. I suppose the breakthrough we get in b39 is confirmation of which ASIO interface is optimal.
Jitter distortion remains the dominant digital scourge. Taming it has big rewards. It never seizes to amaze how the slightest improvement can be readily heard. The math suggests we have to get jitter below 8.5ps (J pp RSS). There's more info here . Playback software that manipulates data for better SQ worries me as such approach at best acts to "compensate" (some vendors prefer using the word ameliorate ) for jitter distortion effects. It's like using a software equaliser.
Cics , thanks for answer. B38 was very fine release . I got crash
on my cMP ( Mobo) so , I rebuild and upgrade ,starting from B38 new set up.
B39 ,I install on same day it was released. Two day ago ,from now, I change it on B38, to compare ,and then back to B39.
On B39 - music just more real in my sistem ,noise level drop dramaticly giving up place for information and making silence pause on track ,realy attractive .
Curently using SE set up , but Bufallo Dac and Tube Buffer connected in
Balanced confg.( use one side now) waiting for balanced tube amplifier for HD800.
Picture later in cMp thread.
Thanks again, Gene.
where cPlay does not respond after stepping the balance several points I found this occurred yet still with this release.In this instance I had selected +1.0R then when bringing it back to 0.0 it locked at 0.2R. The selection continued playback even beginning again cPlay being set to "repeat". I had to reboot the machine.
I did uninstall the previous build before installing this and rebooted before first playback.
If it would be of any help I can try it again and post all the diagnostic reports.
Edits: 10/30/10
Unfortunately I can't recreate the problem on my side. Will give it another try...
I discovered the same problem last night, seemed to lock up but finished playing the selection.
Question/ My keyboard has never really responded to the commands. With this version I get balance and Pause is there something specifically I have to do to get the other commands, volume etc. T.
No worries!
for this feature. I am using the SSE2 version. I did find it occured again. Regardless of how gradually I returned the setting toward zero at .02 the player stopped responding while the media played out. Just figured I'd offer the feedback. May be SSE2 platform specific?
Edits: 10/31/10
yeah !
there is no doubt in my mind that this latest release is the best so far, but apart from it´s return to a less vulpitous tonal character as the 38 indeed is, the transparency and soundstage is just staggering !
the definition in the entire spectrum has increased tremendously.
on my system most noticeable in the top-end airiness, but also attack and transients seem more "flowing"
one of my initial feelings was that piano forte has a slight "edge" in the upper registries, but that could be a disguised problem from somewhere else revealed in this issue.
we are stunned again, cics !
Thanks CICS for another awesome release. My system loves this release. I am getting a very open airy 3d presentation with a very precise but relaxed natural top end.
It's been a great couple of evenings.
Last evening a pair of 6n2p tubes arrived in the mail to replace 6n1p's in my tda1541 dac which caused the sound stage to explode. At about 11.25pm I popped onto your site for the first time in weeks only to find you had dropped this wonderful gift for us all to share. Tonight a wonderful experience getting to know 2.0b39. Thanks again!
Hi cics'people
I'm get on this fantastic carousel 6 months ago, with the 035 version.
From that time I've found that every new release was better than the previous one (I've tried always the sse4 version).
This time, with 039 and in my setting, I've hear a sensible downstep compared to the 038.
With the 038 the sound was live, detailed, with a fantastic "silence between instruments", a lot of headroom and ambient informations and a great detail of high frequencies without any sign of fatigue and a big and tight bass too.
The 039 one seems to me like the 037, very good when listened alone but not comparable with the 038 by the slight "veil" of both instruments and voices, with fewer ambient informations, with a more artificial reproduction.
And this is for me a sensible difference, not a light one.
So I'm come back to the 038
Daniele
PS: I'm very happy with cMP2 but can I hope in a future a little improvement of GUI of the c shell, even with a little little cover's image?
Hi,
I completely agree with you about 2.38.
I believe it is the best vesrion of cplay so far..
I apologize about my previous post on 039,maybe too hasty.
After so large and positive opinions I've tried it again.
This time I've installed 039 in the night and I've leave my transport in repeat around for 10 hours.
I cannot understand why this time 039 is absolutely better than 038.
Could be a program needs to burn?
The only hypothesis I can advance is about some my mistakes in setting of cplay but it seems strange to me.
Conclusion is that 039 is now playng very very well, better than 038.
Daniele
Hopefully you are not using a 'hardware transport'?
Excuse me for asking Theob
what does it means "using an hardware transport"?
maybe I'vnt understand your question.
I'm using always my cMP2 transport (MB GB H57M USB3, i3 540, Corsair SSD , Corsair DDR3 1333 XMS3, linear psu on soundboard,P4 and USB, Seasonic X650 PSU)with a full software and bios cics'setting
Daniele
My mistake I interpreted transport as a 'real time disc reader' like a cd player or dvd player. If you are using cmp^2 then by definition you are using cplay to generate a bit stream.
Likewise for B37, B39 didn't last a whole album for me, before reverting to the previous versions. I am also back to B38.System adversion to odd numbers released???
Edits: 10/29/10
.
Is it not possible that it is not at all due to burn in of the hardware. I suggest its about our brain getting used to the new sound. It is well established that our brain does a lot to filter out the sounds bouncing around the room so we can focus on the primary sound and create a sound stage in our minds. I recall it described as a survival skill so we could tell where the lion is even when we are in a cave with all the reflected sounds we can still tell the location of the danger. I suspect when we go to a new release, it takes some time for our brain to retrain itself. Even with a more transparent and natural release we don't instantly percieve it as a better sound. It takes some time until our brain says... ok so this is the new deal. Now that I've adjusted it's actually really great sound.
This burn in period would be even more pronounced when the change is more pronounced.
Just a thought..
Is it not possible that it is not at all due to burn in of the hardware?
I'd argue that it is, to put it mildly, extremely unlikely that perceived changes in sound over time with a new version of a program such as a music player are due to "burn in".
The only exception I can think of is if the program is run on a brand new computer - the improvement over time in the performance of components such as electrolytic capacitors on both a life-cycle and use-cycle basis is well known and quantifiable. That case is easily dealt with.
We are all more or less aware that the perception of complex phenomena has a temporal dimension even if we don't know how or why. Which of us does not regularly say the likes of "I didn't like it at first but it grew on me"? We are well aware that it's us that's doing the "growing" not the percept. For a trivial example of the opposite, try watching an indifferent movie twice on successive evenings.
If the differences between revisions of the same music player are not subtle - and thus elusive - does it not it rather suggest that the program wasn't that good in the first place? It is hardly surprising if they take a while to detect and longer to assess.
They will be more apparent with some material and harder to spot if the listener is tired or distracted. It takes practice (though if, like some of us, you've persevered with 50-odd versions of cPlay, you've probably had a fair bit of that . . . ). Besides, the performance of a music system changes over time especially if it is sensitive to the quality of the mains supply.
Sometimes, for whatever reason, listeners report that a new version sounds poorer than its predecessor on a particular system even though the new one is "better" going (correctly in my view) by the criterion of efficient coding. Over time, one may either come to like the changes or to conclude that they really are, in this instance, not for the better.
And so on. I think cics and others are right to suggest that people take time to assess detailed changes in a system but the need to do so can readily be found in well known psychological phenomena.
There is no need to resort to any notion of "burn in" as a function of minor changes in coding. Not only is it slightly absurd, it is in all probability untestable.
I'm confused about this statement:
'If the differences between revisions of the same music player are not subtle - and thus elusive - does it not it rather suggest that the program wasn't that good in the first place? It is hardly surprising if they take a while to detect and longer to assess'
Could you say more?
Could you say more?
There's not much to say, only that the better a program like a music player is, the less scope there is for dramatic change. More or less by definition, any changes are going to be subtle.
I confess that I tend to take those "birth-of-a-new-planet" reports about every latest version of cPlay with a pinch of salt though some new releases have been noticeably better. OTOH, I don't buy fmak's argument that the new versions are never better, just different. It is instructive to re-load a version from about a year ago - the differences are certainly marked. I haven't done it for a while though.
Those who say they hear no differences between players (let alone versions) are also of course correct. They don't hear differences.
HTH
Running B39 continuously for 24 hours, tiny buffer, SRC@145, and output@192kHz, I must report that there are noticeable improvement in the SQ over version B38.
Kudos to cic!
well, so smycita and me have had both this strange experience.
My setting is the same of him and the was the same of previous mine( SRC 145 db, upsampling @ 192, tiny buffer, AWE).
But above all it was'nt the first time of a changing of a cplay version but it's the first time I've got this "burn in" experience.
Now I'm curious about this point. Someone can explains me?
thx
Daniele
PS: the only thing I could made wrong is the use of ASIO4ALL (default in cplay)instead of Xonar ASIO bitperfect. Obviously I've tried again and I found a difference in SQ but not so huge like my first 039 test.
Yes, it happens to me too, but more on b38 than b39. It took me a few hours (10?) for b38 to open up and distinguish itself. For b39, I notice a favorable difference right off the bat, but it keeps improving during burn-in.
Purely on conjecture, I surmise that a new improved version would sent additional information through the system, new information that system had probably never handled before. Thus the burn-in period.
"Purely on conjecture, I surmise that a new improved version would sent additional information through the system, new information that system had probably never handled before. Thus the burn-in period." -
It's very probable explanation. I have thought of the similar mechanism too.
I was still intrigued about this strange behavoir.
Another hypothesis could be the incomplete uninstall of previous version. So, after some days, I've changed again 0b39 with the 0b38, using total uninstall software instead the windows uninstaller(http://www.martau.com/).
Now I'm sure about the total removal of old cplay files and.............
0b38 is better in my system or,to be more precise, I enjoy 0b38 much more.
It could be possible that my system is not so good to appreciate the new cplay but now I'll wait the 0b40 one.
Thanks so much cics for your incredible job, you have open my mind in the digital domain and stoked my interest in hifi too
Daniele
Hi,
After some days with 2.39 I switched back to 2.38 and it sounds a lot better to me.
My MB is intel D510MO, 1G corsair ram, Lynx22 sound card, Windows vista32
I use only one cpu (disabled HT and cpu2 in windows)
I set pci latency in bios to 192
default buffers is cplay and asio
windows32priorityseperation set it to 16
and all the other optimizations described by cics..
cic
Thanks for the pointers. Will do just that and see.
Hi Cics,
Thank you for yet another version of cPlay
Some initial impressions and feedback on first hand:
Compared to version 38: version 39 has incredible resolution and is extremely detailed from the low’s till the highs. Has much more information on ambiance and space. Is tuneful and has very ‘fast’ transients. But also at the same time version 39 stays very natural sounding (in my setup).
Because of my harsh sounding living room, the tweeters in the Klein & Hummel O300’s, the Mogami cables and the Lavry DAC, my setup is always ‘a bit nervous’ and sensitive too bright upper mids & edgy highs.
Although the highs in 39 are very articulate, there is no sign of any sharpness or edges. None what so ever. The highs stay very ‘relaxed’ and voices sound very natural.
May be version 39 is somewhat ‘present’ and ‘forward’. But version 39 also encourages me to turn the volume up :-). So I expect too like 39 also in the long run.
As mentioned earlier: I didn’t like 37 very much. To me version 39 is in line with (the very good) version 38 but with lots of ‘good things’ added.
Thank you Cics for your work and passion and version 39.
fully AOB optimized cMP2 PC -> Lynx AES16 -> XLR AES/EBU -> Lavry Black DA10 -> XLR Mogami Gold -> Klein & Hummel O300
In my system version 2.38 is by far the best version.
In fact for the first time version 2.38 is much better than any of the cd players I own.
Thank you Cics!!!
"...2.38 is by far the best version" - May be you have confused the version?
Hi,
no I have not confused the version. I mean that 2.38 is better than 2.39 and any older version I have tested.
I concur as well. I wish I could articulate what I hear as well as you and others, but I know it's right when I read your description.
Mark you said:
'Compared to version 38: version 39 has incredible resolution and is extremely detailed from the low’s till the highs. Has much more information on ambiance and space. Is tuneful and has very ‘fast’ transients. But also at the same time version 39 stays very natural sounding (in my setup).'
I absolutely agree I could not have said better. And as Mark said these are preliminary comments. More to come after a few days.
It seems as if upper mid/high frequency noise level is reduced further below that of version 38. Very nice.
Well done Cics!!
This is about the best I have heard my system...ever!! Whoa, highs are very open, airy, silky smooth. I have been listening to many of my favorite discs (files) and they are magnificient. The sound is in a word very transparent. If something is off, power line noise or source material, you'll know it. If all is right then I hark back to all those dark days I could not get rid of that nasty noise or could not download drivers properly and I say ...yes it was all worth it, if this is the outcome.
I think some of the mixed reactions could be due to the increase in tranparency.
You'd think that after 30 some odd versions there would be no more sonic gains to be gotten but you (I) would be wrong.
Magnificient indeed Cics!!
Thanks to all for the excellent feedback.
B39 is excellent in my setup (with Tiny buffer setting). Everything comes together beautifully.
'Elvis in Memphis' is one of the recordings that has always been excruciatingly bad on the digital medium. While the analog LP sounded fantastic- huge sound stage, many layered instrumentation, with Elvis voice well placed in the mist of it all-the digital versions on CD (I have multiple versions the latest being the Sony Legacy) had always been full of muddy echos where all the sound, including Elvis voice, like being in the next room, with the door closed. cPlay b39 is the first player which makes 'Elvis In Memphis' sounded right! Elvis is now back in the room.
In hifi terms, comparing with b38 which btw was fantastic:
-- better definition and articulation throughout the spectrum
-- the echo effect in the high is now gone
-- elements in the sound stage is now well proportioned with better perspective
-- better high and tighter mid base
Thanks cics for the continuing effort to improve what is already a great player.
I'm a new convert to cplay thanks to the new drivers for the Asus Essence STX but have been using cmp for quite awhile now with foobar.
I have a full cmp2 setup with the recommended i3 + GA-H55M-UD2H + 1GB value RAM. I've tried to set the values in the BIOS where I can that other people are recommending. When I do this I'm getting a stutter when a new track loads when playing music. I have cryptographic services disabled, which is one of the tips I saw for someone having the same problem. I have also noticed this setup is way slower than when I had my e7300 + ga-eg45m-ud2h + 512MB RAM underclocked as slow as I could get it. I figure this has something to do with it. Is anyone else getting this?
Here are my settings:
VCORE =0.6500V
Graphic V =0.6500V
DRAM=1.3V
DRAM Termination V=0.705V
Mem freq is 600mhz and timings are 5-5-5-11. Cpu clock is 9x and QPI is 24x.
hey apokalypse,
I have the same problem on my cMP2.
even on some hi-rez flac files the stutter apears when cPlay is loading the RAM during playback.
a multiple ramloading is indicated by (02) or (03) after the track title.
the stutter appears to be shorter when less upsampling is needed, say when the file is native 192kHz for example.
I suspect the RAM of course or maybe the AWE setting.
maybe we can get cics to add a little "mute" between tracks....
kind regards
It sounds like it is flac.
Try the same file but as a .wav.
I had a nasty surprise when I tried a flac file. I had forgotten it was a flac and kept getting stutters.
Anyhow, here is a post he wrote that explains it
No one here remembers the bending of our minds
no, it´s not only FLAC.
also WAV is affected.
Sorry Playmate,
I thought that might be it.
Is this one very track??
No one here remembers the bending of our minds
I'm the same, wav is also affected, although the stutter is not as pronounced as with flac. Since cplay no longer loads the entire album to RAM before playing I think this post is probably on the money, it's a CPU processing power thing
Set frequency to 44.1kHz and buffer to tiny.
Sooner or later the STX new driver is gonna messed up your OS. I gave up on the ST long ago.
Maybe Daniele(audiodan@tiscali.it)can offer some inputs.
Good Luck!
Hi Apokalypse
I've solved the problems of new Asio 2 beta drivers for my Essence STX enabling in the BIOS both dual core and hyper-threading. In this way I've decreased safer CPU voltage value without any pop, crack or stutter.
SQ of this beta driver is absolutely better than the previous ones or the Asio4ALL. When you join this one with last version of cplay you'll can hear the heaven's trumpets :D!
I've also upgraded the soundcard replacing the originals OPAmp with the discrete Burson. It's easy and absolutely valuable change.( Next step will be add some Elna Silmic capacitors)
Now I'm waiting to assemble the new Gigabyte miniITX H55N-USB3 MB in place of the microATX H57M-USB3.
I'm very curious to compare them and to know if and how can change the SQ with this little and sparing MB.
I hope this will be useful
A full cMP2 with the Xonar Essence STX is a fantastic music transport!
Daniele
Thanks Daniele,
I'll try this trick. I thought enabling both processors would affect sound qualiCertainly a big difference when using cplay+Asus STX vs foobar+Asus STX. ty though? I will try it out anyway thanks.
Hi Apokalypse
don't worry about the worsening of SQ enabling both core+hyperthreading, it's a misleading problem. More impressive IMHO is the effect of new ASIO beta, if they works well, absolutely better than Asio4all or previous versions of Asus driver.
And I confirm you that the Essence with Burson OpAmp and Elna Silmic, plus linear ps on +5 and +12 Molex input, worth both cost and effort.
Next step: GD audio JZ-1 Type 3: (Output 24.576MHZ,use for DVD ,DAC, SB3), low jitter clock.
Essence STX is a very good soundcard, just a bit tricky to squeeze the best from it. But the best is really the best.
Daniele
First, is there a specific version of XP Pro that is the "best"? SP1, 2 or 3? Any other considerations?
Some of the recommended Mobo's seem to be outdated - what is/ are the currently available best Mobo(s)?
In the guide, it is suggested that an external computer be used to rip the songs to the player - this seems to be a bit of a challenge, as having a library of any size will require multiple steps to rip songs, is there a way to optimize an on-board CD-ROM to use with the main PC, to save steps?
Finally, (for now) the use of DSP to EQ certain songs, or even the whole playback system, makes a lot of sense - is there a way to add a DSP module/ solution to the system, again, keeping everything in the same box?
Thanks for the help!
...I find it very convenient to rip on my home computer by saving the file on a usb memory stick, then plugging said memory stick into my cmp computer and saving on the hard drive there. Many of us free our cmp computer of any online function to optimize its sonics. The above ripping / transfering process is very quick. As an example when I rip a 60minute cd into a 600 megabit file (takes about 10-12 minutes on home computer) it takes about 25-30 seconds to transfer to hard disk on cmp computer (assuming you always have one usb controller still active on the cmp computer). It also helps to have the two computers in close proximity.
Edits: 10/21/10
Hey Theob,
I used to do it the way you describe.
Though now I do it differently. I embedded the freedb database in my cmp2 box so you dont need the internet connection and use eac to rip from a usb cdrom drive.
I am not saying that is the perfect arrangement but I was concerned that the regular computer might not get the best rips due to the lack of optimization and power supplies. The usb approach is supposed to be better so i didnt have to put a drive in the cmp2 box.
The only downside I can see for me is that now the cmp2 box can and does have different music than my regular pc, whereas the method you describe kind of had backup built in. Now I just have to pull the cmp2 hdd and update.
I suppose if you have a dynamic record collection it might be a pain to update the freedb database, but in my case, that is not a problem.
The other issue for most would be that accuraterip doesnt work. I think I was only getting that to work on something like 10% anyhow, so not a big issue FOR ME. YMMV.
No one here remembers the bending of our minds
Interesting. How big is the freedata base? Didn't know you can do that. I really don't have dual data bases. I rip and save directly on the memory stick.
hey Theob,
It think the last one was 696mb.
Here is a link to the tar files:
http://ftp.freedb.org/pub/freedb/
The trick if I remember is that you have to unzip it a few times. Just search for unzipping tar and .bz2 files
On cmp2 it goes pretty fast because there is no antivirus to check each file but on a regular pc, yikes.
Hydrogenaudio.com has steps on how to embed it if you search there...but it is easy, unzip the database and direct eac to the local freedb.
How are you backing up if you save direct to stick??
No one here remembers the bending of our minds
Hi DR. When I started cmp^2 I loaded 2 different 3.5" hard drives (now I have 2 2.5" lap top drives) so many of my files are on the 2 original hd's. Also I am mostly ripping my own collection so the cd's are in effect backups. I have a few downloaded files but not many. When I complete doing my collection I guess I'll have to start backing up.
Any suggestions ?
so the cd's are in effect backups . . .
Once you've ripped more than, say, 20 CDs, the effort of re-ripping makes the original CDs a very inefficient backup. You have protection of sorts for the data but it takes a long time to restore. You have none at all for the metadata, which is mostly there thanks to your effort.
Remember that human error and/or OS gremlins are far more likely to be the cause of data loss than hardware failure.
I guess I'll have to start backing up. Any suggestions?
Yes - two of them. Do it now and keep it simple.
Buy a USB drive with enough space to store your collection and then some; copy all your music onto that drive using Explorer after installing, if you're wise, the free TeraCopy plugin.
(OTOH, if you can bear to use the command line, it is quite a bit faster with big transfers.)
That's it. Keep the backup in synch with your main library using e.g. AllWaySynch as and when.
** If you don't backup and you do lose data, please, please let me know as I'd hate to miss the schadenfraude.
D
thanks ...good advice
I think that the absolutely best configuration for cMP2 will be:
- mini-ITX system board Gigabyte GA-H55N-USB3 (advantages of this board were already described).
- CPU Intel Core i3-530
- sound card Lynx AES16e w/o spdif output (or RME HDSPe AIO with SPDIF digital I/O (192 kHz-capable) and with the stereo analog in/output (24-bit/192kHz)).
But if you have PCI sound card the recommended configuration is here http://www.audioasylum.com/forums/pcaudio/messages/7/76249.html.
dear forum,
dearest cics
I´ve already commented on the latest evolution of cPlay, but I feel the urge to extend or maybe rather confirm my excitement on this version.
there has been some very good upgrades indeed, but this latest one is just marvelous !
if subjective comments are of significance to any further development I can only attest that THIS direction is spot on. mostly because the version 38 gives instruments more body and real timbre.
as if there is much more poise to the decay of tones; much more presence and wholehearted statement of performers....this is not a "fat" or "tubed" emphasis, as some feel it being, but just much less pronounced on "attack"
the most pleasing effect on this understated version is it´s ability to keep depth of stage intact at even very demanding complex passages and no putting spotlight details or transients......
probably dull or "tubey" to some, but this version is seriously a major step in cPlay´s development.
congratulations cics !!
& kind regards
Pressing stop button causes an immediate but abrupt end to the sound.
Would it be possible to have a quick fadeout on stop?
Cics?!Thanks.
Edits: 10/02/10
The last few releases (since 2.0b34) wrestled with implementing a specific design objective at machine code (assembler) level. Getting Intel's compiler to do so proved to be a big challenge. Manually coding Assembler was ruled out. Albeit progress was being made, it was slow.
In this release that design objective at assembler level has been achieved! It took a small but important innovation to find this solution. Hence this unplanned release. Your feedback will be most appreciated.
Technical explanation of design objective: Refinements to both DSP and ASIO aims to minimise resource conflict arising from their concurrency. This CPU level conflict can be reduced to an absolute minimum but requires careful implementation.
Refer to ReadMe.txt for important details & specifications.
Change Log (2.0b38 release):
- Minor UI refinement to indicate balance control is engaged (affected balance button icon is changed)
- DSP & ASIO refinements
cPlay documentation can be found at the new cMP² website .
Please REMOVE previous versions before installing cPlay 2.0b38. Normal (SSE2), SSE3, SSSE3 and SSE4 versions are available. Running cPlay on a CPU not supporting the required instruction set will cause cPlay to exit immediately. For best results, uncheck "Timer" setting and test different "Buffer" sizes.
As from cPlay 2.0b33, the installation folder is NOT deleted if one or more plugins were used. All plugin settings are stored in files named with the corresponding plugin name and '.vst' as extension. Subsequent cPlay releases will attempt to use these settings. That is, cPlay releases can be upgraded without loosing plugin settings. On installation of a new release and plugins were previously used, the installer prompts to confirm reuse of the existing folder. Answer 'Yes'.
and I remain impressed by the quality of this version.
cics, I know you have surprised many times before but one wonders if there are any large steps still to come?
I sit at my listening position transfixed at what I am hearing.
Thanks,
I concur. b38 is an amazing release. Especially after fine tuning with the balance control
KC
Hi
thx a lot cics for your last work.The 38 is the best version of cplay I've tried.
I've found his sound a bit richer and fuller body than the previous 37, not less precise or "triode like", in my "all tube" system at least.
The only bias was that your release was contemporary to my assemblage of linear psu for P4 and soundboard, so I need some more time for a final opinion.
In the next days I would post my cmp2, another version of transport with analog output and differences in the hardware composition (It was born like an all-in-one transport)) but absolutely faithful to cics' optimized software instructions. It sounds really amazing in my "audiophile ultra-orthodox and ultra-maniacal" system.
My final though is about my opinion on cmp2: it doesn't sound nor analogic neither digital, It seems to me simply the best way to listen recorded music in home I've listen to so far: this is the true third way!
Cics has killed my absolute audiophile's certainty; another and bigger thanks
Daniele (italian cics' supporter)
dear forum,
always a treat to try any new "upgrade" of cPlay.
while some versions have different sonic characteristics and/or balance that may or may not suit different systems, I have mostly found newer releases better.
when a release turns "fatter" or "leaner" I have the advantage to adjust my digital crossover settings to compensate for this. the real issue is however how the software delivers the musical information and how transparency, timbre and air is handled. -and room information of course !
I also find the 38 version slightly fat, but my most apparent impression is that of transients and the decay of tones.....-wow !
-major improvement folks.
it gives all (especially darker instruments) much more body and individual substance....any musical note is ringing out in a much thorough way now.
....I would not call this "slower", but it definitely seems that more information is passed on.
-as if cPlay is not in the same "hurry" to get on to the next tone !
congratulations to this significant release !
38 is my lucky number! (Well, one of them..)
I installed 38 and sat down to listen. Sounded real good. So, I reinstalled 37 just to check.. Indeed 38 is far better. Timing and coherence is improved. Better sense of presence and very natural sounding. I did not like 37 but I like what you have done here. Well done!
My initial reaction to cPlay 2.0b38 is that I find it a bit fat. Yes to use Rick's analogy it sounds like a triode amp in the positive/negative sense. I love the smooth highs but miss the expansive soundstage/depth of cPlay 2.0b37. Dynamics also suffer a bit. 38 is smoother than 37 I will say though.I am listening to full src/tiny buffer/192 sample rate. I also use ssse3.
Edits: 09/10/10 09/10/10
I thought the term for that was 'phat'!;)
I'm curious to listen to v38 this weekend. I had a few minutes with it last night, but the system was not warmed up at the time and I don't make critical judgements when it's in that state.
Will try both SSSE3 & SSE4 versions.
Later!
Greg in Mississippi
Everything matters!
Edits: 09/10/10
I did a quick bit of listening & some swaps between B37 & B38 (and even B31, an old favorite) last night. I'll repeat them tonight to confirm my impressions and post a longer analysis tonight or tomorrow morning, but I'm back to B37... In my setup, B38 never gelled musically.Greg in Mississippi
P.S. Forgot to mention I did not take the time to play with buffer sizes... I've had best luck with tiny in the past and stuck with that, so I'll try that too when I re-swap tonight.
Everything matters!
Edits: 09/11/10
In the first hour of listening, the sound was way too fat for me, b37 was so detail that I keep wanting to go back. The I realized that the detail is still there, and may be even more, in a different perspective. And after the first hour, the sound seems to open up. Am I condition to the new sound? or may be even software requires time to burn in?
Will do more listening today.
KC
Ok, I'm back to trying B38 again... more successfully this time.
Before, I got a lot more PRAT and sense of the music through B37.
This weekend, I made a couple of tweaks that provided more transparency, detail, and life... I damped the heatsinks and the clamps that hold the various devices to the heatsinks on my amp and installed several Alan Maher CBF's on my cMP's transformers (ATX-20 5v, & both dirty supplies, these are the only ones where I can fit the filters in my current setup) & on the transformers on my amps.
Sound was definitely better with B37, so I thought I'd try B38 tonight again. It now had good pacing and made musical sense... and the additional detailing and textures were nice to hear too!
Morale of the story is... just do it! I've been thinking I should do that damping since I got these amps in late 2006, but since they came from one of the more rabid tweakers around, I figured there wasn't much I could do to improve them further. BZZZZT... WRONG!
cics, I'd say thanks even if I had decided to stick with B37 as I know you are advancing the art, but now I can say THANKS!
Greg in Mississippi
Everything matters!
Theo you should switch to SS4.
Yes there's much better harmonic richness with b38. With SSE4, you should regain some of that dynamics and transients performance. I'm using medium buffer setting.
I will try it.
But when I 1st tried Cplay 37 SS4 I found the same reaction with my system (fat (er...phat) and maybe a little slow) you suggested I try SSS3, I did and that solved the issues. Now if the same relationship still remains for SS4 to SSS3 won't trying SS4 result make my system sound fatter? Remember I am still using the 7200 cpu and laptop drives not ssd hardware. I know how you balance the dsp in all your various cplay versions and I have marvelled at how much that balance matched my preferences. Now that you are into new hardware maybe there is a difference in optimum balance for the old hardware.
At any rate I will try SS4.
SSE4 is very very good. More after I listen for a day or so. I guess my prediction that SS4 would be fatter than SS3 was wrong...it's just the opposite.
More later. :)
It took me a while to get the measure of this release...fiddlin' with this ... tweaking that but I think I have finally optimized the system with cplay 38 (sse4). At the risk of po-ing the 'proportions croud' I find 38 to be a little voluptious. No its not fat (as 38 ssse3), no it's not a-musical, it's just a wee bit on the side of honey dripping smooth. I don't think it is as fast as 37 ssse3 or as open but it is very musical in its own right. It images very well. The tone is very good and perhaps a strength of 38 but the speed and dazzling transients of 37 sse3 are a bit blunted.
Overall another fine release from cics, thank you very much again for your tireless effort.
I listen to 192/tiny buffer, 145 src.
After re-reading the above comments I feel I came off more negative than I intended. Let me say that I would rather listen to Cplay 2.0v38 than any other release. It is not perfect but it is very enjoyable to experience. Some of the variation in comments (from others and myself)are no doubt due to differing balances one uses for individual's bias's.
Cics is it possible to know what the break though was that allowed you to (I hope I have this right) release assemblor language version of Cplay?
This is a basic AMD system/Win7, with the known limitations (no SSE3/4) but the sound is richer.
Whaever it is, my digital Victrola likes it so far. Like you mentioned, there is a degree of voluptuousness that was not there before. At times, with SRC 145, tubes came to mind...not quite but closer.
I am back to liking SRC 145 (@176) while still finding myself switching to SOX for some music. Both exhibit the same change in character but now, either SOX lost some of its sharpness vs SRC, or SRC itself got sharper. Voluptuous & sharper ain't a bad thing.
In any event, I no longer get the impression of softness that SRC used to give me.
The sweet spots changed for this system. I went from Small to Medium as a common ground for SRC & SOX. Also, SOX at 192 no longer is musically "better" than at 176. This makes my life easier because I can't do SRC 145 at 192. With both at 176 it is easier to switch depending on the music.
Still tweaking...
Didn't think this would be the case but am amazed. SQ is as Daniele puts it, a whole new experience. Everything from bass to highs is natural with mids enjoying loads of space for expression. Beautiful.
Ha ha, it seems like my use of the term "sweet spot" in a different context launched a whole new thing on the new balance control. I only meant the "sweet spot" of settings.
In any event, I played a little with the balance control. It is good to have it for some music.
An issue that could turn out to be just my system: using the balance control can lock up my cPlay b38 here. It happened a few times in the course of a week while LOWERING a channel back to 0db. It suddenly shows "-0.2db" and locks up if I went too far down, it seems.
If it happenes to anyone else, here is how I dealt with it (USB sound card):
- cPlay will not turn off, it is frozen...so,
- MUTE the amps
- Launch the Task Manager and use it to kill the cPlay session.
- Unplug the USB card (that's what I have)
- Plug it back after a few seconds
- Run cPlay again. It it looks good...un-mute. ENJOY!
The issue happened on an AMD-based PC running Win7 pro, cPlay at 176 & 192.
Oh! It happened to me too at 0.2dB. Computer hang after that.
Can't even get to task manager so have to turn-off computer using power button.
Can't even get to task manager so have to turn-off computer using power button.
Same here. I assumed it was just me but it does seem to be a genuine "undocumented feature" . . .
Love the "undocumented feature" label!
Gang, let's add a little about the systems this is happening on. It may help Cics zoom in on where the "feature" is.
Cics and kclo, can the same result be obtained by moving your position in place of changing the balance? It should be so, or the "balance" is doing more than just that. Either way, a benefit is a benefit.
That was what I originally thought, since the balance control only adjusts the relative loudness of the channels, there is no need to bother. But prompted by cics's post, and the fact that I always thought that my right channel was a bit lacking in air--not a big deal, and the situation improves after a contact cleanup--there is no harm to try. This is what I did.
Increase right channel by 01.db, then played some recordings that were a bit dull or out of focus to me (e.g., Beach Boy's 'Holland', Elliot Smoith's 'XO', Elvis's 'Elvis in Memphis', Grateful Dea's 'Reckoning'; or any piano music you find dull) The improvement was like adjusting the focus of a video projector. Sharper, fuller, and a deeper soundstage. After some more listening, I now set the right channel to +0.2db.
No, you cannot get the same effect by moving your listening position
Some explanation is needed for this focusing effect, which I have heard myself as well, since it is different from shifting position nearer to one speaker than before (thus farther from the other). One thought is that unlike this or a typical analog balance control which decreases one channel as it increases the other, in cPlay one channel is made louder while the other remains the same, so overall volume is increased.
Follow discussion here on balance control. The "focus effect" is new to me and it would be interesting to hear more from Rick and Mihaylov.
My take: Speaker/Room interaction results in a unique frequency response curve (as each room is different). So when one channel gains some volume (due to room effects) over a wide frequency band, balance control becomes very useful (esp in the mid frequencies). Altering speaker position will change its frequency response curve. I suspect balance control gives us finer adjustment. Things get more complicated when listening off-axis. Another factor is what was done during recording/mastering. In this case each recording may have a "preferred" balance as well.
One thought is that unlike this or a typical analog balance control which decreases one channel as it increases the other, in cPlay one channel is made louder while the other remains the same, so overall volume is increased.
cPlay never increases volume. Volume setting determines maximum volume and balance reduces the affected channel further. For example, if volume is -20.0db and balance is -0.5 (left bias) then cPlay sets right channel to -20.5 whilst retaining left channel at -20.0db. This is similar to a preamp offering 2 (left & right) volume pots.
I am sure the other factors you mention can play a role. Still, according to your explanation, there is a volume effect only it is reducing overall volume with balance adjustments rather than increasing it compared to no adjustment, right? Also I am having a hard time seeing how a very small volume adjustment in one speaker can alter the frequency response curve in the room in any way like moving a speaker. When you move or turn a speaker, the whole geometry of the sound output changes not just volume. In this way moving one's position is more like that than adjusting the balance.
I think you are describing three related but different scenarios:
1. Moving your listening position, you are adjusting your ears relating to the sound field;
2. Moving the speakers, changes the interaction of the speakers output to the space and boundaries of the room, thus changing the sound field; and
3. Adjusting the balance, changes the relative loudness of the channels, thus modifies the content of the sound source, which then leads to a change in the speakers output.
I believe the more upstream the adjustment in the audio reproduction chain, the more effective--in the sense of perceive changes--the adjustment would be.
what was done with the individual recording.
A secondary effect, in my experience, is that it can correct a room placement problem.
I find most of the time "centered" sounds best but there are recordings that benefit from correcting the engineering.
With LP's and the vagaries of disk cutting there could be many more factors to correct, With digital it should be less varied from the actual production of the disk, though I would be foolish to think this is a given.
All controls are a blessing and a curse, like everything else on earth, especially if not used with care.
I think that if "balance" was able to be controlled at the listening position one would more easily hear the effect. I know we cannot have it all but I would prefer to be able to adjust balance at my chair than volume.
Any chance that a choice could be offered? Maybe one could pick a certain diminution of volume and simply adjust one channel up and down? My thought is that this would make the control easier to implement.
Sorry I cannot give any exciting psychology for this. It is simply a corrector in my experience.
Bye,
cics,
Thanks for the suggestion. When I increased the right channel by 01.db, everything slapped into focus. Records that sounded dull before (e.g. Beach Boy's 'Holland') now sound fine.
What a pleasant surprise!
KC
I have enjoyed listening to the evolution of cMP (I am recently back and this is my first audition of a new version since 37); once one takes away one's taste in tonal balance they have offered many insights into the performance. I have not been able to get too worked up about any one version (in any negative sense) since I have always been confident there would be another so one might as well enjoy what has been given and learn from it what one can. I have always trusted cics's ears and figured each step was just that, a step.
Hesitant to make my assumptions of goodness relative, I have found something good in all of them.
38 is definitely more of the single ended triode form of tonality which is very attractive since, in this case, it is not riddled with second order harmonic distortion. Very pleasant and very "musical" in the audiophile reviewer sense of the word.
In my system I find the bass to be magnified in comparison to 37. Tomorrow I will probably turn the subwoofers down a notch. Not much, but a little.
All I could think about was TONE TONE TONE. Maybe clarity has taken a hit in this versions rhapsodic devotion to tone. I can not complain. It is beautiful. I could live with it with no complaints. I do not hear any lack of rhythm, transients sound intact - no fuzziness, no discombobulation, the attack in intact, which has long been one of my favorite aspects of cMP squared.
I hear nothing I do not like.
To repeat, as beautiful a sound as I have heard from a digital source.
I do wish you would give an update on BIOS settings or, at least, assure that nothing has changed. And while you are at it ... are you using an SSD for booting and whose?
Thanks,
Rick McInnis
Thanks Rick & hfavandepas - great feedback.
I do wish you would give an update on BIOS settings or, at least, assure that nothing has changed. And while you are at it ... are you using an SSD for booting and whose?
Not much has changed to 32nm BIOS setup as posted. I'm planning to change site documentation to 32nm once there's more feedback - I see Mihaylov is also going 32nm.
I use a large capacity SSD - SuperTalent (still expensive).
Hi Cics,Thank you for version 38. When I heard the first note’s from version 38, first thing what sprang to mind was: ‘Yes ! Back on track !’
I never could get used too the sound of version 37. 37 had clearly more micro details in mid and highs than the very good 36 version, but somehow I could not listen too 37 very long. 37 was also: ‘in my face’ and very present. Always asking for attention
(Sorry but I don’t have better words too better describe what I didn’t like about 37. )
Especially dance & trance was so unlistenable that I switched too Foobar 1.1. too play my favourite Godskitchen Dance & Trance Anthems.I didn’t post these comments on 37 because I only own (a fairly good) mid-quality-level setup and all the guys with the high-end stuff seem too like 37 very much.
In the same period on 5 august 2010 Lynx released a firmware update for the EAS16, which also had quite an impact on sound quality (same as there latest 3 driver releases in the 18 version range had on sound quality). Unfortunately this Lynx firmware update didn’t do version 37 any good at all. Too my taste 37 now sounded even more ‘in my face’.Ryelands registry setting tip too decrease the USB polling frequency from 1 millisecond too 5 millisecond however did some good too version 37. Things bettered somehow but version 37 still was not my cup of thea.
But with version 38 I feel everything is back on track now. Too me it’s sounds like version 36, but than just a bit ‘better’ again. All the extra micro-details in mids and highs which 37 brought, are still there but the unpleasant ‘in my face feeling’ is gone. Too me version 37 was kind of a D-tour from version 36 too reach version 38.
Thank you for version 38. From now on Godskitchen Dance & Trance Anthems will be played with cPlay again.
fully AOB optimized cMP2 PC -> Lynx AES16 -> XLR AES/EBU -> Lavry Black DA10 -> XLR Mogami Gold -> Klein & Hummel O300
Edits: 09/09/10
Sounds interesting!
I enjoy for a long time using Tranport PC. I decide to translate the Tranport PC project in Turkish, for the people who dont know English. I prepare a how to guide in Turkish based on the Cics original texts and articles. Stereo Mecmuasi, publish the guide. Stereo Mecmuasi is a free e-mag and leading hifi community in Turkey. Editors say more than 1.000 Turkish enthusiasts download the guide in 2 days and a lot of people decide to build their own Transport PC. Once again i want to thanks to Cics for this amazing Project. Link for the project:http://www.stereomecmuasi.com/cihazlar/transport_pc/1_giris.htm
Edits: 09/01/10
I like very much the cic's project too. It is pleasant to me as this project develops, it is very interesting to me.
I have dared without permission cics to be engaged in popularization of its project cMP2 on open spaces of the Russian-speaking Internet having created the site devoted to this remarkable project.
I think it would be very good, if there was a possibility to create the multilanguage interface for cPlay and cMP. I can translate necessary words and phrases from English to Russian if cics will wish to make the Russian interface for his programs. I can add that the player cPlay is rather popular for Russian-speaking audiophiles.
Edits: 09/05/10
Thank you for translating to Turkish. A 1000 downloads in 2 days is amazing! As we get more enthusiasts onboard there's greater chance for insights and experiences from which we all benefit.
I started work on cPlay 2.0b38 which should be released soon. It's an unplanned release but I found something important which allows me to solve a complex design challenge I've been trying to implement. More later...
Wow! Great work there. Must had taken a lot of your precious time to do the translation.
Any reason for setting System Multiplier to 2.50 instead of 2?
Really, I have spend a lot of time for this job, but it was worth.
Unfortunately, my system is stable only when set to 2.50. I think this is because of the KHX8500D2K2 Kingston RAM module... Which do you think of HyperX module is the best?
I see that you are using the 2 modules kit (2x512Mb)? The manual recommendation is the use of a single module for a better sonic performance. Maybe after full Window optimisation you can remove one module and use just one 512Mb module(if it works that way).
My initial built was with a KVR667D2N5/1G which was subsequently replaced by a KHX6400D2N5LL/1G. I don't perceive any difference sonically with either of the said rams.
So I changed the ram to a KVR800D2N5/512 and notice a substantial improvement in the sound quality.
Only late last night I replaced the 512Mb with a Samsung PC2 4200U 256Mb. Didn't get a chance to listen yet if there is any improvement on the smallest available ram.
My suggestion would be to get a lower than 1G capacity ram. Size does matter, HyperX or otherwise.
Best Regards
Hi,
cPlay reports a problem with a VST plugin:
"Plugin could not be processed. Confirm selected plugin is certified with Steinberg"
What to do? What certification is cPlay checking for? How to get this certification?
Best regards, Uli
The error message is given when cPlay detects a program crash in the plugin. This is a bad thing and is a bug that needs attention from the plugin's developer.
Try testing the plugin using foobar. What plugin version you're running?
Hi,
the plugin (VST2.4, 32 bit dll) is running without problem with Foobar, Mediacenter and also on my PC with cPlay 2.0b34 SSE2. Another cPlay user has reported this problem, I guess he is using 2.0b37 right now. But Foobar on his system is again without problems.
Uli
Another cPlay user has reported this problem, I guess he is using 2.0b37 right now. But Foobar on his system is again without problems.
Get the user to delete the ?.vst file in cPlay's installation folder. (? refers to plugin name.) I suspect there could be a conflict in parameters and the plugin (this can happen when newer versions of the plugin are released).
It's me who got this problem...
Maybe this is a Windows 7 64Bit issue, as cplay was not intended for this platform.
Nontheless i tried all current b versions and got the same error message.
It works in Virtual XP under Windows 7, as i just tried to find a way to blame Win 7.
Hi Michael,
I was able to load the latest convolver vst into cPlay. I am using cPlay build 37, SSE4 edition, and windows 7 64 bit.
Not sure if you did this or not, the direction from convolver website say that you must add the installation directory to the path.
Fred
Refer to ReadMe.txt for important details & specifications.
Change Log (2.0b37 release):
- Clipping added when phase 180 and no upsampling is used. Recordings using max negative scale causes clipping error when inverted (e.g. at 16 bit scale -32768 is used instead of -32767). Loud recordings will suffer from this error.
- Implement balance control. See specifications for details. Buttons ( <,> ) next to volume display allows for balance adjustments.
On balance adjustment, volume of affected channel is displayed in elapsed time window. Settings panel displays status information.
- Minor UI refinements
- Minor DSP refinements
- ASIO interface refinement to improve support for Nehalem based CPUs (Core iX)
cPlay documentation can be found at the new cMP² website .
Please REMOVE previous versions before installing cPlay 2.0b37. Normal (SSE2), SSE3, SSSE3 and SSE4 versions are available. Running cPlay on a CPU not supporting the required instruction set will cause cPlay to exit immediately. For best results, uncheck "Timer" setting and test different "Buffer" sizes.
As from cPlay 2.0b33, the installation folder is NOT deleted if one or more plugins were used. All plugin settings are stored in files named with the corresponding plugin name and '.vst' as extension. Subsequent cPlay releases will attempt to use these settings. That is, cPlay releases can be upgraded without loosing plugin settings. On installation of a new release and plugins were previously used, the installer prompts to confirm reuse of the existing folder. Answer 'Yes'.
Revised sample code for Remote Control applications supporting balance control adjustments:
// cicsRemoteApp.cpp - trivial console app that launches cPlay in realtime and sends commands.
#include < stdio.h >
#include < windows.h >
#define REMOTE WM_APP+2
#define RM_PREVIOUS 219
#define RM_NEXT 221
#define RM_STOP 190
#define RM_PHASE 186
#define RM_SKIPBACK 189
#define RM_SKIPFOR 187
#define RM_PLAYPAUSE 80
#define RM_VOLUP 10
#define RM_VOLDOWN 11
#define RM_TRACK 12 //Set "lParam" to track # to be played (1..n)
#define RM_VSTEDITOR 13
#define RM_EXIT 14
#define RM_BALL 15
#define RM_BALR 16
#define CONSOLE ( "cicsRemoteApp" )
char CPLAY[ 1024 ] ( "c:\\program files\\cics Play\\cicsPlay.exe " );
HWND cPlayMainWnd = 0;
BOOL CALLBACK SetMainWnd( HWND hwnd, LPARAM lParam ) { cPlayMainWnd = hwnd; return TRUE; }
DWORD playSomething( char *mediaFile )
{ STARTUPINFO start;
PROCESS_INFORMATION process;
memset( &start, 0, sizeof( start ) ); memset( &process, 0, sizeof( process ) ); cPlayMainWnd = 0;
start.cb = sizeof( start );
CreateProcess( 0, strcat( CPLAY, mediaFile ),
0, 0, 0, REALTIME_PRIORITY_CLASS | CREATE_NEW_CONSOLE | CREATE_NEW_PROCESS_GROUP,
0, 0, &start, &process );
Sleep( 500 );
CloseHandle( process.hProcess ); CloseHandle( process.hThread );
return process.dwThreadId;
}
int main(int argc, char* argv[])
{ if ( !argv[1] )
{ printf( "\nUsage: cicsRemoteApp.exe \" < My Media File.cue|.wav|.flac > \"\n" );
return -1;
}
DWORD cPlayThreadId = playSomething( argv[1] );
if ( !cPlayThreadId )
{ printf( "\ncPlay failed to start.\n" );
return -2;
}
EnumThreadWindows( cPlayThreadId, SetMainWnd, 0 );
( SetConsoleTitle( CONSOLE ), Sleep( 40 ), SetForegroundWindow( FindWindow( 0, CONSOLE ) ) );
printf( "\nCommands:\t1..9=Track, p/n=Previous/Next, \n\t\tf/b=Forward/Backward, s=Stop, \n\t\tu/d=Volume Up/Down, l/r=Balance Left/Right, i=Phase, v=VST Editor, x=Exit.\n" );
for( int cmd = 0; cmd != 'X'; cmd != '\n' ? getchar() : 0 )
{ ( printf( "\n==> " ), cmd = toupper( getchar() ) );
switch ( cmd )
{ case 'P' : PostMessage( cPlayMainWnd, REMOTE, RM_PREVIOUS, 0 );
break;
case 'N' : PostMessage( cPlayMainWnd, REMOTE, RM_NEXT, 0 );
break;
case 'F' : PostMessage( cPlayMainWnd, REMOTE, RM_SKIPFOR, 0 );
break;
case 'B' : PostMessage( cPlayMainWnd, REMOTE, RM_SKIPBACK, 0 );
break;
case 'U' : PostMessage( cPlayMainWnd, REMOTE, RM_VOLUP, 0 );
break;
case 'D' : PostMessage( cPlayMainWnd, REMOTE, RM_VOLDOWN, 0 );
break;
case 'L' : PostMessage( cPlayMainWnd, REMOTE, RM_BALL, 0 );
break;
case 'R' : PostMessage( cPlayMainWnd, REMOTE, RM_BALR, 0 );
break;
case 'S' : PostMessage( cPlayMainWnd, REMOTE, RM_STOP, 0 );
break;
case 'I' : PostMessage( cPlayMainWnd, REMOTE, RM_PHASE, 0 );
break;
case 'V' : PostMessage( cPlayMainWnd, REMOTE, RM_VSTEDITOR, 0 );
break;
case 'X' : PostMessage( cPlayMainWnd, REMOTE, RM_EXIT, 0 );
Sleep( 300 );
break;
default : if ( cmd > '0' && cmd <= '9' )
PostMessage( cPlayMainWnd, REMOTE, RM_TRACK, cmd - '0' );
else
printf( "\nDon\'t understand %i.\n", cmd );
break;
}
}
return 0;
}
It took me sometime to come to an opinion about b37 because my system was powered off for 2 months, and it took 3 weeks of warming up to get back to performance.
With b37, I can no longer dissect it in hifi terms. It just brings me back to the same emotional level as my previous analog setup (Infinity RS-1, Conrad Johnson, Counterpoint, Sota+ET1+Decca Gold). I am still amazed by the fact that a set of medium size speakers that I am using now (Mirage OM6 + Sunfire sub) can bring me back to the good old days of analog. Thinking that there are still a lot that I can do to bring the music server up to spec, it is possible that I can surpass my previous analog benchmark. Certainly something to look forward to.
Thanks cics again for this great piece of work.
I cannot figure it out, but that is not surprising.
Is this possible?
.
even though if there was a way ...
setting this at the listening chair would be really nice.
I would gladly trade volume control for balance control.
Though I can easily see how that is not a simple trade-off.
Still can't get over how great the new software and CPU/MB sound.
I am going to have to work on my analogue set-up - I think this has, at least, equaled it in many areas and exceeded in more than I would have ever thought possible.
Who would have thought so much music could come out of a computer?
Thanks,
Rick McInnis
Sadly my main system is down as I finish a few audio projects like new ICs and a new speaker stand.
So I can just go by my regular computer with cplay and some headphones out of my m-audio revo 5.1's headphone jack...not high end.
But it certainly seems like v37 seems "slower". I am not saying it IS slower, just that there is more decay or something like that which allows MORE stuff to happen in the same period of time.
Anyone notice something similar?
No one here remembers the bending of our minds
Yes I do on sse4 but not on ssse3.
T,
Weird. I am using the sse2 version on this regular pc and that is what I hear from 37 when compared to 36.
looks like when my audio rig is running I need to try sse3 too.
Thanks for confirming that I am not that crazy.
No one here remembers the bending of our minds
My advice... try both SSSE3 & SSE4 versions (at least) and find out what works best in your setup, along with the various sampling rates and buffer sizes.
I installed SSE4 when B37 came out initially. When cics suggested SSSE3 & Theob tried it and liked it, I tried SSSE3 & didn't like it, but only listened for a few minutes. More recently, I got an email from one of my cMP buddies and he was really liking SSSE3 over SSE4. So I tried SSSE3 again. Here's what I wrote him:
"I listened to the current SSE4 cPlay, switched over to SSSE3 & listened again, then switched back to SSE4 & listened yet again.
I strongly prefer SSE4. In fact, if I only had SSSE3 available, I'd be looking for what to change in my system to make it sound better.
SSE4 provides a very 'analog-ish' sound in my setup... very liquid, present, and transparent mids, good, but very slightly recessed highs, and full and tuneful, but not the most detailed and defined bass. I get a more holographic and direct sense to the sound than I have with any previous version... I'd mentioned in posts before getting a sense that each recording is a D-to-D. This is strongest yet with SSE4. There's also a good sense of musical flow / PRAT. It is not as strong as it has been on some previous versions and setups (B31 & the last version of the Linear-Hybrid supply set the pace for this, pun intended), but still very good.
In contrast, SSSE3 added a slight emphasis in the upper mids and less transparency to the mids. Bass was a little better defined in the mid-bass, but about the same as SSE4 in the mid-bass, leaving the bass disjointed-sounded. Really it was disjointed across the spectrum and did not gell for me. All the dynamics & jump were in that emphasized upper-mids.
So back to SSE4, happily."
I'm definitely not saying that one is better than the other or that I'm right and others are wrong... I'm only saying that which works better may be very system dependent.
With all the power supply & other tweaks to my setup, I wouldn't expect it to sound the same as a more stock-ish setup. Very curious which versions Gene, Mihalov, & JackWong prefer.
So give them both a try & see which works better for you. As Grainger Morrison (who also posts on the Asylums, but I've never seen him here) says:
"Remember, YOU are the only one who needs to be happy with the sound of your system."
Greg in Mississippi
Everything matters!
Hi Greg,
After reading your post, I did install SSSE3 and listen it for 30 minute. Of course it was not broken-in , but I have a doubt it will be better in my cMP. I use tube headphone amplifier DV3222 with cPlay Volume control, maybe because of tube ,slightly lifted high on SSE4 over SSSE3 better for me. High, in B37 SSE4 faded very nicely as never before, close to analog. It not so grainy on small level as usually may to be.
So, I am very happy with SSE4 , thanks Cics again, he know it is breakthrough!
Gene.
dear dawnrazor,
in the contrary, I would say.
in the previous versions I have had a disturbing distortion at the very beginning of every track, as if the first buffer sample was not complete (nonsense, I know).
the 037 version does hardly (only very little) exhibit this.
I have however never figured out if it actually is the allocator VST causing this.
from a sonic point I find the new version displaying more room information and is a tad more holographic.
-as if the decay, and the ringing out of guitars and strings give instruments more "space"
....but "slower" ?
:-)
Perhaps that is not the best description.
There is just more stuff there to hear so it seems like it is slowed down so you can hear more. I am not saying the speed is off, just that there is more decay.
No one here remembers the bending of our minds
Hello, everybody. "Been away so long, that I do not remember what it's like..." I wonder that nobody has left any comments conserning WHAT REALLY MATTERS - namely, the sound quality of the b37 release. B36 left me with somewhat mixed up feelings with the highs being slightly subdued (in my oldish setup). Now for the listening. Will report lately if my impressions with biostar mobo are of any interest. Will be using SSSE3 version. Don't see why we need balance control, though. Also, if anyone is interested, b36 creates "impossible" orange condition with asio4all about every 7 minutes and the player hiccups. But The MIDS!!!
Serge.
but once the recommended latency adjustment is made and CRYPTOGRPAHIC SERVICES suspended the sound is amazing.
I know there are always improvements possible but I could live with this forever (of course, a month from now I could feel differently).
For the moment I am staggered by how good this is with the new MB and CPU.
You just might be surprised how important that balance control is when you use it. Either for room correction or correcting the recording, you will find the balance control to be akin to focusing the lens of the camera.
When you get it right there is a very worthwhile improvement.
Excuse my gushing, but it is sincere.
You just might be surprised how important that balance control is when you use it. Either for room correction or correcting the recording, you will find the balance control to be akin to focusing the lens of the camera.
Let me start by thanking Mihaylov for suggesting this very important feature. I consider this another one of those beautiful gems in the cMP² journey. For some reason, I considered balance control only useful for shifting the centre image. Your experiences prompted me to test this for room and toe-in correction. To my surprise, the "focus" aspect really works!
I'm still experimenting and hope to find a "sweet spot". Of course this will differ by recording however room characteristics should be addressed to some extent. On toe-in, JMLab's Nova Utopia Be speakers are very sensitive and 0.1db adjustments allows for finer control. A speaker's off-axis response varies considerably (in the order of a few dbs). I see balance control as a useful way to correct for this (esp in the all important midrange frequency band).
My reluctance on using balance control goes back to analogue days where any added signal overhead causes deterioration. Digital balance control as done in cPlay adds no additional overheads!
kind of a pain, but better than having two pots in series.
You only needed to fiddle with one to get the balance right, so it was not as bad as it seems.
I was surprised how few audiophiles took to the idea of a control for each channel
Yet, again, I am amazed at what has happened with this system since I last used. How the latest software works with the new platform is a giant leap.
Last night, after making some BIOS changes (still working the voltages down) I sat down to listen and thought how bad my amplifier must be sounding before warming up and then realized I had not hit the RIP button to disable CRYPTOGRAPHIC SERVICES. As soon as I did that all became as it should. That damn CS really screws up the sound!!!!
Thanks for making it easy to disengage. Along with many thanks for the rest of this treasured gift.
When you get a chance would you post what you have found to be the ideal BIOS settings. As I have read over the posts there seems to be a lack of specificity as to what is best. Have you come to any conclusions as to what seems best to you?
Thanks, again,
Rick McInnis
It has been two years so I cannot remember what I was hearing before and I have made numerous other changes, new DAC, I2S connection, powering JULI@ with a linear power supply, and of course, the new CPU/MB combo, but I am sure this sounds very fine.
After my LP immersion I was worried I was going to think otherwise, but if anything cMP squared has even more perceived space than the LP. There is a slight upward tilt in my perception of the tone but not to an obnoxious extent.
There are still tunings to be done. I was so excite about all of this working after making so many changes all at once that I just sat back and listened to music that had been unavailable (not on LP) and enjoyed it very much.
I am sticking with what has worked for me in the past, -145 dB upsampler at 192 kHz. Tiny cache is a new choice but that is what I used last night.
I did try no upsampler, but as before, and it was not an extensive audition, with my DAC, letting the computer do the upsampling sounds better than the DAC doing it. I do not doubt others may hear it differently.
Next step - making sure all of the cics recommended setting are in place.
I am still a bit confused about the PCI LATENCY tool. Has this been settled? Does this need to be done? I would greatly appreciate an opinion, even three.
I wonder how many have replaced the clock on JULI@ and if the consensus is that it is worthwhile.
As usual my thanks to cics for what he has started and what he has maintained. This is great fun and this project has inspired so many to think about this problem: how to get music out of a computer.
Now to get linear supplies built.
Thanks to all who have kept this the most vital thread in all of AUDIODOM!
Rick McInnis
Cics, in my set up, I use E7300 ,battery (only 5v Hdd on AC)
after 5 hours break in, I can report - it is breakthrough !
To me it is sound more coherent, more liquid.
Thanks also for balance control, I did order Buffalo Dac
straight after your announcement , it was bottleneck for me.
I happy with your implementation of a balance but I agree
with riboge.
Sounds like you converted from your fully-linear supplies.
THANKS!
Greg in Mississippi
Everything matters!
Hi Greg,
Yes, I am almost fully on battery, still on midle of the road ,
set up not finished. As you know ,I use for Juli@3.3v(2) and MOBO 3.3v(4) battery. If cMP off , relay connect it in one bank for charging.
Charger- just a linear PSU 5v with resistor. I use Gigabyte
EasyTune 5 Pro utility for reading voltage 3.3v on Mobo. Juli@
battery last longer, so if I know Mobo 3.3v , I know Juli@ v not less.
I also use cap for them both, a lot of, will send picture.
For 12v CPU and 12v Mobo - Lead Automobile battery 48 Amp., with tons
of caps ,plus 1 Farad supercap for CPU . In time CPU make upsampling on
192Khz, 15Hz spike rise,on power supply, so Supercap help reduce it on 6db. From start of the project, worst spike was at -20db , now at -52db(everage noise -67-70db) on CPU feed 20mm from Mobo.
Ones again, EasyTune 5 utility show me voltage +12v Mobo.(and CPU)
If Voltage of CPU drop to + 10v ( cMP still work !!! ) Super cap start
flashing by blue light, signaling me , time to stop music!
I use 12v linear , ajusted to max ,13v for charging.
Linear 5 v 6A for Mobo , in use now ,as a regulator only, disconnected from transformer and take feed before rectifier from same 12v battery. I will use separate battery- not enough Amp, just for 2.5hours.
With two 12v auto battery , will try connect 5v HDD by regulator from 12v CPU supply , will see what happen.
BTW I boght Buffalo Dac and parts for Lampizator , hope, I can finish this hardware project and only listen music after all.
Greg, before wright this post , I listen music 2.5 hours non stop, you
gess why I stoped , If I will got improvement from Benchmark DAC 1 to Buffalo 11 , same ,as I got from b36 to b37 I will be happy.
Regards, Gene.
Gene,
First, sorry for the slow reply. Then, thanks for your very detailed reply! You've done a lot of work on your cMP since late last year. I bet it is sounding mighty nice! Would love some pictures.
I'm very intrigued by battery power for these setups. I've got a LiFEPO4 battery to try on my Juli@ digital... I plan to compare it to the Dexa regulator I've been using there along with a Paul Hynes shunt, and a home-built shunt sorta like a Salas (from DIYAudio). I've seen pros and cons on using batteries... my (sorta) plan is to take my AC-sourced power as far as I can, then try battery to see if it improves on a radical AC power setup. My main resistance to going battery is the inconvenience. With the AC setup, I normally leave my system on and playing all the time (except when there are electrical storms in the area). This ensures everything is warmed up and operating at it's peak all the time and also helps break in new parts as quickly as possible as I install upgrades. Enabling this capability would be more difficult with batteries, but Red Wine (see below) has this going with the LiFePO4 batteries, so this is possible.
I did find an interesting link while surfing last night... Red Wine Audio sells some favorably-reviewed amps, preamps, and DACs that are battery power. He just recently changed over from SLA to LiFePO4 and claims some significant improvements... see here:
http://www.redwineaudio.com/products/lfpv-edition#lifepo4
Then I think you'll like the Buff-II. I did a Buff-32 for another inmate earlier this year. I was sufficently impressed by it to get a Buff-II for myself... slowly getting it setup & installed. On the I/V/output stage, I had originally looked at the Lampizator output stage, but after reading some comments on DIYAudio about the design's suitability for that purpose, I'm going to try several different routes. First will be the standard Twisted Pear output stage, but with the release of their new Legato, I may switch to that instead of the Ivy-III. Then there are several other options I plan to try:
- Transformer-coupled using Onetics 1:1 transformers
- Thorsten's current-out DAC tube stage (I understand he'll be publishing an update to his original soon).
- The Pass D1-inspired FET stage by opc on the DIYAudio forums. This is a fairly simple 2-FET circuit that runs at +-50v and takes a heatsink that otherwise would be appropriate for an amplifier! See here:
http://www.diyaudio.com/forums/digital-line-level/154866-new-take-classic-pass-labs-d1-ess-dac.html
- The modified and upgraded 'Mojo' stage from DIYParadise as shown in this thread:
http://diyparadise.com/forum/index.php?topic=1157.30
Paul Hynes had a hand it it's design, so I expect it'll be something special.
Lots to do!
Finally, thanks again for de-mystifying the motherboard startup process for linear supplys. You provided the path for huge upgrades in the sonic potential of cMP (and other similar computer-based music systems).
Later!
Greg in Mississippi
Everything matters!
Congratulations on your multi-battery set-up. It sounds awesome. You will not regret going to a Buffalo. Not that I need to tell you but a 3.3 v battery for its digital supply is also awesome.
Hi Theob,
The fact is, I following yours footstep, DAC 1 => Buffalo 11 , so you
encoyrage me. Tell me please - did you compare Buffalo shunt reg with
battery? Maybe we need to swith on cMP forum.
Thanks, Gene.
If you mean the digital active regulator that came with the Buffalo32s yes the 3.3v LiFePo4 batteries just smoked the active circuit by a lot ...not even close. I think it was one of European inmates that tried batteries on digital and analogue circuits and he (they) said the analogue batteries improved a little over the active circuits but that the digital greatly benefited from LiFePo4 batteries. So I never tried the batteries on the analogue circuits. But with LiFePo4 batteries running both Juli@ and Buf32s it is match made in heaven.
Hi Theob,
Thanks for information, but question arise.
The is a two 3.3v digital feed - VDD_XO and DVCC.
Could you please say , what way do you use and what inmate recomend.
1. Pack of battery 3.3v for Juli@ , VDD_XO and DVcc.
2. Pack of battery for Juli@ and another pack for VDD and DVCC.
3. All separate.
Of cause last better, but maybe diffence not so big.
Thanks, Gene.
I believe I use the battery for DVCC on Buf32s(it is the only voltage input for the digital circuit). I use 3 LiFePo4's in parallel....one set for Buf32s and one different set for Juli@.
I will reserve place for battery in my cMP.
I believe the confusion is this:
- Theob has a Buffalo-32, which has the DAC and I/V stages on one board and only three power inputs
- Gene has a Buffalo-II which has only the DAC on one board with at least one power input, but up to five. This model is designed to use a separate I/V stage
Will write more later, but hopefully this will help.
Greg in Mississippi
Everything matters!
Your notice save a bit my time.
Cics, the balance controls are a real enhancement to cPlay. There needs, however, to be a lasting not just transient indicator of an altered balance. I am unsure of the restraints given the need to fit such an indication into the single line of display at the bottom, so I can only guess what might be possible: 1) when the balance is not at 0 change, the volume indicator might be split to show L/R volume level or the amount of + or - on one or both sides, or 2) you have right and left "carats" to change the balance, so perhaps the side that is increased can be in bold so that when you restart cPlay you can tell if you are at neutral balance or at an altered balance at the least, or some other way to indicate the state of balance.
Next to "Diagnostics" checkbox, you will get "volume (balance 0.6, left -28.1db, right -27.5db, phase 0)". This is also provided in Diagnostics window on startup.
In ordinary use one doesn't get a diagnostic window or need to open settings and yet when you restart or resume cPlay altered balance settings may be present but you don't see that and may not recall it. For the same reason the volume level is shown, etc, the balance should be shown if possible.
On my regular pc I installed cplay.
The bottom of the player has always been cut off and you have to drag the player up to just see the tops of the bottom control icons.
I kept clicking on what I thought were the volume buttons and finally I realized I was affecting the balance. I still cant tell 100% if I managed to get it back to center. Without a confirmation I will not know.
The specs say:
[-6.0,+6.0] in 0.1db steps
Left(-ve), Right(+ve)
(53 bit resolution)
How one starts at -30 with that I just dont get. Why not -.1 or +.1??
No one here remembers the bending of our minds
While adjusting balance you will see the exact value of the balance on the left (where you normally see progress of the playing track). If balance is exactly "neutral" you will see "> <", the indication is only there if you are adjusting the balance.
On the right you see (again only while adjusting) the exact values of the adjustment: "-1.0 L" - "0.0" "0.2 R"What is it I do not understand in your question about too little information?
More interesting to me is why the bottom of the player is cut off on your pc?
Edits: 07/31/10
Af,
I did figure it out, but I still dont get the -30 part. Care to explain??
ANd Riboge is right. You can listen all day and not know your balance is out of wack..well at least from cplay.
As to being off the screen, it always has been on my non audio pc. ALWAYS.
Even with different monitors cplay is off the screen and I cant change it.
ON the cmp2 rig there is no issue.
OH, I think I just had a revelation. The volume numbers are displaced when you do the balance and I think that is what shows up on the left side. I have to check when I am back home.
so maybe it is ok outside of the feedback only when you change the balance.
No one here remembers the bending of our minds
Hey I was right. The volume number gets moved to the left and as you change the balance BOTH sides change. This confused the hell out of me. Originally my volume was at -30. When I changed the balanced the left side read -30.1, then -30.2, etc as the right side showed .1, .2.
Now my volume is at -11 and it does the same thing:
I can tell cics never uses his volume control from cplay :)....it works really well when the volume is set to 0, but if you are using cplays volume control as I am it gets really confusing like the pict. And you can see how the top of the cplay screen is cut off as I had to move it off the screen to see the bottom of the player.
No one here remembers the bending of our minds
It would be best if the state of balance was indicated when you are NOT adjusting, say when you restart cPlay having changed the balance previously. The balance changes are carried over but you don't see that and may not recall it either.
I am still using the E7200 with accompanying mb. 37 sounds mid rangy and slow and heavy to me versus 36. I will give it a few more days to see if I warm up to it. Let me be clear it is subtle and not overt but I feel that at least in my system it is a step backward. A positive comment is that on my most edgy files it sounds less edgy and more acceptable.
Sorry.
I no longer have my 45nm setup and cannot comment on how B37 will perform. You should try different buffer settings and/or B37's SSSE3 version on 45nm CPUs - this will to some extent offset the changes done.
...I am very pleased. The sound not only tightens up a lot in a positive sense but it brings a new level of '...that album now sounds better than ever before...' to many of my favorite pieces. Better transients, better bass (deeper and tighter much more texture) and great ambience,air around instruments. What a difference SSSE3 made! I am running 192/src/-145 db/tiny buffer.I just don't know if its SSSE3 or cplay37 that made the difference. Last time I tried SSSE3 vs SSSSE4 (or is it 2 SS's) I thought SSSE3 was too edgy over time now I find it not only just right but just right and then some. Maybe the balance you came up with for the cpu/mobo is now softer to accomodate. Whatever, I like it.
Edits: 08/02/10
More comments after I listen for a day or so.
Funny how what we listen for and how our systems sound effects our impression of a change.
For me, my setup was tending towards a little thinness in the midrange and warmth areas with B36. I had stuck with B31 for a long time (up until getting the linear supply to work) as it gave me more musicality and PRAT than B32-B36. The linear added a little PRAT and a good bit of detail and I could enjoy B36 then, but as it broke in, it was sounding a bit thin.
Then B37 came out and I got more fullness and warmth. I'ma happy camper now!
Theob, As I've played around with different buffer settings, I find the most detail and least warmth in the midrange with the smaller buffers. I usually get the best results with 'Tiny' and that's what I'm running right now. In the past, going to a larger buffer gave less high-end detail and more mid-range warmth... If you haven't tried it, 'Tiny' might help.
Then I'll have to try SSSE3 and see what I get.
cics, THANKS!
Greg in Mississippi
Everything matters!
I have used tiny buffer ever since cplay 31. I agree with your assessment of smaller buffers tightening the sound. But 37 and even 36 with ssse4 sounded so loose in the bass that I had to change x-over points to accomodate. Now with 37 and sse3 I'm back to my original preferred x-over points. SSe3 sounds very very good with respect to transients/large classical pieces with lots of bass.
Great! Thanks!
can't wait to get it installed and to hear.
Knew you were getting close to something when you "disappeared"!
Hope all is well.
Thanks,
Rick McInnis
One question: I love the sound of cplay. Very clearly.
But I have to say that I want a bit more convenience. Ideally a Sonos or itunes tyoe of GUI with an Ipod or Ipad as Remote-Control.
I think some of the High-End-Producers like Naim or Linn are offering this today, has anyone done some comparison with cplay ? Anything which comes close ?
one have to look at the motives behind cMP & cPlay to capture the current quality of the project.
cMP was developed to strip a lot of windows inherent realtime problems. I believe cics choose windows because of it´s wide distribution and because he is familiar with it.
there is hardly any doubt that a linux based system can be developed much slim´er per se, and this is what linn and naim are doing in their players.
one of the sacrifies for the superb quality is to reduce some of the GUI overheads windows has, so unfortunately you have to develop a linux based system yourself; or pay the dosh for linn or naim...or live with the limitation.
I believe that with an optimized (battery ?) power supply and all the other modfications the cMP 2 system is on par or even superior to the naim.
-haven´t heard the $20k klimax DS.....
kind regards
My apologies if I offend anybody with injecting some Linux discussion; in no way am I trying to hijack a thread called "cPLay"
play-mate,
You don't have to necessarily develop your own minimalist distribution of Linux. And by develop, I'm assuming you mean stripping down an off the shelf Linux distribution (Ubuntu, Red hat) for use in computer audio.
There is a very tiny Linux distribution called Voyage Linux (http://linux.voyage.hk/) designed for small single board computers, such as the PC Engines Alix and the Sokeris boards. Of course, there are install targets for generic x86 as well, should you desire to install it on on your favorite Atom board.
Voyage Linux is based off of Debian and retains the apt package management system, which is especially convenient for installing packages on an embedded machine, e.g., apt-get install mpd alsa-base. Voyage runs off a compact flash as small as 128mb, and the entire OS runs in RAM.
I have been using an Alix 3d2 board with Voyage Linux and music player daemon for close to three years now. The small music player daemon runs on the Alix feeding a USB DAC, and the GUI intensive mpd clients runs on other machines on the network; in my case a Nokia N800 tablet or my Thinkpad tablet.
So, I'm injecting some Linux discusion into the cPLay thread is because I'm very curious how a Voyage/alix/mpd system would compare to the most minimalist cPlay system. I don't run any Windows machines at home, so I can't do any AB testing here. Maybe someone on the thread has done the cPlay/Linux comparison?
I've attached a picture of the Alix 3d2board. Specs: http://www.pcengines.ch/pdf/alix2.pdf
Also, has anybody attempted to install cPlay on an Alix 1d mini-itx board? It does have windows drivers. Link: http://www.pcengines.ch/alix1d.htm
hey nycparamedic,
many thanks for your reply.
there can hardly be any doubt, even with hardcore cMP2 fans, that a simple but effective linux should theoretically perform much better than any stripped down win OS..
the hurdle is rather the compatibillity to ASIO, the avalliable soundcard drivers, and the broad acceptance/ familiarity of windows in general.
-I think even cics will agree !
kind regards
QUOTE: "there can hardly be any doubt, even with hardcore cMP2 fans, that a simple but effective linux should theoretically perform much better than any stripped down win OS"
UNQUOTE.
I would love to think so too but the the word "theoretically" says it all! As long as there are so many items unsolved in Linux and so many different versions of Linux it will never get past that stage! For that matter: I keep wondering why Linux has to change its core so often (are they buggy?) and the biggest problem is that most distributions ate not compatible with each other!
There is a distribution of Linux for just about every job but NONE specifically for bitperfect audio playback like cMP! It makes you wonder if it is possible or....if only Windows users are interested in this subject? Are we (Windows users)a special breed?
hey audiofanaticapart of the "theoretical" advantage any operating system may have, the big issue is the driver of the device the system has to run !
writing software drivers for highly sophisticated hardware is not banal....it is the core of the device performance !!!
-and an easy way for a manufacturer to cut development cost.I´ve been seriously surprised about the quality gain by newer and progressingly refined drivers for my lynx two-b card that have been issued in the last two years (5 updates in that period).
I´m sure that this sort of (costly) policy by lynx is the dedication to a long-term professional market.likewise the qualitative upgrades of cPlay are the result of fine tuning of the DSP driving software.....
.....linux does simply not have that audiophile/commercial driving force on developing or refining the software that drive the D/A chips to the ultimate....its not all asio4all :-)
Edits: 07/15/10
I've been trying to get Linux working as a good audio-pc but so far no luck.
My biggest problem is that I cannot find a Linux audioplayer that can work with CUE files, the ones that say they have support have no full (read faultless) support. Running a program like Foobar under Wine in Linux is not an option!
cMP works very good with CUE and I love the simplicity of CUE files.
If someone can point me to a good distro that does what I want I will try again, for now I'll stick to cMP with Foobar.
I just read in a german hifi-newpaper that the sooloos system which is based on pc-technology, is superior to the Naim-System, which is why I asked as well...but Iguess, I need to borrowa Naim-System and see how it compares, was hoping that someone had the experience already.
hey multiblitz,
no doubt that what meridian is contributing to the market with the sooloos is along the same lines as linn and naim.
that meridian has choosen the windows route does not mean they utillize a full computer operating software (XP, vista or win7), rather a tailored CE system.
the minlogon modification of the cMP system reduces software overheads in a simmilar way.
it will take a little time and we´ll see much more of audioplayers like this in a much more affordable price range.
let me guess : squeezebox will probably offer a 24/192 version in a years time....
could such device output to a proper DAC with a firewire or I2S (forget S/P DIF) you could save a serious load of cash.
kind regards
dear forum,
does anyone of you use a convolution engine with cPlay´s VST.
I´m a little lost on applying/configuring/implementing/mapping digital crossover & FIR filters.
would be great to get it going....
thanks !
i am wondering if this most excellent cPlay can playback music library content by genre and/or by artist. also, can playback be randomized?
thanks for your response, jorge.
hey joerge,
on the website there is a link to a playlist editor. a little java program that should solve your wish.
kind regards
many thanks for the heads-up; i'll check it out...
be well
dear cics
I have decided to take the plunge into FIR filtering instead of allocator.
now is one of the questions if cPlay does upsample before the VST or after.
the convolution engine needs to be programed with a fixed sampling rate....
another of my concerns is about the mapping the different crossover channels:
do I need an additional program like "console" to map directly to the lynx card ?
how would that affect the cMP mode ?
kind regards
cPlay resamples before VST. See here for more details.
Mapping channels should not require console - plugin outputs are shown on VST Settings window.
Hi all,
Currently I’m using XP home as OS for my cMP2 setup.
So I cannot use the AWE-feature in cPlay. But recently I got a XP Professional CD.
Should I go through the trouble of installing and fine tuning an XP Prof install?
Installing a fresh XP Prof install, fine tuning and installing programs etc, ect will take 2,5 – 3 hours.
Are there any inmates who have compared the sound quality (SQ) between XP Prof and XP Home? Is there any?
And if ‘yes’: is it possible too describe, how big the SQ difference is?
More specificly I’m interested in the use of the AWE-function in cPlay.
Is it possible too describe the SQ improvement between using AWE and not using AWE in cPlay?
Mark.
N.B. I’ve posted a related thread on XP Home <-> XP Prof here: http://www.audioasylum.com/cgi/vt.mpl?f=pcaudio&m=76042
fully AOB optimized cMP2 PC -> Lynx AES16 -> XLR AES/EBU -> Lavry Black DA10 -> XLR Mogami Gold -> Klein & Hummel O300
Should I go through the trouble of installing and fine tuning an XP Prof install?
Of course, it depends on how you value your time but I'd suggest that the benefits of playing from RAM with AWE enabled are well worth an evening's effort. It's been a while since I ran without AWE on my main system but I've just done a quick A/B and would say that, yes, the difference is pretty marked.
It's also noticeable on my desktop box where I run cPlay daily without cMP but with AWE enabled. I had cause to compare AWE/not to AWE when trying to cure a fit of "the stutters" a week or two back (fixed by reloading drivers).
HTH
Dave
Hi Dave,
Thankx for reply.
So… in short:
In your setup using cPlay with the special AWE-method of Physical RAM allocation(only possible in XP Prof), gives better sound quality than the conventional method of RAM loading in cPlay?
Mark
fully AOB optimized cMP2 PC -> Lynx AES16 -> XLR AES/EBU -> Lavry Black DA10 -> XLR Mogami Gold -> Klein & Hummel O300
In your setup using cPlay with the special AWE-method of Physical RAM allocation (only possible in XP Pro), gives better sound quality than the conventional method of RAM loading in cPlay?
Hmmmm . . .
Is cPlay loading data into RAM using AWE better than cPlay as a "streamer"? Yes. For me, it's the "conventional" way to do it.
Before answering your question a few hours ago, I re-tested the two setups. FWIW, there's little doubt in my mind that the AWE technique is better than streaming.
Is cPlay with AWE better than cPlay using cMP to load data into RAM? A different question that I can't honestly answer. As I've always had access to XP Pro, as cics advocated the AWE method and as it takes only a few moments to set up, I just did it.
As it'll take you very little longer to install XP Pro and and try AWE for yourself than it would for me to compare the alternatives and give you a (probably useless) report, you can guess what my advice is going to be . . .
cics did explain why he thinks his AWE trick is a better approach than using cMP to perform RAM loading. As I say, I forget the reasons he gave and took them on trust.
I've posted on several aspects of the cMP^2 project where I don't agree with cics. (We can still talk to each other . . . ) It just happens that this isn't one of them. Why not just try it out and report back?
HTH
Dave
Hi Dave,
Thankx for reply.
Since it’s easy too check / un-check the AWE option in cPlay, I thought lets ask other inmates what differences in sound quality they hear in there systems.
But you’re probably right, it’s better to install XP Prof in a partition and test it myself.
fully AOB optimized cMP2 PC -> Lynx AES16 -> XLR AES/EBU -> Lavry Black DA10 -> XLR Mogami Gold -> Klein & Hummel O300
But you’re probably right, it’s better to install XP Prof in a partition and test it myself.
The difference between cPlay and Foobar on identical, more or less conventional setups is IMHO discernible but it's not "night and day".
What I call the "hairshirt" cMP^2 setup was always, as you know, intended to be used on a computer dedicated to audio playback. The cMP shell, the minlogon "tweak", AWE and all the rest are as important as cPlay for getting the best SQ.
Even if you prefer Foobar to cPlay, there's nothing to stop you using it with the cMP shell and doing all the tweaks (except of course the AWE one as Foobar can't use it). In your case that would probably be on your E:\ partition: you just need to edit cMP's .pth file. (With a couple of batch files, it is quick and easy to switch between them.)
Basically, I'm arguing that, if my experience is anything to go by, it would be worth your making the effort to upgrade your "audio only" XP install from Home to Pro and to run cPlay with AWE. It's probably not as big a jump in SQ as "minlogon" but it's still there. Also, if you like using Foobar, there's nothing to prevent you using it on the same optimal (optimised??) setup.
HTH
Hi Dave,Thankx for sharing your experience.
Although using XP Home until now, my I experience is exactly the same as your experience.
Being fed up with re-installing the OS every time when I ‘over-tweaked’ it again, I started using DriveImage XML for image backups and I also started to use 4 partitions. I realized only than, that the use of more partitions, makes it very easy to compare differently tweaked OS’s or compare any other differences in software used like: drivers, players, ect in various combinations. From this I experienced/learned that in my setup:
* already with some basic optimizations (as also given by most pro-gear sellers) very good sound quality can come from a Windows PC.
* use of ASIO, KS or WASAPI is absolutely necessary get good sound quality from Windows.
* the sound quality difference between ASIO, KS or WASAPI is very, very, very small.
Using one or the other depends more on synergy with other components in the audio setup.* when correctly setup: the difference between Foobar and cPlay is (indeed) relatively small.
That’s why I use Foobar most of the time.* cPlay only takes (a clear) lead in a highly tuned and tweaked cMP2 systems as per Cics recommendations.
Only when I’m in the mood for an ‘audiofile trip’ and want to relax on jazz or classical music, I boot into the cMP2 / minlogon partition. Although dance and trance (which I listen 80% of the time) also sounds very engaging (!) through cPlay on the cMP2 / minlogon partition* When I use cMP on different partitions, than ‘C:\’ in the PTH-file is changed accordingly.
* re-sampling and/or up-sampling doesn’t necessarily sounds ‘better’. In almost all cases it only sounds slightly ‘different’. I stopped using it.
I’m drewling over a Prism Dream DA-2 / ATC SCM 25-A setup. Once I have managed too purchase that setup, I will try re-sampling and/or up-sampling again. In my current setup I feel there’s nothing too win on sound quality when applying re-sampling and/or up-sampling.* although sometimes heavily argued, it can easily be heard that software (drivers, players, up sample algorithms, DSP-modules, OS’s, etc, ect) all can have (a sometimes very big) ‘sonic fingerprint’ on sound quality.
That’s why I asked on the forum if there are inmates who compaired XP Home against XP Prof. (although I know the difference in SQ between XP Home and Win 7 home is nihil)* From my own experiences on SQ with different RAM timings, different brands of RAM and using 2 or 1 banks of RAM, I know that DRAM has very slight but clearly positive impact on sound quality.
So I wouldn’t be surprised, if the AWE feature also has (some) positive impact on sound quality in cPlay.But indeed, best is: too try that myself.
Thankx for reply and suggestions.
fully AOB optimized cMP2 PC -> Lynx AES16 -> XLR AES/EBU -> Lavry Black DA10 -> XLR Mogami Gold -> Klein & Hummel O300
Edits: 06/17/10
* when correctly setup: the difference between Foobar and cPlay is (indeed) relatively small.
That’s why I use Foobar most of the time.
* cPlay only takes (a clear) lead in a highly tuned and tweaked cMP2 systems as per Cics recommendations.
Hf, I am wondering where I may have gone wrong if these two above are "universal" truths. I don't use cMP. So far, cPlay has always had an edge over Foobar. In the lastest versions (I am still using 35) the difference is even more marked. In Foobar, I use "file [memory] buffering". In cPlay, I use AWE. (BTW, they do make a difference here, in both).
I started with XP Pro and now I use Win7 Pro, which may or may not have something to do with it. Yet, no manner of tweaking has ever made Foobar sound better than cPlay in 2 systems here. I'd love to bring Foobar closer to cPlay, for the sake of convenience at times.
Hey J,
That has been my experience too. Foobar is decent but not nearly as good as cplay. Compared, foobar is kind of 2d.
As for convenience, that is WHY you use cmp!!!
No one here remembers the bending of our minds
Heck, after using a modified Playstation PS-1 for awhile with no display and no on-device buttons, just a remote with start, stop, skip forward/backward, and fast forward/backward, my cMP with my entire music library, a display, a touchscreen, and a pretty effective remote is the lap of luxury for me!
Greg in Mississippi
P.S. My previous CDP was a home-made Shigaclone and slightly better than the PS-1 in ergonomics with a on-device display and buttons in addition to the remote, but I still had to go find the CD, put on the damper, and put it on the player!
But it had a bad case of the uglies, as you can see!
Everything matters!
.
Is anyone out there using IK Multimedia's ARC room correction software with cPlay?
Please let me know your experience with it.
ok, bought IK Multimedia ARC, took room/speaker measurements and generated system correction vst plug-in.
When I use the plug-in with cPlay I get the following cPlay diagnostics error message when the VST loads:
Parameters do not match.
I am using the digital outputs (channels 3/4) of a Juli@ sound card with ASIO4ALL. The channels seem to change on the VST plug-in screen when I launch the player.
cics - any thoughts on why I am having a problem with this plug-in? I am using the ambiophonics plug-in with no problems whatsoever. Are all CST files created equally?
Thanks,
Jack
I do not have ARC so perhaps this is a long shot.
ARC is limited to 96 kHz sampling rate.
Does your room correction vst plug-in match your playback sampling rate setting?
Thanks Bibo01.
I've tried altering that parameter to no effect. I tried 48 and 96khz.
No change in cPlay response.
Jack
When you get the following error:
VST Error: Plugin parameters mismatch. Plugin aborted (try deleting above parameter file & restart).
Try deleting the ???.vst file being used (file is found in cPlay's installation folder). ??? should be the same name as your .dll you used in settings.
Background: cPlay restores VST parameters on startup. Each plugin tells cPlay how many parameters to load - it seems your plugin does not provide all paramaters when cPlay saves them, or alternatively, you have changed the plugin version and the .vst file must be deleted.
Thanks cics, I much appreciate the response.
I will try this. Does not the file saved in the cplay folder get over-written if it is altered at a later date? I am not pointing cplay to the arc.dll in the cplay folder, it is located elsewhere. Would I be better off to save it in the cplay folder?
Jack
cics, I got the ARC.dll to run. Your solution was the ticket! Thanks!
Another question - does cPlay or any other player that you know of, have the ability to run more than one VST file at a time?
Thanks again.
Jack
hey jack
cPlay can only handle one VST at the time.
it can however handle different outputs with different FIR filters in principle when specified configuration file and syntax.
I´m using convolverVST to map outputs for digital crossovering within cPlay in a still experimental state.
kind regards
play-mate, thanks for the post. I'm intrigued by the thought of using FIR to perform the digital crossover function with my 4 -channel audio card. can this be done using the digital out?
I have a complete digital system, using 4 channels in biamp mode. If i could send different frquencies to each of the stereo channels, that would be great.
Let me know if you want to pursue the dialog here or take it off-line.
Thanks,
Jack
Hi,
I am a big fan of cmp and recommend it often on head-fi.org . I have been wondering, perhaps we should try a new architecture besides computers? I have this iriver h120 mp3 player that has optical spdif output, it runs on rockbox opensource software, which like cmp/cplay, buffers to ram and idles the hard drive to save battery. I am GUESSING an open source mp3 player, even if tweaked for sound quality and higher hard drive and ram capacity, has certain advantages over a computer, 1. less power usage, 2. less software to deal with (especially with rockbox), and no fans, but disadvantages such as 1. weaker processor, and 2. much harder to find/expensive hardware.
Rockbox has an open source project trying to make an mp3 player from scratch using ARM architecture.
http://forums.rockbox.org/index.php?topic=6751.0
http://www.rockbox.org/wiki/LyreProject#Topic_discussions_on_Rockbox_for
If anyone has used rockbox bootloaded mp3 players, they know that it is EXTREMELY EASY to use. If we can create the necessary hardware for an ideal hard drive transport including an lcd screen, maybe we can combine cmp/cplay software with rockbox software, using the intelligent handling of music of cmp with the hardware mapping and UI of rockbox.
Here is a brief explanation of how new hardware is given rockbox bootloader.
http://www.rockbox.org/wiki/NewPort
Note: This is just speculation on my part, I do not know anything about electronics, but I really like the convenience of my h120 mp3 player optical SPDIF output... but with my glass toslink, it is still a bit inferior in sound quality to cmp^2 when my computer is given a power line filter. Thanks for such a great software cics.
Suggestions to improve cPlay/cMP.
1. To add for the cPlay's interface a output balance (if it is possible).
Then it will be possible to use the power amplifier without the preamplifier (that often it is more preferable) and even without a passive volume control (and balance), that quite positively affects quality of a sound.
2. To cancel scanning of libraries at each start cMP (the configuration of libraries should remain in the cMP's configuration) - it will save a lot of time at loading cMP (especially with the big libraries). In library change it will be necessary to press button Refresh.
3. To add for the cMP's interface (and only cMP, but not cPlay) the button "Cover"as the sort preference. I think that it would be very convenient for a music selecting.
Probably all that I have offered is nonsense. In this case I ask to excuse me.
Mihaylov,
CAn you clarify #1? I run direct to my amps without of any need for a balance control.
Do you need a balance control do to room issues???
#3. If cmp or cplay doesnt handle covers, how can you sort by them??
"Do you need a balance control do to room issues???" - yes.
"If cmp or cplay doesnt handle covers, how can you sort by them??" - I want that cMP (and only cMP) handle covers.
This can be developed via Settings panel allowing for 0.5db stepdown adjustments to either L or R channel. Suggested range is 0.0 to -12.0db. Would this work?
It would be magnificent!
I would think no more than 0.25 db per step, less still better. I cannot a imagine a room requiring 12 dB attenuation for either channel!
Between the room and recording balance it would be great to be able to fine tune a recording. I have heard how this can allow a record to snap into focus.
This would be a great refinement.
Have you installed new Core i3 32nm CPU using GB GA-H55M-UD2H mobo? Step change from 45nm technology is significant.
Do you see balance changing from recording to recording? If so, it would be useful to have this on main window (which would offer refinements on overall settings).
Dear cics,
Just saw your post.
I am debating over whether to make the change. Thought I would see if you were REALLY pleased with the new set-up. We all know how new can sound good initially.
I know on LP's the balance can change from recording to recording. I find this to me much more vairable than the room since they are intertwined as soon as the music starts.
Maybe some of the newer all digial recordings have less of this problem than the old analogues (which is what I mainly listen).
I think this is one of those things that you do not realize there is a problem until you have the ability to fix it. With 1n eight of a dB step I suspect we will be able to make a useful improvement to music reproduction.
NEAT!
Of course, having a visual indicator is nice but I do not have my screen showing when I listen so my opinion is not important here.
Hope you are enjoying the madness!
Take care,
Rick McInnis
"...it would be useful to have this on main window" -
Certainly. Then it will be possible to adjust a balance DURING playback.
seeing this answer makes me wonder if I understood what you were asking about the adjusting.
YES, one would need to be able to make the adjustment while listening.
One does hear a literal snapping into focus at the right point.
Thanks for clearing that up.
#1. DO a search for VST balance control. If you find a plugin, cplay should handle it.
Here is one I FOUND but never tried:
http://www.brothersoft.com/rs-balance-vst-319961.html
#2. We can all dream right??
Thanks, but I think that VST plugins are better be not to using at all for sound improvement.
Guess it is a moot point now if cics adds balance control.
THough it depends on the plug in. Some suck but others can improve things like room correction and crossovers.
DR: question for you can you please lay out how you install a vst plugin? Then how you use it? A real simple example (for us simple minded old timers) would be nice.
Hey Theob,
Always glad to help (hey, if I can help my dead beat boss who has the family in town, is off for the next four days starting saturday, and called me at 8:40am today to come to work because he wasnt feeling well (yeah right he wanted 5 days with the family), then I can definately help you).
Cics had a post about this I think, but I'll try to save him some time so he can get on that cover art :).
Lets use the balance one that I linked. Download and extract the file. You will end up with a .dll file:
Then in cplay, click settings (bottom left green lightening bolt??) and click the VST option that is next to ABOUT. Should give you a screen like this:
Click on the vst plugin tab at the top left and then find the dll file for the plugin you want to install:
You will then get the details of the plugin (some wont be compatible but that is rare):
Hit OK and then restart cplay. Note. Cplay remembers the last folder selected, so odds are you will have to change the folder from the vst folder back to your music folder to play music.
Now the vst button next to the stop button should be active and blue now.
Press it and it will bring up the vst controls and settings if the plugin has settings that need to be adjusted. In this example you will get balance control knobs:
There is a program called console that cics details in a post that you can use to string multiple vst plugins together.
Do a google for vst plugin. There are tons to choose from. Things like eq, room correction, crossovers (that is why I pestered cics to add vst...a big THANKS for that c), tube effects, soundstage expanders, headphone plugins, balance, and probably a ton I am missing are all things that can be added to cplay.
This is very very powerful and can add a ton of functionality. Though not all plugins are necessarily good soundwise, and some experimenting will be necessary for the free ones.
D
Oh, and in the settings menu in cplay you will see a BYPASS option next to the vst button. This allows you to listen with the vst bypassed and in the circuit, so you can ab it.
Also the VST check box is there as a vst safety as things like crossover can be deadly if you need to make changes before playing. There is no need to check that for most vst plugins.
Thanks DR!! Your steps are very clear thanks. Looking for vst plugins is cumbersome (i.e., so many with odd names/unclear functionality). I personally will need to comb over these to find someting I want.
I take it that for a crossover one needs a breakout cable from which comes multiple frequency band digital streams which need multiple dacs?
hey theob,
let me see if dawnrazor will at least share the "honour" of having got cics persuaded to implement a VST functionallity to cPlay....:-)
I´ve been using thuneaus digital crossover with cPlay for now 18 month, and since the VST function came avalliable, it´s very straight forward (sort of).
you will need a multichannel soundcard to do this, not a multi-breakout cable!
I use the lynx two B, as also dawnrazor does (his recommendation/thanks DR) but also the RME 9652 will work fine.
these cards have 3 dacs onboard and deliver the analog signal for the different amps.
kind regards
Well I guess I can't do that with juli@. Oh well thanks to both of you anyway.
Hey Theob,
A couple of thoughts about the Juli@.
First off, your profile shows a crossover for the bass right. I dont know how you have things but that usually means running the mains full range. YOu could use a vst plugin crossover to send a crossed over signal to the main speakers and perhaps take some strain off of things.
Another thought is that maybe you could use the analog outs of the Juli@ for the bass and ditch the outboard crossover you have. I am betting the analog outs of the Juli@ just might be better than the crossover you have.
Well I think it at least might require some extra thought specific to your system.
Yes you have it right. Here is my signal flow:
Pc I2S--> Buf32s--> XLR out--> y connector--> one feed to electronic bass crossover, one feed to high freq amp --> speaker external crossover --> speaker.
If I used a vst plugin to send a high pass signal to the speaker yes I could eliminate the speaker crossover but how would I get a full range signal to my bass crossover or even a low pass signal? Now that I said that I guess I would have to use analogue out from juli@. But juli@ only has one set of analogue outs.
Hey Theob,
This is more of an educated guess on my part at this point but I think it is sound.
You should be able to do something like use the digital out AND the analog out:
Pc I2S--> Buf32s--> XLR out high pass to speaker amp
PC Juli@ analog out---> Low pass to sub.
It would come down to the asio drivers, but I am guessing that such a mapping would work.
DR here's the thing: if I use digital (via I2S) out I can't use analogue out because the analogue board needs to be installed into main juli@ board via I2S pin outs plus other pins. So I can have one or another not both. I think I am doomed unless I get another sound card.
d
Hey thanks for your interest in trying to help me. Means a lot.
Hey Leif,
Take all the credit you want. All I know is that we have it now and it has changed cmp2 for the better. Come to think of it, I didnt know what a vst was until you told me :).
Theob...Play-mate details a way to use a crossover and get ANALOG out of the pc.
There is a way to get DIGITAL out of the computer. Using something like the Lynx Aes-16 and its multiple digital outs to multiple dacs is probably the best way to do it...but the cost is crazy which is why I went with the Lynx 2b (what a bargain for 3 high quality dacs).
Also, it might not exactly fit in with the cmp2 concept, but a middle ground between cost and crossovers would be some of the firewire units like the RME Fireface or the Lynx Aurora....
D
I would like to add another suggestion for cMP:
the possibility to select multiple CUE files for playback.
It would be very useful for multiple CDs albums.
Perhaps it is simple to implement at mouse program level.
Thanks.
I have some difficulties with b36 to control the resampling.
It is good to see whether the signal is going through resampling. Problem is that after I switch to 24/96, then I want to play some 88.2 file, I stop the music, went to setting screen change to 88.2 sampling rate. Then select the 24/88.2 record.
But it keep resample from 88.2 to 96. I then try to switch to 44.1 and load some 44.1 songs. cplay keep resample to 96. I can only solve this problem by close cplay and start it over. That is not very convenient.
I have asked for this feature before, is it possible to have a button for "always bit perfect" selection. When cplay plan a record, detect its sampling rate, reset the asio output to that sample rate and eliminate resampling.
Thanks.
Thanks for the beautiful software you wrote, Cplay and cMP !
I am enjoying it every day.
A few questions:
Is it possible to compile the software, so that it's full screen at a native screen resolution of 1280x1024 ?
If I am using the volume control, is it still bit perfect, or only at max (full) volume ?
Thanks for your reply !
This is a problem I've also found using CPlay with a "10 foot interface" - using the PC with a TV and sitting on the couch and using, say, 200dpi and/or a low res screen. I've found lower resolutions sound a bit better. I do Windows programming and this shouldn't be hard, just scale everything to the window size. Not everyone uses an LCD built into the case.
Is it possible to compile the software, so that it's full screen at a native screen resolution of 1280x1024 ?
It's not a compile option. One needs to manually change windows for that resolution. No plans to do this. On this resolution, try setting a higher "DPI" number under display settings > advanced (default is 96, recommended for cMP is 120) to increase window sizes that way reducing the border.
If I am using the volume control, is it still bit perfect, or only at max (full) volume ?
Bit perfect means same audio bits (from player) is seen by the DAC. There should be no loss of bits or mangled bits. So if a player sends all 16 bits of an audio CD, your DAC should receive these perfectly.
cPlay maps everything to much higher resolution (32 bits). Your soundcard re-maps this to the target output rate (44.1 would be 16 bits, all else tends to be 24 bits). So in the example, the original 16 bits are sent bit perfectly to the DAC (CD 16 > cPlay 32 > ASIO 32 > Soundcard 16) as the computation to/from 32 bits is trivial (bit shift by 16). Aside: all CPUs works most efficiently at 32 bits - think of this as our "native resolution".
Digital VC should not the confused with bit perfect delivery. All bits are transferred perfectly. VC simply reduces audio signal level. Each low-order bit reduction translates to roughly -6db volume reduction.
Where the fun starts is how digital VC is done. Early implementations did this using horrible integer math causing rounding errors (volume is logarithmic). This type of calculation causes resolution loss. Another common but inferior approach is to use 24 bit resolution (of 32 bit math). cPlay applies VC using 53 bit resolution (i.e. 64 bit math) onto our 32 bit audio signal. This way there's no resolution loss except for quantisation error (rounding error in the lowest order bit). For cPlay, this quantisation error or noise is at 32 bit resolution which is a theoretical one (noise is at -175db level!). Whether volume is at 0.0db or -30.0db, audio quality is not degraded - you simply get lower volume. When we do digital VC this way, analogue VC is at best equal to but never better. As a test I used my AA Prestige SE player's excellent ladder based analogue VC and compared it to cPlay's digital VC - no difference!
It's not a compile option. One needs to manually change windows for that resolution. No plans to do this. On this resolution, try setting a higher "DPI" number under display settings > advanced (default is 96, recommended for cMP is 120) to increase window sizes that way reducing the border.
I think that's an very ignorant and arrogant attitude of managing an issue that concerns most users that haven't got the Zalman case. There are just 3 rows of code necessary to fit the windows to any screen resolution and you will definately not suffer from worse sound! That is one of the reasons that keeps me staying with Foobar!
Another question: why don't you compile cmp and cplay to one program?
I guess I dont see what the issue is.
Sure, on my normal pc, cplay doesnt fill the full screen, and before I bought my 22" monitor, I had a 17" one and had to often move the taskbar down and cplay up to get to the controls.
But on my audio pc, it is a different story. I just selected the resolution that fills up the 17" monitor I have. No problems.
And I am guessing that cics is trying to have a more open architecture and allow for use of other players so to make it all one program might not be the way he wants to take it.
I can see the same critique coming to him if he did make it one program...
Also you have to remember why cplay was created in the first place. See IIRC cmp had been created and foobar is what was recommended....but foobar is not the best sounding player and cplay was born. While it can be stand alone, it was designed to be used in the context of a cmp2 rig, which IMHO it works great, fills the screen, sounds great.
I already tried the DPI change, but I want it full screen.
Is there any way for me to change it (reg hack or something ?)
Thanks for your explanation about how the VC works, much appriciated !
Don't know how to force a full screen via registry. Closest is to set DPI to give a large enough window and set your desktop colour to black. Although black on LCDs increases power consumption this shouldn't be a problem if its powered from a different AC circuit.
Here's more information on volume control:
@Rankena - I haven't tested whether DPI causes extra load. Given default is 96, setting to any other would require similar GUI scaling computations.
higher "DPI" need more system resources?Or Im wrong.
Hi,
only for better understanding: When cplay begins playing music, is the file (after loading to memory) already upsampled or does cplay upsample "on the fly" (the loaded unchanged file)?
Thanks.
Regards
Thomas
.
For some bizarre reason, whenever cPlay crashes (or is improperly terminated), it causes all my other programs to lose communication with the Juli@ card (games, foobar, itunes, browers, you name them). The only solution I have right now is to reinstall the Juli@ driver.
Obviously this isn't my dedicated cMP transport, but I'm so used to cPlay that I really want to use it on my "everyday PC" as well. Can anyone lend me some insight on why this is happening?
Operating System: Win7
Hardware: H55M-UD2H + i3 530, 2G DDR3-1600 RAM
"We should no more let numbers define audio quality than we would let chemical analysis be the arbiter of fine wines." N.P.
Hi,
Yesterday i started my "real" cMP² installation with the recommended hardware except the USB power seperation / linear psu and a fanless cooler, the stuff just didn´t reached in time.
So after i installed a nlite pre thinned out xp sp2 i tested with latency checker and got very bad results compared with the ons from the tutorial. I get 4 Rows between 5 and 50 ps and then peaks of 700ps realy strange, i set up alot computers so i doubt that this is a driver issue, i´ve installed everything correcty and even device manager is faultless.
Is it possible that it´s the bios ? Is the F6 bios from the gigbyte website the right one or are there another customized version or should i even flash the newest one ( beta ) ?
Btw, i played some songs this morning with the new cplay after finishing 80% of the software optimisations and it sounds awesome already. The question now is try a reinstall of a un nlited xp or flash the bios first to get a somewhat flat latency....
regards
Ramón
I´ve updated to f6, evrything seems fine now. Latency is in between 7ps and 50ps, much much better than before but still not perfect.
Is it just misunderstanding or is there a modded ( custom ) f6 bios ? Where can i get it ?
Hi Ramón,
I build several cMP’s, based on mobo GA-G31M-S2L & Intel E7200 and also mobo GA-G31M-ES2L & Intel E7300.
With either one of the above sets of mobo’s and with either both of the hereafter mentioned sets of tuning tips (set A or set B), you should be able to easily reach a latency of between 4 uS to 8 uS. Without any spikes.
* set of (light weight) tuning tips A:
see cMP website for
- chapter 06: Bios (without the under volting)
- XP optimizations found here: http://www.eye4systems.nl/pdf/tuning_tips.pdf
- chapter 09 Kernel
Or
* Set of tuning tips B:
See cMP website for:
- chapter: 06 Bios (without the under volting)
- chapter: 07 Optimizations: only 7.1.
- chapter: 09 Kernel
So only just checking and to be absolutely sure, I would like too ask:
-> did you already apply completely one of the above mentioned sets of (light weight) tuning tips?
If yes….than I’m out of options. :-(
(May be other inmates can join in with more/better tips).
Mark
fully AOB optimized cMP2 PC -> Lynx AES16 -> XLR AES/EBU -> Lavry Black DA10 -> XLR Mogami Gold -> Klein & Hummel O300
Hi Mark,Yes i´ve applied all optimisations except the Kernel ones. The Bios is optimised except smart fan ( need it until fanless cooler arrives ) and spread spcetrum ( does not exist in f6 bios ). The Ram isn´t optimised either, i´ve bought a nice Crucial module of Micron D9DQT second hand wich is on the way. I used this type of ram years ago for Overclocking and i hope to achieve a cl 2 setting with it. It´s beside d9gmh the best ddr2 ram chips wich was ever made.
Atm i´m running the cpu @ 1v. and 1.05ghz, tomorrow i´ll finish undervolting to lowest possible evaluating what the cpu is able to and do the kernel optimisation stuff of course. The rest is done already except the advanced tweaks. Perhaps i reinstall a newer chipset driver, i used the inf ones on the gb disc but i saw that there are much newer ones available online.
I think latency it´s mainly a bios issue, i need the modified one ( f6x ), i´m running f6 only. I wouldn´t wonder if this is the key to achieve very low latency.
Edits: 05/08/10
Hi Ramón,
Although under-volting and under-clocking is an essential part of fine tuning a cMP setup, it will not bring you’re XP OS at rest. With latency’s between 5 uS and 50 uS and peaks up to 700 uS, your XP is clearly not a rest.
It’s obvious that XP still is very busy doing al sorts of things. (That’s why I asked if you applied the XP tuning). But what makes your XP so busy in your setup ? What is XP doing, that it still has such high latency’s with high peaks?
Is XP still bussy with driving:
- unnecessary hardware ?
- unnecessary services ?
- or unnecessary programs ?
Or a combination of these three. ?
Only already with some (light weight) fine tuning, one can already bring XP at rest to latency levels between 5 uS and 8 uS. But your XP latency levels are still very high (and also with peaks).
So it’s essential that you find out what makes XP so busy in your setup.
If you already applied the XP fine tuning tips, than I’m out of options. But I do suggest you make an extra post for help ( as I don’t know any further).
Good luck.
Mark
fully AOB optimized cMP2 PC -> Lynx AES16 -> XLR AES/EBU -> Lavry Black DA10 -> XLR Mogami Gold -> Klein & Hummel O300
Hi Mark,
My latency is now between 5 and 7us with the f6 bios ( without x the non modded one ).
Now im doing the low level kernel optimisiations and then i do the minlogon. Does ram latency matters for dpl latency ? Within the next days i´ll receive 512mb ultra low latency ram, the ram i have now running is running in auto mode, no timings tweaked yet.
regards
Ramón
Hi RamónOkay that’s already much better. :-)
In my setup RAM latency did not matter for dpl latency. Biggest impact on dpl latency is: how bussy is your XP? With latency’s between 5 and 7 us XP is still not completely 100% ad rest, but it’s fine for now.
At this point I had in my setup already latency’s between 2 and 4. So I think that your XP is still driving some unnecessary hardware and/or unnecessary services. Hope you can find out what XP is still somewhat busy with.
TIP:
After you have applied Minlogon, it’s impossible to modify/change your XP installation if needed.I strongly suggest that you first make a copy of the XP installation at this point with Drive Image XML before continuing with the fine tuning process.
See http://www.techsupportalert.com/partitioning-hard-drives-3.htm.
Drive Image XML (DI XML) is part of the UBCD4Win bootable rescue disk, which makes it really very, very easy to make a complete working copy of the XP installation at this point.
Putting back this XP installation (in case you screw your cMP XP installation at this point) is also really easy with DI XML.
While fine tuning XP, I screwed it up so many times, that I really got fed up with installing XP again and again! With DI XML you never have to re-install XP again if you make a backup from time to time, while XP is still working and in good shape in the fine tuning process..
IMHO it’s absolutely worth the trouble at this moment and at this point, to first invest some time now in creating a UBCD4Win bootdisk and first making a backup copy of your XP (at this point) before continuing.
fully AOB optimized cMP2 PC -> Lynx AES16 -> XLR AES/EBU -> Lavry Black DA10 -> XLR Mogami Gold -> Klein & Hummel O300
Edits: 05/10/10
Thx for the tips Mark,I usally use Acronis true image but i´ll try drive xml ;)
I have no idea what could cause the higher latency, i did the recommended steps bit by bit. Could it be the hdd ? I´m using a WD blue AAKS 640gb, a 3,5" 7200 drive wich was a left over and put it into a scythe quiet drive damping box. I may buy a 2,5" one or take a 1tb WD green when i upgrade to 2 tb drives on my server.
Could the modded bios do the trick ? when i´ve upgradet from f4 to f6 there was a big latency drop, perhaps the modded f6x bios will perform even better. I just have no idea where to find it, the f6 i´m using hasn´t got the spread spectrum option anyway. Would be nice if somebody could upöoad it, maybe you Mark, it would be great.
I´m very surprised that VST works so well, im fiddling with TAL-tube, a free VST tube saturation plugin to emulate tube gear and it works really well. http://kunz.corrupt.ch/?Products:VST_TAL-Tube
I use it for some typical higher pitched digital recordings to give them some more body and musicality. Try it and let me know what you think :)
regards
Ramón
Edits: 05/10/10
Hi Ramón,
Just use Acronis if you’re comfortable with that program.
I don’t think it’s the hdd-driver that keeps XP bussy. XP uses it’s own native driver for HDD’s. No special WD driver is installed (which than could make a difference)
I reach low latency’s with whatever bios version, so I would be amazed if it where caused by the BIOS.
I fried my S2L mobo last September 2009. Since than I use the ES2L version.
But I looked on backups and I still have the F6x and the F6 bios version files. :-) They have the same file size but different md5 checksums. I could email both to you, but you don’t accept email. I do. So mail me your email.
I use only a VST plugin for equalization. Although the sound of tubes is appealing, I don’t want to develop a taste for it. Although it’s a type of distortion that’s highly ‘likeble’ and liked by many, it’s still a distortion. ‘Solid state’ should be able too produce a good ‘likeble’ sound quality too. So I try find a ‘solid state’ optimization route until I have ‘likeble’ sound from ‘solid state’.
fully AOB optimized cMP2 PC -> Lynx AES16 -> XLR AES/EBU -> Lavry Black DA10 -> XLR Mogami Gold -> Klein & Hummel O300
Very nice sound using :
Intel D510MO board (Atom 510 cpu)
corsair 1GB ddr2 dominator
Lynx L22 soundcard
WD 640GB 2,5" hard disk
Silverstone 450W fanless PSUI have tried every operating system and prefer Vista SP2. much more musical than the others.
disabled hyperthreading in bios, all devices including network card, disable high precission event timer, set PCI latency to 128.
I also use Interrupt affinity tool to route soundcard's irq to CPU1 and all others to CPU0.
power consumption is about 15-20W
I can post more details if anyone is interested.I would like to thank cics for his beautiful project.
Edits: 05/08/10 05/08/10 05/08/10
I just bought this board and I'm very impressed as well.For kicks you should try going analog out with a quality 1/8 to RCA adaptor - you might be surprised at how good that is! If you can go USB to the DAC, try that too. As good as the Lynx is, it may be this board being so low power that it will be hard to improve on the jitter with cards, which is why this generation of Atom/board is a game changer, IMO.
It could be that in the era of high power computers, you needed a card like the Lynx or RME to get low jitter, but with a super low power system, not necessarily so.
That being said, I'll try installing my Juli@ and report back. I have to assume the lower power of this Atom system can only help my stock Juli@ perform better.
I vote for the D510MO as bang-for-the-buck-of-the-year winner.
What I love about this board with an ITX case is that it's finally cheap enough and good enough to be attractive to non-audiophiles and might convert a few to the cause.
Dave
Edits: 05/08/10
I use the computer as a transport connected via S/PDIF to my Audionote Kit 2.1C DAC. it is a zero oversampling DAC so I use only 44.1K sampling rate. Oelbach's vga plug is perfect for the Lynx S/PDIF output.
Follow the discussion here . I don't think you can beat the price/performance of this mobo/cpu. Also with Juli@ you should output 192k from cPlay - Atom processor should cope with all resampler options except highest quality (SRC@145db).
@vangelis71 - is there a BIOS option to disable CPU cores? A better option is to use only 1 core with HT enabled. Both threads on a single core run concurrently. Although some bandwidth is lost at the core's Execution Units, cPlay is not affected by this. This way we further reduce power consumption. If you cannot disable a CPU core in BIOS, you can do this via XP's boot.ini file with "/NUMPROC=2" (don't know about Vista).
Hello, I would like to thank you once more for sharing your project with us. No, there is no option in intel's BIOS for disabling cpu cores.
It can be done in Vista using msconfig utility, in the advanced section.
I will give it a try and post my thoughts on it.
I only use 44.1 KHz output, because my Audionote dac does not use upsampling or oversampling.
I am not sure if I understand all of the options of cplay. I am checking different buffer sizes and try to see which is best for me, I also test different asio buffer sizes, but I am not sure about the resampler options. I use the default setting on this.
Intel's bios only let's me to disable devices and set the pci latency timer. I have set this to 128 but higher settings also give good sound.
The resampler is bypassed when you use 44.1 khz recordings and 44.1 khz output.
Edits: 05/10/10
Thanks for the information.
I did enabled hyperthreading and set the cpus to 2 in msconfig.
It sounds nice, maybe the sound is more full.
I am testing the one (1) cpu setting which also sounds very good.
I have currently set pci latency to 192, having more detailed sound.
I've installed today this new beta driver for xonar essence st-x (you'll can find here):
ftp://ftp.asus.com.tw/pub/ASUS/Audio_Card/Xonar_Essence_STX/
PCI_STX_100423_7_12_8_17731_W7ForASIO2_2
they must be bit perfect but......cplay don't work!
the problem is that this driver get a sort of priority on asus audio card so asio4all became ineffective.
Now Cplay can't work at all. I need an help from Cics, as every owner of xonar I think, because I want cplay but now it's impossible to use for me.
Obviously Foobar recognize new asio driver and works well!
My OS is window 7 ultimate
my transport was a sort of cmp2 until this evening.
Daniele
lot of music, not only sounds
@audiodan@tiscali.it
Delete all drivers except ASIO4ALL and cmasioPCIX in the registry - software/asio.Configure Xonar Essence ASIO driver to 32 bits(did mine on foobars).
cPlay should now detect the Xonar Essence ST/STX.
HTH
I think the problem is with 24bit ASUS driver.
cPlay wants a 32bit ASIO driver.
I don't agree with you bibo 01.
With Foobar's settings you'll can find the option for 32 bit, so this release must be true 32 bit.
The problem is that cplay can't see, or can't find, this asio driver.
It finds asio4all only.
I've tried new cplay release 036 and I've found a miserable trick to use it:
when you start your transport you'll launch cplay only(asio4all, obviously) and forget Foobar.
After the foobar's opening cplay don't work and windows shut down the pc!
This is a problem 'cause new asio driver seems better a lot than previous one and joined with cplay it could be the best solution for a lot of us.
lot of music, not only sounds
I rigged a test to fail on ASIO4ALL (for not supporting 32 bit output).
Screen 1: Diagnostics is shown with messages at bottom explaining reasons.
Screen 2: Notification that ASIO Control Panel will be started. This is important as you can change your ASUS driver to support 32 bit output.
Screen 3: ASIO Control Panel (in this case ASIO4ALL).
As far as I know, ASUS driver previous versions did not work with cPlay because they were 24 bit. That is the reason why - I think - cPlay cannot see it and it can only see ASIO4All.
I am not sure I understand the 2nd part of your message. Asus ASIO driver is single client and it can only work in one application at a time. Probably there is a conflict on the audio stack.
PS: are you Italian? :)
I am using the Xonar Essence ST. Previous version of the ASIO driver is version 1.0@24bits. At 24 bits cPlay can still detect the driver but would not start the DSP(errors as postd by cics).
Latest driver can be configured to 32bits but then cPLAY is not able to detect it. With foobar there is no such problem.
Question:When I am installing the new driver, should I do it out of minilogon? My new driver is now installed in minilogon.
Hi smicyta
thanks a lot for your response but I'm coming back.
I've reinstalled old asus driver and cplay sounds greater than ever,easy living with both ASIO drivers. However I think your solution couldn't work because asus driver are embedded and don't permit to open separately Asus Xonar Centre, where you manage output setting and so on, and ASIO setting, so if you don't install Asus driver audiocard don't work at all.
New 0b36 release is a true improvement, not exactly marginal.
The only thingh I've noticed is that CPU seems now working harder than previous.
Now I follow to search the right bios setting of my transport (i3-GigaByte H57-seasonic psu etc.)forasmuch as I listen always with resempling at 24/192, that's an heavier weight for CPU than 16/44 and it needs different setting.
When I finish I'll post my final setting ,waiting for definitive Asus driver......
lot of music, not only sounds
@ audiodan@tiscali.it
It works on my system but I am reverting to Juli@.
I forgot to add that in order for cPlay to detect the ST/STX ASIO driver you would have to rename the Description in the registry to Xonar Essence ST/STX ASIO. But it is a disappointment that the driver is ASIO ver 1.0. I had wrote to ASUS in March this year on the 32bits and ASIO ver 1.0 issue but it seems that they had only resolve the 32 bit issue. I don't know if the STX have ASIO ver 2.0 driver but I believe not.
Hi Smicyta
thanksa lot for your trick but I cannot see your first picture so I ask you:
- which file must be renamed and how
- where is this file
I've tried again new beta driver but it works only with foobar and no seems to me so fantastic. I prefere cplay always!
Daniele
lot of music, not only sounds
@audiodan@tiscali.it
Start Menu > run. Enter "regedit" and click OK.
Select HKEY_LOCAL_MACHINE > SOFTWARE > ASIO > cmasioPCIX.
On the right hand side of screen you would see:
DESCRIPTION REG_SZ Xonar Essence ST bit-perfect ASIO driver (beta)
Change this to :-
DESCRIPTION REG_SZ Xonar Essence ST ASIO
cPlay would be able to detect your driver now.
foobar would be happy with either description.
HTH
Best Regards
Hi Smicyta
I've tried today but cplay doesn't recognize asio drivers also renamed.
I'll wait definitive release, carry on with blessed asio4all.
Thank you for your kind interest
Daniele
lot of music, not only sounds
Hi
I apologize for my scarce knowledge of pc.
Smicyta give us right instructions for new ASIO Driver....
Today, after the assembling of Burson Audio discrete dual op-amp on my Xonar, I've retried to follow Smycita's Instructions about asio2 beta and, finally, cplay works!!
Thx a lot Smicyta
Daniele
lot of music, not only sounds
Glad it works for you. But ASIO buffer settings are not available in cPlay. So the default is 10ms when using cMP + cPlay. You can only set the buffer size in foobar.
Shure, I've seen this strange behaviour of new xonar asio.
cplay open only a blind window of asio setting . I don't know if default latency is 10 ms, I cannot see it. Using foobar's window (this one works)I've choice 1 ms latency, hoping this setting was mainteined in cplay use.
New asio are stunning but when I play music via cplay every track starts whit a prolonged (1-2 sec)disturb, like a pickup jumping and scratching the vynil record. After this music flows free, with only sporadic click.
Buffer is set Tiny and SRC 145 db at 196.
Obviously nothing of this with foobar!
Someone could tell me why?
Thx
Daniele
lot of music, not only sounds
It was mentioned in Asus download site that latency is set at 10 ms.
I, too experienced pops and crackles during start of playback with cPlay. I believe the Xonar Essence sc draw a lot more of cpu processing power than juli@.
Setting the latency in foobar and thereafter playing on cPlay does work but not with cMP. But latency in and out would not be the same i.e. in =48 and out=96.
Now I am back to ASIO4ALL. With juli@ the sound is muffled!
After I've posting my previous message, I've checked the gray windows of asio setting in cplay. If I change the latency with foobar's asio window, these value will be reteined in cplay grey window. I cannot know if is the true operative value. I'll can tray to set asio via foobar at 24 bit to watch if cplay can works or not.
Do you use juli@ like cics's istructions, only like digital output or via analog output?
I use my xonar via rca output, and I've found sq of new asio fantastic, a lot beyond previous asus driver or asio4all too, so this is annoyng myself a lot.
Otherwise foobar works well with this hell's driver.........nearby cplay sq.....
lot of music, not only sounds
Maybe setting in foobar would had locked in the latency @ 1ms.
Xonar Essence cannot do anything above 44.1kHz, that's the reason on start of playback I kept getting pops and crackles. @ 44.1kHz there is no such problem and you would consistently get latencies(in 48, out 96). At higher frequencies settings, the latencies in and out in cPlay keep changing.
The sound is also softer, but with greater transparency when playback using Xonar Essence ASIO.
With juli@ I am using only the digital part through the breakout cable via RCA to an outboard DAC.
Will try the juli@ again when the RCA jack I ordered arrived.
Refer to ReadMe.txt for important details & specifications.
Change Log (2.0b36 release):
- DSP refinements including more efficient resampler performance for both SRC and SoX
- Minor UI refinements including cosmetics. Window title now shows track number, in & out rates, e.g. "[3 44> 192] Vengelis - ...".
- Fix for cue sheet processing when content file exceeds 99 minutes
cPlay documentation can be found at the new cMP² website .
Please REMOVE previous versions before installing cPlay 2.0b36. Normal (SSE2), SSE3, SSSE3 and SSE4 versions are available. Running cPlay on a CPU not supporting the required instruction set will cause cPlay to exit immediately. For best results, uncheck "Timer" setting and test different "Buffer" sizes.
As from cPlay 2.0b33, the installation folder is NOT deleted if one or more plugins were used. All plugin settings are stored in files named with the corresponding plugin name and '.vst' as extension. Subsequent cPlay releases will attempt to use these settings. That is, cPlay releases can be upgraded without loosing plugin settings. On installation of a new release and plugins were previously used, the installer prompts to confirm reuse of the existing folder. Answer 'Yes'.
With all the issues on Hirez tracks here it's nice to know the original file sr. I had several that I thought were hirez and were in fact not. So thank you for this feature.
Ps I'm not sure that all the 'real hirez' files are actually better sonically than a plain vanilla 44 sr file. 16/44 forevever!
Thank you cics! Compare with b34, I like b36 much more. Bass is less boom, more solid and clean while retaining the legendary resolution.
Edits: 05/24/10
Thank you so much cics for this version and indeed all you have done with your tireless efforts to date. I preferred v34 to v35 somewhat -- but am so happy with v36 which is your best release so far and pretty much nails everything across the sonic spectrum as far as I'm concerned. As with v34 I'm thinking you can't possibly improve on this... but you never cease to surprise!
all the best
OE
Agree with everybody's opinion here: better base, better focus, more solid mid-range, and more consistency throughout the frequency spectrum. A major step forward. Every base note in Van Morrison's Moon Dance (Japanese remastered) is now discernible. Yes, big shrill indeed.
I mean thrill, not shrill. Sorry.
Just to chime in I want to say I am thrilled with this latest version. In addition to definition and imaging improvements, I find the bass clearer and tightened and not diminished. Further on my cMP2'd laptop, I can use the medium buffer setting at 145db/192; whereas with 35 anything bigger than tiny brought occasional requests for driver resets mid-file. The Weiss driver sounds very good but doesn't interface with cPlay, especially when in cMP mode, entirely smoothly.
I have had recent mounting stepwise improvements as I started using Weiss DAC202 (replacing DAC2), then got your version 36, and then went to single core (added /NUMPROC=2) as you suggested. It sounds so wonderfully musical and real now!
Hi Cics,
Thank you for yet another very nice cPlay version.
First things that spring to mind (compared with version 35) are:
* The highs.
36 has very b e a u t i f u l highs which suite my setup very well. I really love ehm.
I didn’t like the highs of 35 very much. Even after ‘double buffer’ was disabled, I still found the highs in 35 too much, too ‘loud’ and ‘in my face’.
* There’s better definition across the whole frequency range. More details, more atmosphere and ambiance, better human voice, ect. Subtle improvements everywhere, but especially the mids benefited from this.
* Luckily there’s a little less bass (too my perception) but also the bass has better definition. Same as with the highs. I like this bass level better. A little less, but better.
Thank you again for your hard work. Version 36 really does fit my setup very beautiful !!
This is ‘The One’.
fully AOB optimized cMP2 PC -> Lynx AES16 -> XLR AES/EBU -> Lavry Black DA10 -> XLR Mogami Gold -> Klein & Hummel O300
Yes, bass performance improves handsomely! There's also better integration and transient performance (similar to B34). Having been to another live performance of a genre I'm starting to learn and love, transient performance is vital. Something one appreciates greatly at live performances.
Some background: there were 3 pre-releases before this final release. I'm pleased with the DSP improvement given the low-level nature of changes done. I loathe doing this as full validation of all cPlay DSP components must be done. This includes validating output for both SRC and SoX.
SRC (44> 192> 44):
SoX (44> 192> 44):
First impressions are lasting. In music it is that 1st transient or ictus of the sound that (for me) forms the basis for the ultimate sound/music value judgement. 36 has better sound all around (bass to highs) but it is those transients that are greatly improved. Yes, to me, it reminds of cplay 34 but I have made a lot of changes in the isolation of my system so I can't with any accuracy say if 36 is better than 34 or vice versa but 36 is the best I have heard period.Another area of improvement (maybe another manifestation of the above) is soundstaging. I hear around instruments better in large ensembles. More air, more spacing of instruments, less running together.
Voices are also better. I have an acoustic album of blues by Buddy Guy & Junior Wells titled Alone and Acoustic on which the voices are superb but now better on 36. These 2 are not great singers (Wells is better than Guy) but wow they CAN play their instruments. Just amazing.
I have several sax pieces by Bennie Maupin and Brandford Marsalis and their instruments now sound more like like real saxes in real space. I can listen louder w/o the anoyances of those little burrs on their otherwise smooth sounding instruments.
Bass is now stronger with 36 almost as if I have done more work on improving the acoustic isolation in my hardware system.
First impressions are lasting and my first impressions of cplay36 are all positive.
PS Please keep going to those live music events.
Edits: 05/08/10
.
Yes my ongoing switching affair between 44 and 192 is currently @ 192 full 145 src.
Edits: 05/09/10
Hi @ all,
Just need some feedback if it´s worth a try to Power p4 and p24 with those cheap second hand Dell Power Bricks, for p24 i would use a Pico 150xt feeded from a Dell Da 2 ( 220w. 12V. 18A ), for the p4 another Brick plugged in directly.
http://freenet-homepage.de/palme2008/Dell_DA-2_mit_P4-Adapter.jpg
Those Bricks are used quite often from the silent and powersaving PC people. I already use a Dell brick with pico psu in my htpc and my impression is that is was a stepup in sq there compared to a seasonic s12 i used before.
Kind regards
Follow the discussion here . Instead of using Dell bricks go with a linear power supply like this which is available almost everywhere (and cheap).
Do you have a link where you can BUY it in the states?
Google certainly doesnt yeild any places to buy it that I can see.
Thanks.
@ CicsThx, is it sure that one velleman supply is enough to power both, p4 and p24 or is it better to get two of them ? or do i need a 12v 5a voltcraft to power both p4 and p24 from one psu ?
Or use the velleman only on p4 and keep the Dell for the p24 ?
Sry but i couldn´t really get from the discussion wich way would be suited best.
Btw, i can´t find the velleman psu you mention anywhere to buy here in germany, it seems that they sell it under another brand ( Peaktech 6080 ), see here :
http://www.peaktech.de/productdetail/kategorie/dc-netzgeraete/produkt/p-6080.html
@ dawnrazor
In case you mean the Dell PSU, you can get it on Ebay, look for Dell DA-2 or ADP-220AB ( Delta ). You cannot buy them new anymore.
Edits: 05/02/10 05/02/10
PeakTech 6080 is even better with ripple < 0.5V RMS! Overall it definitely looks like the same lab PS.
Only use one at 3A max load if you sure load for both P4 & P24 is less than 3A (or 36 watts). This is the case when using specified hw and BIOS settings (i.e. undervolting and underclocking) where power consumption is ~20 watts.
For best results use it only for P24. Use "dirty PS" for P4 and all else.
Wow, i´ll buy two of them if it´s useful. They are available in many E-stores over here and costing ca. 50,-€ per unit.The Big Supplies from them, the 6010 and 6020 are interesting aswell offering 2 x 0 ~ 30v 5a and 5v. 3a fixed. But 275,-€.... oO
If i get short on plugs i propably change.
I´m a bit confused now indeed, i thought that powering especially the p4 would make the biggest impact, you say i should power p24 with it instead. If it´s useful on p24 aswell i will buy two units.
So if you sign it off cics i go for Two Peaktech 6080 for p4 and p24 each. Then i need to figure out how i´d connect the cables with the female plugs on the linear psu.
Edits: 05/02/10
I´m a bit confused now indeed, i thought that powering especially the p4 would make the biggest impact, you say i should power p24 with it instead. If it´s useful on p24 aswell i will buy two units.
See discussion here . P24 is the critical supply and ideally a pure linear PS is best. To date this has been challenging to do but thanks to Gene_ and Mihaylov's efforts, we have a pure linear PS .
I would suggest getting only one for P24. All else including P4 can be powered using a standard ATX PS. Make sure its AC circuit is separate from P24.
When P4 only powers the CPU, we can switch to the dirty side. Theo has recently adopted this approach and like others have found sq improvement. I'll update the site to reflect this method of powering as standard.
Allright, you´re the boss cics *g*How can i find out if the AC circuit is seperatet from my ps ? Sry if it´s a nooby question but i´ve no idea. Already available are those two ps wich i can use as dirty supply :
Hiper psu from Hiper media center case http://www.silentpcreview.com/article756-page3.html
The Hiper doesn´t have a dedicated power on button.
And a Seasonic s12II 430w. here the specs :
Technically, the S12II is a dual 12V rails Double Forward Converter circuit design. High grade Conductive Solid Polymer Aluminum Capacitors are used to further enhance performance and reliability. These capacitors offer improved heat resistance and conductivity over the market standard electrolytic capacitors
• 430W ATX power supply
• Multiple +12-V Outputs: Enhanced +12V current capability broadens utilization possibilities
• Super high efficiency (up to 85%): Optimal solution for lower energy consumption, noise & heat
• Active power factor correction (99% PFC): Reduces line loss & power distortion
• Double forward converter circuit design: Advanced topology for highest efficiency
• Conductive polymer aluminium solid capacitors: Advanced ultra-low resistance, high temperature capacitors on the critical 12V rail
• 12cm ball bearing cooling fan: Increases airflow & lifetime and reduces rotation speed and noise
• Smart & silent fan control (S2FC): Smart thermal control to balance noise & cooling
• Soft-mounting rubber cushions: Reduces fan rotation & vibration noise
• Ultra ventilation (honeycomb structure) Minimized airflow resistance for maximum cooling
• All in one DC cabling design: Supports PC, IPC, workstation, server & dual CPU systems
• Universal video card support: Supports new PCI-E video card technologies
• Universal AC input (full range): Plug & run safely anywhere in the worldOr i buy a new one, all in one dc cabling sounds a bit like there´s no seperate AC...
Edits: 05/02/10
I'm referring to the wall AC outlet. Ideally it should be sourced from a separate breaker (or phase would be better). Another approach is a power line filter that provides isolated AC outlets or use a dedicated filter.
Your Seasonic will work as the "dirty PSU".
Don´t know if my Oehlbach Powersocket 404 is enough, it has 8 Plugs with filters but afaik no power conditioning. I think over here it´s a bit different then in the US due to other plugs, here you have two ways to plug in a device and for best sonics you need to know where the phase is on the device and on the plugs of course.
I´m living for rent and can´t mess with circuit modifications here, it has to be a dedicted device like a powerline filter or similar.
The so called Hifi power conditioners are very expensive over here, if there´s anything bargain like like the linear psu stuff buyable from a non hifi normal electronics shop and doesn´t need skills in electronics to modify then let me know please.
Btw, thx for the great support ;)
Hi,
The solution is right at your doorstep. :-) The German company Epcos produces al sorts of main filters. See there website. Th Epcos filter will cost you between 20 and 40 euro's. Depending on the needed current flow.
You can buy them at www.conrad.de. (In The Netherlands one can buy them at www.conrad.nl. So I expect you can also order them at www.conrad.de)
These filters shunt HF noise to the ground of your mains. So you need a mains with a real earth (really grounded) line. In The Netherlands it's required by law that a ground-line is really grounded to the earth. I expect German safety laws even to be tougher. So it’s likely you will have main-lines in your rental with a real ground / earthed line.
fully AOB optimized cMP2 PC -> Lynx AES16 -> XLR AES/EBU -> Lavry Black DA10 -> XLR Mogami Gold -> Klein & Hummel O300
Thx hfavandenpas,Do you mean something like this right ? http://cgi.ebay.de/Starkes-Netzfilter-10A-230V-50-60-Hz-EPCOS-HiFi-/160405925891
I think there´s nothing available already build into a case from epcos. I had a look on the conrad page but at first sight i didn´t found something similar or even the item from the ebay link wich is, to my surprise advertised as hifi filter^^ I´ll keep trying to find this at conrad or similar and a tutorial perhaps to build a unit by myself incl. case.
So if i have something like this epcos thing i can plug my 8 plug oehlbach into it or is it better to plug only the cMP² psu in ?
regards
Edit : Ifound something here http://de.farnell.com/epcos/b84112bb110/filter-10a-chassis/dp/9751939?Ntt=EPCOS+-+B84112BB110
and @ conrad http://www.conrad.de/ce/de/overview/5441641/Netzfilter-Epcos
Edits: 05/03/10 05/03/10 05/03/10
Hi Fidelity Castro,
I first would suggest: relax and read the suggested info which cics pointed you too (after you have red the cMP manual). You need some more knowledge to correctly interpret the information that is given too you by other inmates (I hope you don’t feel offended by this comment)
I want to put my Epcos filter suggestion in too perspective. (yes I mean those you found on the Conrad website. But use the wizard on the Epcos website too choose the right one. Epcos has hundreds you can choose from).
Cics recommends to use an ( as much as possible) clean 230 Volt AC power source for the PSU that will power the P24 pin.
1. Where to get clean 230 V AC power fro the PSU that will power the P24?
Therefore Cics suggests to preferably take this 230 V AC power from a separate (clean ! ) AC mains circuit.
That is: if there are any separate and/or clean AC mains circuits in your house ! But in many cases there are ! For instance: in my house there are separate mains circuits for: the laundry washing machine and laundry drying machine, the electric combi oven and the dishwasher machine. These all have there own separate AC circuits. Depending on the floor plan / lay-out of your home: it might be possible too easy tap from of those circuits.
The floor plan in my apartment, lets me easily tap into the separate AC mains circuit of the laundry washing machine to get 230 V from a clean separate AC circuit.
May be the floor plan / lay-out of your home makes it also possible to easy tap into a clean circuit. (without having an extra circuit installed, since that isn’t possible, since you rent your home)
You can easily see/check how the 230 V AC circuits in your home are arranged. Go and check how the safety circuit breakers are arranged. Every separate circuit has it’s own separate circuit breaker. Check which rooms and/or which appliances have there own circuits.
2. next best options if you can’t tap into a clean separate 230 V AC circuit for the P24 pin.
2.1. prevent that the circuit where you draw the power from, gets spoiled and polluted. (by dimmers, other ATX PSU’s, etc etc. Preventing is better than trying to clean)
2.2. try to ‘clean’ the power that you drawn from a spoiled and polluted 230 AC circuit.
In my place I do all tree.
A. drawing power from clean separate 230 V AC circuit for critical components
B. preventing that a circuit get’s spoiled
C. cleaning AC power before using it.
This is were the Epcos filters do come in. These filters are designed to prevent PSU’s spoiling the circuit with HF-frequencies, from which they draw there power from.
Ad B.
I use a Epcos filter in the power cord to my ‘dirty’ ATX PSU. Preventing the ATX PSU to spoil the circuit.
Ad C.
Since I don’t really know If the separate AC mains circuit of the laundry washing machine is really clean, I also use the epcos filters in:
- the power cord to a 4 piece power socket where 3 linear PSU’s drawn there power from.
- in the power cord that feeds the Lavry DAC
- and also in the power too the active Klein & Hummels O300 speakers.
I you google on: ‘ Jon Risch filter ‘ you will find the schematics of a popular AC mains filter used in audio circles. If you also study the schematics of the Epcos filters, you will find that the schematics are exactly the same. Applying AC Mains filters doesn’t do wonders. They just shunt HF noise to the ground via a small capacitor (see schematics). So: they only work if you have/use a TRULY AERTHED ground !
I can’t recommend too build a DIY AC mains filter, unless you are a professional electrician. The parts you need, have to be specially designed for HIGH VOLTAGE AC use and I also think it’s very risky business ( potentially lethal !!) to construct such a HIGH VOLTAGE filter if that is not your all day professional work.
And since the schematics are the same (!) is much easier and safer too just buy a few Epcos filters. It won’t break the bank.
-> Some (none scientific) testing / probs I did in my home.
I toke two 4 piece earthed power sockets and put one into the other. That way I had 7 sockets too draw power from. I connected all of my off my gear together (psu’s, dac, active speakers) to these 7 sockets. One could easily hear how the ATX PSU was spoiling the power on the 6 other sockets. Using 1 Epcos filter which was placed on the ATX PSU power line bettered things.
After that I, decided to put Epcos filters on all power lines, wether or not the power is drawn from a separate and/or clean circuit (just to be shure).
Hope this helped somewhat.
Enjoy.
Mark
fully AOB optimized cMP2 PC -> Lynx AES16 -> XLR AES/EBU -> Lavry Black DA10 -> XLR Mogami Gold -> Klein & Hummel O300
Thank you very much Mark,
I don´t feel offended by any way because to care really about the electricity feeding my hifi chain is a new approach for me. After some tests with cMP² with an old PC wich impressed mei decided to build a dedicated one as good and as quick as possible, you´re right i should have taken some more time to undestand what cics really meant. English is not my native language and new stuff is twice as hard to understand.
Now, thx to your extensive post i know what is meant, i will check the safety circuit box asap, i even forgot and can´t tell of my memory if there are any seperate breakers..lol
Logically i doubt that any of my circuits here will not be polluted somehow, industry is not far away and possibly the electricity is pulluted before it gets to my apartment. To get real unpolluted power i´would need to install a solar system, buy big akku´s or similar i think. So cleaning and perhaps reconditioning seems to be the best choice if you live in a big city and don´t have possibilities to generate your own power, logically indeed.
I will try it out to feed the p24 with a linear supply taking power from a "clean" circuit and let my ears decide. Logically there would not even be a difference between any well build hifi component. As far as i experienced that doesn´t apply when it comes down to listening, anything can matter perception wise, even the colour of the room...
Back to the Epcos stuff, how did you connect them / integrate them to your power cords ?
Hi Fidelity Castro,I also first tried a cMP build with an old PC. :-) The results where so encouraging, that I build an original cMP with the suggested hardware. But when I did build several more afterwards, I didn’t use the Zahlman again. I used much cheaper, normal PC cases, but with extra damping materials attached (bituminous felt pads ))
I installed the Epcos filter (sifi a series) into the power cord of the Antec ATX PSU in an quick, easy and (dirty) approach. See picture. The load side of the Epcos mains filter faces the ATX PSU.
Van Epcos On the net you will also find people who use 2 Epcos filters in series at various ways. Sometimes with both ‘line sides’ connected to each other. So as a whole the 2 filters have a ‘load-side’ on each outer end.
<- load–line -> - <- line–load -> .
Depending on the circuit diagram this makes sense or less sense .
See circuit diagrams Sifi A or sifi C
Van Epcos
Van Epcos
Depending on the circuit used, the HF frequencies will see a load ahead and than choose the low resistance route through the capacitor and flow too the ground.
Van Epcos I didn’t try this myself, but I can see the ration in that. Coming from the load-side the HF frequencies travelling on the mains line, ‘see’ a resistance ahead (the inductor) and thus will take the route with less resistance through the capacitor too the earthed ground.
Travelling HF frequencies arriving at the line side, will see no resistance (the inductor) ahead and thus won’t take the short-cut route through the capacitor to the ground. So that’s why some folks use 2 Epcos filters, with both line sides connected togetherFor the Lavry DAC I put the Epcos filter in the power cord. Just as with the Antec, but now with the load-side facing the mains net.
For the Klein & Hummels I didn’t cut there power cords. But instead I used the Epcos filters (load side facing mains) in the power cord to a distribution block with 2 power sockets. The two Klein & Hummels power cords I put into this distribution block
The same approach is used for the 3 linear PSU’s. An Epcos filter is put in too the power cord (load side facing mains) that feeds a 4 socket power distribution block.Too prevent from electrical shock (230 Volts !!) I used heat shrink tubing as isolation material on the 4 connection points on the Epcos filter, where the 230 Volt mains lines are soldered onto. After soldering and shrinking the heat shrink tubes in place, I just wrapped a (broad) clear tape around it (less in the middle and more onto both ends), to make the whole of filter + attached AC wires more sturdy and stiff.
The whole setup is hidden away in a sideboard, so aesthetics doesn’t have to be taken in consideration. The only thing that’s on top of the sideboard is the Lavry DAC and one tinny the combined receiver of the wireless keyboard and wireless mouse.
* Putting the whole 230 Volt AC filter thing into perspective.
-> Filters aren't needed if there is nothing to filter. I don’t think ‘all evil’ necessarily comes from the outside. I think most ‘evil’ is already in your own home. Nowadays there is so much gear (dimmers, humming chokes for halogen lightning, lots of other equipment with switching PSU’s, motors in freezers, coolers, ect etc) in our home that pump al sorts of noise into the mains, that it doesn’t hurt if one tries to filter (some of it). When running the audio system on a dedicated mains circuit, there is much less to filter out.-> There’s the same trouble on the other side of the PSU: on the low voltage DC side.
One can go through great lengths in filtering the 230 V ac mains but on the other side of the PSU the 3.3, 5 and 12 Volt DC there is the same trouble.
The MoBo and other sphericals heavily spoil and polluted the low voltage lines. Switchers, rectifiers, opamps, motors, etc ect are all over the place and spoil and pollute the low DC voltage rails from which they draw there power from.Some of these polluters can be easily removed / avoided (motors) or polluters can be put on to an other PSU. This way they dump there pollution into another, less critical PSU. That is why Cics recommends the use of a ‘dirty’ PSU and a clean PSU. By putting dirty components on an another separate ‘dirty’ PSU, one (tries !) to avoid that these dirty components spoil and pollute the low voltage DC lines of a clean PSU.
* Filtering on the low voltage DC side of the PSU.
There are PSU’s that pay much attention and put much effort into filtering pollution that comes in on either side of the PSU (Jung, Sulzer PSU regulators etc).
But there are also PSU’s that hardly pay any attention and hardly put effort into filtering pollution that comes in on the 230 V AC side and comes in on the low voltage DC load side. Anything goes.I think removing or avoiding polluters on the low voltage DC side or putting them on another PSU is a good strategy to avoid there pollution. But there are still many polluters left inside the digital audio workstation / audio playback PC, which can’t be removed.
Of course everything begins with providing good and clean low voltage DC. But I also
think filtering the pollution that is coming from the load which travels around on the low voltage DC lines, is also a good extra strategy. This can be done inside the PSU or at other places along the DC lines.Placing smoothing caps on the DC powerlines to the P4 and/or P24 pin is a nice and simple way of filtering the low voltage DC lines IMHO. So right now I’m heavily interested in finding out if there are more advanced filter methods for the low voltage DC lines. For instance shunting HF Frequencies too the ground. (instead of shunting them to the 0 volt line as with smoothing caps)
Because: what use is building a nice and very clean linear PSU that provides ultra clean dc power too the mobo if there are still many polluters left that ‘poop’and ‘fart’ into this ultra clean dc voltage line, and thus spoiling and polluting it at the same time.I think it has to be a two way approach: providing very clean low voltage DC power and at the same time preventing that this clean low voltage DC power is polluted and spoiled by the load (mobo, ect, ect). Unfortunately I’m very much hampered by my lack of knowledge of electronics.
So I think it’s worth too take a balanced approach and put effort and attention into the whole of the PSU chain. Too the PSU itself, but also too what is happening on BOTH SIDES of the PSU. Too the 230 AC input side as well as too the low voltage DC output side.
As an DIY amateur I can do the following:
* 230 Volt side:
- taking 230 V from a clean circuit.
- prevent the circuit from being spoiled and polluted
- trying to filter pollution with filters (like Epcos) on the 230 Volt AC side* PSU itself
- Using a quality PSU that delivers as clean as possible DC
(low ripple, linear PSU’s, hybrid solutions, ect, ect, )
- Using 2 or more PSU’s: one for clean loads and one or more PSU’s for dirty loads
- using a PSU that has good filtering or cleaning capabilities build inside.* low voltage DC side.
- additional filtering of the DC lines with ‘smooting caps’ or more advanced filters.Hope this long post will trigger some more response from more knowledgeable inmates on more advanced filter techniques on the low voltage DC lines.
Mark
fully AOB optimized cMP2 PC -> Lynx AES16 -> XLR AES/EBU -> Lavry Black DA10 -> XLR Mogami Gold -> Klein & Hummel O300
Edits: 05/04/10 05/04/10
Thx again Mark for your very detailed explanation :)
I found out that there are two usable circuits in my home, actually the Fridge and Washing Machine are using one of them in the kitchen. Another one is heavy current and used for the oven and one for the heater in the bath.
What i will do is connect them both, Fridge and Washing machine to the big circuit and take the current for my Hifi gear from the free one, or if it makes sense only the linear psu until i do some filtering for the other components to prevent pollution.
The Kitchen is right through the wall of my living room where my hifi system is. I propably drill a hole into the Wall and pull a extension cable with power plug through to get the Power from the dedicated Plug in the Kitchen. Should i especially look on hifi like quality cable for the extension or is it disregardable ?
Hi Fidelity Castro,
In my apartment I could do the same. I drilled a hole into the room where the laundry washing and laundry drying machine are kept. I tapped into their dedicated mains circuit. Since the laundry washing and laundry drying machine only are used a few times a week during the night (too benefit from low priced ‘night-power’ between 23:00 and 07:00 hours), it’s essentially a clean mains circuit when the washing and drying machine are not in use.
Be sure not to tap into a 380 Volts ‘power’ mains circuit !!. In Dutch homes one can also have a so called 380 volts ‘power’mains circuit to feed high power electric ovens, electric cooking plates, heaters, ect. I used to have one 380 V ‘power’ mains circuit in my last apartment.
I don’t use expensive exotic loony ‘HiFi quality’ mains cable. If quality manufactures like ATC, PMC, Klein & Hummel, Prism, Weiss, Lavry, Bryston (and many, many others not mentioned) are not using them…..why should I ? I think most of that is sales-talk to let you spend big money. Flame suite on :-).
I ‘believe’ in Ohms law. So I bought mine at the nearest DIY / Home Depot store, but did pay attention too quality used. Choose a cable with wires that are thick enough (low resistance).
It’s also worth to pay some attention to the plugs, (wall)sockets and distribution sockets that you are going to use. Check if these have proper (low resistance) connection terminals for the wires, ect, ect. Also check the already installed wall-socket in your home on this.
I know people who use cable with extra safety shielding. It prevents from electric shock if you accidentally cut into the cable. But it also shields the cable from RFI.. These type of cables are used in industry and outside in gardens ect. You could choose such a cable type if you are worried about RFI. I choose not to use this type of cable. It’s to bulky and sturdy and also the mains wires in the walls of my apartment are not shielded either. So if there’s really any RFI to worry about, than those wires in the wall, will still pick it up.
fully AOB optimized cMP2 PC -> Lynx AES16 -> XLR AES/EBU -> Lavry Black DA10 -> XLR Mogami Gold -> Klein & Hummel O300
Allright Mark, that´s basically my philosophy, only spend extra money if it gives you the extra. So i´ll take a look at homedepot sort like stuff. The Plugs in the Kitchen are lil bit crappy, propably i replace the one from wich i take the current from. A Furutech plug in the kitchen would be a bit exagerated indeed... roflSo you swapped your higher and Naim stuff and went active then, propably i´d do similar if would´t be so impressed with speaker DIY kits. Someone on a budget with hancraft skill looking for speakers i can reconmend to take a look here http://www.audiovideoforum.de/portal.php ( There´s a little english section aswell were you can get info )
The Developer worked for Focal in the 80´s and is one of germanys experienced people in this regard. His new line of speakers ( Atmossphäre ) uses the same Chassis like Magico and Raal ribbon tweeters from serbia and is sold for a ridiculous price in comparison to the established manufacturers if you sonically want to equal that. I only drive the Imagination Two wich is 500,-€ as a kit but burned 3000,-€ speakers in comparisons. The used parts ( Hivi ) maybe seem cheap but this is were the developers skill to modify and create advanced crossovers with mundorf stuff and the usage of effective speaker housing comes in.
@ Ryelands
Thx for the link :)
Edits: 05/06/10
Not a cheap publication given that it's magazine offprints but, for a deal of useful (UK/US centered) information on mains power supplies for audio that you won't readily find elsewhere, see link.
Hi Reylands,Don’t let Fidelity Castro buy that information. Advice him what to do. Advice him what’s possible within a reasonable balance of: time, materials, attention, money ect, versus results / effects.
I think also here the 20/80-rule applies. 20% of the most effectieve inputs will already yield 80% of the result.
I like to illustrated that with my own experience.
From 2000 till 2006 I lived in Amsterdam centre in a newly build apartment complex. The whole infra structure ((waste-)water, electricity, ITC, ect etc) was newly developed since the complex was erected at an old factory locatin at the Amstel river. I discovered that this company http://www.kempelektroniks.nl/General/Company.aspx was only two blocks away from my home. That time the owner had just started his company and still did home instalments of a dedicates audio mains group himself.
So since his company was real close the whole job could be done in less than our + materials.
He adviced: a dedicated mains group, a dedicated mains cable into the livingroom and a dedicated audio grade outlet socket for my Naim system (CDX + Hicap, NAC82 + 2x hicap + snaps, NAP250)Since their was a lot of drilling involved with specialzed drilling machines through thick concrete wall, I the stupid part of the job. This would take the most of time.
When finished the Kemp guy would come over and instal the extra mains group and connect everything together (fuse group, dedicated mains cable and audio grade outlet)It toke me whole afternoon drilling with a lot of dust ect etc and the Kemp guy needed 45 minutes. Labor costs 90 euro. Materials: 280 euro’s. Dedicated mainscable 180 euro’s (15 metres) audiograde outlet + mainsgroup and fuses.
So there I was: A dedicated mains group installed by a professional company. The guy graduated and did his engineering thesis on power conditioning in recording studio. So not a guy who is selling snake oil.
The results ??? Well….. I had a very very hard time to hear the difference.
There where differences but hey where so minimal….. May be the mains in the Amsterdam apartment was already okay. Enough quality for home audio use.When in 2007 I moved to my new apartment in Rotterdam I decided not to have a dedicated audiogroup installed again with a massive shielded cable running around in my living room.
Given my experience in Amsterdam, I didn’t think it is worth all that extra time, money and trouble. Especially if one easily can tap into a reasonably clean mains circuit.So that’s why I’ve gotten very keen on the cost/effectiveness ratio.
Now to the equipment and it’s vulnerability to mains noise etc. If you open up a Naim Hicap, one can easily see how it’s designed, build and constructed. It’s a minimalistic ‘brute force’ design. One hughe coil transformer, 2 rectifiers, and 2 enormous capacitors. Nothing is done to prevent that noise from mains is singing through into the 24 volts side.If you open up Quad stereo equipment one can see that the Quad design takes into account that there might be noise on the mains.
So the design of the equipment is very important if one will hear benefits from exotic mains cable, mains filtering, ect ect.
On this my view on things is: if your equipment is very sensitive (or vulnerable) for your mains or mains cable than put your money into a better design or better brand. But that’s just my approach of things.
Since some friends of mine are musicians and an other one is a graduated audio engineer (SAE) and one is an video editor, a few times a year I’m in a recording studio. To my surprise almost all of the gear used, sounded amazingly good and at a price level of 50%.
So these two things (design issue’s and price/performance ratio) made that I had a good look (and listen) at pro audio gear. But there a also manufacturers who produce for both markets (pro and home) like: ATC, PMC, etc. You can guess: the days of my Naim gear where numbered.
Now back to Fidelity Castro: what reasonable advice can I give him?
1. Try to get clean mains power. Preferably at a good cost/effectiveness ratio (see our posts)
2. Don’t spend (much) money on crazy (over-)priced exotic cables. Instead: spend that amount of money on gear that is better designed and constructed.Now onto your advice for Fidelity Castro. What would you advice him?
Please backup your advice with clear arguments. Not because mine or your arguments are better or worse, or that one of us has more knowledge.
Just for that Fidelity Castro knows on what the advice is based. And also because that’s why there are forums like this to discuss things with one other.
fully AOB optimized cMP2 PC -> Lynx AES16 -> XLR AES/EBU -> Lavry Black DA10 -> XLR Mogami Gold -> Klein & Hummel O300
Edits: 05/06/10 05/06/10
Advice him what to do.
I am not presuming to advise anyone on anything here not least because I'm unfamiliar with local conditions and regulations. I was merely pointing anyone who might be interested to some good information on mains power for audio in a book I bought some years ago, found useful and recently re-read.
Apart from agreeing that it's unnecessary to spend silly money on mains cabling, I have nothing useful to add to what you say except to suggest to that, given a choice between my advice for free and some genuine expertise for £22, I know what I'd do.
Hi Reylands,
I see your point. But reading your posts, you are clearly belonging too the more knowledgeable inmates. All this even backuped by thorough reading ect.
Why don’t you tell others what you did and what your practical experience are? What exactly did you? If you did. Or why did you not. What where the effects in your setup, ect. Others can benefit from such information too.
I have some suggested reading too: http://www.aboutlearning.com/what-is-4mat.html
Mark
fully AOB optimized cMP2 PC -> Lynx AES16 -> XLR AES/EBU -> Lavry Black DA10 -> XLR Mogami Gold -> Klein & Hummel O300
Ideally it should be sourced from a separate breaker (or phase would be better).
At least in the UK, power sockets connected to different phases of the mains supply must be sited (from memory) at least two metres apart, i.e. well outside arms' reach. This is because, while the voltage between any phase and ground is 240 volts, the voltage between two phases is an interesting 440 volts.
In the highly unlikely event of a domestic premise even having a three-phase supply, it would be a brave man or woman who used different phases to power an audio setup even in these days of ELCBs and the like. No electrician would countenance it.
If anyone on a 240-volt supply does decide to go down that route, good luck - you'll need some (and your children rather more).
Dave
This might be of interest to those wanting to try the mythical I2S.
I am hoping someone with familiarity might be able to explain to me how to make this work with JULI@. Greg and sonics, you come to mind first, obviously!
From looking at the attached STUFF can you recommend what goes to what?
http://www.raleighaudio.com/I2S%20Interface%20v2.0.pdf
I am not sure what to think of using HDMI cable. Whether this is a good way to go or not, I have no idea. I do like the idea of not having to be concerned about distance between JULI@ and my DAC card, which I prefer having close to the line stage.
Any help and advice would be greatly appreciated.
Thanks,
Rick McInnis
What I cqan tell you is that like the link I use only 3 leads from juli@ to my Buf32s. I do not understand the link you provided for your dac in that it shows more leads. Also what was most important for me was getting damping resistors (soldered right to juli@ pins) of 22 ohms into the 3 leads. Also I used 75 ohm teflon dielectric wire for my 3 leads (about $50 for 5 feet) and the most important item: ground all the shields from (in my case the outer shields of the teflon wire) at both ends. I use 3 feet of wire.
If you can get bye the what wires do you connect question its the best performance you can hope for IMO.
with it, I figured it out.
Went to TWISTED PEAR's site and the BUFFALO User Manual and it seem this does the trick. Below are the equivalents for those who might want to try the K&K/RAKK implementation
BUFFALO RAKK i2s
DCK BCK
D2 SD
D1 LRCK
I am going to try this out for my return to digital audio and will let you know how well it works.
Hope all is well for you and thanks again for the help.
Bye,
Rick McInnis
glad I could help. with I2S and damping resistors I have not had a 'metallic issue' for 6 months.
though your continuing problem with that is almost supernatural in its inexplicability!
I wish you continued success!
Thanks, again,
Rick McInnis
PS I guess I am in a particularly exclamatory mood!!!
Hi: I have search many articles about soundcard output via coxial digital out.
Which soundcard recommend?Juli@ or Xonar?
About Juli@,ASIO driver is the best quality.
About Xonar,the db value is low very much.
About these two,which one you recommand?Or another?WHY?
Thanks!
A modified Juli@ works just fine with only the digital portion installed. BNC to SPDIF out.
RayBan
YES,I have remove the analog section.
My DAC(Apogee mini) only have spdif,not BNC.
Hi.
Cics, please, are there some kind of limits in the lenght of the files?. I have problems with long files (more than 90 min): the time of the tracks is not show correctly and cPlay jump back to the first track.
It's happens whit files with cue sheet. If I play directly the flac or wav file, it plays fine -but, obviusly, I have no control over the tracks-.
Thank you.
Regards.
Ignacio.
.
Thanks!.
Please, could be possible also an horizontal scroll bar for long track titles?.
Regards.
Ignacio.
Please, could be possible also an horizontal scroll bar for long track titles?.
Why are clever ideas simple ones! And why did no-one else think of it? So, seconded - in the hope it is as easy to implement as it would be to use.
I second too.
Furthermore, I have another little suggestion.
If I open cPlay in Windows mode, its window size is fixed and always behind the Taskbar. Is it possible to max its window, and seeing all buttons at the bottom, without having to auto-hide the Taskbar.
Thanks
I need more details like the actual cue sheet and diagnostics output from cPlay. Hopefully I can recreate on my side.
See this post .
dear cics,
is there any reason for that the sse4 b9 instruction set not been released on the .035 version ?
just curious.....
Hi all,
I have finally set up my dedicated HTPC for CMP. Did everything according to the Cics guide, installed CMP, killed all services etc., except that I use Foobar2000 instead of cPlay. I am using RME Digi 96/8 PST analog out (AD1852) @ HTPC with picoPSU+Linear PSU -> Shanling PH100 amp -> AKG k701.
I was a bit skeptical about all that - and now all I can say is WOW. I Still can't believe that the improvement is due to software alone, after moving from standard winXP+foobar to tweaked XP+CMP+foobar. Everything is so alive now, it seems that I hear real sax/cymbals, lots of 'air' around the instruments, pleasant and not annoying highs.
Ok, back to listening :)
Hello Dexter3D.Can you write step by step settings CMP+Foobar2000.Im trying use this duet,but... rankena@ml.lt Ps.I like cymbals sound too ;)
Are you implying cymbals do not sound right via cplay? If you do there might be something else wrong in your system. Cymbals sound absolutely right as far as I'm concerned in my system.
Awesome DEX that you have joined the club. That was the experience I had too. Amazing that the same hardware can sound so much better when you get the computer optimized correctly.
SOunds like a great system.
THough I dont understand the need for foobar if you are using cmp.
What is up with THAT?
Well, my system is based on low power Via EPIA (VIA C3 cpu), which AFAIK has only SSE support (no SSE2/3). I like epia because my pc draws no more than 15w, and I can use 3A@12V linear PSU at less than half of its power to feed the system via picoPSU. Out of all versions that I've tried (including 2.0b7, b9 which are supposed to work), only cPlay v1.0 works for me... I am yet to compare v1.0 vs foobar.
-
Edits: 03/19/10 03/19/10 03/19/10
AH, that explains it.
I have a similar computer that I used to use and was trying to set it up to be a headphone setup and ran in to that issue myself.
IIRC there is a version like #15 or 16 that is the last that supported sse.
True I never had a linear psu on my via c3, but I wouldn't be at all surprised if the recommended hardware and the latest version of cplay would sound better. It certainly is better than what I could do with the via.
Given your situation, I would try multistream asio player if you can.
dear forum &
dear cics
upgraded the driver for my lynx TWO-B card recently and was quite surprised by the increased definition and transparency this yielded.
most driver upgrades seems to be related to compatibility issues with win7 or the new 32nM technology, so my feeling is that this may also be the case with steinbergs third version of VST.
so my curiosity is therefore to find out if cPlay is using the newest VST protocol ?
there is a surprising "silence" on the new processor generation in our forum..... has anyone tried these out yet or have any first impressions ?
-or are there any technical bottlenecks causing so little activity ?
kind regards
so my curiosity is therefore to find out if cPlay is using the newest VST protocol ?
No. Currently very few plugins support version 3. Allocator for example does version 2. The issue here is not many plugin developers are eager to upgrade to 3...
There's very limited SQ benefit here.
there is a surprising "silence" on the new processor generation in our forum..... has anyone tried these out yet or have any first impressions ?
Whilst I do plan to test the new hw, problem is my current setup sounds very good with B35 (organic, natural and involving). So it may take more time...
dear cics,
thanks for your reply.
I´m not eager to swap for any newer tech HW yet, as also I enjoy the current state of the cMP2.
-but there was just that little "curiosity"
especially have I found new exitement with SOX running on the .35 version.... the previous less musical character of SOX (versus SRC) has definitely to be re-evalued....
kind regards
Refer to ReadMe.txt for important details & specifications.Change Log (2.0b35 release):
- Minor VST refinement
- VST parameters maintained on ASIO startup error. Previous approach reset parameters to plugin defaults whenever an ASIO error occurred at startup.
- Remote control via PostMessage function. External programs can start player and perform control actions like Next, Previous or Stop directly. Useful for headless setups requiring programmed interaction with player. UI frontends can also exploit this new feature. See supplied sample program in release notes.
- Keyboard processing only done when focus is on main window
- DSP, SRC & ASIO refinements
cPlay documentation can be found at the new cMP² website .
Please REMOVE previous versions before installing cPlay 2.0b35. Normal (SSE2), SSE3, SSSE3 and SSE4 versions are available. Running cPlay on a CPU not supporting the required instruction set will cause cPlay to exit immediately. For best results, uncheck "Timer" setting and test different "Buffer" sizes.
As from cPlay 2.0b33, the installation folder is NOT deleted if one or more plugins were used. All plugin settings are stored in files named with the corresponding plugin name and '.vst' as extension. Subsequent cPlay releases will attempt to use these settings. That is, cPlay releases can be upgraded without loosing plugin settings. On installation of a new release and plugins were previously used, the installer prompts to confirm reuse of the existing folder. Answer 'Yes'.
Sample ProgramCopy Remote Control application (cicsRemoteApp.cpp) below and compile.
// cicsRemoteApp.cpp - trivial console app that launches cPlay in realtime and sends commands.
#include < stdio.h >
#include < windows.h >
#define REMOTE WM_APP+2
#define RM_PREVIOUS 219
#define RM_NEXT 221
#define RM_STOP 190
#define RM_PHASE 186
#define RM_SKIPBACK 189
#define RM_SKIPFOR 187
#define RM_PLAYPAUSE 80
#define RM_VOLUP 10
#define RM_VOLDOWN 11
#define RM_TRACK 12 //Set "lParam" to track # to be played (1..n)
#define RM_VSTEDITOR 13
#define RM_EXIT 14
#define CPLAY ( "c:\\program files\\cics Play\\cicsPlay.exe " )
#define CONSOLE ( "cicsRemoteApp" )HWND cPlayMainWnd = 0;
BOOL CALLBACK SetMainWnd( HWND hwnd, LPARAM lParam ) { cPlayMainWnd = hwnd; return TRUE; }DWORD playSomething( char *mediaFile )
{ STARTUPINFO start;
PROCESS_INFORMATION process;
memset( &start, 0, sizeof( start ) ); memset( &process, 0, sizeof( process ) ); cPlayMainWnd = 0;
start.cb = sizeof( start );
CreateProcess( 0, strcat( CPLAY, mediaFile ),
0, 0, 0, REALTIME_PRIORITY_CLASS | CREATE_NEW_CONSOLE | CREATE_NEW_PROCESS_GROUP,
0, 0, &start, &process );
Sleep( 1000 );
CloseHandle( process.hProcess ); CloseHandle( process.hThread );
return process.dwThreadId;
}int main(int argc, char* argv[])
{ if ( !argv[1] )
{ printf( "\nUsage: cicsRemoteApp.exe \"\"\n" );
return -1;
}
DWORD cPlayThreadId = playSomething( argv[1] );
if ( !cPlayThreadId )
{ printf( "\ncPlay failed to start.\n" );
return -2;
}
EnumThreadWindows( cPlayThreadId, SetMainWnd, 0 );
( SetConsoleTitle( CONSOLE ), Sleep( 40 ), SetForegroundWindow( FindWindow( 0, CONSOLE ) ) );
printf( "\nCommands:\t1..9=Track, p=Previous, n=Next, \n\t\tf=Forward, b=Backward, s=Stop, \n\t\tu=Vol Up, d=Vol Down, i=Phase, v=VST, x=Exit.\n" );
for( int cmd = 0; cmd != 'X'; cmd != '\n' ? getchar() : 0 )
{ ( printf( "\n==> " ), cmd = toupper( getchar() ) );
switch ( cmd )
{ case 'P' : PostMessage( cPlayMainWnd, REMOTE, RM_PREVIOUS, 0 );
break;
case 'N' : PostMessage( cPlayMainWnd, REMOTE, RM_NEXT, 0 );
break;
case 'F' : PostMessage( cPlayMainWnd, REMOTE, RM_SKIPFOR, 0 );
break;
case 'B' : PostMessage( cPlayMainWnd, REMOTE, RM_SKIPBACK, 0 );
break;
case 'U' : PostMessage( cPlayMainWnd, REMOTE, RM_VOLUP, 0 );
break;
case 'D' : PostMessage( cPlayMainWnd, REMOTE, RM_VOLDOWN, 0 );
break;
case 'S' : PostMessage( cPlayMainWnd, REMOTE, RM_STOP, 0 );
break;
case 'I' : PostMessage( cPlayMainWnd, REMOTE, RM_PHASE, 0 );
break;
case 'V' : PostMessage( cPlayMainWnd, REMOTE, RM_VSTEDITOR, 0 );
break;
case 'X' : PostMessage( cPlayMainWnd, REMOTE, RM_EXIT, 0 );
Sleep( 300 );
break;
default : if ( cmd > '0' && cmd <= '9' )
PostMessage( cPlayMainWnd, REMOTE, RM_TRACK, cmd - '0' );
else
printf( "\nDon\'t understand %i.\n", cmd );
break;
}
}
return 0;
}
Edits: 03/20/10
Hi,
Which compiler are you using to compile CicsRemoteApp? It doesn't compile under Visual C++ 2008 Express edition.
Thanks,
Alan
Compiler in Intel's cpp 10.1. I suspect the problem is the first two lines:
#include < stdio.h >
#include < windows.h >
where the space between braces and letters should be removed - I added spaces to prevent AA post editor from interpreting to something else.
Any compiler should do - what error are you getting?
Hi all,
Coming home from my holiday in New Zealand and Australia I found 2 audiophile surprises waiting for me. A driver upgrade for my Lynx AES16 (version 18 from 16 march 2010) and a new a Cplay release (version 35 from 3 march 2010).
* Lynx V2 Driver Build 18
In my setup the biggest improvement comes from the new Lynx driver version 18. Whow ! Some inmates call it: minor changes…. But the stereo-imaging is much (!) better (broader and more 3D) and also the sound is less digital and much more ‘analog’. A major refinement in my setup and to my ears.
So everybody with Lynx gear: go and get it !!
* cPlay release: 2.0B35
I have mixed feelings about B35. Treble and upper-mid’s are the same as in build B34 but there is more bass. For my taste there is too much bass. And also the lower mids are more ‘muddy’ and less focussed. Imaging is less focussed.
With Lynx driver version 18 the bass in cPlay version B35 is somewhat better (not so much). And also the lower-mids are less muddy and a bit more focussed. But still much less as in version B34.
* Lynx V2 Driver Build 18 + cPlay 2.0B34 that’s my favourite (for now)
May be I have to get used to the new sound of B35. But right now, for me, and in my setup, the combination of the Lynx V2 Driver Build 18 driver + cPlay 2.0B34 I like best.
Despite I like version B34 better, again thank you Cics for all the hard work. May be version B36 will fit better to my system and my taste.
Side note:
I read posts from inmates who are very sceptical on: if software can make a difference in sound quality. Well, they should hear the differences in my setup after installing the Lynx V2 Driver Build 18 for Windows and a new cPLay release. One must be almost deaf not hearing these ( big ! ) changes in sound quality.
fully AOB optimized cMP2 PC -> Lynx AES16 -> XLR AES/EBU -> Lavry Black DA10 -> XLR Mogami Gold -> Klein & Hummel O300
Looks as if Lynx has a Xmass present. New driver 18j, still evaluating 4 hours burn in. At this point the music has been extraordinary but then it was before. T.
No worries!
hav,
I agree having updated my lynx driver. Do not that it installs differently than the older drivers in that one just has to run the file. The extra right click and install step is not needed. Took me a while to figure that out.
Though I am using 35. I really like it and given the other recent changes and the overall improvement I am afraid to touch a thing at this point!
Cics thanks again.
Try this: cPlay Settings > ASIO Settings > In Lynx ASIO Control panel, uncheck "Double Buffer Output".
Above with B35 is breathtakingly beautiful.
Listening through my K701's incredible sound. I finally have a grasp of DSP buffer settings, sometimes one cannot see the forest for the trees. I was mistaken in my understanding that a smaller dsp buffer was necessarily better and could not reconcile between what I heard and what I perceived. Thank you for your patience with my questions. Listening to Diana Krall large buffer setting, E7500, 3meg L2 cache. The bass has settled right down, the detail is really incredible. Foot pedals on the piano etc., the background is so clear. I have updated the Lynx driver and disabled double buffer. Thank you again, T.
No worries!
Edits: 03/24/10
Eratum 20 march 2010:
* problems with cPlay version B2.0B35 in my setup solved.
Problems:
After the Lynx driver update to V2 18, I had several non-optimal settings in the Lynx mixer:
Synchrolock was ‘on’, up-sampling from 44.1. to 48 was ‘on’, double buffer output was ‘on’, all buffers where still at there large default settings, etc.
After correcting these settings for the Lynx driver to:
Synchrolock ‘of, no up-sampling, buffers to 32, double buffer out ‘off’ cPlay version 2.0B35 now sounds indeed utterly beautiful.
The fat bass is gone and the lower mids are no longer muddy and blurred. Compared to B34 version B2.0B35 has brought (again) a new level of micro details to my setup. Resulting in MUCH (!) more ambiance and 3D and…… SWING !! This cPlay version (B35) really swings ! It’s by far the most ‘musical’ version. It has laser precision and superb timing but without the slightest hint of being sharp, grain, ‘digi-ditis’ etc.
After having all the settings the right way, I like this version (B35) very much. I especially like that little extra swing that my NAIM setup (CDX, NAC82, NAP250) also had. But which wasn’t present in earlier Cplay releases until this version B35.
Thank you Cics for B35 !
fully AOB optimized cMP2 PC -> Lynx AES16 -> XLR AES/EBU -> Lavry Black DA10 -> XLR Mogami Gold -> Klein & Hummel O300
Finally updated my lynx driver. That took me forever to figure out as the old instructions didnt apply.
Then I installed 35.
This reminds me of 31 I think or whichever version it was that was laser precise and no one liked it. Though this manages to retain the warmth and emotion with the focus.
Though it seems like the center image is laser precise but the sides not so much.
Overall it is a step forward I think, and I wont be going back to 34.
Thanks for all the hard work!!!!
Honestly I didnt hear much difference with src and sox, though I think the bass was better with src, though I certainly didnt spend alot of effort in the comparison.
Still running at 1.6ghz to burn in a powercord, so things could change.
hey dawnrazor,
I´m a little surprised that you don´t discern the difference to sox, since I feel it quite different.
sox is still the more analytical choice of the too, but within the ver. 35 it´s actually very good indeed. there is also bit more "kick" and ...eh, "dynamics" so the less present elements of playback and therefore a slight pronounciation of details.
-this I think actually suits our magnepans quite well !
do give it another try.....
Play mate,
I will be giving this an extended audition. Though I have some major things to do that I am focusing on for the next few weeks.
Do tell, how you did your measurements.
'Still running at 1.6ghz to burn in a powercord, so things could change.'
Could you please explain this?
'Still running at 1.6ghz to burn in a powercord, so things could change.'
Could you please explain this?
Se the post below but I changed out a power cord and the system that would routinely boot at 840hz just wouldnt with the new cord. I dont understand it.
So I thought that if I upped the voltage and the speed and ran things that way for a while it would be able to boot at lower speeds. Right now it is at 900 and seems to be booting ok. But I think the voltage is higher than it was. This is really weird to me.
Anyhow since I didnt think I was getting the best at 1.6, I threw that disclaimer in.
ok thanks I now remember the discussion. Just had a senior moment there for a moment.
Unfortunately I think I know what you mean.
The sad thing is that I have led a clean life. If I know 20 years ago that my short term memory would leave anyhow, I would have partied ALOT more....
Up to b34, SRC was my choice no matter how many times I tried SOX.
With b35 things are all the opposite. I am still trying setting combinations but I am running out of them. SOX now is cleaner/clearer than SRC on this PC.
To be sure, I am now using Win7, where b34 lived a short life (I ran it for much longer in XP). I am planning to put it back on and see if this is a Win7 thing after all.
While I get to do this, I figured I'd ask if anyone found similar behaviors.
hey JBen,
I have always thought of SOX being more analytical and maybe also picking up more details, but prefered SRC for it more musical liquid presentation.
-but switching back upon your post here, I DO agree that the somewhat harder edge of SOX is diminished.
there are definitely more details in SOX !
I´ll let SOX play for a while and revaluate.
good observation !
I had a comparison session this evening just to norrow things down a little more on cPlay b35. Play-mate, I'd say your are right on the money. While SRC did not gain or lose, SOX manages to deliver more musical warmth than before and still sound sharper than SRC.
The evening's music mix included classical ensembles, jazz, latin and vocals. SRC does have its more alluring moments but now I have to stay alert for them. I think I stll liked Diana Krall's voice with SRC just a hair more. But with the rest of her ensemble going for SOX, so did I. In the past, bright violins liked SRC; SOX always seemd better at metal percussion. Yet, a Prokofiev violin sonata sounded so sweet and harmonic this evening with SOX that I had to switch on the panel to make sure I had not left the settings on SRC. Then I went on to violin/cello/piano trios, SOX was the better all-around charmer.
Bebo Valdes' "Kabul y Lorenzo" is a recording that pans around instrumental solos. While in cPlay b34, it had been solidly in SRC's camp despite SOX's prowess at its metal percussion. In b35 it became fully SOX-friendly. Almost the full range of its instruments from piano to brass, was better under SOX. The sax was slightly smoother under SRC but saxophones can irritate me easily. Therefore...I jumped to David Sanford's "Man from Mars and then "Tequila". SOX took the honors.
I left large orchestras and bands for another day (I tend to run these from SACDs).
This was with buffers at "small" (perhaps "medium" is good also). In both SOX and SRC, with large buffers I get an enhanced perception of sharpness. It "ain't so musical"...although, at times, it is impressive.
I had cPlay with AWE on, at 176.4 upsampling of 44.1 Flacs and Wavs, SRC=145db SNR. SOX=VHQ, bandwidth=97; Phase=Intermediate 35 with my modded MMGs (bi-amped) and a Denon AVR-3805 driving. 'Puter is AMD, overclocked!!! cpu, NON-optimized for cPlay, under Windows 7. Fanless, low power PS. USB DAC is Musiland 02 or E-mu 0404 USB. This sounds so good already that there will have to be a dedicated cMP/cPlay outfit later this year.
J,
This sounds so good already that there will have to be a dedicated cMP/cPlay outfit later this year.
Good for you Jben!
There is something about using his recommended hardware...at least vs my first attempt using existing Asus/AMd hardware.
I'll try your sox settings and see what I get.
Hey DR!
Last evening, for the first time -- other than brief previous tests -- I tried 192 upsampling of 44 material. I always use 176 for being 4x 44. At 192, I only play the few "native" 192 pieces that I have in the computer.
One thing I found out is that SRC 145 @192 does not work with this computer (or I need to tweak for it, for it did before with b34/WinXP). So, I compared SOX and SRC 121. As far as I remember, it is the first time that cPlay (here, with this poor-man's outfit) captured me at 192 upsampling of 44.1 material.
Relative to my previous set of comments, SRC (now at 121) regained ground vs SOX. I liked both. Head-scratching weird but there was no beer involved. More hours this evening...
BTW, my friend got the Halo A21 for his B&W 802Ds this week. I'll be spending a few hours listening with him this afternoon. Check email late tonight.
I listened all afternoon but to my ear it was not quite right. I switched back to 34 definitely a difference 35 is too sharp in my system. Was wondering what others are experiencing. T.
No worries!
Did you adjust buffer size? When I 1st loaded 35 I forgot (usually @ medium for me, default is auto) to reset it and everything sounded too sharp. When I reset I thought the sound was not sharp enough.
Ater one day playing with buffer sizes I stand by my initial comments. While I think 35 is still more mid rangy to me vs 34 I can find a happy listening window with 35 by either setting buffer size to small or medium. With 34 it was absolutely spot on with medium buffer. With 35 & buffer @ medium there are more mids but definitely not harsh or overly bright. With 35 @ small buffer I get all the extension, dynamics I would want with the system only bordering on being too thin or sharp.
Who knows maybe I'll go back to 34.
After a day playing around I find that my favorite listening programs are in this order:
1) cPlay 2.0b35 @ 44khz/small buffer
2) cPlay 2.0b34 @ 44khz/medium buffer
3) cplay 2.0b35 @ 44khz/medium buffer
I abandoned 192 once again as I find better dynamics/sound stage and bass and highs @ 44.
I was much too harsh in my initial comments after trying all of the above. After listening to some really good source material I had close to what I call a 'religious experience' with cPlay 2.0b35 @ 44khz/small buffer this morning (no disrepect to the religious). I felt transported to Avery Fisher Hall in 1963 listening to Leonard Bernstein conducting the NY orchestra in Copland's Danzan Cubano.
What cPlay 2.0b35 @ 44khz/small buffer also gives me is an almost grain free playback of even some of my border line hard material. I like it.
Once again thank you cics...you have done it yet again. And to all those who find that they might not like 35 try varying the program parameters. You may find new treasure.
Prompted by theob's suggestion, I had tried out the various DSP buffer size settings. My preference is 'medium' buffer. I find 'medium' has more space among the sonic elements (on the other hand, 'small' has more continuity in the sound stage), sparkling high (not sharp), and a more controlled bass. Going from b34 to b35 (both 'medium' buffer) is like having all contacts in the system cleaned by a very good contact cleaner. May be my preference is due to the fact that I listen mostly to contemporary music.
I'm still learning B35. To date my preference is 192k SRC@145db and medium buffer. There were some 9 pre-releases before I released this version. However, B34 is no slouch...
This is different for VST users. Good improvement over B34 should be experienced as more benefits that non-VST users enjoy have been implemented.
Strange day, music was driving me crazy the mellowness was gone a hardness throughout. Re-booted to find my FSB frequency was reset to 1.6. Went in to the BIOs to find several MIT settings were changed. Weird never power down! Reset everything and back to 840 shall listen again starting back at tiny.
No worries!
Normally I had used tiny, 96 SRC! Today, I have some time so I plan to listen through each buffer setting. Question! What specifically does changing buffer achieve and how specifically. T.
No worries!
With both 34 and 35 set at 192k SRC@145db using my Weiss DAC2 via firewire from my as much as possible cMP2'd laptop I get a request for a driver reset every few minutes when set at medium or automatic. There is no problem at tiny or small. I generally have used tiny. Now that you and theob are concurring about preferring medium it is mattering to me, so I am wondering why this is and if there is a way to be able to use medium. Ideas?
I can think of 2 reasons for this:
1. Laptop CPU doesn't have 2 cores or 2 threads (hyper-threading of a single core)
2. Optimise setting in cMP is not "Critical"
Weiss' ASIO driver could also be an issue but I would first check above.
For whatever reason when I use 192/145/medium buffer I get a lazy mouse cursor (which to me indicates some instability) but when I use a smaller buffer or go back to 44 the mouse cursor shows no lag time. Don't have an aswer just a similar observation. But please notice I have gravitated to 44 again ilo 192. Maybe it sounds better on my system because it is more stable.
T,
Are you using flac? I think what you are describing is a flac issue, well it doesnt occur with .wav. I noticed it purely because I forgot I ripped one album to flac and I played it right after one of those albums that was recorded at a lower level...the system was super loud for the new flac album and the mouse froze as you describe. It was rather comical. It only happens with that album.
I use wav exclusively and when I get the lazy mouse it happens for all music files...for the conditions I referenced before.
I use medium buffer and it works fine. That is strange you have that problem. Everything about your saga though Theob has been strange. Though I am impressed with your ability to persevere. Keep plugging away!
"DSP Buffer" is an important setting that affects resource consumption of CPU and RAM. This in turn affects power supply and the ground plane.
Another key factor is how often the DSP thread is used to fill buffers. It's a periodic transaction and hence causes periodic jitter. Every player has this. Larger buffers require longer CPU bursts (especially so with SRC@145db) and therefore creates slower UI responsiveness. Theo's experience is expected - I too get a "slow" or "lazy" mouse with larger buffers. It's entirely by design in cMP² with Optimise set to "Critical". Try setting Optimise to "Player" or "Realtime" and you'll experience different results.
Explanation: when you move the mouse, it requires CPU intervention but with a large DSP load there's a delay as the DSP thread enjoys RealTime priority. When Optimise is not set to "Critical", the mouse action is serviced immediately by the other (most likely) free CPU core (when using a dual core CPU). This unfortunately interferes with our most critical ASIO thread: an optimal design has a CPU Core dedicated to it as this not only circumvents but is superior to having a RealTime OS kernel.
Therefore, different buffer sizes induce (by software) different jitter distortion (both in magnitude and frequency) causing audible differences.
Thank you for your reply on buffer settings. Cics, I have as a practice always used small and then tiny. With 35 after correcting some system errors I have been using auto which I have found to be very detailed and relaxed. With respect to jitter and buffer size in theory would auto buffer offer an advantage over manual buffer settings with respect to periodic jitter. I try to see it in terms of reducing VRMS to the CPU, perhaps I have over complicated my understanding. This is the reason I asked you the specific effect of buffer size. Tiny for me is very accurate and detailed but does not have the same relaxed flow as auto, small and medium seem to have some base emphasis but auto is the most fluid, for my system. T.
No worries!
Auto never uses Tiny.
Either Small, Medium or Large is used based on selected output rate. For example, for output rates at or below 48k, Auto selects Large.
Would auto buffer reduce CPU request as default for monitor refresh rate.
No worries!
MIT reset, new nos Sylvania 6sn7gt/vt231 cracked, (*&^%^^&**^&%). Auto cPLAY 2.ob35/minlogin/l22/24/96/Dac1/300b/PP/SHL5. Diana Washington is in the room!!!!! No worrries,
No worries!
Thanks for the explanation. I guess I was way off on how buffers should affect things.
WIth other players it seemed that with larger buffers the more stable things were.
thanks for the explanation.
dear cMP community,
brief first impression of ver. 035 :
more articulation and more "air" !
-it also seem more details present, as instruments separate better and have more "sonoric" colour.
I might even think there is a bit more "musicality" or "coherent flow"
I have experienced more obvious dramatic upgrades of cPlay, but this is real refinement. probably only a "significant" upgrade on already very high resolution systems.
the devil lays in the detail.
kudos again, cics !
BTW : there is building up significant (controversial) interest upon cMP2 in germany. (bloody germans)
I agree with what play-mate was saying about b35. I however, think the upgrade is significant. b35 brings more focus on the whole sound scape. The high, mid, and bass are all more lucid and the boundaries of the various sonic elements better define. The end result is that more emotion and passion from the recordings.
I don't know how cics does his magic. It seems to me that every upgrade raises the hurdle that would make improvement for future upgrades even harder. It was pretty hard for me to imagine b35 given how good I thought b34 was already. Thanks again for the great work cics.
Initial comments also: while I agree with playmates comments in general I would say 35 sounds like there is more midrange at the expense of extension at both ends...a little brighter/airier in upper mids but not as much depth and transient attack and as much bass as 34.
More to follow.
How can I get the source code for cplay? Have checked sourceforge but I can only find the binaries.
cPlay is the perfect player for me. High sound quality (and ASIO, oversampling and VST-plugin) is precisely what I need. But I would like to use a more "family friendly" frontend than cMP (sorry ;) )
Soren
Source code is available on email (via SourceForge.net) request. See here .
Calling All [cPlay] Headphone Users...
If you are less than comfortable with the way traditional stereo recordings sit "in your head" when listening over headphones, I have a little VST plugin to share with you.
From the brilliant minds at Refined Audiometrics Laboratory comes the free VST plugin HDPHX. Using the Haas Effect, it creates a delayed and attenuated cross-channel mix for each ear, simulating the experience of listening over loudspeakers. Quite effective, and "realistic", if that term can be used to describe domestic audio playback.
No user controls or interface - it's either IN or OUT of the circuit (use the VST bypass button to compare). I'm not sure I'd use it all the time, but it certainly does reduce that odd congested feeling of some music mixes. Obscures a tiny amount of detail too... just like speakers will do.
Read more and download from the link below.
Hope this helps (and doesn't agitate the Audiophilia Nervosa...)
Cheers,
Grant
ps: Thanks again to cics for creating a VST path in cPlay !
That's not a Toy... IT'S A TOOL !!
Another excellent but pricey one I use is Redline monitor that simulates listening with near-field monitor speakers. http://112db.com/redline/monitor/
I have not experience with VST plugins. I tried installing this plugin. The Bypass button is inactivated when the hdphx is listed in the vst window. So how are you and Dawnrazor going back and forth to compare in to out. I am not sure I hear any difference checking and unchecking the vst box but then I'm also unsure I've gotten it installed and functioning.
How do I do it correctly?
Hey Riboge,
That is a challenging plug in for your first try.
It has zero settings so it is tough to hear what changes it makes on the fly. But one can bypass it.
Cplay is a bit counter intuitive in its setting. Instinct would make one check the VST box, but that is actually a VST SAFETY setting...it was added so one couldnt accidentally disable a plugin like crossovers that would be disastrous. VST is just on if you add a plugin location.
So uncheck the VST box, and you will have bypass capability. And like asio settings it is good to set cplay up and then exit the player and restart.
Thanks, I get it now. The checkbox caption should be changed to "VST lockon" or "VST failsafe" to make it clearer.
I'm not one who is bothered by normal stereo thru headphones as some others are, but some correction could be better. HDPHX makes some improvement but not that much that I can be sure or have a clear preference. I'll try the one you suggested next.
I am using pc> Weiss DAC2> Singlepower ES-1> Senn HE90.
I was able to get more of wanted with the following plugin. There is a TON of things you can do and I am just getting the handle on it but it seems to allow for much more of a difference than the one linked before.
Here are the preliminary settings I have stumbled on
as it has been pulled from the official site (although a new version is promised for 2012 - if that is not free, that would explain why the free one has been pulled :) ):http://www.vst4free.com/free_vst.php?plugin=Omnisone&id=519
Edits: 01/09/12
Thanks for suggesting this plugin. It allows one to play with the spatial aspects of playback, useful with speakers as well as headphones. I wish I understood better what the "side signal" is. That term makes it seem like there is a separate signal from left and right channels, but of course there isn't. I suppose it is derived from them somehow. There does not seem to be significant degradation of the signal using it, which is really nice.
Hey Riboge,
Glad you like it. It does seem to offer alot of control and manages to keep the quality.
I'll be the first to say that I dont know what this thing does and how it does it outside of what my ears tell me. But come on Riboge here is what the manual says:
Side gain : it is rather explicit : the volume of the side signal ;
So I cant say what your problem is :)
The nice thing about the vst capabilities of cplay is that there is a whole big world out there of interesting stuff. Most of it is for the pro world, but some are useful, like eq and others.
Here is a one of many many places that have free vst plugins that do all kind of things. When you get something like console you can link several together and cics has shown some examples.
A-Ha, yes - Sorry I forgot to qualify my observations with a signal chain...
Juli@ (BNC-modded) > Benchmark DAC1 Classic (BNC-coaxial IN) > Sennheiser HD 580 phones. Someday the Audio Faerie will leave HD 650s under my pillow...
Cheers,
Grant
That's not a Toy... IT'S A TOOL !!
Hey G,
I have had that plug in in for months. It doesnt really do much in my system. Though I did a blind test where I clicked the bypass button with my eyes closed enough times and quick enough where I didnt know if it was bypassed or on. Then I listened and switched many times. I became clear which was bypass and which wasnt. I stopped after the first time getting it right so my "test" wasnt really conclusive.
My original impression was that it didnt really do much so I just left it in. I had tried other headphone vst plugins and the HDPHX is the best of the ones I tried.
Though my headphone rig is a $50 sound card and $100 headphones so not exactly the most revealing thing around....
I think we all know when starving buffer size is causing "crackles" in the audio stream, but I get a break-up in audio in the first second of a tune when upsampling to 192kHz instead of 96kHz.
I presume it may be the VST driven allocator that causes this, but there is no doubt that the 192kHz resolution is sounding far better.
my lynx asio settings allow to reduce the buffer size down to 64 samples with even tighter quality as result, but the break-up of the first few tones is pretty annoying.
can someone explain this phenomena ?
kind regards
dear inmates,
the cMP2 project shall not become the subject to any political debate, but I believe it´s rather in "cics" spirit, to just spread the word of his exceptional work for high-end audio.
in a specific way there is, however political dynamite in the fact that a do-it-yourself software can challenge the very best in the costly market of digital high-fidelity. not anyone in the industry like to acknowledge that !
but in the same way I would like to announce that cMP2 has got a "reference"-write-up from a major audio magazine in germany : "klang & ton"
-one that is giving "cics" and the project justice, too !
applause !
It's always a pleasant surprise when editorial staff take the courage to break out of their business model. Previously we had Enjoy The Music and now, Klang & Ton . Bravo!
I find through greater participation, we all benefit more. Gene's pure linear PSU implementation is a great example of this.
hey cics,
AA-inmate soundchekk says he has been in contact with the editors of klang+ton sometime in december but dropped out of further involvement due to certain disagreements.
I suppose that there is another chain of enthusiasts than through me. I have no affiliation to the mag.
I have, however tried to stir some promotion of the cMP2 project in a few audio forums here in germany and there must be something moving.......
the article is very objective and will be continuing with a german DIY guide in the next issues.
kind regards
PS - BTW what is the "Gene's pure linear PSU implementation...."
-have I missed something ?
Is there an english translation?
.
thanks for responding
See link below.
Thanks to Soundchekk and of course yourself.
Cics.
No idea if you read my comment in the other thread about this subject.
I still think your approach is the best on a "XP system" regarding
software. Your PC HW tweaks are also pretty much leading edge
if we look at a "pretty standard PC HW solution".
That's why I referred to your solution as one of the possibilities for a wider DIY community (The magazine K&T addresses pretty much 100% the DIY community btw.)
Windows XP has still a huge market share - your solution is still more then applicable to quite some people I'd guess.
As I told you in the past I consider XP based audio a dead end and even not at all the preferred choice for best audio.
As you know I am running a Linux system. We ( me + some AA inmates/friends) compared my setup to CPLAY/CMP2 on the same PC.
My Linux solution just delivered a "lot" more precision.
Shame on me - I just made my solution available to a small group of people by now - my solution does not exist for wider community of people. I just bring this up to tell you that there is more to gain.
However - even my setup is not able to get around all PC obstacles.
My position:
If you're able to slightly tweak a streaming device such as a Squeezebox (100$) and you're able to beat every PC based solution soundwise with that. Because you knock out 85% of the problems associated to a PC with that, I am wondering if we bet on the right horse if a PC gets tweaked to death.
What you're doing with CMP2 is nothing else then trying to make an FPGA like device out of the PC. Does this make sense, if these devices already exist?
Your PC can not be used for anything else then music playback anymore anyhow.
Why don't you just program an FPGA and hook it up to a 3,3V battery. Processing is done offline on the server. The poor OS sound layers
are not even required anymore. Life could be much easier. ;)
Cheers
P.S: Of course I don't know if anything "innovative" is cooking in your kitchen. I'd guess so.
OK. Read the material.
I think what you suggesting will yield excellent results. It's a 3 box approach (Server, Router & FPGA DAC) - you may add a 4th for remote control. With Ethernet, one could have music streamed to multiple rooms from a single music server. Overall this design will grow in popularity where homes are wired for Ethernet (alternatively, one could use a wireless LAN). Streaming 24/192 may be a challenge with a data rate of 1.1MB ps on a LAN (with Gb LANs this may be less of an issue).
Does this design yield best performance? No. The way I see it, this design substitutes SPDIF for Ethernet. We already have a similar concept in USB (including Async USB). SPDIF offers least complexity (as its unintelligent) but requires the clock to be embedded. USB and Ethernet are heavier protocols and allows for local clocking. Put another way, instead of encoding / decoding SPDIF, we do the same with Ethernet (i.e. encode / decode data packets).
The challenge still remains the same in that the overall chain needs to be streamlined and efficient. That is, the FPGA DAC benefits from cleaner upstream components. Issues like having a dedicated or minimalist LAN becomes relevant (less CSCD overheads). Server side changes will eventually affect SQ (although Ethernet does a good job of isolating the streamer).
From a cMP² viewpoint, the soundcard is replaced by Ethernet (we just need an ASIO driver which can be developed). That is, we've done a full circle which gets us back to managing the full value chain. My take:
Highest performance is achieved when classical notions of DAC as “Master” or “Slave” are superseded with the design principle of creating a partnership wherein both DAC chip and Computer Transport are slaved to the XO. Jitter is tamed when our “string” is shortest and audio is streamed under minimal stress.
Imagine a direct I2S feed from Southbridge or Hub chipset...
No it is not the same.
The FPGA is a standalone device. As I said it is not driven by Windows Audio /Coreaudio /Alsa. It buffers 30s of audio-data (as you would send normal PC data) , which will decouple the stream very well.
It can't be compared to your cmp2 approach on a traditional PC. It won't require ASIO/WASAPI and an audio driver, which you know is not very seldom another problem. Doesn't matter what you do on the PC, it won't change the sound situation on the FPGA. In that application cmp2 HW and SW tweaks are absolutely useless ( you need to get your network properly set though). We both know that you can't say this about USB/PCI/FW devices.
Again - I consider this dedicated FPGA nothing else than a PC - stripped down to its absolute minimum at lowest cost and best performance by just doing one job more or less.
Don't forget it has been proven on several systems, against your and my setup, that this is not just a theory.
Just face it. ;)
Cheers
P.S: I won't enter a 24/192 discussion at this place and point. The quantity of 24/192 material is neglectable and its quality benefit more then questionable (especially on resampled material).
How will decoupling and recoupling be done per 30s interval?
Firmware will have Ethernet stack which would establish and maintain network connection (using most likely sockets).
I'm interrested in your alternative solution. Have you posted about them either here or elsewhere? If so please let me know where so I can read.
I used to maintain a Linux Wiki over at DIY-Audio. Since they reshuffled
the forum layout - I never managed ( time and motivation) to build it up again.
I installed the currently running beta systems via remote access by myself. (I couldn't afford endless conversations/support via mail/phone - within less than an hour they
were set) Prerequisite was a pre-installed Ubuntu.
My final goal was to supply a bootable pendrive system. You just plug it into any PC and your PC becomes a Linux Audio machine sourcing the data from UPNP network storages. I got quite far with it.
What scares me is to support such a project. If you make it available to a wider community it will take 90% of your time to handle all the requests. No idea how Cics manages this.
Honestly - since I have my Squeezebox with iPeng up'n running my motivation to finish this project went down to zero.
Let see - if I can find a motivation to reanimate the project.
Cheers
Please don't start something just for me. If its Linux I would need to study that for 6 months before I would feel comfortable enough to try someting like that which you have done.
I know that Bertel and others have been having extremely good luck with squeeze box setups. So there are alternatives to cmp^2. Not that I am chomping at the bit to get out of cmp^2 (in fact I just love where I am now with it). But I am always looking for sq opportunities/different approaches.
Thank you for your response.
> > But I am always looking for sq opportunities/different approaches.
You're not alone. I havn't met very many audiophiles who are just happy with what they have. Doesn't matter what they paid for it - they always want more - and the funny thing is - they always manage to get more.
P.S: I know Bertel.
Hi Play-mate, could you point out in which number/ edition of the magazine it can be found? December 2009/ Jan 2010 or the very latest feb & march?
Thanks
bernie
the latest feb/march issue is the first one.
their intension is to do a series of articles of how to DIY
I don,t know if I am allowed? Listening to my 300b new Russian K40Y's, I have been tinkering. Cmp2 lynx Dac1 SHL5 (25th). Ray Charles, Genius Loves Company, Sound stage/huge, layers of musicians and singers, PRAT. Heaven help us all, thank you Mr Cics. T.
No worries!
A lot of people seem to like Amarra. I searched but it seems no one here has asked the question of whether anyone has compared Amarra to Cplay/cmp? Cplay/cmp is easily the best replay I've achieved on my Sony Vaio/W7/usb/dCS Paganini set up, beating the Paganini transport, and all hte common players. The reason i'd look further is for a good remote interface. In that respect I love JRiver's theater mode. BTW I ran JRiver from within CMP - that seemed to get it a step better. I may have to settle there, but I wonder where Amarra sits. Anyone?
tks
murray
I thought Amarra was Mac-only and I know that cPlay is PC-only.
It might be an interesting comparison from a system-level, but in no way can you compare the players head-to-head.
Greg in Mississippi
Everything matters!
Hey murray,
The reason i'd look further is for a good remote interface.
Can you explain that? What is missing with cmp/cplay?? Personally I find that interface better than any player I have messed with including Jriver.
And I do know of an Amarra cmp2 shootout, though IIRC neither setup was truly indicative of the the best expression of either.
Hi Dawnrazor,
I tend to be someone who works better in pictures rather than text (I'm an engineer - possibly says it all ;-) I like browsing by album covers and will often find my choice of music by spotting one that I haven't heard in a while. I've got around 500 cds ripped to flac, so there isn't much choice to browse by reading all the titles. Also with JR in theater mode you can remotely control the screen from around 10' and view by genre, recent, artist, album etc. All pretty convenient from the seat. Originally I liked JR's sound but now find it a bit thin, and prefer Media Monkey for listening - even though it seems to have less detail. Nevertheless cplay/cmp is way out in front sound wise. I just tried HQ Player too and initial impressions are a very nice clear sound but so far has exhibited bad glitching on my set up. Just my impressions, no doubt others have other views.
best
M
Hey Murray,
That makes some sense. Cmp2 will not give you cover art. I find this to be a thing I was reluctant to give up but once I did it wasnt a big deal especially with the increase in sq I get from cmp2.
While I use the ALL titles mode, there are Artist and Genre views with cmp one could use.
I suppose the thing to do is switch to cmp2 when you do critical listening and just use another player when not listening so critically.
maybe I am off, but I think there is an improvement still to be had. Have you done the video detuning that cics recommends?? This would seem to make viewing coverart from 10 ft very difficult. At least it was when I was using coverart in winamp and began with the cmp2 optimizations.
No, I'm a bit new to this and have just placed cmp on my laptop without any other attention. Results are remarkable and depressing - I spent $$ on the dCS async USB interface on the expectation that given bit accurate sources, it would eliminate the computer from the sq equation. How can a computer or program affect the sound if the data is accurate and just sent across on demand? Beats me but there are worlds of difference (audiophile hyperbole understood of course) between players on this set up. Rats!
M
Hey Murray, I'm curious, does the dCS Async USB interface come with an ASIO driver or are you using ASIO4ALL?
Question, how does Cplay compare to Foobar2000?
And is there Cplay for Windows 7?
Keith
TheAVLinkAdmin
http://www.TheAVLink.com
Hey Keith,
I have been through nearly every player out there including things like wavelab, multistream asio player, sampletude, xmplayer, media monkey, acxo player, foobar, winamp, etc.
Cplay especially when used with cmp and the optimizations cics recommends is my favorite player and to me does the best job...said another way, I am not messing around with any other players now. cMP2 addresses all my needs including remote control, and most of all sound quality.
And another thing is that because cmp and cplay are created to work together, I have noticed that system stability is much better than when I was just using windows and winamp or foobar...
cPlay runs on XP, Vista or 7. Only requirement is instruction set supported by CPU - most modern CPUs support SSE2 at least.
I have not made any direct comparisons with foobar (for flac or wav).
dear AVlink,
I can testify we´re a few of us who been through a few comparisons to cPlay/cMP versus a number of players (foobar included) and there is absolutely no doubt (also amongst strict analog devotees) that the cMP2 system is far superior and "natural" sounding than the rest of the offerings on the market/net !!
set up proper, the breathtaking flow and ease that cMP2 delivers musical beauty is simply not matched anywhere !!!
-if anyone claims anything else, they´re sitting on their ears.......
period !
First off, thank you for cMP and cPlay. Using a modest Thinkpad T40-> USB-> Havana DAC with your suggested XP optimizations is quite a revelation. The sound is pushing beyond my regular transport via SPDIF into the same DAC.
I also really like the ability to change the phase from 0 to 180 depending on the source material. (It is something my ears are very attuned to.) I am wondering how this is achieved within the program. Is this a directive sent to the DAC or a DSP thing? Does the wave form have to be inverted by you in the program before it is sent to the DAC or do you just have to flip one bit on the fly?
Does the wave form have to be inverted by you in the program before it is sent to the DAC or do you just have to flip one bit on the fly?
It's the former where each source sample is inverted in the DSP when populating the DSP buffer. Actual computational cost is very insignificant. As a comparison, showing elapsed time display uses more CPU!
Most interesting. So, for the terminally insane tweaker, there might be a slight performance gain by doing an Invert/Save process in EAC for those files you have found to sound best in Phase 180 and then just playing them back in Phase 0.
Actually I tried this and can't hear much difference. As you say, the processing overhead must be very low.
Thanks again for the great software.
Hi,only for testing some things I wanted to activate AWE in Win7 Home. Unfortunately I have only the home version here, which is coming without the policy-GUIs.
I've installed the "MS 2003 Server Ressource Kit" and could successful activate "lock pages in memory" with ntrights:
ntrights -u [username] +r SeLockMemoryPrivilege
But in cplay-Diagnostics AWE failed. Any idea?
Thanks a lot!
Regards
Thomas
Edits: 01/31/10
Hi guys, first post here.
In the last few days i tried out cMP and cPlay on my Notebook with SPDIF Toslink output and i'm impressed by the quality of sound reproduction compared to other software.
I use Windows Vista Business SP2 and set LOCK privileges (Diagnostics shows that AWE is successful). However the diagnostics say, that for each track there are only between 25 and 50 MB of RAM allocated (287MB in total for 12 tracks).
My question is: If AWE is successful and enough RAM is available (available 1423MB, system 1978MB), shouldn't each track be loaded into RAM completely?
I also tried Vista Ultimate and 7 Ultimate with the same issue. I need to use Vista because i have some HD Tracks with 88.2kHz/96kHz and the Windows XP driver for the Realtek ALC269 Soundchip doesn't support all output frequencies.
Thank you very much for any help.
cPlay allocates between 160MB and 1GB RAM (based on available physical RAM) for either AWE or standard. RAM playback is done using buffers which doesn't conform to tracks - a track could be split into 1 or more buffer loads.
cPlay will never allocate all available RAM - this makes Windows unstable. Also, with very large buffers load time becomes noticeable (esp flac).
cics,
In your cMP2 guide at paragraph "Periodic Jitter - sensitivity analysis" you claim that measured jitter performance is 51ps Jpp RSS using foobar, by following the procedure explained at http://cplay.sourceforge.net/pmwiki.php?n=CMP.03Jitter#PeriodicJitter.
Have you recently measured the cPlay's periodic jitter performance?
I really would like to know how cPlay compares, not Foobar.
Thanks for all your help.
Have you recently measured the cPlay's periodic jitter performance?
I really would like to know how cPlay compares, not Foobar.
No, only with foobar.
Yes I do think measuring cPlay would be useful especially with the new 32nm Core i3-530 and GB mobo hw. All in good time...
Thanks.
I'll be waiting for it.
I have enjoyed my music with cPlay + cMP2 for months after following the instructions of The Art Of Building Computer Transports strictly. But recently I need to re-open the port of IEEE1394, but Windows crashed. Is there any step(s) I do need to 'revive' this firewire function? Thank you for any help.
[How} to 'revive' this firewire function?
We've probably all had similar problems some time or other. Typically, you need to restart one or more services and sometimes, though IME rarely, to reverse the minlogon change. It's down to trial and error. Keep a note of what works because you'll forget it by next time . . .
HTH
Dave
Dave:
Thank you for your advice. It seems lots of work ahead in this coming weekend. :-)
Tak
Hi
Are there recommended settings for ASIO4ALL, with cmp+cPlaY?
I have been using the following but do not know what they are;
ASIO Buffer Size = 128 samples (cplay reports my output is 128, so set set this so that "prefered" matches)
Latency Compensation = 0 samples (both In & Out)
Hardware Buffer = Off
Kernel Buffers = 2
Always resample = Off
Force WDM = Off
I am grateful for any advice!
Thanks
Oli
Your ASIO4ALL setup is not ideal. Try following:
Notes:
- Keep ASIO latency to lowest offered (64 samples)
- Never use any buffers, i.e. set ASIO4ALL to use hardware buffers and no latency compensation or buffer offset. As soon as any latency or buffer is added you get more overheads (added buffer/latency processing).
Thank you cics, those settings are working. cPlay is now reporting 64 for both preferred buffer & ouput latency.
Thanks
Hi,
I'm learning ASIO programming and would like to have cPlay source code as a reference. Thanks.
Please send it to
simmconn AT yahoo DOT com
Regards,
simmconn
cPlay 1.0 Final has been packaged (includes instructions, precompiled libraries and DLL) and is easier for learning ASIO.
cPlay 2.0 will be packaged later.
Dear all
I run a full cMP2 system & have been absolutely delighted with the results. I still sometimes listen to my vinyl on EMT930 turntable but haven't played a CD for months. I only listen to 16/44.1 files going to homebrew TDA1541a crown dac with i/v transformers & tube output stage (lampizator.eu)
Presently, I have an old fashioned Wireless USB keyboard/mouse combo attached to the PC. This goes on the accessory, separately powered USB port which also has the touchscreen attached to it. I am thinking of getting a Logitech DiNovo Edge mini bluetooth 2.4Ghz keyboard for it. Is there any reason that bluetooth will cause any ticks/pops, interference or any more latency delay than present? Presently my latency is 13-14 & occasionally goes to mid twenties if & when I enable wireless card to access lyrics, EAC dBase info etc.
Many thanks for advice & thank you again cics for making listening to music in this way an absolute pleasure. :-)
Regards
Fib
Since you are already using wireless mouse & KB, switching to bluetooth should not be a problem. Ideally, try reducing peripherals - if you can do without the KB that would be beneficial. I no longer use a mouse & KB - only touch screen (idea is to have it close to you). Using alternate USB ports helps a lot but still, we get the added overheads in Windows.
What mobo do you have? Latency (as in DPC Latency) is high at ~14µs.
Thanks for reply, Cics. Presently, my KB & mouse are attched thru a single usb port & I intend to keep it that way with the newer bluetooth combo as well.
I am using Biostar P4M900 Micro775 motherboard & a Celeron Dual-Core E1200/1.60 processor. Obviously, this is not the present recommendation but I made the system right near the very begining of this forum when you recommended these components. :-) I have been so happy that I didnt want to change as it wasnt broken. All the cMP mods are done to the last detail & this thing really sings. Maybe the latency issue is due to processor/MB combo.
Regards
Fib
Please see the attached link.
Alan
Hi,
I can see that volume processing seems to take place before data gets sent to the VST plugin, and I imagine it is the same for upsampling, but I just wanted to check.
Is the chain Volume -> Upsampling -> VST processing?
Thanks,
Alan
cPlay processing sequence:
Source > Resampler > Volume > Clipping > VST > Clipping
"Clipping" rarely takes place - with volume at set to -0.5db or lower, there's almost no chance of clipping.
Hi,
Thus far I have preferred CPlay without upsampling, however I have not played around with the different algorithms available within CPlay very much. Is there a consensus on the best sounding settings for upsampling?
Thanks,
Alan
I still prefer no up sampling. I still find substantial difference between SRC and SOX. Both are excellent and IMHO have come closer together with each recent release. I find SRC slightly more musical but with less air and definition compared to SOX. While SRC has gained definition, and SOX musicality, I still prefer SOX. Could all be psychosomatic but that's how I see it.
I used to be an upsampling fan (192) but as of late I prefer 44. But tiny buffer sounds way better to me than other settings at either extreme of sample rates.
Edits: 01/02/10
What is the best frame buffer size settings in Asus M2NPV-VM bios for cPLAY?Possible options for changes:Disable-> 16-> 32-> 64-> 128-> 256?
Typically smaller buffers sound better. But try the extremes and see what you think.
I also use cMP therefore i want maximize system for both players.
If you're interested I've just posted some observations on the Naim site:
http://forums.naim-audio.com/eve/forums/a/tpc/f/8772903417/m/1642928727
M
I've just posted some observations on the Naim site.Many thanks for the link. Just for the record:
1. The paragraph starting "To assist, cics has written two Java-based programs for generating and managing cue files . . ." and the following one are not quite accurate. The program (Recursive Cue Creator - 2009's Utility of the Year as far as I'm concerned) was written by AA list member Alan Jordan.
2. A tiny quibble. The replacement for the Explorer shell is called cMP (cics Music Player) and the player, as you note, cPlay. It's the two together that were named cMP2, i.e. cMP squared.
Not that it greatly matters but that's surrogates for you. (Give 'em 'alf a bleedin' chance and they'll even go saarf of the river after midnight . . . ) *
Dave
* Don't ask - it's a London thing. See OP's profile.
Edits: 01/01/10 01/01/10
Thx Ryelands,
I would go and correct the post - but for some reason when I tried to update some typos it wasn't updating but REPOSTING!
Happy New Year!
M
Hi,
Can anyone confirm:
If I select output of 44.1 and play a 9624 file it will be output at 9624?
I understand that if I select an output higher than the playing file it will be upsampled.
Thx,
M
If rate is set at 44.1 and you play 96, cPlay will downsample to 44.1.
Thx cics,
I turned on the diagnosis in cPlay and read the output - the equivalent to RTFM!!
M
Hello, everybody! I got a WV album image from a friend for evaluation, but when converted to wav or flac it will play only in foobar. cPlay does not recognize it or the cue (yes, I edited it correspondingly). Even dobleclicked directly, wav won't open, cplay saying it CAN'T FIND IT.
What may be wrong, please tell me.ASIO started successfully.
AUDIO source [ C:\RIP\HORSLIPS\1973 The Tain\Horslips.WAV ]
Genre = Unknown
Artist = Horslips
Album = wavFILE: C:\RIP\HORSLIPS\1973 The Tain\Horslips.WAV
Wav file NOT found!
WHEN I CLICK ON A CUE, cPlay says:
AUDIO source [ C:\RIP\HORSLIPS\1973 The Tain\Horslips - The Tain.cue ]
Cue file 'C:\RIP\HORSLIPS\1973 The Tain\Horslips - The Tain.CUE' NOT found.
Serge.
Edits: 12/30/09 12/30/09 12/30/09
Hey Serge,
can you list your cue file?
It might be a typo, but in your post you write:
AUDIO source [ C:\RIP\HORSLIPS\1973 The Tain\Horslips - The Tain.cue ]
Cue file 'C:\RIP\HORSLIPS\1973 The Tain\Horslips - The Tain.CUE' NOT found.
shouldnt the audio source be .wav and not .cue??
Here is the cue file. The error message I get is written by cPlay when I DIRECTLY CLICK THE cue file, which is really puzzling...
REM GENRE Rock
REM DATE 1973
REM DISCID B509E90E
REM COMMENT "ExactAudioCopy v0.99pb3"
PERFORMER "Horslips"
TITLE "The Tain"
FILE "Horslips - The Tain.wav" WAVE
TRACK 01 AUDIO
TITLE "Setanta"
PERFORMER "Horslips"
INDEX 01 00:00:00
TRACK 02 AUDIO
TITLE "Maeve's Court"
PERFORMER "Horslips"
INDEX 01 01:52:65
TRACK 03 AUDIO......
Also, when I directly click THE WAV file, I get the following:
FILE: C:\RIP\HORSLIPS\1973 The Tain\Horslips - The Tain.WAVWav file NOT found!
Isn't it a mystery? Serge.
Edits: 01/01/10 01/01/10
Serge did you check the music file extensions? Also what is a WV file? Did you mean WAV?
It is an idiotic packing, that some folks use like ape or flack. I have never come across a wv torrent that was not "trouble-packed". Always something is wrong, though after some work with foobar decompressing, cue-splitting and so on, You will end with a normal album with tracks and a cue.
Serge.
Thanks for the feedback...sorry I could not help.
Could be the conversion went bad - cplay gives the wav not found error for faulty files as well as for bad cuesheets. You could try the WavPack frontend for decompressing .wv or use Winamp with plugin, but use the most recent downloads: http://wavpack.com/downloads.html. I've had duds (usually when doing batch conversions), which you can't see from Windows file properties, so I have a quick look with Tag&Rename just in case, and doing the conversion over should fix it.
Hello, happy New Year and thanks for help, but I am such a chainick, I don't know how to use command line utilities. I have downloaded the newest zip, unrarred it but don't know how to use it. Chainick in Russian means a tea-pot - is a rookie, newcomer to some technology or habit like driving a car...
Serge.
Hi Serge
I have found "xrecode II" to be very useful for transcoding between audio file formats. Nice, elegant GUI... handles "any" format you can throw at it... manages metadata... will rip from CD... is continually maintained and updated... and the demo is fully functional (with 10 second nag screen upon opening).
Download latest version at the link below.
Cheers,
Grant
That's not a Toy... IT'S A TOOL !!
No need for command line for the things I mentioned. Download the wavpack-4.60.1.zip file and extract to a wavpack-4.60.1 folder in Program Files directory. Then download WavPack_frontend.zip ( http://members.home.nl/w.speek/wavpack.htm ) and extract to a WavPack_frontend folder in Program Files.
Inside the WavPack_frontend folder is WavPack frontend.exe. Run this. You may get an error telling you to download some additional material (the links for it is at the bottom of the WavPack frontend web page above.) For instance, I had to download the tabctl32 file - click on the link and save it anywhere you want. Extract it to a tabctl32 folder and inside the folder click on the install.bat file. After that, WavPack frontend worked fine.
On the WavPack frontend UI, select the Unpack tab. Drag and drop the files you want to decompress. You can leave all the options in the "General" area unticked. Browse to the correct Output Directory (at the bottom of the UI) and when you are ready, hit the Go button.
*
...to me it is because I have made a typo in the creation of the cue file, name not right, or something silly like that. As a 1st step just double check your cue and wav files.
Edits: 12/30/09
and, yes, everything is right. What really puzzles me, is that cPlay says it doesn't find wav, when I double click The Wav itself...?
Anyway, hello, Theo, and thanks for Your suggestion.
Hi,.
The latest (two versions) of Cplay will not find any ASIO except for my Lynx card. All my other media players see all ASIO enabled hardware. I uninstalled the Lynx driver and now it won't find anything. I am running the latest SSE2 version on XP SP3.Any advice on what to try?
Update: The SSSE3 version works fine on my Windows 7 laptop going after the same ASIO hardware. Still no luck with the SSE2 version though. I uninstalled and reinstalled the ASIO driver but it didn't help.
Thank you,
Alan
Edits: 12/28/09
Make sure all your drivers are in the registry - software/asio.
Also test with asio4all which should list all the soundcards available.
I fixed the problem. It wasn't that Prism Orpheus ASIO wasn't in the registry, it was that old ASIO devices were still in the registry even though I had removed the drivers and software. Once I deleted those entries from the registry, the Prism Orpheus was enumerated by CPlay.
I'm not sure why CPlay was the only media player getting stuck on this.
Alan
ASIO 4 All will see all of my sound cards when run through another media player, but for some reason CPlay isn't even seeing ASIO 4 All.
Wouldn't the ASIO driver have to be in the registry for other media players to see it?
Thanks,
Alan
Hi
I am looking for a way of properly using cplay at 24/192 with a nettop, e.g. Acer Revo. Nettop means no PCI slot, which means no ESI Juli@.
Has anyone tried the m2tech hiface adapter with cplay? It apparently provides real 24 bit / 192kHz by using of proprietry KS optimised drivers (so maybe no good for cplay which I understand needs ASIO).
PS Is there anything equal/better and cheaper than Juli@ for PCI solution?
Thanks
Hey folks,
I happen to have a couple of 1703v3 PH shunt regs available. Is it safe to replace the stock SMD 3.3v reg on Juli@ with a PH shunt reg rated at 90ma? How much current does Juli@ draw from the 3.3v reg?
Thank you in advance!
I use a modified Juli@ where I have cut the power lines on the PCI connector and feed it from entirely separate power supplies. Due to my configuration, I could easily measure everything EXCEPT the 3.3v directly.
But the 5v Digital feeds 3.3v, so the following measurements will be useful. My impression is that the largest part of the current draw on the 5v is the 3.3v regulator and what it feeds. So in short, I think you'll need a higher-current regulator to power the 3.3v. I had corresponded with PH on this and he supplied a regulator with a much higher current capacity. I am very apologetic to say that with work and the challenges of getting the separate DAC cards to work, I haven't had a chance to test his regulator yet. But he can configure them to work at a much higher current level.
I took the measurements with a good milliamp meter placed directly in the feed from the raw DC of the separate power supplies to the regulators, in three different states:
Juli@ powered, computer off:
-9v analog in/out 55.8 mA
+9v analog in/out 89.7 mA
+5v DAC/ADC analog 11.0 mA
+5v Digital 73.7 mA
Juli@ powered, computer booted to cMP^2 mode, no music playing:
-9v analog in/out 69.8 mA
+9v analog in/out 51.6 mA
+5v DAC/ADC analog 84.6 mA
+5v Digital 147.8 mA
As above, music playing (note that the -9v, +9v & +5v didn't change with music on):
+5v Digital 189.2 mA
I doubt this is what you wanted to hear, but I hope this helps.
Greg in Mississippi
Everything matters!
I read that the VST host on cPlay accepts 32 bit plug-ins.
Is this a limitation? I am thinking, for exampe, to Izotope SRC 64 bit...
Does it have anything to do with 64bit floating point plug-ins?
Thanks
Hey B,
I'll take a stab.
SInce I have been about to use the allocator with cplay, and allocator says this in is specs:
#12. 64bit processing throughout. SSE2 instructions utilized.
I think you will be ok.
You are right.
However, in cPlay Guide under VST it says:
"cPlay supports use of a single VST plugin (up to version 2.4.2, 2-in/max 8-out) in 32 bit float precision."
What about then?
Is "Izotope SRC 64 bit" 64 bit float point precision?
Can anybody reply to the above question?!
What is the best VST equalizer?
I am playing around with VoxengoOvertoneGEQ. First impressions on b34 are quite good. I am interested in other opinions.
Refer to ReadMe.txt for important details & specifications.Change Log (2.0b34 release):
- Major VST update:
- Fix minor UI error on 64 bit indicator in VST settings
- Add silence when switching tracks to clear plugin buffers
- Enable speaker setup with plugin versions beyond 2300. Such plugins are automatically configured for 2-in/2-out.
- Extensive revision in VST ASIO interface. New buffer design allows for faster ASIO processing.
- VST to ASIO output setup allows for each ASIO channel to be enabled. This removes unnecessary ASIO overheads when using less plugin outputs, e.g. of plugin's 8-outs only 6 are used.
- SSE2 128 bit processing
- Add plugin 'bypass' button in Settings. On stereo output DSP output is passed through. On multi-output, only the first 2 mapped ASIO outputs are used (for left & right). On mono output, DSP output is summed & averaged for selected mono ASIO output channel.
- B9 compiler option removed to ensure plugin compatibility
- Add VST safety option to limit control. VST editor is accessed through a prompt, bypass is not allowed, and autoplay is not done at startup.
- Other minor (& cosmetic) refinements
- DSP refinements:
- Auto-mute feature delays playback start. Mute is longer (and noticeable) when using a VST plugin (~2 seconds).
- New method implemented for DSP termination improving safety when using a VST plugin
- Other minor refinements (efficiency & ASIO)
- Performing "ASIO Settings" (via Settings) stops playback and a "restart" message is shown. ASIO drivers behave differently on setting changes resulting in unpredictable behaviour (for some).
cPlay documentation can be found at the new cMP² website .
Please REMOVE previous versions before installing cPlay 2.0b34. Normal (SSE2), SSE3, SSSE3 and SSE4 versions are available. Running cPlay on a CPU not supporting the required instruction set will cause cPlay to exit immediately. For best results, uncheck "Timer" setting and test different "Buffer" sizes.
As from cPlay 2.0b33, the installation folder is NOT deleted if one or more plugins were used. All plugin settings are stored in files named with the corresponding plugin name and '.vst' as extension. Subsequent cPlay releases will attempt to use these settings. That is, cPlay releases can be upgraded without loosing plugin settings. On installation of a new release and where plugins were previously used, the installer prompts to confirm reuse of the existing folder. Answer 'Yes'.
Edits: 03/20/10
I was wondering why there seems to be a preference for the sse2 download by approx. 4 to 1. Is there an advantage to SQ or is it MOBO related. T.
No worries!
SSE2 seems to be the new default minimum CPU instruction spec. Rarely will there be a new CPU with less capability. Its also the most common option across chip makers. Unfortunately, AMD's SSE3 is different to Intel's and so on. cPlay is built using Intel's compiler and supports it's version of the instruction set. That means other CPUs from VIA or AMD that offer better than SSE2 capability may not work - its a trail and error thing.
Here's more details on the build differences:
- Using a more advanced instruction set allows the compiler to implement more efficient algorithms and instructions resulting in faster performance.
- If you're using SoX, its important to use SSE3 or better build. SoX unfortunately is not compatible with an important compiler option that avoids an inefficient CPU instruction. SSE3 build was introduced for this reason.
- SSE4.1 instruction set allows for an important efficiency gain in ASIO output. cPlay manually caters for this by exploiting the new instruction set capability (SSE4 ASIO output design differs significantly). This is the recommended version as it offers significant improvement.
I prefer to have just 2 builds (SSE2 and SSE4). Given that SSE4 is not readily available (E7xxx onwards), SSE3 & SSSE3 does offer some improvements (more so with SoX).
SSE3 & SSSE3 does offer some improvements (more so with SoX).
I find a noticeable difference between the SSE2 and SSSE3 versions of cPlay when I've compared them (not that often, to be fair) on one or more of the three cMP2 systems I've built that are used regularly but do not support the SSE4 set.
If you are suggesting that new cPlay versions might not offer the SSSE3 option, I'd ask you to reconsider even though I appreciate the benefit of "streamlining" development and of sticking to "Intel only".
Are any other cPlay users and/or surrogates in the same position?
Dave
Hi Cics,
I have been playing with b34 SSSE3 on a Windows 7 laptop to Firewire DAC. This particular version sounds outstanding.
Thanks for all your work.
Alan
Merry Christmas!
I have been using Cplay from pretty much the beginning and feel that version 34 is the best yet. Thanks to Cics for donating his time to share this great product with the audio community.
Thanks and Best Regards,
Lou
Hi CICS , Thanks for release B34!
In my set up it is not just improvement, it is new level in
all aspect. I think you made some discovery, did find some key point.
I can repeat everything Theob said about B34 , but I think it is yours new beginning.
happy christmas everybody !
the ver.34 of cPlay is indeed another step forward for our favorite player !
on my system the definition, transparency and definition is significantly improved. very sharp and focused. even a notch more dynamic!
complex passages with strings and choir have considerable more air....
it seems that everything just stops much sooner.....
an impressive version !
Hello, cics and everybody! Really, ever after b29 I think that god is in details. This version demonstrates it brilliantly. I am very happy with b34. Also, I'd like to point out that I still use onboard audio and the changes I hear are not small. So onboard audio isn't so bad after all. I wonder, when we come to the point (and fear it) when the changes will be imagined or dubious. A question to cics - Will assigning affinity for cplay in task manager to cpu 1 and all other processes to cpu 0 be beneficial?
Serge.
THis is what I am doing now, dedicate processor 1 to cicsplay so that at least I can prevent those accidental pops and clicks shocks :)
Hello! Don't know if it is a placebo, but it sounds more "freely"? Almost no clicks (two very insignificant ones in 45 minutes with buffer set to tiny). Also I think it is worth it, if the comp is always on. Also I have switched off windows time process (this process is for time-zone and spring-winter time change routine)- very good step. Now task manager shows 37 mb used at startup
Serge.
Edits: 12/25/09
Hi Cics,
Thank you for another cPlay release.
Although much effort was directed to VST, and only minor changes were done on efficiency & ASIO, I do hear significant (!) changes ( I think: improvements !) regarding SQ. Or is the placebo-effect fooling me :-( ?
The mids, but especially the upper mids and highs are (much) more detailed and revealing. At first impression it sounds somewhat ‘HiFi-isch’ but after careful listening, the upper mids and highs are really better. But somehow they also seem to be a bit louder (+ 1 or 2 dB).
Other than this, nothing seems changed. This cPlay release still sounds ‘tuneful’ and is a pleasure to listen too.
May be the effect on upper mids and highs I hear, are related to the Klein & Hummel O300 titanium/fabric dome tweeters. Although the O300 have a reputation of being extremely neutral sounding, I have a love-and-hate relationship with these tweeters. They are extremely unforgiving. And there are also reports on that the Mogamie Gold cables I use, attenuate the highs somewhat. (Aljordan in his Prism Orpheus review. By the way: men, I do envy him. A Prism converter !! A Prism converter will me my next converter. When funds allow)
So, if nobody reports SQ changes in the (upper) midds and highs (although they are not subtle in my setup), than you should probably forget about it. It will probably be a minor SQ change, but one that is highlighted through my setup (Lavry -> Mogami gold -> K&H 0300 tweeter-> a harsh sounding un-damped livingroom (concrete walls, on one whole side a glass wall and a wooden floor).
But…… thankx to cPlay VST compatibility, I can now easily correct the perceived louder upper mids and highs, it with Overtone GEQ. :-)
Mark
fully AOB optimized cMP2 PC -> Lynx AES16 -> XLR AES/EBU -> Lavry Black DA10 -> XLR Mogami Gold -> Klein & Hummel O300
Mark you are right this is exactly where I hear the differences and where my audio priorities are. I think 34 is a significant success. I want to wait for a few more days to report detailed specifics.
Ok this version is superbly transparent. Let me 1st lay out how I am listening:
Cplay @ 44Khz -> Juli@ -> I2s with damping resistors -> Buf32s
I had been toggling between 192 and 44 khz for a while but version 2.0b34 makes 44khz preferred (for me in my system). The one area I disagree with Mark on is whether or not upper frequencies are elevated or not. I think not and they even sound rather lower in level to me. But the upper frequencies sound integrated into the total body of sound (as they do in real life). So all sounds are very very realistic to me. But the air around voices and instruments in a large classical piece is phenomenal! I have what I refer to as my demo pieces I play to sonically test changes in my system---mostly Paul Paray, Antal Dorati stuff on Mercury---50 years old but superbly recorded. They all sound better than I have ever experienced in the past. The imaging and placement of instruments is very precise w/o the highs being separated from the body of sound. All very disarming at 1st and then alas I don't expect anything less from one of Cic's new releases. But I fully thought this was going to be an evolutionary step (i.e., with all the vst changes) but whoaaa was I wrong. Another thing is noise seems to have vanished to infinitesimal levels. Listening to quiet sections is scary. The tonal effects are outstanding with all instruments sounding more like themselves than I have ever experienced. And voices, always a strength in cplay, are well just better...more body and emotion. Well done indeed!!
Ocassionally a poster will request of Cics no more changes after a certain release (undoubtedly because he may have rebalanced his system accordingly) and all of us long time posters will say no Cics please keep on going. I have to tell you if Cics ever wishes to take a rest for whatever reason now may be the time...I just can't see how it can get better than this.
Great feedback!
On 192k output using SRC@145db SNR, try "Medium" buffer setting. Improvement is subtle over previous "Tiny" setting.
Still prefer tiny/44. Although I have been a robust 192/145 src fan, this version of cplay just sounds more right to me at 44/tiny. Using upsampling, while preferred on some files, still sounds too diffuse and while great in the highs is not right to me in the bass nor mids for the majority of my music files. Maybe its because I now have a Buf32s dac hooked up via resistor damped I2S lines from Juli@ and the Buf32s also upsamples to 192. Differences are not big but definitely there in favor of 44 (ironic isn't it that now that I have the metallic issue solved I listen to 44).
I listened again to 44 at medium buffer...now this I like. More later.
BTW thanks for the recommendation of trying different settings.
Listening to 44 / medium buffer I get a more fleshed out midrange from mid bass to upper mids that is ripe in a good way. Woodwinds sound rounder, acoustic guitar sound boards sound wonderful, voices get more real. There are some tradeoffs (at least for me) the highs sound not so quite extended nor as integtrated with the total body of tone, depth is forshortened but the feel of live music or the gestalt of live music as J Gordon Holt described it, is all there in spades. Don't know which I prefer now: medium or tiny buffer @ 44 sr...they are both outstanding!Thank you Cics.
Edits: 01/06/10 01/06/10
cics, thanks for the suggestion of trying medium buffer in SNR. After switching from 'tiny', I found everything has better focus, especially the vocals and the drums. My foot is now tapping to the music.
My wife recently treated me to Verdi's La forza del destino at Vienna's Staatsoper. Link is in German, here's wiki's English version . We were lucky to get prime seats in the sweet spot (balcony directly opposite stage) - Musikverein is reputed to offer better acoustics.This is not my personal preference for music, however there are some pieces I'm familiar with and do enjoy. This was live in the purist sense, i.e. no electronics - all 65 piece orchestra consisted of natural instruments. The sound projected from stage (mainly vocals) was truly impactful!
Some things I noted:
- HF or "treble" seemed lower than found on recordings. In fact rarely were high frequencies present - seems like recording engineers apply extra treble. At first this sounded somewhat "rolled-off".
- There was a solo violin piece which was sadly not long enough but the performance was very emotional and sweet.
- Vocals were simply outstanding (especially transients) - a popular piece was rendered with such beauty and emotion that someone in the audience screamed "Bravo!!!" before it ended (I got the feeling one has to be very PC at this venue).
- Within an hour of the performance and especially after the interval we were simply overwhelmed with the natural harmonics and rich sound. Everything just came together in a very pleasing way - the HF rolloff I thought would be a problem was not the case. Mids were completely glare-free but certainly not lacking in bite or beauty.
- Wind instruments in particular had a pleasing sweetness.
- Overall impression I got was recorded instruments and vocals when reproduced are expected (or tweaked) to sound "fuller" then what they actually render. A tube thing or maybe too much reverbs?
A week later and armed with these insights I wanted to compare how close my setup could get. cPlay is set to output 192k SRC@145db SNR using Medium buffer - this feeds my Scarlatti DAC (using filter 6 - "linear" which offers least interference on HiRes input). System ( equipment used ) takes ~1-2 hours of warm-up to reach peak performance.
Aside: a few weeks back, I experimented with RAM timings and came to the conclusion that "Auto" is best. However, I switched to a 512MB Kingston HyperX module as it offers faster timings (3-3-3-7 over 3-3-3-8). Although 256MB uses slightly less power, the benefit of faster "default" timings seems to be a winner.For starters, B34 is a significant step closer than B33. So much so that the sound at home renders a similar emotional experience with sweet mids, beautifully preserved harmonics and transients that are just soooo close. Mids and Highs as you put it is fully integrated. Likewise, you say it best "The tonal effects are outstanding with all instruments sounding more like themselves than I have ever experienced. And voices, always a strength in cplay, are well just better...more body and emotion.".
Vocals are splendid, emotional and enjoyable but its one area that I feel could improve further. What I heard at the Staatsoper was perfection and it's addictive. I don't know where this will come from but I'm searching...
Edits: 01/07/10
...for the 512MB Kingston HyperX module. Do you recommend inmates to go there? Said another way is this a big pop in sq?
I was wondering whether you were using critical or realtime. T.
No worries!
I only use 'Critical' which is the best setting with ASIO soundcards. When using ASIO4ALL (i.e. soundcard without native ASIO drivers), its worth experimenting with "Realtime".
Here's an explanation of 'Critical' setting:
'Optimize Critical' setting allows for stable low latency playback. This setting optimizes L1 Data & Instruction cache for audio sample delivery which means lower jitter. This is achieved by locking the most critical audio thread to CPU1. Although load on CPU1 will be insignificant, this implementation removes need for a real-time OS and is superior (as a CPU core is dedicated to audio).
The medium buffer truly is remarkable in the midrange. If you can improve upon 34 it will indeed be amazing to those of us that appreciate music in the home.
Again well done and thank you!
I'm using this power supply setup where CPU PS moves to the "dirty" side. This is an improvement (especially when upsampling) over current cMP² specifications (but not the more advanced power mods like a dedicated CPU battery or linear PS).
Hello cics,
I just have a quick question about the 2nd ATX PSU of your setup. Did you remove the fan, adjusted the potentiometer and removed all unecessary cables? Or since it's sitting outside your CMP^2 box and used for the "dirty peripherals", you did not bother doing so.
Thanks
I am using another atx ps to drive cpu, all hd's and my dvd/cd drive and I left the cables and fan intact (in case I ever want to use again) and I get good sonics.
I could try this. I was on a linear ps for p4 then had an issue so now I am totally on my pc ps for all but juli@ digital power.
...I put my spare power supply onto all my dirty units including p4. Yes not as good as an external linear supply for p4 but better than where I was. Now 44/medium and 192/medium are closer with the nod going to 192/medium. I still don't like 192/medium from the standpoint that it uses lots of cpu power and I get a lazy mouse. My system is always bordering on instability. But, here it is, it sounds wonderful...air space and liquidity. With the new ps for p4 almost as good as 44/medium for bass/mids.
Thank you cics again
T,
I got the lazy mouse too the other day. It was at the worst possible time, going from classical to compressed rock. The volume was super high and I had to turn it down fast. But the mouse hung for 5 seconds.
Turns out I made a mistake and put one flac album on my main rig by mistake. Cics confirmed that flac is the reason for the lazy mouse, and I dont get that with .wav.
Are you using flac?
No I am using wav exclusively. If I run 44 the mouse is prompt. When I say lazy what I mean is that the screen cursor is a fraction of a second behind hand/finger input motions to the mouse. So it doesn't cause dropouts or such its just an indicator my system is at the margin of stability.
OH,
I had a different issue. With wav, no problem. But flacs the mouse would hang for 5 seconds.
I wrestled with this a bit. There's a test case where ASIO fails at initialisation. This is rare but can occur, e.g. using an output rate not supported by the soundcard. cPlay will show the error and not proceed.
With a plugin in use, I decided on having the plugin go back to its original default settings when an ASIO initialisation error happens. The other alternative is to have the plugin use its current settings. Although either scenario can be easily supported I felt the current method is more cautious and safe (as the ASIO error may be unusual and result in different behaviour).
What this means is should an ASIO error happen at startup (and not during playback) cPlay will revert to the plugin's default settings. In this case, settings must be redone. This can be done either manually or by restoring the "plugin.vst" file in cPlay's installation folder. That is, its always a good idea to keep a backup of your plugin settings (copy "plugin.vst" to "plugin.vst.bak").
I will only say that this release while not only good could be transformative. I won't say more now. I should really give it a few days to make sure of what I am hearing.
Thank you cics and happy holidays to you and yours.
This is a significant release. My notes are some 20 pages long detailing changes & designs that went into b34!
Although much is dedicated to VST, there are critical low level bits on DSP & ASIO. Item 2.3 of the changelog says "Other minor refinements (efficiency & ASIO)". This is reflective of the code changes done (which is correct) but not its significance regarding SQ.
I'd like to release B34 before end of December. It's primary focus is vst. Plan so far:
- Minor cosmetics & fixes (silence between tracks, UI 64 bit processing flag incorrect)
- Better method for ASIO processing
- Support more plugins (vst version 2300+)
Let me know of other areas needing attention.
B34 is in final testing and has also benefited from a small DSP change. VST changes have been complex and extensive. There's a new ASIO interface which is also more customizable allowing for faster VST/ASIO processing. Extensive SSE2 128 bit processing has been introduced. Items like silence between tracks, more plugin support and improved usability has been done.
Changes suggested:
- Remembering polarity switch (as in digital "absolute polarity" switch) for each track. Not for b34 but its useful.
- Cue sheets in FLAC & better UI. Haven't forgotten this. Idea I'm testing with is providing a "Remote Control" capability in cPlay. A program with whatever UI interacts with cPlay through "PostMessage" api. Nice thing about this approach is it allows for robust headerless playback.
- Saving audio files. There must be a plugin that can do this. Plugin will however run full elapsed time. Direct file save is not for b34.
- Dynamic parameters. This can be done with any cPlay release by directly changing its .ini file before starting cPlay.
- VST Editor button with "prompt". Not sure about this - I really like accessing the editor panel quickly. Point on making Allocator changes is a valid risk. Another option is to have Allocator not offer an Editor (cPlay disables the VST button) or maybe just something that shows all the details but doesn't allow changes (an optional feature in Allocator?). Allocator then offers an offline tool for parameter settings.
- Indicate resampler used or bypassed. This is partly available through Diagnostics. On Stop (and diagnostics enabled), cPlay shows a record of all tracks played with other details including whether resampler was used or bypassed.
- VST bypass capability. Functioning in b34 - this is another Allocator risk so I've put the "bypass" button in Settings panel. Should the VST Editor button also move here?
hey cics,
thanks again for your effort on the VST issue.
the V33 works faultless on "remembering" settings in "allocator", but I think it´s still a valid security to prompt allocator screen to appear before signal is "live".
-regardless of confidence on system !
it´s bloody serious if anything goes wrong on digital XO.....
i think you should implement the VST features as "tick here" switches in configuration settings of cPlay, like samplerate, AWE, repeat and others.
the switch between VST and not, maybe implemented in the .ini file.
just my thoghts.....
You can change these via cPlay ini IF you know what setting suits a given listing in cMP, but presently cMP does not show, for instance, the sample rate of files in each cuesheet. If it did, at least to the extent of distinguishing 44.1 and multiples from 48 and multiples, then one could do this ini adjustment before starting cPlay.
Loooong shot, I suppose, but somewhere down the line I wish I could mark each music piece to automatically use the "phase" that I prefer with it. My Magnepan speakers do react well to this kind of choice.
BTW, I assume that this is really a simulated/digital "absolute polarity" switch rather than "phase" (as in inter-driver phase), right?
Hi Cics.
I only have two wishes:
1. Be able to select the font size or an horizontal scroll bar (to be able to read long track titles).
2. Read the cue sheet information directly embebed in the flac files. These is, no need cue file with flac files.
Thanks a lot and happy christmas !!.
Ignacio.
I have asked before. It would be great to provide option that no resampling is performed.
i.e. Add an option so that I can select "no resampling by cplay whenever possible". If I select this option, cplay will
1/ find out the sampling rate of the audio file
2/ check whether the sampling rate is supported by the audio device
3/ if the audio device support the source sampling frequency, no resampling be performed.
4/ if the audio device do not support the source sampling frequency, resample the source to the default output audio sampling frequency
Is it possible to give us this function? I prefer using my DAC built-in upsample function instead.
Thanks!
Hi cics
Is it possible to implement a "save to file" function in cPlay, whereby the upsampled output can be stored on the hard drive ? This would be a very helpful addition for those wanting to use cPlay's various upsampling routines, from a convenient user interface.
Thanks in advance...
Grant
That's not a Toy... IT'S A TOOL !!
Hi Cics,
My wish list for Christmas :P
1. Support for 640x480 screen as I find 640x480x16 gives best result
2. All settings can be controlled during program initialization So that we can create custom playlists and playback each song at specific SRC or buffer settings, e.g. VOX linear 91 tiny buffer for symphonies, VOX linear 99...
e.g. c:\cicsplay.exe -src VOX linear 95 -samplerate 192 -buffer tiny ... c:\mysong.wav
hey cics,
yeah, that buffer overflow would be very good to get rid of.
one little suggestion :
as digital crossovers are very dangerous when settings are not correct, I´m always careful to check them before applying signal. maybe it would be comfortable to launch the VST application window by a "tick-box" instead of a button. one could choose to keep it hidden or open it everytime cplay is started.
BTW. I have not managed to fix the problem that allocator is not remembering its settings....very annoying !
-are there other inmates who have settled their impression on the SQ of V33 ?
I still find my speakers more acoustic "visible" than in previous releases of cPlay......
having skipped JACK and implemented cMP my system have improved very significant !
thanks a bunch for this !
hey Leif,
That is weird. My settings ARE remembered. I have only ever changed them from cplay.
I assume you are making changes inside cplay and it is still not remembering them??
I wonder what it could be.
And I was mostly dialing crossovers but I did listen to 34 without any vst fun and certainly didnt have any impression that the speakers werent disappearing. It wasnt any different in that regard than other versions.
Did you maintain similar settings for buffer and upsampler quality. And Lynx asio buffer settings??
hey dawnrazor,
well, there are always a reasons for problems.
the annoying lack of remembering the settings in allocator, were of course the switch of the "active" buttons.
the previous settings were configured to use the bass, mid1 and the treble setting, and not like with the VST output the bass, mid1 and mid2......
everything fine now !
silly me !
Hi PM,
Perhaps this is unrelated...
I have a similar problem with J River MC forgetting Allocator settings.
What do you mean by "not like with the VST output ..."?
Is it something in cPlay?
My active buttons are only for bass and mid1.
Thanks
hey bibo01,
I had the trouble because the "active" (the top one of 4 buttons next to the sliders of allocator) button was configured differently when JACK was setting the outputs.
-therefore the allocator reversed to the old settings everytime.
hope it clears the matter.....
Hey Leif,
That kind of explains it. Looks like cplay only remembers the settings you make from the vst button on cplay. This is why I haven't had that problem- I have ONLY made changes that way.
Good thing is you got it figured out, hopefully with no damage.
Cics, I know this is not what you are looking or or what floats your boat, but I am still hoping for a way to see whether to set cPlay to 176.2k or 192k (alternatively, it might be 88.1 vs. 96) for a given album based on its samplerate. I have given up on automatic playback rate setting, but at the least cMP could indicate for each listing(cuesheet) whether it refers to files whose rate is a multiple of 44.1k or 48k. The former, for instance, could be in regular font as now and the latter could be in italic or underlined or different font, whichever would work best. This should not affect playback performance at all but would be quite helpful. (For playlist cuesheets comprising files of differing rates, I guess, the rate of the first entry would have to be indicated or it could default to regular font, if doable.)
I put all of this logic into a version of CplayListEditor that is not available for regular download. You can configure it to start CPlay at a sample rate that is an even multiple of whatever files are in the playlist if they are all the same.
Problem is that I'm not sure if my application can start CPlay in the right manner within CMP. I'll get you a copy if you want to try it.
Alan
Hey Cics,
Vst is a great addition to cPlay and good news that you are making it better.
Looks like you are addressing the few issues nicely.
One thing that might be nice would be a button to turn off and on the vst plugin. While that would be deadly for crossover plugins, it would be nice for those using eqs or room correction, etc. You can kind of do this now, but you have to delete things and start over.
Thanks again for adding vst support!!!
dear forum,
it feels a bit strange to ask the most elementary questions sometimes, but being new to my much adored cMP2 / allocator VST rig, I´m just puzzled how to get my high resolution FLAC files played in cMP2..... is .cue files the only format cMP can handle ?
if so, how to convert these FLAC files ?
thanks for your enduring support !
Hey PM,
try creating cue sheets using Alans program. Cue is just a text file that references the flacs...the files themselves dont change.
Think of a cue as a passport...it references you but isnt you, and you need it to get into the country.
FWIW it took me 20 minutes to figure out why my allocator full version wasnt working. With the cmp2 video settings and my old eyes, I couldnt see the text and that page is different from the trial version. It had "apply" and "edit". There were 4 red boxes before the apply button. I thought this was somehow related to the 4 parts of the crossover....but 20 minutes of dicking around and I finally realized that those boxes were for the serial number to authenticate the version!!!!!
DUH
This 2.0 b33 version is a dream come true for me, cMP2 with VST plugin.
But I have a problem using Convolver VST plugin in cMP mode, when cPlay starts, it just hangs there showing the hourglass icon. While in XP mode, running cMP, it works and sounds really nice. Am I doing something wrong here?
Anyway, thank you very much for what you have done, cics.
/Sopian
In cMP test with both lsass and svchost set to NO suspend.
What's happening is cPlay is loading the plugin but it's not happenning fast enough causing cMP to lock (indefinitie wait). You can fix this by running cPlay once from explorer and thereafter from cMP (the first cPlay start loads the plugin into cache which makes subsequent calls very fast).
Cics, I followed your instructions to set lsass and svchost to NO suspend. It works.
Thanks a lot, you really are the man !
Hi Cics,Thank you for the VST-functionality. Now I finally have a quality EQ function! (my purist hardware setup has no hardware regulation for bass and/or treble)
Reading the few posts on the VST-plug-in in this thread, I found it hard to abstract information from them on how to install effectChainer and the Voxengo Overtone GEQ windows VST plugin. So I wrote how I installed them, to save other inmates some time.
* Installing EffectChainer in the cMP2 setup.
Download EffectChainer and execute the executable installation file.
http://www.acondigital.com/us_effectchainer_screenshots.html
Installing EffectChanier creates this pad on the PC.
?:\Program Files\Steinberg\VstPlugins\Acon Digital Media
The effectchainer.dll VST plugin is in this directory.
Later on you need to point cPlay to this file in the cPlay VST setup window.* Installing Overtone GEQ
Dowload the Overtone GEO zip-file
http://www.voxengo.com/product/overtonegeq/
Unzip VoxengoOvertoneGEQ_18_WinVST_setup.zip (on a normal XP installation).
Copy the Overtone GEQ.dll to the Vstplugin diretory in the cMP setup.
C:\Program Files\Steinberg\VstPlugins\Acon Digital Media
Later on you may have to point EffectChainer to it.* configuring cPlay for VST-plug-ins
The VST button in cPlay’s mainpage is ‘grey-out’ and doesn’t function yet.
First cPlay needs to know where the EffectChainer.dll is located.
Go to the cPlay settings window, and click the VST-button
When in the cPlay VST Plugin settings window, click on the VST Plugin button and point cPlay to the effectchainer.dll file ( in ?:\Program Files\Steinberg\VstPlugins\Acon Digital Media directory).
Restart cPlay* Configuring EffectChainer at first startup.
Go to cPlay’s mainpage and start some music top see if cPlay still works.
The VST button should be green now and functioning.
Click on the VST button and EffectChainer will start.
-> EffectChainer will ask you now if it should scan your whole PC for VST-plugins.
I chose: ‘no’. (because I know myself where the VST plugins are located)
-> Than EffectChainer will ask you to scan for DirectX plugins.
I chose: ‘no’. (because I know myself where the VST plugins are located)
-> than EffectChainer will ask you to do an initial scan of the C:\Program Files\Steinberg\VstPlugins\Acon Digital Media directory.
I chose ‘Yes’ since I put the Overtone GEQ.dll in to the C:\Program Files\Steinberg\VstPlugins\Acon Digital Media diretory.
The EffectChainer window will now come up.* loading a VST-plugin in EffectChainer.
After EffectChainer has started there’s nothing much in the Effectchainer mainwindow other than an empty rectangular area that turns bleu when hovering over it with your mouse. It then turns bleu. Click this now bleu rectangular area and a sub-menu will pop-up. Go to VST in this pop-up menu, than to Accon Digital Media and under ‘Accon Digital Media’ you should see listed both dll’s that are in the Accon Digital Media folder. Chose ‘Overtone GEQ’. The bleu area will now show the VST Overtone GEQ plugin and 3 buttons for configuration. Click the most right button (spanner icon) and Voxengo Overtone GEQ user panel will pop-up.
Now play around a little with Voxengo Overtone. Have fun.A nice cPlay-feature would be that EffectChainer could be set as the standard VST plugin which will start automatically when cplay is started.
I have some trouble (don’t know how) to save a session, so that it pops-up again when EffectChainer is started again next time. Also I haven’t found out how to load Overtone GEQ automatically when EffectChainer is lauched.
Nonetheless I’m very happy with the cPlay VST plugin feature.
Hope this short ‘how to’ saves other inmates some time.
fully AOB optimized cMP2 PC -> Lynx AES16 -> XLR AES/EBU -> Lavry Black DA10 -> XLR Mogami Gold -> Klein & Hummel O300
Edits: 12/08/09 12/08/09 12/08/09
Thanks for documenting the setup.
If only one plugin is used (likely the most common case) do not use a chainer. cPlay remembers all settings and will automatically start it on playback, i.e. no need to edit the plugin and load settings etc.. Using Overtone GEQ directly is not possible in B33 - will be fixed in B34.
Likewise, cPlay does the same with EffectChainer but unfortunately EffectChainer does not offer much by way of auto-loading the setup. That's why I think Console may be a better chainer. Console also offers incredible flexibility with routing. Example:
I just need someone to verify if Console does auto-loading of all plugin settings. cPlay will certainly startup Console with its base settings.
hey cics,
I wrote to jan thuneau about allocator not remembering (recall) the last settings when cPlay exits to change album in cMP2 mode.
he replies that cPlay needs a certain change of its code to recall the settings....
one have to be very careful to cut off the audiostream and correct the settings manually, before every change of album.
kind regards
If you look at cPlay's diagnostics output, you'll see the plugin settings being applied. From my tests, cPlay saved all plugin settings before exiting and then restores them. I'm not familiar with all the Allocator settings - is there a specific problem area?
dear forum,
unfortunately I´m still having problems to get the cMP2 mode running.
I believe to have set everything right, but on launching cMP my albums are displayed correctly, but upon selecting one, cPlay does not launch. it shows following :
back
***album/artist***
back
....and nothing happens.
on selecting cPlay I get an error msg. saying :
"no cue, flac or wav files supplied
usage : cicsPlay myCue.cue
....what´s going on ?
please help.
kind regards
to all problems there is a solution.
english and german windows do not have the same file treads.....the cics memoryPlayer .pth file had to be corrected.
thanks guys
delighted about this VST plug-in, but I get following error message after correct (?) configuration :
cannot find component MSVCR71.dll
-this driver (component) is sitting next to the allocator.dll in the allocator folder. does it belong somewhere else ?
by pointing at the allocator.dll driver, cPlay displays correct setup for allocatot in the VST menu.
what is wrong ?
can´t wait to get this working.....
.
hey cics,
that was just the detail that was missing !
-thanks a bunch.
apart from that the 33th issue of cPlay obviously is different in terms of sonics to the "bloomed" 32 version, I can report that without the JACK my system has now gained further ease and precision. especially the "ringing" of piano has got dramatically improved...
-another reduction of jitter I suppose !
this JACK thing has prevented my adoption of the real cMP2 mode so I´m really, really grateful for this VST version. -that´s gonna be my thrill tomorrow to implement !
my deepest bow of respect !
I would like to know.
Is this a convenience or a sound quality enhancement?
Thanks,
Rick McInnis
VST is Steinberg's standard for applying DSP sound effects to audio streams. Effects can be absolutely anything one desires (stereo to mono, EQs, Stereo to 5.1., etc.). There's a huge selection - one needs to be careful as some have bad quality (bugs and/or process in integer causing quality loss). BTW - another common acronym is VSTi which are DSP components that act like musical instruments.
Here's Steinberg's explanation of a VST plugin:
Essentially, a VST Plug-in is a pure audio processing component, and not an audio application: It is a component that is utilized within a host application. This host application provides the audio streams that are processed by the plug-in's code.
Generally speaking, a VST plug-in it can take a stream of audio data, apply a process to the audio, and return the result to the host application. A VST Plug-In performs its process normally using the processor of the computer; It does not necessarily need dedicated digital signal processors. The audio stream is broken down into a series of blocks. The host supplies the blocks in sequence. The host and its current environment control the block-size. The VST Plug-In maintains the status of all its own parameters relating to the running process: The host does not maintain any information about what the plug-in did with the last block of data it processed.
From the host application’s point of view, a VST Plug-In is a black box with an arbitrary number of inputs, outputs (MIDI or Audio), and associated parameters. The host needs no implicit knowledge of the plug-in's process to be able to use it. The plug-in process can use whatever parameters it wishes, internally to the process, but depending on the capabilities of the host, it can allow the changes to user parameters to be automated by the host.
Allocator is formally called the Frequency Allocator and allows one to eliminate speaker x-overs (and associated nasties). Instead, we use digital processing: stereo is converted into pairs of x-over frequencies (low L+R, mid L+R, etc.) whose amplified analogue signal is then fed directly to a speaker. Really interesting stuff and could very well be the next level of sonic performance. Playmate & Dawnrazor have the most experience with this.
Another plugin of value is Room Correction - don't know of any reasonable ones except for one vendor who sells it for ~$700 and may be limited to 96k.
Hey Cics,
And dont forget that vst is a large world that might at first glance have things that we might not use.
However, on my normal computer I was easily able to use a few of the free vst plugins I found. One I am using is the headphone vst plugin that adds some crossfeed to help with headphone listening and is called hdphx. This is handy because I was planning at some point to build another cmp2 box for ripping, headphones and maybe internet radio.
hdphx
There are also a plethora of TUBE plugins that simulate that old school tube sound. some are free to try and some arent, but maybe that would be a good thing for some systems. Probably not exactly consistent with the whole cMP2 concept, but nice to have that capability.
one of many
And one could use something like Izotopes Ozone to really tailor the sound.
Of course there are many eq plugins that could be used to tame the room.
Anyhow this IMHO gives cplay a whole new list of capabilities and there are probably a bunch of other plugins that could be useful that escape me.
It just worked flawlessly and the only issues were plugins that had multiple channel inputs, and remembering to exit cplay and restarting once the vst plugin was selected.
Thanks again, and is there a way to do more than one vst plugin at a time, or is one as good as it gets?
D
So if one would use Allocator as a crossover in a 3 way system would one need 3 dacs or does allocator output analogue that can be routed directly to speakers? If the latter is true one would really a great power supply in the pc. Is this an oversimplified statement of how it works?I am also truly interested in how one would use vst's to improve latencies.
Also does the Allocator replace a sound card like Juli@, otherwise I do not see how it outputs several pass bands to 'directly feed applifiers to individual speaker elements'.
Edits: 12/06/09
Hey Theob,
your soundcard needs to have multiple digital outputs like the Lynx Aes16, or multiple analog outs like the lynx 2b I have or other pro devices like the RME Firefaces for example.
I bought the 2b for this purpose and cant think of 3 better dacs for 900$.
Though I bet you could use the allocator and the ANALOG AND DIGITAL outs on your Juli@ if you wanted, just map it correctly in allocator. Say send the analog outs to a sub and the digital to your dac and main speakers and just use it as a 2-way sub crossover.
I have been using 3 pairs of outputs from the lynx for many years in a triamped passive config, and the powersupply whether in my hush or cmp2 rig has just worked. But your point is taken. I plan to use a dirty supply on the cpu per cics new instructions.
Anyhow I cant wait till thursday to get it going on my audio rig.
Though I think vst plugins will never improve latency, probably just the opposite. And Allocator will add a ton of latency in order to do its magic...I am hoping that that will be a slight downside compared to the upside of removing my crossovers.
Thanks for the answer but it seems it is more practical with the Lynx. I use I2S out of Juli@ and can either activate digital out (spdif) or analogue out (analogue or I2S)but not both. I sure wish there was another sound card that had the flexibility of I2S, external power and multiple outputs for a allocator type crossover.
Hey Theob,
IIRC the i2s from the Juli@ is a mod right?
So why couldn't one mod a firewire device like the fireface or maybe some of the m-audio devices?
Also, I wasn't aware of the limitation on the Juli@ that limits it from doing both digital and analog. My Lynx does this easy enough.
If you really want to throw money at the problem, get the Lynx 2b and one of those $1000 external pci cases and run it with a nice linear. IIRC a pci card with onboard dacs DOES use I2s :). Then you can have crossovers. Of course it may be best just to use a firewire device like the Rme Fireface.
I2s from juli@ is not a mod, its right there on the digital section (ie., analogue board removed). To implement is a not trivial though(soldering directly onto pin outs). But no I can definitely see the advantages of a Lynx now.
Edits: 12/07/09
I've wrestled with this a lot and decided on a single robust VST plugin design. This allows cPlay to focus on achieving very low ASIO latency performance and no compromises on SQ. There are some refinements to come but I wanted this release out asap to get feedback. So far the design already allows for low latency playback without compromising on using intensive plugins like Allocator ☺
Chaining is desired and is best achieved using an Effect Chainer . cPlay only sees 1 plugin which in turn allows one to chain extensively. Here's an example of effects (EQ + Spectrum Analyser):
So how I'm seeing it is cPlay does the heavy lifting on resampling, vst and ASIO; and a 'Chainer' does the sexy GUI bits, parameter admin and multi-plugin chaining.
I'm hoping a volunteer developer will offer a similar product as above with 2-in/8-out capability. What will be very sexy is sophisticated audio routing using a 'CAD' like tool (think of boxes with I/O stubs that one can freely link). The possibilities are endless...
Hey CIcs,
That looks like what I was thinking of. Jriver does kind of the same thing, you just put the first one on top and the rest in the order you want them to work in.
Thanks for finding this. Ill try it out later this week.
I didnt quite understand playmates folder problem, but I think I had the same thing. Cplay remembers the last folder you were in and so when you set the vst plugin, cplay uses that folder as the new default for selecting music. Not a huge deal, but did you say double clicking would help this??
Thanks again!!
Are you aware of this software http://console.jp/en/ seems to have 2 in and multiple outs as well as combining plug-ins.
"(think of boxes with I/O stubs that one can freely link)"
Seems to have this too.
Another item of interest is Convolver which can apply FIR filters for crossover and room correction from different sources. There is a VST plug-in version. http://convolver.sourceforge.net/
An alternative to Allocator is Audiolense which also has room correction
http://www.juicehifi.com/index.html
Audiolense works with Convolver.
Audiolense is limited to 96khz but its other advantages may out weigh its limitations.
Very exciting stuff!
Thanks again Cics for your time and sharing your knowledge.
PC XPpro TC Electronic Konnekt-8 firewire i/o, 2-Tact 2150's, Genesis 500 modified speakers, Virtual Dynamics & XLO cables
Thanks, I wasn't aware of Console until now. Brilliant! Here's a little setup I quickly put together in cPlay:
Basic 2-in/8-out setup. I've chained EQ > Allocator > Spectrum Analyser (to verify Allocator's 6 outputs). Output channels 7&8 are not connected so I use it for spectrum analysis of whichever Allocator output.
As a plugin chainer its very powerful (but not free, $54) - don't know how well it managers settings and so on (I just used the trial version which doesn't allow saving).
Dude how does one get convolver to work with audiolense?
I have convolver showing up in cplay but which file do I chose here:
The file you need here is the filter file created in Audiolense , a short wav file of 64 to 512 KB depending on the type of filter created (less, default or more correction).
Chris
Thanks.
But I was trying to try its crossovers.
Does it also creat .wav files for use with convolver??
Go to the http below; the info is a little better laid out than the website.
I have not used the crossover function either but it looks like it is done by routing the channels.
http://www.duffroomcorrection.com/images/f/fb/Audiolense_wiki.pdf
PC XPpro TC Electronic Konnekt-8 firewire i/o, 2-Tact 2150's, Genesis 500 modified speakers, Virtual Dynamics & XLO cables
d
Sorry didn't catch it. I don't know the answer since I only use DRC filters .
Chris
kX offers a sexy way to chain things (it's NOT a vst plugin but an alternative EMU soundcard driver). Idea is the left-side is our primary inputs (left & right) and the right-side allows up to 8 outputs. In between one can do crazy things like this:
hey cics,
I´ve found two small "behaviours" of cPlay / allocator VST that possibly could be fixed.
when selecting a new track directly, without selecting STOP first, there are remains of the last buffer signal audible. can that be suppressed ?
the second detail may actually better be solved by thuneau, but when selecting a new album via the "folder" icon of cPlay, the window displays "desktop, documents" folders, and not the specific one that contains my music files. can that be set differently ?
-furthermore the selecting disturbs the crossover configuration of allocator, which are stored and recalled by the same kind of windows/file folder menu.
I´ve listened to this new jack-less setup and there is definitely less distortion (especially on complex material) and noticeable more "flow".....
-maybe it´s a renewed challenge for me to adjust the position of my dipole planars, but in this cPlay V33 the magnepan´s don´t quite "dissapear" as they have previously.
-slightly strange....
kind regards
Play-mate says:
-maybe it´s a renewed challenge for me to adjust the position of my dipole planars, but in this cPlay V33 the magnepan´s don´t quite "dissapear" as they have previously
I agree. I have Magnepan MMGs. Just tried V33 this evening (no VST)...the speakers become more obvious sources in V33. V32 had become the best version for me in this regard, probably more musical overall, as well (but just 2 hours of listening may be too short).
when selecting a new track directly, without selecting STOP first, there are remains of the last buffer signal audible. can that be suppressed ?
Noticed this too. It's internal to Allocator's buffer processing and cPlay cannot change that directly. Instead cPlay must send some silence then switch. (Not a priority for now).
the second detail may actually better be solved by thuneau, but when selecting a new album via the "folder" icon of cPlay, the window displays "desktop, documents" folders, and not the specific one that contains my music files. can that be set differently ?
-furthermore the selecting disturbs the crossover configuration of allocator, which are stored and recalled by the same kind of windows/file folder menu.
Try double-clicking your music file. It causes file browser to remember this last folder you looked at.
I´ve listened to this new jack-less setup and there is definitely less distortion (especially on complex material) and noticeable more "flow".....
Also test with cMP - you should get further improvement. I wouldn't change settings yet until you can get cMP going (initially don't suspend lsass & svshost). Optimise "Critical" is best. What latency are you using? On Lynx you should be able to do 32 samples.
Dear cics,
The crossover thing could be interesting.
Room correction is too much like active suspensions on cars, and elaborate speed controls for turntables, one worries they are just a little late in making their corrections and you end up with tail chasing. If only we were dogs we would find that enjoyable ...
Concerning VST, I was wondering if I was missing something. I think this is something I can do without for the time being.
Almost ready to give digital sound another go. I think I have missed 13 or 14 versions so I expect I will be amazed at what has happened in the interim. Not that I have tired of analogue. There are too many recordings unavailable on LP, what else can one do?
I like the idea of a separate supply for the CPU. I think you are on the right track with this. There might be something gained by the heroic linear approach but I suspect not as much as the trouble would promise.
(I found those little supplies you recommended for five dollars on EBAY.)
Certainly switchers have their problems but thinking about what the components send back to the supply which is then reflected back into the rest of the system; this is the subtlety that defines creative thinking.
I would add that if one does the linear supply they should follow your thought and use a separate linear supply for the CPU. I suspect my battery experiment was likely to have seemed successful more for the isolation of the CPU as to the effects of battery power.
cics, you never cease to amaze and delight me. I wish you the best for the closing of the year and look forward to what you discover in the new year. Even though I have not been using cmp2 I look at the forum everyday to see what is happening. This is one of the few vital, exciting threads in all of audio.
With sincere appreciation,
Rick McInnis
Refer to ReadMe.txt for important details & specifications.
Change Log (2.0b33 release):
- Refinements in DSP and ASIO
- VST Plugin support in DSP processing. Only stereo plugins (i.e. 2 input channels) can be used with up to 8 output channels. ASIO performance is maintained for 2-in/2-out plugins (stereo output). For multi-channel output, best ASIO performance is achieved when latency is 64 samples or less.
DSP processing (including volume control) preceeds plugin. See specifications for more details.
Usage: Set VST plugin via Settings > VST panel. Main panel enables VST button if plugin offers an Editor. Plugin settings are saved.
cPlay documentation can be found at the new cMP² website .
Please REMOVE previous versions before installing cPlay 2.0b33. Normal (SSE2), SSE3 B9, SSSE3 B9 and SSE4 B9 versions are available. Running cPlay on a CPU not supporting the required instruction set will cause cPlay to exit immediately. For best results, uncheck "Timer" setting and test different "Buffer" sizes.
Comments: Both DSP & ASIO components sees subtle but important changes in this release. Changes were informed by the recent cMP PS improvement, B31 & B32. An ASIO change in B32 was reversed whilst other new changes were applied.
If it is 33 or something I have done (e.g., resistor damped I2S lines) I no longer have any issues with this dreadful noise. Hopefully this will continue through 34 and beyond.
Over 7 days noise free confirmed.
Nice!!!
Hi Cics,
In my setup cPlay 2,0_B33 shows much improvement in the whole midrange. From upper bass / lower mids to the upper mids, the whole mid-range has more resolution, more detail, more ambiance, more rhythm , more space. In one word: beautiful mids.
I especially choose the K&H O300 active speakers because of there famous ATC soft mid domes, and cPlay 2,0_B33 makes them shine. :-)
fully AOB optimized cMP2 PC -> Lynx AES16 -> XLR AES/EBU -> Lavry Black DA10 -> XLR Mogami Gold -> Klein & Hummel O300
I still love this version perhaps even more than my original post suggests. But after about 18-20 hours of play a funny thing happened--no more static-ee noise (this is the juli@ 1/2 version of the dreaded metallic noise). I have tried lots of things to solve this and know of only one true solution: all battery power plus external 5 volt power to juli@ sound card (as so diligently discovered by Bertel). I have mitigated the noise a lot by futzing with clock speed and core voltages (a little higher than the Cic's recommended values)and with damping resistors on the I2S lines to the point where it was managable but cPlay 2.0b33 puts me over the goal line. Maybe its too early to declare victory. Time will tell but I'm lovin' it. (Yes clock speed is set correctly).
Not since version 18 has a cplay release been so successful @ minimzing this noise. Subsequent releases after 18 while being sonically superior always had the noise issue. Now we get both in cPlay 2.0b33... brilliant Cics!!!
So question: anybody else experience this?
Theob,
I've been watching your various comments on your Buffalo32 setup on the Juli@ digital card with much interest. I've been working on AKM-based add-on I2S-connected DAC cards since middle summer and am close to having a Buffalo32 card setup for another inmate (who has been very patient as work and a couple of minor illnesses have slowed me down on getting it done for him).
Very curious how you've got your setup configured. Especially curious about the 'damping resistors'. Series or parallel, what value, what
brand & type of resistors?
Series resistors on the I2S lines were key to getting the AKM-based DAC cards working well. In researching this, I found that the Juli@ analog card has 50R series resistors in the I2S connections to its AK4358 DAC. The evaluation boards for the AK4395, AK4396, AK4397, and AK4399 all have 100R in series in the I2S connections, while interestingly enough, the evaluation board for the AK4358 has none. In this thread "http://www.head-fi.org/forums/f46/easy-i2s-juli-pci-sound-card-242285/", 30R series resistors are recommended in a similar situation (Juli@ digital I2S out to an external DAC).
And of course, the Buffalo32 card has 22R's in it's I2S lines.
Right now I'm using 50R Vishay/Dale SMD resistors in the I2S lines. They were significantly better than no resistor and 50R Vishay/Dale regular leaded resistors. Next I'm planning to try 30R & 20R to see if there's a sweet spot.
To help minimize interface effects, I'm mounting both the AKM-based cards and the Buffalo32 right on the Juli@ to keep the I2S lines short... under 1" long in all cases.
I've spent a good bit of effort getting the AKM-based cards to sound like I think they should. Sound was muffled and phasey at first... that improved with different I2S wiring and adding series resistors to the I2S connections. None of the AKM card's I've gotten had these resistors... the AK4397/AK4399 cards were used in applications where the resistors were in the source circuit, the AK4395/AK4396 card just hasn't been used yet, so it wasn't known if it needs them.
I'm currently listening to an AK4399 card and it is equal or better than my modified Juli@ analog card in all parameters, but especially has pulled ahead in detailing, most surprisingly in the bass. But this and the AK4397 cards have some tricks on them that I can't share and I won't be able to post info on them.
So I'm looking at the AK4395/AK4396 card as the next one to try... It's from another audio hobbyist from DIY-Audio, and its all public domain. I'm hoping I can get sound quality from it nearly as good as what I'm getting now with the AK4399... I've seen several postings preferring these DACs to the AK4397 & AK4399. If so, it would be a good add-on DAC alternative for the audio hobbyist looking to setup a high-end cMP^2 at a significant improvement to the stock Juli@ and many external DACs and a significantly lower cost than the Buffalo32... perhaps $100-$300 total depending on the sophistication and complexity of the power supplies and output stages.
Then there's the redesign of the Buffalo32 that should be out in January. From the comments I've read, the Sabre32 DAC will likely better the AKM DACs and the redesigned version should better the current Buffalo32 AND be more 'tweakable'... but at a price. Twisted Pear has not released detailed info on the redesign yet, but I know that the current Buffalo32 that I'm putting together has $800-$1000 in the card, regulators, transformers, and all the supporting parts. Still, for SOTA sound, that is pretty reasonable.
At least by the end of the year, I'll have some pix of the Buffalo32/Juli@ and AK4396 implementations I can post, along with some details on how to duplicate them.
I'd be very interested to see pix of what yours looks like.
Later!
Greg in Mississippi
Hi Greg I'm nowhere close to the type of analyses you apparently have done. I used this DIY linkhttp://www.head-fi.org/forums/f46/easy-i2s-juli-pci-sound-card-242285/index4.html
(This is the same link you referenced)
as the basis for what I have done. I used garden variety 2 watt 22 ohm in series resistors mounted right on juli@ I2S pin outs. Ultimately the values were recommended by audioeng and were just a 'try it and see' experiment for me. You indicate 30 ohms was recommended by Steve but I also remember the mention of 22 ohm resistors (in an email from the OP)which is what I used. Mounting the resistors on the pins made soldering the 75 ohm coax cables (very tiny and delicate) on the ends of the resistors very easy. I don't know if 22 ohms is optimum or not but they work very well for me. Remembering Steve's comments the total of the damping resistors plus the output impedance of juli@ should total 75 ohms (to match the coax I used). So that's 22ohms plus 50 ohms for juli@ or 72 ohms ...pretty close to 75 ohms.
The 75 ohm coax was another idea from audioeng, the cable is expensive ($50 for 3 meters) uses teflon dielectric and is 1.3 meters long in my setup. If you read the link you will find the OP used 50 ohm coax of several more meters long.
I am certainly interested in your efforts to find the optimum resistor value.
What I can tell you is that the damping resistors made the static-ee/metallic noise virtually go away (in combo with cPlay 2.0 33, so don't know which did it). Also sonically my system became more dynamic with more precise imaging/separation of instruments on large classical pieces. Its almost as if a false high frequency ambience was eliminated by the resistors.
Edits: 12/13/09 12/13/09 12/13/09 12/13/09 12/14/09
Theob,
Thanks for the response.
Still would like to see pictures, if you have any. I'll be posting what mine implementations look like as I get further down the road.
For my AKM-based DAC cards, so far the SMD resistors are far superior to the leaded ones. BUT, with a 1" or so connection, the body of the resistor becomes a sizable fraction (30%-40%) of the signal path length and I suspect their inductance becomes a limiting factor. In your setup, with much longer connections, this is likely a moot point. It sure is challenging to install these tiny R's... I make up small adapters out of prototyping board that mount on the plastic housing of the Juli@ signal pin block... Not as pretty as I'd like, but it works.
I was surprised at first that you needed the 22R resistors with the ones already in place on the Buffalo DAC board, but didn't then remember that you had a 1.3m I2S line. The one I'm doing, with the Buffalo mounted on the Juli@, may not need extra Rs... And my experiences hooking up DAC cards so far suggests that it will sound better with the shorter connection too.
But it also suggests that an optimal resistor size will likely be dependant on the connection... What works best for my 1" lines will likely not be the optimal for your 1.3m lines
Also based on the comments in the Head-Fi thread we both referenced, I plan to try twisted pairs for the I2S connections at some point. The only caution I have is that there's no good place to terminate the ground side that's close to where I connect the I2S lines... The nearby screw hole is the closest, but that gives me ground lines 1 1/2 times the length of my signal lines. Still, it might make an improvement... So I'll try it at some point down the road.
I got the AK4396 board from a guy in Germany named Klaus... The board design and some info on the implementation is discussed here:
'http://www.diyhifi.org/forums/viewtopic.php?f=2&t=1399'.
I'd also done some research into an even lower-cost solution for a cMP^2 with a Juli@... Using the Juli@'s analog card, but bypassing all the output circuitry on the card (which includes the chips that do the mixing of the various inputs... They CAN'T sound good!) and taking the outputs from the AK4358 DAC directly downstream, possibly through a simple, high-quality output stage if that's needed. I'm looking at direct via coupling capacitors, a couple of solid-state options (key among them the OPA627), the tubed output stage used by Lukas Fikas on his Behringer SRC2496 mod here:
'http://lampizator.eu/LAMPIZATOR/TRANSPORT/behringer/Behringer.html'
and then the tubed BDT Preamp by John Swenson here:
'http://www.audioasylum.com/cgi/t.mpl?forum=tubediy&m=141601&highlight=bdt+preamp'
and here:
'http://www.audioasylum.com/cgi/t.mpl?forum=tubediy&m=141747&highlight=bdt+preamp'.
Of course, the Juli@ direct-out setup would work best with separate supplies powering the Juli@, but it COULD be implemented without these as a truly budget option or a first step.
I just gotta find time to get it all done! This stuff takes a lot more time than the original hybrid-linear supplies I built up!
Later!
Greg in Mississippi
There is nothing magic about my implementation plus I already admitted many times here and in pms my skills as a diyer are barely average at best and I don't want to validate that by posting pics. Look you, Bertel, Alfred, and Hans are the thinkers here I am just an audiophile who is committed to pc audio trying to make do. What makes it feasible to do I2S lines 1.3 meters long, I believe, is shielded coax cable. You may want to ask Wackyterbacky (the OP on DIYaudio) to explain. The coax plus 22 ohm damping resistors are his idea. Wrt the resistors on the input side of the Buf32s yes they are still there on my setup. Steve mentioned that these should be eliminated but I never tried...maybe someday though.
Whatever I have done is working satisfactorily for me. If I can do other stuff to improve I will.
Now some questions for you if I may: are you saying you are creating several new sound cards based on AK DACS that you may sell? I would be interested in that. However my ideal for a sound card would be: easy implementation of Batteries for 3.3 volts and 5 volts; easy I2S output and no problems with metallic/static-ee noise. If you can do that count me as interested.
You really don't want to see my pics.
Theob,
Ok, I'll give up on the pix. But I'll still post mine by-and-by.
As for the thinkers here, I have to add Rylands... And of course, cics. But you knew that!
As for cards, I'm working on I2S-connected DAC cards to fit on the Juli@... Basically doing the same thing as the Buffalo. I'm working a three-tier approach right now:
Highest Level - Twisted Pear Buffalo32. My goal here is to show how to integrate a Buffalo32 onto the Juli@ in a manner that maximizes its strengths while providing great value. I have high expectations for their new version of this card and for my short-connection implementation, but they will come at a pretty high price, around $1000 for the cards and parts for a hobbyist to then implement.
High Level - AKM-based card, probably using the AK4396 DAC (AK4395 as an alternative). I expect this will provide a high fraction of the sound quality of the Buffalo, but a hobbyist should be able to do it for much less, around $300 for the parts.
Very Good Level - A modification of the Juli@ analog card for 2-channel analog-out-only setup that bypasses a lot of the extra circuitry used for mixing & ADC. A budget, no-external-power version of this may be as low as $20-$40 in parts, but would work a LOT better with separate supplies which would push it into the $200-$250 range.
I AM NOT working on a sound card, sorry, that's well beyond me.
Also, I AM NOT working to produce a commercial product. I'm still of the mind that the fastest way to a million dollars in an industry like this is to start with two million. I'm looking to post designs and instructions that others can duplicate. And offer assistance where appropriate. But definitely not looking to go into business on this!
Later!
Greg in Mississippi
Cool ... yes did not mean to exclude Ryelands and Cics. Good luck!
Still operating w/o the dreadful noise but with a tradeoff (there always seems to be a tradeoff). When I try to lower my clock and / or core voltage the system gets unstable...it boots but collapses in a horrible loud sound after playing awhile. As long as I can can decouple my dac it causes no harm but once I didn't and it burned up my amp (since repaired under warranty). But it looks like Ryelands was correct my problem all along has been marginal stability. Only argument counter is that I have tried 2 mobo's, 2 cpu's with same results.What really improved my static-ee noise I believe was loading my I2S lines with 22 ohm resistors. System got way more stable after this (wrt to the metallics or static-ee noise that is).
Anyway I have a good work around and as long as I run 147 hcc and .875 volts I'm ok.
And I should repeat cPlay 2.0b33 also is the least susceptible to static-ee/metallic noise since
cPlay 2.0b18.What this all really means I'll leave to the computer experts but it at least offers my experience to others who may have this issue.
Edits: 12/10/09
Again, another jump toward perfection. 32 was an amazing leap forward, and 33 is the polished version of 32. I am surprised at the silky rich detail throughout the spectrum and amazed at the jump in low end clarity.
Thank you for sharing this amazing gift!
Files from folder cPlay 2.0b33.pre4 are dated 2009-12-04 .
Actual release is cPlay 2.0b33 with files dated 2009-12-05 . This is the default folder containing the correct download files. If you downloaded files prior to this post then its very likely you have B33.pre-release 4 (i.e. not B33 final version).
Background:
I normally validate by hand all critical areas at ASM (assembler) level for each release. ASM is what the compiler produces and is the stuff contained in .exe files. Note ASM must not be confused with machine code (micro-ops or µops). µops are CPU dependent where different models decode ASM instructions into 1 or more µops (advanced CPU's like Dual Cores also does macro fusion, i.e. combines µops).
A silly oversight on my part overlooked a critical area in the DSP (because it's source code was not changed). That was a bad assumption! Given that a new compiler option is used (to support VST) this caused a change in ASM output. This part of the DSP needs to be very efficient. B33.pre4 has unnecessary overheads here to the tune of some 110 ASM instructions (B33 final has only 8 after doing a small work-around)!
I left "B33.pre4" at SourceForge for those interested in experiencing how unwanted overheads in critical areas can affect sound. Difference in SQ is significant: whilst both offer improved mids, B33 final is much better.
Nice release (cplay 33 12/5). I sense it is more transparent all up/down the frequency spectrum. The hall ambiance is very good without being hard. In fact highs are sweet in the true sense. There is more music in mids and upper bass, again along with that air/ambience.
Thank you Cics.
I am with Playmate!
What a great present! I just bought the allocator package and was going to get going with Jriver and cmp, but now I dont have to!!!
Thanks so much!
Next year my ass :)
Cics you rock!
Hello, Cics and everybody. This release is long-awaited for me, as I didn't much liked 32 - too soft and smooth. Now, this version really shines. Resolution and natural brightness of highs is back. Mids are clean and fleshy, but (in my setup) what really exceeds any expectations is lower midrange to upper bass region. Bass lines suddenly got longer, much more informative and solid. VERY GOOD RELEASE!
Thank You, Cics.
This is another happy weekend.
Serge.
what a christmas present !
installed this newest VST-able version, and can´t wait to get it configured.
need to sleep a little now, but it´s gonna really like waking up to new toys on christmas tomorrow.
for all of us (or, just the few of us...:-) who run VST allocator, this is just groundbreaking steps forward !
thanks sooooo very much !
now the door for cMP2 is open for allocator users !
may I bow for the deepest respect and jump of cheerful joy !!!
thanks cics !
I looked thru the various posts but did not find an easy answer, is there one? My question is: what are good/acceptable jitter values?
I use the dpclat program and measure values like 17 but I do not know how to understand them. I saw some technical explanation but too technical for me!
Anyone who can explain this in simple language?
See thread link below for more details.
I have a problem with cuesheets which has me completely baffled. In my cue sheet I might have track title text that looks like this:-
TITLE "Trio élégiaque No.1 in G minor"
When cPlay reads the cuesheet, what I get on the screen is this:-
Trio élégiaque No.1 in G minor (02)
I have the problem with several different cue sheets and I cannot seem to find ant reason. The coding is ANSI in all cases.
Any clues would be gladly received.
If you mean the (2) at the end, that is added by cPlay to reflect that it has to load the file into memory in two separate segments because all at once it is too large to load. It plays one then loads the other, but you don't hear the transition.
Ah, now it becomes clear. I thought that the (02) was a strange piece of redundant text.
Many thanks for the information. I keep learning!
From what I understand cMP uses AutoHotkey to setup/create its environment.
I cannot find any info on the commands they use and how the screens of cMP are manipulated.
I want to add some functions to cMP using AutoHotkey (1 is switching off the monitor with remote). For that I need to know how cMP uses AUtoHotkey, which commands it uses etcetera.
Where can I find that info?
See post in cMP thread.
Thanks for the answer. This is the AutoHotkey script for the remote, I was under the impression that cMP at startup runs a script as well to change/hide backgrounds and screens? Maybe I am wrong.
Anyway, I am trying to switch off the screen using something like this:
^1::
KeyWait, CTRL
KeyWait, 1
state := state ? : 1 ;toggle state
Gui,Color,0x000000
Gui,-Caption
WinMinimizeAll
Gui,Show, NA w%A_ScreenWidth% h%A_ScreenHeight%
If state
SendMessage, 0x112, 0xF170, 2,, Program Manager
Return
It works like a charm outside cMP, it switches off the monitor with ^1 and with ^1 starts again. I've also tried this with WinMinimizeAll, same thing.
The moment I start cMP and play a song thru Foobar the command ^1 does nothing. If I use a "set focus script, (setting focus to Foobar)it works but ignores the standard background of Windows which is aparently still hyding under the cMP menu?
I just don't get it to work and my latest idea is that if cMP uses similar AutoHotkey commands to setup its environment they might disturb eacht other?
I am fairly new to AutoHotkey so maybe I am just doing something wrong?
If you want to keep the cicsRemote functions, perhaps you could use cics' script, then add your own extra hotkeys, and name it cicsRemote.ahk, compile it with the autohotkey compiler, copy it into the cics Memory Player directory, replacing the original cicsRemote.exe. I made my own crude scripts to do what I wanted (not turning the monitor off though) and did this and it works fine.
see link below
It is working now thanks to all the help I got.
Let me explain.
I use cMP together with Foobar, I wanted 2 things: Foobar fullscreen (without the menu with the red X on top) and remotely switch the monitor off and on while playing music. All this should be done while using as little resources as possible. I run Foobar without any special features or fancy menus as it needs to be as low profile as possible.
The original cicsRemote had functions I did not need, I use Imon as remotecontrol and I can program it perfectly for all the functions I need.
My code consists of 2 files using the program AutoHotkey. I called the first file "cicsRemote.exe", I copied it to the cMP directory (replacing the original one) as cMP loads that file automatically. I did not need the original remote functions. The second file is called "nogeens.ahk" and is copied to the same directory, it is activated by the first file so no intervention needed.
When cMP starts you now have the added option to switch the screen off and on with "CTRL1", the monitor goes in standby.
The same function also works in Foobar.
Also added is that Foobar runs completely fullscreen, no nagging menubars.
Here are the scripts I used:
1) 1st program "cicsRemote.ahk" (needs to be compiled to ".exe":
#Include C:\Program Files\cics Memory Player\nogeens.ahk
^1::
KeyWait, CTRL
KeyWait, 1
Gui,Color,0x000000
Gui,-Caption
WinMinimizeAll Gui,Show, NA w%A_ScreenWidth% h%A_ScreenHeight%
Loop
{
If (GetKeyState("CTRL","P") and GetKeyState("1","P"))
Break
else if (A_TimeIdlePhysical <1000)
SendMessage, 0x112, 0xF170, 2,, A
}
Gui,Destroy
Return
2) 2nd program "nogeens.ahk":
SetTitleMatchMode, 2
; -------- select the desired program to operate on by the following statement
program = foobar2000
; -------- the last assignment statement is the 'valid' one
x=1
y=40
w=200
h=500
foobar_max:
WinWait, %program%
;=======================================================
; maximize the application
;=======================================================
;#q:: ; win-q hotkey, change as desired
winactivate
WinSet, Style, -0xC00000
WinSet, Style, -0x200000
WinSet, Region, 4-4 W1680 H1050
winactivate ; changes made, make it visible
winmaximize ; changes made, make it visible
WinWaitclose, %program%
goto foobar_max
-----------------------------
How to make the files yourself:
copy/paste the text of the cicsRemote.ahk and nogeens.ahk file seperately to notepad and save both files as
"cicsRemote.ahk" and "nogeens.ahk".
Download the program AutoHotkey and install. Compile the "cicsRemote.ahk" to "cicsRemote.exe". Copy both files into the cMP directory (replacing the original cicsRemote file). ALL DONE!!
Good luck!
AF,
Glad you got it working. But it seems like a lot of trouble to waste on Foobar. Why not use cplay?
And what kind of latencies are you getting in dpc latency checker?
When I was using remotes with my rig like the IR keyspan one or a WMCE remote the added latency was crazy. Removing these remotes in favor of a wireless usb mouse helped a ton. I hope you are having a different experience then I did.
That is a great feature to be able to turn off the monitor from the chair. I might even have tried to implement it but a future room change puts the monitor right next to me so I think I can just hit the button. But one never knows, so thanks for figuring that out.
@Dawnrazor:
Hi,
Thanks for the reply! I will let you know about current latency asap, might not be able to check it this week due to work. Latency used to be on par with cplay (11-min, 42-max), what is your score?
I use Foobar as I can tweak it. My menu for example shows info like lossless or lossy, the codec used, decoding time, time elapsed per track/total.
CPLAY has no configuration options at all as far as I know or am I wrong?
Besides that I want to be able to use kernelstreaming as well and cplay only uses asio?
AF,
Thanks for the info.
I have to check but i remember latency being min of 1 or 2 (mostly 2 during the whole test) and a max of 26 or 36 I think. Try removing your remote and running the test with a normal ps2 mouse or usb.
I am a bit confused. If you are using CMP with Foobar, then arent all the menus on foobar really superfluous? But i suppose KS makes sense if you dont have hardware that supports ASIO. But it would be interesting to see how Cplay compare using asio4all which is kind of like KS anyhow.
Thank you for making this cool utility. But I hit some difficulties when I tried to use cplay on an AMD system. My secondary computer is an AMD Phenom II X4 which supports SSE3, SSE4A. But any version higher than SSE2 will not start on my AMD system, and even though SSE2 version is running, it comes with cracking noise. My primary computer uses Core 2 Q6600 and it works fine with SSSE3 version. Both computer runs on Windows 7 X64.
I searched your older posts and found that you are using Intel Compiler, which could be the reason for the problem. Is there any chance for you to give GCC compiler a try, please?
Thank you very much.
Yes, Intel compiler is used (no plans for GCC). The fact that SSE2 operates and others not suggests an AMD processor error. It may also be that AMD reports instruction sets differently which causes cPlay to exit.
Phenom's TLB bug is an issue and requires a BIOS workaround. This could explain your playback issues.
Same issue here though with an AMD X2 Brisbane that reports as SSE3 compatible. I can only run the sse2 version.
Just reporting, not complaining.
Thank you for the explanation.
My CPU is Phenom II and the chipset is 785G, both roll out after 2009 Q3, I don't think they still have a bug discovered in 2007, though.
Anyway, the AMD system is located in my workplace with a certain level of background noise, using cplay on it is a waste, but I just got addicted to listen everything using cplay XD
Refer to ReadMe.txt for important details & specifications.
Change Log (2.0b32 release):
- Low level improvements in DSP and ASIO
cPlay documentation can be found at the new cMP² website .
Please REMOVE previous versions before installing cPlay 2.0b32. Normal (SSE2), SSE3 B9, SSSE3 B9 and SSE4 B9 versions are available. Running cPlay on a CPU not supporting the required instruction set will cause cPlay to exit immediately. For best results, uncheck "Timer" setting and test different "Buffer" sizes.
I started at b31. Now b32 seems to have fixed my only real gripe. I loved b31.
However, Foobar seemengly had it beat in depth of stage and "air". B1 brought improved clarity and detail. The only problem, it compressed the imaging, mainly front to back. It also constrained the width somewhat. With my Magnepan MMGs, depth & overall imaging is terrific via either Foobar (E-mu 0404 USB) or my Denon 3910 player (the Denon is better).
This seems to have been "fixed" in B2. The only reason I am asking is that the configuration of my system has temporarily changed from the time I mostly listened to b31. I still had b31 on briefly after changing the system and did not notice this:
-- when I switched to B2, the immediate impression was that nothing good was lost
-- but now the stage got as good as with Foobar
-- PLUS the definition and placement of instruments became way more solid.
If no one has heard this type of improvement from b31 to b32 I'll have to suspect that the mushrooms I had with dinner yesterday were "loaded".
hey J,
You are right. There is a big difference and I myself heard similar things. Such differences undoubtedly are system dependent.
Did you enable AWE or are you using cplay "straight out of the box?"
Now this is what I am talking about!
Great job. Very musical yet detailed.
Its weird because the image is bigger and with 31 it was smaller and more precise. That precision though seemed to make things less warm. The bigger image on 32 seems to make it warmer.
Is that the trade off? Laser precision vs. musicality??
No tradeoff involved, just a reduction in playback jitter ☺
ASIO change was significant and together with lowest latency (Lynx does 32 samples), results will be impressive (esp mids).
Y.! I have been listening more and more in triode, analog out> Benchmark Dac1> digital out> Lynx L22. I have been able to achieve a latency of 33ms at 96.000 khz with a buffer of 32 using the Lynx L22. Speakers are Harbeth Anniversary SHL5. The lushness of the timbres are really something. Thanks again T.
No worries!
hey cics,
now that my cPlay> Jack> allocator VST> lynxTWO B setup crashed again (due to a new lynx driver installation) my hope for an alternative to the direct ASIO output of cPlay, enters my mind again.
the impossibility of using the JACK within cMP2 mode is still a bummer, so i wonder what could get you to look at this ?
winamp, foobar and (I think) J.River too have the ability to output to a VST client.....
kindest regards
.
I agree. Adding such support would give cplay a whole new world of usefulness and capability.
Thanks in advance.
.
Hello Y.N. I am listening to one of those B&W audio CD's> lossless> ultralinear. I have been burning in B32 for 48 hours. The fret work is palpable the decay is excellent. Thank you again, T.
No worries!
Hi, Please, tell me, can you planning to add support for Monkey Audio files?
.
Another improvement: a bit warmer and more real-sounding and engaging. Ongoing appreciation and thanks to you, cics.
Thank you very much cics! It's always exciting to test new version... As this is now late night in Hong Kong but I just can't wait, so I use AOBCT PC -> AES16e -> Mytek -> Beyer DT990 headphone to listen to wave files. Again SSSE4, upsample 192 SOX Linear 95.
The musicality is back, just after a few songs I can't stop... This is a good version! I will listen further and provide you an update tomorrow.
On the other hand, just an opinion that for every time you release a new version, if you can provide some further details on exactly what you have changed in coding, then we can provide further opinion on what "parameters" we like so for your reference. But anyway as you know I am always a supporter of cplay. Cheers!
Another thing that would like to bring up is that due to the "coarse" of high frequency in B31, I was forced to look for solution and came across the method below... DIY shielding of the sound card using aluminum foil... sounds crazy but it works! It is cheap and effective on high frequency, worth a try:
http://www.overclock.net/sound-cards-computer-audio/571718-how-make-emi-shielding-your-sound.html
To further reduce high frequency distortion, you may also try applying binder clips on edges of your sound card. I still have a fan in my computer and another one in the power supply, but I think anyone with a cmp would still notice an improvement even if you don't have any fans. The force of the compressed spring will dampen mechanical vibrations that although not easily detected, will transmit into and interfere with your electrical signal. Large ones and tiny ones all fit in different places and I have them on everything.. from equipment shelves and interconnects to RAM and sata cables. It's not a pretty tweak, but it is cheap...the more you use the more they show their effectiveness.
Thank you very much texastea006! I know we are a brunch of weird audiophiles... When I was holding my al-foil-wrapped Lynx AES16e walking around in the house, my sister saw it and just couldn't stop laughing... you can imagine how silly it looks.
Anyway, I understand what you mean and I may use cable tie/binder clip to tighten up the cards and cables inside. THanks again for your advise!
http://en.wikipedia.org/wiki/Binder_clip
Two key areas were affected:
- TLBs or Translation Lookaside Buffers when using SRC
- ASIO - this area has had biggest improvement
Overall, B32 is a refinement on the extensive work done in B31. I wasn't planning to release B32 so soon but SQ improvement is significant and felt it warranted a release.
From now I have to be more careful before making comment on any new version, as cplay's resolution is so high that any flaws or noises in my PC will be revealed by cplay mercilessly. Recall that I thought cplay B31 was coarse at high frequency. However, after inspecting my own PC, wrapping up cables and sound card using aluminum foil, the high frequency resumes normal, strings becomes crispy clear without grains.
Thanks again cics :)
THank you cics. Would you suggest using 64bit Win7 instead of XP? As someone claims 64bit Win 7 uses floating point Sound system, but my understanding is that ASIO bypasses everything of Windows Audio. So does it matter?
I would avoid 64 bit OS as it adds more hardware complexity (64 address lines vs 32+4). Main reason for switching to 64 bit is using more than 4GB RAM. This is useful when using large capacity database servers.
.
SQ improvement is indeed significant. The top end haze is again reduced, or in the case of b32, removed, resulting in:
-- better high frequencies
-- more solid mid frequencies
-- more precise imaging, and
-- wider and more 'substantial' sound stage
Thanks again cics. The cMP/cPlay combination means a lot to my listening life.
I didn’t gave my feedback on version B31 because I couldn’t make my mind up on B31. It had more accuray, details and more soundstage than B30, but somehow the scanpeak tweeters in my K&H O300 didn’t like the high’s from B31. B31 as a whole sounded “tensed & stressed”. Version B31 in my setup would be best described as: B30 ‘on steroids’ Or ‘on speed’. Good and fast but somehow not very pleasant to listen to.
But since there were also many very good reviews of version B31, saying the high’s where very smooth and a pleasure to listen too, version B31 left me puzzled about my gear, which I have carefully chosen for their reputation being very neutral and uncoloured gear. I choose to go back too version B30 and occasionally even used the players included in Wavelab lite and in Audition 3 again.
But my worry’s are over now. The unpleasantries of B31 are gone! But all the quality’s of the B31 are still there (more detail, more open, more stage) and are now a real pleasure to listen too. High’s are even more detailed and open than in B31. But probably the most progress has been made in the mid section. B32 has very very nice sounding and very accurate mids. Hmmmm ….. I really, really love the mids in version B32 !!
The D-tour via version B31 was probably a necessity to reach version B32. I’m very pleased with B32. B32 sounds very, very good in my setup. I think it will be tough job to better B32 again.
Thank you cics for B32 !
fully AOB optimized cMP2 PC -> Lynx AES16 -> XLR AES/EBU -> Lavry Black DA10 -> XLR Mogami Gold -> Klein & Hummel O300
Good afternoon,
and many compliments for the development of Cplay.
I have a small issue, related to my HTPC soundcard and the proprietory ASIO drivers that come with it.
In Cplay, the proprietory ASIO 2.0 drivers are visible, but I cannot use them to play music, as an error message comes out telling me that they are 24bit ASIO drivers, while Cplay only accepts 32bit ASIO.
As a solution, I'm using ASIO4ALL, but I'd like to know if this situation is widespread or specific to my setup, and if there's any chance of an alternative solution allowing me to use the original ASIO drivers of my (quite nice) soundcard.
Thank you for the patience...
Ciao!
tranfa
I had exactly the same problem. I've been using winamp and foobar with Asio.
Yes, that's exactely what I'm doing as well...but I tend to like CPlay better for FLAC reproduction....
Should it have a Convolution utility inbuilt, in order to apply DRC filters, it would be the perfect application for me.
Hi carcass93,
thank you for your frank reply.
Tranfa
It is not unique to your setup.Aside from using ASIO4ALL, which is a compromise sound quality-wise, you have 2 options:
1. Give up on ASUS, and replace it with another soundcard;
2. Give up on cPlay, and replace it with another media player.As I understand, cics (developer of cPlay) is not interested in making 24-bit version. By the same token, I'm fairly sure ASUS is not going to create 32-bit version of their driver.
Edits: 11/04/09
If I build the cMP to spec. What complications am I adding to use a Blu Ray capable drive and video card? Does that completely defeat the purpose?
Whatever sounds good...do that.
Probably. Minimizing video is a decent part of this project and you would be going the other way.
Though there ARE some tips for home theater in the build and you can take what parts you want. There are some inmates who have used the principles and made a HT rig and are pretty happy once they got it figured out.
...No, the real killer feature of Windows 7 is scalability. Simply put, Windows 7 does a better job of taking advantage of the available hardware resources than its predecessors. This scalability edge manifests itself in the form of better performance under complex, multiprocess, multithreaded workloads.
Given the same number of CPU cores, Windows 7 runs circles around both Windows Vista and Windows XP. In fact, the results aren't even close: In one multiprocess workflow test, Windows 7 outpaced Windows XP by 250 percent -- this on an eight-core (dual quad-core Xeon) HP Z800 workstation.
This is Windows 7's killer feature. It means that, as customers invest in new PC hardware, they'll be better positioned to reap the improvements in CPU, memory, and chip set performance by deploying Windows 7. It also means that sticking with Windows XP -- ostensibly because it is less bloated and performs better -- is a fool's errand...
Ok to any cplay/cmp uisers out there on Windows 7: do you notice any cpu efficiency running music files? More specifically does cpu % usage drop with Windows 7 versus XP?
Hi all,Summer in The Netherlands is at it's end, so it's time too pick up my indoor hobby again. Which means its time for: Power Supply improvements: part 2.
As already stated by other inmates (Gstew, Reylands, Bertel, Theob, carcass93, Rayban, Sondale, fmak and many others) improving the power supply is a very worthwhile and very rewarding exercise.
Optimizations of the Operation System, Kernel and the BIOS all together do have a nice effect on sound quality, but I was really amazed what BIG IMPACT (!) a linear PSU on the P4-pin had on sound quality in my setup.Note: my setup is a fully optimized PC as per Cics hard- and software recommendations. So my findings are based on a setup like this: see: http://www.cicsmemoryplayer.com. With only 2 exceptions: I do not under-volt the FSB nor RAM . I over-volt both with + 0,2 V.
Improving the power supply to the P4 pin with a linear PSU turned out to be surprisingly easy (and also relatively cheap), Click on picture to see photo's in Picasa webalbum:
cMP2 project: Optimizing power with linear PSU to P4 power connector'
So I now would like to try optimizing the power to the P24 connector.Just before summer (in june) more knowledgeable inmates who already tried to improve the power on the P24 (Bertel, Gstew and others) gave me a lot of suggestions and possible directions.
-> Bertel pointed out the importance of choosing good quality low-ESR smoothing caps but also and the use of low resistance wires (!).
"to keep power supply as clean and quick as possible to keep the rise time between 0 and 1 ("no power" and "power", changing ever so quickly) at a minimum! IMHO this can best be done with caps sized the "right" way, plus cables from the caps to the board which are short, have a lot of internal conductors (that brings speed, e.g. I use AWG 6 which is 133*AWG27) and a very low internal resistance ( <.5Ohm/1000ft, just to be safe - that's in the region of an average PCB's traces)".-> Gstew suggested use a tweaked Pico PSU (extra smoothing caps) on the P24 pin.
So that's what I did (wires and picoPSU) last week, with unexpected and very surprising results !!
* using LOW RESISTANCE internal power wires to P4 and P24.
Although Bertel already had pointed out to me the importance of using low resistance wires, I still wired everything together with standard computer wires (18AWG wire). Why? because its easy and I'm lazy. Standard computer 12 v power wires are constructed of 18 AWG and can simply be bought everywhere and they have already all the right plugs and connectors attached and in place. Regular wire to distribute power inside pc's is: 18AWG = 6,5 ohm/1000ft. Bertel suggested to use 6AWG which is: 0,4 ohm/1000ft. http://www.interfacebus.com/Copper_Wire_AWG_SIze.htmlUntill last Sunday. A suddenly cancelled social appointment left me with nothing much to do on a rainy Sunday afternoon. With electronic stores being closed on Sunday in The Netherlands, I had no other wire at hand than the type of wire that is used to wire the electricity in housing (with solid 2 mm copper conductor).
I braided 2 solid copper wires together until I had 1 meter of braided solid copper cable. With still 1 cm standard 18AWG computer wire left at the P4 plug I soldered the braided solid core copper wire to the P4 plug (a P4 plug has 2 black 18AWG wires + 2 yellow 18AWG wires)
I saw no other way than too solder the 2 mm solid copper wire to the P4 plug than trough leaving 1 cm of the original 18AWG standard wire attached to the P4 plug. The best solution would be to solder the low resistance wire directly to the metal-connectors inside the plastic plug. But I don't manage / don't know how, to get the metal-connectors out of the plastic plug.Click on picture too see photo's of the solderingproces of low resistance multi-stranded wire to the P4 plug in the picasa webalbum.
cMP2 project: low resistance wire between linear PSU's and P4 / P24 power connector
I also braided 1 meter of 2 times 2 mm solid copper wire to also connect the picoPSU with this braided wire to the linear PSU that is feeding the picoPSU.Working with this braided 1 meter 2 x 2 mm solid coper wire is a hassle. The braided wire is not flexible and one has to bend the braided solid copper wire in place all the way from the linear PSU, through the pc case, to the P4 and P24 connectors. Of course it's much easier to use multi-stranded wire (6AWG or similar).
I didn't expect much of this construction because there still was a bottle neck of 1 cm 18 AWG standard computer wire attached to the P4 plug. So I expected the whole wrestle with the stiff braided cable would probably not be worth the trouble.
But WHOW! I couldn't be more wrong! It made big and massive impact!!
Bass, mids, treble everything changed dramatically. (I'm not writing this for effect purposes) Especially drum kicks, plugged strings, or any type of 'explosive sounds', where very much more heard and present. A drum kick, now sounds like a real drum kick !!!
The changes where that big, that I first thought, that there was something wrong. That I only added some kind of reverb like distortion. But after playing more and more music and tracks I started to realize that this was no distortion at all. This is probably how the music was recorded and mastered.
* Optimizing power to the 12Volt rail of the P24 with linear PSU + picoPSU.
- Chosing the right picoPSU model.
It turned out that not al picoPSU are made by the same concept. Some picoPSU models leave the 12 V line untouched and pas it straight on too the MoBo, where other picoPSU models don't leave the 12 V line untouched and regulate the 12 V line.So shit happens and I first bought the wrong picoPSU model. A PW-200-M w/P4-ATX. This model regulates the 12 V line which results in almost no sound quality improvement. So it's not worth the trouble to change a 'smoothing capped' ANTEC ATX PSU for an PW-200-M w/P4-ATX.
By now I have ordered an other picoPSU model and installed it. The model picoPSU-150-XT leaves the 12 v line untouched and passes the 12 V line straight on too the MoBo. Now there is a very nice improvement in sound quality !
However a very nice improvement, it's not as much as with a linear PSU on the P4 pin.
In comparison: a picoPSU-150-XT on the P24 gives about 1/3 of the improvement of a linear PSU on the P4 pin. (So improving the 5 Volt and the 3,3 volt power also needs to be done. Especialy the 5 v line. But thats for PSU improvements: part 3, because I haven't made up my mind yet on how to do it the easy way). But although only 1/3, it's still a worthwhile and recommended improvement !So chosing the right picoPSU model (which leaves the 12v line untouched) is important! Installing a picoPSU model that regulates the 12V line, it's not worth the trouble IMHO.
cMP2 project: Optimizing power with linear PSU + picoPSU-150-XT to 12V rail on P24 power connector'
* First question on internal power wires.
Are there any other inmates who swapped out the standard PC power wire type (18 AWG) and replaced these internal pc-power-wires for low resistance wires like Bertel suggests?
(AWG 6 wire or lower resistance)
Or, are there any inmates willing to experiment with low resistance type of wire between their linear PSU and the P4 pin and/or their picoPSU ? And report back there findings ?* Second question on internal power wires.
If this would be analog audio, I could easily understand what is happening and understand the effect. But what is happening here ? Is this really the effect Bertel pointed at? "keep the rise time between 0 and 1 ("no power" and "power", changing ever so quickly) at a minimum!"Any comments, suggestions, feedback is very much appreciated. Or even better: please try this at home and report findings back here at the forum !
Power supply improvements part 3, will on the P24 5 volt and 3,5 volt lines but I haven't made up my mind yet on how do to this the easy way. Probably there is no easy way.
Mark
Note:
I don't encourage anyone too use solid core 2 mm copper wires. It's a hassle. I just had nothing else lying around at that time. But I do encourage to try multi-stranded 6AWG wire (or any other low resistance wire) between the linear PSU and the P4 / P24 connecter and report back the findings to this forum.
fully optimized cMP2 PC -> ESI Juli@ -> Van den Hul Optocoupler MkII-> Lavry Black DA10 -> XLR Mogami Gold -> Klein & Hummel O300
Edits: 11/02/09 11/02/09 11/02/09 11/03/09 11/03/09 11/08/09 11/09/09 11/11/09 11/11/09 11/11/09 11/11/09 11/11/09 11/11/09 11/15/09 11/23/09 01/18/11
I think I'm almost ready to take this same journey. All this information is very exciting to me. Do you have any pictures you can share?
Also, I'm interested to know, do you have just one linear PSU powering the P4 and the pico or are you using two separate linear PSUs? Also, which brand/model linear PSU are you using? Or did you build it yourself?
Thanks for sharing!
Hi Edward,
Since you asked I made some picture to share via Google Picasa webalbums this week, for your and other inmates convenience .
See added links to Picasa web photo albums in original post.
fully AOB optimized cMP2 PC -> Lynx AES16 -> XLR AES/EBU -> Lavry Black DA10 -> XLR Mogami Gold -> Klein & Hummel O300
Thanks for the pix! So then I take it you are going from the PSU straight to the P4 & P24? And not using any "smoothing" capacitors in between? Have you found that with the linear PSU, the caps are not needed?
Also, I'm intrigued about using lower resistance (higher gauge) wiring. I'm confused though. Correct me if I misunderstood, but did you not say you braided two 12 gauge solid core wires together? In the photo it appears to be a multi-stranded wire. I'm not an expert on these matters nor an electrician, but I am a fan of the concept of solid core cables. So I'm wondering if it would be sufficient to use 10 gauge solid core copper (and try to go directly to the P4 connector without using the 18 gauge stranded wires).
Thanks again for your contributions and information.
Hi Edward,Yes from the linear PSU the wires go straight to the P4 / straight to the picoPSU on the P24.
I’m also not an expert nor trained / skilled in electronics, so we both should be careful not to become: two blind people, who are leading each others way. :-)
But my understanding is that smoothing caps are already employed inside a linear PSU after the rectifier bridge and or Voltage regulator. I don’t know if extra caps outside the linear PSU betters things extra.I’m not trained /skilled in audio electronics and/or digital electronics. So I can’t say that I’m a fan of solid core cables or stranded wire. I just don’t know which one is better for a given situation/application. I just wanted to try Bertels suggestion of using wires with a much lower resistance than the standard used 18AWG wire. At that time (Sunday afternoon) I only had solid (1,5 mm core) copper wire lying around as used in the electricity system in walls in housing. I braided 2 strands of this solid core copper wire together, to get 1 metre cable.
Since the 1 meter cable of 2 braided solid copper wires is really stiff and a hassle to work with, this week I bought some 2,5 mm (outside diameter including isolation) multi-stranded copper wire and made again: a P4 extension cable of 4 times 2,5mm multi stranded copper wire. Soldering of these 4 multi-stranded 2,5 mm copper wires to the four 1cm 18AWG wires at the P4 plug, is what you see on the pixs.
Both wires types (solid copper and multi-stranded) sound the same as used for P4 extension cable / power cable to the picoPSU at the P24. But 2,5 mm multi-stranded copper wire is much more easy to work with.I haven’t received any feedback from other inmates yet on sound quality when replacing 18AWG wire for a wire type with a much lower resistance as Bertel suggests (6AWG or similar).
I hope other inmates do want to try this and report their findings to the forum. Because – also very much to my surprise ! - it had a very, very big impact on sound quality in my setup.As for your question: “So I'm wondering if it would be sufficient to use 10 gauge solid core copper (and try to go directly to the P4 connector without using the 18 gauge stranded wires)”, I can’t answer your question as I don’t know that.
fully AOB optimized cMP2 PC -> Lynx AES16 -> XLR AES/EBU -> Lavry Black DA10 -> XLR Mogami Gold -> Klein & Hummel O300
Edits: 11/09/09
Hi Edward,I bought a simple Voltcraft 12V/2A inear PSU (40,- euro’s) at www.conrad.nl (part number 510172). I have not enough knowledge nor skills to build my own.
Voltcraft is their home-brand and is produced in Chechoslowakia. But one can buy this type of simple linear PSU’s everywhere. For what is worth: the technical specification shows a max ripple of 2,5 mVolts. But I don’t know if there are any other distortions present on the 12Volt output.As Reylands already suggested it really very easy and simple to connect a linear PSU to the P4 socket if one uses a P4 extension cable. http://www.atxpowersupplies.com/p4-extension-cable-adapter.php . Just cut off one connector and attach that side of the extension cable too the linear PSU and put the other end into the P4 connector on the MoBo. That’s all. This will take you no longer than max 15 minutes. And there you go !
If one P4 extension cable is to short (they are about 14 inch / 36 cm which usually is to short), just use 2 or 3 and connect them together to get the desired length.
But using these handy ready made P4 extension-cables comes at a sonic price as I discovered this weekend. They are constructed with AWG 18 wire. And as inmate Bertel already suggested to me, using low resistance wire (for instance AWG 6 or so) might be far better. And Bertel was o so right !!
If one uses low resistance wire as Bertel suggests, than there’s again a big sound quality improvement.I think many inmates did the same thing like I did, and wired everything together with standard AWG18 wire. So they are not getting the maximum sound quality improvement possible. That’s why I decided to post my findings and encourage other inmates to try the same and report back there findings on sound quality.
Right now I have 2 linear PSU’s outside my pc-case to power the cMP2 PC and 1 standard ‘dirty’ ANTEC Earthwatts ATX PSU mounted inside the PC case. The ANTEC now takes care of the job that originally was indented for the Granite PSU’s.
I don’t like the Granite solution to feed the USB-port, HHD and DVD-drive. When using the Granites one has to bring 230 AC wires inside the PC-case to power these Granites. I don’t like that in terms of noise, hum, RF-polution, etc.
On the other hand I don’t worry very much about ‘horror stories’ on how ‘noisy’ it supposably is inside a PC case. A PC entirely runs on 12/5/3,3 volt DC, so I don’t worry at all about ‘RF-noise’ inside my PC-case. I just don’t like 230 AC wires and unshielded 230 AC -> 12V DC converters inside my case.Having 2 linear PSU’s was not a deliberated choice, it just went this way because I first optimized the power on the P4 pin. I found this to be so rewarding that I decided to also give the P24 a try. So I had to buy second linear PSU 12V/5Amp to feed a picoPSU on the P24 (Using a picoPSU model that leaves the 12V line untouched !)
But when using 1 linear PSU it will have to be a big one because it will have to provide 7 or 8 amp total. May be there is also some benefit from using 2 separate linear PSU’s. They cannot influence each other, ect.
If you buy a linear PSU be sure to buy one with low ripple specs and passive cooling (without a noisy cooling fan).
Furthermore I do suggest that you first carefully read the ‘Advanced’ section chapter 12.b on http://cplay.sourceforge.net/ and their subsequent links to other inmates here on the forum. Some of those inmates are far more knowledgeable than I am. I just read all their recommendations and suggestions and than try something in practice and report back my findings to the forum.
If you don’t mind an extra PSU box beside you PC-case, than a linear PSU on the P4 is a must do. When using P4 extension-cables as suggested by Reylands, you are done within 15 minutes.
It even gets much (!) better, when you don’t use the ready made P4 extension cables (with AWG 18 wire), but when you use low resistance wire (something like AWG 6 wire as Bertel suggests) .Succes and happy listening.
fully AOB optimized cMP2 PC -> Lynx AES16 -> XLR AES/EBU -> Lavry Black DA10 -> XLR Mogami Gold -> Klein & Hummel O300
Edits: 11/03/09
I have made a 12v linear psu to power up the 12v p4 line on my cMP. The unit is about 2 feet away from the load (p4 pin). I have used 10,000uF caps for the filter capacitor & 1000uF for the output capacitor after the 7812 regulator. Since the wire from the output cap to the p4 pin is roughly about 2 feet long (and twisted), do i need to place a small cap close to the load????
Vellerman's PS1503SB (for ~$90) looks very impressive with RMS ripple specifications of ≤1mV and ≤3mA!
According to manual: for stable voltage supply you need to set current output to Max (3A), turn on and adjust voltage to 12V. Anybody tried this?
PS - I've added a link to your excellent post for Hybrid P24 power supply (Advanced section).
Hi Cics,* using one linear PSU (instead of two)
When I first red your suggestion: “how about 1 linear PSU of 12/3A ? ”, I thought: well a 3Amp linear PSU is not enough. Other inmates on this forum reported that the current draw on the 5 volt P24 line alone was between 4 and 6Amps. If one would ad this up to the also reported:
- 0,5 Amps for the 3,3 Volt
- 0,5 Amps for the 12 Volt
And for the P4:
- 0,4 / 0,5 amp during boot (0,3 – 0,4 while playing music)
That would make a total of:
- P24: 6A + 0,5A + 0,5A
- P4: 0,5A
Total = 7,5 A.
Say 8 Amps to be on the save side.But it’s seems that 4-6 Amps trough the 5 Volt rail on the P24 pin (as other inmates have reported) isn’t correct.
Since I now use 2 linear PSU’s, of which one is feeding the picoPSU at the P24 pin, it’s has become very easy to measure the current flow through the 12 Volt line that is powering the picoPSU. So that’s what I did this afternoon, after reading your suggestions too use only one 3 Amp linear PSU. My measurements gave the following readings:
The current flow on the 12 Volt DC line towards the picoPSU is:
- while booting up: between 1,5 and 1,7 with peaks to 2,0 amps
- while starting XP: between: 1,75 and 1,85 with peaks to 2,0 amps
- when Xp is at rest: 1,75 Amps
- when starting a program: 1,8 – 1.85 Amps
- While playing music: 1,76 – 1,79 AmpsSo when using only one linear PSU, the unit would have to deliver:
- P24: max 2 amp
- P4: 0,5 amp
makes a total of: 2,5 Amp.
So your suggestion of a linear PSU at 12V/3 Amp would indeed be enough.
I indeed would like only one linear PSU for space and energy consumption reasons. But also because only than both the P24 and the P4 will get exactly the same voltage. And there will also be no a voltage difference between the 2 black minus poles of each PSU and the ground.
On the 12V/2Amp linear PSU the voltage between the black minus pole and the grnd always was: 0 Volt.
But on the 12V/5Amp linear PSU the voltage between the black minus pole and the grnd always was: -0,003 / -,021 volt. No matter what I did, I couldn’t bring it also back to also: 0 volt.
If this where analog audio, one wouldn’t want this difference between two PSU’s. I don’t know if this also matters when powering digital data processing, but this didn’t give me ‘ease of mind’.I now connect both lines feeding the P4 and feeding the P24 to one linear PSU as you suggested. The 12V/5Amp one. It works fine. The safety-circuit hasn’t shut it down. Not when switching on the PC (run-in currents), booting up or when playing music. On first ear, I doesn't make any differences in SQ, But I will do some carefull listening on that in near future.
* low resistance wire and linear PSU’s with better specs.
Unstill now I haven’t had any feedback from inmates who also changed from 18AWG wire to a wire with much less resistance and how this improves sound quality in their setup.
That’s a pitty, because the sound quality in my setup improved really, very, very much !!
And also the MoBo liked it very much :-) I now can cold boot without any failure at only 130 mhz FSB speed. And warm boot at even 125 mhz FSB speed !If low resistance wires have such a (real big !) impact on sound quality (in my setup) than I probably should have paid much more attention to linear PSU specs, such as very low Voltage ripple and very low Amp ripple, when I ordered one. I more or less bought the first low cost linear PSU, with reasonable low Voltage ripple, I came across.
I think the Velleman linear PSU you suggest, looks like a good buy. It’s only € 57.50 euro and has much better specs than the 12v/2A one that I bought for € 40,-. And it’s looks better too! Thankx for your suggestion. Although I won’t buy another linear PSU in the coming months. All funds will be directed to my upcoming 1 month holiday in New Zealand and Australia in February 2010.
I hope other inmates will soon report their findings on linear PSU’s with better specs and swapping their internal 18AWG power wires for low resistance ones.
fully AOB optimized cMP2 PC -> Lynx AES16 -> XLR AES/EBU -> Lavry Black DA10 -> XLR Mogami Gold -> Klein & Hummel O300
Edits: 11/08/09 11/09/09
Thanks again for the info. And it brings up another question. On your measurements of 2 amp (P24) and 0,5 amp (P4), I assume this is based on underclocking and undervolting? Does it make any difference if the CPU is run at full speed with default volt?
For example, if I get a PSU at 3A, but then experiment with lowering FSB and voltage further and then go too low to the point it will not boot - then I will have to reset the CMOS/BIOS which will boot at defaults. Will 3 amps still be enough at default speeds?
Hi Edward,Good point. your assumption is correct. 3 Amps is not enough with the BIOS set at: ‘Load optimized Defaults’ :-(
I connected the multi-meter again and loaded BIOS settings: ‘optimized Defaults’. Then I repeated the booting process several times. The highest peak I saw on the multi-meter while the BIOS was starting the PC was a short peak of 3,44 Amps. One of these booting sessions I recorded on camera, which can be seen here:
During this boot up session you can see a short peak of 3,33 Amps in the 18 / 19 second (is fase where the BIOS is starting the PC). Another short peak of 3,44 Amps can be seen later on when (a normal full version of) XP starts up.So a linear PSU of 3 Amps is only okay if it handles / tolerates short peaks of 3,5 Amps.
Too be on the safe side a Linear PSU which has to feed both the P4 and P24 should be cappable of supplying: 4 Amps.I also was curious so what the maximum Amp need of the system was. So I played a HD-movie (1920 x 1080) in a normal XP version with Windows Media Player Classic Home Theatre Version (1 Gb RAM installed). The PC than needs 3,75 / 3,85 amps for playing a high-definition movie
I also played the same HD movie in Windows 7 with Windows Media Player 11. In Win7 the PC needs 3,90 / 4.00 for playing a High Definition movie.So a linear PSU capable of providing 4 Amps is enough for the BIOS to start the system with setting 'Optimized Defaults'. It's also (just) enough for the system to play a high definition movie.
* PC system:
-> Gigabyte GA-G31M-ES2L, 1 RAM HyperX, E7300 Intel Processor.
* PSU’s:
-> the linear PSU 12V/5A powers: P4 = processor. P24 = MoBo, Lynx AES16 digital audio interface and integrated VGA on MoBO (1920x1080).
-> Antec 430 ATX PSU powers: USB, HDD and DVD.
fully AOB optimized cMP2 PC -> Lynx AES16 -> XLR AES/EBU -> Lavry Black DA10 -> XLR Mogami Gold -> Klein & Hummel O300
Edits: 11/12/09
nt
Will 3 amps still be enough at default speeds?
I posted some measurements with the Gigabyte G21 board back in April. They were buried deep in one of the cMP^2 threads but the key bit was:When playing flac-format music data without upsampling with the CPU clock at 140MHz, VID at 0.75000, PSU at precisely 12v, EIST off, 512 MB Kingston "ValueRAM" and a "Full Monty" cMP2 setup, the motherboard and CPU typically draw about 1630 ma.If you reset to defaults, the current draw will get quite close to 3 amps. Running flat out 24/7 with a 3-amp unit is cutting things a bit fine but, given you're going to be resetting fairly quickly, my guess is it'll be just fine.
HTH
Dave
I have problem displaying multi-byte Chinese characters in cMP and cPlay; though the characters are displaying quite nicely in MS notepad when I edited the cue sheet. I believe a friend of mine has no such problem in his setup so I think may be I am lacking the appropriate Chinese font file. Can someone help me with this problem and what font is cMP and cPlay using?
Unicode / DBCS support has to be added...
Thanks for the info; and I finally got it after a few trail and errors.
This is what I'd done in addition to having Asian Language added to Widows XP:
1. Save the cue sheets with Chinese characters as ANSI files. There is a warning when one tries to save files with Chinese characters to an ANSI file, but one has to just ignore it.
2. Go to Control Panel> Regional and Language Options> Advanced, and set the "Language for Non-Unicode Program" to the language one needs--in my case "Chinese (PRC)"
Then everything just works perfect.
I already had added Asian language support in XP. Chinese character input and display works in the OS (file and directory names ext.) as well as applications like Notepad. In cMP and cPlay, the Chinese character in the cue sheets, after edited in through Notepad, are display as garbage. May be I don't have the required Chinese font file? What is the font used in cMP and cPlay?
I am newbie and haven't tried to test this but someone on head-fi has stumbled upon an issue with cPlay's resampler on his Windows 7 64-bit computer. Is the resampler being used improperly or is this just a bug from using windows 7?
To quote: There appears to be no change in volume below -80 db & the output is very dirty below -60db. there appears to be no dither used to clean it up as it sounds just like the early 14 to 16 bit ladder converters with early digital recordings. Massive distortion in the lower 2-3 bits. This seems to me to mean that there is some bit truncation occuring in the software of the Cplay program. This is true no matter the settings.
http://www.head-fi.org/forums/f46/cmp-cplay-media-player-444784/index6.html#post6094470
Thank you.
cPlay outputs 32 bit audio to the soundcard driver. On startup, if the driver doesn't support 32 bit, cPlay reports an error and will not allow for any playback.
Driver converts 32bit format down to 24 (or 16) either at sw or hw level and may apply dithering if needed.
Based on the problem experienced, that driver needs to be either reinstalled or may have a bug. There could also be an issue with 64bit Windows - here, cPlay must be run in 32bit mode.
cics, I have not been able to get cPlay to play .wav files that I have ripped from DVD-A's using DVD-A explorer although i have been successful in playing high rez .wav files ripped from DVD-V's.
Is there something different about the DVD-A .wav files that is not compatible with cPlay?
Thanks,
Jack
Look at "Diagnostics" messages. There should be a reason given for why the wav file was rejected.
cics, the error message I am geting from these wav files is:
Wav file is NOT supported.
Only 16, 24 or 32 bit (PCM or IEEE754 Float) uncompressed stereo (at any sample rate) is supported.
You can use SoX to check your file. Download "MS-Windows executable", unzip folder and copy the 2 dll files (cygwin1 & cygomp-1) to c:\windows. Command from a command prompt is:
c:\mysoxfldr\sox.exe --i "c:\my long name folder\My Wav.wav"
Output would be something like this:
C:\> sox --i "c:\music\01 Track01.wav"
Input File : 'c:\music\01 Track01.wav'
Channels : 2
Sample Rate : 44100
Precision : 16-bit
Duration : 00:02:39.73 = 7044240 samples = 11980 CDDA sectors
File Size : 28.2M
Bit Rate : 1.41M
Sample Encoding: 16-bit Signed Integer PCM
C:\>
It may be different wav format and SoX may help in converting it back to the old wav format.
Try following:sox "MyTrack01.wav" "NewTrack01.wavpcm"
rename "NewTrack01.wavpcm" "NewTrack01.wav"
Test cPlay with "NewTrack01.wav".
In cMP Settings, set RAM Load to "No" and both Suspend svchost & lssas to "No". What soundcard are you using?
Thanks cics. I"ll try your suggestion but I am pretty certian that the files I ripped from the DVD-A's were the stereo track. I will follow up if my SOX output confirms this.
Another problem I am experiencing id that when I try to use cPlay while in CMP-2 it appears that CMP/cPlay recognizes and loads all of my .wav and .cue files but I cannot gey them to play. I have no problem playing the .cue files when nCMP is not loaded when I select hem directly from my music file directly so I know it's not the files themselves but perhaps the path to them?
I seem to have a similar problem when I create a playlist with cPlaylisteditor. I cannot get the saved file tp play in cPlay.
Thanks again for any help you may be able to lend.
I have a question that might seem stupid to those with more than my beginner’s knowledge. I apologize.
My question concerns upsampling in cPlay. Do I need this, given that my DAC upsamples? Is there any potential downside to upsampling in cPlay and then again in the DAC?
Thanks to all for the information. I know that my ears ought to be able to tell me what works and what doesn't, but sometimes it's good to get some outside verification to save on some of those terrible moments of self-doubt.
I can’t disable the upsampling in my DAC, so I’ll just continue to experiment with the setting in cPlay.
This is not a stupid question at all. And the answer is not so cut and dry. It depends on what sample rate your DAC upsamples to and how good it is. Some people prefer the upsampling in cPlay and others think their DAC upsampling sounds better (or no upsampling at all). The *opinions* on this subject will run the gamut.
The simplest answer I can give you is to try it both ways and see which sounds best to you.
Hi DeDe
Just to continue Edward's "it depends" answer... I run the output of cMP/cPlay from the Juli@ card via S/PDIF into a Benchmark DAC1. The DAC1 resamples ALL incoming datastreams to 110 KHz (this being the SR that the engineers feel is the sweet-spot of performance for the DAC chip they use). There is no user option to alter this behaviour - it's the internal architecture.
Curiously (or maybe not...) The change in upsampling with cPlay is readily apparent in my system, from no-upsampling, to 2x SR to 4x SR. Although the DAC1 converts it "down" to 110 KHz, I find setting cPlay to output at 192 KHz gives me the most pleasing (and seemingly accurate/ revealing sound).
Other's preference will certainly vary... so the "correct" way would be as Edward says - whatever you like best. Figure that out, and enjoy your music guilt-free !
That's not a Toy... IT'S A TOOL !!
I find this hard to understand. I too output to a dac1 digital aes3. the dac1 input is 24/96, how do you input at 192. When I output above 96 the dac does not input the signal, Lynx L22/Cplay advanced no power mods. T.
No worries!
@tsearay: I connect to DAC1 using S/PDIF. Though mine is original issue DAC1, it has been factory upgraded to full 192 KHz capability.
Does the Lynx22 send 192 KHz in single-wire AES mode, or dual-wire mode ? It must be single-wire AES for DAC1 to work at 192 KHz on XLR input.
Hope this helps.
Cheers,
Grant
That's not a Toy... IT'S A TOOL !!
AH. You dont output higher than 96k on the Lynx.The Juli@ though can do 192.
And have you tried the analog outs of your lynx? That WILL handle 192k.
Edits: 10/20/09
What leads you to say this? My lynx aes shows an output setting of single wire 192 as well as dual wire 192. I have no way to test it presently, but it appears it is in its ken.
Yeah, for the AES, but we are talking about the L22.
Look at the spec sheet for the AES 16 it DOES have support for 192, but not the L22. Look under Digital I/O in the lynxs ( :) )below
http://www.lynxstudio.com/pop/product_file.asp?i=25
and for the l22
http://www.lynxstudio.com/pop/product_file.asp?i=24
Your right, I missed which Lynx you were talking about and hadn't realized they differed in this respect.
I wasn't sure myself!
Looks like the 16 came later and supports 192. Thats a great thing for those who have the cards.
Ah! Thank you! you are correct the digital out on the Lynx is limited to 96. I have tried the balanced out but prefer the digital to DAC. I have been struggling with "to upsample or not to upsample" to the dac, myself. I get a lower latency at 96 than 41 but really hard to tell if there is a benefit in SQ. AES/XLR throughput to balanced amplification. T.
No worries!
I thought it was worthy of a semi-celebration.
Heh-heh... Thank You Rick for the kind thought. Let us find a suitable container and pour a half-glass...
Cheers,
Grant
That's not a Toy... IT'S A TOOL !!
New cPLAY user that has only used versions 2.0b30 and 2.0b31 so far. On both these version cPlay does not fit in the maximized window (it is just cut a little short) nor does it fill the full width and height of the monitor.
I have used a 17" and a 19" LCD at differant resoutions with no prevail. It really isn't bothersome but was wondering if others have had the problem and if there is a possible fix.
This is on a allout cMP² minus ps mods.
Brad
to fix that:
In cicsMemoryPlayer.pth, set CUE_PLAYER to start with %M (Maximised) instead of %N (Normal). It helps only when video resolution is 1024x768.
i must be using that resolution because it worked, though I dont remember choosing that resolution.
Basically it made the bottom of the player longer so it filled the screen.
I must have something turned off that won't allow me to open the .pth extension???
Brad
d
Duh me thanks, that worked but only with resolution at 1024 x 768 as you said.
Brad
.
Had no luck with dpi setting
Brad
What resolution do you want to use? Even if it is more than 1024x768, you should be able to get close to filling the screen with the right dpi, try with %N if %M isn't doing it, and maybe you can make up the difference with adjustments to display size on the monitor itself. Or make it a bit smaller and have a nice coloured border (desktop colour).
If you are working with Process Explorer or Explorer, or any embedded apps, and you are using mouse exclusively, it is useful to not have fullscreen, so you can easily click between them and cPlay, and then not forget and leave them running while you listen to music.
It looks great at 1024 x 768 with M changed as Dawnrazor recommended and the desktop set to black using it as a border since it doesn't fill the entire monitor as you recommended. Also hiding icons behind cPlay and taskbar set to auto-hide.
Thanks guys
It also may be graphics-card related.I have my cMP2 in a Zalman case and use the "itzy-bitzy "embedded" monitor". I had this issue with the two GA-EG45M-UD2H motherboards I've used in varying degrees... they differed in how much of the window was off the monitor.
OTOH, no issue with either the GA-G31M-S2Ls or GA-G31M-ES2L I've used.
All were as fully optimized standard cMP2 setups as we know how to do today.
Greg in Mississippi
P.S. Ryelands, sorry, have had your cable for about a month now, but too busy to do anything more than take it out, look at it, say 'yup, that's a pretty cable', and put it back. SORRY!
Edits: 10/15/09
[I] was wondering if others have had the problem . . .
We all have the same problem - except for those who use the itzy-bitzy "embedded" monitor. (The display resolution is fixed for that format and is not scalable.)
. . . and if there is a possible fix.
Not so far. As you say, it's not a deal-breaker - just tell your friends that all the best people call Mozart's 41st symphony "The Jup".
Dave
Thanks
Just looking for opinions of what has worked best for you.
Thanks, Brad
If the files are already tagged (downloaded from Linn, HDtracks, etc), I simply use the foobar cuesheet creator plugin. For those that aren't, I recommend Recursive Cuesheet Creator by alrjordan. It's smart and very easy to use.
"We should no more let numbers define audio quality than we would let chemical analysis be the arbiter of fine wines." N.P.
Edits: 10/11/09
Added foobar cuesheet creator to components and it worked great, Thanks
Brad
.
"We should no more let numbers define audio quality than we would let chemical analysis be the arbiter of fine wines." N.P.
I should note that I used this post as a guide.
Brad
Please note that cMP/cPlay have gone through substantial improvements, and the incompatibility "[n]one of these multiple track cue sheets can be read by cPlay" observed by seger no longer exists.
"We should no more let numbers define audio quality than we would let chemical analysis be the arbiter of fine wines." N.P.
Refer to ReadMe.txt for important details & specifications.
Change Log (2.0b31 release):
- DSP refinements
- Threads optimisation for use with cMP's "Realtime" Optimise setting. Recommendation of "Critical" setting remains however significant efficiency gains makes "Realtime" a strong option.
This allows for complex decoding (i.e. FLAC) to occur without affecting playback in cMP. Depending on soundcard ASIO driver, "Realtime" setting could be better.
- SRC efficiency improvements reducing CPU load and L2/L3 cache footprint
- Minor refinements
- Added support for "Tiny" DSP Buffer Size. Previous buffer options remains the same with "Auto" using either "Small", "Medium", or "Large", i.e. "Tiny" must be manually selected.
"Tiny" buffer option reduces L2/L3 cache footprint enabling lower specification CPUs to be used with resampling. The 2MB L2/L3 cache requirement remains to produce highest quality 192k output.
cPlay documentation can be found at the new cMP² website . cPlay 2.0b31 offers significant SRC processing improvements and remains the reference resampler. Those not using SRC should re-evaluate this. "Tiny" buffer setting is also highly recommended (especially in combination with SRC).
Please REMOVE previous versions before installing cPlay 2.0b31. Normal (SSE2), SSE3 B9, SSSE3 B9 and SSE4 B9 versions are available. Running cPlay on a CPU not supporting the required instruction set will cause cPlay to exit immediately. For best results, uncheck "Timer" setting and test different "Buffer" sizes.
Putting things into perspective
Changes are readily observed with each new release. What follows is a technical explanation of why this is so. Modeling (Periodic) Jitter Theory for a wide range of J pp using a frequency sweep of the audible range and beyond yields the following:
Practical implications are profound as such Jitter Distortion damages sound quality:
- Sideband distortion rapidly increases with input frequency, i.e. high frequency sounds are most affected. This is seen in the rapid rate of distortion increase as frequency rises. Critical harmonic sound information in this HF band (delivered by tweeters) is distorted. This has detrimental effects such as veiling, poor transients and poor tonal decay.
- Most important insight is the extent to which J pp must be reduced.
- We see that for high J pp (150..350ps and beyond) an audio designer is not well rewarded for Jitter reduction efforts, i.e. despite large scale drop in J pp , distortion levels remain stubbornly high. Modern DACs (including some soundcards) have an excellent noise floor of -118db or better. This implies that jitter distortion arising from HF input (above 5kHz) will be well above such noise floor. For example, an input tone of 10kHz and J pp at 150ps would yield sideband distortion of -112.6dbFS. Ouch!
Interestingly this explains why some reviewers unexpectantly express disappointment with front-ends offering ~200ps J pp performance. That is, the noise floor is polluted with excessive sideband distortion.
This may also explain why some vendors have become disillusioned with Jitter and have abandoned it's role as a major source of audible distortion. Sadly, others have yet to understand how to measure such distortion (which can only be done correctly at the analogue outputs).
- Implications for poor setups with J pp above 150ps suggest sound quality improvements will not be easily experienced. Perhaps this explains why some object to for example RAM size, quality & setup to have sound quality impact.
- Much to our relief, distortion drops rapidly below 150ps J pp . As can be seen, for each 50ps drop, distortion reduces in ever increasing drops. That is, we see an exponential decay. Consider for example 50ps J pp , any improvements here will have factors more benefit than with same improvement at higher J pp . This suggests high quality setups like a fully specified cMP (whose measured jitter performance is 51ps J pp RSS using foobar) would be significantly better at revealing improvements.
- For the ideal case, jitter performance results in sideband distortion for all input frequencies below the DAC's noise floor. A noise floor of -130dbFS, requires jitter performance below 10ps J pp ! Intel's new 32nm Nehalem platform is a good step forward.
Rarely does theory and practice meet so beautifully. The efforts of Julian Dunn who laid the foundations for understanding Jitter and its challenges is truly remarkable.
Can you add .APE (monkey Audio) files support, may be in new versions?
cPlay 2.0b31 Released ,plese!
I just got around to installing 31 today.
It is amazing how different versions can change the sound pretty radically.
mixed is a word I would use like a previous poster did.
SOme things like the resolution is dramatically improved and you can hear more into the recording. That comment about a light being turned on is so true.
The soundstage is a bit weird on my rig with 31. The center is so prominent now that it kind of excludes the other areas of the soundstage., but the center does sound really in the room.
Warmth is missing compared to 30. I miss that. And as I was listening to 30 and thinking how I missed some resolution, and now that I have it with 31, I miss the warmth.
I just put it on tiny and that is what I listened to. But maybe that isnt the best setting for my cpu?
Are there any settings that can get me "30.5"?
I am using a latency setting of 32 with tiny and src 145db set at 192k.
And do take this with a grain of salt. I just got my cmp2 up and running a month or so ago and now I have new amps and in the middle have changed speaker positions a bit. SO my memory could be off, but I did listen with 30 and then install 31 and listen.
If I had to pick it would be a tough choice. I think i lean toward 31 but on some recordings I miss 30.
Thanks
Sometimes lowest is not "best". On my system, latency setting of 48 sounds the most powerful and "precise", but falls short in terms of "warmth" and treble smoothness. I found that 128 sounds the most pleasant, natural, and the bass is better than any other source I've heard in my system.
Incidentally, latency setting of 128 is also what cics recommended for 192k. He said, "for 192k output, latencies to 128 samples (ASIO Hz of 1500) will be excellent. This assumes there's no "dead bands" in PLL, i.e. PLL is inoperable for certain jitter frequencies."
My other settings are: Tiny, SRC 145db, 192k, AWE, Timer Off
"We should no more let numbers define audio quality than we would let chemical analysis be the arbiter of fine wines." N.P.
It just was more musical and 31 often left me uninvolved even though I was hearing more.
But I never tried 128. The cplay diagnostics reported that the Lynx liked 64 so I tried that and 32 and tiny, small, and medium. It just never sounded as good as 30. Perhaps I should give 128 a try.
I'm not sure if it's a glitch, but when I switch between different latency settings (i.e. 48 and 64), the cPlay diagnostics *always* report my current setting as "preferred". Hmm...
"We should no more let numbers define audio quality than we would let chemical analysis be the arbiter of fine wines." N.P.
Hey, I was on 32 and it reported 64 was preferred so I tried 64. So I dont think I have that issue.
she needs to make up her mind!
Quick update, I now prefer the 48 sample setting after remaking my I2S cable with RG179 coax. Digital audio works in mysterious ways...
"We should no more let numbers define audio quality than we would let chemical analysis be the arbiter of fine wines." N.P.
SOunds like!
I tried 128 with tiny, and that does seem to be much better. Not sure i will stay with 31 but right now it is pretty good.
Glad you like it!
In one of his posts, theo stated that "my system has many degrees of adjusting tonal balance ... [so] when I make changes I always prefer to go the way of better dynamics then adjust the tonal balance accordingly."
I happen to have very similar beliefs when it comes to choices of equipment/playback software. I believe that certain qualities, such as warmth, tonal balance, even better soundstaging, can be had through relatively simple tweaks. On the other hand, attributes like detail, dynamics, and low noise floor are more difficult to obtain. Oh, and I also happen to prefer to do things the easier way!
"We should no more let numbers define audio quality than we would let chemical analysis be the arbiter of fine wines." N.P.
This combo at src/192 is nothing short of amazing, I just love it. Imaging, dynamics and a sense that a light is turned on at the back of the soundstage---so much more to hear but smooth, very smooth. Congratulations.But I have some questions regarding some anomalies. I get zero metallics but I do get an occasional what I call a 'static-ee' noise which goes away via going to another track or changing sample rate. Others get this as well. I got the same noise when I was running juli@ analogue outs. If metallics were a 10 as an issue the static-ee is a 3: far fewer and easier to fix but annoying just the same and does not require a reboot. So upon throwing this around in my head I believe it may have the same root cause as metallics but since the system is no longer using juli@ drivers it is just manifested as
different. But assuming this is correct here are some more observations: when I get the static’s my cpu usage is typically 50 % with spikes up and down. But since Sunday when you asked me to run 48/tiny I have gotten zero static’s so I thought it was the combination of 48 / tiny parameters that was doing it (and maybe it is) but I also checked cpu usage and I'm back down to 25%. This same phenomena occurred back in June when I was running spdif out to the Benchmark: for a wonderful week I was getting no metallics while getting 25% (when prior to this I was getting 50% cpu usage and lots of metallics). So I think there is still an issue with juli@ (albeit minor with I2S out) that is somehow associated with cpu usage jumping around from 50% to 25%. This am I had to reboot after being on for 3 days and upon reboot I got 25% cpu usage but with spikes up to 45-50%. After being on continuously for a while the spikes started to settle down in the 35-40% level but come back to a higher level after not running cplay for a while but with the pc still continously on.Edit: my mistake on the cpu usage bit, I found out my pc reset unbeknownst to me my bios hcc to auto and I got clock speed of 1600 when I reset back to 840 cpu usage goes back up to 45-50%. When I reboot I normally sense a reset of bios when It takes long to boot up, I missed it. I hate computers. I love cplay/cmp.
Not to throw cold water on my glowing review of cplay 31 @ tiny/48 (and it is nothing short of magnificent) what in your opinion could be happening here with the noise issues and cpu usage which seems to be correlated. Look if I had to live with this situation as it is now I would be happy for a long time. I'm just trying to understand if juli@ is fatally flawed and whether I should pursue another sound card. I was tempted by the Lynx but w/o anyway to run battery power or I2S out I fear the Lynx may be too limiting now that I have experienced LiFePo4 batteries and I2S out.
Edits: 10/20/09
Which type of cable(s) are you using for the I2S? I currently use Cat5e for my I2S -> DAC connection. While I can't complain about the sound quality, it certainly couldn't hurt to get "proper" cabling. I'd love to hear your thoughts on this.
Thank you in advance!
"We should no more let numbers define audio quality than we would let chemical analysis be the arbiter of fine wines." N.P.
It is 75 ohm coax with teflon dielectric. I believe it is bg 179 (or something like that). It costs about $40 for 4 feet of it. Its very delicate but fairly easy to solder right onto juli@ pins.
Thank you! Sounds like the cable (or something similiar) in the link below. Can you please confirm?
Oh, and did you ground the braided shield or make twisted pairs with the cable?
"We should no more let numbers define audio quality than we would let chemical analysis be the arbiter of fine wines." N.P.
yes it was rg not bg 179. I used the cable as a coax: soldered the center connector to juli@ pins & grounded the shield to a juli@ pin on the 10 pin connector. I twisted all the shields together and fastened a wire from the twisted bunch of 3 shields to a ground pin on juli@.
Thank you theo.I remade my I2S cable yesterday following your instructions, and it sounds good so far. Overall, it sounds cleaner, more detailed and resolving. However, the presentation also becomes a bit bright and forward, making vocals less enjoyable. I wonder if this is due to the lack of burn in. Does my experience with the RG179 coax sound familiar to you?
"We should no more let numbers define audio quality than we would let chemical analysis be the arbiter of fine wines." N.P.
Edits: 10/28/09 10/28/09
Not sure...I was breaking in a new sound card as well as the I2S cables. Give it a few days then pass judgment. But yes it did mellow out eventually.
Edits: 10/29/09 10/29/09
Thanks, I think it already improved noticeably during the past 24 hours (I left the system on the entire time). By the way, the 30 AWG conductor of the RG179 seems *really* fragile. I used some hot melt glue on my Juli@ to ensure I don't tear the I2S cables off by accident. Ugly, but it works!
"We should no more let numbers define audio quality than we would let chemical analysis be the arbiter of fine wines." N.P.
Edits: 10/29/09
why not solder?
Glue on top of solder. I'll post a pic when I have time.
"We should no more let numbers define audio quality than we would let chemical analysis be the arbiter of fine wines." N.P.
Edits: 10/29/09
I was thinking of using a usb cable for I2S, just a thought :).
Brad
After intensive listening and adjusting (in a tri-amped system one has many tonal balance options) I have currently settled on medium buffer setting with juli@ panel @ 64 latency. I also needed to adjust subwoofer and midbass levels several iterations to get it right. With 31 I get great balance at low levels but at FFFF levels the highs seem to come forward and be prominent (not edgy) . Edit: ...Subsequent listening and I find this phenomena has diminished...I can understand why many have mixed emotions re 31 because w/o the adjustments I have, I may not have been able to dial in an acceptable tonal balance. I am now listening to cmp set to Realtime (rather than Critical) and will evaluate this based on your comments Cics. But what really does this do versus the Critical setting? My memory is that Critical was optimum, is that still correct for juli@?
Edit...Critical is better...
Edits: 10/17/09 10/17/09
Optimise set to "Critical" is the recommendation. "Realtime" can work with a minimal ASIO driver - one needs to test this. I prefer Critical with my Lynx.
It seems that you are changing too many things at once (and you have a new Juli@). Sound will change with more burn-in. Try this: in cPlay, set buffer to "Tiny" and set Juli@'s latency to 48 samples - gives this at least 48 hours.
You are right I believe my juli and my teflon coax (for I2s) are breaking in. I had been listening to 64 latency and tiny buffer for last several hours and its great(I mean absolutely MAGNIFICENT). Right now per your suggestion I switched to 48 latency and will evaluate this for awhile. When I loaded 31 I could not run 48 latency with tiny buffer (it stuttered) so I believe it was breakin all along.
But so far after several days the sound is simply the best I have ever heard in my system. The anomaly with the highs (dynamics shifts back & forth) was simply a breakin phenomena. The imaging is nothing short of superb. Later I will comment on the tiny/48 combination with cplay 31.
I have been down for 2 weeks. Events had overtaken me. In an attempt to implement I2s I hooked up my LiFePo4's backwards and fried juli@. Took me two weeks to get a new juli@. In the interim I lost my 96 year old mother...lots of stuff to take care of. I got my juli@ and just now loaded cplay 31 (still with spdif into Buffalo)...wow and I'm not even to 140 host clock control nor 3 3 3 5 on the memory. When I'm down for a while I have to creep back to optimum hcc and timings. But nevertheless 31 is great. Can't describe it yet but it sounds very very natural. Lesson learnt for me and hopefully for others KEEP POLARITY CORRECT ON BATTERY SUPPLY. More on 31 later, Initially very very natural
Hey Theob,
Sorry to hear about your mother. That is sad to hear, and kind of puts this hobby into perspective.
Thanks for keeping us updated and for the tips on the Juli@. One of the nice things about that card is that it is such a great performer for the money, so accidents are recoverable.
One of the reason i kind of have to run my lynx stock...cant afford to screw it up!
Thank you for your kind words. I appreciate it. I can't afford to screw up the Juli@ BUT I did, hopefully nobody else will.
For those of you who have followed my issues (& I had a lot) I could not get spdif to work on my juli@ after a dreaded metallic. So necessity becomes the mother of invention...I had my 75 ohm teflon coax to try I2s but 3 weeks of trial and error I could not get the mechanics to work. But with help from inmates (bertel and wackybytacky) I finally got it to work. Big shout out to Sonics without his i2s sketch I could not have corrected my incorrect hookup to Buf32. Alfred and Robert you are the man (er men)!!! Anyway I2s from cplay 31 is a significant step up in soundspace (wider and deeper) separation of instruments, fantastic (ie bigger) dynamics and lo and behold no metallics (omg what did it take me ---10 months from identification to solution). I'm running 192 with buffer set to auto and juli@ latency set to 64 (tried 48: too dry, tried 128: not enough air, 64 just right). Anyway 31 is unbelievably natural, open and dynamic. I'm not sure I am set on all the parameter setting but I am so glad to be back into 192 music! Thank you cics for 31: a masterpiece!
ps Robert: I think I am not going to try those 22 ohm resistors to damp the I2S signals...just a little gun shy now. I'll wait till you measure juli@ output impedance.
For those of you with Buf32s be carefull and learn from my mistakes if you decide to go I2s. Yes it was worth it but it took 3 weeks. The journey may not be for the faint of heart.
Sorry Theob. I was looking at things relative to the price of my Lynx.
If it is any consolation I think most of us have done some boneheaded things to destroy audio gear.
Why within the last year I toasted my Lynx card by wrapping it with tinfoil. Who would have thought it would have overheated? The damn pci slots end up making things upside down anyhow...
Fortunately Lynx replaced the op amps for a reasonable fee.
Forgive me, but I think I remember that we tracked down your metallics to your juli@. Do I have that right?
Yes the metallics are caused by juli@ drivers (required when using spdif out). However when going I2S no drivers are required and so no metallics.
I used to have a professional sound card (not Lynx, but the RME AES-32) in my cMP as well.However, ever since I found the I2S of the Juli@ outperformed my RME's AES, I sold the RME and never looked back. Then again, I didn't find the stock SPDIF out of the Juli@ impressive at all. The RME was significantly better than the Juli@ unmodded.
"We should no more let numbers define audio quality than we would let chemical analysis be the arbiter of fine wines." N.P.
Edits: 10/14/09
Hey FP,
I am using the dacs on the Lynx as I will soon be triamping and need 6 dacs basically. The Lynx is fantastic for the money.
Actually I ran out to my external dac and while it was pretty good, I thought the Lynx was everybit as good, probably a bit better given the price difference.
Are you doing DSP crossover inside the PC? If so, what software are you using for that? Or does the Lynx software allow you to do that? Pardon my ignorance. I am running six channels active but do the crossover in a modified Rane RPM 26z. That is fed by a Mac/iTunes/Amarra/Offramp but am reading this thread to see if I want to try doing it with cPlay/Lynx instead.
Hey B,
The plan is to use the thuneau Allocator (www.thuneau.com) with cmp/cplay.
The hold up has been the funds for new identical amps. I get them tommorrow and I can then start in earnest.
Play-mate and Bibo01 are ahead of me in implementing the Allocator with the Lynx. Hopefully I'll catch up soon enough.
Nice Job! With a couple of mentions of mixed reviews and 2.0b30 being so good, I was not sure what to expect. I was initially pleased to see that 31 clearly had not taken a step back in any regard. It seemed at first that there was an increase in detail and clarity. CMP2 has likely gone well past the ability of my speakers to resolve all that it is putting out, but as I listened over a couple of sessions it is clear that while it is a subtle difference, I am hearing much deeper into the music. It is again time to re listen to my entire collection. For my system, SOX has been the clear winner, but with this release the gap has closed to the point that I can easily flip between SOX and SRC with no loss of air and detail and enjoy the subtle advantages of each on given track. Its now closer to flipping phase. I run on a laptop with a celeron with very small cache, and Tiny buffer really works well.
Again thank you for sharing this great audio achievement.
I have the new release for some hours running and I am amazed, that after the last, great version again some subtancial qualities are obvious. Again more details. As stated by someone already, lyrics are easy to follow. Voices in the background are clearly separated. I am used now, to let the system run in for several hours, before I give a statement. Thanks to sics for this version. Within a few days I will have my buffalo-dac running and can not wait for the outcome. Walter
Tarkowsky wrote:
As stated by someone already, lyrics are easy to follow.
You're absolutely right - but it may not always be a good thing.
My wife was amused to discover last night (over 40 years after first buying the record!) that the lyric for Chuck Berry's "Tulane" has one word in a line that is not what the usual sources say it is. They usually give it as "Get me out this funky jail". They're wrong.
Worse, this morning, I was playing "Purcell in The Ale House" (an album of 17th Century "catches" and "rounds", i.e. light songs for private performance, often by young men, in "noble" houses).
I knew HP wasn't among the angels whose praises he so eloquently rang in his music but I wasn't prepared to hear staid English choral group Pro Cantone Antiqua singing a hitherto-indecipherable line (no lyrics in the sleeve note) to a song called "Once, twice, thrice I Julia tried, the scornful puss as oft denied" was in fact "Kiss my arse" sung repeatedly in the most delicate harmony one could wish for.
Whatever next?
Cics, thank you for your clarification re latency. It did become clearer under closer study. b31 has reduced latency to 55 using tiny with the L22. My impression of b31 is somewhat mixed, with the greatest regard to that which I can comprehend but never achieve by myself. The instrumentation is very clearly balanced but does not have the room interaction of b30. Base music has lost some of the string decay, which is to say slightly smeared. For example Patricia Barbers Mythologies. I have switched back and forth several times and am still burning in b31. Thank you again, T.
No worries!
congratulations again, cics....!
there is a remarkable difference in this release :
its very obvious that the sound is drier and less resonant. especially on multiple brass arrangements there is clearly more precision. maybe a tad less glossy and shiny, but equally transparent.
beautiful !
I so adored b30. But after 3 weeks in-depth listening, I had two reservations (SOX, HQ, 48K, Small Buffer) :
-- there was a haze in the high frequency that could inject extra space in some recordings. The resulting 3D effect could be quite attractive but ultimately felt false. For vocals that were mixed way at the back of the sound stage, the images are fuzzy.
-- for some recordings (I am thinking of 'Astral Weeks' as an example, the Japanese remastered version) the emotional impact of the instrumentation was not quite there when compared to the LP version (again Japanese vinyl version). The attack of instruments such as congo, high hat, guitar etc. were soft and I couldn't feel the human behind the playing. Please note that the emotional content of the instruments is a big thing in 'Astral Weeks'
With b31, I now switch to SRC (tiny buffer, 48K, 121db) and the above too 'problems' are mostly gone! The high end haze is no longer there and the emotions are mostly back in 'Astral Weeks". And the bass is now more solid.
I find myself listening to the music rather than the hifi, finally, after all these years in the digital world.
Thanks Mr. cics.
In the past year (plus) with the many cPlay releases, the difference in SQ has ranged between subtle to a little more noticeable (and I'm a "Sharpener"), but (I may be exaggerating to say this) but wow, the difference between Tiny and Small is about the most dramatic difference I've heard. And I haven't even compared it to Medium and Large yet.Dawnrazor, you'll probably appreciate this: I started to listen to "Loveless" by "My Bloody Valentine" and was floored at how I could finally hear the lyrics in "When You Sleep" (without it being smeared by all the guitar distortion).
A usually "dark" (pardon the pun) sounding "Masquerade" by "Clan Of Xymox" now sounds more intelligible, and a usually "bright" sounding "Four Calendar Cafe" by "Cocteau Twins" now sounds less edgy.
Edits: 10/08/09
edward wrote:. . .the difference between Tiny and Small is about the most dramatic difference I've heard.
Judging by your list of favourite music, there's not much risk of our getting in each other's way in a music store but, having just experimented with these settings, you're certainly not far out - the difference is very marked.
(What I haven't done is compare v31 set at Tiny with earlier versions which, of course, lack the setting.)
Meanwhile, it's worth visiting the link below.
Dave
Edits: 10/11/09
Hey Edward,
I unfortunately havent had any time with 31 on the big rig yet. I was really digging 30 after my recent cmp2 build.
Hard to imagine it being better but cics always amazes.
I can tell you that over the years recordings that i thought were crap have actually turned out to be some of the better ones as my stereo has improved.
I can totally appreciate your comments and hope soon to share in them.
Thanks for posting this and I'll see if I have similar results.
There is a line by the Church in "Destination" on their starfish album that I have always been unable to hear. I hope this version reveals it, but I just think that might not be intelligible any how. We'll see.
D
Well, Dawnrazor, I know you and I share the same musical taste, and some of those 80s recordings that we are talking about are just dreadful. This past year has been rewarding though, as I have picked up "remastered" versions of Dead Can Dance, The Cure, Tears For Fears, Yaz, Joy Division and there are also remasters of Skinny Puppy and Cocteau Twins out there and Bauhaus is releasing a remastered version of Mask (and hopefully the rest of the catalog will follow). The Yaz box came with a DVD that included 24/48 versions which sound awesome. In fact all these remastered versions I mentioned sound awesome. I'm hoping that now that all these artists are doing remasters, that they digitally mastered at 24/192 and that these will some day be available. And I do thank NIN for paving the way, and offering "The Slip" as a free download (in 24/96 no less), but his music is so distorted anyway, that having the 24/96 version doesn't really offer you that much more.
But here's hoping that more of the 80s discs follow suit. (And I know my wife will pay good money if and when Duran Duran's entire catalog is remastered and/or available as hi-rez).
E,
That is awesome info. i had no idea all these remasters were around. i might have to buy some music for the first time in a LONG time. (OK I did buy that Sisters of Mercy "First And Last And Always" release that was supposed to be like the original vinyl release that was somehow different than the original cd release. But it wasn't remastered.)
Thanks a bunch for sharing.
FWIW your wife is better than mine in this case. When she saw me adding the cure cds to my shopping cart at amazon she said something like "you already have those cds, who cares if it is remastered. Its the same old song anyhow..."
I was nice enough to have added the new Michael Buble cd she has been dropping hints on, so I said, "Then why do you want that Michael Buble cd? Its mostly just covers of the same old songs..."
My wife is the perfect wife in almost every way. Audio is her one flaw...
Ironically I have enough pcs lying around that I was thinking of making a rip box to see if there were better ways to rip than I did years ago. Now I will have even more reason to go to the trouble!
Thanks again!
.
.
Hi Cics,
Thank you very much for the b31 update!
My feeling for this version is quite mixed: This version's accuracy and sound stage is the best thus far. Sound stage is so accurate that I can feel the singer standing in front of me. More micro details are revealed when playing back complex classical (orchestra) music. The other advantage to me is richer bass but not boom (well I only use JM Lab micro Utopia so other experts who have floor standing speaker will provide more relevant comments). However, the sound is "too accurate" that musicality is affected, especially when compared to b30. Therefore this version is good for classical but a bit coarse for female vocal. Therefore, I need to check "small buffer" in stead of "tiny" in order to smooth out the sound a bit. Other settings remain the same (upsample 192, VHQ, Linear 95, no alias). I still miss the female vocal in b30.
Thanks again for the good work!
Try adjusting juli@ panel latency. As I adjusted from 48-> 128-> 64 I found the impact on dry or etched to detailed and liquid very profound. I prefer 64 but you may like another setting.
You should try SRC, as cics recommended. Having experimented extensively with SoX settings in the past, I have a strong feeling that SoX tends to outperform SRC in certain areas, but the latter is simply more pleasing and musical overall.I will report back after I have the chance to properly evaluate the b31 release (new DAC still burning in).
"We should no more let numbers define audio quality than we would let chemical analysis be the arbiter of fine wines." N.P.
Edits: 10/08/09
Cic's wrote: Refined setups will readily reveal sound quality changes with changing latency. Its best to use lowest stable latency. A good programmer would argue that latency is a non-issue for playback. "Just set it to highest level as we get less context switches which is more efficient...". This is not correct for best SQ (Sound Quality).
After months of playing around with Cplay/cMP/Juli@ settings i have to conclude that in my config high buffer settings sounds better. When i choose small buffer in cplay or the smallest latency at the juli@ panel, the sound looses musicality en smoothness. Maybe its a bit more refined, but less enjoyable.
What is your experience with latency settings?
For 192k output, latencies to 128 samples (ASIO Hz of 1500) will be excellent. This assumes there's no "dead bands" in PLL, i.e. PLL is inoperable for certain jitter frequencies.
B31 will bring important SRC processing improvements. Would be very interesting to see how it works in your setup.
Does this apply to all cmp users, or Lynx card owner only? Do you recommend setting 128 samples @ 192k for everyone?
Thank you!
"We should no more let numbers define audio quality than we would let chemical analysis be the arbiter of fine wines." N.P.
It's not specific to Lynx cards but directly relates to output rate. At 192k, latency of 128 gives 1500 ASIO buffer loads likewise, at 96k, latency of 64 samples gives 1500 ASIO Hz. The idea is to ensure this periodic "interference" does so above the PLL cut-off.
I find with Lynx cards, its best to disable its Synchrolock feature (digital PLL for external clock sources).
Thank you for the explanation. Forgive my newbieness, but what about those of us using I2S over SPDIF? Should we go for the lowest latency possible, or stick with 128 samples like you suggested?
"We should no more let numbers define audio quality than we would let chemical analysis be the arbiter of fine wines." N.P.
You need to test which latency (32..128) is best - this depends entirely on the PLL performance specific to your equipment.
Thanks, I'll do so when I get the chance.
"We should no more let numbers define audio quality than we would let chemical analysis be the arbiter of fine wines." N.P.
Edits: 10/08/09
I am having some confusion with setting latency. I have the Lynx L22 sound card using digital out to a dac. When I review the Cplay's Guide for ASIO drivers it indicates an ideal setting for the Juli@ of 48 in "Buffer details" and "Latency". How exactly does one set latency. My buffer details are( min 32 max 1024 preferred 32 granularity -1), Latency is (input97, output 72). Is there a way to adjust these settings. My PC is a Cplay dedicated modified with the exception of advanced mods, Gigabyte ES2L mobo. T.
No worries!
Your setup "(min 32 max 1024 preferred 32 granularity -1)" is perfect. The key item to look for is "preferred" which is set to 32 - lowest for Lynx.
Latency is adjusted using cPlay Settings > ASIO Settings > Lynx Mixer settings .
I noticed this at DIYAUDIO in the commercial section:
http://www.diyaudio.com/forums/showthread.php?t=151630
It seems to be on the right track. First thing one thinks about is powering it properly.
I personally would prefer for conversion to be done outside of the computer but one cannot be too sure of their assumptions!
Looks interesting. Although there are a few points I'm unclear about.-The driver appears to be a WDM driver with a patch mixer that will route other driver interfaces (ASIO) to their driver. Is it just me, or does this sound like the same thing as ASIO4ALL? IOW, it does not sound like a true ASIO driver. Or maybe I'm just mis-reading the ESI description. I'd like to know if it's truly ASIO (and 32-bit).
Edit: Doh! I just realized this is from the makers of the Juli@ So drivers should be fine, right?
-Also, they claim supporting up to 24/192 however I've seen other products make this claim and then not support 24/176. I'd like to confirm that they indeed support all rates (44.1, 48, 88.2, 96, 176.4, 192).
Can anyone answer these questions?
Edits: 09/27/09
> The driver appears to be a WDM driver with a patch mixer that will route
> other driver interfaces (ASIO) to their driver. Is it just me, or does
> this sound like the same thing as ASIO4ALL?
> IOW, it does not sound like a true ASIO driver.
> Or maybe I'm just mis-reading the ESI description.
> I'd like to know if it's truly ASIO (and 32-bit).
You are just reading things into the description that aren't there. The ASIO driver is integrated with the WDM driver. if I had written the drivers, that is what I'd have done.
You just select ASIO output from your player. You do not need to patch anything with DirectWire. ASIO output to the AudioTrak Prodigy HD2 has been flawless for me in a Win XP system with J. River media Center 11/12/13/14.
Outputting at 32 bits from J. River MC seems to work.
> -Also, they claim supporting up to 24/192 however I've seen other
> products make this claim and then not support 24/176. I'd like to
> confirm that they indeed support all rates
> (44.1, 48, 88.2, 96, 176.4, 192).
I played material in J. River MC at various rates.
44.1/16 - works
88.2/24 - fails (error message from JRMC.)
96/24 - works
176.4/24 - works
192/24 - works
There are several threads on op-amp choice on the head-fi computer audio forum.
http://www.head-fi.org/forums/f46/
Bill
Thanks for setting me straight on my ASIO confusion Bill. I guess I just never heard it described as integrated with the WDM driver before. (Or I'm just not paying close enough attention).
I still find it curious that 88.2 doesn't work and yet 176.4 does. (Or is this more common than I realize - and I'm just being dense today).
That is weird. SO far I only have direct experience with 2 cards, Lynx and M-audio Revolution. Both could handle anything between 44 and 192.
I am able to output at 24 bits / 88.2 KHz to the ESI Juli@ and it plays.
ESI makes both the Juli@ and the AudioTrak Prodigy HD2. The console apps look similar and the driver files have similar distribution of function. I haven't a clue why 88.2 doesn't work for the AudioTrak card.
I will point out that supporting rates higher than 48 KHz for playback was mostly a checklist feature with little use for consumer audio applications until recently. Both Juli@ and the AudioTrak cards were designed several years ago.
Right now, there is just one 88.2 download I have any interest in acquiring (Mackerras and the Scottish Chamber Orchestra playing Mozart symphonies 35-41). Linn sells this set as 44.1 Flac downloads or as 88.2 downloads for considerably more money. I'm not keen about paying more for a higher quality format that will be converted to a different rate to be played on my office system. The effect has been to keep me from buying these performances in either format.
Bill
Found this review:
http://ixbtlabs.com/articles2/multimedia/prodigy-hd2-gold-page1.html
Thanks for the link.
Odd that under ASIO it shows 88.2KHz as "Not Supported", yet 176.4 IS supported. And it does not support "Output Ready". But it looks like it is ASIO 32-bit.
Maybe I am missing something here, but that doesnt seem like a card that fits into cics principles, especially cmp2.
The lowest latency supported I saw was 256. Wouldnt we want a card that did 32?
Are we still talking about the plain AudioTrak Prodigy HD2 card?
> The lowest latency supported I saw was 256.
> Wouldnt we want a card that did 32?
I normally use a setting of 256 for latency. I tried 48 just now and it worked fine. That is the lowest setting in the Prodigy control panel.
I use the drivers written for this card and supplied by ESI/AudioTrak. It is possible to use generic Via drivers but that might require re-flashing the firmware on the card.
The fancier versions of this card may come with different firmware and work only with the Via drivers.
Bill
Hey Bill,
Yeah, I think that is the one...whatever was in the link.
Anyhow, it is good to know you can go lower. Maybe 48 is ok?
I just set my Lynx to the lowest setting which is 32 and that works great.
I guess the easy solution is still not there.
Lots of compromises that are not very attractive.
Oh, well. Thanks,
Anybody getting an ocassional dropout on cplay/cmp? Seems to have increased in frequency the last few weeks. If not what could it be, dirty connections? This is not a cpu interrupt hiccup its something else I believe. Sometimes it manifests itself as a stutter. Anybody else experience this?
Edits: 09/17/09
I don't know if it's the same thing as what you are talking about, and I've been waiting to mention it to "cics" until I tried to better diagnose it myself, but here's what's happening to me.
I don't know if it's a new issue, because I've only recently (in the past month) been resampling with SRC. Previously I used SOX or no resampling. But I think my issue is related to SRC. What happens is that when a song transitions to it's next segment or sometimes to the next track, then it stutters. (What I mean by "segment" is how the file is broken up into segments and delineated by the number in parentheses). I thought it was related to latency/buffer by my sound device so I haven't said anything before, because I was playing around with it. But one thing I did confirm, was that the stutter always happened at the same time. So I played a track that had (2) segments and it always stuttered at the same time, and then I put a 512MB RAM chip in so now the same track was not segmented and it played without a hitch. Changing the latency on the sound device did not seem to matter, and I was not experiencing any other dropouts at the lowest latency setting. FWIW I am using a USB device at the moment, so I thought USB was the culprit.
Any ideas?
This only applies to upsampling with SRC@145db. This stutter is eliminated when running cPlay with cMP. Also within cMP, Optimise should be set to Critical or Realtime.
Here's a test that I always run for gapless:
Transitions are totally seamless within cMP (except when Optimise is set to Player). Without cMP, its difficult to eliminate.
Yes, of course, I am running cPlay within cMP and Optimise is set to Critical. (And I'm booting straight into cMP). I've also made sure Cryptographic service is not running.
Do you have any other suggestions? I'll keep poking around to see if I come up with anything.
Well, I figured it probably wasn't the same thing as Theo and I think I confirmed it's just my POS USB device. I tried a few other things and then went back to the latency settings and cranked up the latency to 512 and now it plays seamless.
I'm broke right now, otherwise I would have bought a PCI card by now. I've been using the USB device because I need to listen with headphones on. (I only have time to listen when my wife and son are asleep - IOW I'm nocturnal).
Well, let me know if you have any other ideas to improve my current situation.
I don't know if my stutters/dropouts occur at segment transitions but I will try to observe. Lately I have not been using upsampling and I get the problem just the same. This is a fairly new problem...did not get it before. It is unpredictable...does not always happen.
Edits: 09/18/09
Edward I was popping fairly often this am so I checked Control Panel and that stubborn "Cryptographic Service' had re-enabled itself. So I stopped it and my stutters stopped (at least for a 60 minute file playing). But sometimes this phenomena just doesn't occur so I cant be sure it is root cause or not but check it out.
Cryptographic Service should be set to "Manual" but Windows will start it whenever a new device type is plugged into USB or doing a detailed query on a device. It's always good idea to check that only the 2 required services are running (RPC & Plug n Play). Some soundcards may require a 3rd service (Windows Audio).
With "Cryptographic Service" running, there's a notable deterioration in sound quality!
Wow, thanks for another great tip. After disabling 'Cryptographic Service' (and a few others like 'Spooling Service') through Administrative Tools, there is a definite SQ improvement such that some previous CD's that I thought were inferior (to the LP sound) transfer are now OK. Some of the missing emotional elements are back. Once again, thanks.
I found that the services do come back if you just set them to 'manual'. What I have to do is to set them to 'disable' in 2 places:
1. In the 'Services' window in 'Administrative Tools', and
2. Right click on the specific service, go into Properties, hit the 'Log On' tag, high light the item in the 'hardware profile', and then hit the 'Disable' button.
After power down and reboot, the services remind disable.
Refer to ReadMe.txt for important details & specifications.
Change Log (2.0b30 release):
- DSP refinements
- Optimisation for Intel's Nehalem platform (Core ix CPUs)
- ASIO Settings activated on invalid latency or output
- Other low level refinements (SRC & kernel)
cPlay documentation can be found at the new cMP² website .
Please REMOVE previous versions before installing cPlay 2.0b30. Normal (SSE2), SSE3 B9, SSSE3 B9 and SSE4 B9 versions are available. Running cPlay on a CPU not supporting the required instruction set will cause cPlay to exit immediately. For best results, uncheck "Timer" setting and test different "Buffer" sizes.
...I also got a Buffalo 32s running the digital power supply off of 3 LiFePo4's in parallel and yes I confounded several changes and I revoiced my system after all this new stuff was broken in. Only thing I haven't done is I2s with the Buffalo directly off of juli@. Several things right off the bat:
...yes Robert you are right the Buffalo is heads and shoulders better than the Benchie.
...the unprecendented highs,dynamics and tonality I am getting is making me (yes its a cliche) listen to everything all over again. This is what being an audiophile who tweaks is always waiting for: something that breaksthrough that sonic limit barrier. I love it!
... the number of Lipo cells one uses for either juli@ or Buf32s has a striking impact on sonics. The amount and kind of bypass caps one uses on the Lipo's also has an impact. I use 2 (3.3 volt 2000 milliamp hour cells)on Juli@ and 3 on Buff32s. Don't know whether this is an absolute optimum but it suits my system best. Also I use 6-10K uf electrolytic caps, 30 uf of polypropylene and .047 uf polystyrene caps in parallel with the LiFePo4's.
Now here is the real biggy for me I'm listening to cplay @ 44khz and prefering it to 192. You guys know I have been an unabashed supporter of 192 with -145 src but not now. Listening to the Benchmark from 44 was painful, it needed 192 to sound decent. But the Buf32s is so smooth I prefer it at 44. The airiness is less but still abundant and the precise sense of placing images on a classical music piece is a lot better and the bass and mid incisiveness is undoubtedly better. I had my wife listen to both and she said she preferred 192 because it sounded more mysterious than 44. To me 192 now sounds too ethereal (too impressionistic) or imprecise but pleasant. I could still go either way since I am not declaring dogma here, I'm just saying this is where I am right now. When I go I2s everything may change. Needless to say my metallics are gone @ 44khz. I still get (but less frequently) dropouts like Edward but I have traced it to my method of transmitting battery power to the digital supplies. So I think I can eliminate it.
After all of the above cplay 30 is wonderful: less edgy, less bright, more tonally and dynamically correct than its predecessors. No I did not go back and listen to my new setup with cplay 29 and won't. I love cplay 30.
One further thought. Since my system has many degrees of adjusting tonal balance (triamped, electronic x-over on bass) when I make changes I always prefer to go the way of better dynamics then adjust the tonal balance accordingly. Cplay b30 has best dynamics yet. Way to go Cics.
Because I had been away for 2 weeks so the system took a few days to warm up before I can give a real listen to the new version. What I find:
-- smoother high
-- much better high mid frequencies. Such that a lot of the recordings with weak middle, e.g. Paul Simon's 'Graceland', are much more listenable now.
-- tighter bass, I think,
-- the sound space is much more expansive. So now for majority of the recordings the listener is IN the sound space, instead of sitting in front of it, with sounds coming to him in all directions; when previously it only happened to a few rare recordings with lots of space information.
Thanks again cics, for bringing back the fun of listening to recorded music at home. Nowadays, with cPlay, I find myself too busy listening to miss the sound of my long dismantled LP system.
Hi Cics,
Thank you (again) for your work and this new version. There are many subtle differences with the last version which all together make this version a very nice one. I’ll try to describe them.
There is a whole new level of details. Percussion, cymbals, brushes, ect have a very accurate ‘attack’ and are very very detailed. But without being scharp, hars or edgy. The combined titanium/fabric Seas dome tweeter in my setup easily tend to sound somewhat sharp and harsh. But no trace of that with this version !
The new level of detail also lets me hear ‘deeper in to’ the music (i.e. the mix / recording). Individual instruments are better separated and there is more soundstage and image.
This B30 is a very nice and enjoyable version ! Trance, dance, jazz, classical, al types of music sound very good !!
LynxL22 dig i/o XLR AES/EBU -> Lavry Black DA10 -> XLR Mogami Gold -> Klein & Hummel O300
Could not agree more. Again I am surprised at how you continue to raise the bar. A level of realism has emerged that is quite surprising. Particularly in female vocalists. I have listened to some of my favorite Jazz and now opera, and in both cases its not only a relaxed realism, but a larger measure of emotion is evident.
Each time you have released a new version, its like someone drops off a new higher priced transport at my door with a note saying here, try this!
Great work, and again thank you.
In terms of SQ, the latest cPlay release is the best playback software that I'm aware of (compared with pro audio software such as Wavelab, Sonar 8, Samplitude, etc). And thanks to the cMP^2 "package", my computer transport now puts all my absurdly (relatively speaking) expensive CD transports to shame.
Thank you for making "hi-end" audio more affordable and enjoyable.
Wonderful Ver 30. Etheric sound... non tiring... Best timing
Thank you Cics
Question for Cics! A conundrum my main pc had a MB flatline, I had located an e7200 and an S2l but know I question the purchase. I was going to swap out my PC audio asus and 6300 as my output is native digital 16/44.1, lynx l22 to a benchmark dac 1, but I was restricted from using sse4. My question is with the new platform should I wait and what cpu would you recommend with the new Nehalem platform. Everything was easier yesterday. Lastly my impression of b30, most noticeable was the sound stage was both deeper and higher, more of the recording stage was present(Paolo Fresu and Uri Caine/Think) T.
No worries!
...but I was restricted from using sse4.
My question is with the new platform should I wait and what cpu would you recommend with the new Nehalem platform.
Hello, cics, and everybody. I am glad, that with this version resolution comes to a new level with much of the upper mids' dip finally flattened. (at least in my humble setup). The upper bass also got a minor facelift and overall sound is back to the "snappy" level of b28, brass and vocals are very good, indeed. Placement of instruments more pronounced.
Every time new version appears, I wonder, how much is still hidden from us by conventional cd-players. I would compare this version with Wavelab demo as far as the resolution goes, only less grainy and without unnecessary sparkle. IMHO, this is the way to go.
Thank You very much. Also, very interested about forthcoming processor and corresponding mobo.
Serge.
I agree Serge, I too hear more resolution and better separation (especially noticeable to me with vocals).
Thanks again "cics".
Does this imply that the "iX" will become the new ideal platform for cMP^2?
New 32nm Core iX is certainly worth testing. The change made in cPlay could also benefit certain AMD CPUs.
Intel has made significant improvements in Nehalem and with the right BIOS we could see power consumption drop to just 10 watts doing full 192k with SRC@145db SNR. We should see Core i3 32nm CPUs in the next few weeks which can use DDR3 RAM (which is better than DDR2 using just 1.7V).
Could this be a better platform? Not sure as the current platform is magnificent.
Read the test results in the link. I think that the cMP^2 platform has a (soon to be) new champion. Imagine the underclocking/undervolting potential of this beast!
"We should no more let numbers define audio quality than we would let chemical analysis be the arbiter of fine wines." N.P.
Hi cics--still evaluating version 30 which is difficult because I just upgraded to a Buffalo32s and it is still breaking in. Anyway question on the Nehalem platform (Core ix CPUs) will they fit in the current recommended gigabyte mobo's or will they require new mobo's?
Core iX CPU's require a new mobo.
There are a few 'if this or that' on how they will be offered/implemented.
After a week of Buffalo 32s and cplay 30 I have to say I agree with the sonic attribute comments made by others. While I am not entirely free of metallics I can say they are much more infrequent than with the Benchmark. The depth of soundstage is extrraordinary and the high frequency timbre is a bit laid back (I think I will try 44 sampling rate again as Alan suggests). I find I am turning my electrostatic panels amp up to compensate and the thing is very very hot. I have been listening to 192 src -145. I don't think I have the optimum combination yet. But while I wait for my battery clips to run LiFePo4's for the digital power supply I will not make a final judgement. Still coming will be an I2s connection to replace my spdif.
nt.
Maybe my two cents can be of interest here too since I'm coming the same path that Theo has chosen (I have replaced my Benchmark DAC1 USB with, well, two Buffalo32S, since I more or less destroyed the first one...):For me the sonic difference, especially in transparency, depth and width of soundstage, but also tonal correctness and just being so "right" is HUGE compared to the Benchmarke, it just floored it in basically any aspect IMHO!
I must add that I run a fully optimized cMP2 system that runs entirely on battery with caps, and the Juli@ connects to the Buffalo32S via i2s (not SPDIF as in Theo's case, never tried this). BUF32S currently is also running on battery power, but for the time being not optimized at all since I am waiting for the TWEAKER version of it. I too feel that this is not the optimum yet, but it is exceptionally and so very clearly the best setup I had so far. Although BUF32S is a (very good) compromise to implement the ESS9018 chip, I can highly recommend it over many DACs and definitely over the Benchmark. (BTW I have not had any metallics in this setup, unlike with the Benchmark, although I meanwhile run the system at 140MHz and 0.76875V)
And BTW: It DOES make a HUGE difference in my setup when I feed BUF32S with 24/192 data upsampled with SRC instead of 16/44.1, it is so much better in again almost any aspect.
Regards,
Robert
Edits: 09/15/09
Hi Robert I think I finally have the Buffalo set up right and I am torn between 176 and 44 khz (src @ -145, hcc @ 145, cpu volts at .8 something). Running it at the higher sample rates (either 176 or 192) gives me an ethereal sound that is quite good in imaging but 44 gives me more incisesiveness while still being smooth in the highs, bass and lower mids are definitely better @ 44. But I don't have my battery clips yet to run my digital ps off LiFePo4's and I'm not yet I2s. The Buffalo is better in the mids and bass than the Benchmark and has more depth but if I didn't have trouble with cplay/Benchmark metallics I could live with the Benchie. I suspect my impressions will get way more favorable for the Buffalo once I further optimize. Metallics happen at about a 50% less frequency with the Buffalo (@ 176 that is, @ 44khz no metallics as cpu usage is 2%).
Speaking of further optimizing the Buffalo have you considered or done anything to tweak it?
Hi TheoInteresting how different our findings (or is it taste already?) are - I would not want to go back to the Benchmark now that I have got to know the Buffalo, and I wouldn't want it to do anything other than 24/192 :-)
No, as said I haven't done any optimizations or tweaks to BUF32S yet (apart from running it from battery which is not yet optimized in any way) since I'm waiting for the TWEAKER version to become available. This I think will be much more suitable for tweaks and optimizations since it (hopefully) will have no regulators, no output stage etc. thus being a much leaner and more precisely to handle component compared to the current BUF32S which is quite an integrated device (of course a very good one).
Well, that said, it may well be that I still start playing around with it and doing some tweaks and mods if the wait for the TWEAKER gets too long and/or I get too impatient, but I don't really have plans to do so yet ;-)
Regards,
Robert
Edits: 09/15/09
Still assessing. But so far equal to or better.
nt
thanks
This 2.0b29 release is the best for me by far!!!.
Until now, I have always found some problems in each of the different versions. Not about the resolution but on the purity of tonal timbres of the instruments, especially in the treble that I always seem too prominent and wheezing. I couldn't understand why.
Not with b29!. I believe that this is a wonderfully well balanced version between resolution and tonal purity.
Cics, I am nobody to tell you what you have to do. But I think, that now, as some have said in previous messages, it is time to develop a library interface to allowing us to develop diferents browsers.
And, also, I think that would be urgent that cPlay could read the embedded cue in the FLAC files. This would free the need to use cue files with flac, greatly simplifying the management of audio files. I think that this does not require much work: rather than read the content in an external CUE file, you can read the relevant section into the FLAC file. They have the same format.
Regards.
Ignacio.
I respectively disagree. I believe cics should do what he does best--seek out the next level of transparency.
How do you know when you have reached the final level.
OR is it a case of developing it until the sound no longer bears any resemblance to the initial recording.
To further complicate matters Vista is at the end of its cycle,7 is waiting in the wings,yet the software is still being developed for XP.
In its present state I fail to see much of a future,more so when there are other players that offer far more with comparable sound that will work on the latest platforms.
If the sole intention of cplay is to be a minority niche player then it has achieved it,but will not achieve much more.
Hi Flipper,
In its present state I fail to see much of a future,more so when there are other players that offer far more with comparable sound that will work on the latest platforms.
What players are you referring to that offer comparable sound?
You don't. Your comments reminds me of a quote I believe made at the beginning of the 20th century.....'all things to be discovered or invented...have been discovered or invented already...' or something like that. Well that did not pan out did it.
I will repeat my bottom line.
If the sole intention of cplay is to be a minority niche player then it has achieved it,but will not achieve much more.
With every new version it gets bigger,give it another year and it will be the same size as the software you now run down.
I use Foobar playing via Asio into an RME Digi96/8 PST then into an aria headphone amp feeding modified Sony CD3000 phones.
The detail/speed/resolution of this rig far exceeds some of your speaker setups.
For anyone to pick any variation in sound between the 2 players,they would indeed have to have Golden Ears.
It seems that those that consider themselves to be audiophiles are not happy to use a player that is mainstream,but will instead extol the virtues of a player where you can not even vary its size.
I am afraid that in its present configeration you audiophiles are welcome to it.
The reason others are not coming into the fold is because it has nothing to offer above others.
Complexity is not even worth a mention,if you consider it to be complex then you have never had much to do with PC's.
There are thousands of people like me that have been using PC's as there music source for years.If cplay had something to offer above the others it would be welcomed with open arms.
It has Great potential as a player,do not burden it down with a lot of settings that have no impact on the sound.
If these settings are having an impact on the sound,then they are altering the sound,you may as well use a SW equaliser.
Most of you would find more of a variation in sound quality by changing your cards around.
No one wants a player with bells and whistles,there are plenty of those as it is,but they do want a player that they can customise.
The Interface is everything as far as SW is concerned if you want it to be adopted by more than the few on this forum that use it.
Flipper,
If the sole intention of cplay is to be a minority niche player then it has achieved it,but will not achieve much more.
I am certain cics doesnt care. I dont think he wants it to be the next foobar or winamp. hell, he doesnt even seem to want any recognition for his behemoth of an effort.
With every new version it gets bigger,give it another year and it will be the same size as the software you now run down.
I use Foobar playing via Asio into an RME Digi96/8 PST then into an aria headphone amp feeding modified Sony CD3000 phones.
The detail/speed/resolution of this rig far exceeds some of your speaker setups.
For anyone to pick any variation in sound between the 2 players,they would indeed have to have Golden Ears."
Personally I doubt that cplay will ever get that big. Take a look at its growth:
Compare that to foobar:
Neither are that big but it does seem that foobar has grown more, and cplay comes with things included like asio upsampling, etc.
Maybe it is your headphones but I hear a difference primarily in the sound stage of Foobar, winamp and cplay, and that was a long time ago. Foobar and winamp were close and produced 2 d images and cplay was more 3d with more space between the images. I dont know, but it was pretty clear and I certainly back then was of the cover art gui crowd, but my ears told me that cplay sounded better.
It seems that those that consider themselves to be audiophiles are not happy to use a player that is mainstream,but will instead extol the virtues of a player where you can not even vary its size.
I am afraid that in its present configeration you audiophiles are welcome to it.
The reason others are not coming into the fold is because it has nothing to offer above others.
Complexity is not even worth a mention,if you consider it to be complex then you have never had much to do with PC's.
There are thousands of people like me that have been using PC's as there music source for years.If cplay had something to offer above the others it would be welcomed with open arms.
It has Great potential as a player,do not burden it down with a lot of settings that have no impact on the sound.
If these settings are having an impact on the sound,then they are altering the sound,you may as well use a SW equaliser.
Most of you would find more of a variation in sound quality by changing your cards around.
No one wants a player with bells and whistles,there are plenty of those as it is,but they do want a player that they can customise.
The Interface is everything as far as SW is concerned if you want it to be adopted by more than the few on this forum that use it.
All that you write is 100% true. But what you clearly are missing is that cplay was not designed to be a stand alone player!!! If you have only installed cplay and that is your only experience you will conclude as you did. However, cplay was designed to be used in conjunction with cmp and the principals in cics The Art of Building COmputer Transports.
In that context and with cmp, the interface is dramatically different. The file browsing window goes away and cmp is used to pick artists albums playlist, etc. It is a completely different experience.
If you open up cplay and move the mouse scroll wheel, it doesnt move from song to song, but with cmp it does. This is just one of many examples.
I thought the cplay interface sucked and sizing was just one of the many things wrong with it, but with cmp and with the principles of the whole build, the interface is great. There is no need to change it IMHO. Try using foobar or winamp or album player SOLELY with a remote. I certainly couldnt get it to work, but cmp2 totally works with the mouse as a remote and no need to move it around like a mouse....just the scroll wheel and the left click button.
Build a cmp2 and you will see for yourself, and it will be easier to hear differences. Heck just getting the extra psu for the hdd and usb is huge for sq.
And whats up with the audiophile hatred?
I agree with theob; cics should do what he does best with his product.
I use .WAV and FLAC users could write their own interfaces if they desired.
The pursuit of SQ should take precedence over GUI and other user refinements. Someone could step-up to the plate and assist cics in his endeavor.
RayBan
Hey Ray,
I agree with you and theob and I was once an eye candy guy who (LOL) wanted pretty pictures.
However I would love it if cics took a break and fixed the documentation or updated the cmp2 website.
I am a huge fanboy but the documentation is so out of date now that it really is hard to recommend cmp2 to anyone but the savviest computer person, and even then you have to keep checking for the latest settings.
Most probably dont care because they have their cmp2 rig running, but man there should be alot more people embracing this approach and the documentation stops them...
I struggled with that very fact.
I started with no computer skills, few tools but a love of music, so I ventured where most would not go.
A few kind souls, (Ryelands, seger, GStew, AlanJ and RickMc) provided great assistance and guidance. Daily (hourly) checking at AA was also of great help as the search feature allowed me to go back in time to see how others had solved their individual problems.
I agree we all owe cics a huge debt. I wish I had skills that could be useful to him but sadly I don't
I can say I started from scratch, crashed and burned as others have, got up and continued onward. I am certain dedicated others will join the challenge and do the same!!
RayBan
I also agree with theob that SQ is of prime importance and everything else is secondary. Unlike some users here I have no particular loyalty to cMP/cPlay but I do want the best and I sincerely appreciate all of cics hard work and dedication.
I thought that new documentation at the end of the year (December 2009) would be timely since Windows 7 is officially out in October. Actually cics should be asking for contributors instead of shouldering the entire burden himself.
With some volunteer help, it's looking much better and hopefully would be done in a month or so.
Awesome.
If you want I plan to build my rig this week and wouldnt mind some proofreading if you need it. I was going to do that anyway for guide I will be using.
Of course, SQ is the first. But, what is SQ?. Transparency?. I don't think so. I believe more in naturality and fidelity, then less distortion and tonal balance.
Resolution and tonal balance is the key. You prefer resolution but I think that the key is the balance between both. And this b29 is the most perfectly balanced of all versions.
My main source of music is live music: classical, jazz and opera. When I listen a live concert, sound is not so "microscopically" transparent as we expect from our equipments. This level of transparency is both unreal and unnatural.
RayBan say: "I use .WAV and FLAC users could write their own interfaces if they desired.".
Are you sure?. How?. It's impossible write an interface if we don't have a interface library.
In Spain, we have a proverb, it could be translate as: "The best is the enemy of the good". That say that an obsessive search for the best never allow us to distinguish the good things that have already achieved.
Of course, the title of my last message was a friendly nod. But I really believe that the SQ goal has been reached with this b29 version. Then, what I proposed to Cics was a reflection moment. I believe that is time to think about everything that has been done so far, the goals achieved and what is really needed to get that this great project might be useful to everyone.
Indeed, as mentioned above, a new documentation would be good. Also, this thread and the cMP thread are excessively long. A new thread with a compendium of the essential from this threads is also necessary. And, of course, the interface library to develop news users interfaces.
I am not criticizing anything, quite the contrary. I am only asking for think the way forward from now.
Regards.
Ignacio.
"naturality and fidelity, then less distortion and tonal balance": nice recipe for SQ.
"The best is the enemy of the good" - what is the original Spanish, if you please, it's a real gem. As you explained it, it is a perfect antidote to the obsessiveness of audiophilia!
"what is really needed to get that this great project might be useful to everyone" - and I think the essence of this would be more flexibility and choice for the user, in terms of both functionality and interface. You have the programming skills that could really help, so I hope you get a chance to put them to use on cMP and cPlay.
.
You are right!.
The original sentence is: "Le mieux est l'ennemi du bien". In spanish we usually say "Lo mejor es enemigo de lo bueno". This is a sentence so much used, and has become a proverb. I know that is also very used in Argentina.
Regards.
Ignacio.
Thanks for the Spanish - sounds best to me. Voltaire may have been the first to use it in print, in Italian 8 years before he used it in French ("Il meglio è l'inimico del bene"), but I wouldn't be surprised if it was already a popular expression that he picked up.
I agree the thread is quite long and a new one should be considered.
If one were to follow cics instructions attached with each new release, the documentation necessary is mostly there.
I used the search feature to locate info I needed (as well as kind assistance from Ryelands, GStew and others) to solve problems I had as a newbie.
RayBan
Does anyone know if the file browser in cPlay supposed to be functional when the computer is fully optimized? Mine goes into an hourglass and stays stuck in the File-Open window when I click on the file folder in the cPlay main screen.
Daniel Gauthier
Montreal Canada
Gigabyte GA-G31M-S2L, E7200, 2 x 1 gig HyperX LL Memory, Antec SLP-450WR PSU, WinXP sp3 on Compact Flash Drive, Jili@ Analog Out direct to Electrocompaniet Ampliwire II Amplifier, Apogée Convergence Speakers
Untick the "cMP" option in cPlay, untick both "Suspend" options in cMP (svchost and lsass) and disable cicsRemote. Then you can use the file browser in cPlay from within cMP.
There is a slight reduction in SQ because of the unused "Suspend" options. cicsRemote doesn't affect SQ, but it ruins most of the hotkey functionality of Windows and media players (apart from cPlay of course).
I'm not willing to compromise SQ so I guess that option is a no-go.
Daniel Gauthier
Montreal Canada
Gigabyte GA-G31M-S2L, E7200, 2 x 1 gig HyperX LL Memory, Antec SLP-450WR PSU, WinXP sp3 on Compact Flash Drive, Jili@ Analog Out direct to Electrocompaniet Ampliwire II Amplifier, Apogée Convergence Speakers
Yes if you are not in cMP mode. In cMP mode, it hangs.
There is an option in cPlay for 'cMP'. If you select that, file browse is disabled.
Were you in cMP mode or regular Windows when you tried to browse?
Greg in Mississippi
I was in cMP mode.
Daniel Gauthier
Montreal Canada
Gigabyte GA-G31M-S2L, E7200, 2 x 1 gig HyperX LL Memory, Antec SLP-450WR PSU, WinXP sp3 on Compact Flash Drive, Jili@ Analog Out direct to Electrocompaniet Ampliwire II Amplifier, Apogée Convergence Speakers
I have a 24/96 flac of about 10 minutes, for some reason as part
of the album cue sheet cplay can't load it, though as an individual file
it is loaded fine.
I can solve this problem by adding an index to the cue sheet to "cut"
the song, but, wouldn't it be better if this will be done automatically
by cplay (as it seems to be doing to the individual file)?
Oh, almost forgot, b29 is great, thanks for all this hard work.
Settings:
192k, SRC 145db, AWE, no timer
I have seen a few folks listing the PHASE they have SET their cPLAY "to"
and must make a comment.
PHASE is not a setting like the others, well, maybe some folks do change ALL for each recording and for those with such an intrepid nature you have my full respect. BUT, PHASE is different. What other setting can be selected from the remote mouse? PHASE is absolutely dependent on the recording you have selected.
Sometimes the PHASE is vague and either setting sounds the same as the other but there are many recordings, the truly good recordings will reveal a best setting. One does need to learn to listen for this but one will learn and hear the difference. It does make a difference.
This is not, ideally, a set and forget setting. For the best sound it must be USED.
I feel much better now.
Bye,
Rick McInnis
Yes its different per recording. I would think it also helps when a system component (DAC, amp or x-over) that inadvertently inverts phase in which case a phase "bias" tends to be preferred. I hope this makes sense.
it really does not matter how your components are handling phase or what is the resulting phase at the end of the chain, the recording determines the final best position for phase.
Most loudspeakers have phase confusion within themselves which some would say eliminates the need for phase switching. In my experience, there is always a best setting for most anything but Phil Spector type productions.
If you have not sensitized yourself to this it would not matter where you had the phase switch since it is going to change back and forth with individual recordings regardless of the selection of components for playback.
It is a worthwhile skill to learn. Don't think this is one of those A/B things. One must, at first, listen for awhile in one position and then try another, going back and forth. You will eventually hear the difference. By this, I do not mean you are going to "get it" the first listening session. I usually find my preferred position early into a recording but many times I decide I had it wrong and change back. And sometimes I do not find a best position at all. The vast majority of recordings do have a best position.
You will only experience anything approaching absolute phase by learning to hear it. Listening critically to voice is the best place to start. We hear voices in phase all day long.
Bye,
Rick McInnis
Hello, and thanks for clarification. I'll try today to find my best phase for the Jethro Tull wavs that I use to evaluate new versions of cplay. I know the songs' every tiny "bit", so, I hope, it would be relatively easy. Before I read Your post I thought it was some strange exotic add-on.
Serge.
That the cMP2 setup had a non-intrusive phase adjustment was one of the things that attracted me to it. Back when I played a lot of vinyl, I setup dual inputs on my phono stage, in-phase & out-of-phase. Each album got marked with a preferred setting (right alongside the preferred SRA setting... but that's another contraversy).
To learn how to 'hear' phase, I recommend starting with some recordings with single singer or instrument. I find voice and drums to be very easy to hear correct phasing on. Listen, flip the phase, continue to listen. You'll find a preferred setting. Then try more complex recordings... but be prepared to be frustrated. Many pop recordings are not phase-consistent... for example, the lead vocals will be in phase along with the drums, but the piano, guitars, bass, and backup vocals will be opposite. They seem to be better nowadays, but many of my 70's & 80's recordings are like this. And this is why learning to hear phasing works better with simple recordings.
This is a key parameter to getting it right in my book.
Greg in Mississippi
Hi, I just discovered this site by the friendly advice of another HiEnd guy from my city, and reading through the endless postings about cMP/cPlay I have become very curious, I have decided to try to build one myself in order to bring my mostly digital playback closer to true audio fidelity.
Here is my planned configuration:
1. Motherbord: Gigabyte GA-G31M-ES2L (bought it today new)
2. RAM: 256 MB DDR2 533 Infineon HYS64T32000HDL-37-A (is this one possible?)
3. HD: WD Scorpio Blue WD5000BEVT (SATA300), 500GB, 2,5"
4. cpu: Intel E7200, maybe boxed first in order to try the OEM fan first.
5. PSU: Antec Earth Watts 430W
6. ESI Juli@ or even my older LynxOne via S/PDIF to my external DAC.
Not yet sure about the following:
7. Case: Don't want to invest much before not being sure that the whole thing sounds really better than my old setup. So would it be possible and save to leave all "naked" first, I mean PSU + MoBo with CPU and RAM + HD outside any case? Or does it need to be "grounded" or however it is called (I do not know much about this stuff)? Otherwise before investing into a Zalman 160XT or such, I'd rather take a second hand HTPC case from Ebay for 50 or 70 bucks.
8. DVD drive: I haven't yet understood how to transfer all those Gigabytes of music to the cMP: Better with a DVD drive or better by USB2 or FireWire? I have read that some people disable both - but then how to connect to the cMP to the normal PC?
Of course, the most convenient would be to use a USB- Stick with 8GB or so. But I have understood that USB is better disabled...
All the other tweakings like more than one PSU, battery powered soundcard etc may follow once I hear that the effort starts to pay off in terms of SQ.
OK, if you see something that in your experience will not work well (hardware related), I'd be glad if you'd reply.
Thanks in advance,
Bernie, from Berlin
1-6 is fine. Just remember that initial install of Windows requires more RAM (1GB should be good). Only switch to 256MB RAM once Windows optimisations are done. Also, install Windows on a small partition (5GB will do) and the rest of your drive (495GB) becomes another partition where you store your music library.
7. No need for an expensive case, some use a wooden box that offers sufficient venting.
8. It's best to rip using EAC in secure mode on your home PC and transfer files to your cMP² via USB drive, memory stick or temporarily remove & connect drive to sata port.
Hello all,
It has been a long time since I’ve contributed here, but I’ve followed along and enjoyed most of your suggestions and the improvements made to cplay over time. Cics may remember me as suggesting process explorer, autoruns, and some random windows settings a couple years back.
I just installed minlogon for the first time (finally), and although the transition seems to have worked and the benefits to cplay are obvious, Zoom Player will not play my media files. All I can do is launch the program and select a file to open; and that is when the program hangs until I reboot. Does anyone else have zoom player functioning with minlogon? I use my CMP for watching movies just as much as listening to music and would like to be able to use only one Windows partition with minlogon for all media types. If I must switch to another application, what else are you using?
Thanks,
Mike
Hello Mike,
It's good to hear from you and congratulations on Minlogon.
With all my old parts I built a separate dedicated cMP HTPC with minlogon - zoomplayer works intermittently. I settled with PowerDVD. Works for my basic needs.
I have two power supplies I bought directly from granite digital a year ago, but I’m wondering if they are not the same one everyone else is using and whether I received the correct product. It only powers up one type of drive I own. As soon as I plug the sata end of the power connector to my Western Digital 3.5” 400GB hard drive, the drive starts spinning. I then turn the power on to the computer and everything is fine. I assume this is normal operation. My other drives buzz very faintly, but do not spin up. So I’ve been using my computers power supply to run a drive for all this time (a carnal sin by now)
Can someone answer why none of the following drives power will up:
Laptop drives:
Fujitsu MV2120BH 120GB Rating 5V - 0.6A
Fujitsu MHV2200BT 200GB 5V 0.55A
3.5” drives
Western Digital WD10EACS 1TB 5VDC – 0.7A, 12 V – 0.55A
Western Digital WD5000YS 500GB 5VDC – 0.7A, 12 V – 0.75A
The power supply I bought from Granite Digital says FINENESS POWER power supply and has some German writing on it as well along with these specs
INPUT 12V 1 AMP
AUSGANG 12V - 2000mA (Ausgang means exit)
Output 5V – 2000mA, and I thought this should be sufficient for any of these drives.
I still have the invoice and it reads:
Serial # 7130 Universal Drive Mech Power Supply $19.95 each
So, shouldn't these drives power up given their power requirements being lower than the power supply limit?
More importantly, did I receive the wrong product?
Thanks,
Mike
Thanks,
Minlogon wasn't really as hard as I feared, but I wasn't about to try it unless I had some backup software and a few extra drives to play with. Getting married and having a kid plus changing jobs just got in the way. Still can't update the hardware though.
I think I'm actually getting zoom player to work after all. I think the problems I am intermittently having are more to do with my backup program and zoom player (which I have to completely reconfigure..along with the windows settings). Every time I reboot I see some different things.
I going to post another question about the power supplies I bought direct from Granite digital (even though they say FINENESS POWER power supply on them)
They don't power up my laptop hard drives or my WD 3.5" drives except for one 400 GB drive.
I thought I read almost all of the cMP/cPlay posts but either I missed or forgot what I read.1. What is the difference between cMP and cMP squared (cMP2)?
2. I have cMP/cPlay running on two different computer that have extra memory in case my or your program needs change. I run the sse4b9 version on a Intel Duo Core 2 E7400 with 4 GB ram. I run the sse2 version on an AMD Athlon 64 x 2 Dual Core 4400+ with 2 GB ram. I don't know why I chose the sse2 version for the AMD and I'm not sure that the sse3b9, ssse3b9 or sse4b9 versions would not run. The Readme installation text mentions a CPU-Z program to determine which version will run but I don't recall seeing that program nor a clear explanation of what to sonics expect with each version.
3. Likewise for many of the other settings, it seems hard to choose settings without established or suggested recommended settings and what each setting is likely to bring to the table. Thus in your experience do you prefer the highest upsampling or no upsampling, a small or large buffer, AWE on or off.
I sometimes like to explore as much as the next guy but sometimes I like an established benchmark upon which to base all mods and tweaks. So can you give us your best and perhaps update the December 2007 document by the end of the year.
Edits: 08/08/09
I have a more or less comprehensive "check list" for setting up cMP2 systems (cMP2 = cMP squared, i.e. cMP and cPlay working together).
Some others on the list say they found it useful. If you want a copy, e-mail me and I'll send it.
Regards
Dave
Refer to ReadMe.txt for important details & specifications.
Change Log (2.0b29 release):
- DSP refinements
- Gapless processing enhancement
- Volume processing efficiency improvement
- DSP buffer optimisation (when using AWE)
- SRC efficiency improvement
- Other minor changes
- Minor ASIO refinement
- UI timer processing improvement
- Fix for Windows Kernel error (in memory management)
- Increased tracks/splits per cue sheet to 160 (from 120)
For best results, uncheck "Timer" setting and test different "Buffer" sizes. Details on SoX settings and output measurements can be found under Release 2.0b26 Notes .
Please REMOVE previous versions before installing cPlay 2.0b29. Normal (SSE2), SSE3 B9, SSSE3 B9 and SSE4 B9 versions are available. Running cPlay on a CPU not supporting the required instruction set will cause cPlay to exit immediately.
For cMP² users, review these BIOS changes (underclocks FSB, CPU & disable additional USB items). New settings are in effect allowing for lower power consumption (below 20W). This release allows for further CPU underclocking under maximum load (192k SRC 145db SNR output) where CPU Host frequency can be lowered to 140 giving CPU frequency of 840MHz! This setting is only recommended for WAV files (for FLAC, use a minimum of 150). Recommended Gigabyte mobo GA-G31M-S2L is no longer available, use either GA-G31M-S2C, GA-G31M-ES2L or GA-EG45M-UD2H ( advanced settings discussion ).
I have a question about the choice of mother board since I am going to build a dedicated music server per cics's recommendation. Do the boards GA-G31M-S2C, GA-G31M-ES2L have the same flexibility in BIOS configuration as GA-G31M-S2L? And if yes, do we need a special version of BIOS?
The GA-G31M-ES2L does not have the option to turn off Spread Spectrum,otherwise all the other settings are available. My testing suggests that option would be useful, but not critical... you can setup a good cMP2 with the ES2L, it is just a little better with the S2L and Spread Spectrum turned off.Greg in Mississippi
P.S. Don't know about the S2C.
Edits: 08/16/09
This is very helpful
Hi again,
I'm having some issues with b29 that I never had with previous versions (I was using b27).
The problem is strange and I don't know how it happes, but after a while of playing music and loading files I get this strange metallic sound. It's not just tics or pops, it's all over the place. I tried quitting cPlay and reloading the CD but it was still there. When I played with the settings (I switched from SRC145 @ 192 to SRC145 @ 44.1) the metallic was still there and really louder. Same problem switching to SoX.
The good news is a reboot solved the problem!
The first time I got the problem was the day I installed b29 (ssse4), the metallic appeared after 2 or 3 CDs. At the time I thought I had to reboot due to the new install so it didin't bother me. But then, a week or so passed, I played over 30 CDs, and this morning, it was back?!?
It's not a huge problem since rebooting seems to do the job but I wonder what might be causing this.
Anybody else had this issue?
Thanks
Etienne
Yes I guess you can say I've had it quite a bit with my Benchmark usb dac (which I connect via bnc cables to juli@). If you listen to 96 you should be ok. I've tried for 8 months to get rid of it and while I have ocassionaly gone a few weeks w/o it, it always comes back at rates of 176 or 192. I have just ordered a Buffalo 32 which I hope (when fed with a 96 khz stream or lower) will fix this.
Cics and others believe it may be an interaction between juli@ drivers and certain dacs or that it is just a juli@ problem. While it may very rarely go metallic with juli@ analogue outs it is easily fixed with a restart of the track (no reboot is necessary).
Hi Theob
I'm glad that we're two with the problem. Maybe together we'll be able to fix it!!
I really doubt it's the DAC because a reboot of the PC fixes the problem... it might have something to do with the RAM. Next time the metallics appear, i'll try to clear the RAM without rebooting the PC just to see if that helps!
By the way, this issue never occured with b27 which I had running since its release! Maybe Cisc can point out what were the major changes in the code between those two versions.
I too am using the BNC connection of the DAC1 (via a RCA to BNC cable (using the breakout cable with the Juli@)). Have you tried the Optical connection? What drivers revision are you using.
I'd really like to know if anybody else is using the Juli@ paired with a Benchmark DAC1 and experiencing the same issues. If others are using this combination without any problem then we'll be able to cross out the DAC, or even the Juli@ and we 'll be able to compare our configurations. I'm also curious to see if anybody else is using the digital out of the Juli@ at 192 paired with any DAC whatsoever.
Thanks
Etienne
Yes, I am using the Gstew modified Juli@ on a BNC to a Benchmark DAC 1 Pre at 192. Just installed 29 and ran it all night before my trip.
I'll run it again when I get home late tomorrow.
Got stuck with some counterfeit mem (256) which turned out to be only 155 so cPlay wouldn't play. Had to insert my Kingston 1GB XLL back
RayBan
Ok thanks for the info.
Could you please play as much music as possible, without rebooting the computer, and give us feedback next week?
Just let us know if you get this weird metallic sound.
Anybody else with this setup?
Thanks
Etienne
Have you tried the Optical connection?
Yes it doesn't fix it.
What drivers revision are you using.
1.05
I have tried everything: New hd (cpu, ps, juli@, hd's, mobo), reloading windows--all of which did not solve the issue.
I do know that I had less metallics with cplay 27 and no metallics with cplay 18. But since 96 src upsampling never gets metallics and since cplay 29 @ 96 sounds better than cplay 18 @ 192 I leave it at cplay 29 @ 96.
Hi Theob,
Today, the metallic sound came back again!
I really wanted to figure out why it happened and how to fix.
I tried clearing the RAM ... the metallic sound didn't go away.
I tried to exit cMP+cPlay and reload everyting ... the metallic sound didn't go away.
I tried to load the file via cPlay only ... the metallic sound didn't go away.
I went in DEVICE MANAGER, disabled the Juli@ Controler and enabled it ... the metallic DID go away!
I'm pretty sure it has absolutely nothing to do with the Benchmark DAC1, since disabling and enabling the Juli@ did the trick...
Maybe it's a driver problem with the Juli@...
Anyways, I'll do more research the next time the metallic occurs.
Thanks
Etienne
Very good analyses! I have just had it though and have already ordered the Buffalo. Good luck.
So did I but it's still not here yet!!!
RayBan
Hello,
Thanks a lot for this amazing release!
I really see an upgrade over the last version I used (b27)!
Plus, this release completely solved my "Not Enough Memory" problem. I had an issue with SP3 and 256MB of RAM. My GA-EG45M-UD2H has 34 MB of "Shared Memory" and there wasn't enough available memory to play a file via cMP/cPlay, my workaround was to kill ciscRemote everytime I loaded an album. But now, there's plenty of available memory and everything's working like a charm!
My settings are SRC 145db @ 192k.
Timer disabled
AWE enabled
Buffer size small
I'm still wondering what level I should set the volume control! Does it affect anything? I didin't hear any difference but I didin't do an exaustive testing. Should I set the level to 0 and control the volume via my DAC?
One last thing... What's the impact of each version on CPU load?? Should I use sse4 just because my CPU supports it?
Great Work!
Etienne
Thanks again!
cPlay's digital VC offers 53 bit resolution (64 bit double precision) throughout. I use this as my main VC over dCS' Scarlatti DAC (digital VC). In your case, I would test which VC to go with, but definitely avoid using both (i.e. one of them should be at full volume). If your DAC does digital VC, use cPlay instead.
CPU load definitely changes a little although same amount of MACs (Multiply Add Computations) are done. Best option is SSE4 when your CPU offers it as this employs a key CPU instruction in the most critical area of cPlay.
Shared memory (34MB) on mobo can be lowered - check your BIOS under video setting ("On-chip Frame Buffer Size") under "Advanced BIOS Features" (I think) and set to 8MB (note: do NOT use the lowest 1MB option).
Thanks for the info.
And i've looked everywhere, there's no "On-chip Frame Buffer Size" for the GA-EG45M-UD2H motherboard. At least not in the latest bios rev. F3.
Anyways, it doesn't really matter because I don't have the "Not Enough Memory" error anymore, even with 256MB of RAM.
Etienne
Curious... why not use the lowest 1MB option for the On-chip Frame Buffer Size?
TIA!
Greg in Mississippi
One of the best Version of cPlay. It feels as if I upgraded my amplifier.
Cannot decribe in words but music sounds as it should sound.
THank you Cics for this Jewel of a Audio Player
Firstly, thank you for the feedback - its excellent. B29 is a breakthrough (again)! In particular, violins and piano are rendered beautifully. There's great balance and emotional energy. I was hesitant to do this release given the many changes done/required. Anyway, this after b25 where I thought more improvements would be unlikely/slow...
I need to know what is your preferred settings: output rate, buffer size, resampler (for SoX other details as well), phase (0/180), awe & timer setting.
Well I've been playing with a "loaner" RME Fireface UC for the past month, so these are my (finally) definitive results based on that playback device.
-Output Rate: 192KHz
-Buffer Size: Small
-Resampler: SRC @ 145db
-Phase: 0
-AWE: Yes
-Timer: No
A couple notes about the RME Fireface UC. It is similar to a Fireface 400 only with a USB input (rather than a firewire input). The USB protocol is BULK mode (as is their firewire version) which is supposed to have similar benefits as asynchronous mode. And contrary to many beliefs, that USB is limited to 24/96, this does 24/192 no problem. With one caveat. I think the USB implementation is too CPU intensive, because I was not able to keep the FSB @ 140MHz with the CPU speed at 840MHz when playing upsampled 192 (with SRC @ 145). I could keep the CPU at 840 when playing upsampled 24/96 (with SRC), and could play 24/192 upsampled with SOX at this speed as well. However when playing 24/192 upsampled with SRC @ 145 I had to increase the FSB to 150 and the CPU at 1.2GHz. But playing 24/192 (with SRC @ 145) was so much better than anything else that it was worth it.
And "cics" did not ask this, but I have a couple additional "tweak" preferences to add: I prefer to set the SATA mode (in Integrated Peripherals section of BIOS) to "Enhanced" rather than "Non-Combined". This requires that you install Windows with the AHCI driver. And finally, I formatted my one HDD with two partitions. The OS partition is default, but the music partition is formatted with a cluster size of 64KB (rather than the default 4KB - and I also tested 16 and 32). Both these HDD "tweaks" result in a deeper soundstage with better separation and three-dimensionality. IME.
Cheers.
B30 is in the works and should see some CPU load reduction but I doubt it would help your USB setup.
I've tested 64kb block size but not with AHCI driver under "Enhanced" SATA mode - this is definitely worth trying!
Another very fine release. thank you!
My preferred settings are:
44.1
buffer large
SoX VHQ Phase-minimum, bandwith-97 alias off
Timer on
AWE off
Phase 0 volume 0.0
running Toshiba laptop with CMP, winxp with all the recommended mods
asio4all out to a peter daniel NOS DAC==> lightspeed attenuator ==> Peter Daniel Chip amp; no interconnects.. hardwired with silver and gold
I enjoyed version b28 so much and I thought that it is not going to be advanced any further in the near future. I was so wrong. Thank you so much Mr. cics for enhancing what is already a very fine player. b29 brings more detail and refinement comparing to b28 such that it is like seeing a better focused picture. Left to right balance is better and a wider sound stage. Better dynamic too. A significant version indeed.
My setting is :
Output -- 48K (the max acceptable by Devilsound DAC)
Buffer -- Small
Unsampler -- Sox, VHQ, Intermediate, 97, alias
No Timer nor AWS
btw, I have never been able to set AWS because I could not get lock privilege following the procedure in your document. I cannot find "User Rights Assignment" after hitting the 'add' button. Any indication why? I am using sp3 instead of sp2.
This setting can only be done under the "Professional" version or in Vista's case, "Ultimate" version.
Navigate step 7's tree hierarchy until "User Rights Assignment" is available in left pane of MMC editor.
Now I know why. I think its because I have the Home Edition of Windows XP. So the AWE option would have to wait until I replace my temporary server (running on an Aspire netbook) by something more substantial in a couple of months time.
Thanks.
It works like a charm.
Daniel Gauthier
Montreal Canada
Gigabyte GA-G31M-S2L, E7200, 2 x 1 gig HyperX LL Memory, Antec SLP-450WR PSU, WinXP sp3 on Compact Flash Drive, Jili@ Analog Out direct to Electrocompaniet Ampliwire II Amplifier, Apogée Convergence Speakers
- Output rate 44,1, SRC 145 db
- Buffer small
- Fase 0
- AWE On
- Timer OFF, also I often minimize the window during playback and switch off the monitor.
When I start CMP I have task manager running and as I don't use my Zalman now and have a monitor, I kill sicsremote and set cmp priority to realtime.
This version is very clean and the sound is getting lighter with the last 2 versions.
Thank You very much. Serge.
I need to know what is your preferred settings:
- Output rate 44,1 (Sabre ES9018 DAC seems to be insensitive for upsampling)
- Buffer large (more emotional)
- Resampler none
- Fase 0
- AWE OFF (More listening is needed for prefering 'on' with Cplay B29)
- Timer OFF
need to know what is your preferred settings:
output rate:
prefer 176 or 192 although I am very happy with 96. I tried going back to cplay 18 which never goes metallic (@ 192) but cplay 29 @ 96 is way better
buffer size:
prefer small altho could go with any of 3
resampler:
prefer src @ 146
phase (0/180):
prefer 0 but 180 works on some files
awe & timer setting:
prefer both on
I meant to comment yesterday, but...This version is just superb. As someone else said it is more airy and detailed and yet the music is more intense and affecting with imaging that is quite real. Congratulations, cics, and thanks.
Hi Cics,
Personally I like this version over v28 alot because (1) I feel less bass booming, and therefore more airy and details. (2) Emotionally when I listen to my favorite track and I can feel the mood of the singer much easier, that I had goose bumps when listening to the song. Congratulations.
Hi cics, thanks for this other very fine release.
Just one concern though: my "available ram" has decreased dramatically from previous version (b28).
I use one 512M stick and the maximum track size I can load dropped from about 9 minutes to 5 minutes for FLAC @ 44.1. and about 2 minutes for 24/96.
Is it due to a new way of treating data, and why you increased the track/split to 160 ?
Should I get a 1GB stick or is it just fine this way ?
Sonically speaking, b29, so far, is a new breakthrough.
Thanks
Martin
No need to go with 1GB RAM. Track split switching is done after 5 minutes playback. I would recommend going smaller to 256MB (split size is ~2:15) when media is largely 44.1k. This switching doesn't interfere with playback however flac does demand heavy cpu load and it may be a good idea to increase cpu speed (from low of 840MHz). Its better to increase from 6x to 7x instead of increasing cpu host frequency (leave this to lowest stable value ~140).
I use 256MB RAM.
A new memory model is used in b29 (to avoid some low level Windows kernel issues).
Thanks for the fast answer.
I'll try to get my hands on a 256 ram module, hard to find now, I fear.
My favorite settings for now: CMP2, 96k output(max. for my DAC), I hesitate between small and medium buffers, SRC@145dB, AWE enabled, Phase 0, no timer.
Phenomenal sonic balance, tight bass and smooth highs. And incredibly large soundstage, out of this world !
Is CPU speed X6 enough for 96K output, if I get no glitches ?
Superb work, once again, Mr. cics
Thanks again.
Martin
Is CPU speed X6 enough for 96K output, if I get no glitches ?
Yes, x7 would be needed for 176k or 192k output.
This is the first version of Cplay where i experience a benefit with AWE turned on. In previous versions AWE was not my favourite setting. AWE gives me now more precize sound, without getting edges.
Thanks Cics for your continously refining the code
I was able to get metallic free sound (@ 192/145/small buffer) from cplay 27 but I am not so fortunate with 28. Tried everything from slowing clock and lowering upsampling to 176. Even tried listening uniquely to music files off of flash drive. For whatever reason while listening to anything higher than 96 I get 40-50% cpu usage which always goes metallic. At 96 khz I get 25% cpu usage which is stable. 176 begets 40% cpu usage which varies + or - 4% (this is the most stable greater than 96 cpu usage I can get). Anything that can improve efficiency on future cplay releases would be appreciated.
For me as a new user of cPlay I´m interested in the related hardware. Which Intel CPU do you use? E7200, E7300 or E7400? Which soundcard?
Ruediger
All by the book...Intel 7400, juli@ soundcard plus all other recommended items. Only things I don't use are wireless mouse and touch screen hardware.
I´m new with cPlay and still a foobar user with a large music archive. The music has been ripped by EAC in flac format for single title without cue sheet. Is cPlay able to play flac files without cue sheet?
Ruediger
Ruediger asked:
Is cPlay able to play flac files without cue sheet?
Not individual files as shown in your screenshot but Al Jordan's Recursive Cuesheet Creator makes cPlay- and Foobar-compatible cuesheets for the largest of libraries in (literally) moments. See link.
Dave
Ingenious it works. The Recursive Cuesheet Creator is a fine tool. :)
Hi,
In my CPlayList Editor application, I have written some functionality to control CPlay across the network. Within my application, I create a run-time instance of CPlay.exe with the temporary playlist as an argument. This works fine.
As part of the handle to CPlay, I could send information to Cplay if it supported a standard input stream. It doesn't appear that CPlay supports an input stream. Instead, I've been able to control CPlay by creating an instance of Java's Robot class, which basically allows a person to send key events and mouse events to Windows. I've been able to add support for next track, previous track, fast forward, rewind, volume up and down, pause / play and phase by sending the appropriate key events to the CPlay application.
This only works when I am controlling CPlay over the network, because the Robot class only sends key strokes to the windows application that has focus. When I send a playlist to a remote machine, my application creates an instance of Cplay on the remote machine, and CPlay naturally has windows focus at that point. I can send whatever keystrokes I want to the application through the Robot class. Obviously, it doesn't work when running locally because my CPlayList Editor would have focus instead of CPlay. It doesn't really need to happen on a non-networked setup because CPlay has the same controls anyway, so it is a non-issue.
However, even across the network, I don't the Robot class is a very graceful manner of controlling CPlay. The reason being that if some other system event caused another window to appear and get focus, my application would then send the keystrokes to that window instead of CPlay. This hasn't happened yet, but I imagine it could. There is no manner in Java to get a list of running non-java windows application and have an application grab focus.
So, is there any other manner of getting information to a running instance of CPlay? It would be perfect if CPlay had standard input stream capabilities, as I could then send the same keys to the input stream instead of the inherent ugliness of using the Robot class. In this manner, I would know if my handle to CPlay still exists, and wouldn't have to worry about window focus or sending keystrokes to the wrong application.
Thanks,
Alan
In your case, I imagine it would be combination of FindWindow and SendMessage with WM_SETFOCUS.
I'll need to give this some thought on how best it can be achieved. cPlay uses direct keyboard input. Standard input stream may be the best way to do this.
In cMP mode with full optimisations and Minlogon, there's no chance for other dialogs grabbing focus (cMP hides until cPlay exits). Only exception is when an error occurs.
cics,
it would be good that all user interface's services (displaying and keyboard functions) were completely separated from the main engine of the application and to communicate them via a messaging API -as ussual in Unix aplications-.
This will facilitate us the development of new interfaces, without having to worry about the main engine.
Regards.
Ignacio.
Regards.
Ignacio.
Good idea. Skin is a separate component and a simpler version can be implemented as you described, i.e. a simple DSP message processor. This could be cPlay as a DLL version. Your thoughts?
I believe that cPlay must work as a service, getting his usual configuration paramaters from a configuration file and, while running, wait for messages from the user interfaces. A DLL must be a static component, not an active component.
The first step would be to fix the functions of cPlay and the interfaces and to design the messages protocol between both.
There are many ways for this messages exchange: you can use memory buffers, exchange registers, FIFO queues, etc, if you want to maintain the integrity of the design. But -and I know that you prefer not initiate the communications services-, I believe that best would be use a tradicional IP port. This standardizes the design and makes it independent of the architecture. In this case, for example, I could get a very simple PC exclusively dedicated and optimized to cMP/cPlay and the client interface in any other system: windows, linux, mac, etc, without consuming resources of cMP/cPlay. This, also, would facilitate the use of external storage and release cPlay the need to control it.
Even better: what if we released cPlay of the tasks of the DSP and it is limited to handle the audio you are serving, decompressed and upsampled, from an external computer?. I am absolutely sure that the resources and power consumption of communications services are much lower than the DSP are. You yourself has admitted that cPlay sounds worse when must to decompress the FLAC files. Let alone deal cPlay control audio and outsource all other tasks.
Now, I have got some time. I could help in the design development. Send me an email.
Regards.
Ignacio.
Regards.
Ignacio.
Unfortunately some win32 apis must be used for best performance in cPlay (although this is limited and can be ported without too much effort). My preference is to limit interaction to just the kernel. I'll meditate on a possible solution, perhaps just a simple .lib that you can use...
Refer to ReadMe.txt for important details & specifications.
Change Log (2.0b28 release):
- DSP refinements (reduce low level overheads & enhance threading)
- ASIO refinement (support for rare soundcard types removed)
For best results, uncheck "Timer" setting and test different "Buffer" sizes. Details on SoX settings and output measurements can be found under Release 2.0b26 Notes .
Please REMOVE previous versions before installing cPlay 2.0b28. Normal (SSE2), SSE3 B9, SSSE3 B9 and SSE4 B9 versions are available. Running cPlay on a CPU not supporting the required instruction set will cause cPlay to exit immediately.
For cMP² users, review these BIOS changes (underclocks FSB, CPU & disable additional USB items). New settings are in effect allowing for lower power consumption (below 20W). This release allows for further CPU underclocking under maximum load (192k SRC 145db SNR output) where CPU Host frequency can be lowered to 140 giving CPU frequency of 840MHz! This setting is only recommended for WAV files (for FLAC, use a minimum of 150). Recommended Gigabyte mobo GA-G31M-S2L is no longer available, use either GA-G31M-S2C, GA-G31M-ES2L or GA-EG45M-UD2H ( advanced settings discussion ).
PROBLEM FIXED: The WiFi showed up in the USB device list. I disabled the WiFi Radio and the problem went away. Sorry for the false alarm.
Cics,I am trying out 2.0b28_sse4b9 (same thing happens on other 3 versions) on a Dell M6400 laptop running 32-bit XP PRO SP3 with a new EMU 0404 via SPDIF.
Every time I select content to play in cplay via the SPDIF out, I get a error that the driver requested a reset and to restart the player. This does not happen if I select the analog outs.
If I put it into diagnostic mode, it will hang indefinitely on "Loading...". I can not kill the task and have to reboot to recover from it.
I have uninstalled and reinstalled both EMU and cplay.
Any ideas on what is causing the driver reset when using the SPDIF out ?
Does cPlay work on 32-bit XP Pro SP3 ?
I will get 1 or 2 seconds of garbled audio before it stops. Both the EMU window and the DEQX window show the rate changes are detected if I try different rates (e.g. 44.1, 48, 96).
I made the changes to allow locking of memory and have the buffer size set to auto.
M6400 Laptop -> USB -> EMU 0404 -> SPDIF -> DEQX HDP 3.0 -> amps -> speakers.
2.53GHz QuadCore Intel Core 2 Extreme QX9300
Dual Boot: 4GB - 32-bit XP Pro; 12GB - 64-bit XP Pro
RAID0 (2 x 320GB drives)
MMX, SSE, SSE2, SSE3, SSSE3, SSE4.1, EM64T
Thanks,
Tim
Edits: 07/25/09 07/25/09 07/26/09 07/26/09 07/26/09
There are cPlay users on XP SP3 and have not had similar problems.
The fact that it works on EMU's analogue outs is very strange and seems to suggest a problem with EMU's driver. Do you have the latest drivers installed?
Cics,
"The fact that it works on EMU's analogue outs is very strange and seems to suggest a problem with EMU's driver. Do you have the latest drivers installed?"
I was only able to get it to playback without throwing the error when it was set to analog outs and may have had other configurations incorrect. This observation may have been a red-herring because I did not have the TRS/XLR analog cables at that time so I could not hear anything from the stereo rack. I just saw that player count up with no external audio. I would dismiss this observation.
I believe I have the latest drivers. I am using the drivers that came with the EMU install discs. I did not find any newer drivers, but I may not have looked in the proper places.
I spent a day trying different software and driver permutations. I had NetJack2 installed and then de-installed that. I started blowing away processes and still no go. I tried ASIO4All and that didn't work. I tried other players including Foobar2000 and J. River Media Center. J. River would randomly output clear content, garbled content, and silence. The next day I went out and bought a pair of TRS to Male XLR interconnects from Sam Ash Music Supply to try the analog outs. Still no go with the analog outs going into my pre-amp.
After all of this, I started looking at the USB entries in the device manager (versus the audio drivers). When I noticed WiFi was listed as a USB entry, I disabled the WiFi radio from the tool-tray and everything started to work (CPlay, Foobar2000 and J. River Media Center).
BTW, I am comparing a Mark Levinson #390S CDP XLR digital out against the EMU 0404 SPDIF digital outs going straight into a DEQX HPD 3.0. So far the ML #390S still has the edge, but I haven't tried doing any PC mods, EMU power supply mods or fancy USB cable mods yet. It works so well straight off the bat (after disabling the WiFi radio) that I will be building a dedicated PC around it. My goal is to replace the ML #390s with a better sounding network player.
Is it possible to launch CPlay in realtime such that it will play 1 song and then quit? When I use the .bat file, it does not exit after playing. I would like to make a light weight network command interface to feed it a song at a time and possibly support other controls so a tablet or laptop could remotely control it without the graphics overhead of VNC or other remote desktop software. This would also offload the database searches, artwork, artist, etc. displays to the laptop or tablet PC. The only thing that would go to the playback machine would be the cplay command with the filename (and possibly other exposed commands).
Thanks,
Tim
Is it possible to launch CPlay in realtime such that it will play 1 song and then quit? When I use the .bat file, it does not exit after playing. I would like to make a light weight network command interface to feed it a song at a time and possibly support other controls so a tablet or laptop could remotely control it without the graphics overhead of VNC or other remote desktop software.
"...BTW - I had the ML390S directly into my ML33H's..."
Kindred ML Spirits.
Has your cMP / CPlay implementation bested you ML390S yet?
My amps are a ML336 for bass and Jeffrey Rowland Model 10 for (mid+highs). Formerly with the DEQX, Magnepan 3.6's and Martin Logan Descent, I had 11Hz to 20KHz +/- 5dB in room response. I have since swapped out the 3.6's for 20.1's and don't have the new numbers dialed in yet.
Tim
Hi cics
I am just starting to try out your cPlayer/cMP, it looks a great concept, thank you.
I am trying to get cPlayer working first. I have AMD 4850e (SSE3, confirmed by CPU-Z). However, neither of the following versions will launch;
cPlay_2_0b28_sse3b9_setup.exe
cPlay_2_0b28_ssse3b9_setup.exe
The following do run fine;
cPlay_2_0b28_sse2_setup.exe
cPlay_1_4_Final_SSSE3_setup.exe
Do you have any tips. Otherwise, which of the last two would you suggest I use?
Thanks
Oli
System;
Windows XP Home (SP3)
Gigabyte GA-MA78GM-S2H rev1 (latest BIOS & chipset drivers installed)
AMD Athlon™ X2 Dual-Core 4850e
Trust 5250 (CMIDriver-1.2.5-bin-x86)
ASIO4ALL
I assume you're able to install the "SSE3B9" version but cPlay doesn't start.
This may be an issue with AMD 4850e (very unlikely though) as cPlay SSE3 is fully tested on a Pentium 4 SSE3 capable processor. What I suspect is the "B9" feature that's affecting your system. Any chance of using XP SP2 or SP3 Pro?
cics
I tried rolling back to SP2 - same problem. Yes, I can install SSE3B9, but it will not run.
Thanks
Oli
I have found that only SSE2 versions of cPlay will run on three AMD AM2 processors. I have tried SSE3 without success on a 4000+, 5000+ black, and 5400+ processors.
BTW - I find that the latest versions of cPlay with SoX VHQ Intermediate Phase seem to sound best in my two all digital systems using Juli@ soundcards and s/pdif output.
Jack
Hi Cics,
Whow this version 2.0b28 really, really swings !
Very enjoyable…..
Every previous version added (more or less) better highs, mids, lows, 3D, ect.
But this version…whow….
This version ads the last missing ingredient: PRAT !
Pace, rhythm and timing.
Although i’m not playing any musical instruments myself, i’m blessed too be surrounded by many friends who do/can and some music-producers. They all have one simple rule to judge sound quality: “do you want to turn it louder or do you want to turn it down ?”
Well….. with this version I want too turn the volume up !
The more volume, the more it swings.
And it doesn’t get aggressive or harsh… or boomy.. or …..
The seas tweeters in my Klein + Hummels really like the highs of this version
The mids are beautifull two.. Saxofoon ect sound wunderfull !
Cics, I must thank your daughter for messing up your CD-collection.
Otherwise you wouldn’t have produced such a nice player. .
Thank you for this version.
LynxL22 dig i/o XLR AES/EBU -> Lavry Black DA10 -> XLR Mogami Gold -> Klein & Hummel O300
I have been spending the entire weekend working on the cMP2 + cPlay system and as a result have locked up the G31M-S2L quite a few times with just about every combination of settings.I know have a 2 inch thick binder with all the documentation and all the relevant posts littered with post-its!
Here are the settings that seem to work best to keep a stable system:
CPU Clock Ratio: x7.0
Fine CPU Clock Ratio: +0.0
CPU Host Frequency: 175 mhz
SPD: 2.0
FSB DeOverVoltage: -0.15
CPU GTLREF Voltage Ratio: 0.566
CPU Voltage Control 0.875CPU-Z reports the following:
Core Speed: 1225 mhz
Multiplier: x7.0
Bus Speed: 175 mhz
Rated FSB: 700 mhz
Core Voltage: 0.864v
Memory Timings: 3-3-3-8 (Set to Auto in BIOS)
CPU Temperature: 48 deg. C.I have done almost all the Windows optimizations (including AWE) except for Minlogon (For which I am waiting to have finished with Windows changes).
AWE configuration instructions are not included with the 2.X versions of cPlay postings. Is it still used? cPlay diagnostics shows it is successful.
For some strange reason I can't set the CPU Clock Ratio to 6, the board refuses to go lower than 7, it locks up and resets the CPU Host Frequency to Auto.
I'm currently using 2 sticks of 512 meg Kingston Value RAM and the memory seems to be the most touchy part of the system. Setting the Fine CPU Clock Ratio to +0.15 and the CPU Host Frequency to 170 mhz resulted in a BSOD related to RAM. Next thing to try is the Hyper-X of which I have 2 sticks of 1 gigs each. I need at least 1 gig of RAM as I listen to a lot of classical music with tracks of more than 30 minutes. I wonder which is the best, 1 stick of 1 gig or 2 sticks of 512 mb (to enable the dual channel)?
I also tried to lower the PCI-Express Bus Frequency to 90 mhz and that caused instabilities and a BSOD in Windows. Is this setting still recommended?
As far as the sound quality is concerned it's amazing. The Vangelis Mythodea album sounds superb, very open, and I hear many details I'd never heard before. The voices sound so natural and the bass response is very strong but not muddled. If anything this version highlights my slightly too bright tweeters, but the sound is never aggressive.
Daniel Gauthier
Montreal Canada
Gigabyte GA-G31M-S2L, E7200, 2 x 1 gig HyperX LL Memory, Antec SLP-450WR PSU, WinXP sp3 on Compact Flash Drive, Jili@ Analog Out direct to Electrocompaniet Ampliwire II Amplifier, Apogée Convergence Speakers
Edits: 07/12/09 07/12/09 07/12/09
Your experiences suggest EIST is enabled in BIOS. It's best to use just one RAM module (either 512 or 256). Are you using the F6x BIOS?
PCI Express should be left to its default value.
After checking and changing for above, leave VID at 0.875V and reduce CPU Clock Ratio to 6x. If stable then look to reducing CPU Host Frequency followed by CPU voltage VID. All else looks OK.
I've only been able to get back to the player a few days ago. I replaced the RAM with the Kingston Hyper-X 1 gig LL module and today had the time to try different settings. I've finally got to where I should be! The RAM made all the difference in being able to lower the clock speed and the lowest voltage I can get to is 0.775 volts. Any lower and the board refuses to start. I am finally satisfied.
CPU-Z now reports:
Core Speed: 900 mhz
Multiplier: x6.0
Bus Speed: 150 mhz
Rated FSB: 600 mhz
Core Voltage: 0.752v
Memory Timings: 3-3-3-7 (Set to Auto in BIOS)
CPU Temperature: 38 deg. C. (10 deg cooler!)
Now all that's left is the min-logon which I will try after Ghosting the compact flash.
I've also installed an external RAID box that sits in the next room connected by an eSATA cable. The box is accessed from my other PC via USB and has 1.5 terrabytes online. I now have access to all my music on the player and it sounds absolutely wonderful.
Daniel Gauthier
Montreal Canada
Gigabyte GA-G31M-S2L, E7200, 2 x 1 gig HyperX LL Memory, Antec SLP-450WR PSU, WinXP sp3 on Compact Flash Drive, Jili@ Analog Out direct to Electrocompaniet Ampliwire II Amplifier, Apogée Convergence Speakers
Reduction in temperature is a very good indicator of lower power consumption. You're very close to the optimal setup.
Use Seger's method (for both SP2 or SP3) to implement Minlogon. You should also reduce RAM size to 512MB.
Hello cics.
EIST is disabled and I am using the F6x BIOS. I think the problem preventing me from lowering the CPU clock is the RAM. I will replace it later today and see what happens.
Daniel Gauthier
Montreal Canada
Gigabyte GA-G31M-S2L, E7200, 2 x 1 gig HyperX LL Memory, Antec SLP-450WR PSU, WinXP sp3 on Compact Flash Drive, Jili@ Analog Out direct to Electrocompaniet Ampliwire II Amplifier, Apogée Convergence Speakers
Has there really been a recommendation to do this?
From cics or from a pasticpant?
Took me by surprise.
I would think it futile trying to slow your machine down until all optimizations are done, most assuredly including MINLOGON, and even then I have found one needs to "ease" into the lowest possible settings, especially CPU voltage.
Bye,
Rick McInnis
I actually don't remember where I saw the PCI Express thing. It was written on a post-it stuck in my printouts of the important threads.
All optimizations except Minlogon are done, I'm going to do a full backup before trying that to ensure I don't trash my Windows install.
Daniel Gauthier
Montreal Canada
Gigabyte GA-G31M-S2L, E7200, 2 x 1 gig HyperX LL Memory, Antec SLP-450WR PSU, WinXP sp3 on Compact Flash Drive, Jili@ Analog Out direct to Electrocompaniet Ampliwire II Amplifier, Apogée Convergence Speakers
CICS.. Once again thank you!! First I did not expect another release, and second I thought it would be minor fine tuning.
This release is a significant enhancement. On my system, SOX pulled out way ahead of SRC. I have yet to experiment with all the buffer sizes. I am running without any up sampling feeding a good quality NOS DAC from a notebook computer driving CMP. The clarity of sound improved while musicality preserved. The sound stage on SOX took a step improved in 3d. I am hearing details in music never before heard. In my opinion on my system SRC sounds like it is processing the sound as you hear with up sampling where SOX is giving full transparency. In any case it is excellent and a surprise gift late last night when I realized it was available and plugged it in.
The bass less muddy and the little haze is gone. More space in the sound stage. May be that's what HP called transparency.
Hello, Cics and everybody! I wonder at your silence! This is a wonderfull new version, indeed. The lower region got so clear and dimensional, the sense of pace and rhythm is so pronounced! I think it is a great achievement overall, and it sounds so clean and HEALTHY. I stopped using SOX.
Serge.
More detail with more musicality! Micro dynamics and vocals are wonderful.
Are you using SRC@145db?
Its a long time ago that I posted something. I am very busy in my job but try to keep up with the development of what is happening here. Also I have changed my hifi system with the help of alfred (sonics) completely. Only the speakers survived. This morning I installed the new player and was imeadetly hooked by new details and natural sq. For me especially voices and piano, but also drums and bass have a new quality I didnt expect to be possible. Thank you cics. Like several times before you made my day. Walter
Sonics (i.e., Alfred) is one of those very knowledgable yet sharing kind of gems on Audio Asylum that makes this such a great place. I too am a student of his.
cics wrote:
"More detail with more musicality! Micro dynamics and vocals are wonderful."
Well, here's an "unsubstantiated assertion" to get 'em going.
You're right.
Running v 28 into an as yet only partly-configured "headless" Fit-PC2 and a NOS DAC, I don't think I've ever heard music reproduction like it, at least not in my home.
As you suggest, unaccompanied choral music, which often on disc can be somewhat flat and lifeless, is sounding uncannily real.
I have no idea how my setup compares to high-end this or mega-buck that (and don't much care) - what I'm hearing is very good and that'll do me.
I haven't commented on a cPlay upgrade for months but this time at least you are definitely to be congratulated.
Dave
Yes! I tried different sox settings, but somehow they all seem to be compromising one aspect in order to push out another, that leads initially to a WOW, but soon to a "don't believe" condition. SRC is more natural to my ears.
Yes, You are right about microdynamics and musicality, every minute detail is here and has a right feeling and shape. Even my incorrigible Robert Plant - Mighty Rearranger started sounding more smooth. Interesting that Your player is Not smooth, but it brings so many details that such terms as sharp, harsh, smooth are not relevant any more. I am not trying to flatter You, I have a vast experience with very smooth and liquid players - accuphase 65, 67 and 500. Well, They are smooth and tame and very "nice" sounding but at the expense of the hf details. Your player sounds smooth when musicians want it, and, I hope, it says all. Thank You for this constant pleasure.
Serge.
It is going to be a fine weekend!
Edits: 07/11/09
Thanks yet again. I'll say the same thing about this release that I've felt about every release. It is yet again another increment in improved clarity, diminished smearing, and improved 3-dimensionality.
Initially I love 28 sonically--sounds more transparent top to bottom esprcially top. However the last couple of days I have been back into metallics. When I run src/146/192 I get 50% + 7% cpu usage and ultimately the annoying ringing. I get it even with sox. Anyway I notice that in Task Manager performance I get a red plot showing up. When it surfaces (running 96 src) the total % cpu usage shoots up 7% to 33% (or way over 50% for 192) . I looked up Performance Help description in windows and it indicates the red can reflect amount of cpu consumed in kernel operations. So I assume this is what I'm seeing. However when I ran for 2 weeks perfectly I got 25% cpu usage with src/146/192 now I get 50% + 7% which eventually goes metallic (I don't remember getting that much % usage variation). So question what do others get for cpu usage with cplay full out? I see this as a major indicator of how stable my pc runs. If I can get back to a steady 25% usage I feel confident I will run good. Any ideas?
Very nice except for the garish interface,the colour's and buttons on the previous version look a lot cleaner than the new ones.
I run it on vista[32" monitor and it looks crap].
Hey Flip,
I agree with you about the looks. But none of the pretty players sound anywhere near as good as cplay on my rig.
having stated that I thought one could get a pretty interface and great sound, I have been sorely disappointed.
Cplay is now what I use and at aleast on my 17" monitor cplay looks fine.
Also if you do the video optimizations ONLY cplay will look good on your screen....
Have switched back to b25,that great big ugly red button is to much for me to handle.For a player to be great it has to look good as well as play good.
When I want a bit of eye candy I switch to Foobar.
Hey Flip,
Keep in mind that Cplay arose due to inadequacies in other players like Foobar. It was designed to fit into the memory player concept where OS bloat and many other things are treated.
If you do the necessary video optimizations to maximize sound (like 8bit color) no player is going to look good. I had winamp with pretty cover art and well it looks like crap in 8bit color.
I have to hand it to cics for staying with his audiophile sound first principles even at the expense of looks....
Hey, my dear beloved wife chose those beauties!
I dare not change them for a while unless I'm willing to loose some liberties (like implementing more refinements... b29).
Please do not get me wrong,I love the program.
There are many who use it on there main pc's without all the os tweaks.
If you want to see what it looks like run it on Vista on a large screen.
I have just had another look and I will stick to b25,maybe some time in the future there will be an option to resize those buttons
No offense taken whatsoever.
Each cPlay release is tested on a Vista machine - the 'spartan' look is much better than in XP. These machines have normal 32 bit color resolution. What res are you using?
Running 1920x1080 on 32",like I said b25 looks great,but would look a lot better if I could change the size of the task bar/text.
I do realise that some people operate it from a distance so they need large buttons/text etc.
I have it open alongside Foobar to compare and the text and buttons are about 3 times larger.
Well, because of the cheap USB device I'm playing with at the moment is limited to 24/96, that's what I'm running at, so my information may not be entirely helpful to compare to your 24/192. But maybe it is at least some indicator that might help you.
I just listened to an entire album and the CPU graph was pretty consistent throughout. It only spiked an additional 5-10% (on CPU 1) when loading the next track, but this seemed to be dependent on the size of the track. In the Task Manager, under View, I enabled "Show Kernel Times" that way I could see the red graph as well as the green.
So running at 24/96 with SOX set to VHQ, Minimum Phase, 99 Bandwidth and Alias on (and CPU at 840MHz):
The overall CPU usage bounced between 10-15%
But looking at the graph for each CPU: CPU 1 Green was 14-22% and Red was 1-6%
CPU 2 Green was 3-8% and Red was 2-7%
I hope this information is useful to you.
thanks it tells me my cpu is laboring maybe unneccesarily
just as mysteriously as it started 2 days it has now just ended---I'm back @ 25% for a full out cplay.
How about a link to the source code? I am learning to code ASIO using C++. This source coce would make a great tutorial.
I second that. This project made me very interested in ASIO coding.
.
cPlay 1.0 Final has been packaged (includes instructions, precompiled libraries and DLL) and is easier for learning ASIO.
cPlay 2.0 will be packaged later.
Could I solicit your thoughts and experience in connection with passive cooling for the E7300 CPU on a Gigabyte Gigabyte GA-G31M-S2L mother board. The Thermalright SI-128 originally suggested by Cics is discontinued. I'd welcome some recommendations for other passive coolers which have been found to work okay.
Thermalright AXP-140 Low Profile 6-Heatpipe Aluminum & Copper CPU Heatsink
RayBan
Thanks for this suggestion.
I just realised that I stupidly put my original post in the cPlay string when I ought to have put it in cMP. I've posted it again in the correct string and I apologise to all for this dual post.
Anybody using cPlay with an Empirical Audio Off Ramp 3?
I loaded the latest version of cPlay and it sees the ESI Julia and ASIO4ALL in my music server but not the Off Ramp on USB. I'm certain the Off Ramp is recognized in the system (Foobar playing). I tried changing USB ports to see if reassigning it would do something and it didn't help.
A solution would be greatly appreciated. I'm very much looking forward to getting cPlay working.
Thanks.
You guys are using asio4all with the offramp? I thought Steve had his own drivers and didn't require 3rd party. My Offramp turbo 2 uses the mtransit drivers, no need for asio4all (which I don't think sounds so good).
The issue is getting the Offramp to work with cPlay. It must go via an asio driver to work with cPlay, hence the use of asio4all. Steve supplies directsound2 for use with Foobar 0.8.3 and Offramp, but this will not work with cPlay. Is the mtransit driver that works an asio driver?
The Maudio Transit driver only works with the older Off-Ramp units. The new units use Centrance USB interface which is plug-and-play with Windows drivers. Same as the Benchmark DAC-1, Belcanto 24/96 and PS Audio.
If this does not work with Off-Ramp, It will not work with these others either. It think the problem is ASIO.
steve N.
Yes, no asio4all on my setup. Cplay sees the unit an an m audio transit. FWIW, I think cplay sounds superior to the Nugent recommended foobar 0.8.3/directsound 2.
Edits: 07/10/09
I used earlier versions of cPlay (those up to early spring) with offramp 2 via asio4all. cPlay sees asio4all(not offramp) and asio4all sees the offramp. If you get this and it still doesn't play, try doing a clean reinstall of asio4all.
Got it working. I uninstalled ASIO4ALL and reinstalled it with the offline dialog, found the Off Ramp there, then cPlay/ASIO4ALL showed the options for the Off Ramp.
My first install of ASIO4ALL excluded the offline dialog. cPlay was seeing only the association to Julia.
Thanks!
For anyone considering a linear PSU + picoPSU
In the process of optimizing the power for the P24 connection, I did some current measurements on the 12 Volt DC powerline that feed the 12V Micro Power Supply from Mini-Box. Model: PW-200-M (200 Watt).
In some posts current draws on the 5 V lines of the P24 connector off over 4 Amp’s are reported (On a cMP optimized computer transport). So just to be sure that enough power can be delivered, I bought an 200 Watt version (PW-200-M) and also an linear bench/lab PSU capable of delivering up to 6 Amps at 12 V.
After measuring the current into the 12v Micro PSU the whole seams a little over dimensioned.
During boot-up the current to the 12V Micro Power Supply reads between 1.6 to 1.95 Amps with one short peak on 2.1 Amp.
During playback (or idle) it's 1,28 Amp. At a constant value.
This value is only a little influenced by bus speed, monitor resolution / color settings, type of sound-card. Ect. The LynxL22 draws 0,1 amp more than the Juli@ (without the analog part)
So in my cMP-setup the P24 pin uses/consumes:
- At start-up: 1.6 – 1.9 Amps. 1 peak at 2.1 Amps. x 12 Volt is about: 20 - 25 Watt’s
- when playing / idle: 1.28 - 1,35 Amps. x 12 Volt is about 15 – 16 Watt’s
If anyone considers buying a picoPSU for the P24 connector:
the smallest picoPSU version of 60 Watt, should be enough (for a cMP optimized PC).
Also a small linear bench / lap PSU (with low ripple specs) capable of delivering 2,5 Amp’s should be enough.
PSU-wise my cMP setup now looks like this:
- P4 pin: 1 linear bench / lab PSU output: 12V DC / 2,5 Amp max
- P24 pin: 1 linear bench / lab PSU output: 12V DC/ 6 Amp max -> Mini-Box PW-200-M. (with smoothing caps on all 3 voltages (3,3/5/12)
- USB, HDD, DVD, etc: ANTEC AW 430
Pin 16 (PS_ON) on the ANTEC P24 connector is grounded to pin 17 to start the Antec. The P24 pin isn’t used to feed the MOBO P24 connector. (The picoPSU now does that). Only the ANTEC’s molex connecters (5V + 12V) are used to feed USB, HDD and DVD as explained in Cic’s AOB manual)
Sound Quality <-> costs
ANTECH AW 430 PSU costs about 47 euro (in The Netherlands)
A picoPSU costs 37 euro + 30 euro for linear bench / lab PSU = 67 euro (in The Netherlands)
IMHO the extra 20 euro’s are definitely worth the extra costs.
I hope the placebo-effect is not fooling me but I think the picoPSU + Linear PSU sounds better than the ANTEC AW 430. Not a gigantic difference, just a little better. More space, better imaging, better treble , etc. Everything a little better.
And…… as with the power optimization for the P4 connector: its very, very easy to implement. Anyone can do it in 15 - 30 minutes.
Are there any other inmates that can comment on SQ-improvements when using a picoPSU (versus a 'normal' swithing ATX PSU) ?
Mark
N.B. How the currents measurements were done:
1 linear bench / lab PSU -> multi-meter -> Mini-Box PW-200-M.
cMP2 optimized as per Cics recommendations: all BIOS optimizations done (no EIST function) + all software + min logon. GigaB G31M-S2L, E7300, Core Speed: 840 MHz (140x6), FSB 560 MHz, 1Gb Kingston HyperX, at 3-3-3-5, FSB not undervolted, CPU GTLREF ratio at 0.65, CPU Voltage at 1.0000V (just enough for allowing passive cooling)
LynxL22 dig i/o XLR AES/EBU -> Lavry Black DA10 -> XLR Mogami Gold -> Klein & Hummel O300
Yes, I'm using the Pico with a Daniels inspired Linear PS built into the case along with a Hammond 193L inside also.
I added AR power cables and a GStew modded Juli@ digital at the same time so the sudden increase in SQ was difficult to attribute to any single change.
No fans, no switching PS, runs cool and sounds great.
Ryelands inspired me to order Element twisted pair speaker cables and interconnects as my arthritis prevents me from making my own as he has so capably done.
RayBan
Refer to ReadMe.txt for important details & specifications.
Change Log (2.0b27 release):
- Minor ASIO refinements
- DSP refinements (new advanced design implemented) giving efficiency improvements
- Disable keyboard processing when file browsing
- Faster 24 bit WAV processing (improving RAM load times)
- Compiler optimisation done (for FLAC)
Details on SoX settings and output measurements can be found under Release 2.0b26 Notes .
Please REMOVE previous versions before installing cPlay 2.0b27. Normal (SSE2), SSE3 B9, SSSE3 B9 and SSE4 B9 versions are available. Running cPlay on a CPU not supporting the required instruction set will cause cPlay to exit immediately.
For cMP² users, review these BIOS changes (underclocks FSB, CPU & disable additional USB items). New settings are in effect allowing for lower power consumption (below 20W). This release allows for further CPU underclocking under maximum load (192k SRC 145db SNR output) where CPU Host frequency can be lowered to 140 giving CPU frequency of 840MHz! This setting is only recommended for WAV files (for FLAC, use a minimum of 150). Recommended Gigabyte mobo GA-G31M-S2L is no longer available, use either GA-G31M-S2C, GA-G31M-ES2L or GA-EG45M-UD2H ( advanced settings discussion ).
Hi,
I have a little suggestion related to audio file loading using cuesheet.
Would it be possible that cplay be modified in the following way:
- If cue file mention FILE "foo.wav" WAVE
-- try to load foo.wav from current directory and if it fails fallback to the loading of foo.flac and if it fail display an error.
This could be usefull for people using flac as modification of cue files generated from EAC or other software would be necessary anymore.
If you don't have time for this, I would be interested in sending you a patch for that feature if you can send me the source code to pbmtp AT free DOT fr
Thanks for this software
What will be most useful is a playlist tool. I'd like to remove the "File Browse" button and replace it with a more general playlist tool button. Alan has built a standalone cue sheet creator / playlist tool in Java.
Idea is when you hit this button, an elegant playlist tool takes the place of the old file browser. One can either select a file / track or build a new playlist and send it to cPlay for playback.
This will need to be built in c/c++ using either WTL 8 or native WIN32 api (preferred). Also, avoid a resource editor.
I ripped the whole cd into one large image FLAC file + a CUE file using EAC. With the cplay "File Browser", I just load the CUE file and play the music. That works very well with the existing cplay file browser.
If we move to "playlist tool" instead of file browser, is it possible to create play list selected tracks from these CD images base on the information of image CUE file? That would be ideal. I don't want to rip all music again into "one file per track".
s
YES,
You CAN create a cue sheet playlist. Alan Jordan who posts on the computer asylum created such a program. I am pressed for time, but doa search for somthing like "cue sheet playlist creator" or "recursive" and look for posts from his moniker "aljordan".
It is a fantastic program.
using cuesheets? Playing the WAV file, playing the FLAC file or displaying an error.
Rip one of your cd with EAC in flac mode, ask EAC to create a .cue file.
Once it is done load the cue file with cplay and you will get an error as inside cue files have .wav extension whereas on disk they have .flac extension.
To fix you need to open the cue file with a text editor and replace all occurence to .wav with .flac.
My idea of trying to load .wav file and if it fails trying to load .flac file would fix such issue without the need of editing all of your .cue file
.
I had upgraded cPlay from b15 to b27 (I know I am a bit lazy but I was away from my main system until a week ago) and the different is huge. The sound I am getting now is very close to the analogue (LP with Infinity RS1) sound twenty years ago. Thanks to cics for getting me close to what I have been missing for so long, the emotion of the music.
I have been trying out the FLAC option in cPlay last night and am totally puzzled! The sound is quite different from WAV from which the FLAC file was derived. I expected that the difference would be minor and that WAV would sound better. On the contrary, the difference is quite noticeable on the get go and what is most surprising, FLAC sounded better, more analog! (There is one case-Joni Mitchell's big band version of 'Both Sides Now' which I haven't decided yet on which is the better version. But the difference is there.)
Does anyone have the same experience or am I just delusional? And can someone explain the difference in sound? I thought the FLAC file was first decoded when load and that would not be any difference at all.
I have given the system another listen this morning and discovered another recording that I prefer WAV over FLAC: Van Morrison's Astral Weeks (Remastered)--Japan import only. My guess is that:
-- there is more high frequency content in WAV playback (easily detectable but why?)
-- less high frequency in FLAC renders some recordings more analogue sounding
But why is there difference between WAV and FLAC playback?
Jumping from b15 to b27 is massive! Analogue together with recreating that natural "un-electrified" sound remains the only reference - doing 192k SRC output of 44.1k material to an external high-end DAC is very special and pure...
Depending on mobo / CPU / RAM setup and power supply, FLAC will have an impact. cPlay decodes all FLAC material into RAM before playback (thus ensuring exact same audio data). FLAC decoding is however intensive ito CPU load. This intensive decoding process should be seen as a severe electrical storm being unleashed. We get a large PS load being generated. How your PSU and mobo setup handles this will affect SQ thereafter. All my music is stored in WAV thus avoiding this issue. A way to reduce this PS spike (which can last several seconds) and its "after effects" is to cap your CPU, i.e. set maximum CPU frequency to lowest stable level ~900MHz.
By turning off alias in SOX, I find the 'electronic' sound of WAV files playback deminishes
I have settled on using juli@ latency of 64 and it in combination with cplay 2.0b27 has been favorable. I can now lower my host clock to 144 and voltage to .85 and I get excellent sonics with sox @ 192 and very few metallics. I'm still not sure about buffer size in cplay. I like small the best for the pristine highs but large is not bad either for the good mids, bass. But I'm confused on what is theoretically the best for cpu usage and sonics. There was a lot of discussion in AA pc audio on this but I was lost in all the debate. Can anybody clarify this?
Relative to 27's sonic impact vs 26 I believe Serge has nailed it for me as well.
Also try SRC and Medium buffer size as host frequency of 144 should not be a problem.
ok will do
I have been running src and 192 sample rate(64 latency in Juli@) for a day and a half ---digital out into my Benchmark---no metallics and the rabbit sounds wonderful!!! I don't know what happened but I am at 140 host clock control, .85 volts in bios and it is rock stable. I always thought that power supply had something to do with it. Now I have cpu on a linear supply, all hdd's on gd's, no fans and I guess the only thing running off the Antec is digital power and analogue power. So with digital out that is a very low load on the antec. Maybe this is the reason, maybe not but any way real nice!!!!
Major thanks to cics for requesting I try it.
Have you removed Juli@'s analogue PCB?
latency now at 48, analogue board off. Please remember I was getting some metallics from juli@ analogue outs with sox. Now nothing @ 64 latency.
Edits: 06/22/09 06/22/09
Got to shut down , going out for awhile. No metallics aftwer 1 hour.
m
.
it goes an hour or so then metallicca. sorry I changed parameters midstream but I have been planning/building the battery supply for some time.
Alfred these are LiFePO4 from a usa company in california.
...they are everything you said. Soundstage becomes large, hardness is eliminated, timbre is spot on.Thank you for this maybe best tweak ever on cmp^2.
One interesting phenomena on the metallics... I am running 50% cpu usage with 192/-146 src but I think I remember usage in the 30% range yesterday before the batteries with src. I get mid teens with sox. But I thought cplay 2.0b27 was an improvement on efficiencies. Can someone who is running 27 with 192/-146 src get a read from task mg on cpu usage? Certainly 50% usage was consistent in the past with metallics for me so that is not a surprise.
ok now I'm getting 25-27% cpu usage with src @192, what the heck would cause it to jump to 50% before??
I have gone 1 hour src / 192 no metallics @ 25% cpu usage. But but the sonics of the battery are overwhelmingly great---several discs I heard as if the first time. One is a classical piece wherein the depth gets 30-40% deeper. Another one with 2 female vocalists singing 'Walk away Renee' it sounds as if they both stepped forward several feet into my room. The air around them is uncanny (you can hear around them). Alfred thank you again!!
Maybe after I modded my juli@ for batteries it takes several hours to settle. I believe now I can use cpu usage as an indicator of whether metallics will come. If its 50% they will come if 25% they will not.
2 + hours no metallics
6 hours on src 145 / 192 sample rate / 48 latency on juli@ and small buffer on cplay---no metallics. The metallics I suffered from earlier was probably due (I'm surmising) to the soldering on juli@ not settled in yet and the power supply being off while I modded the sound card.
It was really unusual that I was getting 50% cpu usage (when metallics ocurred) and now I'm getting 25-28%.
My first impressions after listening for a number of hours, very very impressive!
This is superb. I have a CD which was recorded while I was in and around the orchestra videotaping, and I remember the sound very clearly. This version is the first time where I really feel I am back there in that same hall. Every instrument is clear and distinct and I can tell exactly where it is on the stage, exactly where it was in reality.
CICS, you have performed quite a feat here, thank-you for all the hard work. This has to be one of the few ultimate ways to listen to digital audio.
Daniel Gauthier
Montreal Canada
Gigabyte GA-G31M-S2L, E7200, 2 x 1 gig HyperX LL Memory, Antec SLP-450WR PSU, WinXP sp3 on Compact Flash Drive, Jili@ Analog Out direct to Electrocompaniet Ampliwire II Amplifier, Apogée Convergence Speakers
- Have you tried using just 1 RAM module (reducing memory to 1GB)?
- How is music accessed if using CF drive?
- Any mods on Juli@? By ESI's own admission, Juli@'s 114db SNR DAC is not the best (they instead went for better ADC). The big advantage is using 192k cPlay output which bypasses AKM's internal 192k interpolater. This way, SQ is exceptionally good (including digital VC) from this "average" AKM DAC chip. Very interesting.
1. Have you tried using just 1 RAM module (reducing memory to 1GB)?Actually I was mistaken, I'm using 2 x 512megs of Kingston ValueRam, The 1gig modules are in another PC at the moment, but as soon as I have a chance I want to swap and test different configurations of memory to see what sounds the best. I think I wouldn't go below 1gig as I listen to a lot of classical music and I like the possibility of loading an entire movement into memory.
2. How is music accessed if using CF drive?
Only Windows and software (C drive) and a spare backup drive of a total of 4 gigs are on Compact Flash, the music is on a 2.5" 120 gig hard drive at the moment, but it's much too small. I'm thinking of getting an external SATA box to put 4 drives in RAID 0. The box would be mounted on the other side of the wall in the next room.
3. Any mods on Juli@? By ESI's own admission, Juli@'s 114db SNR DAC is not the best (they instead went for better ADC). The big advantage is using 192k cPlay output which bypasses AKM's internal 192k interpolater. This way, SQ is exceptionally good (including digital VC) from this "average" AKM DAC chip. Very interesting.
No mods yet, I plan to power the card through a dedicated 5 and 3.3 volt linear supply and cut the power connections to the PCI bus. I would love to get an external DAC, say like the Apogee, or the Benchmark since people seem to get very good results with them, but it's not in the budget for a while yet.
I am also using only an Antec PSU to power the computer itself, no mods there yet, but I'm planning to add caps and later to use dedicated linear supplies for everything.
I've got these monster power supplies made for the Meridian-1 phone system that gives all the necessary voltages, I want to try that first, if it works you could probably get them cheap on eBay, but they are heavy, about 25 pounds!
Daniel Gauthier
Montreal Canada
Gigabyte GA-G31M-S2L, E7200, 2 x 1 gig HyperX LL Memory, Antec SLP-450WR PSU, WinXP sp3 on Compact Flash Drive, Jili@ Analog Out direct to Electrocompaniet Ampliwire II Amplifier, Apogée Convergence Speakers
Edits: 06/22/09
Thank you Cics! I agree with other fellows :) I feel the focus of sound has improved, resulting (1) less booming (2) deeper and wider sound stage (3) I can turn the volume louder without losing the sound stage or feeling fatiqueHowever for some reason I miss the "booming effect" in v26... Because I am using bookshelf speaker haha
And I have deleted all my up-sampled files (300MB+ each) and now only rely on SoX.
Edits: 06/18/09
Sox is available to use as a standalone program
http://sourceforge.net/project/showfiles.php?group_id=10706
May be you can compare offline upsampling as well
Hi everyone, I have been on another planet for the past 6 months but am now catching up on all the posts, as well as updating my system to the latest versions.About the composer name issue in cue sheets, there is a tag specifically for this in the specs called SONGWRITER. Me and my friend have been working on this issue and are transferring huge collections of music to hard disc, I've got now over 1000 CDs with the same number of cue sheets to edit and this is a big issue, not to mention managing the GENRE tags as well.
It would be nice if there was a way to support this eventually in cMP and cPlay to make it a little easier to navigate a big collection of music.
I've included a part of one of our cue sheets as an example below as well as the cue sheet standard description for that tag.
Frodan
REM GENRE Classical-Symphony
REM DATE 1977
CATALOG 4988005503473
REM ORCHESTRA BPO
REM FIRM DG SHM-CD
SONGWRITER "Brahms"
PERFORMER "Karajan"
TITLE "Symphony No. 1 in C minor Op. 68"
FILE "Brahms - Karajan - Symphony No. 1 in C minor Op. 68 - 1977.wav" WAVETRACK 01 AUDIO
TITLE "I. Un poco sostenuto - Allegro"
SONGWRITER "Brahms"
ISRC DEF058702021
INDEX 01 00:00:00TRACK 02 AUDIO
TITLE "II. Andante sostenuto"
SONGWRITER "Brahms"
ISRC DEF058702022
INDEX 00 13:22:02
INDEX 01 13:26:00
SONGWRITER
Description:This command is used to specify the name of a songwriter for a CD-TEXT enhanced disc.
Syntax:SONGWRITER [songwriter-string]
Parameters:songwriter-string - Name of songwriter. If the string contains any spaces, then it must be enclosed in quotation marks. Strings should be limited to 80 character or less.
Example:SONGWRITER "Paul McCartney"
Rules:If the SONGWRITER command appears before any TRACK commands, then the string will be encoded as the songwriter of the entire disc. If the command appears after a TRACK command, then the string will be encoded as the songwriter of the current track. Note: If your recorder does not support CD-TEXT, then this command will be ignored.
Daniel Gauthier
Montreal Canada
Gigabyte GA-G31M-S2L, E7200, 2 x 1 gig HyperX LL Memory, Antec SLP-450WR PSU, WinXP sp3 on Compact Flash Drive, Jili@ Analog Out direct to Electrocompaniet Ampliwire II Amplifier, Apogée Convergence Speakers
Edits: 06/17/09
I don't know if you are aware of the interim solution I have been using which is to list composers from the flac files' Composer tag as genre in the case of classical files. Al Jordan has included an option for this in his Recursive Cuesheet Creator. This accomplishes much though not all of what you are after. Under genre display, if one selects a given composer then his works are listed by performer-title.
If you want to use Songwriter in the cuesheets, you have to have a way of entering composer data from the music files into this and then a way of displaying Songwriter as you have described or some other way.
Next cMP release (much later) will look into this.
I have a few suggestions on how to implement it in cMP2.1. Have an ini setting to turn it on or off, Songwriter=True.
2. If it's on, you could have 4 buttons along the bottom for searching:
IF Songwriter=.F.
- Show buttons as they are now.ELSE Songwriter=.T.
- IF button = GENRE --> After selecting a GENRE offers a list by
- - SONGWRITER, after selecting a SONGWRITER, offers a list by TITLE,
- - then offers a list by PERFORMER.- ELSE IF button = TITLE --> After selecting a TITLE, offers a list by
- - SONGWRITER, after selecting a SONGWRITER, then offers a list by
- - PERFORMER.- ELSE IF button = SONGWRITER --> After selecting a SONGWRITER, offers a
- - list by TITLE, after selecting a TITLE, offers a list by PERFORMER.- ELSE IF button = PERFORMER --> After selecting a PERFORMER, offers a
- - list by SONGWRITER, after selecting a SONGWRITER, offers a list by
- - TITLE.
- ENDIF
ENDIFBTW I love the fact that with a keyboard one can type the first few letters of a search and cMP goes directly to the first entry that starts with those letters. It makes things a little easier when navigating though a collection of 1000 titles!
Daniel Gauthier
Montreal Canada
Gigabyte GA-G31M-S2L, E7200, 2 x 1 gig HyperX LL Memory, Antec SLP-450WR PSU, WinXP sp3 on Compact Flash Drive, Jili@ Analog Out direct to Electrocompaniet Ampliwire II Amplifier, Apogée Convergence Speakers
Edits: 06/22/09 06/22/09 06/22/09 06/22/09 06/24/09 06/24/09
Hello, Cics, hello everybody! We are happy to get one more wonderfull version. This one is less "intensive" as 26, but much cleaner. Soundstage is very detailed, and the instruments and their respective lines are naturally separated, which was not so evident before. Also the highs are on a new level of intelligibility, being sharp and "furry" but smooth. Like sable fur. With this version SRC is gaining up, especially for me, with Sox my best setting was 97 minimal no alias. Overall a shining new development of my only player. I also have to report that, even after minlogon, I had been having short dropouts-hiccups appr. every 12-17 minutes with 26, while with 27 I experienced only 1 and much less pronounced. So, if any changes were done in this area, they must have been beneficial as well. I really enjoy 27.
Serge.
My question to Cics - can You tell me how does volume control affect the resolution? I am afraid to touch it, it is always 0.
First, I'd like to thank CICS for his wonderful contribution to audio. I agree with Serg in his observations on 27. It is wonderfully musical while remaining precise. On my system much better bass. Wider presentation of sound stage. I am running on a dedicated lap top with most of the cmp mods applied. Perhaps with less availble cpu power the code optimizations have an even bigger impact.
Last section of this post explains modern digital VC. cPlay offers 53 bit numerical precision, i.e. no resolution loss. I'm currently using this digital VC [-24db .. -30db] into DAC.
Analogue VC only has advantage when preamp's SNR betters DAC's. Juli@'s DAC offers a SNR of 114db - not sure if you can get an affordable preamp that better this!
This is an excellent software volume control.
I am using the lightspeed attenuator posted on diyaudio and find it a wonderful companion to cmp/cplay. Very affordable and to my ear allows the full potential of cplay to come through. No question that with cplay at 0.0 and the lightspeed handling -40db of attenuation the sonics are much more complete.
Dear cics,wow, 840 MHz and my system is still running absolutely stable and without any dropouts or so - maybe I should even try to get lower... ;-)
I had settled on CPU Host Control voltage of 0.76875V, couldn't go any lower - tried again today and it seems as if i can go lower now - quite a surprise, I had to do a CMOS reset before when going any lower - is the system "burning in" or somehow getting adjusted to lower voltages over time...?
Anyhow - look forward very much to see how 2.0b27 will perform in a few days - I really believed it when you said "no more new releases in mid-erm", and here we are with (i) a release that has implemented a new and very promising upsampler (I'm not using it currently due to Buffalo32S though) and (ii) a new release with major refinements re footprint and performance - thank you so much cics, life is good :-)
Best,
Robert
Edits: 06/15/09
Lower CPU frequency allows for lowering CPU VID. In my setup, I have VID @ 0.73750V (one notch lower than before). At this level, CPU temperature (at max load) as measured in BIOS is 35°C and through RealTemp its 46°C/45°C (core 0/core 1).
No, sorry, I have to correct my yesterday's post - a cold boot (system was off overnight) is still not possible with VID values below 0.76875, even with CPU Host frequency now at 140 MHz.
When doing a cold boot before, the system as kind of "standard procedure" always had failed to come up at 900MHz so BIOS always reverted CPU Host Clock Control to "Disabled", System Memory Multiplier to "Auto" and DRAM Timing Selectable to "Auto", which I then always reset to "Enabled", "2.00" and "Manual". A warm reboot was always possible without that procedure if the system hadn't been idle too long (~5-10mins).
So it really seems I can't get any lower than 0,76875V. I would have loved to do so though, since every notch or two down brought what I considered still another (although small) improvement in relaxed audio reproduction with a tad more transparency and less distortion.
Or did I miss something?
Hi cics,
not very important, but your new buttons are not looking very well on (my) different screens. Want the old ones back :-))
Regards
Thomas
I currently have one 2.5" laptop drive but am going to need another soon. I was thinking of a SSD hd. I know cics recommended keeping all hd's of same type. But what is the risk of mixing lap tops with ssd?
As far as I know none. My cmp2 & cPlay system boots off of a 4 gig Compact Flash card and the music is on a 2.5" laptop drive.
Daniel Gauthier
Montreal Canada
Gigabyte GA-G31M-S2L, E7200, 2 x 1 gig HyperX LL Memory, Antec SLP-450WR PSU, WinXP sp3 on Compact Flash Drive, Jili@ Analog Out direct to Electrocompaniet Ampliwire II Amplifier, Apogée Convergence Speakers
Fro,
I know I can search for that info on the net, but in the interest of time and clarity, can you say more about how you have your 4Gb CF Flash Card setup as the boot drive and how it's connected into the computer? Also, does it connect via IDE or SATA?
Finally, what type of sonic differences do you hear with that as your boot drive versus a regular drive?
TIA!
Greg in Mississippi
Hi Greg
i use the same device . You can easyly change cards with different configurations and windows is separated from my music storages.And Cheap!!!!.they are connected with an CF to SATA Adaptor.It's not better sonically than an external SATA Drive with external PSU , but very practical.
As sonics said it's using an SATA to Compact Flash adapter, and the only advantage I see is less vibration from a hard disk - there's enough of that with the one with the music, and it makes the drive that has the music dedicated to it - ie put Windows on a different drive, so that random Windows disk access won't interfere with cPlay's access to the music drive. I'm planning to have the drives in a separate box later but in the meantime there's only one 2.5" disk spinning.
Daniel Gauthier
Montreal Canada
Gigabyte GA-G31M-S2L, E7200, 2 x 1 gig HyperX LL Memory, Antec SLP-450WR PSU, WinXP sp3 on Compact Flash Drive, Jili@ Analog Out direct to Electrocompaniet Ampliwire II Amplifier, Apogée Convergence Speakers
Edits: 06/19/09
Can you guys tell us which brand and model CF card you are using? Some of my research on this subject is saying a fixed one is recommended.
Brad
I use a Kingston 4 gig elite series compact flash card. I have had no problems so far and have rebooted the pc at least a hundred times.
Daniel Gauthier
Montreal Canada
Gigabyte GA-G31M-S2L, E7200, 2 x 1 gig HyperX LL Memory, Antec SLP-450WR PSU, WinXP sp3 on Compact Flash Drive, Jili@ Analog Out direct to Electrocompaniet Ampliwire II Amplifier, Apogée Convergence Speakers
For your Kingston Elite what file system did you format with, FAT32 or NTFS?
Brad
NTFS
Daniel Gauthier
Montreal Canada
Gigabyte GA-G31M-S2L, E7200, 2 x 1 gig HyperX LL Memory, Antec SLP-450WR PSU, WinXP sp3 on Compact Flash Drive, Jili@ Analog Out direct to Electrocompaniet Ampliwire II Amplifier, Apogée Convergence Speakers
Thanks Daniel for your help.
I had sucess with this CF adapter
http://www.syba.com/index.php?controller=Product&action=Info&Id=838
and your recommended Kingston Elite 4gb CF card. Also I think the drives shows up as fixed since its location is under hard disk drives in MY Computer.
What I like about this route is if a new revision comes out you could leave this CF card alone and install the new revision on a new Kingston Elite which are very cheap!!!
Oh yeah, boot up time is right under 30 seconds without any optimizations or bios changes yet.
Thanks again, Brad
I'm glad it worked well! I have an older version of that CF card so it takes longer to boot.
Daniel Gauthier
Montreal Canada
Gigabyte GA-G31M-S2L, E7200, 2 x 1 gig HyperX LL Memory, Antec SLP-450WR PSU, WinXP sp3 on Compact Flash Drive, Jili@ Analog Out direct to Electrocompaniet Ampliwire II Amplifier, Apogée Convergence Speakers
I bought mine from this guy and paid extra for priotity shipping, it arrived in 3 days.
Brad
ebay store
http://stores.shop.ebay.com/Yogi-Computers-Inc__W0QQ_armrsZ1
Thanks, that's allot cheaper than the ones I was looking at.
Brad
The places I have looked for the kingston elite showed they were out of stock so I explored some other options.
I found a new open box CF to sataII that included a Seagate ST1 5gb CF for 20 bucks shipped on ebay.
Link to adapter
http://www.syba.com/index.php?controller=Product&action=Info&Id=838
This adapter has the option to use 3.3v or 5v with a jumper with the source power from a floppy molex (wish it had a sata power inlet).
Link to Seagate CF
http://www.seagate.com/staticfiles/support/disc/manuals/ce/ST1%20Series/ST1%20Series/100325578h.pdf
and a review
http://www.xbitlabs.com/articles/storage/print/seagate-st1-5gb.html
Didn't know this till I looked it up that the seagate is an actual HDD at 3600rpm with 2mb cache.
I went ahead and purchased a 8gb ST1.2 for 17 bucks to try out too. At these price points I am hoping this works out. I have yet to build the cMP yet but have purchased all of the parts and am just waiting for arrival. Will let yall know how this works out.
Brad
Just a follow up of the seagate drives. After installing the OS on these drives I kept getting an error before Xp loaded saying "Your System has no paging file, or the paging file is too small". From what I could find the fix for this would require the purchase of an MFT editor so I ditched the effort. The other negative with this drive was it showed up as removable storage. Time to sell these drives on eBay.
Brad
Another solution is to get the Sata harddisks external with external psu.
I use the 'Sharkoon Sata quickport DUO'. It can has own psu and can handle 2 hdd's. Very easy.
Sorry if this has been answered before, but I was wondering what was the number besides certain track number when I load a CD into cPlay?
example :
01 26:50 Feierlich. Misterioso (04)
Whats the (04) ?
It's not in my cuesheet...
One other thing, will it cause me any trouble if I sometimes have ISRC codes in my cuesheets?
Thanks
It's the amount of pieces the track was split into, in order to fit into ram. In this case, the 26:50 track was split into 4 chunks.
I can't help you about ISRC.
Have a nice day
Hi,
I made some updates to CPlayList Editor: See the attached link for information.
Alan
aljourdan,
I've worked with your recursive cue creator in the past and it works wonderfully, placing the cue files in the same directory as my artist/albums.
Using the cPlaylist editor in the same fashion, and placing the files in the same directory, I cannot seem to get them to play with cPlay either by loading them directly from the directory or by selecting the optoion to play the list right from the cPlaylist application.
Any ideas on what the problem might be?
Thanks,
Jack
Thank you Alan - great job :-)
Regards
Thomas
I guess I should have looked before I leaped.
So I guess one if it does do a good job of collecting the information you would still need EAC to make a cue sheet. Not the worst thing to have to do BUT one wonders why this thing won't do it for you.
Oh, well ...
Of course, it I am incorrect I would greatly appreciate being corrected.
Rick wrote:
". . . you would still need EAC to make a cue sheet."
Al Jordan's freeware Recursive Cuesheet Creator (RCC) makes cuesheets that work with cPlay and (at least for me) Foobar.
It's far quicker and more reliable than writing cuesheets with either EAC or Foobar. Once I was familiar with it, I deleted all my cuefiles (search for *.cue) and re-wrote the lot. It took less than five minutes.
And while you're at it, take a look at the photographs on Alan's homepage.
Best
Dave
and the photographs are certainly worth viewing.
What an interesting fellow. I have bookmarked both.
I will report if I can hear any difference between the EAC and db... files.
Thanks again for the help,
Rick McInnis
Nope, you are correct. Sorry.
They have been "just about to come out" with cue sheet support for years now.
Eac is still the one to use IMHO for a cmp box.
Dear fellow inmates,I would be quite interested if anyone of you is playing SACD recordings through cMP2 - somehow... I know that SACD is typically 1/2822.4 DSD with a totally diffrent physical layer etc., but I've seen so many creative and brilliant solutions here that I would not be surprised, and thought I'd just ask, maybe I can learn once more... ;-)
Many thanks
Robert
Edits: 06/03/09
it has been said the main reason SONY developed it was to eliminate the ease with which we can play with REDBOOK.
I start this comparison with the caveot... your mileage may vary. Watching the oversampling vs no-oversampling discussions currently taking place just reminds me that how a particular piece of gear or tweak will sound to a person will vary based on their listening bias and their system. So this is MY comparison... you may hear other differences and come to different conclusions.My setup is a highly-tweaked cMP... 7 separate linear supplies, one each for P4, ATX-24 (provided by a modified PicoPSU), HDD, touchscreen/USB, Juli@ digital, Juli@ DAC analog supply, and Juli@ analog output stages (see my threads "PSU Follies" and "Revenge of the Power Supply Follies"). I've applied all the customizations and blazed the way a bit for the GA-EG45M-UD2H... I still need to write a report on setting up that m-board. I'm using the analog outputs of the Juli@, which are performing very well with some mods (see my thread "Juli@ follies"). I used a 256Mb stick of ValueRAM at 3-3-3-5 on both motherboards, preferring it to both the 512Mb Mushkin and 512Mb HyperX RAM. It's cased in a damped Zalman with everything taken out or off that is not needed to run the system... for example, I've even removed the door that hides the front ports.
The rest of the system is a pair of modified Eminent Technology LFT-IVs planar magnetic speakers driven by Electronic Visionary System's 500M's (modified B&O ICEPower 1000ASP modules), a home-built shunt attenuator using a Vishay series resistor and a silver-contact Shallcross switch, and various cables, most of them hard-wired at one end. AC filtering is provided by a couple of Hammond chokes, one for the amps and another for the cMP. I'd also had a few days off for the US Memorial Day holiday and I spent some of the time catching up on my system tweak backlog... rebuilt the speaker crossovers (now all film-caps), braiding a Rylands-inspired extension cord for my dirty supplies, rebuilding my m-board and dirty supplies with larger filter caps (15kuf-> 47kuf), plus a few tweaks to the m-board supply regulators. These all helped make the system a bit more natural and revealing for the comparison.
On the GA-EG45M-UD2H... I did a new WinXP install for it, disabling the un-needed devices in the BIOS before the install to minimize un-needed devices that needed to be disabled later. I've also gone a bit farther on some of the other customizations, going after all networking drivers I could find in Autoruns and disabling all USB devices except the one providing the connection for the wireless mouse and touchscreen. My latest BIOS settings for this board are on the cPlay 2.0b25 thread and as I said above, I'll publish a lessons learned post in the near future. Also in preparations for this comparison, I updated its BIOS to the latest F3 version from the Gigabyte website, which didn't add any new settings or make any sonic difference as far as I could tell.
To setup the comparison, I configured two harddrives (same models ordered and received at the same time) with a Windows setup for each board (this is made easy by having a separate system partition on the drives AND using imaging software to make and restore system partition backups. Then when I put in the G31, I spend a few hours first updating (to F6), then re-configuring the BIOS (this was not the board I used for my cMP before, but the one out of my ripping machine) and performing some of the additional system configuration tweaks I'd learned setting up the EG45M. But ultimately, I need to do a fresh install of WinXP for the G31M... it was my first successful cMP setup and and I now know some better ways to do the install and configuration that will make it a slightly cleaner install (that may sound a little better too, I hope)... but see below, it was pretty good without doing that.
And I stuck with cPlay 2.0B25 for the comparison to limit the variables, as B26 had just become available and it looks to need some tweaking to perform optimally with SOX (my next project!).
So enough of the preamble... how'd they compare?
On the side of the GA-EG45M-UD2H, as I'd mentioned in my preview, it has a very robust sound... bass and midrange sounds are strong and full. This was noticable from the start and was enhanced the BIOS settings were moved towards optimum. But also as I mentioned in that preview, the highs and level of details lagged behind the G31M.
On the side of the GA-G31M-S2L, first, even though the EG45M has a very 'robust' sound to its bass and mids, the G31M does not sound deficient in those ranges... it's just that the EG45M has a stronger sound there. And the G31M did have more delicate and detailed trebles which provided an increased sense of details and a "see-through' quality to its sound... as I have mentioned in the past, the cMP with this board gives even average recordings a sense of being a direct-to-disk recording, something not evident with the EG45M. But I was surprised by what I had not remembered about its sound... I get a much more engaging and emotive sound out of the G31M. PRAT and drive are better... I find myself being draw more into the music and wanting to listen to more and more recordings... something I didn't get with the EG45M. Bass was very detailed and tuneful and the overall sound is very dense with details... yet all are there and separate from each other. During the setup and breakin of the EG45M, I'd gotten used to its sound and forgotten about that aspect of the G31M's sound. I should say that the EG45M is not bad, but in my setup, with my listening bias, it does not provide the magic that the G31M does.
While doing the comparison, one thing I noticed was the difference between processor chips... I kept the one I've been using in my cMP in the EG45M and used the one from my ripping maching in the G31M. That one would not allow as low of a VCore setting as the other... .77500v or so was about as low as it'd go. So I swapped processors between the boards... no significant sonic differences that I could tell (one is a 7200, the other a 7300) except that the 7200 goes down to 74375v and may go a bit lower and this provides some subtle, but important benefits in the darkness of the background.
I did try turning back on 'Spread Spectrum' on the G31M board to see how much of a difference it makes in the sound... it made a noticable dent in the purity and clarity of the highs and details, but it didn't make the G31M sound like the EG45M in the treble. If I get time over the next week or so, I'll try duplicating the EG45M's processor speed settings on the G31M to see if that is the cause of the differences. The G31M will give a stable processor speed of 900Mhz while the EG45M bottoms out at about 1200Mhz... this along with the absence of a setting to turn off Spread Spectrum are the two most significant (AFAIK) omissions in the EG45M BIOS settings. cics provided a few welcome suggestions to the EG45M BIOS settings which are noted in the cPlay 2.0b25 thread that eliminated some of the other omissions I could see.
Bottom line... the GA-G31M-S2L is staying in my cMP, I find it more musically satisfying and engaging. I put my $ where my heart is... there's a vendor with a few of these boards on Ebay and I purchased another as a spare. I really, really hope that we can get an update to the BIOS of the GA-EG45M-UD2H board as the delicacy and PRAT of the G31M combined with its robustness in the bass and mids would be killer... but for me, the G31 has the magic, the EG45M does not.
Greg in Mississippi
Edits: 05/31/09 05/31/09 06/01/09
Carcass,Thanks for filling me in on this. DANGED, another setting we need on this board!
Have you tried using 3rd-party software like Double-Dawg (http://www.mark-knutson.com/dawg/) to make these settings changes?
cics,Thanks for the suggestion. I will give that a try on the GA-EG45M-UD2H (and also try the Mushkin and HyperX RAM again), likely this coming weekend. And of course, I'll report my findings.
Greg in Mississippi
Edits: 06/01/09
Ok... put DoubleDawg on the cMP and with the G31M on my fully-configured cMP, there are no other PCI connections shown except for the Juli@ card.
That's a good thing, right?
Greg in Mississippi
You need "PCI Latency Tool" - link below.
This latest version 3.1 is not the best, in the way it 1) doesn't always show the soundcard itself, and b) installs a device under Hidden devices. But it's still useful, since it allows you to see all onboard devices - I guarantee you'll see at least 5-6 of them, more if USB is enabled etc. Not all of them are programmable, those that have default setting of 0 are most likely not.
I actually use version 2.7, which should be still available somewhere on the 'Net.
Ok... I installed the PCI Latency Tool 2.7 on my cMP tonight and it showed several (6, I think) more PCI devices, but all already had a latency of 0. Only the sound card was at 128.
A couple of them were devices I'd disabled while doing the cMP optimizations.
Am I good now?
Greg in Mississipp
So, is it true for both G31 and G45?
I only tested the G31 last night. I'll publish results with the G45 when I get it back in the system... now looking like a week or two out.
Thanks for confirming that I've got the PCI Latency setup correctly now and also thanks for educating me on how to do that and what tools to use.
My only disappointment is that I didn't have a sound quality improvement to gain!
Greg in Mississippi
Thanks for the info.
I've located what is supposed to be v2.7... will try it tonight and report results.
Greg in Mississippi
I'm suggesting this as this mobo forces a higher RAM speed and your current 3-3-3-5 setting may be too aggressive.
At the BIOS level, that is correct, both m-boards provide a single setting for PCI latency.
Of course, if you're using one of these boards for a suggested cMP configuration, that is not an issue as you have just one PCI card being used.
Is this an issue for how you're using it?
Greg in Mississippi
... also setting latencies for all other (on-board) PCI devices. In my system, based on "old-fashioned" Biostar mobo, you can easily hear an improvement from setting latency for couple of devices to 0 (directing them to release bus immediately), in addition to setting sound card to 128.
As I understand from some other posts, neither of these two boards allows setting PCI latency individually by device using PCI Latency config tool - although they allow setting it for all devices in BIOS (cics recommended 128).
Is that correct?
Refer to ReadMe.txt for important details & specifications.
Change Log (2.0b26 release):
- Increase support for large WAV files to 4GB (from 2GB)
- SoX 14.2.0 resampler implemented. Settings allow for VHQ (Very High Quality) or HQ (High Quality) converters. Important resampler effects includes phase, bandwidth and (above bandwidth) aliasing.
- DSP and ASIO refinements
- Added SSE3 B9 version for CPUs supporting SSE3 but not SSSE3. SoX resampler is not compatible with a compiler optimisation. This impacts SSE2 and can be avoided by using SSE3 or higher versions.
- UI allows for most settings ( excluding ASIO, Output Rate, Buffer & AWE ) to have immediate effect. "Apply" button in Settings applies new settings during playback enabling easy comparisons.
- Improve UI (new icons and enhanced Settings window)
- Enhance gapless processing and background loading
Please REMOVE previous versions before installing cPlay 2.0b26. Normal (SSE2), SSE3 B9, SSSE3 B9 and SSE4 B9 versions are available. Running cPlay on a CPU not supporting the required instruction set will cause cPlay to exit immediately.
For cMP² users, review these BIOS changes (underclocks FSB, CPU & disable additional USB items). New settings are in effect allowing for lower power consumption (below 20W). Recommended Gigabyte mobo GA-G31M-S2L is no longer available, use either GA-G31M-S2C, GA-G31M-ES2L or GA-EG45M-UD2H ( advanced settings discussion ).
SoX 14.2.0 Settings
SoX implementation in cPlay enjoys full optimisation as that for SRC (Secret Rabbit Code). This includes 128bit DSP processing and other optimisations. Only VHQ (Very High Quality, 175db rejection aka Stop Band) and HQ (High Quality, 125db rejection) converters are supported. SRC at 145db SNR (154db rejection) and 121db SNR (120db rejection) remains as before. SoX VHQ offers better than 170db SNR performance (see measurements below). Other SoX resampler options available in cPlay are (as per SoX Manual):
- Phase (0-100)
All resamplers use filters that can sometimes create `echo' (a.k.a. `ringing') artefacts with transient signals such as those that occur with `finger snaps' or other highly percussive sounds. Such artefacts are much more noticable to the human ear if they occur before the transient (`pre-echo') than if they occur after it (`post-echo'). Note that frequency of any such artefacts is related to the smaller of the original and new sampling rates but that if this is at least 44.1kHz, then the artefacts will lie outside the range of human hearing.
A phase response setting may be used to control the distribution of any transient echo between `pre' and `post': with minimum phase (0), there is no pre-echo but the longest post-echo; with linear phase (50), pre and post echo are in equal amounts (in signal terms, but not audibility terms); the intermediate phase (25) setting attempts to find the best compromise by selecting a small length (and level) of pre-echo and a medium lengthed post-echo. Note that phase responses between `linear' (50) and `maximum' (50..100) are rarely useful.
- Bandwidth (90-99.7%)
Band-width is the percentage of the audio frequency band that is preserved. A resampler's band-width setting determines how much of the frequency content of the original signal (w.r.t. the orignal sample rate when up-sampling, or the new sample rate when down-sampling) is preserved during conversion. The term `pass-band' is used to refer to all frequencies up to the band-width point (e.g. for 44.1kHz sampling rate, and a resampling band-width of 95%, the pass-band represents frequencies from 0Hz (D.C.) to circa 21kHz). Increasing the resampler's band-width results in a slower conversion and can increase transient echo artefacts (and vice versa).
The -s `steep filter' option (99.0% bandwidth) changes resampling band-width from the default 95% (based on the 3dB point), to 99%. Band-width values greater than 99% are not recommended for normal use as they can cause excessive transient echo.
- (Above Bandwidth) Aliasing
Aliasing above the pass-band is allowed. For example, with 44.1kHz sampling rate, and a resampling band-width of 95%, this means that frequency content above 21kHz can be distorted; however, since this is above the pass-band (i.e. above the highest frequency of interest/audibility), this may not be a problem. The benefits of allowing aliasing are reduced processing time, and reduced (by almost half) transient echo artefacts.
Resampler Measurements
All output files are generated from cPlay. Both 16/44.1 and 32/44.1 1 kHz test tones are used:
Upsampling 16/44.1 > 192 (SoX -vLb99.7 on left and SRC 145db right):
A closer look at the bandwidth option using 16 bit test tone (noise floor ~130db):
SoX at 99.7% offers all frequencies up to 22kHz (max is 22.05kHz). SRC offers ~96% and cannot be changed.
Upsampling 32/44.1 > 192:
Downsampling 32/192 > 44.1:
SoX SNR is better than 170db whilst SRC is better than 150db (notice small artefact ~150db above 20kHz).
Phase measurements done using a 2kHz (2 cycles per ms) 24/44.1 test tone. Transient is a sudden volume increase (50db) from -90dbfs to -40dbfs for 5ms:
SRC output at 192:
Both pre- & post-ringing is equally distributed.
SoX output at 192 (minimum phase):
No pre-ringing but greater post-ringing. Overshoot and ringing level is greater than SRC (due to higher rejection db, higher SoX bandwidth and no aliasing). Note that:
- SoX non-linear (non-50) phase setting affects HF phase response of original input signal (i.e. a trade-off is made wherein increasingly higher frequencies undergo greater phase shift). Intermediate setting (anything between 0..50) offers better balance regarding this trade-off.
- Observed ringing is beyond human hearing when "min(input, output)" is 44.1k or larger.
- Ringing effects can be further reduced when "alias" setting is enabled. This results in a trade-off wherein alias artefacts above Nyquist cut-off is added, e.g. for 44.1k input and 192k output, distortion above 22.05kHz occurs. These levels are low (below 100dbRMS) and not in the audible range.
Hello, everybody! I know that many people like resampling to 96, or even 192. As I work in graphic design sphere, I often come across images, resampled by other people, who do it to conform to printshop standards. As our wave files can easily be represented, and they are (in professional programs like samplitude), as a black&white image, it is very easy to understand with the above picture, what resampling does to original image (wave). On the picture You see a detail of my cat. To the left - non resampled original (16 bit 44,1 kHz), to the right the same image (resampled 400 % - read to 96 kHz) The arrows show artefacts of resampling. Take a close look. Resampling was done in Photoshop, the king of professional image processing, so it can be compared to Sox or SRC.
If you watch carefully you will certainly find more. To better see the difference, save the image and zoom in on it in picture viewer program.
This is by no means to discourage You from resampling, I thought it might be interesting to represent the fact this way
Serge.
IMO the pleasures are not dubious. Similarly, if something looks or sounds worse then IMO the shortfalls are not dubious. Only if one cannot discern whether something looks or sounds better is anything dubious.
The good thing about cPlay, as I understand it, is that upsampling is an option that can be bypassed. You may find that your DAC has better playback at 24/96 than 16/44.1 or 24/192. You may also find that software upsampling is superior to hardware upsampling and perhaps you prefer software upsampling with a NOS DAC or no resampling with a NOS DAC. On the other hand do you really want to resample in both software and hardware and does it sound better?
It always seems to come back to implementation is everything.
Take a low-res pic which would be our 44.1k. Then take a hi-res pic (@ ~4x) to be our 192k version. Yes the upsampled results are as per the hi-res picture if strict bandlimited interpolation is done (SRC does this; I'm not able to verify SoX but their site suggests so).
At issue with the upsampled hi-res pic would be added artefacts ("ringing" or more correctly "Gibbs phenomenon"). New information is added wherever we go from low to high intensity (i.e. discontinuities). This means at these boundaries the low intensity area now contains information that's NOT visible to us. This would be in the UltraViolet (UV) spectrum. Aside: Radio, TV, Microwave & Infrared are lower down the spectrum to visible light. Beyond UV are harmful X-rays and Gamma-rays.
So does this invisible artefact affect the visual? Experimentation with SoX confirms this to be true. SRC offers no settings but has a redeeming feature in that the pre-ringing (same for post-ringing) is just ~0.3ms in duration. SoX offers settings that reduce the pre-ringing but at the expense of changing some aspects of the visual (phase shift). Although the pre-ringing is short, the levels are relatively high (some 36db lower than transient peak) bringing into question how well your speaker's tweeter can handle this ultra-sonic load. Should this be at or near the break-up or resonance frequency, your tweeter is most likely to loose the plot...
Hey serge,
Forgive my ignorance, but I thought all this depends on the dac. Some handle different rates better dont they?
Hello! Yes, I think that generally resampling takes place within dac but not in case of cMp-cPlay, I believe, where it is done by src or sox and then fed to the dac to be used for conversion unaltered. The process for resampling can be different depending on the math. In graphics, which, I repeat, is very similiar to wave, the math can be bicubic, bilinear and some others, also different dithering (even the terms are the same) processes are involved like bayer, fatting, floyd-steinberg math, diffusion, dispersion... Next, if I remember correctly, some DAC manufacturers employ tracing - the process that converts our wave into scaleable peak which is closest to analogue (according to their decision and or understanding). IMHO one of the best but hardest ways to achieve the best sound is to try to read and convert to analog every bit to the last one, but I am not an engineer nor am I a program developer, so I may be mistaken, but resampling does NOTHING to bring us closer to truth. I will prepare more pictures with dithering and vectorization (bezier) examples, used in graphics that closely resemble what is happening to the wav within the dac or resampling program (if it is interesting). Also, once again I want to say that I am not trying to discourage anybody from resampling, nor am I ungratefull or trying to say that Sics is doing something wrong.
Serge.
what the heck does that have to do with anything. You are listening to a live event through speakers , in a different room, with different amplification, at different volume, at a different distance, with different cabling, in digital, with different DAC"s, and different computer chips, different versions of cPlay (or other players), oversampling DAC's....etc.
It's a friggin representation of an event.
You should know that with your photographic endeavors. Fidelity is a goal but it is more than pure specifications.
If every system sounds different what is TRUTH?
Now remember that I am sensitive to this being a professional photographer. I hear comments like "is that the way it really looked?" all the time. I try to explain that the slide film I use responds differently than the eye to a scene and the Ilfochrome has it's own character which is dependent on the illumination etc etc. Even the most accurate photograph is a representation and fine art is an "Artist's" representation.
So everything is but a representation and higher bit depth and sampling rates are the way of the futuure. Right now the playback hardware is ahead of the media. Non oversampling DAC's will probably never be made again in light of HD audio. So we live with it and find better ways.
Bob
www.PlateauLight.com
D,
I try to explain that the slide film I use responds differently than the eye to a scene and the Ilfochrome has it's own character which is dependent on the illumination etc etc.
I think that is what I was getting at. I thought some dacs like getting info at say 24/96 while others might be better off at 24/192.
I mean the initial wav file. Nothing more.
Serge
I am a graphics pro also.I agree with your assessment however I combat the fuzzy uprezzing by grossly over rezzing and then downsampling.
It is the equivalent of sampling to 352.8 then do a DC offset and downsample to 192.
Of course this is in Photoshop and I have no way of doing it for audio.
Just and interesting idea
Bob
www.PlateauLight.com
Yep, I agree, by upsampling to 400% and then viewing the result in 25% image size, Serge has in fact do two resampling processes. I can usually accept images downsampled using Photoshop internal function, but for upsampling I prefer it done with external plugin such as Genuine Fractal Print Pro. I prefer editing wave file with Wavelab software, but for upsampling, I prefer SRC.
Sopian
yes I agree upsampling de-focuses the sound just like it does in the picture. but I recall what j gordon holt said when somebody told him that solid state devices sound more accurate than tubes to which he said then I like tubes better than accuracy. I go through periods where I like 16/44 better than 24/192. Its all an illusion.
My schoolboy's limited understanding of the vastness of Hindu culture, I do wish I could take the time to truly understand, but the Hindu concept of ILLUSION is one of the most important concepts in coming to "understand" the world.
All is illusion and one makes of it what they will. This hobby is about illusions within an illusion, no one knows what is really TRUTH, it is illusive and its discovery will do no one any good, or harm for that matter, for it is meaningless. Especially in this context.
We try to find something that allows our consciousness to experience an event, an event we most likely did not attend in a hall we most likely have never entered, through microphones and recording equipment we will never own, blah, blah, blah.
I think steppe had a clever insight, but not clever enough. I know there is no "bad" intention on his part, it was interesting, just not very useful. DHT for ME's analogy was far more accurate, to my mind.
This is an interesting discussion but it leads nowhere. With such bare bones as RED BOOK we have to find a way to fill in the blanks and a best version of the "blank filler" will come. Just as with the LP, no one heard what we are able to hear now with LPs at the beginning of digital era, some amazing refinement occurred after what was the end of the LP's commercial life. Who knows what will happen with the CD, but all of us have lots of them and we would like to hear what is on them as best we can, as it was for those with large LP collections twenty years ago.
My fear is that the future, except for a few specialist exceptions, is moving towards even less information, not more. In ten years we might be pining for the CD to return. I would bet someone will find a way to upsample/extrapolate this information into something much better than we would have ever thought possible.
Bye,
Rick McInnis
Hello, Theo! You have found exactly the right word - de-focuses, it softens, washes away the edge. Unfortunately I cannot post a bigger image - or tiff image, where everything is much more explicite - it won't fit on the screen here. Also an old CRT display (which I cherish) is much better at such subtle differences. If we zoom to highier percentages the "96 kHz" right side picture looks awfully washed-out.
I personally prefer no resampling .
Serge.
I just compare SoX VHQ Linear 99 vs Izotope, 44.1 -> 192, and I feel that SOX online upsample is better than Izotope offline upsample. SoX has lower distortion and background is quieter. So I get a better imaging and resolution is higher. I use Lynx aes16e-> Mytek 8x192-> Beyer 990. Result is the same when I switch to Chord+MUBE.
My perception has always been that online upsample cannot beat offline... so I will try a more days. If SOX is confirmed successful then I can delete all the 192 up-sampled wave files (300MB+ each). SoX is not a CPU hog. My E7200 runs SoX without pops and clicks at 800Mhz (Asus P5KPL-CM) and heat sink is cold as room temperature.
Otherwise b26 sounds just as musical as previous versions. I guess this is the result of good ASIO integration.
Gee I thought B25 could not be improved!
Just received an email from a cMP² user on a similar experience as yours, i.e. Online betters Offline.
Which filter do you use with Isotope?
Sorry ackcheng I need to tell you later... My main PC just doesn't boot anymore after I insert the E7400 :( I need to fix it...
With SOX, how the sample rate conversion is performed?
Is it done on the fly?
Or is it first complete the conversion and then play?
I think it may sound better if it is first converted to the target sampling rate before it is loaded to the memory and play.
Thank you.
Given that there's less RAM processing with realtime upsampling, it would be better. It would be interesting to compare batch converted SoX file vs realtime upsampling.
How good is SOX? How SOX compare to leading sample rate conversion software such as SARACON by Weiss if I use it to upsample from 44.1kHz to 96kHz?
The SNR is one of the best. SARACON uses linear phase filter which SOX allows. If you look at the comparison site, SARACON is not that good afterall
Being a lazy man who does not much want to find out for myself; I follow the leader.
I hope this is the one that will keep me from wearing out my cartridge!
Thanks,
Rick McInnis
For starters, I would remove the preamp altogether (maybe give that cartridge a well deserved rest).
I've settled with B26 SSE4 B9 with large buffer setting (E7xx processors have 3MB L2 cache - think of this as a free resource so exploit it). I use a single Kingston 256MB ValueRAM set at 3-3-3-5.
SoX is very good and certainly has the upper hand with transient performance. On some recordings / genres it may be preferred. SRC offers better tonal decay and does a marvelous job of preserving those delicate harmonics. Overall SRC (@145db SNR) is no slouch on transients but its harmonic purity makes it a perfect choice for long term listening. It's like tubes without the haze.
Nothing would make me happier than to not have to depend on the tweaky turntable and cartridge!!!
Nothing like a click of the mouse and then music is presented.
I am looking forward to hearing if that CD ripping software makes an audible difference.
I have yet to install #26 so I am confused about some of what you said. I guess one can go back and forth between the SoX and SRC and make a choice? All of this is attractive even though I will look forward to their being a unified, this is best, configuration. At the same time I will take delivery of an absolute best phono cartridge!!!
I admit, I am lazy, though once one finds a way to optimize you are compelled to use it. At one time I wanted to believe polarity control was ridiculous, until I learned to hear it.
Have you implemented some of the power supply options the faithful have offered? Still using the onboard DAC of the JULI@ or back to the AERO CAPITOLE?
If I was a cat I would be dead.
Thanks!
Rick McInnis
To clarify, I prefer SRC@145db - harmonics & tonal decay is superior to SoX. If I use SoX then I set it to intermediate:30, bw 96% & no alias. Other settings are also good like minimum, bw 97% and alias.
Setup still the same (Scarlatti DAC used with 2 cMP²: RME / Toslink / 96k & Lynx / Dual AES / 192k. AA Prestige SE & Scarlatti Clock collecting dust).
I use Juli@ for development & testing. It still offers best value for money but I wouldn't rely on it to better your TT (and new cartridge). Perhaps doing GStew's mods together with I2S feed to ESS Sabre's 9012 Stereo DAC will suffice. Clean and isolated power to Sabre's digital and analogue voltage inputs is a must.
interesting that you chose large buffer vs small (opposite of my choice). I'll try large. Is there any cpu load differences between large/small buffer? I have several resampled sox files I will listen to both at same parameters and post.
I got the loudest version of metallics ever today while listening to juli@ analogue outs. Who knows maybe I have a real ghost in my machine. I listened all day yesterday and only ocassionally got slight static-y sound when loading ram sometimes--not gross. It was not constant but when I rebooted the 'load static' went away for most of the day. Today the metallics completely obliterated the music. Had to reboot and it went away. I'm running cpu volts set @ .85 VOLtS (READS .800 VOLTS ON CPUZ), 157 host clock control. Maybe I'm at the limits of my machine. Really thought I was going to be ok with sox. The gross metallics started when I tried to go to a prior track during cplay playback. BTW when I compared pre prepared 192 sox files with realtime sox upsampling in cplay the sonic results were almost identical. could not specify a preference. Cpu usage never went above 28% while running sox 192. I can't understand why this monster defect keeps coming back.
The first test involves increasing Juli@'s latency to a maximum 128 samples (that's the maximum I would recommend for 192k output). I suspect you have tested this and got metallics.
Second test is to never jump directly to a track, instead hit Next till desired track.
I want to use Juli@'s I2S output to ESS Sabre's 9012 Stereo DAC (133db SNR) but this problem bugs me.
Did not like 128 latency but I have settled on 64. Sounds good plus I seem to be having fewer metallics as my linear ps on p4 settles in. Also I'm running sox at 192 with alias box checked. This means no filtering of aliases right? If right that means lower cpu load.
I have tried no mouse navigation within a file and for 2 days or so I've gotten 1-2 metallics a day--not bad. Now I will try 128 latency with the juli@ control panel (in past I've preferred 48 sonically) but I assume this reduces cpu load.btw sox upsampling within cplay is better (by a small margin sonically) and infinitely better for parameter setting and disk file usage. I've just saved 15 gigs of hd!
so thanks again for a great cplay 2.0vs 26 release!
Edits: 06/04/09
I ended up setting cpu voltage higher (.896 on cpuz) and host clock control to 153 and it seems to be better (soundwise and metallics wise). Thank you ray konkle for this suggestion.
I got metallics a couple of times while trying to change tracks with the mouse (sorry I just forgot). And although I had to reboot it seems not as bad while running sox at 192.
Also I implemented a Power One power supply (12 volts. 1.7 amps thank you sondale, mark, robert, ray and gstew) for the p4 and it sounds really good (I'm sure it has more breaking in to do) but it also seems more metallic resistant. I have to figure out where to put it now.
But all in all I may be getting a Lynx 2b. Hate to abandon the juli@ but it is really a distraction to reboot all the time. If I can get less than one a day or so metallics I could live with the juli@. We'll see.
I tried only the 'next' or 'previous' buttons or keyboard for navigating within a file all day yesterday and no got metallics (more below in response to Bertel). When I was actively working as an engineer doing problem solving we used the technique called 'Is/Is not' map. It simply was the listing of conditions that are associated with problem occurence or non problem occurence. With cplay version 2.0v18 I never got one experience of metallics ever-over maybe hundreds of hours of use. I went back to it several times just to listen to music during 'heavy metal' days and it never failed in this way. Yes cplay versions higher than 18 are way way better sonically but all were subject to metallics. Not trying to blame cplay or cmp this is just a statement of fact that may shed some light on what is happenning.In addition to only using keyboard for within file navigation I am currently @ 157 hcc (was 155 the day of the big metallics)--so far after 1.5 days ok.
Edits: 05/31/09
Cics,
I can already comment on the second test:
My version of the metallics occurs even when cplay just continues playing and moves to the next track - then (at random) metallics can occur or not. I restart the track, sometimes once more, sometimes again, rarely again, and sooner or later the track starts without metallics. I could discover no pattern whatsover yet to give me a hint on possible causes (other than the combination of latencies maybe).
However, since I am currently focussed on getting the system optimized for Buffalo32s first, it isn't an issue for me for the time being, i feed it with 16/44.1 since it does upsampling anyway. Once that is running fine I'll try 24/192 and will report.
Best,
Robert
you said '...My version of the metallics occurs even when cplay just continues playing and moves to the next track - then (at random) metallics can occur or not. I restart the track, sometimes once more, sometimes again, rarely again, and sooner or later the track starts without metallics. I could discover no pattern whatsover ...' --------- was very close to my experience before I went 2.5" hd, fanless cooling except ----when metallics were bad (real bad) only a reboot would make them go away. Now that I am into 2.5" hdd, fanless cpu cooling and using sox I get metallics much less frequently and I get the kind that are less than 'real bad' (except for other day). yesterday, with using only keyboard to navigate within a file I got none. Also with sox I have not yet gotten any at auto track changes or loads.
Robert I found out that cpu voltage and speed settings affect my metallics. I am right now at .85 volts (as set in bios) and 157 host clock control. I'm sure your equivalent but very low settings may impact your metallics.
what's very frustrating to me is I can hear more purity @ 150 hcc but I get the problem more frequently plus sometimes I get instabilities (system shuts down). So finding the sweet spot is perhaps different for all of us. One thing though I believe impacts metallics, is power supply purity. I am using the P4 Ryelands cap mod which is great sonically but I'm not sure it makes my sys more 'metallic' resistant. Others with linear external ps P4 feeds may want to comment how it may effect metallics. As I recall you use a linear ps for P4 so maybe it just depends on voltage/speed levels plus whether you use juli@ or not.
one last thing I have a benchmark too but I like juli@ analogue out better. When using Lynx did I understand correctly that you can only go digital out with an AES cable? Is there no i2s out with Lynx?
Theo,
I am at 0.76785 (not sure about the last three digits, can't look it up since I am not at home today) and host clock at 150MHz, that's stable but lower values don't work (have to restart from BIOS). At that values everything was fine with Lynx AES-16.
IMHO power supply purity does not help to cure the metallics - I am on pure battery power with well-sized caps (so I think) for P4 and P24 and got the metallics just the same as when I power everything with the Antec (I have tested that already). Maybe things are different when I power Juli@ externally as suggested by Alfred/sonics, will test in the next few days.
However, my task for the next few weeks is to go from Juli@ via i2s to Buffalo32S and optimize power supply (and possibly also caps and resistors?) in that chain. Since Buffalo32S does upsampling I output 16/44.1. Once that is running fine, I will switch to 24/192 and 24/178 (thanks Jean) and see if the metallics are back and how stable and effortless Juli@ can perform. Will report.
And yes, as far as I know you can only go out digital from Lynx with their breakout cable that provides you with an AES/XLR plug which I ran into the Benchmark DAC1, worked absolutely stable and reliable. I tried both to find any possible way to (i) go out i2s and (ii) power the AES16 externally, but failed with both (probably could find a way for (ii) but (i) is impossible AFAIK, Lynx techs have confirmed this). Otherwise I would happily stay with Lynx without any doubt.
Cheers
Robert
thanks for info.
Ok I am trying not to ever use my mouse again--using keyboard to navigate. I was planning on getting a buffalo32 also but I guess you're saying metallics might interupt that too. I will try juli@ latency at higher values than 48 and see what that does (I know 48 sounds best). does that help alleviate cpu load?
Theo,I'm afraid this doesn't really help to solve the issue, but I thought I'd just share:
I am a late Juli@ convert, have been running my highly optimized system (at least so I believe) until last week with Lynx AES16. I left everything as it was (e.g. 256MB ValueRam, E7200 with CPU voltage at 0.76875 I believe, all optimizations done etc.), just took out the Lynx card, uninstalled the driver, plugged in Juli@ and installed the unified driver 1.05 - and instantly and for the first time I had the pleasure to get to know your metallics - it's dreadfull...
Since I'm currently going Buffalo32S (which does resampling and reclocking, something I definitely don't like...) I went back to 16/44.1 and everything is fine. So this is just to say that a completely stable and (for me) close to perfect system tipped over just by exchanging Lynx with Juli@. I definitely have to have a Juli@, since Lynx can't do i2s and external power (and cics said its ASIO implementation is so fine), but I don't really have a doubt that it is either Juli@'s driver that screws up, or Juli@ hardware - now the latter was my theory before you said that you have the metallics even through Juli@'s analogue outs, I now expect to get them even over i2s, still to verify, if so it can't be the digital out part. Ah yes, and I had the idea of experimenting with PCI latency setting in BIOS, don't know if that could make a difference, haven't tried yet but could make sense.
Just sharing another experience...
Robert
Edits: 05/29/09 05/29/09 05/29/09 05/29/09 05/29/09 05/29/09 05/30/09
thanks Robert. Maybe the solution is the Lynx. I know its expensive but I bought power cords more costly. What is you theory about pci latency? Do you think something other than 128mb might impact it. Are you running a ssd hard drive? I am happy to hear you report it. If more people taslk about it the chances of a fix go up.
Theo,Lynx is definitely a solution - if you can live with XLR output to say the Benchmark DAC1 (as I had before, that was good and so stable and absolutely hassle free) and do not fancy i2s (like I do now to feed the Buffalo32S). On the other hand, once you go Buffalo32s you only need 16/44.1 since it does upsampling and reclocking anyway, so there will not be a problem. I am intending though to build my own solution based on ESS 9018 chip later that year (8ch dual mono configuration) so I've got to find a solution to this, but fortunately that's still far enough away ;-)
Well, my "blaming" PCI latency is in no way scientific or based on any facts yet, just a gut feeling - my metallics occur whenever I start a track (when cPlay set to SRC @ 145.68db and 192.0), it's either there or it isn't, or maybe it also starts with glitches or dropouts - I just have to restart the specific track, it either reoccurs or is gone, no pattern foreseeable. I'm just guessing, but looks to me as if processor's output and soundcard input are not in sync, and that as far as I understand it can only be driver or something "on the way", during transport, i.e. PCI and its latency I guess. As I said, just a gut feeling, haven't had the time (and need due to 16/44.1) yet to really check, one would need to simply try for some time with a setting other than 128. However, my metallic issue was definitely introduced with Juli@ and its driver, everything else has been there and unchanged before and has worked perfectly stable and without any glitches or dropouts or even any distortions noticable.
And yes, I have a SSD only.
Cheers
Robert
Edits: 06/03/09
Seems unlikely to me.
I know I have a DAC with the BURR BROWN upsampling chip and my set-up sounds MUCH better with the computer making the calculations.
It seems implausible to me that any chip could do a better job than a powerful microprocessor and these very clever upsampling software choices.
Have you tried it both ways?
I had the metallic problem when using the computer after 8 weeks of being turned off and installing #25. After it settled in it never happened again. To be honest this was only over a two days period.
After Theo has exchanged so many things I am seriously thinking it could be the mouse. Are you using the cics recommended LOGITECH? I had my temporary problem when making some quick mouse clicks (with the recommended device). Could a somewhat recalcitrant mouse send confusing commands to the computer creating this condition? I know I am making a stretch, but this has gone on so long there must be something less than obvious at work.
Bye,
Rick McInnis
I would have never thought a mouse could be a source of problems but I use a p/s2 type mechanical roller ball mouse with a 6 foot extension (to reach my listening chair). Is this a problem? If I could navigate the music library with my keyboard I'd get rid of the mouse.
when I thought about how the problem was triggered in my set-up it isn't completely ridiculous. A baffling solution to a baffling problem?
I would think you would find this easier to use:
http://www.newegg.com/Product/Product.aspx?Item=N82E16826104080
This is what cics uses (or did use when he recommended it to me). I have always tried to use what HE was using to assure I was getting the same end result. One would think a hardwired mouse would be less likely to misbehave, BUT at the same time you have this antenna in your room and no telling what could be happening.
Even if it doesn't settle the problem it is a pleasure to use.
I know you might be starting to think I work for NEWEGG as much stuff as I have told you to buy! My email address is not a ruse.
I sincerely think this is something worth trying. After all you have tried EVERYTHING else. (If this works I think Stephen Jobs should give me a job at APPLE even though I hate their machines)
Bye,
Rick McInnis
always wanted to try it (but didn't you get this problem with the logitech?). Rick I cannot thank you enough for always trying to help to solve my problems.
Edits: 05/30/09
and that was in the settling in stage after eight weeks of being turned off and then restart and install #25/disable EIST and trying to lower the processor voltage too quickly/
After being on 24 housrs it did not happen again.
And it was, each time, started by some fumbling with the mouse. Hitting multiple buttons quickly.
At least you could get rid of that wire!!!
Rick McInnis
Rick,yes, I've had this confirmed by the designers of the Buffalo32S, the oversampling FIR filter is also clearly mentioned in the data sheet (although they don't really disclose how the upsampling is done and to which frequency), the ESS9018 chip has what they call the Time Domain Jitter Eliminator which basically is upsampling and reclocking, as with your Burr Brown a technique that is commonly used. The ESS9018 is highly praised for exactly this feature, it is said to be very efficient and to do things very well, but I agree with you, I have yet to hear a solution where a hardware upsampler does a better job (or at least not worse) than a software upsampler like SRC or SoX.
And yes, the chip is designed with the ability of turning this feature off, it requires setting a few registers via i2c, but the firmware used by the Buffalo32S designers does not offer that functionality, you need a custom firmware for that. Surely the Buffalos are a very fine piece of audio design and manufacture, but I believe the chip can do even better. Since like you I do not want the DAC to do the upsampling, plus I do not think that our cMP2 environment produces too much jitter anymore that needs attenuating (the measurements cics had done in March last year were already impressive, and so much improvements have been made since then), I will build my own DAC later this year with two ESS9018 chips in dual mono configuration (recommended by ESS for best audiophile performance) and compare - at least that's the plan ;-)
Best,
Robert
Edits: 05/30/09 05/31/09
Robert,
I'm planning same thing.
I was reluctant to select Sabre unit over TI 1704 for the same exact reason you mentionned, ie internal upsampling / downsampling.
This technique allows the widest compatibility with the different sampling rates sound format availlable (everything from 32k up to 192k)
Although this is common practice in Silicon SRC and good to rmove jitter of common source, here we have a machine and a software that provide us with minimum jitter at the PCI Sound Card output which mean (if well implemented) at the DAC Input.
I'm thinking for my dac implementation to try to achieve as low jitter as possible all the way down to the analog which means the use of a Master Clock close to the DAC chip.
The problem here if one wants to avoid too much complexity implies to select a single as universal sampling rate (178K4 or 192K).
Should you select 178k4 it would be perfect for original files sampled at 178K4 and close to perfect with 88k2 and 44k1 providing that you select the right settings in Sox (that i suspect may be full system dependant).
So the question is what will be the standard sampling rates (if any) for Hi res files pick the right clock for that standard and accept imperfection for other sampling rates.
Now that you explained us that Sabre can be forced not to use its internal SRC, it comes back to interrest to me because of its ability to work with 32 bits resolution.
So, if you move on with this project and if we can share energy, you can count on me.
Best,
Jean
Jean,absolutely great to hear you're heading down the same path - or rather up :-)
Exactly, the first assumption I'm making is that we in our cMP2 environment have eliminated or reduced jitter to such an extent that the downside of on-chip jitter reduction techniques, e.g. hardware upsampling and reclocking, more or less clearly outweigh the benefits of doing the upsampling in software and closely coupling the DAC via i2s - in my expectation this is superior, but we'll have to test and see. IMHO the ESS9018 is currently the best DAC chip around, but need to properly re-evaluate.
Secondly, I'm also with you that it is key to use the best master clock available on the DAC. As far as I can see, that is NewClassD's Neutron Star (http://www.newclassd.com/index.php?page=36), Lars has done an outstanding job with this.
Thirdly, the question of selecting one "standard" sampling rate for everything and optimizing for this is easy for me I thought - since all of my playback is through cPlay, I intended to use 24/192 (or rather its 32bit incarnation through ASIO2) through SRC or SoX exclusively. Your comment on 176k4 though suggests that this probably might be the best choice (another assumption I make for me is that most important for me is data derived from red Book CDs, thus 16/44.1, that's what I focus on), probably wil go for that, need to get more info on this, and then choose clock and settings etc. accordingly.
And lastly but most important for me, I'll be absolutely glad to sync with you and join forces once I get started with this part of the project - look forward to this very much! Please PM me (if possible) so I have your coordinates.
Best,
Robert
Edits: 05/31/09 05/31/09
Hi Robert,
Glad to hear from you.
I've selected 176k4 as the best for CD material for the exact same reason mentionned earlier.
Up to now i could only test on my reference system (i'm hifi dealer) with 44k1 and 88k2, to be honest i didn't even try 96k.
But i suspect that upcoming hi res material will be 192k ...
Reagarding clock i was also interrested in Audiocom stuff. Also praised as the best one in town.
The next important thing is power supply and IV conversion.
I will be playng soon with different PSU for evaluation.
My problem with Sabre is how to get it, as i don't want to use Buffalo, buffalo one not any more availlable and at least no more support, buffalo two is too expensive if you take into account that i consider changing IV and PSU. Second big problem Board design.
That's why i am on the conservative approach thinking to use TI 1704.
Best regards,
Jean
Jean,you're probably right that over time there might be a shift to 192kHz material - for the time being for me this still is almost irrelevant since by far most of my music ("old" one I have as well as "new" one I buy) is 16/44.1 because it comes on CD. Let's see how SRC vs SoX as part of cPlay develop, I'd very much be interested in cics' opinion on what sample rate he sees best for Red Book material (176 vs 192) and for what reason (SNR? etc.)
Audiocom is certainly good - need to ask Lars why he's better :-) My knowledge in clock technology and architecture needs to be expanded and updated before I will decide on which to choose.
Well, power supply - you certainly have seen that I fancy battery plus proper capacitance - can only recommend that, need some extremist spirit though ;-)
Buffalo32S IMHO is not expensive at USD 469 at all, given the top-of-the-hill solution you get with it. I'm currently setting it up as my reference, then quite likely will go with 2 x ESS9018 evaluation boards as a next step (not a bargain at 2 x > EUR900), and then take the experience of those two to get to some "final end-all-search" DAC, who knows ;-)
Cheers
Robert
Edits: 05/31/09 05/31/09
I know we will all look forward to your implementation.
Bye,
Rick McInnis
thanks for your response. I'll try different pci latency values.
very nice, the ability to swap sox parameters on the fly is very usefull for finding a 'personalized sonic sweet spot'. I still prefer M for phase but am now experimenting with bandwidth. Seems like allowing aliasing is a no brainer (for me anyway). I don't have a handle yet on overall sonics (but very very close to 25) so far.
Excellent release! I like it!
There is an interesteing article on SRC and filters here. The author compared minimal phase, mixed phase and linear phase filters
http://tech.groups.yahoo.com/group/acourate/files/Uli/
I got it finally
can't sign on to link can you copy / paste here?
I am not sure how to post the white paper as it is a pdf and is quite large. If you send me your email, i can post it to you
I usually let one of my music servers play 24/7, which means that I prefer that it play my entire music library unattended.
Since cPlay requires cue sheets, I thought I could just combine the cue sheets into one large cue sheet to contain my entire library. But it seems that when I combine too many cue sheets (I don't recall the exact number - perhaps two dozen cue sheets representing 24 albums or 300 wav files), cPlay reports an error and fails to play. If I split the combined cue sheet into two separate cue sheets, cPlay plays each cue sheet perfectly.
Is there a way to have cPlay play a large music library unattended? Can cPlay play all the cue sheets in a particular subdirectory folder?
To play all your cue sheets, setup a batch file that calls cplay for each cue sheet (in cPlay settings, set playback to exit):
@echo off
"c:\program files\cics play\cicsplay.exe" "cue sheet 1"
...
You can list all your cue sheets into a file using "dir *.cue > allcues.bat" then edit .bat file as above.
Why are you doing this?
Thanks!
Answer to why are you doing this?
1. I enjoy listening to music
2. It’s similar to playing the radio all day and night, except it’s my collection of favorites
3. No further effort from me is required unless I want to hear something different
4. Too much music, too little time – unattended, unlimited playback means I hear more music
5. The music collection playback may be structured, but my attendance during any given 24 hours is random – thus I am pleasantly surprised or pleased with what I hear
6. I am used to continuous or random playback features in MediaMonkey and iTunes
7. From the Audio fanatic Arthur Salvatore “Optimizing Digital Components -IMPORTANT ADVICE: Never* turn off any digital equipment. It usually takes around 2 days of playing for it to "break in" and sound optimum. Once turned off, you have to go through the entire cycle all over again. Accordingly, do not judge digital equipment unless it has been on and operating for at least 48 hours.”
8. Because I can and it’s so easy
does the SNR setting have an effect if you are at 44.1?
If input media is 96k and output is 44.1 then downsampling is done using selected SNR. If they match match, resampling is totally bypassed.
Hi all,After reading posts of Gstew (linear PSU), Ryelands (extra caps) and others on power supply (PS) improvements I decided to try their recommended PS improvements on my cics memory player.
I can really recommend GStew (linear PSU) and Ryelands (extra caps) suggestions on PS improvements. All the suggested improvements on the PS made (sometimes big) improvements in sound quality (SQ) in my setup.
First my setup:
Zalhman case, GA-G31M-ES2L, E7300, 2Gb HyperX, Samsung 500Gb, passive nexus cooler, Lynx AES16 -> xlr AES/EBU -> Lavry Back DA10 -> xlr mogami gold -> Klein + Hummel O 300 active studio monitors.
DVD, HDD, USB separately powered as per cics recommendations.
But i didn’t like the thought of running 230 AC (!!) Volt wires inside my PC for powering the Granite PSU’s . So I choose an different way too power them. I use an external desktop PSU. (Search for number 998343 – 89 in the search box at www.conrad.nl. It’s not possible to directly link in to their site)Like Ryelands suggests, if one buys a few of these extension cables
http://www.maplin.co.uk/Module.aspx?ModuleNo=34109
http://www.maplin.co.uk/Module.aspx?ModuleNo=98854
It’s really easy to swap suggested PS improvements in (and out) place. Swapping in and out, also makes it very easy to hear the improvements in sound quality (SQ).* Adding a linear power supply the easy way ( very much recommended !! )
The SQ improvement I most recommend is a linear PSU on the P4 pin. Because GStew decided to build his own linear PSU, its looks very difficult too implement. But if one simply buys a 12 V (low ripple !) linear bench power supply / linear lab power supply, than its realy easy and simple too implement. Within 15 minutes the job is done !
And whow ! What an improvement difference in SQ.
Recommended ! Thank you GStew for the suggestion and inspiration.A simple, but good (low ripple ! ) 12V linear bench power supply / linear lab power supply will cost about 30 – 50 euro and a P4 extension cable 3 euro’s or so.
About the aesthetics. Almost all simple linear power supply’s look like this one: http://www.maplin.co.uk/Module.aspx?ModuleNo=231 .
So very ugly. But since the PC, the external desktop PSU and the linear PSU are stored in a board. I don’t mind.* Adding 4700 uF caps (to each of) the +12 V P4 PS lines on the ATX PSU.
(see also Ryelands post) If powering the P4 pin with a linear PSU is not an option, than I can realy recommend adding caps to the P4 power line of an ATX power supply. This also gives a nice improvement in SQ. And - I really hate to say this- I do hear a very slight (!) difference in SQ between ordinary (brandless) caps and audiophile (Audyn) caps.
I soldered the caps directly to a P4 extension cable. Again one might think it’s a hassle. But again it’s surprisingly easy to solder the caps directly to the P4 extension cable and swap the P4 extension cable in place (preparations, soldering, swapping in ect will take 40 – 50 minutes). Because it’s easy to swap the P4 extension cable in and out, one can easy hear the SQ differences.
Note: the SQ differences where much (!) bigger on the Zahlman PSU (with I used first) than adding the caps to the ANTEX Aerth watts 430 PSU P4 PS-line (which I use at the moment)
Note: adding extra 'smoothing' caps to the linear P4 PS-line, made no difference in SQ at all !
* Adding 1000 uF caps to the P24 PS lines.
Al though its more time consuming it’s also easy to solder caps (to the P24 power lines.
It took me 2 hours to complete the job. It’s a precise and a bit boring job. Soldering so many little caps to al those wire’s.
Luckily no caps are needed on the ‘power good line’ (grey), ‘power supply on’ (green) , ‘standy line’ (purple), and the white line (- 5 V).
Swapping in and out the capped P24 extension cable gave only some (not much) SQ improvements.Summary:
1. The biggest SQ improvement in my setup was: a linear PSU on the P4 pin.
(a must do !! Very much recommended)
2. When not using a linear PSU for the P4 pin but a common ATX PSU:
ad the caps to the P4 pin as Ryelands suggests.(a very nice SQ improvement)
3. If you have the time: you can also put caps on the P24 lines. But that only gave minor improvements
-> Final thoughts on power usage and power supply (sink <-> source)
As already suggested by some other inmates here (I can’t find there posts anymore because I forgot there names): lowering the power consumption might give better sound quality because of less demand on the PSU and the need to deliver less power to various parts.Hence: optimizing the power supply could make the need for lowering the power consumption less important or even irrelevant.
Personality I don’t hear any SQ differences in my setup, when the CPU is running at 0,8 volt, 0,9 volt of 1.0 volt or 1,15 Volt. In my setup, it al sounds the same to me. Now mind you: I do hear minute differences in sound quality in my setup:
- when I leave my keyboard connected tot the PS2 connection on the motherboard,
- between various cPlay releases
- after the kernel optimizations,
- when giving priority to background services, disableling services,
- low busspeed to get faster RAM timings
- when adding caps and or adding a linear PSU,
- when powering DVD, HDD and USB separately, etc, etc
I my setup I can hear all these minute differences in SQ.
But there's only one I don’t hear: that’s under-volting the CPU. In my setup (with 3 PSU’s) I don’t hear any change in SQ….Do other inmates really hear SQ differences between various levels of under-volting?
PC fully cMP2 optimized -> Lynx AES16 -> XLR aes/ebu -> Lavry Black DA10 -> XLR Mogami Gold -> Klein + Hummel O 300
Edits: 10/31/09
allow power from a linear supply to flow.
I had suggested this many months back though I did and do not have the motivation to give it a try. Nor the expertise to know which rails would need to remain attached to the P24.
It does seem to me this would make the process rather easy in comparison the approach mentioned.
Bye,
Rick McInnis
Rick,
I'm really sorry, I have been to quick on this... I re-read your posts on this and know understand that your suggestion wasn't to use relays to simply do the "ordinary" switching over to the clean sources once the system is completely powered up, but to use the ATX PSU to do the exact timing and switch on the respective lines, coming from clean sources, by relays at exactly these points in time where the ATX PSU powers up those lines itself, intended to give power to the mainboard but now actuating the respective relays that thus give clean power to the mainboard at this point in time, right?
Clearly a very elegant approach, my congrats - sorry that I didn't get it the first time I read your suggestion.
On the other hand, as far as I can see at the moment I'm afraid there isn't much gained:
- You still have an ATX PSU around - in your concept not used to power up but to actuate the relays, but still it needs to be there, both our concepts can't get rid of it.
- What's more, in my concept I can turn it off once I have switched over, but when actuating the relays it has to stay up, doesn't it?
- You have to provide the "special" rails in a clean version, i.e. linear or battery, too. 5V is already there, that's not the problem, and you can turn them off too once everything is up, but you also have to provide a -12V line (which I have powered by the Antec during boot-up and simply switch off then).
- As far as I know relays wouldn't provide the ramp rates (don't know how critical that is, could be irrelevant or covered by kind of the "rise time" the relay needs to close, would need to test)
I like the elegant concept of kind of "externally" do the timing and sequencing (that's why I was looking for a module or circuit that would do that as any ATX PSU has designed in), while my switching off/over is rather brutal. But I think I would really like to combine both ways and have the switching off and over done by the push of just one button quite elegantly with relays, I think that'll suit me best.
Again, many thanks for bringing that back into the discussion!
Best,
Robert
Robert,
I agree with all of your concerns and was aware that this would be the compromise - you cannot get rid of the thing entirely, my conjecture is that you do not need to. I cannot think of any uncompromised audio component - there is always something that cannot be ideal. The key is to find which approach allows the least interference from the weak link.
If all of the relays are identical (respond similarly) I think the ramp rate concern goes away since the delays would be consistent across the board, the timings would still occur in the correct sequence, yet, all would be delayed the same amount of time. A phase shift ...
I would worry about switching between the power supplies; this could eventually take its toll, like my experience with destroying a GB MB with turning the HDD on and off while the board was "on". I thought I was saving wear and tear on the HDD when I would leave the same selection to repeat over and over for burning in purposes. I did save wear and tear on the HDD!!!
Back to having to retain the switcher - let's face it: these things are not fatal to sound quality or none of us would be where we are. No need to fear the switcher, just minimize it.
I wish I was motivated to pursue the idea. At this point I am back to analogue. I do not think even a perfect power supply will get me where analogue takes me effortlessly (but noisily it is true). I mean the sound is effortless. I have spent many hours getting the system right.
I have a forty years collection of LP's, if I was younger and did not have all of these I would be far more motivated to get digital to sound really good.
Sorry to be so pessimistic but the leap in sound between my old Theta Jade and cMP is HUGE and important but the crevasse between analogue and digital is planetary in size and I do not think the trickest power supply in the world will get us there. I wish, I wish ...
I look forward to hearing what you come up with!
Rick McInnis
Rick,thank you again for your very much appreciated thoughts!
Regarding ramp rate: I agree to what you say, that if the relays are all more or less identical the timing and sequencing, although delayed a bit, should still work fine. However, the ramp rate is a different issue - it is described in the ATX12V Power Supply Design Guide as such: "There must be a smooth and continuous ramp of each DC output voltage from 10% to 90% of its final set-point within the regulation band, while loaded as specified in Section 3.2.3. The smooth turn-on requires that, during the 10% to 90% portion of the rise time, the slope of the turn-on waveform must be positive and have a value of between 0 V/ms and [Vout, nominal / 0.1] V/ms. Also, for any 5 ms segment of the 10% to 90% risetime waveform, a straight line drawn between the end points of the waveform segment must have a slope ≥ [Vout, nominal / 20] V/ms." I am not sure how strict and restrictive the Gigabyte MoBos are on that, but it sounds not trivial to me and might spoil the elegant relay solution. I for my part want to get the switcher out of the system, and I can the way I do things by just switching it off after the rails are switched over to their clean sources - but that surely is personal taste, or probably extremism and spleen are better words to describe it :-)
Your comment about putting the MoBo at risk by doing the switching over an extended period of time is something I indeed fear myself. It is also for that reason that I consider using relays, I think I remember that they don't always have to produce "spikes" like a switch does (I believe, haven't confirmed yet) when switching over from one source to another, but can allow for a smooth turn-on if designed accordingly. Definitely need to look into that.
Although I am not an analogue guy myself (I bought my last Thorens 23 years ago and "went digital" completely some 13 or 14 years ago, for the theory that digital by its binary nature is always more "right", just as I prefer solid state over tubes for quite the same reason), this is one of the things I definitely want to do when finally some day I consider my personal cMP2 incarnation good enough (the power supply is just one, although IMHO important, area of construction to get there): compare with one or two reference turntable solutions! I look forward to that probably quite revealing experience a lot! :-)
Best,
Robert
Edits: 05/29/09 05/29/09
I do remember reading all of that stuff and being suitably deterred.
Then there is the picoPSU ... when you look at that thing it is hard to believe that the ramp rate is being controlled by that! And when you look into the ANTEC, though far more complicated, STILL hard to believe there is that kind of circuitry contained.
I think all of this is controlled on the board. The supplies supply voltage and nothing else. I believe that the relays would have nothing to do with this happening properly.
The only thing I think one needs to retain the SWITCHER for is the absolute first moment of turn on when there is communication between the 5 volts rail and, I think, the -12 volts rail, after that I think all of the complicated actions are controlled from the MB.
Please excuse the fact that all of this is conjecture. I practice empirical computer architecture as an avocation.
Then there is the other side of the supply question; instead of getting rid of this mode of operation getting someone like the fellow at HYPEX to design an audio grade switcher.
I enjoy the conversation.
Bye,
Rick McInnis
Rick,
am I repeating myself? Ido this with pleasure - many thanks for your highly appreciated critical thoughts! :-)
Well, picoPSU and Antec doing the ramp thing: You don't need big and complicated for this, at least not at an outside glance - e.g. look at Actel's Fusion (a one-chip solution, see http://www.actel.com/documents/Fusion_PIB.pdf) or Linear Technology's LTC2928 (even smaller and "simpler", see http://cds.linear.com/docs/Datasheet/2928f.pdf), there's FPGA's and small stuff that I can very well imagine being on such a tiny piece as the picoPSu - the thing is we don't know and are guessing... :-) We could find out by simply trying.
However, I'm comfortable with my switching off/over solution for the moment and for the time being have put off the efforts to find a solution for the sequencing (and possibly ramping) issues.
I do want to get rid of switchers though - not so much for practical reasons but for purity and integrity of the concept - strange extremist position, I know, but "black and white" works best for me :-)
Best,
Robert
Rick,
yes, you are absolutely right, using relays would be a very elegant way to do the switchin off and over (as described above). I have thought of that too but have not used relays before, have no experience with them and (most important) do not know yet "what they do" to the signal. However that's definitely something I will look at next, if this switching is best done with high quality switches or if relays do a better job in signal integrity while not adding any harm to the digital environment (another DC user...).
Thank you for bringing this back to my mind and to our attention in this context!
Robert
how do you power up your lab 12 volt supply to p4? is the lab device turned on first then the p24 power supply or the other way around? this is a a very interesting way to do it. also which amp ratimg supply do you use?
Theo,hope you don't mind if I answer this: It doesn't matter at all which one (P4 or P24) you turn on first - the power-up process is initiated with the push of the power button on the case, so once you do this both P4 and P24 need to be powered on.
Or in other words: Turning on power for P4 or P24 DOES NOT power up the PC - this is done by pushing the button, that causes the PC to power up.
Clarification note: It is exactly that power up process that requires proper sequencing and ramp rates. I have not been able to achieve that "manually" (i.e. switching various lines in a specific sequence etc.), that needs to be done by a module that has the logic to get the timing and ramping quite exact. For this reason I power up with P24 powered either by Antec or by picoPSU, and when the system is completely up I switch over to the "clean" supplies. (P4 does not need any sequencing or ramping etc., you just power up the PSU of your choice and you're good.)
Hope that helps
Robert
Edits: 05/27/09
thank you very much Robert I appreciate both your and Mark's contribution to the power supply saga on cmp.
Mark,
Short response...
I can't comment on Robert's 'SERIOUS' capacitance add-on or separate manually-switched linear supplies... they sound very interesting, but I just haven't gone there yet.
But I can comment that a Pico-PSU with some capacitance mods powered from a good-quality linear supply... it is a greater leap over a 'good' computer PSU (Antec 430) with run-of-the-mill added capacitors than that supply with capacitors was over the stock Antec.
So from the perspective of a fairly simple-to-implement and fairly safe (for your motherboard) solution, a modified Pico-PSU with a good linear supply is a significant upgrade from the Antec-430, stock or casually-modified.
BTW, for mods to the PicoPSU, I'm using 2 Black Gate 1000uf/25v NX replacing the caps on the PicoPSU boards and have added a Black Gate 2200uf/6.3v NX across the 5v and 3.3v outputs of the Pico. I'm not tied to these particular caps (the Mundorf sound very interesting), but I do know these work and know that some others will not work as well.
Also implementing a 12V linear supply is a pretty easy thing if you can solder and follow a schematic. After gethering up all the parts, I did my first setup with dual supplies (12v P4 & 12v into PicoPSU) over a weekend. I just rebuilt them with different filter caps and a different regulator configuration in an afternoon.
Robert,
You've provided a lot to chew on, especially your comments on the Mundorf caps and your research into bettering the ATX-24 supplies. I'll spend some time reading your sources and also get some Mundorfs to try.
Greg in Mississippi
P.S. Mark, here's an interesting article on the value of 'audiophile' caps:
http://gboers.xs4all.nl/daisy/home/g3/139/measure/capacitor-comparison.html
Hi Mark,
many thanks for another valuable contribution to our "Special Interest Group power Supplies" :-)
Although I try to achieve similar results with battery instead of linear power supplies, I hope I can further help by commenting on a thing or two.
Surely the most important recommendation to follow is to add extra capacitance on both P4 and P24 to achieve a clean, steady and quick supply of power in order to support the digital circuits' and components' power demands best possible. Since there is either "power" or "no power" (think of a square wave signal), the ideal power supply has to follow this almost infinitesimally small rise time as good and as quickly as it can, and capacitance (especially the true "low ESR"-like types) support that very well.
My brand of choice to reach that goal is definitely Mundorf (http://www.mundorf.com/english%201.1/kondensatoren.htm, see link below). Their MLytic SI capacitors offer very low inner resistance of somewhere between as little as 11 to 55mOhm, depending on which size you choose, and do so at 100Hz which is where it is most needed in terms of (non-)interference with audio signals, plus a design that is targeted on audio use with lowest internal noise etc. I have received excellent results with them.
In terms of sizing the capacitance I have gone a slightly different route, depending on the current draw on the respective lines:
- On the "low current" lines, i.e. 12V on P4 and 12V and 3.3V on P24, I use 5 x 2.200μF of the 63VDC version of Mundorf's MLytic SI. Since it is crucial how quickly the caps can release the current stored, I chose that value, in my perception they are unmatched for that purpose. I topped them off with a 47μF Mundorf TubeCap (unbelievably low ESR of only 7mOhm!) plus a 1μF Mundorf Supreme as bypass cap. The latter keep the loss tangent as low as possible, shifting possible interference very high into the MHz range where the PC's digital circuits aren't bothered anymore and thus can operate in a clear and electrically quient environment.
- On the "high current" line, i.e. 5V P24 (4-5A is already quite a high current for a computer environment), i choose 5 x 10.000μF Mundorf MLytic SI 63VDC. They offer an ESR value of 20mOhm (with 5 in parallel going down to 4mOhm in total) plus an unrivalled ability to quickly release the current required even in such quantities as required on a 4A draw. These are also followed by the 47μF and 1μF caps for filtering purposes.
Although I prefer to also have the source as clean and AC-free as possible (hence battery for me), in my experience with this capacitance setup it is gradually becoming irrelevant which power source you have "behind" them filling the caps as the first-hand power reservoir, as long as the AC interference isn't becoming too big.
Regarding your remark on P24's Pins 11/12 and 23/24:
You're right, as chipsets, components and interface cards required more and more power, P20 was extended to P24. However, there's not a dedicated correlation of Pins to consuming component or groups of components (like "Pin 11 is used to power this and that chip on the MoBo") - in fact all Pins for all power rails are just multiple pin-outs for one single rail of every voltage. This has been done to reduce interference by the PSU's cables, plus to keep their operating conditions modest by using them kind of "in parallel" and thus reducing both their internal resistance and the current that has to come across them. (You can easily "proof" that for yourself by putting an ohm meter to various Pins of e.g. the 5V line, the resistance between them is 0)
For us that results in the fact that we do not have to care about which Pin powers which component and how we best cap that or such - get a decent capacitance setup (e.g. like the one described above) per rail and you have done everything you can. The fact that those single rails are then subdivided on the MoBo with legions of sub-regulators etc. is a completely different sad story that adds interference to a clean power signal provided, but I'm afraid there's not much we can do against that (other than starting to look for and unsolder regulators on the board... *turns in horrror*) ;-)
Best,
Robert
Hi Robert,Thank you for your response. It’s very helpful.
Optimizing the Pin 24 is my last and ‘final frontier’ in my hobby project of building cics cMP2 memory player.
I know battery power comes most close too ‘the ideal power supply’ but I do not find it practical. So I started looking for the next best way to power the P24 pin.
Since powering the P4 with a linear PSU gave a significant improvement over conditioning the power output line on the Antec ATX PSU to the P4 pin with (audiophile) caps, also started looking for a better way to power the P24 pin than with adding caps.
Thank you for explaining how cap parameters are important for conditioning dc power that will be delivered to digital circuits that process audio-information.
I read to many posts where approaches from the analog audio domain are simply copied to the digital audio domain.
Now I know ‘digital audio’ doesn’t exist and that all ‘digital audio’ simple obeys to ‘analog’ physics and electricity laws. But what is valid and relevant in the analog audio domain cannot simply be copied too the digital audio domain.
So I am always very skeptical when I read in posts, that one has to use audiophile caps (true in the analog audio domain) when conditioning the dc power output of a PC PSU for a PC which is processing digital audio information.
I have an engineering degree but not in electronics. I only have some basic knowledge about AC, DC, caps, coils, electronic circuits, etc, ect.
But that’s just enough to understand the concept that when a square wave signal is turned into a shark-vin signal, a jigsaw signal or a mountain/valley signal, that the time on which it’s considered to be a ‘zero’ or a ‘one’ will differ. Or that threshold levels which decide if something is a ‘one’ or a ‘zero’ should be steady, the importance of small rise and sink times, ect, ect. But I have no knowledge at all of, on how these concepts are brought in to practice in real-life electronic products.
But nevertheless these shortcomings in know-how and knowledge, my challenge is to improve the dc power delivery to the P24 pin beyond just adding caps.Since a lot of ‘google-ing’ didn’t shed any light on why there are five red +5 volt wires, eight (yes: 8 !) black ground wires, 2 orange +3,3 Volt wires and two yellow 12 volt wires. I did expect it too have some very good technical reasons. But now I guess it has to do with evolution and backwards compatibility ect, ect.
Already A few months ago I also red the formfactor.org text on the ATX power-up sequence ect, ect, So I understand why it’s difficult (Impossible ?) to power-up from battery’s.
When google-ing a also stumbled opun this (for me) very informative text on how a ATX PSU works at a conceptual level
http://www.xbitlabs.com/articles/coolers/display/psu-methodology.html
It explains why all ATX PSU have troubles with load balancing between 3,3 v, 5 volt and 12 v lines and have irregular distribution of the output voltages when the load is unevenly distributed, ect. It all comes down to that a ATX PSU only has one single voltage regulator but it has to regulate 3 different voltages lines. The bottom line of the paragraph is: an ATX can only regulate one voltage good, the other 2 voltages are regulated within an ‘acceptable range’ with some ‘trics’. Literally the text says :
“this regulation is far from perfect and gives rise to the most common problem of computer PSUs – irregular distribution of the output voltages.”Sow how can one feed the P24 pin with ‘good’ +3,3, +5 and +12 v dc power?
- 2 ATX PSU: one master PSU and one slave PSU. It’s easy to implement.
Each of the ATX PSU is only doing one voltage. The loads on each line will not influence each other. But to do with the 3rd power line?- 3 ATX PSU’s ? each doing there own voltage?
Than I will have 5 !! PSU’s supplying the cMP2 memory player.- Pico PSU’s?
How is load balancing done by the pico-PSU’s?
Since a pico-PSU already is fed with 12V DC, it may be more easy for a pico -PSU to balance the 3 loads. But that is just more of a question. I really have no clue on how a Pico PSU works and how it is is balancing the loads between + 3,3V + 5V and +12V. May be pico-PSU’s do a beter job in load balancing. I don’t know.May be a better way of supplying power to the P24 pin could be:
- a dedicated ATX PSU’s for each voltage on the P24 pin
- or a dedicated pico-PSU’s for each voltage on the P24 pinBut since adding extra caps to the power lines on the P24 lines only gave a small improvement. I ask my self: is it worth the trouble to better the power-supply on the P24 pins with extra dedicated ATX PSU’s. Because there may be only a small potential SQ improvement to be achieved.
When adding 2 of 3 dedicated extra PSU’s might not being worth the trouble (and costs) How about using a linear powered 12v pico-PSU (instead of an ATX PSU)?
GStew reports that his pico-PSU powered by a linear power supply sounds better (than powered by an ATX PSU?) So what I also would like to know is:
is it a significant improvement in SQ when switching from a (good) ATX PSU (with smoothing caps on the powerlines) to a pico-PSU ?Can anyone report on the SQ improvement between an (good) ATX PSU and a pico-PSU?
Robert thank you for your response.
Al 3,3V, 5 V and 12 V lines are bundled together behind the P24 connector.
So paying special attention to feeding lines 11/12 and 23/24 is less relevant. One would only connect 2 PSU’s to 1 voltage.LynxL22 dig i/o -> Lavry Black DA10 -> Mogami Gold -> Klein & Hummel O300
Edits: 05/26/09 05/26/09
Hi Mark,wow, what a post! ;-) As usual, let me try to add my 1.8 cent (inflation takes its toll):
Interesting - optimizing P4 and P24 was my start, to get the basics right, and there's sooo much more to do (DAC, amps, HF shielding etc.) ;-)
You're right, battery power isn't always totally practical, but (i) you can get it organized well, respecting load cycles and using a good charging concept, and (ii) as a perfectionist I myself am way into impracticability too often anyway ;-)
So your goal is to achieve an as-good-as-possible power supply for P24 with power derived from AC mains (by the way, you don't want ANY AC traces in your digital pure DC environment, that's also what the caps take care of). I went to great lengths to really pin down which change in PS does what, so I compared
- Antec EarthWatts vs picoPSU powered by Antec EarthWatts (without any caps)
- Antec EarthWatts vs picoPSU powered by battery (again without caps)until i found out that once power is up I can easily switch the main rails (3.3V, 5V, 12V) to whatever alternative PSU I favor, so I compared
- Antec EarthWatts vs picoPSU (powered by Antec EarthWatts or by battery) vs "pure" battery without caps, and
- the same WITH caps.Let's cut a too long story short: Once you add caps (and I would like to repeat that for what I have understood the capacitance I have detailed before seems to suit the purpose best) it's really getting close to irrelevant what you use "before" them as a source.
Always keep the square wave in mind: the most important task is to keep power supply as clean and quick as possible to keep the rise time between 0 and 1 ("no power" and "power", changing ever so quickly) at a minimum! IMHO this can best be done with caps sized the "right" way, plus cables from the caps to the board which are short, have a lot of internal conductors (that brings speed, e.g. I use AWG 6 which is 133*AWG27) and a very low internal resistance ( <.5Ohm/1000ft, just to be safe - that's in the region of an average PCB's traces).
One more comment on the number of wires: you may have seen from my P24 measurements that current draw on 12V and 3.3V lines is quite low (always < 500mA), whereas current draw on 5v line is quite hefty with up to 4-6A (almost high current in a PC environment). That explains the number of wires - 2 for 3.3v and 12V usually is more than enough, whereas BIG current is drawn across the 5V line (MUCH more in a gaming PC or such of course), so getting more wires in parallel reduces the current flow, thus temeprature and internal resistance and so on - you get the picture. The GND ones just complement.
Also one more comment on ATX power-up sequence: As said elsewhere, I tried to build some module that would do this power sequencing and correct ramp rates for me, but (although it would be very elegant) I have put that back for the time being: I simply power up either with picoPSU or Antec (haven't decided yet finally) and when power is up swith off the "special" lines (-12V, Power_Good and 5V_standby) and switch over the main lines to the clean power sources. Keeping the picoPSU powered by battery or linear is not an option since it is a switching PSU in itself (although a rather good one when powered itlsef with "clean" power) - it's definitely not doing load balancing but is just as much a switching PSU to maintain the various power rails as your average ATX PSU, with all its drawbacks too well known.
In your case - I'm sorry to say that - a "clean power source" for every rail indeed would be one linear PS per voltage, there's no way around that if you're heading for the optimum - you simply HAVE to keep them separate with strict and exact power regulation on the proper voltage. (If we're talking about extremes, I'm providing a separate battery PS for every voltage rail for every component, just to be safe, I don't want to think about interference of any kind. There's not many voltages involved anyway, so a quite ok charging concept for all of those is doable. And by the way, you say in your last line "One would only connect 2 PSUs to 1 voltage" - yes, that makes great sense, I have in some places connected 5 or more batteries in parallel to get an internal resistance of equal to or below 2mOhm - think of the square wave, speed and impulse is key!).
My recommendation really would be: Don't think about 2 / 3 / many ATX PSUs (they're switching PSUs, that's crap for our purpose, forget about them!) - start up with one (ATX or pico, doesn't matter), and when done, turn off the "special" lines and switch to your 3 linear PSUs for the 3 main voltage rails that feed the proper caps, then we're talking :-) Maybe a bit of overkill, sure - but isn't that what we are here for, striving for excellence and sonic bliss?! :-)
Oh, almost forgot to mention: Modification on the P24 12V rail (as detailed above, I used switches in setups with and without caps to determine exactly which mod leads to which sonic result) yielded minimal sonic results, 3.3.V was clearly perceivable, and 5V is HUGE - sure, that's the main powerhouse! Interestingly enough, the sonic effects weren't "more/less bass" / "more / less transparency" or such, but less distortion, less fatigue (I thought that was a myth and lack-of-words before I clearly heard it myself), a sonic picture clearly and perceivably more "right" and correct, all in all absolutely worth the effort, for me one of the major contributions for a sonic experience just "right" and "natural" - I'm sure you'll find out yourself! :-) Enjoy!
Best,
Robert
Edits: 05/26/09 05/26/09 05/26/09
OS-CON type caps (made by Sanyo, Nichion, and many others) have extremely low ESR (under 20mOhm at 100kHz). These caps are used on high performance motherboards for the same reasons proposed here. They do seem somewhat difficult to source though...
Edits: 06/13/09
Hi Robert, Gstew,Thankx again for your information. It’s now more clear for me which routes I might want to follow in optimizing the dc delivery to the P24 connector
you wrote: "Let's cut a too long story short: Once you add caps (and I would like to repeat that for what I have understood the capacitance I have detailed before seems to suit the purpose best) it's really getting close to irrelevant what you use "before" them as a source."
1. So (high quality) caps are still a viable option.
But I made calculations on the cost for low ESR caps from Mundorf. It costs about 222,75 euro (12V VP4) and 321,50 euro (+5 V at P24). I think the costs involved are to high versus the resulting effect. It has to be done let costly. Low ESR caps (model’s NSP or NPC) from NIC Components Corp. (ESR 7mOhm) cost about 1 dollar.2. Looking at Gstews contribution:
A Hybrid solution (mixed pico / ATX ) also springs to my mind:
- the picoPSU (with extra caps) fed by a linear dc 12V supply for letting the pico deliver only the 5 volt.
- the ANTEC (with extra caps) for delivery 3,3 V and 12 Volt.
(don't know if the ANTEC wil start but i'll give it some dummy load on the 5 volt, so it has someting to chew on)
3. Your recommendation to use 3 linear PSU's.
But than there will be 5 PSU boxes (!) in my setup.
It’s going to look real messy in and around the Zahlman-case.
:-(I think I start with Gstew’s suggestion.
A second linear PSU that is feeding dc 12 V into a pico-PSUAnd after that I think I will try to make a hybrid solution and see how that impacts SQ.
I also going to look for super low ESR caps. But at a more cost/effective proposition. As you explained, I see the need for low ESR.
LynxL22 dig i/o -> Lavry Black DA10 -> Mogami Gold -> Klein & Hummel O300
Edits: 05/27/09 05/27/09
I used Greg's plan and built a Daniels derived 12VDC Linear coupled to the Pico w/Oscon caps.
Also have a GStew modified Juli@ with regs, BG caps and BNC out. Made a large difference on the positive side.
Have the Hammond L193 inside and AR power cables. No fans.
Sounds clean and sharp.
RayBan
Mark,glad that the information shared was helpful for you.
However I have to comment on your calculation for pricing the suggested capacitance:
For P4 (based on how I have implemented it) you would need 5 x MLytic SI 2.200 63VDC + 1 x TubeCap 47uF + 1 x MCap RXF 1uF. looking this up online in Mundorf's price list, I get to 5 x EUR 7.99 + 1 x 24.90 + 1 x 5.49 or EUR 70.34 in total. Although this is not exactly cheap, it is (i) way below the value you have come to, and (ii) IMHO well worth the money spent :-)
For P24, you need 2 x the above plus 5 x MLytic SI 10.000 63VDC + 1 x TubeCap 47uF + 1 x MCap RXF 1uF. This amounts to 2 x EUR 70.34 plus 5 x EUR 16.90 + 1 x 24.90 + 1 x 5.49 or a total of EUR 255.57 if you want to add capacitance to all 3 rails, or EUR 114.89 for the 5V rail only.
Remember, you don't have to put caps to every red or orange or yellow line, these are just parallel wires that re-connect "underneath" P24 anyway - connecting all red and all orange and all yellow cables to the positive pole to the 1uF cap for the respective voltage rail is enough. (I guess that's how you came to these high values.) And please do have a look at the link Greg provided re cap comparisons, the inner structure and architecture of the caps is just as important as the relevant parameters like low ESR etc. as clearly shown and measured there, I fully support that.
Cheers,
Robert
Edits: 05/27/09
Hi Bertel,
I toke a while but after my holiday two weeks ago, I was able to pick up again the process of optimizing the power supply to the P24 connector.
As a start I ordered a mini-box a 12 V Mini-Box PSU, model: PW-200-M. and a linear bench / lab PSU capable of putting out: 12V DC/ 6 Amp max
As for a start I put some Audyn smoothing caps on all 3 voltages (3,3/5/12) with I had laying around. But the don’t do much. With or without, I can’t hear any difference. I will have to order real good quality caps as you pointed out already in an earlier post. :-)
I also did some current measurements before the 12 V Mini-Box PSU PW-200-M on the 12 Volt DC line that feeds the 12 V Mini-Box PSU PW-200-M.
I can’t copy your current measurements into the P24 connector. The current into the 12 V Mini-Box PSU PW-200-M on start up on my multi-meter varies between 1.6 amp and 2.1 amp max. And during play back it’s a constant of 1,28 amp – 1.35 amps. Depending on FSB speed, grafics, soundcard used ect.
I also ordered a picoPSU (90 watts) yesterday. Some inmates report that the 12V line on these smaller picoPSU’s are just 'passed through' in these models.
The PW-200-M isn't just passing through the 12 Volt line. If I vary the input voltage up and down, the PW-200-M stabilizes it to 11.93 volts. So the PW-200-M isn’t just passing through the 12 Volt line.
If the smaller picoPSU’s realy just pass through the 12 Volt line, than they have a different concept than the PW-200-M has. Since the smaller picoPSU's are real cheap (37 euro’s) I ordered one. I’m curious if this different setup also makes a difference in Sound Quality
If you plan any new current measurements in the future, please post the results if you like. You’re readings are almost twice mine. May be my multi-meter (4 month old but just a cheap one of 75 euro’s) isn’t working properly.
LynxL22 dig i/o XLR AES/EBU -> Lavry Black DA10 -> XLR Mogami Gold -> Klein & Hummel O300
Hi Robert,
grrr :-(
you'r right.
I mis-calculated that way.
Thank you !
At 70 euro's, I indeed think costs are worth the effect.
Thank you for correcting
LynxL22 dig i/o -> Lavry Black DA10 -> Mogami Gold -> Klein & Hummel O300
Robert,
My current setup uses a Power-One 12v / 1.7A supply for the P4 (fitted with several Black Gate 1000uF caps), a un-modified Pico 120 psu (currently changing the caps on another pico and putting together a linear 12v to replace the switcher) and a Power-One 5v 2A supply for the HDDs.
I am about to change it putting all the psu's in another case underneath the Antec case. So having read the various posts it seems to me that I should try your idea of switching the pico lines off and switching in three other linear psu's. So would I be correct in saying that for the P24 supply there would be effectively 4 psus, a 12v to power the picoPSU, another 12v, a 5V and a 3.3V psu? The first 12V (on the picoPSU) would stay connected with the 12V / 5V / 3.3V switching in and out.
Where do you have the switches? I assume that the picoPSU would not be plugged directly into the board but into a socket that feeds into the switch.
I have tried batteries on the P24 but it is the charging that made me change back to the Switcher/picoPSU; if changing over to batteries then I would like to leave the batteries in situ and when the pc is switched off they would connect to a charging circuit. For the moment I have too many other projects to pursue this idea - I am sure that a search of the web will give me what I need to know.
I also need to power my Juli@ from batteries - this one is high on my list as I have to break out the SPDIF and I2S signals to send directly to the DAC.
And of course then there is the time to listen to the music!!
Thanks for your information.
Alan
I am certainly impressed by your industriousness on the power supply saga. So I take it a 12 volt 2 amp supply is ok for p4? Also is the 12volt supply adjustable?
theob, the Power One linear supplies have a trimmer pot on the board - never tried using it though!
Alan,
your setup for P4 and the HDDs already sounds very good (maybe you might want to consider adding an additional cap like my 47uF Mundorf MLytic SI to the P4 line for supplying the current peaks most quickly plus a 1uF type for shifting all remaining AC fractions and interference into oblivion...? ;-) )
Regarding PSUs for P24 and position of the switches: Ok, I'm afraid I need to answer that again a bit too detailed... ;-)
Concept/layout:
- I haven't decided yet what I finally will use for powering up the system before switching over, the Antec or the picoPSU.
- You are right, when I use the picoPSU, I plug it into a P24 extension cable to get all the required lines "out of" the picoPSU.
- Once the system is completely powered up, I switch of all "special" lines (-12V, Power_Good, 5V_standby) with a 3-pole toggle switch right "before" the P24 connector.
- The capacitance detailed above is also sitting "before" the P24 connectors, the 1uF cap being the last in line and connecting all red / yellow lines for the 5V / 12V rail with its positive pole and a number of (in my case 4 for the 5V rail and 2 for 12V) black ones with its negative - you get the picture?
- 3.3V is different - I use a pack of 5 LiFePo batteries that supply exactly 3.3V for a long time and at an internal resistance of 2mOhm - no need for any regulators and caps here, direct is best possible :-) (I use an identical one to power Juli@ as per Alfred/sonics)
Switches:
- The switches for 5v and 12V are "before" the caps, i.e. between PSU and caps, thus their interference is irrelevant (I believe).
- Well, there is a switch between the battery pack for 3.3V and the P24 pins... Up to now it's still an ordinary toggle switch, but will upgrade to some "audiophile" switch (like for choosing source in a pre-amp, not decided yet which, internal resistance is the key)
- Another upgrade to come is getting rid of the P24 extension cable and soldering low resistance cables (thick ones with many wires for speed plus copper mesh for keeping HF interference out) directly onto the caps on one side and into gold-plated Molex P24 pins (got them thru Digikey) on the other. Well, you can't get too extreme, can you? ;-)
- All in all, this allows a very "pure" environment when everything is switched over or off - decent capacitance supplying clean power as quickly as possible to the various lines completely separated, fed by exactly regulated and also quite clean voltage from batteries. Probably sounds like overkill for most, but since current and voltage is basically everything and the starting point for any signal, it's my preferred point of interest, and in audiophile terms it absolutely pays for me :-)
Ah well, your thoughts about the number of linear PSUs to power pico: Well, you could get along with 3 linears for 3.3V, 5V and 12V, using the 12V one also to power the pico from the start and switch off the pico when you have switched everything over and the pico is idle then. But remember to size this 12V linear right - the 12V line only draws <200mA but needs to provide 4-6A while powering everything through the pico... I (of course) use a seperate battery... ;-) But surely that's not necessary.
Charging the batteries:
- I disconnect the 3.3V LiFePo packs and charge them elsewhere. not a hassle for me, once charged the last for days so that's a nonissue for me
- All the other batteries are in fact charged in situ (I like that) by manually connecting the required chargers. Not an elegant concept, I know, but I like to have the chargers completely out of the way for "serious" listening sessions, I don't have too much time for them unfortunately anyway (1-3hrs every second day on average) so again for me that's pretty manageable - and my "pure extremist" approach (no switch is the best switch) is supported :-)
BTW: Breaking out i2s from Juli@ to DAC - interesting! What DAC are you using? I'm going Buffalo32S at the moment... ;-)
Best,
Robert
Robert - I will digest your words over a glass of wine this evening - but as a quick response if you look for 'rebirth of the juli@' you will see all the details for the I2S breakout - as soon as I have done this (slightly differently as i will remove the Analogue board) I will take a photo and post.
I am also putting together the Buffalo 32s - hopefully I should have it in the system (using SPDIF) tonight or tomorrow. I am really looking forward to using this dac.
Alan
Robert,
I think I understand what you are doing and where, one question springs to mind - you say that you are using a 12v battery for the ATX24 which you switch to once the pc is up and running. Is this one SLA or is it several (is it even an SLA?) also (OK I cannot count) with the 3.3v you do not use any regulator as per sonics - what about the 12v?
Since I have a few spare Jensen / Black Gate / BHC capacitors I think I will first of all attack the setup as it is (i.e. no change of psu) and then look at the switching off / over of the supplies.
All of this will be after I have the Buffalo 32 settled in - all psus for that now up and running just a question of a few connections (and possibly putting it in a box).
Alan,not sure if I understand your question right - I probably was unclear on whereto I switch once the PC is up and running, right?
Let's assume for powering up the PC, I have the Antec Earthwatts connected to P24 (caps are in between on 5V and 12V lines). Well, "connected" isn't really precisely what it is - every line currently is going through a switch, two switches in total, one for the "special" lines (-12V, Power_Good and 5V_standby) and another one for 3.3V, 5V and 12V and the respective GND wires.
When powering up, all switches are so that every cable coming from the Antec has connection to its Pin on P24. When everything is up and running, I switch off the "special" lines. I then switch over the three main lines to their battery sources:
- For 3.3V I have a LiFePo battery pack (5 cells) directly connected, i.e. the wire from the switch that connects to the orange cables is connected to the positive pole, another wire that is now connected via the switch to two black lines goes to the negative pole of the pack. No regulators in between, no nothing - the pack is charged at exactly 3.3V and has a capacity of 2.3Ah so that they can keep exactly this voltage for a long time (due to the quite low current draw on this line of ~ 300mA) and give full advantage of their low internal resistance.
- For 12V I use a set of three preconfigured NiMH packs at 6V each (just so easy to handle and charge, of course you could use any 6V batteries that you prefer), so 18V in total, and feed them into a voltage regulator to bring them at exactly 12V. I know that most probably consider that as overkill, you usually use a 12V battery for a 12V circuit, don't you - but a 12v battery almost never is at exactly 12.00V, whereas I prefer to feed the digital circuits on the mainboard with an exact and always identical and reproducable voltage of 12.00V. (BTW I use the same setup for P4 to get to exactly 11.93V which has been reported to be the best sounding voltage by Bernd and Theo and others). Again, positive wire from switch to positive output of the regulator, negative wire to negative output.
- For 5V I currently use 4 8V VRLA Pb batteries (can't use the NiMH ones or other "weak" batteries because for 4-6A and regulated you need really strong ones, and these are). Each goes through one regulator (same as above) each to get to exatcly 5.00V. Again positive wires to positive pole etc.
Is this explanation halfways clear...? I promise to take a few pictures and post them - once I tidy up the still very experimental mess... ;-) That'll take a few weeks.
And you're right - what I described (except from the 3.3V line) all takes place "before" the capacitors, so they are really the most important part re quick and clean supply of decent power to the demanding circuits. By adding the spare caps you'll most likely will get substantial improvements already.
Cheers,
Robert
Edits: 05/28/09 05/28/09 05/28/09
Bertel,
I am trying to put together a diagrammatic representation of the various psu mods using batteries and switches that you have posted - I always find this helps me understand things. I will send you a copy before posting - to make sure I have understood.
I am trying to source the batteries and I thought I had a reasonably priced source for the LiFePo4 batts until I checked and one of them is dead! So after much searching through all the Radio Controlled Planes/Helicopters I think I found some decent ones - Graupner / Racing batteries at http://www.acemodel.co.uk/batteries/a123-lifes04-/1100mah-3-3v-single-cell/prod_199.html although they describe them as LiFeSo4!Are these OK? If not where do you get yours?
I can find many sources of 6V NiMh packs - but the only source for 8v VRLA batteries is China! Given that this is supplying the 5v work-horse line then the other styles of batteries do not have enough power. I suppose I could always use one of those Marine batteries - big enough to run a boat!
I have to say that searching for batteries has been quite an experience - I might just take up Radio-controlled Planes as a hobby - well it would get me out in the fresh air.
Any help/sources would be appreciated.
Alan
Alan,
Radio Controlled electric airplanes and helicopters is my other main hobby, so maybe I can help. Where are you located?
Greg in Mississippi
Greg,
I am in the UK in a town called Farnham (Surrey); I made my comment about Helicopters and Planes in jest - however I have now started looking at the various planes available. It seems that there is quite a selection from ones to use indoors to hi-performance outdoors types.
Near where I work is a very large playing field owned by the Army and there are quite often people flying planes - despite being close to one of the largest Army Barracks in the UK!
Regards
Alan
look at that handsome guy. that is a big plane!
Robert,
Thanks for the clarifications - I will spend some time now planning what to do - I have quite a lot of 6v SLA's sitting around not doing a lot, same as the capacitors; just need a few more regulators.
On the Buffalo side I have been wondering about how to put it into the PC without any interference - this would make the I2S leads as short as possible.
Photos would be great - but I would not like anyone to see my PC as it is - just a heap of wires - transformers open to the world - so when yours is all neat and tidy please publish.
Alan
how difficult was the buffalo32 to assemble? which ps did you get for it? nice.
theob,
The Buffalo 32 board itself comes built/tested all that is needed is to solder in the connectors - left out just in case you want to hard-wire it.
I bought three psu's so that I could drive each analogue channel separately - they come with all the components identified - there are documents on the website as to the board layout etc - very easy to build, the only thing that has to be done is to twiddle the 20-turn trimmers to get the voltages set as required.
All in all - very easy - hopefully it all works when I connect them tonight.
I am very excited for you--having ability to go I2S out to the Buffalo. Please let me know how it sounds. I am very interested in going this route too, just waiting for better time to finance it (2 trips to finance 1st).
I bet it is so.
I think Mr. Nugent is one of the important voices in the world of PC AUDIO. I do not own any of his devices but I am always looking over his shoulder. I am always interested in what he is working on. He is committed to the USB and we, mostly, are not. Nonetheless, no one has done more tinkering and thinking than he has.
Look at this thread:
http://www.audioasylum.com/forums/pcaudio/messages/5/52276.html
Nothing like improvements at the absolute beginning of the chain to make the biggest difference.
I went ahead and bought the package of drive and software instead of seeing what I get from the ebay seller. For an extra eighty five dollars I like knowing I am getting what the software writer intends.
If it is worth it looks like I will have to re-load all of my files, but if it sounds better it will be worth it.
I doubt this is going to romance me away from the turntable but it could make a significant difference.
Bye,
Rick McInnis
Hey Rick - I'm not trying to diminish what Steve is recommending there, but I just want to point out a couple other things of note. When Spoon (of dBpoweramp) chose the Teac drive as the best one for the RipNAS, he was not saying it is the best drive period. His criteria was to find a drive that was an IDE laptop drive with slot-loading. So of those drives, this Teac was the best. However there are desktop drives that are just as capable. I have a Plextor Premium and Plextor 760 which he said was equivalent to the Teac.
Also, in Steve's comparison when comparing the different Led Zeppelin rips, he said they had different checksums, so of course they sounded different. And apparently to him the Teac results sounded better. But let me just warn you that simply re-ripping your music with a different drive is not going to necessarily be a jump in SQ. If you have damaged CDs (as I suspect Steve's LZ disc is), then the Teac may give you better accuracy (and different checksums) as your current rips (and indeed sound different). But just to share my experience, my CDs are flawless, no fingerprints or scratches or anything, and I have compared some rips between EAC and dBPoweramp and on different drives and they end up matching AccurateRip and have identical checksums - and they sound exactly the same.
I'm not trying to come off as a skeptic here. YMMV. But if you already have a decent desktop drive and have correctly used dBPoweramp, then you may not be gaining too much with the Teac (except perhaps to better rip damaged CDs).
Edward,
If only everyone commented the way you do. Of course, the apologies are appreciated but unnecessary when one makes reasonable points in a reasonable way.
I do not doubt I probably jumped too quickly. Nonetheless, I was needing a new CD R drive so at worst it is not as if I have bought something I do not need. Maybe I did pay too much ...
I will see if I can tell any difference and report my findings. We are all too anxious to find the golden goose of digital audio. I was not expecting a dramatic change just hoping for another increment but how great it would be find something that would make a leap similar to cMP!!! Even one-quarter as much ...
Thanks for tempering my intemperate enthusiasm.
Rick McInnis
... but I keep getting a 'not enough ram error' in cplay. Before I put my old kingston 1 gb ram back in is there some criterion for determining not enough ram? I got one 6 minute file converted that plays ok but that took 9 loadings. Is 10 the limit?
.
...and once I got the hang of it the sox upsampler is not bad. It's not on par with SRC but it's good. Also I found that I was more successful converting to flac files than wavpcm-> wav. But let me say sox is good, very good, but I would have to say that SRC is really a lot better in upper mids, highs. There is more texture, nuance, details in a full orchestra with SRC. With sox the same piece is smoother, warm but an almost 'cover-up' of details occurs. Some may prefer sox to src because it is warmer (and I know those who are willing to throw the baby out with the bathwater but I'll take the baby and bathweater every time).All in all though I prefer sox at 192 versus src at 96. A very good choice as an alternative to the metallic(in my case) src.
Question though on the sox parameters how do phase choices (other than I) affect sound? Just curious if anyone has tried.
Edits: 05/20/09
I have listened to several files with both M (minimum) and I (intermediate) phase called out in the sox conversion command and I believe I prefer M. With M the resulting sonics are closer to SRC wrt to getting the initial ictus or impulse of the sound right. Plucked strings, fast percusion and imaging within a large orchestra sounds better or more realistic to me. Reading about minimum or linear phase suggests that minimum has less pre ringing but more post ringing in a asymmetrical way. Linear phase is more balanced wrt to pre and post ringing but has more pre ringing ( than M) which is a bad thing imo. I is in the middle but I think pre ringing is the more important parameter.Has anybody else tried this?
Edits: 05/21/09
theob,
I have not tried it but I will give it a spin over the weekend - very interesting findings. Have fiddled about with any of the other settings?
right now with sox I use gain -1 (or -2) rate -v -s -M 192k and I put data in flac files.
with cplay I use 192 rate, dsp buffer at small. I'm still running juli@ analogue outs.
I'm really liking sox-ed files at gain -1 (or -2) rate -v -s -M 192k. when I go back to cplay 96 rate 146 upsampler and listen to original 44khz files it sounds more etchy to me in the highs. maybe I'm getting used to sox-ed files. Its very nice but if I continue to convert my files at my current rate I'll be needing another 2.5" drive soon.
Obtw when running flac 192 files I'm getting under 10% cpu usage (key to avoiding metallics).
Hey Theob,
Forgive me for jumping in and not reading alot of the prior posts, but it seems like this is not a fair comparison as you are really comparing an offline upsampler against a realtime one.
It would be interesting if cplay would allow for creating files using src, or comparing sox when it is integrated for realtime upsampling.
probably not a fair comparison but a real either/or for me. are you saying one is intrinsically better than the other?
As you've noticed, offline upsampling is much less CPU intensive.
YES, and IIRC that might be a big deal with Theo and his brightness.
Anyhow, it would be nice to do a fair comparison either way.
that is sounding really good - looks like you are back to enjoying the music!
:)
A Cautionary Tale :
Hi All (...and especially you Juli@-with-BNC-sockets-added folks)
Last fall my Juli@ underwent a modest transformation, as can be seen here -> http://www.audioasylum.com/forums/pcaudio/messages/3/39956.html
Having compared the S/PDIF out from both connections - new BNCs vs the stock breakout cable - and declaring the BNCs a clear winner (...is that a pun ?), I turned my attention to other things.
The final step of that listening test was completed only last night ... ie: remove the breakout cable !! - Duhhh !!
I was listening with my teenage daughter, Jane (no... not her real name) and we both instantly heard the improvement... and she is not encumbered with "audiophile" filters to cloud or colour her perception.
Jane hasn't heard the cMP since the release of B22, and it's become a tradition to celebrate a new version of cPlay with a performance of Tubular Bells II from Mike Oldfield. Every time there is a new cPlay, we come to know and appreciate this recording better.
(1992: WEA CD 90618 - producers: Trevor Horn/ Mike Oldfield/ Tom Newman. One of the most stunning examples of Studiocraft, in the service of Music, that I know !!)
We began with B22-@-121/192KHz (my P4 can't do 144 SNR), just to calibrate our ears. Track 08 - Weightless ... very nice.
Next up was B25-@-121/192KHz (same track). MUCH better... remarkable that TB II still has more secrets to be revealed.
OK, with so much opinion on both sides of the NOS 44.1 vs 192 upsampling, we listened again (B25-@-121/44.1KHz).
Nope... not in this system. Sixty seconds into the song, Jane turns to me making a sour-puss face ("I don't like it"). Neither did I. Missing in action was the incredible sense of width & depth - confined now to the space between the speakers (and no longer reaching back into the neighbour's house !!). Gone was the sparkle of the HF, the minute dynamic contrast of some percussive lines, and the separation of instrumental voices we heard before.
Quick !! back to 192... and The Magic returned. At this point I noticed the pigtail still hanging out of the Juli@. Hmmm... "OK Jane, just one more test before we get to hear the whole thing" (remember the impatience of youth...).
AGAIN... within the first sixty seconds of play, the benefit of losing the breakout cable was plain to hear. Jane said the "different layers" were more distinct, and that all the parts "were easier to hear". The "audiophile ears" agree.
Yes, this simple tweak should have been made back in the fall... but in a way I'm glad it happened like this. It's another lesson in support of the old maxim: EVERYTHING MAKES A DIFFERENCE.
Cheers,
Grant
That's not a Toy... IT'S A TOOL !!
Uh-huh. Just "left it there"... of course "there" being in a corner on the floor where you never go unless something's fallen there - or broken.
Out of sight, out of mind. And the Mind is too full of other noise ! Gee, it's like getting a new version of cPlay for free... er, well... ahead of the rest of the group. Or am I just catching up ? Yes, THAT's what the fuss is all about. Well done and three Loud, Appreciative Cheers for version B25 !!!
Better quit now.
That's not a Toy... IT'S A TOOL !!
Grant,
Are you saying you had connected the BNC to the board, YET, had left the breakout cable attached?
I can only assume this is the case, yet, I am amazed you left it "there" for so long! Yet, I know the feeling, who has not done this same thing in the audio eccentric's feverish must listen to it now after all of this work syndrome. Just leaves the pleasure of hearing another veil lifted. Not in the ARABIAN NIGHTS sense ...
Still trying to muster the courage to get a copper mat.
Bye,
Rick McInnis
Hello cics and cplay users,
First of all, let me say that cplay is the best audio player I know.
Thanks cics for all your hard work.
That said, there is one problematic side to cplay, the visual side.
Even a slight aesthetic improvement without a complete gui makeover
will make a big difference. Since I understand cics is very busy, I have
found a software "Resource Hacker" :
http://www.angusj.com/resourcehacker/
that allows one to change icon buttons inside executable files.
Tiny step, but, one step for man ... :)
Anyway, cics, if you need any help I am willing to contribute my
programming skills and little time I have for this great project.
.
http://en.wikipedia.org/wiki/Hidden_Markov_model
As a way to improve upsampling, I was wondering lately about using similar
patterns in the track, to thicken the original sample rate.
Here are some small steps even more worth taking:
1. In cMP, show the samplerate of each cuesheet so one knows whether to change the cPlay playback rate: either by showing the number or using different colors for different rates in the cuesheet names
2 In both or either cMP or cPlay, allow enlarging the window or decreasing the font size to allow more of the album, title or cue title to display
3. In cMP, since the artist displays already in "All", substitute or provide option for Composer display instead of "Artist"
Dear Theo and others who have heard Theo lament his metallic affliction,
I heard it today.
I installed 25 today in the SSE4 version since cics seemed to say this was his best work. I have never installed this version (instruction set) ever before.
At the same time I am trying to lower the Vcore voltage and probably too quickly since the machine has not been running for six weeks previous to yesterday. I, of course, have disabled EIST. Did that yesterday and was able to run at .80 volts without a problem.
Today, I took the voltage down to 0.7625 (or something like that) and I get the blue screen. I turn the computer off and attempt to re-start and there is nothing. I re-set BIOS and incrementally increase the voltage until the machine will run. Something like 0.80 and request a selection and I hear the sound he has been speaking of. It is horrible.
I re-start and increase the voltage to 0.825 and ask for the same selection (all within "start cMP") and it is back to normal until I accidentally hit a button on the mouse and it begins again. I re-start and all seems to be fine so far. Though I did not try to antagonize the machine. I will let it run constantly for the next few days and see if it settles down.
NOW, I am wondering if it has something to do with SE4? Does changing the instruction set take some getting used to for the CPU? Excuse my processor mysticism, but I have never had this happen before.
I cannot remember all of the details of Theo's travails, but I wonder if there is any connection with this?
Could disabling EIST have anything to do with it? Since I have just done those two things, they are the differences in a previously trouble-free set-up I cannot help but wonder. Again, one wonders if a system gets used to a certain set-up and rebels when "these things" happen? Of course, it sounds ridiculous to me too but in the world of audio, it seems, anything can happen. When little atomic squiggles can produce astonishing sound what is one to say?
More than anything I wanted to assure Theo he is not damned or cursed or any of that other stuff OR I am, also.
Other than that, I do find 25 SSE4 to be very attractive and superior to the sound I was getting last night with the previous iteration and the other instruction set, even though, in my case, it required a higher voltage. Still very little information extending beyond the outside of the loudspeakers. Within, it is very good. Like a loving microscope on the recording. Tremendous detail/precision in imaging within the scope of the lens. This is sound one could easily live with as long as they do not ever listen to a better turntable. blah, blah, blah. I know ...
It has occurred to me that cics has never told us what this mysterious abbreviation stands for. Dear cics, any chance of letting us in on the secret?
As always my best wishes to cics and his family who must be given credit for allowing him to tinker and tweak this amazing device so all of us may enjoy listening to the CD.
And best of luck to Theo for keeping the metallic wolves at bay.
Rick McInnis
thanks rick. I began going metallic with cplay 2.0 15 and above, only exception was 18 which never went metallic. Eist on or off never mattered. Going lower on voltage and speed seems to have an effect. I can only go to .85 volts (set in bios) which reads .80 something in cpuz. When I go lower than host clock control of 157 I get instabilities. And of course running 44 sample rate is no problem. I seemed to think it was related to the revisions associated with asio 'refinements or changes' (I'm saying associated with not caused by).
Also playing music files off of flash drives helped a lot. I believe it is somehow associated with power supply or power line noise. I guess if I ever go to ssd's or more battery produced power supplies will that probably help too.
Anyway I have concluded that 44 is better sonically in my system than 192 so its a moot point now. But if I could get sonics of 25 with non metallic performance of 18 I would be a happy enough camper to try 192 again.
Theo and all,
I can play through entire albums and nothing goes wrong. I can stop a selected album and ask for a new one and all is fine.
It seemed to happen (and it was only twice) when I accidentally hit one of the mouse buttons. After I re-started all was fine.
I think it can only be because we are running our machines at such a low level. There is really nothing else that could explain it. Considering the above it would be hard to believe there is any stress on the power supplies. My ANTEC must be loafing along. If anything one wonders if there could be a problem from the PS running below the level it was designed to handle.
Since I am needing a new MD for the home computer I will try the board recommended by abysstw.
As far as the upsampler is concerned, for many of us, including myself, if the computer does not do the upsampling your DAC chip will. Most of the current available devices contain these chips. Not using the upsampler just passes the task to the internal DAC which, in my case, does NOT do as good a job.
Those having DACs with chips that do not do this could very well experience better sound without the upsampler, but I have no idea.
When this happens it sounds so awful I figure if we could find a way to make it repeatable at will it would become the next rap sound sensation only to find its way across the spectrum of POP music and then Phillip Glass could attempt to formalize it and we would ultimately hear it in concert halls around the world.
Don't say I didn't warn you.
Bye,
Rick McInnis
Hi theob,
You may want to try out the GA-EG45M-UD2H.. The sound is beautiful, and my metallics problem completely went away after I replaced my old GA-G31M-S2L with the new board. I'm able to go down to .75v without any resets, hiccups, glitches or whatsoever. With my old motherboard, going below .85v was not possible without allowing the machine to "warm up" for at least a couple hours.
outstanding I went thru the review of this mobo (link shown else where under pc audio) and this thing looks like the real deal. could you get all the bios settings in cics specs (plus obviously the new ones associated with voltage, speed and timings)? how low a host clock control did you get?
do you know if cics is going to try this?
it looks very promising!!!
The BIOS of the EG45M is quite flexible. It offers some new settings, but a few of the old ones are missing.
I'm not sure whether cics plans on trying it personally, but he seems to think highly of the new motherboard (see cPlay 2.0b25 release post and link below). Greg (GStew) says that his GA-EG45M-UD2H has already arrived, so you may want to hear his thoughts first.
Even with my current (probably non-optimal) BIOS settings, the new setup sounds significantly better than my old one with the recommended settings applied.
If you do buy an EG45M, please spend some time trying out the various settings, and help the community realize the MB's full potential.
I'm still in the process of loading & customizing it's setup... I did discover that you cannot activate Windows XP AFTER you've done most of the customizations (through the autoruns disables and eliminations), so I started to reload Windows XP again 9pm last night.
I did listen to it partly optimized in a garden-variety case with a stock non-optimal PS and a not-separately-powered Juli@ for awhile yesterday afternoon. The sound was promisingly-good. I hope to have it fully setup by Wednesday evening.
One trick I tried that worked out well was to disable things like the on-board network card BEFORE loading Win XP... that way it doesn't find them during the installation, doesn't load the drivers, and at worse reduces the number of things you need to remove later... and at best makes for an overall cleaner install.
don't understand, hanssatink said that he did not have to reload windows. pls say more about this.
And while the new motherboard did function without a new load of WinXP, I figured that the cleanest and likely best-sounding setup would be with a new, fresh, and clean install.
With the trick of disabling un-needed devices in the BIOS before loading Windows, I'm getting an even cleaner install than my 1st one. Plus with the experience of loading a cMP the first time behind me, this time is going rather quickly.
So you can use it without a fresh install and re-customization... but for the cleanest installation and the greatest probability of the best sound quality (and lack of problems, like your 'metallics' issue), a fresh install is best.
Greg in Mississippi
I'll probably do a clean install and customization just to be sure I'm getting the best possible SQ. I also plan to further reduce the windows footprint with XPLite this time around.
thanks for explanation
Isn't it the real world that windows only loads device drivers wich are presented by the hardware. Example; when i disable the onboard soundcard, usb or lan in the bios, windows won't load the drivers after a restart. Sound quality is imho not influenced by the size of windows installation, but the sum of processes running. Enable/disable hardware before or after installation make no sense i suppose.
Honestly, I don't not really sure that's the only impact of non-used drivers. If they show up in Windows somewhere (like in the Device Manager), they at least add to the size of the Registry. Other impacts... I don't know, I'd be interested in input from those more versed in Windows internals than I am.
One thing I have learned from doing this project is that there is a lot about Windows that I don't know and that things I didn't think made a difference do. So I'm playing it safe by making my system as clean as possible.
Greg in Mississippi
did you have any trouble fitting everything in? are you using the thermalright fanless cooler? It looks like it could be a tight fit.
No trouble at all. I use the Coolermaster Gemini (see pic). Not sure whether the thermalright fanless cooler will fit. Can you post the dimensions of your cooler?
Thermalright SI-128 SE Main Features:
Technical Specifications
Model number: SI-128 SE
Dimensions: 125 x 145 x 91.5mm (LxWxH)
Materials: Copper base, 8mm copper heat pipes, and aluminum fins
Finish: Nickel plated
Weight: 510g (heatsink only)
Intel support: Socket LGA775
AMD support: Socket AM2
Recommended fan: all 120mm
The gemini seems to take up alot more space than the SI-128SE, so I think you'll be fine. Not 100% sure though.
Oh - I am glad someone else is getting the metallics - not really - but it is good to know that it is not something that we are all doing wrong.
I know that on my setup the metallics are only at 192khz and it always starts as cPlay is reading the next track. Since I have my hard drives on a separate psu it can only be either a fight between cPlay and Juli@ for processor time or for power.
I will try the new mobo but I want to experiment a bit more with the Juli@ sound card psu (Sonics battery mods) to eliminate the possible psu conflict. I have also wondered about the screen and keyboard being permanently attached (especially the screen as it is also being updated at the start of each track and I have noticed it skip the display of a second or two - e.g. it jumps from 0.02 to 0.4)
Having said all that I am currently running at 96khz very happily.
thats fairly similar to my experience.
Refer to ReadMe.txt for important details & specifications.
Change Log (2.0b25 release):
- ASIO output interface refinement
- Minor DSP refinements
- Fix for soundcards using latencies of odd samples (rare)
No further releases are planned.
Please REMOVE previous versions before installing cPlay 2.0b25. Normal (SSE2), SSSE3 B9 and SSE4 B9 versions are available. Running cPlay on a CPU not supporting the required instruction set will cause cPlay to exit immediately.
For cMP² users, review these BIOS changes (underclocks FSB, CPU & disable additional USB items). New settings are in effect allowing for lower power consumption (below 20W). Recommended Gigabyte mobo GA-G31M-S2L is no longer available, use either GA-G31M-S2C, GA-G31M-ES2L or GA-EG45M-UD2H ( advanced settings ).
Tried SoundForge Seattle for V25 and all downloads were corrupt, switched to San Jose-success.
WOW, what a difference between V22 and V25. Soundstage is immense!! Bass somewhat subdued but still there.
Listening to Alan Parsons and can really, really hear the difference.
I recently made big changes in my Cmp machine: Linear PS, Pico PS, filtering caps, Hammond L193 and a Greg Stew modified Juli@ so changes weren't subtle. But still going from V22 to V25 is stellar!!
RayBan
Can't agree more with the magnificence of 2.0b25!
Did you modify the Pico itself, by chance? I found that replacing the stock caps with a BG NX 1000uf/25v improved SQ noticeably. Adding "output filter caps" to the 3.3v and 5v rails seems to also have a significant effect on SQ.
I added 470 Oscon's.
Having made so many changes at once (not recommending anyone do so) masks the ability to judge which had the most (any) impact either way.
RayBan
Hi Cics,
When I was building my cmp2 rig, I got to a point where there was a dramatic improvement and a 3d flesh to things. But I made more changes and some where along the way that improvement was lost a bit.
But v25 brought it all back.
This is a very big leap forward.
Nice job!
Hi Cics,
So I have settled on b25 SSE4 as well. Needless to say the resolution of B25 SSE4 is the highest, surpassing wavelab, Samplitude and Saw Studio, Musicality can be adjusted using DSP buffer. Again if resolution is your objective of cplay then I think you have done 300%. I support the idea of taking a rest for the moment. After all it's been a full year of hard work.
I am very very satisfied with my current system. I will go back to study how to write a playlist program.
(FYI My PC/Audio configuration can be found at http://canhtpcbeatcd.blogspot.com)
Thanks again cics for a wonderful year of surprises after surprises. Greetings from Hong Kong.
First I downloaded both cPlay 2.0b25 and cMP 1.2. Then I installed cPlay just to make sure it worked like the earlier version 2.0b22 I briefly tried. Needless to say I was amazed at how well version 2.0b25 sounds, way better than anything else I've heard.
Now the rub is I've never used cMP before and I wasn't sure whether you install cMP before cPlay or vice versa. I tried cMP first and found that I had few computer controls and I had to unplug the computer to even reboot. So then I install cPlay first and then cMP which seems to work but I'm not sure if that's the correct sequence. I thought I read "The art of building Computer Transports" but I must be tired since the document wasn't clear and I thought it to be out of date with a focus on foobar and not cPlay.
Please forgive my eagerness to achieve quick success but I'm excited and would be ecstatic if I knew the correct sequence for cPlay, cMP and how to bring a starter library of about 300 WAV and FLAC files into cPlay so I can relax and enjoy the rest of this weekend.
Z
cMP's user manual explains all the operational details needed (see Installation & User Manual (18 pages) - be sure to select the second PDF document). Sequence (easiest):
- Install Soundcard & ASIO drivers
- Install cPlay - test
- Install cMP - test in XP Mode thereafter switch to cMP Mode
cMP works on video resolutions of 1027x768 or higher. Pressing 'X' exits cMP (there's different behaviour depending on mode you're in - all explained in manual).
I had pretty much given up on ever seeing a response. So I did some exploring, found what I needed from earlier posts and worked things out to my satisfaction. The biggest impediment to the installation was actually finding the Installation & Users Manual, since as some earlier posts indicated that links to it were broken. Even though I eventually found it on my own, thanks for re-posting the link with an explanation for others that may have similar difficulties.
I believe any perceived difficulty is somewhat of a good thing because it forces you to learn. Similar to other DIY efforts where one's own sweat and trials and tribulations can lead to greater rewards than what one could ordinarily find or afford in commercial audio.
The biggest problem I have so far is with the cMP and cPlay websites. Its just too slow to navigate quickly. If there is a better way please let me know.
The other somewhat minor problem I have is the apparent need to re-rip my CDs that were ripped to the AIFF format for iTunes on a Mac since cPlay does not support AIFF. Thus I am using EAC to rip to WAV files since I also need to create cue sheets for cPlay. In the meantime I can still use cMP and MediaMonkey to play all my AIFF, FLAC and WAV files on my XP computer music server or just use cMP/cPlay to play the few dozen CDs I've ripped so far.
Use the Windows exe version (Mac is experimental).
Command would be something like this:
sox "input.aiff" "output.wavpcm"
Then rename the output file to "output.wav". Thereafter, use Alan's (AudioAl) playlist utility to create cue sheets.
and there was me thinking theob was having a laugh (Men In Black??) with his 'best of the best' - nope - no joke - this is astonishingly good. Running at 96khz I am getting music which is hard to describe (to do justice to it) - at the moment I am listening to a Janacek String Quartet, The Kreutzer Sonata, the power and emotion is unbelievable, the bows hitting the strings is visceral. Earlier I listened to John Lill playing Beethoven Sonatas not only with power but with delicacy; Eleanor McEvoy was standing a few feet away and the bass playing was superb.
Like all the others I can only thank cics for his hard-work and his willingness to share his findings with us all - very unselfish and generous despite our many 'complaints'.
Have a good 'rest'.
cPlay 2.0b25 is the most delicate/detailed w/o hardness mids to upper mids I have ever heard period! I am listening to J P Rampal's Japenese Melodies for Flute and Harp and I am thunder struck. This is beatiful!!Clearly best of the best.
Also I continue to try different dsp buffer settings and I have gravitated back to auto. But if I have something that sounds a little hard or etchy I reset dsp buffer to small and it is greatly improved (I'm talking 16/44 files). But if small buffer size is used where it should not be then bass, depth of field perception and bass dynamics suffer. When I use the small setting to improve a marginal file I simply raise volume 1.5 db or so and it brings back some of the losses in bass and dynamics.
But I also note (per appleteapot) that the small setting is overwhelmingly beatiful on female voices (try some of channel classic pieces).
Again thanks cics!!!
Btw I'm using sse4.
Many call hobbies pastimes. I have been listening to cPlay 2.0b25 for over an hour and it seems like less than 10 minutes. I can only state that one: I have the right past time and two: it is enhanced when listening to cPlay 2.0b25.
Edits: 05/08/09
Reason: SQ in B25 will be difficult to improve on, especially SSE4 B9 version.
That is very good to hear! You do deserve a rest--your recent burst of creativity has been dazzling. I'm still trying to get a handle on 25 (so far its all been good).
Edits: 05/07/09
Thank You, Cics! It is a wonderful player and it is a wonderful hobby. I think I understand You.
Serge.
P.S. Hardware will change and times will be different. There is always new morning after night.
I think we’ve been spoiled with cPlay, at least, I perceive it to be the most detailed and clear sounding player (in my system and environment).
I would also like to thank for all the work U are doing and also express my anguish if the development of this outstanding player is over.
I’ll be using it for some time to come!
Thank you for all you have done for me in terms of helping me build cmp^2. You are a scholar, a humanitarian and a great audiophile. Please don't keep away from this website. Your comments are always brilliant.
(Please say it ain't so).
The following are my current settings.
1) CPU Vcore at 0.75v
2) CPU Termination Voltage at 1.05v
3) CPU Reference Voltage at 0.669v
The .703v in my previous post was a typo.
I'm sorry to hear that there are no further planned releases of cPlay. Then again, most of us didn't expected another 25 releases after cPlay 1.4. So we can always hope, right?
Just looked at the GB manual.
Suggested settings:
- CPU Clock Ratio=6x
- Fine CPU Clock Ratio=0.0
- CPU Host Frequency=150 [CPU-Z to show core speed of 900MHz]
- MCH Frequency Latch=150
- System Memory Multiplier=2x [CPU-Z to show Memory speed at 150MHz, 1:1 ratio]
- DRAM Timings Selectable=Auto [CPU-Z to 3-3-3-?]
- CPU Vcore=0.85000V or lower [Make sure "CPU EIST Function" is disabled in Advanced BIOS Settings]
- CPU Termination=1.05V (or lower if available - this is the same as FSB "De-overvoltage" at -0.15V)
- CPU Reference=trial n error (this replaces GTLREF)
- MCH Core=lowest possible setting (the northbridge voltage?)
- DRAM Voltage=1.800V (or if possible, lower - test this)
- CPU Clock Drive=700mV (found under Advanced Clock Control)
- Other: Disable PAVP Mode (found under Advanced Chipset Features)
cics,
Thanks for the suggestions. I tried these settings today and here's my results:
CPU Clock Ratio=6x - Same as my previous settings
Fine CPU Clock Ratio=0.0 - Not changable, but automatically set to 0.0
CPU Host Frequency=150 [CPU-Z to show core speed of 900MHz] - I tried this and went back and forth a few times between 150 and my previous setting of 133. I marginally preferred 133 with it providing a bit more midrange presence and body, but need to spend more time comparing the two with lowered CPU Vcore voltages.
MCH Frequency Latch=150 - The only options available are 266Mhz, 200Mhz, 333Mhz, 400Mhz, Auto. I'm currently using Auto. When I used 200Mhz before, 2.66C was the lowest System Memory Multiplier available and it did not achieve a FSB:DRAM ratio of 1:1. When I tested it before, I preferred the Auto/2.00B setup over the 200Mhz/2.66c setup.
System Memory Multiplier=2x [CPU-Z to show Memory speed at 150MHz, 1:1 ratio] Options are Auto, 2.50A, 3.00A, 2.00D & B, 2.40B, 2.66C, 3.33C, 4.00C, I have set to 2.00B
DRAM Timings Selectable=Auto [CPU-Z to 3-3-3-?] - I'm using 3-3-3-5 with Kingston ValueRAM
CPU Vcore=0.85000V or lower [Make sure "CPU EIST Function" is disabled in Advanced BIOS Settings] - I had previously disabled "CPU EIST" and gotten down to .75625v. I went down three more steps over the day today, but got an interesting instability... the system locks up and I get a high-pitched squeal out of the speakers. Luckily, it hasn't blown a tweeter yet! I'm back up to .75000v now and checking to see if it stays stable there. BTW, I was not getting this before... either lowering the CPU Vcore further or setting the CPU Clock Drive down to 700mV caused this.
CPU Termination=1.05V (or lower if available - this is the same as FSB "De-overvoltage" at -0.15V) - I was already at 1.050V, the lowest available setting
CPU Reference=trial n error (this replaces GTLREF) - I was currently at 0.669V and tried 0.654V, the only lower setting available. I marginally preferred 0.669V, but I could not articulate why. Difference was subtle.
MCH Core=lowest possible setting (the northbridge voltage?) - I was already at 1.100V, the lowest available setting
DRAM Voltage=1.800V (or if possible, lower - test this) - I was already at 1.800V, the lowest available setting
CPU Clock Drive=700mV (found under Advanced Clock Control) - I changed this. I did not listen to the different it made, will have to go back and do that later. Also, if my instability behavior stays, I'll set this back to 800mV to see if it fixes it.
Other: Disable PAVP Mode (found under Advanced Chipset Features) - I had disabled this when first setup this board. I also set DVMT Memory Size to 128MB. the lowest available setting.
The major changes from my previous setup were setting CPU Host Frequency to 150 (from 133) and setting CPU Clock Drive down to 700mV. The Bios settings do not allow an MCH Frequency Latch of 150.
All of these settings still result in the following:
In the BIOS...
CPU Freq shown as 800Mhz normal, 133x6 current
Mem Freq shown as 533Mhz normal, 267Mhz current
And in CPUZ...
Core Speed 1192.6
Multiplier x6.0
Bus Speed 198.0
FSB 794.8
DRAM Freq 198.7
FSB:DRAM 1:1
3-3-3-5
That the CPUZ Core Speed is still shown at 1200 or so makes me think that there are still some limitations in the BIOS. I very interested in seeing what the updated BIOS Gigabyte is providing Abysstw will do. I'd also like to have the option to disable Spread Spectrum... These two things should give us the same level of flexibility we had with the GA-G31M-S2L BIOS.
Sonically, from memory, my impressions are that this board adds a nice sense of 'solidity' and micro-dynamics to the sound up and down the spectrum, but with the current settings, the treble and fine details were better with the GA-G31M-S2L. The GA-EG45M-UD2H is not bad, but the GA-G31M-S2L was magical for me. But this is based on long-time-recollection (three weeks!) which is notoriously faulty, I really need to put the GA-G31M-S2L back in the system and listen a bit to make a valid comparison.
So progress is being made... and many will prefer this Mobo the way it stands. It's still a mixed-bag in my book... but the things it does well are so good it's worth working with it!
Greg in Mississippi
Hello,
I just started building a memory player and I bought the GA-EG45M-UD2H.
I did a little testing yesterday with all the settings just to see if I could underclock everything with success.
Here are my BIOS settings, with my results in CPUz.
I started (just like GStew) with the settings listed for the GA-G31M-S2L : (see post below)
http://www.audioasylum.com/forums/pcaudio/messages/3/36343.html
Then, I whent in the MB Intelligent Tweaker and entered GStew's settings :(see post below) http://www.audioasylum.com/forums/pcaudio/messages/5/51752.html
At first, it didin't work (multiple boot) so I had to change somme settings in order to be able to boot :
CPU Host Frequency: 133 MHz
DRAM Timing Selectable: Manual
CAL Latency Time: 3
tRCD: 3
tRP: 3
tRAS: 6 (With my 512MB ValueRam, when I had tRAS to 5 the BIOS would not boot)
CPU Vcore: .9v (still testing this setting)
CPU Termination: 1.100v (still testing)
CPU Reference: 0.688v (still testing)
CPUz 1.49 tells me :
CoreV : 0.864V
Bus Speed : 196 MHz
Rated FSB : 785 MHz
MEMORY : 3-3-3-6 with Command Rate 2T
Just to let you guys know, I couldn't disable USB2.0 because my mouse wouldn't work without it (strange!)
I don't have the Juli@ plugged in yet and I don't have the Granite PSU's.
others : CPU E7200
psu : Antec EarthWatt 430
RAM : 512 MB ValueRam
BIOS Rev : R3
I'm not sure which setting will make a SQ improvement because I didin't configure windows/cMP/cPlay yet!
Thanks
Etienne
Etienne,
Two things...
First, the CPU VCore setting is one that is best to sneak up on... your .9 volt is a safe starting point. After you have your machine fully configured and running cMP, you can drop it by a step or two, boot the machine, run it for an hour or two, then if it seems stable, drop it by another step or two and repeat. I'm at .75625v now and at the edge of instability... I got some minor, occasional playback glitches when I first set it this low, now that it's been running a few days it's pretty stable and I'll try one more step down soon.
Second, I experienced the same issue with the USB mouse with USB 2.0 shut down in the BIOS. But after I loaded the Touchpad software for the Zalman touchscreen, I could shut down USB 2.0 and the mouse would work ok. If you have the recommended Zalman case with touchscreen, this is a good path forward. If not, I'm sure there is a driver that can be loaded to make it work, but I haven't spent time trying to track it down.
Good luck. Let us know how it works out!
Greg in Mississippi
P.S. Theob & all, I'll be posting a few impressions of the sonic differences soon. Now that I've got the new board as configured as I think it can be with the currently available BIOS settings and it's had a week or so to break in, I want to put in the old motherboard and listen a bit before publishing detailed impressions. But in short, it sounds very good in some ways, but I think additional BIOS options and/or configuration settings will be required to get the best performance out of it.
great cant wait to hear your sonic reaction vs other gigabyte mobo!
I went to set up the old board and found that my spare Thermalright cooler was a different model. I wanted to have the cooler already mounted to try and minimize the time and effort to swap between them.
I have a second cooler of the same type as the one I have alreay on order and it should be here Friday... hopefully I can do the comparison over the weekend.
Greg in Mississippi
P.S. I'll also post some comments on the latest BIOS settings suggestions from cics tonight.
thanks
Good post! I'm interested in your sonic reaction once you get fully onboard with cmp^2.
Vcore 0,75?
What processor do you have?
.
Since yesterday i use a GA-EG45M-UD2H too with E7400
The normal CPU core voltage is 1.24
I am now at 1.1 volt. Don't know yet what the lowest possible is. 0.75 doesn't work.
What is the normal voltage of the E7200?
Great Board by the way. It burning in at the moment :-)
How is it sounding? Any installation issues? What about support for a crt type display (looking at back panel it does not appear to have the normal multipin connector)?
Use this board with a E7400 @ 0,9 V and 7x multiplier (1,8 ghz)
The boxed intel cooler is very quiet. No need for expensive cooler. rest on the board is passive cooled.
Installation is simple. No jumpers at all. This board has a DVI and HDMI connector. Video quality is excellent. After HW installation i'll started up with previous XP on harddisk. I had luck. XP started and i could easily update chipset drivers and other drivers. No need to install a fresh XP and do all the tweaking again.
Bios settings are huge and complicating (for me) I followed the guide from Cic's. Most settings are equal. Disabled as most as possible to keep a running system. Maybe i'll find time to post all the settings. (seperate thread?)
The sound is very clear, open and airy with Cplay SSE4. Highs are little bit metallic, but out of the box not bad en very promissing. The system is now burning in for a least 48 hour.
One great benefit of this board is the self-repairing Bios. When the system doesn't boot anymore because of tweaking the bios, you don't have to reset the Bios. The pc starts a few times on itself and then booting again. So playing with the bios settings is easy.
I spent some time tonight working with the BIOS settings of my GA-EG45M-UD2H.The overall state of my cMP is that it is a fresh Win XP install (SP1) that is mostly customized/optimized through component removal, configuration, device & services disabling, AWE, and Minlogon (I'll post some updates and tips based on my experiences doing that later, likely next week).
I've based my target BIOS settings on the ones listed for the GA-G31M-S2L board.
In the MB Intelligent Tweaker (M.I.T.) settings page, here's the best settings that I've gotten so far:
Robust Graphics Booster - Auto (default)
CPU Clock Ratio: 6X (lowest available)
Fine CPU Clock Ratio: 0.0 (default)
CPU Host Clock Control: Enabled
CPU Host Frequency: 133 MHz (I set this as low as 110, but it sounded over-etched at anything lower than 130 and I may ultimately prefer it higher than 133)
PCI Express Frequency, all the Advanced Clock Control settings, and Performance Enhance all left at default values
(G)MCH Frequency Latch: 200 MHz
System Memory Multiplier: 266 C
DRAM Timing Selectable: Manual
CAL Latency Time: 3
tRCD: 3
tRP: 3
tRAS: 5
Advanced Timing Control Settings left at default values
CPU Vcore: .7625v (I started at .850 and have been slowly setting it lower, trying to get to abysstw's .75v or cics' magic .74375v)
CPU Termination: 1.050v
CPU Reference: 0.669v
MCH Core AND DRAM Voltage left at default values
According to CPUZ 1.49 (not sure I want to install 1.51 with it trying to make ASK.com my homepage!), this gives me a Core Speed of 1188, Bus Speed of 198, and Rated FSB of 791. This compares to a core speed of 900, Bus Speed of ????? (sorry, I didn't record this and can't easily check it anymore), and FSB of 600 for the GA-G31M-S2L board.Not bad, but I have the feeling it's not really optimized yet. Especially since I can't get it down to match the Core Speed and FSB of the previous board.
Thoughts?
Also, the setup above is with the Mushkin memory. With either Kingston memory (ValueRAM or HyperX), I can't get good settings. When I put one of these modules in, the board does the multiple-reboot that signifies it is adjusting the BIOS settings because they didn't work. After it is done and finally comes up, it ends up with Core Speed at 1600, Bus Speed at 266, and FSB at 1066 and the RAM timings have been set to something like 5-5-5-12 (sorry, I din't record those). Thoughts on why this is happening too? Maybe I need to do a 'Load Failsafe Defaults' and 'Load Optimized Defaults' before putting one of these modules in?
TIA!
Greg in Mississippi
Edits: 05/12/09
Try setting MCH Frequency Latch to Auto and Memory Multiplier to 2.00D. IIRC, that lowers memory speed to 200Mhz and FSB to 600 or so.
How does it sound like with your current settings?
I think the UD2H probably prevents bus speed to go below 197MHz. Right now, my CPU Host Frequency is set to 200 since the core speed wouldn't go much lower than 1.2G anyway. Do you hear sonic improvements by lowering the CPU Host Frequency, even if the core speed doesn't (appear to) change?
I'll call Gigabyte to confirm whether the 197MHz "limit" is intended, and if there's a way to circumvent it.
***** EDIT! I WAS WRONG!!! The FSB stayed at 800... I could not read the small print on the Zalman screen. See more info in post below! *****Just before leaving for work this morning, I tried your trick of setting MCH Frequency Latch @ Auto and Memory Multiplier @ 2.00D (2.00B worked too) and it did get the FSB to 600. I didn't look at the memory speed, but did notice the Core Speed went up a bit to 1206 or so.
I didn't get a chance to listen this morning and only did a bit of listening when I was messing with settings last night. The sound showed promise, with good solidity to the sound along with mucho details, but I wasn't totally blown away yet. I did think I heard a difference with various Host Frequency settings, with settings below 130 sounding too etched, but as you indicated, the Core Speed did not vary as I made those changes, so it may have been imagined.
I should have a little time to compare the Freq Latch @ 200/Mem X @ 266 vs Freq Latch @ Auto/Mem X @ 200 this evening, along with listening to various Host Frequencies from 200 and below AND I'll try to get the Kingston memory working with a BIOS reset to Failsafe Defaults, then Optimized Defaults, then set it up again.
Much to do yet.
Greg in Mississippi
Edits: 05/13/09
Have you tried using the Gigabyte EasyTune6 tool instead of BIOS to make the changes? I talked with Gigabyte's technical support, and the engineer said the program may solve our "problem". However, I couldn't install the program on my fully optimized cMP machine, so it'll have to wait until I have the time to reload XP.
I suspect that you'll notice SQ improvements from the new settings. Memory Multiplier @ 2.00 brings FSB:DRAM to 1:1, a parameter that - IMO - is critical to SQ.
I didn't try the EasyTune6 tool and my cMP is well beyond where I can easily install a program like this too.
I'm skeptical that it'll do anything for us that we can't do with the BIOS settings, but I'm willing to be wrong on this. I'm just not likely to be willing to go through the work to get it installed on my machine at this time.
Did you get any more details on what it might do to help us get the machines optimized?
On the setup with FSB:DRAM @ 1:1, I'll listen tonight and let you know.
Greg in Mississippi
I didn't get anything else that's helpful, unfortunately. She asked me to take screenshots of my BIOS (I assume she means photo...) and EasyTune (cpu-z doesn't qualify) for them to see the "discrepancies" and find a solution. I know, sounds like customer service BS to me as well. I highly doubt that EasyTune would display core/bus speeds that differ from cpu-z.
Anyway, I'll call them again later today, and see if I can get anything helpful.
LOL... Cust Support BS indeed. They really don't know how to do what we need! (IMHO)
AFAIK, the main issues I see are:
1. Setting the processor speed down below 1200 or so.
2. Setting the FSB down to 600... I was wrong, your trick with MCH Frequency Latch @ Auto and Memory Multiplier @ 2.00D didn't set the FSB down to 600, it stayed at 800, I just couldn't read the numbers on my small Zalman screen!
3. Getting access to some of the missing options such as disabling Spread Spectrum.
Maybe cics can work his magic and get a BIOS update released!
The picture of a rabid audiophile is me sitting in front of my cMP at lunchtime eating a sandwich and checking BIOS settings!
I did listen for a couple of minutes... still promising, but not quite there yet! I'll try to get the other memory sticks to work this evening.
Greg in Mississippi
More time tonight tweaking the GA-EG45M-UD2H BIOS settings.
I started out focusing on two setups... First, the one that worked the best last night:
Settings:
CPU Host Freq 133 MHz
(G)MCH Freq 200Mhz
Sys Mem Multi 2.66C
Giving:
Mem Freq 800 / 356
And in CPUZ:
Core Speed 1188.6
Multiplier x6.0
Bus Speed 198.2
FSB 792.2
DRAM Freq blank
FSB:DRAM blank
The other settings were the same as the baseline settings from last night.
And the one that gave the best settings today based on Abysstw's advice:
Settings:
CPU Host Freq 150 MHz
(G)MCH Freq auto
Sys Mem Multi 2.00B
Giving:
Mem Freq 800 / 300
And in CPUZ:
Core Speed 1187.5
Multiplier x6.0
Bus Speed 198.0
FSB 791.4
DRAM Freq 198.0
FSB:DRAM 1:1
The second was better in objective terms... Better definition, more silence between notes, good composure when the music got complex, just as expected with the 1:1 ratio, but the highs were a bit etched. The first was more listenable with a softer and more delicate treble and I found my head nodding in time with the music more with it, but it was not as defined.
Remembering that I thought I'd heard sonic differences with various CPU Host Frequency settings even though the Core Speed didn't change with differnces in the Host Frequency, I tried a Host Frequency of 133 in the second setup. Hmph... The strengths of that setup was retained, but the highs lost much of the etched-ness and it was overall more listenable. That's what's playing now and while I still don't think this is the ultimate setting for this board, I could listen to it long-term now.
I'm very curious to see if others can duplicate my results or if it was my imagination.
I didn't have time to work out the issues with the ValueRAM and HyperX tonight... That'll have to wait until Sunday or later as I’m away for business for a few days.
Greg in Mississippi
Back home after being in the DC area for the weekend, I played with the cMP settings just to unwind a bit.
1st, I did two of the last customizations remaining... I checked to see if the soundcard was sharing an IRQ (it didn't) and then disabled the un-used USB controllers. BTW, this produced a noticable improvement in transparency... well worthwhile.
Then I tried to compare the ValueRAM vs HyperX vs Mushkin memory sticks. With the BIOS settings I was using early last week, both the ValueRAM and HyperX would not get a clean boot when installed. Now the ValueRAM worked ok... but I still had BIOS errors with the HyperX. I didn't take the time to further troubleshoot the HyperX, but did go back and forth between the ValueRAM and the Mushkin... and I preferred the ValueRAM. I heard better definition up and down the spectrum, but most noticable in the bass. Tonal balance was a bit different between the two, with the Mushkin sounding a bit more forward in the mids and lower treble.
I would judge that whether you liked one or the other could be very dependant on what one considered important and what worked best in one's system... and since they both cost less than $20 ea, if you are so inclined, get one of each and compare yourself.
Maybe this weekend I'll spend more time with the HyperX. In the meantime, I'm very curious what the BIOS update will bring.
Greg in Mississippi
I just convinced Gigabyte to give me a "special" BIOS version. I'll let you guys know if it fixes the problem.
Hello Abysstw,
I know I'm replying on an old post but:
any news on that special BIOS for the GA-EG45M-UD2H?
Did you actually receive it?
Are you able to get bus speed to go below 197MHz?
What about spread spectrum?
Thanks
Etienne
abysstw,
Any news on the how the 'special' bios worked?
Inquiring minds & such!
Greg in Mississippi
That's GREAT!
See if they can give us the option to disable Spread Spectrum at the same time!
Thanks!
Greg in Mississippi
P.S. Contact me via my Asylum profile so we can talk Juli@ mods.
nt.
Keep us posted as the mobo burns in relative to sonics. did you use the easy tune facility or utility to adjust speeds/voltages?
Did the voltage adjustments manually in the bios.
It does not seem to be available in the UK (advertized but not stocked) so I will have to be a late adopter.
Glad that you like it :)
I think the normal voltage of the E7200 is 1.1v. I'm out of town right now, so you might want to reconfirm this with other inmates.
I compared a 512Mb stick of HyperX to a 256Mb stick of ValueRAM in my supercharged cMP2 Saturday evening with a rematch Sunday evening so I could test it out with b24 too. To try and cover all the options, I ran the HyperX at various timings, 3-3-3-7, 3-3-3-6, and 3-3-3-5, along with the manufacturer's recommended 4-4-4-12 and a wild-card 4-4-4-5.
In short, I liked the HyperX best at 3-3-3-5, but still prefered the ValueRAM at 3-3-3-5. The HyperX was very clean and didn't do anything wrong, especially at 3-3-3-5, but the ValueRAM gives me deeper and darker silences between notes and a wealth of texture and density to the sounds that added to the sense of realism along with a greater sense of presense and vibrancy in the midrange
As for the 44.1 vs 192, what I heard provided me with more insight as why I prefer 192 in my setup... it's that wealth of texture and density to the sounds which is present with 192 and diminished with 44.1 (and it grades down pretty linearly as you select the intermediate sampling rates between these two). The metaphors that come to mind... "average recordings start to sound like direct-to-disk"... "it can be so involving that I'll stop what I'm doing and just listen to the music"... "For the first time, I'm hearing all of each note... before now, the beginnings, ends and much of the fine textures in the middles were missing."
In my setup, 44.1 is very, very good. But in my setup, 192 is exceptional!
Greg in Mississippi
Reading the specifications of the ValueRAM they generaly need less power than Hyper sticks, as peer Cics recommandation the lower power give the best sound.
You can check on www.valueram.com, the specifications for valueram and hyperX is available for all part number.
trend in system architecture and whether one finds upsampling preferable.
I am still in analogue mode so I have no opinion to offer on #24. When I get tired of listening to my LPs and want to hear stuff that is only available on CD, I will return to cMP.
Nonetheless, with my system there was much more ease to the sound with the upsampling. My system needs no additional sharpening and if anything would benefit from intentional softening. I do not doubt there is something like this at work.
Based on the fact that my system is very high efficiency (edgar horn titan II with largest round horn and TAD 4001 and TAD 2001 (w/ small horn)as super tweeter) my guess is that similar systems would also sound "better" with the upsampler. My experience with direct radiators of low to middle efficiency would lead me to believe that owners of these speaker would not mind a little additional "sharpening" of the sound. If the upsampler smooths the tops of those sawtooth waves, those remaining tips could actually be euphemistic with the direct radiator and its comparatively more difficult job of loading a room. All of this is nothing more than conjecture but it would be interesting to see if there is any discernable trend.
Now I will get deluged with reports of 44.1 partisans telling me of their horn loudspeakers!!!
Bye,
Rick McInnis
Rick,
As always, good to hear from you!
Well... I can provide some contrary data here... my system is based around a pair of fairly in-efficient planar magnetic (and fairly rare) Eminent Technology LFT-IVs.
For me with my setup and these speakers, the additional detailing provided by the 192 setting adds sharpness to the beginning of sounds... it adds the initial impact and rise time that is present in real life (jingle your key ring), but so often missing in reproduced sound. I don't hear extra hardness, but the beginning of each note is more clear and distinct with the 192 setting... adding 'micro-dynamics' and giving my planars a taste of most horn-based systems do very naturally.
My 2 cents!
Later!
Greg in Mississippi
P.S. Don't be such a stranger!
Hey Gstew,
My magnepans dont suffer from bright treble, but they do act as radio antennas! Thus it is possible for rf to get into the amp and modulate the signal. This kind of sounds like the "digital" sound. Putting the chokes really helped add warmth and also a bit of top end resolution.
For about $12 you could really improve your emminents if they are anything like the maggies. I certainly found a great improvement:
http://db.audioasylum.com/cgi/m.mpl?forum=mug&n=139415&highlight=octamom+dawnrazor&r=
very interesting! do you think this concept might work with electrostatic tweeters (like martin logans)?
I really dont know. BUt I am finally going to hear Al's system on Friday and I can ask him for you.
Have you tried powercords on the Logans?
About 5-6 years ago I made some powercords and speaker cables for a guy who had some logans. He was supposed to use the cords on his pre and his cdp (hey, it was about 6 years ago...) and I got this call from him and he was frantic.
Apparently it was too much trouble to get behind his pre and cdp so he just tried the cables on his Logans.
The difference apparently was pretty big and shocked him (but not in the dangerous way). He said something like "...but they aren't even in the signal path, how can they make a difference?"
Anyhow, you may have already tried some, but if not give it a shot.
d
I have gotten good results with pcs and conditioners with my Logan Requests: improved clarity, detail and tighter bass. I haven't choked 'em, though.
thanks yes I have experimented with pc's on logans, not very much but some. I use the hammond inductor on many of my circuits as well as a running springs ac conditioner on my panels.
I do use an inductor, cap filter on my woofers and they improve the bass a lot. so I am always interested in these type of low cost/high impact mods.
oh, well.
Greg,
Nonetheless, a planar speaker is unlike other direct radiators because of its ability to reproduce a large wavefront without all of that frantic motion of an un-fron loaded cone.
I am an admirer of Bruce Thigpen's thinking and products. I have maintained an ET-2 arm for twenty years keeping up with his upgrades, not yet getting the carbon fiber arm wand.
I have never heard his loudspeakers though I am very familiar with how they work.
I came to horns from electrostatics. The large planar loudspeakers are excellent. I have a room that would overwhelm any of them, or at least, anyone of them I could afford.
SO, I do not think we are far off in our comparative assessments. Those speakers will follow the waveform and therefore, no surprise to me, you would benefit from the smoothing from the up-sampler.
Thanks for pointing out this discontinuity in my thinking.
Now if only we would hear from some other cMP users.
Bye,
Rick McInnis
I must confess I've been running ET arms since 1986 or so starting with an ET-1. While my vinyl source is not currently setup, I have a tweaked ET-2.5, use a high-pressure pump with serious smoothing tanks, and regulate the pressure to about 20 lbs. I do want to try the large storage tank / pump off setup when I get it running again.
Then on the planar side, I've been running Acoustats, Magnapans, Martin-Logans, and now ET speakers pretty much exclusively since 1982. The only speakers that have tempted me away from them were a pair of Spica TC-50s and I still have a pair of them around. I got the LFT-IV's for a song about a year ago... they replaced a pair of Gallo Ref-3's and I strongly prefer the LFT-IV's, even with an almost 4x price differential.
I never had a pair of direct-drive amps even tho I've lusted for them for years, but did have a pair of original CLS's and still have a pair of highly tweaked 1+1's... but my heart is currently with ET speakers and in addition to the IVs, I also have a pair of LFT-VIs and LFT-VIIIs.
TOO MUCH GEAR!
Greg in Mississippi
I promise not to tell Barney Frank how profligate you are. My, my, my, you should be ashamed!!!!
Sounds like you are having fun. OH, I did it again. I am really going to get you in trouble, it seems.
Maybe you should make it correct by calling your collection a museum of planar loudspeakers? The world does need one, after all.
Bye,
Rick McInnis
PS if it was not OK to go "off-topic" I would have had most of my posts deleted.
I had an ET 2 tonearm for about ten years. I drove it off my sears air compressor which had a 20 gallon reserve. I would start it up (sounded like a lawn mower in th house--wife hated it) then I would set the regulated pressure at some value (say 3 or 4psi) and it would run for 4 or 5 hours. Man that was great fun and wow did it sound good that way. Sigh and then I sold all my original vinyl stuff (vpi, ET2, Koetso cartridge) and even half of my vinyl. Now I have an old tecniques sl210 integrated table, pivoting arm and some mm cartridge. Haven't listened to it in 4 months.
I have Martin Logan Monolith III's biamped.
Your comment on Eminent brought back some memories and hence this post.
I, too, use a BIG air compressor. I have got a manifold to optimize this so I do not get 4 hours out of my large tank, more like thirty minutes but it is far enough away as to not be a problem. High Pressure and high flow make for a STIFF bearing.
I had a pair of MARTIN LOGAN CS-1's that were powered by a dedicated tube amp (Joe Curcio designed and built) that was once part of the original ACOUSTATs. No transformers between the output tubes and the stators.
When I got a large listening room these speakers did not have a chance unless I wore them like headphones. They are classics, though.
Sorry you got rid of all that nice stuff!!! My LPs are a part of me even though some are borderline unlistenable. They are my biography.
forgot to mention I heard Bruce Thigpen's fan generated subwoofer down in Las Vegas 2 years ago. Tremendous ability to reproduce bass down to a couple of hertz (he had a scope there to prove it). But it didn't seem to have very good dynamics. Maybe it needs more work. But he is very innovative and his stuff will be among the classics. Love to talk to him --- very bright young man.
Rick,
If you mean confess to owning a horn-loaded speaker then I may own up to having a pair of Gedlee Abbey (clone) speakers which have a 12 inch waveguide. For the most part I would describe these as efficient and transparent speakers; they are powered by a Peter Daniels Patek-style gainclone amplifier.
They have shown clearly differences in the cMP setup and until recently I have always preferred upsampling at 192k - easier to listen to / better detail. However I have of late been running at 44.1k and I have been pleased with the sound; with cPlay 24 I am not sure (after one evening) whether I prefer 44.1 or 96 or 192. Whatever the first listen to 24 put a smile on my face and that was the initial listen at 44.1.
I will let the system settle in for a few days before deciding which I like.
On the analogue side I have not played any records for about six months being so busy with setting up cMP2; it might be a good idea to just leave the system as it is but I have not started modding the Juli@ card and its power-supply.
So much to do so little time.
Refer to ReadMe.txt for important details & specifications.
Change Log (2.0b24 release):
- ASIO startup refinement
- ASIO output interface replaced
- Minor DSP refinement
Please REMOVE previous versions before installing cPlay 2.0b24. Normal (SSE2), SSSE3 B9 and SSE4 B9 versions are available. Running cPlay on a CPU not supporting the required instruction set will cause cPlay to exit immediately.
For cMP² users, review these BIOS changes (underclocks FSB, CPU & disable additional USB items). New settings are in effect allowing for lower power consumption (below 20W) as well as kernel changes. Recommended Gigabyte mobo GA-G31M-S2L is no longer available, use either GA-G31M-S2C or GA-G31M-ES2L.
due to major turntable modification in progress.
I cannot say with specifics what 24 is doing in comparison to its immediate predecessors, but there is no question this sounds very good but (here comes the big surprise) it is missing that enveloping musical cocoon effect that good analogue does so well. The image between the speakers is precise in three dimensions but there is not much of anything to either side of the speaker. Of course, this could be my DAC's fault. I am just reporting what I hear. I hope I would be considered the last person with a desire to savage cMP. I believe these are the tired old digital shibboleths and I am not saying anything that hasn't been said a thousand times.
Nonetheless, as fine as cMP is, it still is not the missing link that I so sincerely want it to be. cics has made immense strides in making digital music reproduction enjoyable; don't get me wrong I can listen to this and not be miserable, it is just the whole time I am thinking, "I can't wait to get that turntable working again".
Something I discovered in the last few weeks: with analogue I was interested in listening to unheard music which, for some bizarre reason, had not seemed interesting to me while on a steady diet of digital. I was buying lots of CD's but they were of music I had on LP.
With a CD of something I was familiar with on LP I assume my brain was able to fill in what wasn't there (upsampler cannot predict what isn't there at all) and I was able to enjoy the sound since it did sound MUCH better than the old THETA JADE and GEN Va by an order of magnitude. But as the old story goes one still cannot put Humpty Dumpty back together again.
Another thing is that I want to listen REALLY loud with digital and find nothing attractive about this with analogue. There is much greater dynamic range with the LP and one compensates with digital by turning it up LOUD to get some impact, I surmise.
To be fair, I have MUCH more money involved in the analogue set-up and when one makes a value judgment, ESPECIALLY considering cics's saintly generosity, it damn well BETTER sound better.
I am glad to have such a good back-up system. What else can one do for recordings never released on LP?
Even with all of the tweaking cMP requires to get started it is nothing compared to what is needed to get a top performing LP set-up.
And digital is quiet, one cannot dismiss this. It is a fine attribute.
Maybe cics has one incredible trick up his sleeve, and maybe there is a DAC that can change water into wine. No one will be more effusive in their praise of such a thing if it appears. But, excuse me for repeating it and I know it is said too often for many, but there is still a LONG way to go. Analogous to the line about the more you know the more you find you do not know ... I think this is happening with cMP and other digital advancements. After the glow wears off, even though one can sense a great improvement from what has come before, the "latest device" seems to point out even more clearly than you perceived previously just how much ground there is to make up. The audio tortoise and the hare ...
Cordially,
Rick McInnis
(admitted analogue partisan)
Hi Cics,
After your suggestion on the "DSP buffer" setting I start to notice the sonic difference for the 3 "DSP Buffer" settings (Small, Medium, Large). When I set to "Small", the resolution is highest, and when I set to "Large", it is very musical and comfortable to listen to. So I use Small buffer to listen to general pop, medium for male vocal, large for female vocal. This gives me alot of power to adjust cplay to suit the music :)
Appleteapot! Thank you for bringing this up. I discovered more music in my files. I won't say which setting I like best but it has moved my system closer to what I expect out of my music. There's more experimenting to do ,but here is yet another thing to adjust to make cplay better. Hey I know it was always there -- but I just discovered it.Thank you again (and obtw thank you cics!)
Just like the old vinyl days -- adjusting the cartridge vta for every record. And I mean that in a positive way.
Edits: 05/06/09
Then I wonder whether I can tweak the Lynx buffer along with DSP buffer...
At the same DSP buffer level, The lower Lynx latency is always better, sounds more accurate. So I still stay at 32 at the moment.
Hi Cics is it possible to change the DSP buffer setting so numeric values setting say 0-1000 so that we can freely tweak the character of the sound?
At small I get a very very fine grain in the highs whichs makes the highs very very finely separated --- great air around everything almost a swimming liquidity. At 1st I just loved it and was going to set it at small and forget it. But, alas, there is a tradeoff. The bass becomes weaker and the soundstage moves forward and dynamics suffer. It sort of like a 'tube' filter. So I use it sparingly where I can or where I need it.
You are right it is so nice to have. I tested only with 16/44 files. Small is killer for 24/96 or higher data files (I always match sampling rate on cplay to the data now).
Thank you very much...B24SSE4's musicality is comming back!! Tears are in my eyes... To me, previous verions resolution is so high that I can hear background noise.
B24SSE4's background noise is not as high (resolution is lowered a bit?) However, musicality has improved and I can enjoy the music more, especially female vocal. I then listen to male vocal immediately, tears in my eyes when I listen to Paul Potts. So to me this version is a very very pleasant change.
(But just watch out in subsequent versions, too much smoothness will start affecting pops and "strong stuff". )
Thank you so much!
Edits: 05/03/09
+1 but there is more for me. I loaded 24 last night (is there a prize for 1st downloader?) and it was ok. I liked it but didn't get excited. This am , in my 5:00-7:00 powerline sweet spot I began to here more into the music. I must say that I had made a couple of changes as well. Yesterday I put my audio pc on hockey pucks versus on the floor--always a good start for isolatating the source. Anyway it sounded better with 23, little fuller in the bass, more air in mids and up. After loading 24 I felt intuitively I could do more. So I reduced my hockey pucks from 4 to 3 and then put an aurios bearing (the floater type) between my case and the pucks. Wow what an omg moment! It was like the fog lifted. I could see into dense classical orchestrations. Highs were less bright, bass tighter and deeper, mids were glorious. I have many household tasks to do today but I don't want to start. I want to listen to everything I have.
I feel 24 (sse4) is more transparent but not in a 'in your face/offensive' way. I know some of this is coming from my better isolation but w/o the details/smoothness of of cplay it wouldn't be there.
An example of what I'm hearing: one of the telarc samplers has an excerpt from the 4th movement of beethoven's 9th and where the male bass comes in is very dramatic. I could always hear air around his voice but now it sounds as if there is a cushion of air all around him. Yes I sense I can hear behind him. Sounds weird but this is my impression.
There are other examples of voices where I can now clearly understand words better and have a sense the weight of these voices/chests radiating in space.
Just simply wonderful, thank you cics--and all this at 44khz.
i can't say it better in english.
:)
you keep talking I appreciate your words, especially on mods. wish I could speak german like my kids. my wife speaks italian, spanish, and some french and english, seems like I am the only one language talker in the family.
In the last version ( 2 0b24_ssse3b9) the sounds are a bit closed .
I think the soundstage of v23 is better than v24
Cics, please check into this.
I removed original release post until files can be downloaded.
riboge wrote:
"Link to download is not working. Cics, please check into this.
It's often several hours before the cPlay links work but they usually seem to sort themselves out. I've no idea why . . .
D
I had been on the fence about this, liking either way but recently I conducted some listening tests when my power line was the cleanest (5:00am -7:00am) and after I went completely fanless. What a revelation! You know when you change something and everything just falls into place how you want to listen to all your faves? I have tested for 3 days but today I finally concluded I'm a 44khz guy. Initially I thought 192 was airier in the highs but after several days listening I conclude 44khz is better defined in the bass, mids and while not as deep in the highs as 192 is definitely more true to the actual sound of music in the hall. 192 has a sort of impressionistic feel to the sound, a little out of focus but pretty. 44khz is the real thing and obtw I'm listening to cplay 2.0b 23 sse4 which by definition has no upsampling period. I often thought that my 24/96 files sounded better @ 96 than upsampled to 192 and could not seem to reconcile that with the fact that I believed 192 was better for 16/44 files. Now with 44 khz with cplay 23 I finally (at least for now) conclude that right here is my best sonics. At other times of the day (when the powerline is noisy) I think I prefered 192 because, being a little out of focus, it masks the noise induced/subtle hardness. 44 at these times still sounds good but not preferred to 192. But in the am, when I do my serious listening, 44 khz is much much preferred.
Getting back to making some breakthrough on ones system, when this happens for me I really just put on some music and listen all the way through w/o flitting around to favorite cuts or passages. And that's what I have done for 44khz the last few days. Yes it does solve my metallica issue but I am still going to try some type of ups particularly one that can allow battery operation for several hours (so I can get that morning sound all day long). Btw does anyone know of a UPS that will allow this (very feasible I think with cmp^2 operating at 20watts or so)?
Cics: juli@ analogue outs now is so far superior to juli@ digital outs through the benchmark even with its better dynamics in the bass. Juli@ is better for resolution from the mids up. I guess I have to try the Sabre like Alfred (Alfred how does it feel to be on the leading edge of dac's, power suplly mods et al).
Anyway just thought I'd share this experience. Those who have not tried 44 setting on cplay 23 please do, you may be surprised as I was.
theob wrote:
Btw does anyone know of a UPS that will allow this (very feasible I think with cmp^2 operating at 20watts or so)?
The 50/60 Hz regeneration circuits in most UPS devices are very poor quality (at least for audio, which is not what they were designed for) and the power they supply is likely to be significantly noisier than the mains. The technique has been tried before but never AFAIK successfully.
Dave
thanks dave
Theob,
Your post is a great testament to how we always need to challenge our assumptions... and as a corollary, check our settings.
At the end of last weekend, my cMP did not reboot after putting in some Juli@ changes. So I had to reset the bios and then reapply all the bios updates. That was a good thing... I'd missed that the preferred setup could go lower in CPU voltages with EIST disabled and I got it all the way down to .74375v. So I got good sound out... but the next few days and then after I did a further Juli@ mod, it didn't sound quite up to snuff.
I went back and checked the bios settings and saw that I'd missed setting the CPU Host Frequency. Set it right and the magic was back... mostly. Then I ran CPUZ and saw that my RAM settings were on 'Auto' and had defaulted to 3-3-3-7 instead of the 3-3-3-5 that seemed to work so well in my setup with the ValueRam.
So just to be sure, I compared a number of different settings options...
1. RAM timings 3-3-3-7 vs 3-3-3-6 vs 3-3-3-5 (I'd been using 3-3-3-5)
2. Juli@ outputs 1/2 vs 3/4 and having monitoring on or off (I'd been using 3/4 with monitoring on)
3. All the sampling options from 44.1 to 192 (I'd been upsampling to 192, which is why I'm putting this in your thread, Theob!)
I ended up with was 3-3-3-5, Juli@ outputs 1/2 with monitoring off, and upsampling to 192.
While I hear a slight increase in bass and midrange solidity with the lower sampling rates (and it seems to transition pretty evenly as you go up from 44.1 to 192), I get an increase in treble details at 192 with no loss of focus. And with the replacement and add-on power supplies in my setup, the bass and midrange solidity already starts out pretty darned good, so this increase in treble details makes a greater difference in my setup and to my ears.
My lessons from this week's exercises are:
1. As new versions of cPLAY come out or I put additional modifications in to my cMP, it's worthwhile to re-check the effect of the various optional settings.
2. If something does not sound right, check the settings. Having the CPU Host Frequency not set correctly AND my RAM timings different than what I consider optimal made a significant difference in my cMP's sound.
3. The sound quality of a cMP is MUCH greater than the sum of it's components... hardware, software, and settings. They all play together to produce what we hear... and if they are not set to the optimum mix, you will hear something that is significantly poorer sounding. Those people who run cMP/cPLAY on a laptop with 1/2 of the optimizations are not hearing what this setup can do. Heck, I was not hearing what this setup could do with one of the most highly-evolved cMP's around and just two settings off!
Greg in Mississippi
have you tried no upsampling for prolonged periods of time? maybe with all your mods 192 is better. but did you have any trouble getting 150 host clock control? the bios resetting itself to auto host clock control and resetting itself to auto for memory settings is something that always happens to me when I lower cpu voltage or speed. but you are right one should I always recheck all assumptions/settings.
Hey Theob,
No, I haven't done no-upsampling for prolonged listening. When I go back and forth, I try a piece of music with a fairly wide range of sounds... listen for about 1-2 minutes, then change the setting and listen again. But I'd say you have a point, that I should listen to 44.1 for awhile and see how it sounds.
OTOH, a main learning that I have from this is that some options and settings, like the amount of oversampling, have no 'right' setting and instead are very dependent on the system and a person's preferences.
Also thanks for the heads-up on the bios resetting itself. That may be why my host clock control was not set right... When I was resetting up my bios settings last Sunday, I had input everything, then was lowering the CPU voltage by steps and rebooting to make sure it work. When I was done, I did not go back and check that nothing else had changed... but I bet that the host clock control and memory settings had been reset to Auto during that process.
Danged bios thinks it's smarter than we are!
Later!
Greg in Mississippi
I does grow on you, doesn't it!
I've had little time for careful comparing, and I'm in the mood to simply enjoy music lately, and that seems to happen more with no upsampling.
Probably been said, but it is more than coincidence that your distortions started happening after the RAM-loading methods in cPlay got a lot more complicated (version 2.0). The performance graphs in Task Manager/Process Explorer are a lot busier now.
yes you are right. although I will probably still try to solve the issue I am very happy listening to no upsampling now. who knows maybe I will get bored after awhile and want to go back.
hi theob
glad to hear your problems solved.
to your question:
buffalo , batteriemods (with CMP/CPlay) etc. lead to a very,very laid back kind of listening.
Its possible to stay a while only listening with great pleasure, but as you know curiosity is going on and we still don't know what's really behind the curtain.
So if i compare the last 20 or 30 years i remenber many situations with the same feeling of beeing arrived, but far away of the perfection of todays possibilities.
Imagine a situation of real perfection-- horrible thougts.
wise words, you are right I don't know whether I'm a music lover that dabbles in the hardware or a hardware tinkerer who dabbles in music!
Given that cMP sets up the pc and allows for the selection of music from a library (which is digital) and that cPlay then takes that formatted digital file and translates it to a (for want of a better word) music stream - what then does the Soundcard do if the output is going to an External DAC?
Is it converting the music stream to another format (I2S) for transmission to its own DAC or an external device?
If so why can this not be handled by software e.g. cPlay - or is this what the professional software packages are doing?
Digital output stream (I2S or SPDIF) is produced at hardware level thus requiring a soundcard. Mobos have onboard sound hw and in some cases offer digital outs over the normal analogue outs.
Refer to ReadMe.txt for important details & specifications.
Change Log (2.0b23 release):
- DSP and ASIO (including startup) refinements
- Compiler optimisation
- UI refinements:
- Keyboard support for Pause / Play using 'p' or 'P' key
- Option to start without autoplay. Adding line 'NOAUTOPLAY' to ini file causes no playback on startup. Desired track must be manually started
Whilst B22 may be a detour of sorts it nevertheless provided huge insights for this release. B23 is a new chapter...
Please REMOVE previous versions before installing cPlay 2.0b23. Normal (SSE2), SSSE3 B9 and SSE4 B9 versions are available. Running cPlay on a CPU not supporting the required instruction set will cause cPlay to exit immediately.
For cMP² users, review these BIOS changes (underclocks FSB, CPU & disable additional USB items). New settings are in effect allowing for lower power consumption (below 20W) as well as kernel changes. Recommended Gigabyte mobo GA-G31M-S2L is no longer available, use either GA-G31M-S2C or GA-G31M-ES2L.
I'm listening to SSe4b9 and the mids are palpaple (ie rich, ripe, woodwinds sound round), bass is deep, soundstage is vast with lots of hall sound. Any remnants of upper mids harshness is gone. This is a true combination of detail, warmth of previous versions plus more.
More comments later.
B22 was important from the viewpoint of what more musical information can be had (i.e. more detail and resolution). Jitter was masking this information in previous versions. B22's bright presentation also meant jitter distortion was unacceptable.
B23 is a "fresh start" (hence a new chapter) on critical areas in cPlay. Changes are extensive and incorporates new designs. There were some 6 internal releases between B22 and B23! There are more areas I want to research but this requires much more time.
SSSE3 version renders more presence and detail whilst SSE4 is smoother. These differences are subtle. In my setup, SSSE3 B9 reins supreme (pure & natural).
Hi cics,
i think you have resolved the remaining problems of 22. SSE4 is now useable
and smoothness is comparable with the earlier SSE3 Versions.Low freq. are dryer ,space is the same, overall increased details.
Thanks
Perfect and well balanced high-end sound it is.
I send my respect to you.
This new release (2.0b23) is quite superior than the old versions.
Thanks and best regards,
Hi cics,
I think I found a small bug with new pause/unpause mode.
- Open one album
- Play some music
- Click on folder icon to open another .cue
- Folder selection pops up if you click on 'p' to browse on first folder starting with 'p' playback pauses.
pb
Keyboard actions is processed at player level even when browsing for files (i.e. keys like '.' for Stop, 'p' for Play/Pause, '[' for Previous, ';' for Phase, etc. will be processed).
Thank you very much cics! I use SSSE3b9, large buffer and 3-3-3-9. This version is smoother than b22 while resolution is one step forward. Works wonderful when Lynx ASIO buffer is set to minimum (32). So I think this version is great for strings, drums, pops.However, Samplitude still excels in smoothness. Samplitude sounds more soft and relax but about 20% less detail. So I use Samplitude to playback violin, soft female vocal...
I try to give both likes and dis-likes instead of just praising -- I hope that will give you a better picture and more information --- IMHO :)
Edits: 04/29/09
I have 2 cMP² (one using RME 96k Toslink, other using Lynx 192k Dual AES). Both have same setup but different CPU voltages (0.85 vs 0.74375) and RAM (512MB HyperX vs 256MB). For B23, I find using the 512MB HyperX RAM to be superior. I have swapped out the 256MB with another HyperX 512MB. Kingston's HyperX range does not offer 256MB modules.
SSE4 should be the smoother option.
Thank you cics! I have just tried b23 SSSE4 and this is even better than SSSE3B9. SSS4B9 is certainly smoother than SSSE3b9.So I have compared with Samplitude again using 24/192 materials, compare only using female vocal (as I am confident cplay will win in all other areas, so I just concentrate on female vocal and violin) and feel that
Priority 1: Resolution/Details:
SSSE4 is 20% above SamplitudePriority 2: Rhythm/Musicality/Dynamics
Samplitude is 5% more "musical" than SSSE4, I can concentrate in music easier.Priority 2: Smoothness
Samplitude is 3-4% smoother than SSSE4In short, now I use cplay b23SSSE4b9 more than Samplitude 10.1.
I "guess" Samplitude has some sort of buffering to enhance dynamics/musicality, therefore it is unfair to pick on SSSE4b9 because SSSE4b9's resolution is so high. I will further try CMP+b23SSS4b9 tonight.
I will try to source HyperX DDR2-800 512M (256Mx2), but I am quite doubtful whether i can find it now... Would Hypex DDR-1066 256Mx2 do the job?
Edits: 05/02/09 05/02/09
Do you think I should go to a lynx soundcard to solve my issue? I believe I have a fully implemented cmp^2 system (fanless, 2.5" laptop drive, all windows optimizations etc) but still get the 'metallics' with juli@. I know you mentioned that maybe my problem is a juli@ design defect but was that speculation or was it fact?
No. Given that you have re-built the unit using alternate parts suggests the problem is elsewhere. I recall you noting improvements when changing power supply. Is it possible to try a UPS (Uninterrupted PS)? The problem could be caused by spurious noise in your AC like abnormal voltage spikes.
quite an innovative idea. I will look into this
cics / theob,
I have finally heard what I believe theob calls 'metallics' and my cMP2 has separate psu's for ATX24 (picpsu with switcher psu) / P4 (linear) / HDDs/SSD (linear), I am also using a Juli@ soundcard - but using the analogue outputs. The problem occurs at 192kz - once it starts the only way to stop it is to return to 44.1 kz.
In the past I have suggested pooling information to see if a possible cause can be found - the readings are from cpu-z and together with a hardware list could be put into a speadsheet.
I have no problems in collating this info as it will help both theob and myself and anyone else who suffers from this malaise. It should highlight any areas in the setup of theob and myself - to me it is worthwhile as my music listening experience with digital has become better and better.
[it has been a while since I switched my vinyl system on - someday I amy compare them again]
Regards
Alan
hi alan when I get this noise on juli@ analogue outs I switch to another track then back again (perhaps several times) and then I'm ok. When this happens though if I switch to digital out I get hard core metallics through the Benchmark which only goes away with a reboot. If you have an external dac to try this, please do to see if you replicate my experience. I'm going to try an uninteruptable battery power supply to see if this fixes it because my problem always ocurrs at high power line noise with cplay @ 192 / 146 src (which is the highest cpu usage). When I went to 2.5" drives the frequency of the problem was reduced significantly. When I ran music off a fd I rarely got the problem (I think I got it once). When I listen to cplay 23 @44
Khz I don't get the problem (btw cplay 23 is very very good at 44: better bass and more transparent mids but 192 is airier).I'd be happy to do a cpuz register dump to see what we get. just let me know. and for whatever reason I can run cplay 18 all day with no metallics (any rate, any power line condition).
Edits: 05/01/09
"B23 is a new chapter..."
cics, can you elaborate what that means? What breakthrough you have achieved in this release?
Are there any plans to incorporate AIFF playback capability in cPlay? I know that I can convert file formats or re-rip, but I find the inherent ability to play AIFF, FLAC and WAV audio file formats to be a desirable feature for any player.
What are the development plans for cPlay?
B23 will cover following:
- Keyboard support for Pause / Play (use 'p' key)
- No autoplay: adding "NOAUTOPLAY" in INI file will disable auto playback. It's a good feature to have (prevents unnecessary loading of first and next tracks) if frequently selecting other than track 1.
NOT supported:
- AIFF & OGG/OGA (request via SourceForge email) formats - not much demand on Windows. It's better to convert to WAV (no impact to SQ)
- Dynamic output rates cannot be done easily due to design (integration with cMP and Kernel)
B23 is fully operational with above changes. More testing / enhancements being done...
Hey Cics,
Thanks for number 2. It will be nice to be able to disable autoplay. Sometimes many cds have a different volume and switching from a soft cd to a louder one can be interesting. Its nice to have time to adjust the volume.
Any idea when/ if cplay will be able to create upsampled files and not just realtime sampling?
Refer to ReadMe.txt for important details & specifications.
Change Log (2.0b22 release):
- Compiler optimisation done (aggressive parallel settings in effect)
This release was expected later. Revised compiler settings used in 2.0b21 caused specific behaviour that results in edginess or aggressive mids. This is related to a specific CPU instruction which is avoided with additional (parallel) settings. Although the solution was easy to implement (hence the quick release), its implications are significant. cPlay 2.0b22's stellar performance is unlikely to be improved upon!
Please REMOVE previous versions before installing cPlay 2.0b22. Normal (SSE2), SSSE3 B9 and SSE4 B9 versions are available. Running cPlay on a CPU not supporting the required instruction set will cause cPlay to exit immediately.
For cMP² users, review these BIOS changes (underclocks FSB, CPU & disable additional USB items). New settings are in effect allowing for lower power consumption (below 20W) as well as kernel changes. Recommended Gigabyte mobo GA-G31M-S2L is no longer available, use either GA-G31M-S2C or GA-G31M-ES2L.
Cics:
B22 SSE2 is the first version of cPlay that sounds good on the two AMD - based systems in my home. Previous to this release, I'd thought that Foobar2000 w/SRC and Wavelab Demo 5 sounded much better.
I always thought the frequency balance of earlier versions was tilted toward the dark side with too much bass emphasis. The sound was somewhat lifeless and sterile. Others that had listened to older versions in my systems agreed. I really was hoping that the software would be head and shoulders better than others due to it's intrinsic differences but it was a dissappointment in my systems - until now.
With this build, things are certainly more alive in my system and the top end now has the shimmer and sparkle that seemed to lack in the older cversions. I'm not sure what you've done differently this time around but this version is likely to stay on my hard drive for a while.
Thanks and keep up the great work!
Hi cics and all,
(Purely IMHO) Cics you know me I have been a whole-hearted cplay supporter since the first version. Each time you release a good version I can't stop myself from praising it and announcing it in our local forum. However, recently in our local forum some CAS users have slowly and quietly switched from cplay to Wavelab 5.0 Demo.
It is no doubt that cplay 2b22 has much higher resolution (estimate 30% more) than Wavelab 5.0 demo, however, when I listen to female vocal or violin, Wavelab gives better rhythm or musicality. Weird is that after i listen to Wavelab, cplay sounds much better in details but quite "flat"... I wonder whether cplay 2b22 resolution is "too high"...? I have no clue why this happen.
(I playback native 24/192 files using both programs and so no real-time upsampling required)
anybody use SAWStudio?
try it
Hi appleteapot (and other Wavelab Believers)...
If you want to compare DAW applications to cPlay, be sure to have a listen via Magix Samplitude v10.1. You can get a demo version from the Magix website (linked below). While you're at it, be sure to grab the PDF manual.
MANY Professionals declare the Samplitude playback engine is superior to Steinberg's... in my experience, I'd have to agree. Used as a file player, I've found it to give me a realism on original high-rez recordings that Steinberg software can't match (Wavelab and Cubase/Nuendo).
Happy Configuration !
Cheers,
Grant
That's not a Toy... IT'S A TOOL !!
Thank you very much for recommending Samplitude! I downloaded Samplitude 10 demo and have listened to it for a couple days. The installation of Sampliutde is quite simple, but configuration takes a long time (thanks for recommending the PDF as well). UI of Samplitude is very fancy. As learned before, I have turned off all the fancy animations, buffer, or optimizations that are not required during playback. Eventually I playback only using the File Browser within the "Manager". The ASIO integration of Samplitude is very good as it can display in real time the number of lost ASIO samples, and merged into ASIO control panel's details. Samplitude is more tolerable in stress tests.
The music balance of Samplitude is very good. It has good balance between "musicality" and "resolution". Thus the sound of Samplitude very "real". Speed is neither rush nor layback. Rhythm/Musicality is in between Wavelab and Cplay. I like using Samplitude to playback female vocal.
I use B22 SSSE3b9 version of Cplay. At after 40+ versions, I am not sure if any playback software running on Windows XP can beat cplay in terms of resolution, let alone CMP+Cplay. Cplay has the best resolution among 3 software. Therefore, I like playback symphonic materials and male vocal using cplay.
In contrast, Wavelab 5.0 Demo has strongest musicality and rhythm, but resolution is lowest among the 3. And if Samplitude is rated "real" in musicality, then Wavelab seems "exaggerated".
I am not trying to bash cplay in anyway. Instead, I am just thinking how cplay could do better. My personal opinion is that after some 40+ versions, I think cplay surely excels in resolution. And with minor fine tune I think Cplay is ready for big time to stand up and compare directly with the big players like Samplitude/Saw/Wavelab in professional audio field. But of course it is up to the design objective of cplay :)
Apple,
Vincent Sanders of VRS soutions was one of the first to preach that the player software made a huge difference and that the pro software was some of the best.
It seems that his preference for players has evolved like this:
wavelab-> samplitude-> izotope
YOu may want to give izotope a try. I know I will but I have to agree that cplay is fantastic and I am hoping that the next version will give that musicality that I heard in v20 and the resolution in v21. I am not sure v22 has quite captured both.
I think cics is figuring out what makes one version sound different from the other and ideally once he knows for certain, he would incorporate a setting option that would allow one to choose which type of sound one wanted. Kind of like the filter settings in some cdp/sacd players. I am picturing a warm setting, a resolving setting, and a balanced setting.
AFAIK this would be a first in a player and allow the audiophile to find the sound that bests complements their system.
Or if the setting isnt possible then maybe 3 versions of the player.
The really good news is that I think he is figuring all this out.
Can you confirm which izotope player is he using? It seems to me that izotope (ozone) is not a stand alone player but a bunch of plugin?
Hey Art,
Here is where I downloaded it
Hi ackcheng,
All I know in PC version of Izotope RX Advanced the playback is available. Cheers!
I can't comment on Steinberg's Wavelab 5.0 demo.
Optimal setup for native 24/192 is important:
- Use B22 SSSE3 B9
- Set DSP Buffer to Large (experiment with this especially if no upsampling is done)
- RAM timings is critical. HiRes material works RAM harder whilst CPU load is insignificant (the opposite applies to 44.1k material where RAM is worked less but heavy CPU load is needed for realtime 192k upsampling). Start with Auto timing then try various options (3-3-3-? with ? set to higher numbers).
You will get different results - let us know.
Thank you cics! I just tried using AOBCTized PC, E7200@1.2Ghz, HyperX 2G 3-3-3-9, Lynx AES16e-> Mytek 8x192 -> Beyer DT990.
Your recommendations do help a lot, I am now on b22SSSe9, large buffer, RAM 3-3-3-9. Lynx buffer is 32 (0.16ms). Sound is smoother then before. THe difference could be heard when playback flute sound. I will listen further and let you know...
When I read your source code on 1.0 but I still don't quite understand how it works... my wild guess is that the playback is essentially a loop take a chunk from wave file, perform calculation, feed into ASIO, take a chunk again, perform calculation, feed in ASIO... etc until end of loop and finished reading the wave file.
I wonder whether "smoothness" is related to timing? Is there any timing control involved in feeding in the right data at the right time slot to the sound card? Or does timing all depends on sound card and ASIO, cplay program just process the loop to read the wave file?
Smoothness is related to jitter.
Timing control is indirectly driven from soundcard's XO. Player sees timed interrupts for sound data buffers. Smaller and more frequent buffers are preferred ( more information ).
Hi,
What i would like to mention is that i have the idea that a lack of resolution maybe gives the effect of better musicality. Kind of same experience i have with the coupling cap between dac and amp.
When i remove it there is super resolution, but flat mid's. When i put one in there is more joy in listening, but lacking 3d and resolution.
Are you sure that Wavelab's code is responsible for better rhytm and musicality?
In B22, SSSE3 (and SSE2) ASIO interfaces are completely new. This is the biggest area of change when comparing to B20.
In my setup, I now get better results with SSSE3. Although the core design is the same as SSE4, in SSSE3 a key SSE4.1 instruction is emulated.
Would be interesting to hear your experiences...
I really love SSE4 it is lush in the midrange but upon further listening to SSSE3 it is more precise. While commenting yesterday that it was drier than SSE4 I may have mis-stated the reality, its very extended/precise (best word I can come up with is precise) highs and upper midrange are superb. For guys old enough to remember, SSE4 reminds me of an Audio Research SP6 preamp sound while SSSE3 reminds me of a John Curl preamp sound - better bass/highs/better depth/pinpoint imaging.
So I think after going back / forth I prefer SSSE3 now. The highs on cymbals / harpsichord/ tambourine are very life like. Dense orchestrations are better separated/articulated on SSSE3. SSE4 has some artifacts that are sometimes beguiling and sometimes not. Its upper midrange can sometimes offend.
But I really do appreciate cplay 2.0b 22 SSE4 and may ocasionally go back to it for its lush midrange.
Both have there strengths - mids in SSE4 is better.
Best scenario is to achieve resolution and detail of B22 and warmth of B20.
I have 44.1, 88.2, 96 and 192 material from rips and downloads, as I assume most of us do. I would like to play the first two at 176.4 and the latter two at 192. So currently one has to know which each potential selection is and then, in cMP mode anyway, select something to go to cPlay, change the rate there, close it to go back to cMP to click the selection which then opens cPlay again to play it. It would be a big convenience improvement if one could select the bitrate along with the music selection in cMP or, even better, if cMP or cPlay could read the bitrate and set itself accordingly, to an even multiple of the selection's rate (multiple=2 or 4 up to a limit of 96 or 192). This could be an option in one of the programs along with a choice to bypass dsp (multiple=1).
Edits: 04/24/09
Hi Cics,
Thank you very much for your continuous effort in making such a wonderful gear even better.
May I have a humble request for the new version of cplay: Add a keyboard switch to pause a playing track?
Hi cics, a humble request for the new version of cplay: Is it possible to incorporate a switch to disable autoplay when an album is double-clicked? What I mean is that only the content of the album is shown, so that we can choose the track we want to play instead. It occurs very often that the track we want to play in an album is not the first track, but since cplay autoplays it, we have to wait until cplay finishes loading the second track before we can move to other track.
Anyway, thank for this wonderful work you have done.
Regards,
kalimasada
cics wrote:In my setup, I now get better results with SSSE3. Although the core design is the same as SSE4, in SSSE3 a key SSE4.1 instruction is emulated. Would be interesting to hear your experiences...
In my setup (standard cMP2 box, NOS 36-way DDDAC, Class D amp), the SSE4 version of b22 sounds very good indeed, the best yet.
Last might, I tried the SSSE3 version as requested. Ouch! After less than an hour, enough was enough. It had all the musicality of a dead fish - a shrill, cold, lifeless sound. The difference was very marked.
Reinstalling the SSE4 version restored sound quality, suggesting it wasn't just a gremlin.
I've been using b22, SSE2 version on a desktop computer for several days and it sounds good too so the perceived difference with the SSSE3 version was not expected.
Best
Dave
Edits: 04/23/09
I'm trying SSSE3 right now and my initial impression is that its a bit tighter in the bass,drier in the highs. Don't know if I like it yet but its still very good. I just might like SSE4 better....initially. More later.
That's very intriguing. I'm very happy with SSE4, but I'll do a serious comparison after giving my new mobo a few more days to burn-in.
For those interested, the new MB I'm referring to is the GA-EG45M-UD2H, the only Intel board with integrated graphics in the Ultra Durable 3 family (I erroneously linked the EP45-UD3P in the posts I made a couple days ago).
Initially, the G45M was barely listenable. The SQ was *bad*, and I felt that even my daily use PC sounded better. However, after leaving the system playing in the background for a few hours, I couldn't help but notice that the bass sounded detailed, and extension at the low end bettered my previous GA-G31M-S2L setup. Intrigued, I powered off my amp and decided to give the MB an entire day before making any further judgements.
I'm glad that I gave the UD2H a second chance. The treble and mid-highs still sound a bit harsher than the G31M, but the improvement in bass is easily noticeable. What really excites me, however, is how the new setup seem to "unleash" energy in the music, which makes recordings sound more "live"ly and natural. I expect the SQ to improve after more burn-in. I also intend to work on optimizing the BIOS settings during the weekend.
I strongly encourage the daring (or bored) inmates to give this MB a chance. You won't be disappointed :-)
Abysstw,
Any more to report on the GA-EG45M-UD2H motherboard. How's it sounding now that it's had more time to break in?
Very curious about this board... I like some of the technical thoughts behind it.
Greg in Mississippi
Hi Greg,
After further BIOS optimizations and more time to break in, the harshness in the sound had completely disappeared. To my ears, the EG45M is a significant step up from the G31M. I feel that there's no reason to stick with the GA-G31M-S2L if it's possible for you to get a GA-EG45M-UD2H.
The following are the key BIOS settings(IMO), in case you decide to buy an EG45M in the future:
1) CPU Vcore at 0.75v
2) CPU Termination Voltage at 1.05v
3) CPU Reference Voltage at 0.703.
Thanks for the BIOS settings info.
My GA-EG45M-UD2H came in yesterday. I plan to mount it in my ripping machine, load up XP, perform the optimizations, then burn it in there for awhile. How long did it take to come on song in your setup?
TIA!
Greg in Mississippi
Thanks for having faith!
I think it took at least two to three (full) days of break in before it's listenable. Then again, I performed the BG NX mod on my Pico (thank you for your inspiration) right before I installed the MB, so I'm still not sure whether to blame it on the MB or the BG NXs.
Instead of reloading XP, I simply replicated the system drive on the old HDD (painlessly, thanks to Acronis TrueImage) to my new SSD, and plugged it on the new MB. And it worked! I only had to disable a few stray components through device manager instead of performing the full set of optimizations.
Please see link for my current BIOS settings, the old post has a typo.
P45 option was a non-starter for me as this requires a separate video card.
Using G45 chipset and the excellent choice for setting lower voltages could allow for lower power consumption. Current setup uses 20W or less - would be interesting if you could post measures (44.1-> 192).
The best feature of this mobo is improvements in the ground plane. Ground noise (including bounce) adds to jitter. Micro-ATX form factor also helps.
Thank you for the kind feedback. I'm out of town right now, so measurements will be done this weekend (if I get the chance).
I think that the enhanced power traces and higher quality components* are also worth mentioning.
*As per GIGABYTE website: Lower RDS(on) MOSFETs, Ferrite Core Chokes, andJapanese Lower ESR Solid Capacitors
best highs and upper mids yet. well done! I'm sure I am not optimized for sq, can only get 160 host clock control and .9 core voltage. Have all other new bios/windows settings.
Does a new mobo and cpu, once burned in, allow lower settings? Only have a few hours on cpu but more on mobo.
New version shows just clean sound comparing with 2.0b20 release.
But I feel something lack of noble nuances and its sound balances
even though the 2.0b22 release to be upgraded with higher specs.
On the other hands, the previous version(2.0b20)it was close to
the high-end audiophile with good harmonic balance I thought.
(natural and noble sound when the rate 44.1 and upsampling with
121.33 db)
Anyway, thank you very much for your lavors.
B22 has amazing balance.
Voltage stability is specific to each CPU. Make sure EIST is disabled in BIOS - this would allow for lower VIDs (you should get to 0.85000 at least).
Have you switched to laptop drives yet? Standard 3.5" drives can cause issues you have been experiencing.
I have been running a 2.5 " lap top drive for 42 hours with both os and music files replicated on same. Interesting results:
First much better sq
Second the harsh noise, metallics occurs much less frequently. During 1st day I got metallics after 3 hours, second day I got 18 hours clean listening. When I get the noise on analogue out from juli@ all I do is restart track and it clears up. On digital outs I still have to re-boot. Since I now prefer juli@ analogue it is for all practical purposes a non issue now as long as this type of stability continues. It does seem to get better with time.
I am able to get down to 162 host clock control and .83 vid (on cpuz) and it certainly sounds better the lower I can get it. Hopefully with additional time on new mobo/cpu I'll be able to go lower.
Hopefully I can now get back to improving sonics through passive cpu cooler and additional tweaks to juli@ hw.
Theob,
Glad to hear you have gone to the 2.5" laptop HDDs. As I'd said in our last email exchange, I think that's a key part of your 'metallics' issue.
On the CPU cooler, I have a spare one I can lend you if you'd like to 'try before you buy'. I would recommend that one over the stock Intel cooler without a fan... you don't want to burn up a processor.
On the Juli@, I did some additional comparing of a stock Juli@ to the modified one. The modified one is more musical with a much more detail in the treble region while the trebles are in better balance (slightly lower in level) with the modified card than the stock card. I'm very impressed with the 'analog-ness' of the trebles right now. The midrange natural-ness is very impressive too, especially on voices.
I still haven't had the time to power the Juli@ from separate power supplies, but did make a few steps in that direction this last weekend by getting the supplies mounted in my cMP.
Hopefully this coming weekend will be a bit less crazy.
Greg in Mississippi
Thanks on your offer but I have a thermalright coming (the one recommended by cics).
Congrats on your juli@ mods
no have not tried laptop drives yet. will do now.
with the current cmp specs non-upsampling sounds very nice too. In fact I'm choosing non-upsampling more often. I'd say it varies with the music and my mood, but while upsampling is silky smooth, non- is fresh and "live" sounding.
I bring this up because, if I recall correctly, your distorted sounds only occur at high upsampling. The problem is as you have said many times, a result of the interaction of cPlay and the system. You can go on replacing hardware until you find a combination that is free of distortion at 145/192, or cics can look at cPlay to find a solution, or you could accept things as they are and save a load of time, money and grief, and just maybe end up with the best sound anyway!
I must say that I am a fan of non up sampling. I use a NOS DAC and find that any level of oversampling in cplay kills the dynamics and presence of the sound.
I just upgraded cplay from v18 to 22. Initially although i found a much sharper defined sound there was somehow too much treble and it seemed that while sharper sounds it somehow lost the natural life and became mechanical. I switched the SN from 148 back to the lower setting and to my ears have regained the natural harmonics that I loved in v18. Is this possible or am I imagining things. Most interested in other opinions...
Regardless, I must say thanks to CICS for this revolutionary system of audiophile sound on a budget.
Thanks for sharing your impressions. Unlike you, I have preferred cPlay with upsampling from its beginnings, but recently the non-upsampling setting has caught my attention. Subjectively (no time for careful a-b analysis lately), disabling EIST and getting clock speed lower and tRAS down to 6 (5 was no improvement for me), and RAM down to 256MB, were the big improvements for me, and non-upsampling playback has begun to sound better than ever. The voltage-reducing exercise has produced a much smaller increment of improvement, IMO. Time, and much more listening, will tell - maybe my tastes or my ears have changed!
I often thought of the solution you mention and I may end up there but even though I have bought a double of many of my components I still spend a heck of a lot less now versus when I was into sacd players, turntables etc. but this whole thing is a sheer quest for me. I really want to solve this problem. I love 192/146 cmp^2. I tell my wife that it keeps me off the streets and out of trouble. And with all the doubles I can soon build another computer for her.
Hey its a hobby, its a past time. Its supposed to be fun. Building, problem solving has always been fun for me (I admit a little frustrating for me this time).
Yeah, you want to beat this thing - more power to you!
But my point about the non-oversampling sound quality matching the oversampling with the latest setup was serious, and I wonder what you and other inmates hear in this regard.
ok good point. I will try and report back here later.
tried 44khz while running juli@ analogue out. It sure is closer to 192 than I remembered. This morning I liked it better @ 44 but last night I liked it better @ 192 but very close. Thought that 192 was more airy-er last night and this am it sounds as if 44 has more weight/solidity. So I could go either way.
T,
So I could go either way.
Just a bit too personal for this forum :)
Refer to ReadMe.txt for important details & specifications.Change Log (2.0b21 release):
- ASIO output refinement:
- Small refinement for SSE4
- SSE2 and SSSE3 versions revised to be closer to SSE4
- Minor changes (ASIO interface and test output functions)
- Minor fix affecting downsampling (error was given)
- Compiler optimisation done
Please REMOVE previous versions before installing cPlay 2.0b21. Normal (SSE2), SSSE3 B9 and SSE4 B9 versions are available. Running cPlay on a CPU not supporting the required instruction set will cause cPlay to exit immediately.
For cMP² users, review these BIOS changes (underclocks FSB, CPU & disable additional USB items). New settings are in effect allowing for lower power consumption (below 20W) as well as kernel changes. Recommended Gigabyte mobo GA-G31M-S2L is no longer available, use either GA-G31M-S2C or GA-G31M-ES2L.
Edits: 04/15/09
Hello, Cics, hello, everybody. My impression of b21 is mixed. It is slightly more detailed than 20 but on my orthodox Biostar setup I got highs that are too straightforward. I wouldn't call this version agggresssive. Just plain hi-fyish. Some good things happened to the "backest" plane, it is much more pronounced, no more channel deficiences, when one channell contains predominantly back-side information and is muffled, while the other is in-your-face. Overall LF region is much more informative. But what I was upset with is the planar shift of some sounds.
Example On the first track of Porcupine Tree - Deadwing I am used to hear the sounds of a tram slightly offcenter and to the far back. b21 brings it suddenly dead center and in your hands, literally forward. The effect is not small and reminds the manipulations with phase that are used by studio engineers to re-place the instruments.
Another example is track 13 Bends Like a Willow by Jethro tull J-Tull dot com. Here we have fantastic interplay of cymbals and the drummer is the main figure of the show, b21 completely de-Structs this shiny piece. B21 sounds grainy only on Robert Plant & TSS - Mighty Rearranger, but so does b22. I am yet to hear ANY cd- or memory player that can cope with this album. B22 sounds closer to 20, it also has fine back plane but not as prominent as b21, probably because those "phase-shift" manipulations are not present and details and instruments returned to their places.
These are my observations, but probably it happened because Biostar mobo and the then recommended processor do not support sse4... Also starting with b20 I am getting dropouts every 7-8 minutes. On b22 dropouts occur 2-3 times per std album (45 minutes) but are more like hiccups (if I spell it correctly). But b22 really sounds very well. It is not edgy or grainy, has right "smell" of the sounds. And what really delights me is the back plane.
Always thankfull and attentive, Serge.
ALSO, as my music is 99,9999999% 16-44, I would be glad if an option to load the whole album into memory would return
Edits: 04/17/09
Since I just returned to my DC place with a different system I can't make an accurate comparison of b20 and b21, but you surely have still never failed us. This version sounds terrific!
Edits: 04/15/09
Started on b22 which should take strengths of both b20 & b21... this would be perfect.
More feedback on b21 is needed for b22's development.
I am now listening once again to juli@ analogue out(got rid of more devices per ryelands to reduce tics/dropouts) and just compared 20 to 21. Yes 20 seems less edgy in mids/upper mids. Not a lot but its there and I prefer to listen to 20. But there are other aspects of 21 I like but I can't pinpoint it yet.
Yeah, 21 has more resolution but 20 is warmer and easier to listen to on my rig.
Sadly now both are hard to listen to. 20 because I know I am missing resolution and 21 because I am loosing some warmth and musicality.
22 please hurry...
.
s
Hi! Thank you for making this wonderful software! I never thought there can be improvement playback through memory and cPlay really surprised me with its supreme playback quality.
I just have one question:
Is it possible to add support of UTF-8 decoding for CUE?
Because in non-Latin speaking countries such like Japan and China, most track titles are coded by non ASCII code. For example if I want to read Japanese cue on a Chinese system, saving a CUE file using UTF-8 is the best way to keep the original information and avoid possible file path issue.
Does it work with XP ?
Any benefit over 1.04 ?
thanks
Martin
Thanks guys.
I think I'll stick to 1.04 for now.
Installed it in second system - works fine with XP. Don't notice a difference, but have not a-b'ed with 1.04 on main system yet. It is even more of a pain to uninstall than 1.04. I helps if you remove its folder from the directory from which it was first installed when you need to uninstall and reinstall a different version.
I tried with my new windows reload. Seems ok not sure I can specify differences with 1.04
tried 1.05. it works with XP. Till now no differences.
See cPlay release update post for details.
Subtle but important changes are made lowering CPU VID (applicable to systems not able to reach lowest VID) and CPU Frequency (now down to 900MHz). This is achieved by disabling EIST. Additional devices are also disabled.
Thanks cics for updating this!
Just once more for clarification: If EIST works with the optimized settings for CPU VID etc (as in my case), isn't it better to go with EIST enabled to handle L2 cache and thus reducing power consumption? Sure if it doesn't work (as I think was true in your case as you reported) it should be disabled because the benefits from lower VID etc. outweigh those of L2 handling, correct?
Thank you,
Robert
.
It is possible to use mp3toys external player feature with cplay. Mp3toys has a pretty good playlist capability and you could use it as the gui, build the playlists and let it send them to cplay.
This of course is for non critical listening (for cmp2 users)
There are a few caveats though.
1. The way cplay is designed it only works with mp3toys when cplay is not running. If it is running and mp3toys sends it a song, you will get an error from cplay about only one instance allowed....BUT if you set cplay to EXIT after it is finished, it will work great. SO if you are the type of person who changes songs midstream then you will get this error. Just exit cplay and select the song again. I am envisioning this for the inmate who wants to put on some music while "doing chores"
2. Mp3 toys doesn't handle cue sheets, but it does .wavs and flac. So if you have ripped to cues and one big file, you will get it to do a playlist of ALBUMS. If you did individual files, then you can do playlists of individual SONGS or ALBUMS.
3. I have no idea if Mp3toys works in on a cmp2 rig. I dont care to put this on my audio rig, but for those who would really enjoy playlists or randoming amongst their albums with a "set it and forget it mentality" this might be a nice thing...if it works. Now if you arent doing cmp but like the sound of cplay but not its interface, then give this a shot.
4. It is not a great option for Lynx cards or any card that has single client ASIO drivers. BOTH players will "grab" the card even though cplay is the only one sending information. But cplay will report that the asio driver is unavailable. This is how my lynx works and the only way around this is to turn on the onboard audio, or try to set something like play 7 as the main soundcard. I did this successfully years ago but dont know with the new drivers. It probably works and it does work on my M-audio card. Dont know about the juli@.
The trick is to tell cplay to exit when it is finished and to set the delay in mp3toys to wait long enough for cplay to finish and exit before it sends the song to cplay.
Here are some shots and some settings:
I just wanted to share that I'm playing music with cMP/cPlay with SRC at 145db - 24/192, feeding via USB an E-MU 0404 USB
it is configured as a dedicated partition in a laptop, multiboot, shared with my everyday installation so only software tweaking
I can only say that the sound is very good, but I don't have a good reference to compare it to, so anyone in Lima-Peru interested in an audition just let me know
LAPTOP
Dell Latitude E4300
Processor: Intel Mobile Core 2 Duo SP9400 @ 2.4GHz, 1066 FSB, 6MB L2 cache
Memory: 4GB DDR3 (2 x 2GB PC3-8500F (533MHz)
cPlay DIAGNOSTICS
cPlay 2.0b20 (SSE4 B9)
Name (ASIO E-MU 0404 | USB)
Channels (inputs: 2, outputs: 2)
Buffer details (min: 384, max: 9600, preferred 384, granularity 192)
Sample rate (192000)
Latencies (input: 613, output 1189)
DSP Buffer (18432 samples of 96.0ms/10.4hz/500.0hz)
WAV file size = 560 MB
Rate = 16/14100 (2 channels)
samples = 146860056 (55:30)
AWE successful
RAM allocation (317 MB) successful
SRC (145db, 192,000.0)
ISSUES
USB and eSata external HD can not work together (cPlay has no sound), now using external HD via Firewire
Little fan from laptop is very noisy, probably disaligned, needs service
How are you possibly getting 24/192 playiback via usb? Are you sure the external card is not downsampling to 24/96 or lower?
look for posts from Christine Tam and Gordon Rankin about EMU 0404 USB and 24/192 capabilities for technical detailshttp://www.audioasylum.com/forums/pcaudio/messages/43607.html
http://db.audioasylum.com/cgi/m.mpl?forum=pcaudio&n=41815&highlight=marvel+gordon+rankin&r=
it seems that EMU 0202, PRE and 0404 USB area capable as they share the same custom drivers
I just got this powerful enough laptop to run cPlay on 145db 24/192 and did the test last night, guess cPlay diagnostics confirm it
edited to add links
Edits: 03/19/09
He gives a nice plug for cplay/cmp sourced from Sound Forge (come on Bill its sourced from Cics, just say it). But then he says
'....While experimenting I've also found that using the motherboard's own SPDIF output to my pre-pro for decoding, rather than my Juli or M-Audio soundcards, there seems to be a slight improvement in low volume information. This leads to an improvement in ambiance recovery, which is the opposite of what normally should occur with a top-grade soundcard. Possibly the ability of the two programs to load the digits to RAM rather than reading them directly from the spinning hard drive actually removes the need for a soundcard. Try it and let me know what happens in your system...'
Is he talking about the gigabyte audio system? I guess we can all try by going into bios and re-enabling.
But has anybody here tried this?
While results are slightly surprising, they're definitely believable.
Looks like the only options left are RME, Lynx, and onboard audio devices!
I don't think so. Xonar D2X is a very good basic for mods: SPDIF fiber optic out (galvanic isolation), separate good (!) power supply and so on ... and the card is working :-)
Thomas
I'm not claiming anything beyond that.
Juli@ I2S/BNC is more than good enough, while Xonar coax doesn't cut it at all.
and pick ONE moniker and stick with it.
It's a rule here.
Thanks for playing along, Chris
Even that is not necessary - clueless statements are perfect giveaway, CHIPZAHOY.
Unless there's another moron like you in Taiwan, that is...
Who apparently believes their clueless statements should be taken seriously, while others are not entitled to their own opinions. Your comments will be ignored from now on.
Hello, theob! I have already written about the mobo's native soundcard in conjunction with Cmp-cPlay. I use the first recommended Biostar mb and it has Realtech "hi definition audio" on board. Now that I cannot afford to get the Lynx, especially as I want Lynx and Lynx Aurora 8, I have to make do with what I have. I used Creative Audigy 2 platinum something (24-96 capable) for some time but after minlogon crash I started using onboard Realtech. I can say that Realtech eats Audigy for breakfast. It is much more transparent and airy in highs and no glaring, no crackles during playback, but it is really strong in the lows. They are very well defined, clear and really Low even on smallish speakers. You may want to read my humble comment on 2-18 o 2-17 where I say that the air appears, but probably due to the fact that lows are more clear. So it goes.
Serge.
thanks Serge I think I'll eventually try but only after I solve my hard metallic overlay issue (see my other post today under cmp).
Oh, theob! I was hoping You have already settled this sad handycap in playback. I am by no means computer guru, but I am afraid You will have to eventually reinstall windows. I know what a "pinass" it is, but I usually grip my courage and perform this soft surgery immediately.
Serge.
I almost forgot about that I will now do this---its cheaper than a new cpu.
I have just finished going thru the steps to setup minlogon (disabling WFP option) - a silly question - do I have to re-enable WFP or is that part of the minlogon setup?
I have yet to reboot - I await a response!!
No - if you re-enable WFP, it may well crash on re-boot. At least it did when I put it back on. Meanwhile, I've run with it disabled for several months on three different systems.
While I've yet to have a problem implementing the change using the procedure described by seger, I'd urge that you do NOT do it until you have a (tested) image file of your system partition.
Best
Dave
Dave,
Thanks for the advice and warning - I have been trying to find some software that I can load from usb memory stick to do backups and restores (I do not have a cdrom nor access to the web on this pc) - a right pain!
I will first back out my changes and take an image.
Alan
I have left WFP disabled for awhile too. Not a problem yet.
After failing miserably to get any form of backup I decided to go for it and implement minlogon as per seger's instructions. Worked first time no problems - in fact so far all my other problems (clicking / stuttering) have now disappeared.
I am using the SSD purely for OS and the HDD for storing the music - works fine - so I will not be trying to copy files from one to other - just play from the HDD.
So once the the switchers swapped out that should be that.
Alan
fantastic, congratulations!
This is the strangest thing I have had happen in PC audio.
I am in the middle of a cmp2 build, and so far have done the bios tweeks up to disabling services. I think I still have autoruns and the kernel tweaks to do. Maybe it will change.
But things are weird to be sure. Bigger buffers are needed than without the optimizations, stutters at 44.1 and 88 but 176 works fine!
I checked the speakers and there is nothing wrong with them. The lynx mixer is set right but the center fill is gone.
This isnt exactly the correct hardware as I am using what I have around but I have never had this problem before.
Cplay had always produced a smaller more defined image with more space between the instruments and my speakers are probably as far apart as they can be, but this is a bit crazy.
Any thoughts on what is happening here?
This device is likely to cause the problem.
d
d
How can upsampling be bypassed at all?
How can upsampling be bypassed at all?
In the bottom LH corner of the cPlay window, there's a little green icon. Click on it to access the cPlay settings dialogue and select the sampling rate that matches that of the target file (typically 44.1). As yet, this has to be reset manually if you change from e.g. CDs to hi-res files.
During play, cPlay displays the number and sampling rate of the current track and the output rate at the top left of the window thus: [3 44> 44] (for track 3, 44.1KHz, NOS).
HTH
Dave
Are there perhaps any plans of implementing the bypassing of upsampling in a new release? Like the previous posters i find it quite a hassle to match the sample rates all the time. Other than that i'm good with it ;)
Yes I'm aware of this request - it's not something that's easily done given how cMP works with cPlay and the underlying kernel.
Just curios, why is SRC (@145db SNR) avoided? Here's some tests on cPlay and SRC:
Quality of upsampling (aka reconstruction filter) in DAC (DSP or chip) is inferior.
Couple of cases in point would be offline upsamplers like iZotope and Adobe Audition, and highly advanced upsampling DACs - like, for instance, Accustic Arts DAC1 MK4 with its 66bit/1536kHz algorithm.
Yes I saw... But without discussing again the pros and contras on upsampling: Would be nice to have the possibility to (really) bypass every kind of DSP :-)
Only to hear for a while the one and the other way ...
If you set cPlay to the same rate as the source there will be no upsampling or downsampling.
I saw... but so I have always go to settings when I take a different file. I use 44,1, 88,2 and 192kHz files... Not very comfortable :-)
Yes, I want to have bypass button as well. I never use upsampling. Changing settings to play different sampling rate audio files is not convenient.
Dear fellow inmates,
I know that there has been word about getting i2s from Juli@ a looooong time ago (the beginnings of this discussion seem to be already in neverland...) - I am aiming to get i2s from my Lynx AES16, somehow...
Anybody have any idea, hint, experience even or seen anywhere any bit of information on this...? Anything is highly appreciated - you see how desperately I want that... ;-)
Many thanks!
Robert
Refer to ReadMe.txt for important details & specifications.
Change Log (2.0b20 release):
- ASIO output refinement:
- Bypass inefficient CPU instruction
- Advanced interface for SSE2 and SSSE3 (SSE4 uses new instruction) for maximum efficiency
- Added "MINIMISE" option for INI file. When added to INI file, playback occurs with minimised window.
- Minor fix affecting diagnostics
- Reinstate function to ensure only one instance for playback
- Upgrade to latest compiler (additional optimisation done)
- Added SSE4 B9 build option allowing for maximum performance with latest Intel CPUs. Explicit SSE4.1 instructions are used
- SSSE3 option removed (advanced versions uses B9 optimisation)
Please REMOVE previous versions before installing cPlay 2.0b20. Normal (SSE2), SSSE3 B9 and SSE4 B9 versions are available. Running cPlay on a CPU not supporting the required instruction set will cause cPlay to exit immediately.
For cMP² users, review these BIOS changes (underclocks FSB, CPU & disable additional USB items). Recommended Gigabyte mobo GA-G31M-S2L may not be available, new suggested motherboard is GA-G31M-S2C.
I am using a P5B VM mobo E6300 processor no problems 96/145 ssse3b9 not compatible with sse4. Will a E7xxxxx or E8xxxxx processor upgrade resolve this issue. T.
No worries!
.
nt
No worries!
.
First, thanks cics for an amazing player!
I'm using cplay on a quite low-end setup: thinkpad t500 (xp pro sp3) w P8400 cpu & ssd (transcend 128 gb sata, but in compatibility mode) & audigy 2zs (usually with koss porta pro)
so far i was using 2.0b19 SSSE3 with great results ,
tried 2.0b20 today
2.0b20 SSE4 works fine (145.68 db & 96Khz upsample), but the treble emphasis is quite prominent and not pleasant compared to 2.0b19 SSSE3 i'm inclined to attribute it to the audigy2zs & portas (for the record: hendrix & fela kuti)
SSSE3b9 crashes at playback start. tried various settings, didn't help
is my h/w incompatible with 2.0b20 SSSE3b9 ?
also, repeated installing and uninstalling of different versions can not hamper performance ?
Try SSSE3 version with ASIO4ALL instead of Creatives' Audigy drivers.
Also, can you test SSE2 version on your setup. This will give me useful insight into the crash you have with SSSE3.
SSE2 same results
cPlay 2.0b20 (SSE2), licensed under GNU GPL
ASIO Technology by Steinberg Media Technologies GmbH
ASIO is a trademark and software of Steinberg Media Technologies GmbH
SRC (libsamplerate-0.1.5) licensed under GNU GPL
(see 'note *1' in specifications located in readme.txt for code changes done)
ASIO Driver Initializing...
Name (SB Audigy 2 ASIO 24/96 [FFC0])
Player ASIO version (2)
Driver ASIO version (2)
Message (No ASIO Driver Error)
Channels (inputs: 2, outputs: 8)
Buffer details (min: 192, max: 32768, preferred: 4800, granularity: 8)
Sample rate ( 96000.0)
Output Ready? Supported
Preparing buffers... successful
Latencies (input: 4900, output: 4800)
Hardware supports additional optimization - enabled
ASIO Driver initialized.
DSP starting...
DSP Buffer (57600 samples or 600.0ms/ 1.7hz/ 20.0hz) allocation successful
ASIO Driver starting...
Input 1 [ Line-In 1/Mic L ] Active (0) type (32bits)
Input 2 [ Line-In 1/Mic R ] Active (0) type (32bits)
Output 1 [ Front L/R ] Active (1) type (32bits) > Left Channel
Output 2 [ Front L/R ] Active (1) type (32bits) > Right Channel
Output 3 [ Rear L/R ] Active (0) type (32bits)
Output 4 [ Rear L/R ] Active (0) type (32bits)
Output 5 [ Front C/Sub ] Active (0) type (32bits)
Output 6 [ Front C/Sub ] Active (0) type (32bits)
Output 7 [ Side L/R or RC ] Active (0) type (32bits)
Output 8 [ Side L/R or RC ] Active (0) type (32bits)
ASIO started successfully.
AUDIO source [ F:\~FLAC~\bd1965-06-15 - Highway 61 Revisited Again\Track08.flac ]
Genre = Unknown
Artist = Track08
Album = flac
FILE: F:\~FLAC~\bd1965-06-15 - Highway 61 Revisited Again\Track08.FLAC
FLAC file size = 27MB (decoded during RAM load, RAM for 'Wav data' is allocated)
Wav data size = 42MB
Rate = 16/44100 (2 channels)
Samples = 11229036 ( 4:15)
RAM allocation (88MB) successful
DSP Started.
Play starts after RAM load (available 1467MB, system 1975MB).
Volume (-30.0db, phase 0)
SRC (145db SNR, 96000.0)
Both SSE2 and SSSE3 use the same ASIO interface (SSE4 is different). Have you tested with ASIO4ALL?
i didn't try asio4all yet (unfort. i can get back to it only in a few days)
tried all available asio settings ('creative asio', 'sb audigy', 'sb audigy 24/96'), only the last manages to initialize, and then crashes:
cPlay 2.0b20 (SSSE3 B9), licensed under GNU GPL
ASIO Technology by Steinberg Media Technologies GmbH
ASIO is a trademark and software of Steinberg Media Technologies GmbH
SRC (libsamplerate-0.1.5) licensed under GNU GPL
(see 'note *1' in specifications located in readme.txt for code changes done)
ASIO Driver Initializing...
Name (SB Audigy 2 ASIO 24/96 [FFC0])
Player ASIO version (2)
Driver ASIO version (2)
Message (No ASIO Driver Error)
Channels (inputs: 2, outputs: 8)
Buffer details (min: 96, max: 32768, preferred: 2400, granularity: 8)
Sample rate ( 48000.0)
being set to 96000.0
New sample rate ( 96000.0)
Output Ready? Supported
Preparing buffers... successful
Latencies (input: 2500, output: 2400)
Hardware supports additional optimization - enabled
ASIO Driver initialized.
DSP starting...
DSP Buffer (28800 samples or 300.0ms/ 3.3hz/ 40.0hz) allocation successful
ASIO Driver starting...
Input 1 [ Line-In 1/Mic L ] Active (0) type (32bits)
Input 2 [ Line-In 1/Mic R ] Active (0) type (32bits)
Output 1 [ Front L/R ] Active (1) type (32bits) > Left Channel
Output 2 [ Front L/R ] Active (1) type (32bits) > Right Channel
Output 3 [ Rear L/R ] Active (0) type (32bits)
Output 4 [ Rear L/R ] Active (0) type (32bits)
Output 5 [ Front C/Sub ] Active (0) type (32bits)
Output 6 [ Front C/Sub ] Active (0) type (32bits)
Output 7 [ Side L/R or RC ] Active (0) type (32bits)
Output 8 [ Side L/R or RC ] Active (0) type (32bits)
ASIO started successfully.
AUDIO source [ F:\~FLAC~\bd1965-06-15 - Highway 61 Revisited Again\Track11.flac ]
Genre = Unknown
Artist = Track11
Album = flac
FILE: F:\~FLAC~\bd1965-06-15 - Highway 61 Revisited Again\Track11.FLAC
FLAC file size = 77MB (decoded during RAM load, RAM for 'Wav data' is allocated)
Wav data size = 120MB
Rate = 16/44100 (2 channels)
Samples = 31669680 ( 11:58)
RAM allocation (244MB) successful
DSP Started.
Play starts after RAM load (available 1578MB, system 1975MB).
Volume (-30.0db, phase 0)
SRC (145db SNR, 96000.0)
gonna try sse2 now
I am running 2b20 SSSE3 on a Thinkpad t500 in xp pro sp3 as fully adjusted as possible in cMP mode to cics recommendations with no problem.
SSE4 also works fine. As I have reported I find the balance of treble and bass different between these, but the SSE4 treble seems to me beautiful but just a wee bit too much on my system and the bass a bit soft--nothing I would have noted if I had not compared it to the SSSE3 version. Saying that the asio implementation is different between these hardly seems to explain how this can effect different parts of the frequency spectrum differently.
.
I am away from the setup with Lynx card since late Dec, so have not yet tried it there with the latest versions.
thanks, so i'll try it again, I didn't try cMP so far (i am quite satisfied with the switch to ssd so far)
what soundcard/dac r u using?
i am thinking that the different asio implementation might be responsible for 2.0b20 SSSE3 not playing
thanks, so i'll try it again, I didn't try cMP so far (i am quite satisfied with the switch to ssd so far)
what soundcard/dac r u using?
i am thinking that the different asio implementation might be responsible for 2.0b20 SSSE3 not playing
You seem not to have a user account on record. I discovered this clicking your screenname seeking to email this to you.
I use Asio4all usb out to EA offramp turbo 2 to Accustic Arts DAC 1mk4.
Hey C,
I couldnt find that motherboard at New egg. Will this one work?
Give them a try
d
.
I cannot get a cuesheet with 102 tracks to work. I get "medea" at the bottom and a freeze. You raised the limit to 120 recently. Is it lowered in this latest version?
.
Thank you Cics !! You say very humble ‘slightly better’. But I agree with Gagnem “– BY FAR – the best version”
In my setup cPlay 2.0B20 SSE4 b9 now definitely stands of the crowd! Previous cPlay versions sounded very good in my setup. But I also could live very happily with the build-inn player in Steinberg’s Wavelab Lite and also with Beatport SYNC on more modern music. But with cPlay 2.0B20 SSE4 B9 things really start going. Whow what a player !!
Now there’s real “Depth, hall ambiance, imaging & dynamics “ like Theob already wrote. And also “There's a solid feel to all the music that also honours the nuances, the small sounds as well. Everything seems more together”. I couldn’t say better.
I don’t think the SSE4 version treble region sounds sharp or emphasized. The combined titanium/fabric dome tweeter in my K+H O300 speakers rapidly tend so sound sharp with poor sources and bad mixes. But with the SSE4 version they sound very natural and luminous
A very, very fine player. Nothing comes close anymore in my setup. Thank you cics for such a wonderful product.
ESI juli@ or LynxL22 dig i/o -> Lavry Black DA10 -> Mogami Gold -> Klein & Hummel O300
I said that the treble area sounded prominent in 2b20 SSE4 only RELATIVE to SSSE3! You could just as well say the bass in the latter seems prominent compared to the former. Have you tried the latter and compared? I'd be interested in what you find.
Very, very nice, Mr. cics. I was pretty darned happy after the PSU upgrades I did this weekend... and then this gift. The treble region especially is luminous, very detailed, but not emphasized and very natural sounding. Micro dynamics jump out and the naturalness of the sound sometimes has me think my wife is making some noise near the stereo setup.
Thank you, very very much.
Greg in Mississippi
Thank you cics for adding the minimized feature!
And thank you for this great version. Whenever you release a great version I can't stop myself from listening my favorite songs all over again until 2AM. You've done it again :)
Hi cics:
I want ask a question.
The cPLAY run under x64 OS is better than x86 OS?
There are other features available but haven't seen anything that would benefit audio.
Hi cics:
When use 145 SNR,upsample to 176.4,do you know the integer operation and floating point,
which is most?
I see the x64 vs x32,x64 floating point performance is much better than x32.
Hi cics:
Recently I read some article about x64 vs x32
(http://www.start64.com/index.php?option=com_content&task=view&id=2140&Itemid=104)
to summary up,there is two point:
1.addressing to use more than 4GB memory.As you mention it,it's no use is our HTPC case.
2.More efficiency arithmetic.I don't know this will improve.Becase when ise SNR 145 or upsample,I think it will need a lot of operation.
So if on 64x OS,will these operation will run faster?
If the benchmark used pure x86 instructions then yes 64 bit version will offer much better floating point math performance. This is a result of the x86 CPU architecture using the FPU (floating point unit, 80 bits) to perform all fp math (stack based and inefficient).
cPlay & SRC uses fp math. Difference is cPlay (which also applies to SRC) does NOT use this FPU. All fp math is done via XMM registers (SSE2 instructions which extends x86 architecture), i.e. same as 64 bit OS. Hence performance is same.
Hi cics:
Thanks for your reply,so I don't need x64 OS!
Cics: I have a question if one has SSE4 capability and SSSE3 capability is it correct to assume sse4 is better?
Both are exceptionally good and have different implementations. SSSE3 has tighter lows, whilst SSE4 offers slightly better resolution in mids and highs. You need to test these to see which one performs best. In my setup I prefer SSE4 but can gladly live with SSSE3. This may be different in other setups.
I find the different sound of these two implementations fascinating. I hear what I take to be the same difference cics and Theo describe. SSSE3 has a tighter/crisper bass which seems more prominent compared to the crisper more prominent treble in SSE4. How can this--an experienced difference in frequency response and tonal quality--result from what is different in the software playing it when both are so accurate and beautiful? If one were softer or clearer than the other throughout, it would be more understandable. But to be different in different parts of the frequency spectrum needs explaining. Can anyone?
I myself on my setup like the balance of the SSSE3 better, in contrast to Theo. I find the treble harder/sharper and too pronounced, however beautiful, in the SSE4. Cics calls this being clearer. This seems to be a matter of taste and equipment but not perception. We all hear the difference. My equipment: Ridge Street Audio's usb to offramp turbo 2 aes to Accustic Arts dac1mk4 to Gamut's top preamp and amp and L7 speakers. I find European equipment tends to have the same qualities that cics and theo like in the SSE4 version, so perhaps combining them is too much in my case. But the question remains: I can see how speakers might make this difference, but how does one cPlay implementation versus another do it?
Whilst both versions are efficient, their ASIO implementations are very different at the physical level (CPU instructions). Yes, the exact same bits are sent to the soundcard but the manner in doing so is different.
This results in slightly different (and very low) jitter distortion at the DAC's output which is audible. CPU instructions affect power consumption by way of complexity (not just CPU cycles but also internal components) and timing (what is done when).
Yes, but how is it different for different parts of the frequency spectrum?
It seems that 2.0b20 sse4 is a bit more efficient (cpu load in 35-45% range) than 2.0b19 ssse3. Also are both versions essentially the same dsp instructions with sse4 just more efficient? Or said another way are their further optimizations for sse4 to come?
Anyway great release!!
...I think I like the balance of sse4 better but both are fantastic and both run about same cpu usage!
Edits: 03/03/09 03/03/09
Is there anything that I need to do to run SSE4 as opposed to SSSE3?
run cpuz to see if you have sse4 capability. if yes just d/l and run. that's what I did and it worked. I believe the intel e7200 cpu is sse4 capable.
Thanks theob - will give it a try tonight.
For 24/96 output, this is -BY FAR- the best version ever.
Simply outstanding !
Thanks again Mr. cics
A sonic wonder, best yet. Didn't think 19 could be improved upon but it has been now. Highs are exquisite. Depth, hall ambiance,imaging & dynamics are all improved. There's a solid feel to all the music that also honors the nuances, the small sounds as well. Everything seems more together. All in all very very well done Mr. Cics!
I've been playing 192/146 off of hd and no metallics yet after 1.5 hours(but I won't say for sure until after several days playing).
As a sculptor, I like to think of it as you step by step chipping away the jitter and whatever from the perfection of the music that emerges more and more.
Edits: 03/02/09
Dear cics (and others),like a few others I am completely SSD in my cMP/cplay system - I run the OS and cMP/cPlay from a 30GB OCZ Solid Series SATA II 2.5" SSD.
My question now: Since this drive offers both an eSATA and an USB 2.0 port - what would you recommend?
Version 1:
eSATA cable to SATA II port of recommended Gigabyte GA-G31M-S2L (with external power supply from Granite PS to the SATA power connector of the SSD), orVersion 2:
USB 2.0 cable instead to Gigabyte's USB header (with external power supply from Granite PS either also to the SATA power supply of the SSD as above or with recommended USB mod to connect USB 5V line and ground line to external Granite PS)?Can't really figure out which would be better, sonically and technically, causing less traffic and interference on mobo and so on - if there is a difference at all... The beauty would be to have a even "purer" design with USB only and no SATA at all in version 2 :-)
Thank you,
Robert
Edits: 03/02/09 03/02/09 03/02/09 03/02/09 03/02/09 03/02/09
This is more so when OS resides on it.
...my first impression: The sound has more body, more 3D less thin, mre natural. It is subtile, but noticable,wont go back even though SSD is slower in transferring into memory (V1B19)
Maybe this is a pipedream, but I wonder if the hot ticket might be a setup where the operating system is on the SSD and the music library on a standard HDD. Then when you want to play a recording, a power management(?) facility is used to spin the HDD up for load to the SSD (and it is loaded to memory from there for playback) with the HDD powering down or at least spinning down (in a power save mode?) after the HDD to SSD load).
Not sure if that's easily doable tho.
Greg in Mississippi
Thank you for sharing! Do you imply that the music files are now stored on the SSD as well? Or is the SSD only used for your OS?
I know that SSDs seem to be well accepted by computer audiophiles, but I feel a bit hesitant about switching over, unless the SSD offers a noticeable improvement over a well-treated HDD. My current HDD is isolated, while also externally mounted and powered, so - I believe - the vibration, heat, and power draw issues normally associated with hard drives seem minimal in my system.
I use it to store the files as well.
Sounds like you made a pretty substantial investment there. May I ask a couple more questions?
Did you give your HDD any "special treatment" (such as external power/mounting)? What about the SSD that you're currently using?
I'd love to make the switch to SSD, but I want to make sure that it's worth all the hassle.
I have not yet tried an SSD but I have tried Flash memory - I was not convinced. I have ordered a 30GB OCZ SSD which will have to do for the moment - however having 250GB of music it means I will have to load the music for an evening (or two) onto the SSD and then set that up as the music folder.
So what would be needed in this case is for someone (cics???) to modify cMP so that when an album is selected it is transferred to SSD automatically - until of course SSD becomes much cheaper.
One thing to note with SSD is that there are two versions MLC(multi-layer) and SLC(single-layer) - the SLC is supposed to be much faster than the MLC - and of course it is more expensive! As far as the speeds go both MLC and SLC are faster than the 2.5 inch HDDs.
Curiously looking at the figures for OCZ products the speeds seem to be very close with the MLC apparently faster!
Since the one I am buying does not state SLC then it will be MLC. I will post my findings when I have this setup.
I had that very same thought about a future cplay 20+ version.
Let me know how the ssd works out
Interesting idea, but wouldn't the frequent transfer of large music files cause MLC SSDs to fail even faster? I plan on using the SSD (when I get one) for the OS only, but I might use the remaining space to hold a few favorite albums. I *think* that a dedicated, non-spinning OS drive may lead to noticeable sonic improvement, but that's something for us to find out.
Dear cics,
since you are using the Lynx AES16 card yourself:
I am about to install my Lynx AES16 card to my cMP/cPlay system. There are two versions of the newest driver (2.0 rev 16) available - version 1 as an "old" NT4 version and version 2 as a "new" WDM version.
Which one would you recommend? I am a bit under the impression that "old is good", i.e. NT versions maybe not causing as much overhead and such, but it's a feeling without knowing anything in particular... Which version do you prefer, and why...? ;-)
Many thanks
Robert
Are you referring to LynxV264 vs LynxWDM?
My understanding is that they are for different processors (AMD or Intel). For Intel, LynxWDM is used.
Dear cics,
hm, I was reading on Lynx' "how to" page on new drivers (http://www.lynxstudio.com/support_faq_result.asp?q=148) about the two versions of the driver - in fact, on their download page is only the WDM version. So on their website by going downloads -> Select Windows + AES16 + drivers, you only get to choose L2WDM016.exe which includes support for WDM (32 & 64-bit), ASIO (32 & 64-bit).This Version 2.0 Build 016 (June 27, 2008) is WDM only, so that's the right one.
Thanks for helping me to see clear, finally... ;-)
Robert
Hi,
A while back I wrote a program that will recursively work through a music library and generate cue sheets for any directory found that has wav or flac files in it. I wrote the application so that I could more easily use CPlay. The application is only useful if you rip your collection to individual wav or flac files per track, and have a directory structure similar to Library Root\Artist\Album\Songs or Library Root\Genre\Artist\Album\Songs.
I've made some updates to the program and have made it available on my web site. The application now has an option to get its information from tags embedded in flac files when flac files are found. It also has an option to prepend the genre to the resulting cue sheet file names so that the cue sheets can be ordered by genre-> artist-> album instead of just artist-> album. The application also now provides a status area so that processing can be monitored.
There is a link to the application at the bottom left of the web page linked below. There are no help files in the application, but its use is explained at the web page.
Alan
windows installer error trying to install java for Recursive Cue Creator CICS modified T.No worries!
Edits: 02/28/09
If you've followed all of cics recommendations, then you have turned off the Windows installer service among other things, so this result is to be expected. I, for one, could not find the right few things to turn back on that would make java installable. You need to run the Recursive Cuesheet creator on a different system/partition and bring the cuesheets resulting over to the cMP one.
Edits: 03/01/09
Hi,
I gather that you are trying to install Java on a CICS Memory Player, and the windows installer for Java failed? I do not run a CMP box so I am not sure what the problem would be. It seems to me, that the spirit of the CMP is to shut down all of the operating system services not needed for audio. So it is possible that Java does not run on these machines.
Alan
If you have trouble installing a new program on a cMP2 box, try setting the Windows Installer Service to Automatic and rebooting. It works for me.
Best,
Dave
Al,
Many thanks for both versions of your applet - it has made me start sorting out all the erroneous / misplaced tags - I had always meant to do this but kept putting it off.
I have had a couple of runs through trying different combinations - it is really great to know that once my tags are sorted it will be a quick job to create the CUE files.
So when does the one for us lovers of Classical Music come out, i.e. the one that uses Composer and other tags?
Alan
Hi Alan,
Another member requested that in the case of classical music and reading tags embedded within flac files, use composer instead of genre when it is available. I built that for him but didn't include that behavior as standard because it didn't seem globally applicable. However, if you can describe what tags are needed or send me an example of a cue sheet, I'll see what I can do.
Alan (there I go, talking to myself again)
Alan (you can call me Al)
Just finished re-doing my tags and cue sheets - excellent - the only problem I had was loading Java onto my music pc; so I did all the work on my main pc and copied the cue files - of course the addresses were all wrong (i.e. E:\music\Beethoven... when what I wanted was Beethoven...).
I searched the web for a recursive editor and gave up; I transferred all the cue files to my music pc and thought 'oh well' editing time.
I use Pspad for the most part and by chance found it already had multiple file replaces - so 5 minutes later all my cue files point to the right place - some thing about pigs and what they wallow in!
There is obviously something wrong with some of my tags but now I have your software and Pspad I can easily change the tags and get them onto the music machine.
And one weird thing - I could never get cMP to detect the GENRE lines - now I can - so another plus!!
Many, many thanks
Alan
Can this program be used to do replacements in the title tags of multiple files? I am seeking a way to abbreviate classical titles to fit on the cPlay screen: such as using "sym" for "symphony", "conc" for "concerto", etc. Could you explain where to look in the program for doing multifile replacements?
I have not tried amending the tags using PSPAD only the CUE files. If you look in the menu bar you will see SEARCH, click on this and just over half way down there is 'Search Replace in files'.
I suppose that once you have created the CUE file then you could alter the text - however you would have to be careful to not amend the Paths in the CUE file.
If you set the options to 'Always ask before amending' then it would prompt you each time it was going to change Symphony to 'sym' or whatever - a bit more tedious but at least it is quicker than typing it in every time.
In fact I think I might try that myself.
Thanks for the response. The problem with doing replacements on the cuesheets instead of the file tags is that when you have more files and want to run the CuesheetCreator again you get cuesheets with unabbreviated titles once again. A better solution would be for the RecCueCreator to do the abbreviating each time it runs. This could be in a classical option which uses composers for genre and a series of abbreviation of the title tags.
So what is really needed is a tag editor which allows for multiple file changes. So off to google!
Yup! I had already despaired of finding such. If you do, please let me know. The problem is that the tag editor I have been able to google either don't do flacs and/or don't have search/replace functions for a given tag like title.
Edits: 02/24/09
Have a look at Tag&Rename. You can load your entire Classical folder at once, and swap tag fields between Composer and Genre (or any other) - it's quite a powerful utility. Pity it doesn't do cuesheets.
Thank you for this very usefull program!!
nt
...reload windows and / or replace processor. So if I replace processor do I have to reload windows? Assuming yes then is there anyway I can back up all my windows optimizations/registry settings/and stuff like minlogon? Or to I have to trudge through all this stuff like I did last time.
New info on problem: reloading juli drivers, a new juli soundcard, new power supply (Antec 430 Earth) did not help problem. I can run all day with 146 src/ 96 upsampling (at about 20-23% cpu load) and no issues. If I run 176 or 192 at 123 src (less than 20% cpu load) I get the problem. So I don't know if a new cpu will help.
Anyway, anybody with a comment would be appreciated.
Use CloneGenius and make a backup of your main C: partition which contains Windows, et al.
Then just reload the backup when you have finished your upgrades.
I crashed my machine when doing the cics tweaks and had to reload everything, one at a time. Clone Genius does it very quick.
Ryelands was kind enough to instruct me on this important step. (Thanks Dave)
RayBan
RayBan I have a question. If you are trying to reload Windows because you think you have something is corrupted in your current version of Windows then CloneGeniuos will only reload that corrupted version of Windows. Correct? But if you are installing a new mobo or cpu use of CloneGenious is appropriate. Correct?
That is correct.
If you create a clone from a corrupted partition, then you would just be copying the "bad" back to the machine.
Making a clone from only known good loads is the only way to go. (In fact I am presently making a "Clone" of partition C:\ before running AutoRuns or any other potentially dangerous steps/tweaks!!)
If Windows is corrupted (mine was) I started from scratch: i.e, Format disk, load Windows, and on...... Then make a Clone prior to any major changes. If it crashes, Blue Screen of D, then just re-boot from the Clone and start again.
Good luck.
RayBan
thank you
I have been going over all my control panel/bios settings to make sure I have things set up right. Here are some observations:
...found out I did not disable NDIS drivers, disabled them and caused a no boot situation. I got the following error message...'stop:0x0000007E (0x0000005,0x805E49BC,0XF9A4A6A0,0XF9A4A39C)...'. I don't know what this means if anything. In the original cics notes it says you can't disable NDIS drivers if you use graphics or networking cards which I do neither so I'm confused why I cant do this. If anybody knows I would appreciate some feedback on this. I did get a boot upon entering Safe Mode and re-enabling these drivers.
...I was not using juli@ unmapped through Windows, plus there were some NTFS file processes I did not disable in the registry-per cics. So I unmapped juli@ disabled these processes but still get the problem.
... I can go 24/176 for an hour or so before 'metallic' starts then I reboot and I'm ok. the 'metallic problem' always occurs at the beginning of a track where I seem to lose the juli@ drivers (I know because I once tried listening with some drivers missing and it sounds similar). So with this true I thought the problem was/is associated with juli@ setup or some file transfer process. Also I always have had(since my first days with cmp/cplay) a problem when switching tracks fast but initially it would just stutter or hang up cplay which was fixed through a reboot and I lived with it. Now I get the metallics. Maybe it has something to do with my hd.
... I found several other issues in control panel settings/fixed them...still get the issue. One thing I could not do was set the juli@ audio device volume to max as requested in cics spec. Does this mean anything?
Anyway if anybody has any comments on this please post them here (except for you Dave you have very very helpful in holding my hand through this process you have already given your feedback).
For those who might suggest that I reload windows...I am planning to do but am reluctant to do this because I would be w/o good cplay/cmp fully optimized music for a few days. I eventually will once I get through the bios/control panel stuff.
Edit: Found the answer in AOB... silly me
Edits: 02/25/09
Ted: You could get another HDD, load ALL of the necessary programs, drivers, etc. using another computer.
Then make a clone backup from that new config. for safety.
Shut down your machine, change to the new HDD and re-start.
That's why I now have two identical HDD's (one in the machine running) and one in the box just waiting in case I need to install it to recover from a failure, crash or situation like you are experiencing.
If I make a change, I re-install the backup with the clone backup installed. It's like a RAID only both are not in the machine at the same time.
I have bought a spare MoBo, 7300, PS, et al, just in case of disaster or unavailability. I have installed and burned them in also.
It should not take long to start from scratch from XP on up. I did it in a few hours. Once you have a list of what you need to install in what order it goes fairly quickly.
Do AutoRun/Minlogon last!!
Just an idea.
RayBan
Let me understand what you are saying. Since I need a new hdd anyway this could be a good idea.
So you are saying get another hdd, put into different computer. Load Windows sp2, cmp, cplay into it. Swap this hdd into audio pc then see if this fixes the issue. Is that right?
What about all the bios drivers, I have a different mobo in home computer vs audio computer. I'm sure there are other things that will be different as well. But is this kinda what is meant (assuming I can compensate for all the differences)?
Btw thanks for your input.
Yes, exactly. Load all of the hardstuff properly, in order, defrag after every load until you get to machine specific stuff. Then move it to the cMP2 machine, startup, do the BIOS tweaks, reload Juli@ & 1.04 drivers and see how it works. If OK do the rest and hopefully your metallic issue will be history.
Then you can format the other HDD, reload the same way and have a spare ready with most of the loads already installed. If new changes come out then you can put them on that HDD, swap, and be back on-track quickly.
I also got a Seagate external HDD for building/transfering music files/directories. Also for when new software/methods come out I can implement and listen at the same time.
Use the cMP2 machine for pleasure not experiments.
RayBan
Love your suggestion but just a few more queries
...If I load the new hdd into my home computer yes I can load xp into it, all the cmp/cplay stuff but since the old computer has a different mobo/different drivers I need to load these in order for this computer to work. Then when I transfer the hdd to new computer it has the gigabyte mobo which needs dif drivers that are not on the new hdd. So I guess I can safe start it and Qflash f6x drivers into from my flash drive...ok. But will the flash drive work w/o gigabyte drivers?
...I'm not seeing it through. Look I love the idea and I'll do it if I think it can work.
Once the BIOS drivers were loaded, they stay loaded so if you diod it once you're done, they will still be in the MoBo. F6x. I did it so I know they stay there even if you reload everything.
RayBan
ok thanks
...recognizing that I could not go for 20-30 minutes w/o getting the metallics @ 192/145 I just went 14 straight hours with no metallics by running music cue files off my 4 gig flash drive!! Not only no metallics but load of data on the fly off the flash drive was so fast that I got zero, let me repeat, zero mini-tics or glitches for the 14 hours. Finally a breakthrough.Anyway I now know my Western Digital hdd has to be replaced. But ss drives sound so good I will look at these too (I know they are expensive).
Anyway what do all of you recommend on hdd's and ss hdd's? What specs are critical --read and write or just read speeds? What about cache how big or do not worry about? I want to hear from you computer guys out there, I need your advice. I now know it is certainly important for my system.
Edits: 02/27/09
I use 3 X 1TB Western Digital RE2-GP.
http://www.wdc.com/en/products/products.asp?driveid=385
and am very satisfied. i.e. no glitches, no metallic sound, and very quiet and cool operation.
I output at 96K and most of my files are in FLAC.
Have you tried booting from a flash drive and let yor music files on internal HDD. You might also consider a Compact Flash to SATA adaptor, for your system partition.
Good luck
Martin
thanks
Is the Intel Core 2 Duo E7400 Processor a viable replacement for the 7200 in a cmp/cplay dedicated audio pc? Does anybody know? It looks to have the same socket #, big L2 cache but maybe a little advanced vs the 7200 on processing.
Faster and cheaper, seems like it would be.
Other than the slight increase in speed it seems to be the same.
One wonders if this is the latest evolution of the original 7200 and when stocks are cleared of 7300, this will be IT.
Yes thats what I thought too, Rick. Btw did you do the p4 capacitor mod or did you do the full blown linear ps mod?
Theo,
As per Alfred's post on this, I used a piece of perf-board to make two banks of capacitors for the four pin connector. I used numerous small caps to make each bank.
Just cut the cable and connect the two plus and minus wires accordingly.
I added capacitance on the 24 pin connector also. Here I carefully stripped the wires; two cuts with a wire stripper and then a razor blade to cut between the two cuts to expose the wire, then soldered in a capacitor. Use the "map" within the MB manual. I tried to connect a capacitor to each voltage rail and each ground, though there are more voltage rails than grounds so you have to double up on a few of those.
I finished up with electrical tape.
There is one negative 12 volts rail that is for starting up which I did not think needed a capacitor.
Since we have never figured out whether JULI@ is using a 3.3 volts or 5 volts I thought it was important to include all rails on the 24 pin with the one exception above.
I used the ELNA audio silk electrolytics that DIGIKEY sells. There have been many recommendations. I bought these because they were not expensive but seemed to be a premium construction. The PANASONIC FC's are probably an excellent choice for this and with these one could get more capacity per unit.
Have fun!
Rick McInnis
excellent. I've only done the p24 mod--nice sq pop. I just suspended lsass and svchost nice sq improvement---thanks!!
Hi all,Does anybody knows how to launch Beatport SYNC audioplayer successfully within cMP?
http://www.native-instruments.com/index.php?id=beatportsyncWhen I try (according too cMP’s manual) on startop cMP / PC hangs. Though foobar2000 works fine but doesn't play mp3's in cMP-mode :-(
In our home we need a player on our HTPC that can also play mp3. Otherwise war breaks out.
Although it’s easy to swith between the XP and the cMP environment by restaring,
In practice this means the HTPC is always in XP environment too facilitate audioplayers capable of playing mp3’s. The cMP environment is used less and less.
I only use it one in a while to listen to classical music with cPlaySince Beatport SYNC is by far the best sounding audio player (after cPlay off course) so I would like too launch it within cMP.
(ow, by the way: don’t judge Beatport SYNC for the fact that it’s a (free) player made for professional DJ’s. First have some good listening sessions! Than judge !!)
So does anybody know how to start Beatport SYNC within cMP ?
Thank you.
ESI juli@ -> Lavry Black DA10 -> Mogami Gold -> Klein & Hummel O300
Edits: 02/16/09 02/16/09
I just downloaded and played with Beatport SYNC - interesting program. Tried it out in my office cMP setup with desktop speakers, and sounds good, but really can't judge audio quality accurately here.
This player has a disadvantage in cMP (same thing that happened with XMPlay)- it looks like you won't be able to get RAM loading or load albums from cMP library, but you can use it anyway and play from the HDD in cMP and get the benefits of a highly optimised environment.
Go to Program files > cics Memory Player > cicsMemoryPlayer.pth. Open with Notepad and edit the CUE_PLAYER line. Enter the full path to Beatport SYNC.exe in place of the path to cicsPlay.exe that is there by default.
When you go back to cMP, it behaves differently than it does with cPlay or Winamp. It doesn't matter what folders you have added to cMP library, or what album you open, or whether RAM load is ticked Yes or No: Beatport SYNC will open, but the album you selected will not be loaded. Never mind, just use Explorer in the tree at the left and find the album you want and play it as normal in Beatport.
Hi Seger,
Many thankx for your response!
Now I know for sure it’s possible to launch Beatport SYNC from within cMP. I must have been doing something wrong or made a mistake. I installed Beatport ‘in’ cMP exactly as you described (as it also is described in the cMP manual) but on startup my PC hangs. I’ll try too install again without mistakes. Thank you !
* Beatport apperently also does some kind of RAM-loading.
In the readme.txt with goes with the program, in paragraph ‘3 BEATPORT SYNC Features’ under ‘preferences’ can be red:
“- Deck Preferences/Transport: Deck caching 8-256mb”
and also
“- Deck Caching which provides user configurable options for users to pre-allocate for BEATPORT SYNC to load tracks directly into memory.”
Sow it looks like Beatport Sync also does some kind of RAM-loading. May that’s why it also a ‘good sounding’ audioplayer.
So far only 3 audioplayers ‘sound’ really good to my ears in my setup:
- cPlay;
- Beatport SYNC (my favorite one)
- and the player inside WaveLab-lite (a free editing program from Steinberg)
I clearly hear differences which too me appear as being ‘better’. (Hopefully no placebo-effect hopefully and hopefully also not a matter of taste too which one is grown accustomed too over the years).
I’m not really sure, if my setup is capable too let me hear all the subtle differences. It’s good but definitely not the best out there: cMP PC -> Lavry Black DA10 -> Klein & Hummel O300
What do the guys think who own the real good stuff? (equipment like: DAC’s from Weiss, Prism, Lavry, dCS, ect and who listen through outstanding monitors from origins like: PMC, ATC, Geithahn, Focal, K+H, ect). What quality differences do they hear between PC audioplayers? Anyone?
ESI juli@ or LynxL22 dig i/o -> Lavry Black DA10 -> Mogami Gold -> Klein & Hummel O300
I can't find the deck preferences/transport setting in the program's settings though I see what you quoted in the readme file. Have you been able to find/set this cache size?
Hi Riboge,
I can’t find it either.
I guess it’s only accessible when you buy the full version from Native Instruments.
But nevertheless I do believe there’s ram-loading.
One can see that a track is loaded for being analyzed for BMP-rate (beats per minute) and automatic gain-adjustment.
But ehm, how do you rate the sound quality?
ESI juli@ or LynxL22 dig i/o -> Lavry Black DA10 -> Mogami Gold -> Klein & Hummel O300
This post may also help. Either %C or %F could work with your player for selected album.
Hi cics
Thankx for your response and tip.
Through the post of Seger I already knew that I must be doing something wrong as for Seger succeeded in launching Beatport in the cMP environment.
I just found my error: I forgot the last “ in the PTH-file
Also many, many thankx for sharing your idea’s and work on cMP and cPlay.
In November I ‘stumbled uppon’ your AOB Computer Transport post. Already for years I’m reading Martin Walker’s articles in soundonsound on how to optimize XP for audio purposes. But the articles of Martin couldn’t win me over to switch to a PC as my most important audiophile quality source.
But your tips on also to optimize the hardware-part (under-clocking, DRAM-timing, getting clean power supply on the mobo and processor, extra power supply’s for spherical’s like hdd, dvd, usb, no power lurking from the mobo for processor cooling, etc, ect,) finally gave me the last push I needed. With the help of your instructive and informative cMP Installation Guide & User Manual it was an exiting journey to build my first (HT-)PC transport.
I enjoy the cMP transport and the quality sound that it produces every day ! Thankx you very much.
Mark
ESI juli@ or LynxL22 dig i/o -> Lavry Black DA10 -> Mogami Gold -> Klein & Hummel O300
...what a big surprise!. I'm sure I don't have the right toslink cable but it is glass and it is made by Amphenol. I liked it right off the bat but I don't know whether I like it better than bnc @ 192 versus toslink @ 96. Biggest advantage is upper mids/highs are very clear. Soundstage is as if a new light is shining upon it. But bnc @ 192 sounds a bit better in bass and dynamics. Time will tell. I do my serious listening at 6:00-9:00am in am when power line is least noisy. I'll see.
I listened last night (in a high noise powerline environment) and toslink was way better (altho still not as good as bnc at 192 for bass & dynamics). This am (in a quiet powerline noise environment) it was closer but I still preferred toslink because of what it does to upper mids/highs (clearer/deeper soundstage more air around violins, plucked strings are to die for) so I'm sticking with this for awhile. I went back to 2.0b9v19 today at 96 (I don't seem to have a metallic problem yet) and its better yet. Toslink that is.
This stuff is really good in the highs, anyway just thought I'd share.
which supposedly will accept a 192kHz input.
So, which cable are you using, Theo? One thing for sure the cables seem to be ridiculously cheap, no complaint there. So you think all AMPHENOL cables are the same? I have not looked into this very far.
My DAC has a transformer at the input so I had assumed I did not need this but hearing what you say you are hearing and trusting your ears, I am going to see what it does.
I can't go too wrong for forty dollars, approx 19 the adaptor and the rest for the cable.
What length cable are you using and where did you get the cable?
I hope you are happy now (!!!?)
Bye,
Rick McInnis
my cable was a 2 meter $29 one. Don't know which model it was, but it was called Amphenol Ultra Series - Premium Optical Glass on my purchase order. But while it sounds good I'm not sure there isn't better like the Audioquest link 5 (or something like that) that goes for quite a bit more.
Does your BENCHMARK DAC not allow this?
The specs for recent production say it will.
cics can only use 96 to his DAC's limitations. If that is why you have only tried 96.
Give it a try,
Rick McInnis
two reasons: with 2.0bv19 I can only do 96 at higher frequencies I get the metallic issue plus juli@ only does 96 (per the esi owners manual).BTW the Benchmark does do 192.
Edits: 02/15/09
Theo,
Thanks for setting me right.
Now, I wonder whether I wasted eighteen dollars?
Bye,
Rick Mcinnis
but try it @ 96 (you can run 192 into it but I suspect it downconverts to 96) it sounds different and quite good in areas I mentioned above.
I have optimized follow ACB 0.3.
CUP is AMD 64x 3800
sound card is JULI@.set to 145db SNR.
and I use analog out.
when I set RATE to 176.4,it play OK.
but when I set to 192,it play abnormal.
What's happen?Does the CPU not power enough?
CPU could be the problem.
CPU load is very low with 176.4k output,under 20%,
L2 cache size is 512K
I'm surprised you've got so far with just 512KB L2 (assuming you upsampling 44.1 to 176.4).
Today I install XP SP2 32bit version.
same hardward,software, after ACB.There is a strang thing.
The CPU load is up to about 40-50%,my cpu is two core,the only one cpu load is high,the other is low.
I don't know it is because OS VISTA vs XP,or 32 bit vs 64 bit?
I will try install XP SP2 64bit version
YES.
I just use cPLAY(no install cMP),and upsample to 176.4 is OK.
I don't know is:44.1*4=176.4 ,the calculate is simple than 44.1=> 192
Hello cPlay users.
I've been using cPlay for a few months and like its sound quality. I'm not so fond of it's user interface and have developed my own for it, which allows me to choose music files, create a cue sheet and launch cPlay to play them.
My question (or request for cics) is, can cPlay be made to launch minimized? Currently it launches maximized and steals focus from my interface application. If not, an option to do this would be GREATLY appreciated.
Thanks!
@multiblitz,
My UI is actually a plugIn written for an HTPC using Meedio so it is not stand alone although I could develop a standalone UI.
With my plugin, you can select tracks and then a CUE file is created and cPlay is started. Inside Meedio, you jump forward or backwards and cover art is displayed too.
Keymaker
Could you shre your UI ? would be great.
I asked the question few months ago as I wrote a small VB to play a .m3u playlist.
In VB i can open a cplay process minimized:
myProcess.StartInfo.WindowStyle = ProcessWindowStyle.Minimized
However it didn't work after some version 2.x. It wouldn't play when the cplay process was minimized.
It would be great if there is an option in cplay that would not display any UI at all during playback.
.
I don't want to use another application to create and manage CUE play list.
Is it possible to add simple UI enhancement.
ADD button to add an audio file to the play list
DEL to remove an audio file from the play list
SAVE button to save the play list
UP/Down Button to move the position of an audio file in the play list
Good suggestion but these seemingly simple functions require full playlist edit capability, e.g. Add would require a browser (cue, wav or flac) followed by specific track selection, etc..
If we go this far we might as well do a proper playlist tool which I want to keep outside of cPlay.
I'm always anxious about toggling minlogon and would prefer to stay within.
Theo,
You are going to reformat your disk: there will be nothing there, no WINDOWS, no MINLOGON (which is like a subset of WINDOWS).
This is akin to eating your cake and having it, too. No one has figured out how to do that either.
I have found every time I have had to do this I thought I did a better job so there is a silver lining of sorts.
The last thing you want to is any kind of shortcut. Make it as perfect as possible to eliminate the possibility of the installation being a cause of future problems.
It isn't that bad.
Bye,
Rick McInnis
... listened to 2.0bv18 sounded great. Got curious and reloaded back in 2.0bv19 and it sounded magnificient but I quickly got into the heavy metallic overlay. I hate to keep throwing $ at this w/o a clue. So what are my options: power supply (new), cpu (new) other? Any ideas anybody? The only thing different in my audio pc is the ps. I'm using a 550 watt agi.
Edits: 02/09/09 02/09/09
hope you have WINDOWS on its own sector.
At least, that is what I would do if this problem was mine.
Something has just gone wrong and I bet, though you know I do not know, this is the only way to be assured it is fixed.
It just occurred to me; have you taken your DAC to another system and verified it is not the problem? Just wondered.
I have dealt with much coincidence lately with my line stage problem (now fixed to its author having one of those AH HA moments) but before the obvious was made obvious we, and especially I, made many erroneous assumptions as to where the problem lay.
Just to be sure before formatting that HDD sector!
Best of luck and sympathy,
Rick McInnis
yes I can do that but 1st a review of the facts:problem only occurs on version 19 not 18 (i've had 18 for more than a week and it plays perfectly). 5 minutes into 19 problem occurs. Problem always goes away with reboot. The dac is solid. I've used dac with other sources and it never had this problem.
I've swapped out power conditioners, sound cards... problem is independent of these items.
problem also occurred on 15 16 17 but not versions 14 and prior nor on 18.
Let me be clear I don't believe that the problem occurs solely with these versions of cplay(all you other guys are using it w/o a problem). I believe something in my system interacts with 19, 15, 16 17 to create the issue--like power supply or cpu or mobo. It always sounds like a sound card driver issue when it occurs. And if I reload 18 (after listening to 19 and problem has occurred) and I don't reboot problem is still there until I reboot then the system is problem free on 18 for as long as I listen. So when it occurs I either lose my juli drivers or something happens to asio. When I get the antec ea 430 that will eliminate another variable.
Edits: 02/10/09 02/10/09
Hi theob,
As you may have seen I get this popping sound at the beginning of tracks when running FLAC files, as far as I am aware only one other person gets this. It seems that only one or two people get this metallic sound.
So what is the cause of these differences?
I have been reading through the various threads which concern Memory sticks, Measurements of various configurations, User perception of the effects of changes and I am a little confused.
I do not subscribe to the view that it is down to the unpredictability of computers - it is I feel down to the usual problem with computers - combinations of events / states.
Since I believe that people perceive these differences I was thinking about collating setups from the various cMP2 users; they would send me the setups of their pcs (e.g. cpuZ prints, cMP / cPLAY diagnostics, hardware (especially memory used) and I would try to analyse this data to see which was important in sound quality.
Some cMP2 users have posted details of measurements taken - take this with the above and soon I would think there would be a body of data to match with subjective assessments of changes.
I do not know what cics thinks about such an idea, but I would like to put together a questionaire based on the cMP2 installtaion which would allow people to easily supply the data.
Obviously the more people the better the result.
It does not help you theob - I hope it gets sorted soon.
Alan
I think it is a great idea to not only id root causes of problems but also to optimize sq. I am with you 100%. Let me know what data you need to start and I will happily be one of the first to supply same.
You know all of my training/experience in problem solving always gets down to what's different and/or what changed to cause the problem. I am sure there is something that will pop as being (as we called it) the red x in this problem. Also I am very happy with listening to 2.0bv18 as I have gone through 8 days of consecutive listening w/o the problem occurring--not even once. So you say what is my problem...why not just be happy with 18 and be done? Well I heard the brilliance of 19 and I won't quit till I can get it.
The only thing that worries me is if my mobo needs to be replaced. As I understand there not be many left out there to get.
I will go through the various postings and try to get a questionnaire together by the weekend.
I know this is tmi but if you just ignore the register dumps you can see cpu, mobo, memory data
-------------------------
CPU-Z version 1.47
-------------------------
Processors Map
------------------------------------------------------------------------------------
Number of processors 1
Number of threads 2
Processor 0
-- Core 0
-- Thread 0
-- Core 1
-- Thread 0
Processors Information
------------------------------------------------------------------------------------
Processor 1 (ID = 0)
Number of cores 2 (max 2)
Number of threads 2 (max 2)
Name Intel Core 2 Duo E7200
Codename Wolfdale
Specification Intel(R) Core(TM)2 Duo CPU E7200 @ 2.53GHz
Package Socket 775 LGA (platform ID = 0h)
CPUID 6.7.6
Extended CPUID 6.17
Core Stepping M0
Technology 45 nm
Core Speed 1020.0 MHz (6.0 x 170.0 MHz)
Rated Bus speed 680.0 MHz
Stock frequency 2533 MHz
Instructions sets MMX, SSE, SSE2, SSE3, SSSE3, SSE4.1, EM64T
L1 Data cache 2 x 32 KBytes, 8-way set associative, 64-byte line size
L1 Instruction cache 2 x 32 KBytes, 8-way set associative, 64-byte line size
L2 cache 3072 KBytes, 12-way set associative, 64-byte line size
FID/VID Control yes
FID range 6.0x - 9.5x
max VID 1.013 V
Features
Thread dumps
------------------------------------------------------------------------------------
CPU Thread 0
APIC ID 0
Topology Processor ID 0, Core ID 0, Thread ID 0
Type 01008001h
Max CPUID level 0000000Ah
Max CPUID ext. level 80000008h
Function eax ebx ecx edx
0x00000000 0x0000000A 0x756E6547 0x6C65746E 0x49656E69
0x00000001 0x00010676 0x00020800 0x0008E39D 0xBFEBFBFF
0x00000002 0x05B0B101 0x005657F0 0x00000000 0x2CB43048
0x00000003 0x00000000 0x00000000 0x00000000 0x00000000
0x00000004 0x04000121 0x01C0003F 0x0000003F 0x00000001
0x00000004 0x04000122 0x01C0003F 0x0000003F 0x00000001
0x00000004 0x04004143 0x02C0003F 0x00000FFF 0x00000001
0x00000005 0x00000040 0x00000040 0x00000003 0x00022220
0x00000006 0x00000001 0x00000002 0x00000001 0x00000000
0x00000007 0x00000000 0x00000000 0x00000000 0x00000000
0x00000008 0x00000400 0x00000000 0x00000000 0x00000000
0x00000009 0x00000000 0x00000000 0x00000000 0x00000000
0x0000000A 0x07280202 0x00000000 0x00000000 0x00000503
0x80000000 0x80000008 0x00000000 0x00000000 0x00000000
0x80000001 0x00000000 0x00000000 0x00000001 0x20000000
0x80000002 0x65746E49 0x2952286C 0x726F4320 0x4D542865
0x80000003 0x44203229 0x43206F75 0x20205550 0x45202020
0x80000004 0x30303237 0x20402020 0x33352E32 0x007A4847
0x80000005 0x00000000 0x00000000 0x00000000 0x00000000
0x80000006 0x00000000 0x00000000 0x0C006040 0x00000000
0x80000007 0x00000000 0x00000000 0x00000000 0x00000000
0x80000008 0x00003024 0x00000000 0x00000000 0x00000000
Cache descriptor Level 1 D 32 KB 1 thread(s)
Cache descriptor Level 1 I 32 KB 1 thread(s)
Cache descriptor Level 2 U 3 MB 2 thread(s)
MSR 0x0000001B edx = 0x00000000 eax = 0xFEE00900
MSR 0x000000E8 edx = 0x00000465 eax = 0x2B0D2F0E
MSR 0x00000017 edx = 0x04000000 eax = 0xA984C90F
MSR 0x000000CD edx = 0x00000000 eax = 0x00000800
MSR 0x0000003F edx = 0x00000000 eax = 0x00000000
MSR 0x000000CE edx = 0x000C490F eax = 0x10100709
MSR 0x000001A0 edx = 0x00000044 eax = 0x60870489
MSR 0x000000EE edx = 0x00000000 eax = 0x867D0300
MSR 0x0000011E edx = 0x00000000 eax = 0xBE702119
MSR 0x0000019C edx = 0x00000000 eax = 0x883C0000
MSR 0x00000198 edx = 0x060C490F eax = 0x0600060F
MSR 0x00000199 edx = 0x00000000 eax = 0x0000060F
CPU Thread 1
APIC ID 1
Topology Processor ID 0, Core ID 1, Thread ID 0
Type 01008001h
Max CPUID level 0000000Ah
Max CPUID ext. level 80000008h
Function eax ebx ecx edx
0x00000000 0x0000000A 0x756E6547 0x6C65746E 0x49656E69
0x00000001 0x00010676 0x01020800 0x0008E39D 0xBFEBFBFF
0x00000002 0x05B0B101 0x005657F0 0x00000000 0x2CB43048
0x00000003 0x00000000 0x00000000 0x00000000 0x00000000
0x00000004 0x04000121 0x01C0003F 0x0000003F 0x00000001
0x00000004 0x04000122 0x01C0003F 0x0000003F 0x00000001
0x00000004 0x04004143 0x02C0003F 0x00000FFF 0x00000001
0x00000005 0x00000040 0x00000040 0x00000003 0x00022220
0x00000006 0x00000001 0x00000002 0x00000001 0x00000000
0x00000007 0x00000000 0x00000000 0x00000000 0x00000000
0x00000008 0x00000400 0x00000000 0x00000000 0x00000000
0x00000009 0x00000000 0x00000000 0x00000000 0x00000000
0x0000000A 0x07280202 0x00000000 0x00000000 0x00000503
0x80000000 0x80000008 0x00000000 0x00000000 0x00000000
0x80000001 0x00000000 0x00000000 0x00000001 0x20000000
0x80000002 0x65746E49 0x2952286C 0x726F4320 0x4D542865
0x80000003 0x44203229 0x43206F75 0x20205550 0x45202020
0x80000004 0x30303237 0x20402020 0x33352E32 0x007A4847
0x80000005 0x00000000 0x00000000 0x00000000 0x00000000
0x80000006 0x00000000 0x00000000 0x0C006040 0x00000000
0x80000007 0x00000000 0x00000000 0x00000000 0x00000000
0x80000008 0x00003024 0x00000000 0x00000000 0x00000000
Cache descriptor Level 1 D 32 KB 1 thread(s)
Cache descriptor Level 1 I 32 KB 1 thread(s)
Cache descriptor Level 2 U 3 MB 2 thread(s)
MSR 0x0000001B edx = 0x00000000 eax = 0xFEE00800
MSR 0x000000E8 edx = 0x00000121 eax = 0x2BA16E8A
MSR 0x00000017 edx = 0x04000000 eax = 0xA984C90F
MSR 0x000000CD edx = 0x00000000 eax = 0x00000800
MSR 0x0000003F edx = 0x00000000 eax = 0x00000000
MSR 0x000000CE edx = 0x000C490F eax = 0x10100709
MSR 0x000001A0 edx = 0x00000044 eax = 0x60870489
MSR 0x000000EE edx = 0x00000000 eax = 0x867D0300
MSR 0x0000011E edx = 0x00000000 eax = 0xBE702119
MSR 0x0000019C edx = 0x00000000 eax = 0x88410000
MSR 0x00000198 edx = 0x060C490F eax = 0x0600060F
MSR 0x00000199 edx = 0x00000000 eax = 0x0000060F
Chipset
------------------------------------------------------------------------------
Northbridge Intel P35/G33/G31 rev. 10
Southbridge Intel 82801GB (ICH7/R) rev. A1
Memory Type DDR2
Memory Size 256 MBytes
Channels Single
Memory Frequency 170.0 MHz (1:1)
CAS# 3.0
RAS# to CAS# 3
RAS# Precharge 3
Cycle Time (tRAS) 5
Command Rate 2T
MCHBAR dump
-----------
Base address 0x0FED14000
Size 4096
0 1 2 3 4 5 6 7 8 9 A B C D E F
000 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
010 E0 15 0A 3D 00 00 00 00 00 00 00 00 00 00 00 00
020 01 30 01 00 00 00 00 00 00 00 00 00 00 00 00 00
030 06 5A 1F 00 10 28 90 01 00 00 00 37 17 44 03 23
040 08 80 03 0A 00 00 0F 00 00 00 00 00 00 00 00 00
050 01 00 84 00 02 01 8F 00 00 00 00 00 00 00 00 00
060 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
070 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
080 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
090 55 50 00 00 00 00 00 00 00 00 00 00 00 00 00 00
0A0 00 00 40 12 00 00 D8 00 00 00 00 00 00 00 00 00
0B0 00 00 D8 00 00 00 00 00 00 00 00 00 00 00 D8 00
0C0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
0D0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
0E0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
0F0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
100 00 00 00 01 00 00 00 00 00 01 00 00 00 00 00 00
110 3C 6C 00 00 00 99 00 A9 48 10 00 00 00 00 00 00
120 5F 7F 8D 50 07 E0 1F 40 05 1F 80 00 10 92 00 00
130 82 36 43 C8 07 26 C0 29 C0 51 F2 7D C0 51 F2 7D
140 00 00 00 00 F7 60 1C 00 00 00 00 00 63 F8 DB 01
150 0F 00 80 07 00 00 00 00 00 00 00 00 00 00 00 00
160 00 00 00 00 1B 00 00 00 1B 00 00 00 1B 00 00 00
170 1B 00 00 00 FF 01 34 01 00 00 00 00 00 FF FF FF
180 00 01 00 00 00 00 00 00 00 00 00 00 01 0F 00 00
190 20 00 00 00 00 00 00 00 01 34 00 00 CC BF 00 00
1A0 33 00 00 00 0F 10 00 00 00 00 00 00 A8 04 00 00
1B0 00 00 00 00 70 00 00 00 70 00 00 00 70 00 00 00
1C0 70 00 00 00 70 00 00 00 70 00 00 00 70 00 00 00
1D0 70 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
1E0 03 18 55 00 8A 00 00 00 F3 00 00 01 00 03 00 00
1F0 C0 80 00 00 02 04 06 08 20 1F 00 00 00 00 00 00
200 04 00 04 00 04 00 04 00 03 00 00 00 00 00 00 00
210 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
220 17 11 00 58 01 02 24 0C 00 00 00 00 00 00 00 00
230 00 00 00 00 00 00 00 00 00 00 00 00 89 7A 0D 00
240 00 11 00 01 10 23 01 00 00 07 C1 0E 00 11 30 00
250 51 02 12 66 24 00 68 34 46 85 06 12 06 05 A5 15
260 91 3D 17 0B 78 00 0C 13 00 2E C5 CF 33 BF 02 03
270 00 1E 55 01 01 88 08 00 81 18 14 88 41 00 00 00
280 00 00 00 00 00 00 00 00 00 02 04 08 10 20 40 FF
290 1C 09 F2 04 00 00 00 00 01 00 00 00 68 06 00 00
2A0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
2B0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 01
2C0 F0 C5 0C 00 00 00 00 00 00 00 00 00 00 00 00 00
2D0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
2E0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
2F0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
300 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
310 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
320 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
330 00 00 00 00 00 00 98 01 00 00 00 02 00 66 00 00
340 00 80 00 40 5A 1A 00 00 00 00 00 00 00 00 00 00
350 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
360 00 00 98 01 00 00 00 02 00 66 00 00 00 80 00 80
370 92 12 00 00 00 00 00 00 00 00 00 00 00 00 00 00
380 00 00 00 00 00 00 00 00 00 00 00 00 00 00 CC 00
390 00 00 00 02 00 33 00 00 00 80 00 80 92 92 00 00
3A0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
3B0 00 00 00 00 00 00 00 00 00 00 CC 00 00 00 00 02
3C0 00 33 00 00 00 80 00 80 92 92 00 00 00 00 00 00
3D0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
3E0 00 00 00 00 00 00 20 02 00 00 00 02 00 88 00 00
3F0 00 80 00 40 5A 1A 00 00 00 00 00 00 00 00 00 00
400 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
410 00 00 20 02 00 00 00 02 00 88 00 00 00 80 00 40
420 5A 1A 00 00 00 00 00 00 00 00 00 00 00 00 00 00
430 00 00 00 00 00 00 00 00 00 00 00 00 00 00 98 01
440 00 00 00 02 00 66 00 00 00 80 00 40 5A 9A 00 00
450 2D 00 00 00 2D 00 00 00 00 00 00 00 00 00 00 00
460 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
470 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
480 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
490 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
4A0 63 F8 DB 01 63 F8 DB 01 C3 01 00 06 00 00 00 00
4B0 00 00 FF FF 00 00 00 00 00 00 00 00 00 00 00 00
4C0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
4D0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
4E0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
4F0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
500 00 00 00 00 00 00 00 00 9B 00 00 00 35 00 00 00
510 4D 84 00 00 BC 04 00 00 80 04 00 00 72 0C 00 00
520 00 00 00 00 00 00 00 00 01 00 00 00 55 55 00 00
530 66 77 00 00 24 09 00 00 24 09 00 00 00 0C 00 00
540 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
550 01 00 00 00 06 70 00 00 0F 00 00 00 05 00 00 00
560 01 00 00 00 06 60 00 00 0F 00 00 00 05 00 00 00
570 01 00 00 00 06 50 00 00 0F 00 00 00 85 00 00 00
580 01 00 00 00 06 40 00 00 0F 00 00 00 00 00 00 00
590 01 00 00 00 06 30 00 00 0F 00 00 00 00 00 00 00
5A0 01 00 00 00 06 20 00 00 0F 00 00 00 00 00 00 00
5B0 01 00 00 00 06 10 00 00 0F 00 00 00 00 00 00 00
5C0 01 00 00 00 06 00 00 00 0F 00 00 00 00 00 00 00
5D0 0C 00 00 80 00 00 00 00 F8 D7 01 EE 18 00 80 38
5E0 EC EC 05 0A 07 38 00 00 00 80 07 00 00 00 00 00
5F0 00 00 00 00 0F 00 00 00 00 00 00 00 00 00 00 80
600 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
610 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
620 00 00 00 40 00 00 00 00 00 00 00 00 00 00 00 00
630 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
640 00 11 00 00 00 00 20 00 00 00 80 08 00 11 41 00
650 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 10
660 01 08 00 00 00 00 00 13 00 30 0C C0 30 18 02 00
670 00 00 55 00 01 00 00 00 82 20 26 08 62 00 00 00
680 00 00 00 00 00 00 00 00 00 03 07 0F 1F 3F 7F FF
690 08 0E 02 04 00 00 00 00 00 00 00 00 00 00 00 00
6A0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
6B0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 01
6C0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
6D0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 04 01
6E0 00 00 00 00 00 00 04 01 00 00 00 00 00 01 00 02
6F0 00 00 00 00 00 01 00 02 00 00 00 00 00 00 00 00
700 00 00 00 00 02 01 00 08 00 00 00 00 00 00 02 08
710 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
720 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
730 00 00 00 00 00 00 20 02 00 00 00 00 00 88 00 00
740 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
750 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
760 00 00 20 02 00 00 00 00 00 88 00 00 00 00 00 00
770 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
780 00 00 00 00 00 00 00 00 00 00 00 00 00 00 20 02
790 00 00 00 00 00 88 00 00 00 00 00 00 00 00 00 00
7A0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
7B0 00 00 00 00 00 00 00 00 00 00 20 02 00 00 00 00
7C0 00 88 00 00 00 00 00 00 00 00 00 00 00 00 00 00
7D0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
7E0 00 00 00 00 00 00 20 02 00 00 00 00 00 88 00 00
7F0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
800 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
810 00 00 20 02 00 00 00 00 00 88 00 00 00 00 00 00
820 00 00 00 00 2B D4 70 00 2B D4 70 00 00 00 00 00
830 00 00 00 00 00 00 00 00 00 00 00 00 00 00 20 02
840 00 00 00 00 00 88 00 00 00 00 00 00 00 00 00 00
850 36 00 00 00 36 00 00 00 00 00 00 00 00 00 00 00
860 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
870 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
880 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
890 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
8A0 63 F8 DB 01 63 F8 DB 01 C3 01 00 06 00 00 00 00
8B0 00 00 FF FF 00 00 00 00 00 00 00 00 00 00 00 00
8C0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
8D0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
8E0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
8F0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
900 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
910 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
920 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
930 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
940 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
950 00 00 00 00 00 70 00 00 0F 00 00 00 00 00 00 00
960 00 00 00 00 00 60 00 00 0F 00 00 00 00 00 00 00
970 00 00 00 00 00 50 00 00 0F 00 00 00 00 00 00 00
980 00 00 00 00 00 40 00 00 0F 00 00 00 40 00 00 00
990 00 00 00 00 00 30 00 00 0F 00 00 00 40 00 00 00
9A0 00 00 00 00 00 20 00 00 0F 00 00 00 00 00 00 00
9B0 00 00 00 00 00 10 00 00 0F 00 00 00 40 00 00 00
9C0 00 00 00 00 00 00 00 00 0F 00 00 00 00 00 00 00
9D0 09 00 00 80 03 00 00 00 3C C0 01 FF 18 00 80 3F
9E0 00 00 00 00 07 38 00 00 00 F0 07 00 00 00 00 00
9F0 00 00 00 00 0E 00 00 00 00 00 00 00 00 00 00 00
A00 00 00 00 00 00 00 00 00 00 00 00 00 01 00 00 00
A10 00 00 00 04 00 01 00 00 00 00 00 00 00 00 00 00
A20 00 00 00 00 00 00 00 00 00 00 00 E0 00 00 00 00
A30 18 06 00 44 00 00 00 00 00 00 00 00 00 00 00 00
A40 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
A50 00 00 01 00 00 00 00 00 00 00 00 00 90 01 00 00
A60 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
A70 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
A80 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
A90 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
AA0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
AB0 00 00 00 00 00 00 00 01 00 00 00 00 00 00 00 00
AC0 00 00 00 00 00 04 00 00 00 00 00 00 00 00 00 00
AD0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
AE0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
AF0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
B00 00 00 00 00 00 00 89 00 00 00 00 00 00 00 00 00
B10 00 00 00 00 00 00 89 00 00 00 00 00 00 00 00 00
B20 00 00 00 00 09 19 00 00 89 00 00 00 00 00 00 00
B30 00 00 81 00 00 00 00 00 00 00 81 00 00 00 00 00
B40 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
B50 08 00 08 00 08 00 08 00 08 00 08 00 00 00 00 00
B60 10 00 15 16 00 00 80 00 00 00 00 BD BD 00 00 00
B70 41 C6 81 FD 41 00 03 18 08 00 00 00 00 00 00 00
B80 00 00 01 01 02 02 03 03 04 04 05 05 06 06 08 0A
B90 0C 0E 11 14 17 1A 1D 20 23 26 29 2C 2F 32 35 38
BA0 03 03 04 04 05 05 06 06 07 08 0A 0C 0F 12 15 18
BB0 1B 1E 21 24 27 2A 2D 30 33 36 39 3C 3F 3F 3F 3F
BC0 37 47 37 47 37 47 37 37 47 37 47 00 0F 09 00 00
BD0 21 1D 12 16 3F 3F 00 00 25 3F 25 3F 0F 09 0F 09
BE0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
BF0 C6 18 00 00 FF FF 03 85 03 03 03 03 11 11 11 11
C00 10 20 00 00 1F 1F 5B 6D 80 00 00 00 00 00 00 00
C10 00 00 00 00 00 00 00 00 00 00 00 00 00 80 00 00
C20 F0 A1 0A 10 00 00 00 00 00 00 00 00 00 00 00 00
C30 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
C40 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
C50 0F 0F 0F 1E 00 00 00 00 00 00 00 00 00 00 00 00
C60 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
C70 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
C80 01 01 01 01 00 00 00 00 00 00 00 00 12 00 03 00
C90 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
CA0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
CB0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
CC0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
CD0 00 00 00 00 00 00 00 00 00 00 00 FF 00 00 00 00
CE0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
CF0 00 00 00 00 03 00 00 00 00 00 00 00 00 00 00 00
D00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
D10 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
D20 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
D30 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
D40 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
D50 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
D60 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
D70 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
D80 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
D90 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
DA0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
DB0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
DC0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
DD0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
DE0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
DF0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
E00 00 00 00 00 00 00 00 00 40 00 00 00 C4 01 00 03
E10 00 35 20 00 00 00 00 00 00 00 00 00 00 00 00 00
E20 00 00 00 00 00 00 00 00 00 00 00 00 C0 6D A9 01
E30 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
E40 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
E50 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
E60 FF FF FF 0F 00 F0 00 00 0F 00 00 00 00 00 00 00
E70 00 00 00 42 C4 9E 31 00 20 F0 00 01 00 00 00 00
E80 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
E90 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
EA0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
EB0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
EC0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
ED0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
EE0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
EF0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
F00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
F10 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
F20 FF FF FF 00 00 00 00 00 00 00 00 00 00 00 00 00
F30 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
F40 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
F50 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
F60 00 00 00 00 00 00 00 00 00 00 00 00 00 00 FE FF
F70 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
F80 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
F90 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
FA0 02 00 00 00 02 00 11 20 55 35 82 00 00 00 10 40
FB0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
FC0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
FD0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
FE0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
FF0 00 00 00 00 CB 5A DB 06 4C 02 00 00 00 00 00 00
Memory SPD
------------------------------------------------------------------------------
DIMM #1
General
Memory type DDR2
Module format Regular UDIMM
Manufacturer (ID) Kingston (7F98000000000000)
Size 256 MBytes
Max bandwidth PC2-4300 (266 MHz)
Part number 9905273-003.B02LF
Serial number B6CC224D
Manufacturing date Week 02/Year 09
Attributes
Number of banks 1
Data width 64 bits
Correction None
Nominal Voltage 1.80 Volts
EPP no
XMP no
Timings table
Frequency (MHz) 200 266
CAS# 3.0 4.0
RAS# to CAS# delay 3 4
RAS# Precharge 3 4
TRAS 9 12
TRC 12 16
Dump Module #1
0 1 2 3 4 5 6 7 8 9 A B C D E F
00 80 08 08 0D 0A 60 40 00 05 3D 50 00 82 10 00 00
10 0C 04 18 01 02 00 01 50 50 00 00 3C 28 3C 2D 40
20 25 37 10 22 3C 1E 1E 00 00 3C 69 80 1E 28 00 00
30 00 00 00 00 00 00 00 00 00 00 00 00 00 00 12 C7
40 7F 98 00 00 00 00 00 00 04 39 39 30 35 32 37 33
50 2D 30 30 33 2E 42 30 32 4C 46 00 00 00 09 02 B6
60 CC 22 4D 00 00 00 00 00 00 00 00 00 00 00 00 00
70 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
80 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
90 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
A0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
B0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
C0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
D0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
E0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
F0 39 39 30 35 32 37 33 2D 30 30 33 2E 42 30 32 4C
Monitoring
------------------------------------------------------------------------------
Mainboard Model G31M-S2L (0x1F6 - 0x9CE9BC)
LPCIO
-----------------------------------------------------
Vendor ITE
Model IT8718
Vendor ID 0x90
Chip ID 0x8718
Revision ID 0x5
Config Mode I/O address 0x2E
Dump config mode register space, LDN = 0x4
0 1 2 3 4 5 6 7 8 9 A B C D E F
00 00 00 00 00 00 00 00 04 00 00 00 00 00 00 00 00
10 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
20 87 18 05 10 00 40 3F 00 41 88 00 00 01 00 00 00
30 01 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
40 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
50 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
60 02 90 00 00 00 00 00 00 00 00 00 00 00 00 00 00
70 00 02 00 00 04 04 00 00 00 00 00 00 00 00 00 00
Hardware monitor
-----------------------------------------------------
ITE IT87 hardware monitor
Voltage sensor 0 0.85 Volts [0x35] (CPU VCORE)
Voltage sensor 1 1.87 Volts [0x75] (DDR)
Voltage sensor 2 3.44 Volts [0xD7] (+3.3V)
Voltage sensor 3 5.13 Volts [0xBF] (+5V)
Voltage sensor 4 12.54 Volts [0xC4] (+12V)
Voltage sensor 8 3.26 Volts [0xCC] (VBAT)
Temperature sensor 2 21°C (69°F) [0x15] (CPU)
Fan sensor 0 874 RPM [0x304] (FANIN0)
Dump hardware monitor
LPC Register space, base address = 0x0290
0 1 2 3 4 5 6 7 8 9 A B C D E F
00 13 10 80 00 FF FF FB FF FF 87 54 09 87 04 FF 00
10 FF FF FF 36 D7 FE 7F C2 03 FF 00 FF FF FF FF FF
20 35 75 D7 BF C4 01 4C FF CC C9 FC 15 80 FB D9 D9
30 FF 00 FF 00 FF 00 FF 00 FF 00 FF 00 FF 00 FF 00
40 7F 7F 7F 7F 55 FF 5F 74 2D 40 9C 22 FF FF FF FF
50 FF 1C 7F 7F 7F 50 F6 F6 90 F6 05 12 60 00 00 00
60 00 14 41 23 90 03 FF FF 7F 7F 7F 00 00 7F FF FF
70 00 14 41 23 90 03 FF FF FF FF FF FF FF FF FF FF
80 00 00 00 00 00 00 00 00 00 00 00 00 00 00 02 00
90 FF 00 00 00 FF 00 00 00 FF FF FF FF FF FF FF FF
A0 00 00 00 00 00 00 00 FF FF FF FF FF FF FF FF FF
B0 FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF
C0 FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF
D0 FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF
E0 FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF
F0 FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF
Hardware monitor
-----------------------------------------------------
Intel Core 2 Duo E7200 hardware monitor
Temperature sensor 0 40°C (103°F) [0x3C] (Core #0)
Temperature sensor 1 35°C (94°F) [0x41] (Core #1)
Dump hardware monitor
PCI Device List
------------------------------------------------------------------------------
Host Bridge
bus 0 (0x00), device 0 (0x00), function 0 (0x00)
Common header
Vendor ID 0x8086
Model ID 0x29C0
Revision ID 0x10
PI 0x00
SubClass 0x00
BaseClass 0x06
Cache Line 0x00
Latency 0x00
Header 0x00
PCI header
Subvendor ID 0x1458
Subsystem ID 0x5000
Int. Line 0x00
Int. Pin 0x00
Capabilities
Vendor Dependant Capability
Offset E0h
Dump
0 1 2 3 4 5 6 7 8 9 A B C D E F
00 86 80 C0 29 06 00 90 20 10 00 00 06 00 00 00 00
10 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
20 00 00 00 00 00 00 00 00 00 00 00 00 58 14 00 50
30 00 00 00 00 E0 00 00 00 00 00 00 00 00 00 00 00
40 01 90 D1 FE 00 00 00 00 01 40 D1 FE 00 00 00 00
50 00 00 30 01 09 00 00 00 00 00 00 00 00 00 00 00
60 05 00 00 D0 00 00 00 00 01 80 D1 FE 00 00 00 00
70 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
80 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
90 10 11 01 00 00 00 00 00 FF 03 00 00 00 0A 79 00
A0 04 00 00 01 00 00 80 0F 00 00 70 0F 00 00 60 0F
B0 00 10 00 00 00 00 00 00 00 00 00 00 00 00 00 00
C0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
D0 00 00 00 00 00 00 00 00 49 00 00 00 00 00 00 BE
E0 09 00 0B B1 62 00 A5 2B 63 1C 16 00 00 00 00 00
F0 00 00 00 00 00 00 00 00 90 0F 10 00 00 00 00 00
VGA Controller
bus 0 (0x00), device 2 (0x02), function 0 (0x00)
Common header
Vendor ID 0x8086
Model ID 0x29C2
Revision ID 0x10
PI 0x00
SubClass 0x00
BaseClass 0x03
Cache Line 0x00
Latency 0x00
Header 0x00
PCI header
Address 0 (memory) 0xF0100000
Address 1 (port) 0x0000E400
Address 2 (memory) 0xE0000000
Address 3 (memory) 0xF0000000
Subvendor ID 0x1458
Subsystem ID 0xD000
Int. Line 0x0A
Int. Pin 0x01
Capabilities
Message Signalled Interrupts Capability
Offset 90h
Power Management Capability
Offset D0h
Version 1.1
Dump
0 1 2 3 4 5 6 7 8 9 A B C D E F
00 86 80 C2 29 07 00 90 00 10 00 00 03 00 00 00 00
10 00 00 10 F0 01 E4 00 00 08 00 00 E0 00 00 00 F0
20 00 00 00 00 00 00 00 00 00 00 00 00 58 14 00 D0
30 00 00 00 00 90 00 00 00 00 00 00 00 0A 01 00 00
40 09 00 0B B1 62 00 A5 2B 63 1C 16 00 00 00 00 00
50 00 00 30 01 09 00 00 00 00 00 00 00 00 00 80 0F
60 00 00 02 00 00 00 00 00 00 00 00 00 00 00 00 00
70 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
80 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
90 05 D0 00 00 00 00 00 00 00 00 00 00 00 00 00 00
A0 11 11 00 00 00 00 06 03 00 00 00 00 00 00 00 00
B0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
C0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
D0 01 00 22 00 00 00 00 00 00 00 00 00 00 01 02 00
E0 00 00 00 00 00 00 00 00 00 80 00 00 00 00 00 00
F0 12 00 03 00 00 00 00 00 90 0F 10 00 00 08 5E 0F
PCI to PCI Bridge
bus 0 (0x00), device 28 (0x1C), function 0 (0x00)
Common header
Vendor ID 0x8086
Model ID 0x27D0
Revision ID 0x01
PI 0x00
SubClass 0x04
BaseClass 0x06
Cache Line 0x08
Latency 0x00
Header 0x81
PCI header
Primary bus 0x00
Secondary bus 0x01
Int. Line 0x10
Int. Pin 0x01
Capabilities
PCI Express Capability
Offset 40h
Device type Root Port of PCI-E Root Complex
Port 1
Version 1.0
Physical slot #2
Presence detect no
Link width 0x (max 1x)
Message Signalled Interrupts Capability
Offset 80h
Subsystem Vendor Capability
Offset 90h
Subsystem ID 0x5001
Sub. Vendor ID 0x1458
Power Management Capability
Offset A0h
Version 1.1
Extended capabilities
Virtual Channel
Offset 100h
Root Complex Link Declaration
Offset 180h
Link Entries # 1
Port Number 1
Dump
0 1 2 3 4 5 6 7 8 9 A B C D E F
00 86 80 D0 27 07 00 10 00 01 00 04 06 08 00 81 00
10 00 00 00 00 00 00 00 00 00 01 01 00 C0 C0 00 20
20 F0 FF 00 00 F1 FF 01 00 00 00 00 00 00 00 00 00
30 00 00 00 00 40 00 00 00 00 00 00 00 10 01 00 00
40 10 80 41 01 C0 0F 00 00 00 00 10 00 11 44 11 01
50 00 00 01 10 60 05 80 00 00 00 00 00 00 00 00 00
60 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
70 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
80 05 90 00 00 00 00 00 00 00 00 00 00 00 00 00 00
90 0D A0 00 00 58 14 01 50 00 00 00 00 00 00 00 00
A0 01 00 02 C8 00 00 00 00 00 00 00 00 00 00 00 00
B0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
C0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
D0 00 00 00 00 00 00 00 00 00 00 11 80 00 00 00 00
E0 00 00 C7 00 06 07 08 00 00 00 00 00 00 00 00 00
F0 00 00 00 00 00 00 00 00 86 0F 01 00 00 00 00 00
100 02 00 01 18 01 00 00 00 01 00 00 00 00 00 00 00
110 01 00 00 00 01 00 00 80 00 00 00 00 01 00 00 00
120 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
130 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
USB Controller (UHCI)
bus 0 (0x00), device 29 (0x1D), function 0 (0x00)
Common header
Vendor ID 0x8086
Model ID 0x27C8
Revision ID 0x01
PI 0x00
SubClass 0x03
BaseClass 0x0C
Cache Line 0x00
Latency 0x00
Header 0x80
PCI header
Address 4 (port) 0x0000E000
Subvendor ID 0x1458
Subsystem ID 0x5004
Int. Line 0x17
Int. Pin 0x01
Dump
0 1 2 3 4 5 6 7 8 9 A B C D E F
00 86 80 C8 27 05 00 80 02 01 00 03 0C 00 00 80 00
10 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
20 01 E0 00 00 00 00 00 00 00 00 00 00 58 14 04 50
30 00 00 00 00 00 00 00 00 00 00 00 00 17 01 00 00
40 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
50 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
60 10 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
70 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
80 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
90 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
A0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
B0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
C0 00 27 00 00 00 00 00 00 00 00 01 00 00 00 00 00
D0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
E0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
F0 00 00 00 00 00 00 00 00 86 0F 01 00 00 00 00 00
USB Controller (UHCI)
bus 0 (0x00), device 29 (0x1D), function 1 (0x01)
Common header
Vendor ID 0x8086
Model ID 0x27C9
Revision ID 0x01
PI 0x00
SubClass 0x03
BaseClass 0x0C
Cache Line 0x00
Latency 0x00
Header 0x00
PCI header
Address 4 (port) 0x0000E100
Subvendor ID 0x1458
Subsystem ID 0x5004
Int. Line 0x13
Int. Pin 0x02
Dump
0 1 2 3 4 5 6 7 8 9 A B C D E F
00 86 80 C9 27 05 00 80 02 01 00 03 0C 00 00 00 00
10 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
20 01 E1 00 00 00 00 00 00 00 00 00 00 58 14 04 50
30 00 00 00 00 00 00 00 00 00 00 00 00 13 02 00 00
40 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
50 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
60 10 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
70 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
80 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
90 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
A0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
B0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
C0 00 27 00 00 00 00 00 00 00 00 01 00 00 00 00 00
D0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
E0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
F0 00 00 00 00 00 00 00 00 86 0F 01 00 00 00 00 00
USB Controller (UHCI)
bus 0 (0x00), device 29 (0x1D), function 2 (0x02)
Common header
Vendor ID 0x8086
Model ID 0x27CA
Revision ID 0x01
PI 0x00
SubClass 0x03
BaseClass 0x0C
Cache Line 0x00
Latency 0x00
Header 0x00
PCI header
Address 4 (port) 0x0000E200
Subvendor ID 0x1458
Subsystem ID 0x5004
Int. Line 0x12
Int. Pin 0x03
Dump
0 1 2 3 4 5 6 7 8 9 A B C D E F
00 86 80 CA 27 05 00 80 02 01 00 03 0C 00 00 00 00
10 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
20 01 E2 00 00 00 00 00 00 00 00 00 00 58 14 04 50
30 00 00 00 00 00 00 00 00 00 00 00 00 12 03 00 00
40 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
50 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
60 10 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
70 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
80 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
90 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
A0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
B0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
C0 00 27 00 00 00 00 00 00 00 00 01 00 00 00 00 00
D0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
E0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
F0 00 00 00 00 00 00 00 00 86 0F 01 00 00 00 00 00
USB Controller (UHCI)
bus 0 (0x00), device 29 (0x1D), function 3 (0x03)
Common header
Vendor ID 0x8086
Model ID 0x27CB
Revision ID 0x01
PI 0x00
SubClass 0x03
BaseClass 0x0C
Cache Line 0x00
Latency 0x00
Header 0x00
PCI header
Address 4 (port) 0x0000E300
Subvendor ID 0x1458
Subsystem ID 0x5004
Int. Line 0x10
Int. Pin 0x04
Dump
0 1 2 3 4 5 6 7 8 9 A B C D E F
00 86 80 CB 27 05 00 80 02 01 00 03 0C 00 00 00 00
10 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
20 01 E3 00 00 00 00 00 00 00 00 00 00 58 14 04 50
30 00 00 00 00 00 00 00 00 00 00 00 00 10 04 00 00
40 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
50 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
60 10 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
70 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
80 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
90 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
A0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
B0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
C0 00 27 00 00 00 00 00 00 00 00 01 00 00 00 00 00
D0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
E0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
F0 00 00 00 00 00 00 00 00 86 0F 01 00 00 00 00 00
USB 2.0 Controller (EHCI)
bus 0 (0x00), device 29 (0x1D), function 7 (0x07)
Common header
Vendor ID 0x8086
Model ID 0x27CC
Revision ID 0x01
PI 0x20
SubClass 0x03
BaseClass 0x0C
Cache Line 0x00
Latency 0x00
Header 0x00
PCI header
Address 0 (memory) 0xF0180000
Subvendor ID 0x1458
Subsystem ID 0x5006
Int. Line 0x17
Int. Pin 0x01
Capabilities
Power Management Capability
Offset 50h
Version 1.1
Dump
0 1 2 3 4 5 6 7 8 9 A B C D E F
00 86 80 CC 27 06 00 90 02 01 20 03 0C 00 00 00 00
10 00 00 18 F0 00 00 00 00 00 00 00 00 00 00 00 00
20 00 00 00 00 00 00 00 00 00 00 00 00 58 14 06 50
30 00 00 00 00 50 00 00 00 00 00 00 00 17 01 00 00
40 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
50 01 00 02 00 00 00 00 00 0A 00 A0 20 00 00 00 00
60 20 20 FF 01 00 00 00 00 01 00 00 00 00 00 00 C0
70 00 00 D5 3F 00 00 00 00 00 00 00 00 00 00 00 00
80 00 00 00 00 01 00 00 00 00 00 00 00 00 00 00 00
90 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
A0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
B0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
C0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
D0 00 00 00 00 00 AA FF 00 FF 00 FF 00 20 00 00 88
E0 00 00 00 00 DB B6 6D 00 00 00 00 00 00 00 00 00
F0 00 80 00 09 88 85 40 00 86 0F 01 00 86 17 02 20
PCI to PCI Bridge
bus 0 (0x00), device 30 (0x1E), function 0 (0x00)
Common header
Vendor ID 0x8086
Model ID 0x244E
Revision ID 0xE1
PI 0x01
SubClass 0x04
BaseClass 0x06
Cache Line 0x00
Latency 0x00
Header 0x01
PCI header
Primary bus 0x00
Secondary bus 0x02
Int. Line 0xFF
Int. Pin 0x00
Capabilities
Subsystem Vendor Capability
Offset 50h
Subsystem ID 0x5000
Sub. Vendor ID 0x1458
Dump
0 1 2 3 4 5 6 7 8 9 A B C D E F
00 86 80 4E 24 07 00 10 00 E1 01 04 06 00 00 01 00
10 00 00 00 00 00 00 00 00 00 02 02 20 D0 D0 80 22
20 F0 FF 00 00 F1 FF 01 00 00 00 00 00 00 00 00 00
30 00 00 00 00 50 00 00 00 00 00 00 00 FF 00 00 00
40 00 00 00 00 00 00 00 00 01 00 00 00 00 12 00 00
50 0D 00 00 00 58 14 00 50 00 00 00 00 00 00 00 00
60 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
70 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
80 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
90 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
A0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
B0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
C0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
D0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
E0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
F0 00 00 00 00 00 00 00 00 86 0F 01 00 00 00 00 00
PCI to ISA Bridge
bus 0 (0x00), device 31 (0x1F), function 0 (0x00)
Common header
Vendor ID 0x8086
Model ID 0x27B8
Revision ID 0x01
PI 0x00
SubClass 0x01
BaseClass 0x06
Cache Line 0x00
Latency 0x00
Header 0x80
PCI header
Subvendor ID 0x1458
Subsystem ID 0x5001
Int. Line 0x00
Int. Pin 0x00
Capabilities
Vendor Dependant Capability
Offset E0h
Dump
0 1 2 3 4 5 6 7 8 9 A B C D E F
00 86 80 B8 27 07 01 10 02 01 00 01 06 00 00 80 00
10 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
20 00 00 00 00 00 00 00 00 00 00 00 00 58 14 01 50
30 00 00 00 00 E0 00 00 00 00 00 00 00 00 00 00 00
40 01 04 00 00 80 00 00 00 81 04 00 00 10 00 00 00
50 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
60 8A 80 83 8B D0 00 00 00 85 80 80 87 00 00 00 00
70 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
80 00 00 08 34 01 08 0C 00 91 02 0C 00 00 00 00 00
90 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
A0 20 06 00 00 38 00 00 00 13 00 00 00 00 03 00 40
B0 00 00 F0 00 00 00 00 00 55 55 55 59 00 00 00 00
C0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
D0 33 22 11 00 67 45 00 00 80 C0 00 00 00 00 00 00
E0 09 00 0C 10 A8 00 24 00 00 00 00 00 00 00 00 00
F0 01 C0 D1 FE 00 00 00 00 86 0F 01 00 00 00 00 00
IDE Controller
bus 0 (0x00), device 31 (0x1F), function 2 (0x02)
Common header
Vendor ID 0x8086
Model ID 0x27C0
Revision ID 0x01
PI 0x8A
SubClass 0x01
BaseClass 0x01
Cache Line 0x00
Latency 0x00
Header 0x00
PCI header
Address 4 (port) 0x0000F000
Subvendor ID 0x1458
Subsystem ID 0xB002
Int. Line 0x00
Int. Pin 0x02
Capabilities
Power Management Capability
Offset 70h
Version 1.1
Dump
0 1 2 3 4 5 6 7 8 9 A B C D E F
00 86 80 C0 27 05 00 B0 02 01 8A 01 01 00 00 00 00
10 01 00 00 00 01 00 00 00 01 00 00 00 01 00 00 00
20 01 F0 00 00 00 00 00 00 00 00 00 00 58 14 02 B0
30 00 00 00 00 70 00 00 00 00 00 00 00 00 02 00 00
40 07 A3 00 80 00 00 00 00 01 00 01 00 00 00 00 00
50 00 00 00 00 00 10 00 00 00 00 00 00 00 00 00 00
60 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
70 01 00 02 40 00 00 00 00 00 00 00 00 00 00 00 00
80 05 70 00 00 00 00 00 00 00 00 00 00 00 00 00 00
90 00 00 1F 10 80 01 80 40 00 00 00 00 00 00 00 00
A0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
B0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
C0 00 00 00 00 05 00 00 00 00 00 00 00 00 00 00 00
D0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
E0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
F0 00 00 00 00 00 00 00 00 86 0F 01 00 00 00 00 00
SMBus Controller
bus 0 (0x00), device 31 (0x1F), function 3 (0x03)
Common header
Vendor ID 0x8086
Model ID 0x27DA
Revision ID 0x01
PI 0x00
SubClass 0x05
BaseClass 0x0C
Cache Line 0x00
Latency 0x00
Header 0x00
PCI header
Address 4 (port) 0x00000500
Subvendor ID 0x1458
Subsystem ID 0x5001
Int. Line 0x0B
Int. Pin 0x02
Dump
0 1 2 3 4 5 6 7 8 9 A B C D E F
00 86 80 DA 27 01 00 80 02 01 00 05 0C 00 00 00 00
10 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
20 01 05 00 00 00 00 00 00 00 00 00 00 58 14 01 50
30 00 00 00 00 00 00 00 00 00 00 00 00 0B 02 00 00
40 01 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
50 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
60 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
70 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
80 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
90 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
A0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
B0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
C0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
D0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
E0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
F0 00 00 00 00 00 00 00 00 86 0F 01 00 00 00 00 00
Audio device
bus 2 (0x02), device 0 (0x00), function 0 (0x00)
Common header
Vendor ID 0x1412
Model ID 0x1724
Revision ID 0x01
PI 0x00
SubClass 0x01
BaseClass 0x04
Cache Line 0x00
Latency 0x80
Header 0x00
PCI header
Address 0 (port) 0x0000D000
Address 1 (port) 0x0000D100
Subvendor ID 0x3031
Subsystem ID 0x4553
Int. Line 0x14
Int. Pin 0x01
Capabilities
Power Management Capability
Offset 80h
Version 1.0
Dump
0 1 2 3 4 5 6 7 8 9 A B C D E F
00 12 14 24 17 05 00 10 02 01 00 01 04 00 80 00 00
10 01 D0 00 00 01 D1 00 00 00 00 00 00 00 00 00 00
20 00 00 00 00 00 00 00 00 00 00 00 00 31 30 53 45
30 00 00 00 00 80 00 00 00 00 00 00 00 14 01 00 00
40 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
50 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
60 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
70 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
80 01 00 01 04 00 00 00 00 00 00 00 00 00 00 00 00
90 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
A0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
B0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
C0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
D0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
E0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
F0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
DMI
------------------------------------------------------------------------------
DMI BIOS
--------
vendor Award Software International, Inc.
version F6x
date 07/11/2008
DMI System Information
----------------------
manufacturer Gigabyte Technology Co., Ltd.
product G31M-S2L
version unknown
serial unknown
UUID 00000000-00000000-0000001D-7D7466AD
DMI Baseboard
-------------
vendor Gigabyte Technology Co., Ltd.
model G31M-S2L
revision unknown
serial unknown
DMI System Enclosure
--------------------
manufacturer Gigabyte Technology Co., Ltd.
chassis type Desktop
chassis serial unknown
DMI Processor
-------------
manufacturer Intel
model Intel(R) Core(TM)2 Duo CPU
clock speed 1020.0 MHz
FSB speed 170.0 MHz
multiplier 6.0x
DMI Memory Controller
---------------------
correction 8-bit parity
Max module size 1024 MBytes
DMI Memory Module
-----------------
designation A0
size 256 MBytes (single bank)
DMI Memory Module
-----------------
designation A1
DMI Port Connector
------------------
designation PRIMARY IDE (internal)
connector On Board IDE
DMI Port Connector
------------------
designation SECONDARY IDE (internal)
connector On Board IDE
DMI Port Connector
------------------
designation FDD (internal)
port type 8251 FIFO Compatible
connector On Board Floppy
DMI Port Connector
------------------
designation COM1 (internal)
port type Serial Port 16450
connector 9 Pin Dual Inline (pin 10 cut)
connector DB-9 male
DMI Port Connector
------------------
designation COM2 (internal)
port type Serial Port 16450
connector 9 Pin Dual Inline (pin 10 cut)
connector DB-9 male
DMI Port Connector
------------------
designation LPT1 (internal)
port type Parallel Port ECP/EPP
connector DB-25 female
connector DB-25 female
DMI Port Connector
------------------
designation Keyboard (internal)
port type Keyboard Port
connector PS/2
DMI Port Connector
------------------
designation PS/2 Mouse (internal)
designation Detected (external)
port type Mouse Port
connector PS/2
connector PS/2
DMI Port Connector
------------------
designation USB (internal)
port type USB
connector Access Bus (USB)
DMI Port Connector
------------------
designation USB (internal)
port type USB
connector Access Bus (USB)
DMI Extension Slot
------------------
designation PCI
type PCI
width 32 bits
populated yes
DMI Extension Slot
------------------
designation PCI
type PCI
width 32 bits
populated no
DMI Extension Slot
------------------
designation PCI
type PCI
width 32 bits
populated yes
DMI Extension Slot
------------------
designation PCI
type PCI
width 32 bits
populated yes
DMI Physical Memory Array
-------------------------
location Motherboard
usage System Memory
correction None
max capacity 2048 MBytes
max# of devices 2
DMI Memory Device
-----------------
designation A0
format DIMM
type unknown
total width 64 bits
data width 64 bits
size 256 MBytes
DMI Memory Device
-----------------
designation A1
format DIMM
type unknown
Software
------------------------------------------------------------------------------
Windows Version Microsoft Windows XP Professional Service Pack 2 (Build 2600)
DirectX Version 9.0c
Resources
------------------------------------------------------------------------------
Memory I/O Space, BA=0x00000000F0100000
Port I/O Space, BA=0xE400
Memory I/O Space, BA=0x00000000E0000000
Memory I/O Space, BA=0x00000000F0000000
Port I/O Space, BA=0xE000
Port I/O Space, BA=0xE100
Port I/O Space, BA=0xE200
Port I/O Space, BA=0xE300
Memory I/O Space, BA=0x00000000F0180000
Port I/O Space, BA=0xF000
Port I/O Space, BA=0x500
Port I/O Space, BA=0xD000
Port I/O Space, BA=0xD100
Memory I/O Space, BA=0x00000000D0000000, size=0x1000
Port I/O Space, BA=0x408, size=0x4
Memory I/O Space, BA=0x00000000FEE00000, size=0x1000
Memory I/O Space, BA=0x00000000FED14000, size=0x1000
Memory I/O Space, BA=0x00000000FED1C000
Port I/O Space, BA=0x480, size=0x3C
Port I/O Space, BA=0x290
Port I/O Space, BA=0x2E
phew!
I can strip out the register dump and repost or put in a spread sheet whatever is convenient. Want to communicate off line on this?
For the moment offline would be better - I have already ripped the various dumps out of the text - I will digest it tonight over a glass or two of wine.
Is this better
CPU-Z 1.49 report file
Processor(s)
Number of processors 1
Number of cores 2 per processor
Number of threads 2 per processor
Name Intel Core 2 Duo E7200
Code Name Wolfdale
Specification Intel(R) Core(TM)2 Duo CPU E7200 @ 2.53GHz
Package Socket 775 LGA
Family/Model/Stepping 6.7.6
Extended Family/Model 6.17
Core Stepping M0
Technology 45 nm
Core Speed 1020.2 MHz
Multiplier x Bus speed 6.0 x 170.0 MHz
Rated Bus speed 680.2 MHz
Stock frequency 2533 MHz
Instruction sets MMX, SSE, SSE2, SSE3, SSSE3, SSE4.1, EM64T
L1 Data cache (per processor) 2 x 32 KBytes, 8-way set associative, 64-byte line size
L1 Instruction cache (per processor) 2 x 32 KBytes, 8-way set associative, 64-byte line size
L2 cache (per processor) 3072 KBytes, 12-way set associative, 64-byte line size
Chipset & Memory
Northbridge Intel P35/G33/G31 rev. 10
Southbridge Intel 82801GB (ICH7/R) rev. A1
Memory Type DDR2
Memory Size 256 MBytes
Memory Frequency 170.0 MHz (1:1)
CAS# Latency (tCL) 3.0 clocks
RAS# to CAS# (tRCD) 3 clocks
RAS# Precharge (tRP) 3 clocks
Cycle Time (tRAS) 5 clocks
Command Rate (CR) 2T
System
System Manufacturer Gigabyte Technology Co., Ltd.
System Name G31M-S2L
System S/N
Mainboard Vendor Gigabyte Technology Co., Ltd.
Mainboard Model G31M-S2L
BIOS Vendor Award Software International, Inc.
BIOS Version F6x
BIOS Date 07/11/2008
Memory SPD
Module 1 DDR2, PC2-4300 (266 MHz), 256 MBytes, Kingston
Software
Windows Version Microsoft Windows XP Professional Service Pack 2 (Build 2600)
DirectX Version 9.0c
Gosh - that is much better - hopefully tomoorow I might have some thoughts
Alan
... that p4 voltage was running at 12.54 volts (thanks to Sondale for suggesting to look at cpuz dumps). And I can not adjust my 550 watt agi ps lower. So I put back in my old Agi 430 watt ps and adjusted p4 pot until 11.58 volts. This gave me a nice sq improvement(less edgy as others have reported). Don't think I have fixed my metallic overlay with 2.0bv19 yet. But with 2.0bv18 I get stability (wrt to metallic overlay) and now improved sq versus other ps at 12.54 volts.On a side bar I ran all day yesterday with 2.0b19 at 146 src and 88khz and no issues. So I think my 'metallic' problem is related to cpu load. Ihaven't decided whether 2.0b18 at 146 src and 192 is better or worse sq wise than 2.0b19 at 146 src and 88khz. With my old 430 watt agi ps sounding good I'm hoping that the antec (coming next week) sounds better and improves stability wrt metallic overlay. We'll see. If that doesn't work then its reload windows (which is in its own subset of hd independent of music data).
Edits: 02/12/09 02/12/09 02/12/09
theob
email up the creek, several successes - I downloaded cpu-z HW Monitor - seems more accurate - 12v line is at 12.8v, so I will have to sort it out.
I think I have found a way round the clicking noise!! Will let you know if this is definitely the case asap.
Going through your cpu-z dump allowed me to reset several Bios settings and improve the SQ - even managed to get cPlay running at 192/145.68.
So I need to get a better psu.
fantastic---just looking at comparitive data helps doesn't it. Where did you get cpuz hw monitor?
Edits: 02/12/09
same place as cpuid - just google cpu-z monitor - it offers two versions 32-bit & 64-bit. It allows saving of previous details - I have not tried comparing results taken at different times.
got it thanks but I don't get all the voltage data. how did u get it? do u have to set up parameters or something in that other file that comes with cpuz hw monitor?
It only shows a limited amount of data - but enough to confirm the cpu-z was not measuring the voltages correctly. You probably have to buy the PRO version to get more data!
I have not tried messing about with the file.
ok I will (when I get home) dump all 3 screens of cpuz and put in a file
Hi all. I recently finally bought an Antec Earthwatts 430 PSU and I'm having stability issues and I was wondering if anyone else is experiencing the same things.First off, after letting it play for a few days (as is) I started playing with the 12v pot inside the case (as others have suggested). It initially measured 12.12 (on my multimeter) so I first lowered it to 12.00 (and I did indeed notice a difference in SQ). So a little later I decided to see how low I could go and I got it to go down to 11.90 and again the SQ improved very nicely. So I let it just play the same album for awhile. Then I was able to switch playlists a couple times, but one time I was scrolling down through the list (holding the arrow key down) and then the PSU just shuts off. This has been repeatable. Now the other bugger is that I can't even boot up with the pot set to 11.90. I have to set it to at least 12.20 now in order to get it to boot up. Then I have to let it play a little while before I can adjust it back down to 11.90.
Second, Rick had mentioned that he was able to disconnect the fan. When I disconnect the fan it will not boot up. Even with the 12v pot set to 12.2. What's the secret?
So all those using the Antec Earthwatts 430 PSU, can you please share your experiences and stories and whether or not you've been able to accomplish what I've tried (while maintaining stability).
Thanks.
-edward
Edits: 02/08/09
edward wrote:Now the other bugger is that I can't even boot up with the pot set to 11.90.
It might just be that the cause of your anomaly is your processor, not your PSU. Overclockers (not my scene but their forums are often well informed) report quite wide tolerances between devices.
I use a linear supply for the processor chip on my PC audio system as it makes a marked difference in SQ. If I lower the voltage from 12 to 10.5 volts, it still works (and boots) fine.
However, as the change, to my ears at least, neither improves sound quality nor (to my surprise) reduces power draw, I now leave it at 12 volts.
Might I suggest that you try adding around 10,000 uF of smoothing to the P4 12v line and set the supply at 12 volts? My hunch is that your system would sound better and be less sensitive to voltage levels.
It will certainly do no harm and costs little. Two list members who repeated the mod report positive results.
Just a suggestion,
Dave
Edits: 02/08/09 02/08/09
First off, let me just thank Dave for the inspiration to improve the power delivery, which in turn also seems to have been a solution to my situation. I hope you all find this as enlightening as it has been to me.
FWIW this setup is on the Biostar MB and the E2160 CPU.
With Dave's help I managed to solder a board of capacitors together (for the P4 connector) and I had my first smile of joy. The improvement was very nice indeed. So after I let it play for awhile I lowered the 12v pot to 11.90 and let it play all day (to burn in a little and warm up). Then I decided to shut it down and see if it would boot back up (with the lower voltage). Lo and behold it booted up fine. At that point I had my second smile of joy.
The next step was to shut it down for the night and see how it fared the next morning. Well, the next morning it would not boot up. I had to increase the 12v pot once again. Frown. I repeated the exercise of the day before, and once again I was able to get it to boot up fine (with lowered voltage) but only after sufficiently warming up. I was feeling ambitious so I then tried shutting it down and booting up with the fan disconnected. No go. Frown again.
After I thought about it for awhile, I thought of Dave running the P4 from a separate PSU than the P24, so I thought maybe in here lies the solution. I had tried running separate PSUs a long time ago and never got it to work, so I never thought of it again until now. Now I realize (with a tip from Dave) that you need to jumper the green and black pins on the P24 to get it to power on. Doh. So I gave it a go with the Antec (and caps) on the P4 and my spare Magnum on the P24. Success! My third smile of joy. And another nice improvement in SQ. Bonus.
So again I shut it down for the night to see if it would still work on a cold start the next morning. I left the voltage as it was and the next morning it booted up just fine! The solution! And scrolling around in cMP did not crash it. So ambition in place again, I shut down and now checked to see if I could boot up without the fan. Success again. My fourth smile of joy!
Just for good measure I tried it once again the next morning from a cold start and it still works!
So the moral of the story is that the PSU wasn't the issue after all. I think the interaction with the MB was being the bugger.
Interesting side note: with the Antec being the only PSU, I set the 12v pot to 11.90. Without changing the pot and using the Antec only on the P4, the voltage was now measuring 11.76. I'm not sure what this means, but I just re-adjusted back to 11.90 and left it there.
SQ has taken a nice big step forward. But ambition being what it is, I'm now ever more curious about performing the "cap mod" on the P24, as well as powering the P4 from a linear PS and moving the Antec over to the P24. We'll see what the future holds.
Thanks for listening. I hope someone out there finds this story useful.
Fan's BLDC motor has a magnet that causes interference.
Also, its a good idea to remove those cable ties.
Thanks.
I have a question on your 'installing another ps for p24---jumpering the green and black pins' where exactly are these green / black pins? On the mobo or the p24 connector?
Could you say a little more on this please?
Theo - I'm referring to the green and black wires on the P24 cable (coming out of the PSU). At the connector, take a little wire and insert one side into the green and the other side into any closest black (into the holes of the connector on the side that would touch the MB). This is the "on switch". So make sure when you attach this wire that the PSU switch is OFF. As soon as you switch it on it will boot up.
And just to be clear, this is done on the PSU that is only going to power the P4 (CPU).
thanks
one more analogy rick (and anybody who might be reading) listening to 192 thru bnc is akin to good solid state amplification. listening to 96 thru glass toslink is akin to good tube amplification.I listened yesterday for 14 hours thru 96/toslink and rediscovered many of my favorite files (mostly Mercury Detroit Symphony Paul Paray). It was just breathtaking: the ambiance, air, detail w/o smugding (thanks to 2.0b vers19 cplay) especially way back in the orchestra rears.
you gotta try and give your self a day to reaquaint your self to the new sonics, you will be rewarded I believe.
Edits: 02/16/09
What more needs to be said?
I would greatly appreciate hearing how you have done this.
Thanks,
Rick McInnis
Rick McInnis wrote:
What more needs to be said?
Not a lot - it's easy to do. See pic. Note (a) that I've yet to tidy things up (b) that someday soon I will and (c) that the Power-One unit shown (which has a 5.1 amp capacity but was to hand) is much bigger than necessary.
As I've noted before, the E7200 processor chip draws (from memory) ~1.4 amps on power up, falling to about 0.7 amps as underclocking kicks in.
John Swenson's recent design should therefore work a treat here unless you wanted to run the CPU continuously at high speeds where it might need to be beefed up slightly.
The Power-One's 1.7 amp (i.e. one transistor) version of this design would also be fine.
If I were not so lazy, I'd run voltage sense lines from the unit to the MoBo side of the added capacitors as the Power-One design allows for them.
I've not tried the linear PSU without the added caps as I wanted it to be easy to switch from the Corsair to the Power-One box.
NB that, if switching supplies over, you must leave the units off for a couple of minutes first. If you don't, the spark is "interesting".
The difference in sound quality is readily perceptible when driving a NOS DAC but I've not tried upsampling.
Hope that helps,
Dave
Hello Dave,
I just ordered two pairs of 2200uF/16V Sikorels and ten 1000uF/16V Panasonic FMs. Hopefully the caps will arrive by Monday, and - with the help of a friend who knows how to solder - I'll be able to apply the mod to my system, and post a brief report.
Being a typical Type A person, I already have my eyes set on the advanced linear psu "mod", even though I'm still waiting for the caps to arrive :P
I don't know how to build a linear PSU, but I can pay a local DIYer to make one for me, as long as I can give him a good idea of what I'm trying to do (which I can't as of now). Could you please enlighten me with a bit more insight on building/modding a linear PSU to supply the CPU? A detailed description, a photo, and/or a simple diagram would be greatly appreciated.
Thank you,
Joy
Joy wrote:
Could you please enlighten me with a bit more insight on building/modding a linear PSU to supply the CPU?
Sorry, I can't add much to what's already on the thread. (Please note that I make no claim to being the first to try this idea though I'm sorry to say that I don't recall who I pinched it from.)
I've already discussed the choice of PSU at (tedious) length and posted a picture of the (over-specified) unit I'm using earlier in the thread.
You could also use John Swenson's recent design if you modify it for 12 volts and possibly beef up the heatsinking. Any competent electronics DIYer should know how to do that. It's certainly a good design.
However, it might be simpler and cheaper to use an off-the-shelf unit such as a smaller version of the Power-One unit shown in the picture. John's design might sound sufficiently better to justify the extra work and cost but I cannot say. AFAIK, no one has yet compared the two in this context.
WARNING I have tried this mod only twice. First, briefly, I did a go/no go on an elderly Athlon 64 motherboard (which I could better afford to lose if it all went pear-shaped) and, second, on a Socket 775 board with an E7200 chip underclocked to 840 MHz.
I have not tested it for more than an hour when running the 45nm technology E7200 flat out (it was fine) and I have not tested it on more powerful chips.
I suspect that if you carefully measure the current drawn by an alternative processor and use a sufficiently powerful PSU, it would be fine but I do not know for sure.
As I said before, running cPlay to the likes of JS's "satellite" player seems the next stage for me so that's about as far as I'm going for now.
Note that if you buy or build a 12 volt PSU to run the processor, you could use it to power a Fit-PC at a later stage so you have little to lose.
Hope that helps. Meanwhile, I look forward to hearing how you get on with the first stage of the "upgrade".
Dave
Dave,
I am assuming you are using the "standard" supply for the other voltages?
Do you just turn both of the supplies on at the same time?
Do you use the linear supply for all of the 12 volts rails or just he "four pin" one?
I was thinking one might need to use the switcher to activate a relay to "turn on" the 12 volts?
Hope you do not mind my inquisition!
Thanks,
Rick McInnis
Rick wrote:I am assuming you are using the "standard" supply for the other voltages?
Yes - see the Corsair box at the top of the picture. It is connected to the ATX-24 lines via a little paddleboard with added smoothing caps.
Do you use the linear supply for all of the 12 volts rails or just the "four pin" one?
Just the four-pin one, the so-called P4-12 line. I believe that the power draw on the on-board 12v lines is pretty small though I've not bothered to measure it.
I had envisaged making a linear supply for the ATX-24 lines as well but now feel that something on the lines of John Swenson's "satellite" player is the more attractive route in the longer term so I'm leaving it "as is" for now. This arrangement is easy to implement but effective.
BTW, a fellow inmate has devised a more elegant way of adding smoothing to the ATX-24 lines. He has promised to post details in due course.
Do you just turn both of the supplies on at the same time?
I turn on the 12v linear supply first, then the mains supply for the computer and then the computer's on/off switch.
That said, it doesn't seem to matter what sequence I use. Note that the current draw on the linear supply when the Corsair is switched off is too low to measure.
Hope that helps,
Dave
Edits: 02/09/09
Thanks very much for being one of the intrepid!!!
I will start with the 4 pin.
I wish someone had a map of how power is distributed in MB's and even more how it is distributed within the GIGABYTE board many of us are using.
Thanks very much for your help,
Rick McInnis
Yes indeed Dave, I finally bought some caps and P4 connector and I will be putting that board together this week. We'll see if that changes the stability.
My CPU is actually underclocked and running at 600MHz. (It's the E2160) You think this makes a difference?
I am sorry to hear your troubles. I think there is something wrong with your supply.
Mine has been trouble-free.
I have had plenty of glitches while using cics's system, mainly with the MB's, but through it all the ANTEC has never flinched. I have had it for close to a year now. Still waiting to get my rebate check. As an aside, I waited six months for the check, put it in the bank and it comes back saying PAYMENT STOPPED. I had never heard of such a thing, but then I have rarely bought anything that has a rebate offer. Went to GOOGLE and found I was not alone and how to get it fixed. Still waiting for check number two, though. It was well worth the price without the rebate.
These things do happen. I would say send it back though I know you have spoiled the little sticker and might not be able to make your supply look virginal. Lucky the things are ridiculously cheap. All of this computer stuff is compared to "audio" components. Nothing like massive production runs to make our toys cheap.
Bye,
Rick McInnis
Hi Rick,
According to your own experience, at what voltage does your Antec offer the best SQ? Do you set it close to 12V, or as low as possible? Mine is currently set to the lowest possible (around 11.4 as shown in BIOS), and while it sounds better than default, I'm not sure that it's set at the optimal level.
Thank you in advance!
Cheers,
Alex
though I know there is always a best level I have not done this.
I simply turned it down all the way and liked what I heard.
One of these days.
Sorry,
Rick McInnis
I turned it down all the way as well, but the voltage of my 12V rail seemed to decrease significantly more (11.4V) compared to what Bernd reported (11.9V), and I remember him saying that SQ started to deteriorate when the voltage drops below 11.9. Can you please check the BIOS to confirm the voltage of your 12V?
Thanks Rick!
Alex
My, agreed, unreliable BIOS voltage reading is barely below 12 volts which would seem to be suspect.
This weekend I will simply measure it at the 4 pins connector.
I will let you know.
Rick McInnis
Alex - I wouldn't rely on the BIOS voltage reading. Depending on the MB it can be very inaccurate. The more reliable way is to measure it with a digital multimeter. For example, when I get a reading of exactly 12.00v with my multimeter, it shows around 11.85 in the BIOS.
Also, BTW, my pot doesn't stop after a quarter (or even half) turn like Bernd suggested. It seems to keep turning. But if I go below 11.85 it just shuts off.
Rick - so are you indeed running with the fan disconnected? And what do you have your 12v pot set to?
and pot set to its lowest point.
Just unplugged the fan. Did not make any difference to my unit.
bye,
Rick McInnis
Hi rick do you have a special cooler unit on your cpu, if so which one?
Theo,
It might have to do with the fact that I have all of this in a big box.
No heating at all. It is usually on ALL of the time unless the power gets cut off during the day for some reason out of my control.
I was nervous initially. Case is warm, that's all.
Even in the summer I had no problem.
Rick McInnis
I wonder if my unit being "newer" has anything to do with my "issues".
Anyone else out there that bought their Antec within the past month having better success than I am?
...I'll let you know how it works out for me. However in an effort to try to tweak the p4 voltage I put in another agi ps with an adjustable pot on it. I initially saw cpuz voltage for p4 at 13.6 volts. I adjusted all the way to the other stop and it would not boot. So I tweaked it until I got 11.58 on cpuz and it sounds pretty good (better than my previous ps at 12.54). So any way I'm leaving it there until I get my antec. I know the cpuz levels are not accurate but close enough until I get the antec.
got the antec, did the ryeland's p4 cap modification and the system sounds great. I adjusted the antec pot all the way counterclockwise and I get 12.19 on cpuz but 11.82 with my vom. System boots ok, sounds great. So no problems here. I'm running cplay @96. I can get up to 176 w/o my metallic issue but, alas, 192 is still a problem with 2.0b19. I guess sooner or later I will do a windows reload but for now I'm listening to & enjoying music.
Refer to ReadMe.txt for important details & specifications.
Change Log (2.0b19 release):
- DSP refinement: Shift workloads around to enable maximum ASIO output efficiency (improvements in ASIO CPU load and output latency is achieved)
New DSP components is used to achieve this complex change
- Changed "INDEX00_GAPLESS" to "INDEX00_IGNORE". When added to INI file, INDEX00 entries are ignored (pregap is added to previous track)
Please REMOVE previous versions before installing cPlay 2.0b19. Normal (SSE2), SSSE3 and SSSE3 B9 versions are available. Running cPlay on a CPU not supporting the required instruction set will cause cPlay to exit immediately.
For cMP² users, review these BIOS changes (underclocks FSB, CPU & disable additional USB items). New settings are in effect to best exploit above capabilites. Recommended Gigabyte mobo GA-G31M-S2L remains available.
I really hate to say this. I loaded 19 and initially it sounded wonderful as Steppe says below. Guess what came back---the dreaded hard metallic overlay. Not even 20 minutes with 19 and there it was again. I went 6 days (some of them 24 hours) and no hint of hardness let alone the dreaded metallic overlay. Now its back and so I reloaded 18 and will listen to that until I can figure something out. I have a new juli@ card coming this week so maybe that is the problem.
Sorry cics to bring some bad news to the great sounding version 2.0bv19. I sure hope the new juli@ will solve this. So much for my theory on the Hammond 193L inductor.
Theob,
Just to let you know, I'm not having any issues with a 193L in my MoBo AC line.
In addition to your sound card, you also might want to try another power supply. If you don't have one... and you're in the US and willing to ship it both ways, I have an extra Earthwatt I'd be willing to lend.
Greg in Mississippi
P.S. And my initial listening agree with the comments on 2.0B19... more natural, more body.
I now know it was not the 193L causing my problem. I have a new juli@ coming this week which I will try. If that doesn't work I'll try the ps. I have an AGI 530 watt unit. Is your spare an Antec, if so I'm interested.
It's strange I've gone all week with 2.0bv18 and not one iota of a problem. 20 minutes with 2.0bv19 and I got the dreaded metallics. So I'm back into 18 for regular listening.
My spare is an Antec Earthwatt 430 watt. I keep meaning to put it into my ripping machine to burn it in a bit just in case I need it... it's yours to try if you'd like.
Let me know after you've tried the 2nd Juli@
Greg in Mississippi
thanks for your gracious offer however I always wanted an antec 430 so I just ordered one. new soundcard didn't help. hopefully new ps will.
Very welcome. I hope the new PSU solves your issue! Motherboard or processor are the other two remaining options and they are more expensive... in time if not in cost.
This type of thing (and Rick's experience with Gigabyte motherboards) was part of why I got many spares when I started gathering all the parts to do a cMP-Squared. That and what might happen when I get deeper into modifying things.
Greg in Mississippi
ok
Hello, everybody! It was not more than a fortnight ago, that Cics said it would be hard to improve the sound of 2-17, but he had done it with 2-18 and he did it now. This time the improvement is very important (to me) - it's all about tonality of the instruments and voices. It is as if we start eating a steak which is a steak, not steam-cooked piece of beef.
The dynamism improved, attack is even faster and with more details than ever. Though improvement is most notable (on my humble setup) in mids and mids-to-highs, I believe that it is lower region that is cleaner now, and it allows instruments and voices to breathe.
I should call it a MATURE version of cPlay. It really sounds mature.
Thank you, Cics for another fine weekend.
Serge.
i have tried to load a cuesheet file into cplay by cMP.
the playlist was shown on the cplay screen but the song could be played... no response when i clicked the play button.
same problem happened when i tried to use cplay in read time mode.
cics or any person can tell me what the problem is and how to solve it?
also how to use cMP to load cue sheet files into wavelab?
many thanks
Ken
Hong Kong
i have tried to load a cuesheet file into cplay by cMP.
the playlist was shown on the cplay screen but the song could be played... no response when i clicked the play button.
same problem happened when i tried to use cplay in read time mode.
cics or any person can tell me what the problem is and how to solve it?
also how to use cMP to load cue sheet files into wavelab?
many thanks
Ken
Hong Kong
.
cics,
i don't have sound card.
i only used eeepc as source with asioall, then via usb cable to feed benchmark dac.
i have no problem to use cplay and wavelab but problem come out when adding cMP or running cplay in real time mode.
Ken, HK
.
use "Player" setting in single core...
i don't understand. can you elaborate?
many thanks
Ken
.
cics,
i have set "Player" in setting of cMP.
playlist was loaded into cplay but still no response when i click the play button.
.
did you toggle your input selector on the benchmark to get the usb input? I know this is trivial but just meant as a kind suggestion.
nt
If you are using a USB-connected DAC, check that Windows Audio Service runs after cMP is launched - some USB DACs need it.
Dave
Hello, Cics and everybody. My stupid question is as follows: In the cPlay readme file I saw that its instruction set is sse2. You wrote that sse2 processors will no longer be supported. At present sse4 is announced in latest Intel's processors. Will it somehow affect the future development of cPlay. What this instruction set does? Another question is: Does volume control, built inside cPlay affect the resolution and, if Yes, from what figures?
Sorry for taking Your time.
Serge.
From 2.0B8, SSE2 or higher is only supported.
I haven't seen anything that would benefit audio when using SSE4 (as yet...).
VC is applied /calculated using 53 bit resolution (double precision). The limiting factor for resolution is the input media (16, 24 or 32 bit). SRC upsampling calculations are also done using 53 bit resolution (with coefficients at 24 bits resolution).
I finally took the plunge and changed my soundcard to the Juli@; after much bad language (I could not get it to work at all) I found cics suggestion about removing and resetting Juli@ drivers/devices - worked a treat.
So what did it sound like - OK (first 15 minutes), Perhaps it was worth the change (30 minutes), Oh now I see what everyone is on about (1 hour).
I have set cPlay / Juli@ to 96k the same as my E-MU, once settled in I will up it to 192k. I am using the unbalanced analogue outputs for the moment - I will wait for a good (also not expensive) 24/96 USB to come out before going over to the usb output.
I have installed 2.0b18 and I like the sound - however one problem that I have had recently still has not gone away.
Nearly all my files are FLAC and when cPlay loads the next track I get a popping sound at the beginning of most tracks. cics suggested upping the cpu frequency - I am now at 1.66GHz and the noise persists. I have created some WAV files and there is no problem with the sound - however the only info I get in cPlay is track number and time.
I have tried various things to get a cue file but I do not think that is problem - it is the tags. I would prefer to keep the FLAC files - so any suggestion as how to get rid of the popping sound would be welcome; otherwise I will have to sort out my WAV problems.
However - since the popping problem did not exist in the versions which loaded everything into RAM before starting I was wondering about two ways of resolving the issue.
First would be to go back to the last version which loaded all tracks (or as many as it could) - this I would not really want to do as I think that even in my short experience of cMP2 it has improved a lot. The other alternative would be to have loading switchable, i.e. there would be a choice between load a track or load an album.
Any thoughts on the feasability of this cics?
One thing I am sure of is that this has totally changed my view of digital-based music.
Alan
It's a problem related to FLAC decoding being CPU intensive.
Unfortunately, its not possible to use the version 1.x way.
Try increasing latency to 64+ samples.
Refer to ReadMe.txt for important details & specifications.
Change Log (2.0b18 release):
- DSP refinements:
- Reduce low level Windows overheads
- Max cue tracks supported increased to 120 (from 80)
- Compiler optimisation done
- Added "INDEX00_GAPLESS" option for cicsPlay.ini file. When added to INI file, INDEX00 is processed without gaps (i.e. previous track ends at INDEX00 and next track starts here).
- Other minor changes
Please REMOVE previous versions before installing cPlay 2.0b18. Normal (SSE2), SSSE3 and SSSE3 B9 versions are available. Running cPlay on a CPU not supporting the required instruction set will cause cPlay to exit immediately.
For cMP² users, review these BIOS changes (underclocks FSB, CPU & disable additional USB items). Recommended Gigabyte mobo GA-G31M-S2L remains available.
Hi cics
Thanks for working again on the "INDEX 00" issue in cPlay. After playing with v2.0b18 a while, I must speak up again, with a change request.
This time - to express it in a more concise way - let me use a Program Requirements approach, more like the specifications you might receive in an IT development shop.
ASSUMPTIONS :
a) this issue affects playback under the following conditions:
- cPlay is involked in CUESHEET mode
- the "album" has been ripped as a single-file CDImage (WAV or FLAC)
- the accompanying CUE file contains "INDEX 00" entries for tracks other than TRACK 01 (typically created using EAC)
b) The accompanying CUE file conforms to standard cdrwin syntax, as defined here -> http://digitalx.org/cuesheetsyntax.php (the one exception, being the cPlay-specific "EOC" command, introduced in v2.0b15).
c) "INDEX 00" specifies the starting time of the track "pregap" (see digitalx.org).
d) "INDEX 01" specifies the starting time of the track data (see digitalx.org).
e) Playback of a CDImage file will occur in the same way as a hardware player plays a CD. No part of the music file will be skipped or omitted, unless the user operates the transport controls to change track order.
f) cPlay transport controls will position to "INDEX 01" of the next or previous track when operated.
g) Individual track duration (displayed for each track in the cPlay window) will be calculated from "INDEX 01" of the current track, to "INDEX 01" of the next sequential track.
OBSERVATIONS :
cPlay's new "INDEX00_GAPLESS" option does not fix the "INDEX 00" Handling issue reported in the past.
Activating the option in cicsPlay.ini file causes track times to display inaccurately in the track list window. To experience this yourself, edit a CUE file such that an "INDEX 00" value is 2:00 less than originally specified. Play the two affected tracks with/ and without the "INDEX00_GAPLESS" option, and note the track times displayed.
cics will know how this is happening.
CHANGE REQUIREMENTS :
a) BYPASS all "INDEX 00" commands in the CUE file. When parsing the file, treat these commands as if they did not exist.
b) Calculate track times from "INDEX 01" of the current track to "INDEX 01" of the next sequential track.
With the greatest respect - cics, please make these changes to v2.0b19. I can think of no reason for cPlay to refer to the "pregap" track position.
Ever. It just doesn't apply to what we're doing with playback from a computer.
...and I can't wait to get off this particular soapbox :)
Thanks again for all the wonderful music in my home !
Cheers,
Grant
That's not a Toy... IT'S A TOOL !!
Here's the official INDEX writeup:
Description:
This command is used to specify indexes (or subindexes) within a track.
Syntax:
INDEX [number] [mm:ss:ff]
Parameters:
number - Index number (0-99)
mm:ss:ff - Starting time in minutes, seconds, and frames (75 frames/second).
Note:
All times are relative to the beginning of the current file.
Example:
INDEX 01 00:00:00
INDEX 05 02:34:50
Rules:
All index numbers must be between 0 and 99 inclusive. The first index must be 0 or 1 with all other indexes being sequential to the first one. The first index of a file must start at 00:00:00.
INDEX 0 Specifies the starting time of the track "pregap".
INDEX 1 Specifies the starting time of the track data. This is the only index that is stored in the disc's table-of-contents.
INDEX > 1 Specifies a subindex within a track.
This is the default action for cPlay. That is INDEX00 is a track's pregap whilst INDEX01 is the actual start. This also infers that the previous track ends at INDEX00. Hence, there will be parts of the CD data that is ignored. Maybe I'm missing something?
Hi cicsIt seems this may be the underlying issue.
In many cases, INDEX00 does not define the end of the previous track - there is valid audio in the "pregap".
In CUE file mode, cPlay skips over the audio BETWEEN the INDEX00 and INDEX01 points (ie: the "pregap" space).
Over the years, CD mastering engineers have used INDEX00 in a way that differs from the original Redbook intent, allowing them some control of what the listener hears. This has given us wildly varying "pregaps" on CDs... depending on what the producers are trying to achieve.
As an example, consider a live recording (continuous audio) - where the INDEX00 points are used creatively - where there is applause from the previous song and introductory remarks before the band plays the next... In this case, cPlay skips over the pregap (which can quite often being 10 to 60 seconds long), and we miss that section of the show completely. cPlay should not do this.
One reason for using INDEX00 creatively this way is the "random playback" feature on hardware players (think: CD changers). In "programmed play mode", the machine will play the current track up to INDEX00 of the next track, and then begin playback at INDEX01 of whichever track is next in the program (not necessarily the next sequential track on the disc).
For our live album example, it means you skip over the talking bits and the next selection starts immediately with music. This is exactly what you want at your party - no quiet talk talk stuff to spoil the flow of music.
HOWEVER, when the same player is not in programmed mode (ie: continuous sequential play of the entire CD) it plays "right through" the pregap sections, as if the pregap was part of the previous track. (and it usually is - artistically speaking).
In practical terms (and for cPlay specifically), a TRACK should be defined as the audio that spans INDEX01-to-next-INDEX01 points. That's it. IGNORE INDEX00 completely and we'll get to hear all the music.
TRACK DURATION should also be calculated in the same way.
I hope that's clear... if not I'll try again.
Thanks for caring.
Cheers,
GrantThat's not a Toy... IT'S A TOOL !!
Edits: 02/04/09 02/04/09 02/04/09
Interesting point from this is how to interpret the pregap (INDEX00 to INDEX01). Is it part of the previous track (as in a post gap)? My interpretation is its not.
Ignacio has this view .
Any other views?
Hi cics (...and other Affected Parties)
Interesting point from this is how to interpret the pregap (INDEX00 to INDEX01). Is it part of the previous track (as in a post gap)? My interpretation is its not.
The real-world answer is "It depends...". It depends on what "they" did when the CD was mastered.
The only thing you can count on, is the inconsistent use of INDEX00 commands in CD mastering
-OR-
"the great thing about Standards is... there's so many of them "
Course of Action
Ignacio and I see eye-to-eye on this one... Ignore INDEX00 completely, and abandon the concept of pregap.
To expand on the idea...
Who are the Affected Parties
Any cPlay user who answers "Yes" to ALL of the following conditions:
- plays file by the CUESHEET method
- the file is ripped as a whole-CD image (ie: all tracks in one big file) - usually by EAC
- the CUE file contains INDEX00 commands
- the INDEX00 commands exist for any track OTHER THAN Track 01
If this is not YOU, then the issue should be of no concern.
If THIS IS YOU... then you may notice something when listening continuously through your album. It depends on a few other factors:
- was the CD mastered with "long" breaks between songs, or do the songs segue directly from one to the next ?
- how long are the pregaps for each track ?
Why Does This Matter
Currently, cPlay skips over the pregap information (never mind the INDEX_gapless option). If there was something there ... you just missed it. Even if there's nothing there (ie: silence), you missed the message as the producers intended you to hear.
Many albums are "sequenced" with great care and attention paid to the timing between songs... tempo, long fade/short fade, mood of adjacent songs... all this goes into deciding "when" the next song should begin. To play the album without these bits, is to listen without fidelity to the original.
While it may be an onerous exercise, a study of such EAC-generated CUE files will be instructive.
You'll discover CDs where not every track has INDEX00 defined - only some of the tracks on the disc. (What's up with that ?)
You'll find the pregap length varies widely - from the Redbook standard of exactly 2 seconds - to much longer (as noted in the live recording example from the previous post).
I freely admit that some CDs are really no more than a collection of tunes - where inter-song timing has no relevance. Ripping these to one-song-per-file format does no artistic harm. However, many of the albums I enjoy are complete works of art - from beginning to end - and I appreciate the skill and talent that goes into creating such a whole (...and for this reason I prefer to rip albums to a single file).
An Impractical Workaround...
...would be to manually edit all such CUE files. Delete or REM-out the offending INDEX00 commands. I was forced to do this on some prized albums I could not live without... but it's wrong, wrong, wrong. The player should play the album as recorded by the artist - without jiggery-pokery required from the user.
In conclusion :
Please ignore INDEX00 commands completely, abandon the concept of pregap... and all will be well with cPlay (on this issue at least !)
Thanks for reading...
Cheers,
Grant
That's not a Toy... IT'S A TOOL !!
"Ignacio and I see eye-to-eye on this one... Ignore INDEX00 completely, and abandon the concept of pregap."
Yes, I agree.
For my cPlay use, I usually delete the INDEX00 lines in the cue files.
Cics, I have now some free time to development. I could help with the interface. Please, send me a message.
Regards.
Ignacio.
"INDEX00_GAPLESS" will change to "INDEX00_IGNORE" in B19. This will conform to pregap being part of previous track.
Development:
Definitely need a playlist tool. Something that allows one to build and manage playlists (select tracks from cue files, files, etc.) and do playback. You mentioned something about doing this already. UI to DSP interface is very simple (separate components). Cue Sheet with new EOC directive makes for a perfect playlist spec.
Use WTL for UI and Intel's ICC compiler (10.1 or higher). Any development studio will do. Plan is to stop after B19 then do the dreaded D stuff (documentation) and packaging for final release.
and add more context.
As you know I am keen on playlist capability, so I am pleased to see you mentioning it and giving it the priority I also think it deserves.
It appears to my ears, and from your comments also, that cPlay has reached a degree of perfection, and that now you are able to consolidate, resting briefly on well-deserved laurels.
My question then focuses on cMP - will you be interested in giving playlist capabilities to cMP so that any media player used within it will benefit - I would like to see this, and it would make cMP a more widely used vehicle. At the moment the playlist capabilities of media players used in cMP are to a great extent thwarted by the limits on Windows functionality they depend on (quite rightly for the sake of the sound quality). My immediate request was for the ability to RAM load full-address cuesheets from cMP, which I cannot do at present. The full benefits of cMP would then be available for any media player when using full-address playlists, which is what is necessary at the moment. Your comments will be much appreciated.
now that I've tried the Recursive Cuesheet Creator, a java program that can create full address cuesheets from your music library folders written and offered to us by Aljordan, I see that this would be easy to take advantage of. His program full path cuesheeted by 560 album library in a minute or two.
You can remap the default "Refresh" button to this "Playlist" utility. Another alternative would be to set "RIP" to playlist utility (that way cMP waits until you exit and automatically does a refresh showing newly created cue sheets...).
What about the point about cMP ram-loading via full path cuesheets when using other players?
I see that as for cuesheet maker it could be used as you say though it is most useful for converting one's whole collection of music files to cuesheets at once, so doesn't have to be available every time one uses cMP.
This gets worse when using single WAV's with EOC directives in cue sheet.
Each GB takes ~20 seconds.
Anyone who wants to use Winamp, or such like, with cMP will be using more RAM anyway - 1GB would be normal. My experience is that a wav playlist or album of normal size (~500MB) RAM loaded to Winamp from a normal cuesheet in cMP takes about 10 seconds to load up. Cuesheets with the EOC directive would only be used with cPlay, so are not an issue with any other player. Any playlist for a player other than cPlay would be made up from multiple-file albums and cuesheets only.
There is no problem RAM loading into Winamp from cMP using normal cuesheets, located in the same folder as the music files. Why would a full-address cuesheet create additional RAM loading problems compared to a normal cuesheet? It's the same amount of data, and finding the files takes the computer no time at all.
From the point of view of cPlay, RAM-loading from cMP would of course be a nonsense, as cPlay has its own very sophisticated RAM-loading methods. (Btw, I haven't noticed any problems at all with EOC tracks in my playlists for cPlay - they work great.) But for other players, RAM-loading the whole cuesheet, while primitive compared to cPlay, is a large improvement in sound quality compared to playing from the HDD with cMP (which is still better than playing outside cMP.)
A cMP upgrade to handle full-address cuesheets would allow other media players used within cMP to benefit from RAM-loaded playlists, not possible at present. In spite of cPlay's superior SQ (imo), enhanced by less RAM, there are reasons to use other players in cMP as well. I don't believe there is a SQ issue for cMP with a full-address upgrade. Unless there are insurmountable programming obstacles to full-address cuesheet capability for cMP, I can't see why you would have a problem with this.
-
I do not have a problem with cplay as is.
Love your player !!
How can one organize his files alphabetically . I rip cd's on one computer and add them to my Vista computer via a USB HD . If I look at the files in explorer they are alphabetical but when I click on the folder file in cPlay they are listed as , last files added first . Which makes it much harder to find what I'm looking for .
Thanks so much ,
Joe
I should have changed the title line .
Thanks Joe
.
This is I think the best release ever, as I can't recall any other better versions. Best accuracy and details of all cplay versions. Great job!
I just got my cMP-Squared player to the point where I could put it in my system and listen... got it configured physically, got a few recordings ripped, got most of the optimizations done (including Minlogon & Awe), got sound coming out, it's a good start.
I played it last night and today until I saw a new release of cPlay was out... time to do a cPlay install.
But the last thing I'd done with the player earlier today was to set cMP in 'cMP' mode for startup... and I'd started it up that way and left it.
Now it won't come out of 'cMP' mode. I've tried every combination of setting it to 'WinXP' mode and restarting it that I can think of, I've looked for registry switches and .ini files that set the shell, I've tried rebooting in safe mode, I've tried uninstalling cMP while in 'Explore' mode... and when that didn't work, I tried renaming the cMP executable and restarting. Of course it didn't come up that way because it was still trying to use cMP as the shell, so I connected that harddrive to another computer and changed the name back... and reinstalled cMP to fix it.
I was able to get cPlay 2.0b18 installed while in 'Explore' mode, but I will want to get it back into Windows XP someday. Suggestions? I am not adverse to doing a reinstall, I'm thinking I'll want to do that one of the next few weekends so I could go through the optimizations more cleanly (now that I've gone through the process once).
How do I get it out of cMP mode?
TIA!
Greg in Mississippi
P.S. SQ is good, on par roughly with my newly-built Shigaclone with more detail and better hi frequencies, but the clone wins in the bass and mid-bass punch category. Still, I've got a lot of optimizations yet to complete... for example, I'm just running through the Juli@ analog outs, the only DAC I have running right now is the NOS DAC I use with the clone, but it won't go to a sampling rate over 48. I do have a Xonar Essence STX to try at some point. I'm thinking the Xonar with it's separate PS connector or the Juli@ on a modified PCI header so I can power either from a separate linear supply will be hot tickets... the Xonar direct, the Juli@ either direct or into that super DAC I ordered that won't be here for a couple of months.
I also have other physical optimizations to complete... it's all still powered from a single AC line, cable clutter has not been reduced, CPU and PSU fans are still active, etc.
greg-- the xonar won;t work in cplay-- cplay requires 32 bit asio, and the xonar is 24 bit only!!
it is a great sound card though, and you can run winamp in cmp, and use carcass's asio recompile -- there are posts here that will explain how, if you want to try that.
And I forgot to ask... does the new STX driver (dated 1/23/09 or so) improve the situation with cMP/cPlay?
Greg in Mississippi
Hmmm... I'm confused. I thought the Xonar would work with cMP/cPlay using ASIOall?
That's from this post:
http://www.audioasylum.com/cgi/t.mpl?f=pcaudio&m=40718.html
BTW, just saw your pix of your PSU for your Xonar. Very nice.
Greg in Mississippi
After setting 'XP Mode', do a shutdown (restart will go into a lock state) for changes to take effect.
Also worth checking if file "explorer.exe" is in c:\windows directory.
You have important optimisations (clean power, etc.) outstanding and I'd suggest doing these before modding Juli@.
cics,I've done as you suggested... I now have the 'dirty' GD PSs on a separate circuit with a dedicated AC filter (did make a slight positive difference), fanless cooler on CPU, and all extraneous cables removed from the Antec along with unplugging the fan. I did ease the rest of the way down to .85000 V on the CPU Voltage Control and have tried both recommended CPU GTLREF Voltage Ratios, settling on .566. And after making these changes and checking the CPU temps, I disabled the rest of the temp monitoring in the BIOS.
These all have made useful and significant improvements to the SQ, lowering hash and harshness, making the spaces 'blacker' which brought out more details, and improving dynamics and bass extension and power. It sounds pretty darned good, especially considering I'm just using the analog outputs of the Juli@.
Looking forward to getting the 256Mb memory, trying external DACs, and beginning to tweak the system (adding voltage rail smoothing caps, damping, etc).
Again, many thanks for this groundbreaking system sir!
Greg in Mississippi
Edits: 02/08/09
Hi Greg,
What methods were used to remove the extraneous cables from the Antec PSU? Did you unsolder the wires, or simply cut them short? If you cut the wires, about how much were left in length?
Thanks!
Alex
I unsoldered them. Not too bad of a job for someone who's comfortable soldering, it took a moderately hefty iron (40 watt) and I went back over the unsoldered areas with solder wick to make sure I didn't cause any solder bridges.
The cables for each hole were bundled into a group with a crimp fitting. I had to cut two of the cables for the motherboard loose from their bundle and solder them back in.
I removed all the cables except for the two motherboard connectors. Since I have a spare GD PSU, I can use that to power disk or CD/DVD drives that I'll use when copying things over or reinstalling the system.
I did it this way so I could reverse it if I needed. The areas of the board where the cables connect for the various voltages are pretty clearly marked and the cables are all color-matched and codes, so it will be pretty easy to reverse even without instructions or pictures.
But I would only recommend this for a person moderately comfortable working with circuitry and skilled in soldering. A good solder bridge or putting the two cables back in the wrong place would have dire impacts on your motherboard & processor... and possibly your harddrive.
The novice modifier's alternative is to clip them off... I'd leave as short of a stub as I could manage, no more than 1/2".
YMMV. And unless you are comfortable doing this, don't.
Greg in Mississippi
Thank you for the detailed explanation!
Right now, I simply use two TEAK clamp filters to reduce the antennae effects coming from the unused cables, and the difference in SQ is quite noticeable. I'm quite sure that unsoldering the cables will improve the SQ even more, but if that's not an option, how much difference should I expect from cutting the cables short? For if they're about 1/2" long, I won't be able to use the filters anymore. Please let me know what you think.
Cheers,
Alex
You are welcome.
And I'm not sure I can give you a good answer to your question.
Here's my gut feel... no cable (that now can't pickup or radiate) is better than a cable where you've reduced the pickup or radiation with a filter clamp. And if I were cutting them off, I'd use a pair of toenail clippers and trim them flush with the board. With care, you can leave the ones that are connected to the motherboard cables connected.
The downside of cutting them is that un-doing it is harder.
But I can't really give you a good feel for how much difference you'd have between no cables and filtered cables... sorry, since I haven't compared them or even used the clamp-on filters.
I will make you an offer tho... if you do decide to cut them off AND you decide you liked it better with the cables and the clamp-on filters, send me the PSU and cables and I'll solder them back on for you.
Greg in Mississippi
Hi Greg,
What a kind offer! If I were still living in the states, I'd unashamedly send my Antec and ask for some professional "un"soldering.
Can you please elaborate on the precautions to take when cutting the cables? Should I leave all extraneous cables that are connected to the 24-pin connector intact? I'm away from home right now, but from memory, I don't think that any of the them are connected (maybe I need to remove the chassis?) to the motherboard cables. Or did you simply mean that I should be careful and not remove the motherboard cables by accident?
Thanks!
Cheers,
Alex
LOL.
I'd love to do it for you, soldering is fun!
What I meant was to be sure and not disconnect cables that you didn't intend to disconnect. If you are going to do what I did, you'll end up with all the cables for the two connectors that go to the motherboard still attached to the supply... the 24 (I think) pin large connector and the small 4-pin connector.
All the other cables that connect to the other connectors... IDE and SATA drive power connector cables along with the floppy drive power connector cables... I pulled out. In at least one case, they were bundled with cables that went to the motherboard connectors... that's where you need to be careful and snip the right cables and not the ones that power the motherboard. What I removed ended up being a large hank of cables that were bundled together... I'll try to post a picture of the removed cable hank tonight.
As far as precautions while doing it, you want to give the PSU some time to be sure it's discharged after removing AC power before clipping wires... 1/2 hour if you really want to be safe, I waited about 10 minutes. Then if it were me, I'd pull the motherboard connectors and their cables to one side, the drive and miscellaneous connectors and their cables to the other side, and wire-by-wire, use a pair of toenail clippers or small dykes to clip each wire as close to the board as possible.
Take it slow and good luck!
Greg in Mississippi
cics,Thanks for your response too. I had seen the comment about restarts not working with SMBus Controller disabled before and I was not trying to do a restart, just shut it down... still, all it would do was restart itself back into cMP mode.
Last night when I went home, I did go through the steps just as described by Theob and it still did the restart cMP loop in both regular and safe modes, whether I had diagnostics selected or not. Explorer.exe was in the Windows directory, so I tried renaming it and pasting in a copy from another PC... not sure if it would work, it was an XP Home machine. It didn't work, but after I renamed the old explorer.exe back (while the drive was attached to another PC), it started working ok again. Not sure what happened, but now I have a means to repair it if need-be. I will generally leave it restarting in XP mode for now, so as not to tempt fate.
SQ was up a good notch last night. I expect it was the combination of 1) burning it in during the day, 2) having cPlay 2.0b18 in, 3) setting the CPU Voltage Control down another notch to just above the magic .85000v, and 4) oddly enough, using the 3/4 outputs on the Juli@. I wanted to try my NOS TDA1543 DAC again, but thought I noticed an SQ improve just changing that setting (before I disconnected the outputs and hooked up the DAC). I went back and forth a few times and it seemed to stay, so I left it that way.... and the cMP-Squared setup is now close to neck-and-neck with the Shigaclone in bass extension and power while improving it's lead in detailing and delicacy AND moving just ahead of the 'clone in dynamics. Very nice.
It's funny getting used to a setup like this... I keep wanting to go find the CD to put it in the player. I do like that bit of convenience and I have at one backup of my rips so far, so I feel pretty safe... better when I have 2-3.
I am having to get used to the remote... since the processor is so busy with upsampling, I have to remember it may have a bit of a delay responding to commands. First day I had it on, I kept selecting another track when I was trying to change the volume.... and I also had a hard time going back out of play-mode, but I have that keying combination down now. I did start out with a piece-of-junk freebe mouse, but replaced it with a high-bux MS laptop mouse... cute form factor, it folds up for travel which also turns it off. I'll have to borrow a Logitech mouse to see if it works any better.
To be clear, what I've already done in my setup is:
1. Physically configure a cMP machine... Zalman case (with all the fluff cables and boards taken out), Gigabyte mobo, e7200, 160Gb Seagate laptop HDD, Antex Earthwatt on the mobo, 2 GD supplies for the HDD and the USBs and touchscreen. I did remove the DVD-ROM drive after heeding your suggestion and setting up a separate ripper machine... good suggestion, I'd already burned-out my original DVD-ROM drive by not letting it cool down between rips. I got a cheap, small case and put two DVD-ROM drives and hooked up my external CD burner to it... now I just cycle through all three and each gets time to cool down before the next rip.
2. Performed the basic AOB configuration, including application and WinXP component removasl, system pare-down including devices, services, and Autoruns disables, sound card, video card, boot.ini, and WinXP configuration with no networking, and set AWE on and installed Minlogon. I also performed the cMP installation / configuration additions, Gigabyte-specific BIOS changes, and any other configuration changes I can find on the monster threads... BTW, any ETA on an updated manual?
I have it hooked up in my system with using either the analog output of the Juli@ at 192 or a NOS TDA1543 DAC that only goes to 48. I need to do a good shootout between them, but with the SQ improvements last night, I think I prefer the Juli@ analog. I don't have the GD PSU's hooked up to a separate AC line yet, but they share a local Hammond inductor with the CPU and the amps have their own inductor... and unlike Theob, I don't seem to have a problem with that.
My next steps are:
0. Order and install the now important 256Mb Kingston ValueRAM.1. Take the CPU Voltage Control down those last few steps and then try the different CPU GTLREF Voltage Ratios.
2. Run an extension cord into the room so I can run the GD PSU's on a separate circuit. I'll also try a separate Hammond choke on that line to see if it makes any difference. I'm real curious about this as I am currently running an all SMPS system... Ice amps, + the cMP.
3. Install the fanless CPU cooler. If the temps look ok, disable the final temperature-monitoring facilities in the bios.
4. Pull un-needed wires out of the Antec and see if it'll work ok with the fan disconnected. Also try adjusting the voltage while I'm in there as someone suggested.
5. Start experimenting with D-to-A alternatives. First, I'll try the Xonar direct both run off the Antec PSU and a GD PSU (I have a 3rd one). I've got a couple of the cheapie Cirrus DAC CS4397 DACs from Ebay that I need to Lampecurize and what should be a pretty good implementation of the new 32bit AKM coming sometime in March. I'll try all of these off both the Juli@ and the Xonar AND only then start looking at a separate linear PS for those cards and modifying the Juli@ to power it that way.
6. Start experimenting with power. I'm curious how a PicoPSU with a good linear PS before it will sound (ala Peter Daniel's setup). Also it will be in line along with the Antec for SMPS mods... I've done a few of these now and seen some other implementations and have gotten good results with them... upgrading caps, rectifiers, and AC input filtering.
After all that, I want to play with selective vibration control and later build a purpose-built wood and carbon-fiber based enclosure for the machine, using a small mobile touchscreen for the display (inspired by the hints Rick McInnis has dropped about his setup).
And if it sounds as good as I heard last night before doing all the above AND not having fully run-in, I can't imagine how it will sound afterwards!
Really, really, REALLY thanks for all you've done to make setups like this possible!
Greg in Mississippi
Edits: 02/03/09
if you get some hardness drifting in and out plus if you get stuck trying to change tracks within a file it may be your 193L. remember I started out ok and problems were intermittent and only became manifest after several weeks (like 5).
dc-dc converters are switchers, too.
Just because you feed it with a linear supply will not make it a linear supply. I wish it were so.
Rick,Thanks for the reply... good to hear from you.
Yah, I know it's a switching device. But I have hopes about it being a useful option. First, the 12V linear supply does the lion's share of the voltage conversion where in a traditional PSU, a high-current switching circuit does that... possible reduction of noise. Second, the 12V rails will be directly powered from the linear supply, definite reduction of noise. Third, it is a moderately simple device compared to a traditional PSU and lends itself to 'some' modifications, such as what Peter Daniel showed in his DIYAudio thread.
Dunno if it really will prove to be useful or a dead end. Easy to try, tho.
Also I saw the comment from cics on the PicoPSU not having the full 24 pins for the Gigabyte moboard. The reviews say that many boards will operate ok without those last four pins... and they can be added via an adapter. One data point is Peter Daniel's setup, where he's currently using a PicoPSU on the Gigabyte board without those pins connected.
Honestly, I think a full linear supply will be the hot ticket here, but until someone engineers same (my EE skills are not up to that, being a self-taught electronics tweaker), this type of experimentation will have to do.
Later!
Greg in Mississippi
Edits: 02/08/09
try clicking on cmp icon (bottom left on cmp screen) click ok start xp and then keep on clicking ok (Windows xp boots on next start up),ok on cmp settings, ok on cmp diagnostics (leave cmp diag always on to boot out of it), X on bottom rt of cmp screen, type No for Restart after System Shutdown, Ok on Cmp diagnostics----wala you're out of cmp into xp.
Theob,
Thanks for your response. That is what I've tried, in both regular and safe mode. I've also tried it with various combinations of cMP and cPlay diagnostics on and off. In all cases, cMP immediately restarts and it never goes to WinXP. I forgot to mention that I cannot shut the computer down normally now... I have to power-cycle it.
Other thoughts?
Thanks!
Greg in Mississippi
dunno always works for me. u sure u have cmp diagnostics on yes?
I concur in the others' raves about the spatial qualities of playback with this version. Also, it seems as if there is a greater delicacy of the individual sounds, which I take to be due to reduced distortion (much as with jitter reduction). The 'excitement' of the last version or two may have been this distortion still there.
One small issue with it: the display lag has returned, at least with high rez (?bigger) files. Often music starts while the 00 or the end time of the last track remains for a few seconds.
First thanks to cics and others for their input and knowledge. I have been following these threads for quite some time. I just installed 2.0b18, 192/145 the SQ is really quite exceptional. I did the V.03 mod to a P5B-VM, Zalman, L22 with the exception of the ram clock. Is there a how to somewhere as the BIOS is not as simple. I had the opportunity to build some video editing suites in a past life so I have some howto. When I play wav through cMP2 explore it spawns my samplitude, do I assume that samplitude is still playing through cMP. Lastly, I have 2x 1gig of the KVR is there a benifit to downsizing the ram. My apologies for such a long post. Really, the separation in the sound stage is quite exceptional. Thanks again, T.
No worries!
Not all BIOS pgms are created equal. Few mATX boards allow the degree of optimisation of the GA-G31M-S2L F6x. Don’t know what degree of optimisation P5B-VM allows but do your best to apply as many as you can as per the cics spec. Don’t understand your question re: ‘samplitude’. As for RAM, the rule of thumb is: less is best.
Hello, Cics! You did it again. Another wonder. This version is very dimentional, the soundstage is highier, wider and much deeper. It lacks some of the "electrified" intensity of silky highs, present in 2-17, but sets new standards in longtitudal materialism. It is most apparent with low level details in the lower region. Instruments, leading in the background, are more prominent, and their position is clearly defined. However, I would rather call this version more "tame", lacking the drama of 2-17 and though it sounds "faster" I'd recommend it for more "in camera" music. Thank You, Cics. Each new version is something I am eagerly waiting for every weekend, and it somehow helps to escape the realities of crisis.
Serge.
p.s. Of course, all this is very subjective, especially with my humble setup.
Edits: 02/01/09
I am listening to 170 host clock control and at 192 khz/146 src, all other settings per spec and so far no issue. More importantly , to me, the subtle hardness I experienced in my sys with 15, 16, 17 is gone. Highs are delicate and steppe is right spatial cues are very very good. Soundstage sets new record for depth.
We'll see how it holds up with time but 5 hours consecutive lstening and no problems. So far so good.
The ease of music just flowing is palpable. Hope my problem is gone but I'm working on another juli@ card to , if nothing else, just have it for mods as per rick's suggestion. Or if my problem is the card hopefully the new one will be ok.
Since my problem kept getting worse and did not seem to consistently respond to trying different computer bios settings I figured that I had to recheck all assumptions and go back to a timeline of changes to my system and try to id a change that could be the culprit. So Sunday morning I shut off my pc and removed my 193 L Hammond inductor from my pc power cord. This tweak is used to great affect in 3 or 4 different places in my system and is recommended by Alan Maher in Tweakers asylum. Anyway that is when 2.0bv18 also became available so I changed 2 things at once: new cplay version and no inductor on my pc power cord. I know the changes are confounded as statisticians say. But the problem (especially subtle hardness) vanished! Also the descend into a gross hard metallic overlay no longer happened. I left system on all day and reset all parameters in bios to the recommended levels for host clock control, clock speed etc and no problem. System playback was best I have had, ever! System has been playing straight for 28 hours & sounds great. I believe the issue has gone. I have posted in Tweakers Asylum to see if this makes sense to Alan. We'll see. We'll also see if problem really is gone. Also in last 28 hours I have tried to trick system by quick track changes within a file and nada. So this may be it.What is unclear to me is why the inductor created no issues until 2.0b15 (except for the quick track changes within a file). Maybe as others have said the new cplay versions are so agressive in achieving great sonics that under a highly loaded cpu this is more of a problem now. I don't know.
So for others that use this wonderful tweak (hammond 193L inductor) on their pc power cord. Take it out--for now until I learn more.
Edits: 02/02/09
i wouldn't rule out power conditioning for the audio pc, but would be very careful about which i used. from what i have been told by people whom i trust, and who have designed very good power conditioners,the filters in conditioners are tuned for a specific load, or kind of load, and get 'detuned' when they are connected to something they don;t expect. this is perhaps why power conditioning is 'system dependent'.
being a retailer limits my ability to discuss particular brands, but i've had great luck with both of my pcs which play audio, and have them wired with good power cords, and connected to audiophile power conditioners.
I found out the hard way. But the Shunyata Hydra and Kimber Kable power cord (forgot model but its the one with device in middle)works nice now. No metallic issues since I removed Hammond 193L inductor.
I know I tried plugging the music computer into a PS AUDIO ac regenerator, the 300 watts version. The computer would not even turn on.
The computer certainly did not need 300 watts to run.
I found the JON RISCH power line filter to make my music computer sound better and did no damage whatsoever.
That transformer was simply a bottleneck in power delivery to the computer. Not enough to keep it from operating but ...
I love mods maybe too much but this is a perfect example of how carefully one need to proceed with them. I am as bad as any as wanting TOO MUCH any mod to effect an improvement. We have to retain some sonic cynicism.
Glad you are back to normal.
My line stage is still making HORRIBLE noises after running for a few hours. Its designer thanks he may know why. This time, I did nothing. No mods, none!
Bye,
Rick McInnis
... very difficult for me to diagnose. When I 1st started using the 193L on the computer I was into 2.0bv6 or 7 and everything was ok to me. It wasn't until I hit v 15 that the problem just magnified any hardness issue (then ultimately the hard metallic overlay) but with 16 and 17 the hardness got better but was still there. So I lucked out by trying to remove the 193 L. The 193 L is a great tweak for non computer audio devices. Maybe I'll look into the John Risch stuff.
Tried it with 256 value ram and got an error in cplay (need 256 ram at least). Apparently the 256 value ram shows only 248 available and doesn't make the cut.
Edits: 01/31/09
thanx
Hi guys,
I came across this thread looking exactly for this - a computer based transport! I tried to read it all but it is kind of overwhelming. I am very eager to start and I need a piece of advice - so far my list of the hardware includes:
1.Intel E7200 CPU
2.1Gb Kingston KHX6400D2UL/1G RAM
3.GA-G31M-S2L Gigabyte MB
4.Antec Neo-HE-430 PSU or the Earthpower 430
5.Case - looking at the Zalman, but not in a hurry, open MB and PSU is OK for now, will try to attach an LCD.
6.Card - I have a Lynx L22 card so I would like to keep it at the moment and at least try, if you have failed to make it work please let me know.If the Juli@ is necessary I will get one.
I will really appreciate it if you give me some advice.
Really thank you! I did search the thread again, I am starting to build the system, will post again soon!
Agree with steppe's comments (below).I will add the following:
1.Intel E7200 CPU
Yes, this is recommended.2.1Gb Kingston KHX6400D2UL/1G RAM
Would be OK. Also experiment with 256Mb Kingston ValueRam (very cheap). See what you think.3.GA-G31M-S2L Gigabyte MB
Yes, this is recommended.4.Antec Neo-HE-430 PSU or the Earthpower 430
Both would be fine. Earthpower 430 is recommended by some here. Steppe's recommendation for a fanless PSU is also a good one.5.Case - looking at the Zalman, but not in a hurry, open MB and PSU is OK for now, will try to attach an LCD.
Agree, case is really a secondary consideration.6.Card - I have a Lynx L22 card so I would like to keep it at the moment and at least try, if you have failed to make it work please let me know.If the Juli@ is necessary I will get one.
Juli@ is fine but not necessary. The advantage of it is that you will know it will work and there will be lots of support for it here. I believe the Lynx is also very good so you ought to try to make it work. I go USB out to EA Freeway2 and then to external DAC; so there are options in this regard.
Edits: 02/01/09
Hello! Search for CMP post from Cics, read carefully, get pdfs, read again, decide for yourself if You will use a touchscreen as per AOBCT recommendations, read about cmp2 and hardware updates, prepare a clean xp sp2 install. Cplay is a player to be used from within Cmp. I tried to use cplay with or without cmp and can tell you - the difference is very big. Don't forget to do all optimizations to OS, get a fanless PSU...
A lot of work, but DEFINITELY worth it.
Serge.
I'm still struggling with my 'metallic issue' in cplay 2.0b17 in a cmp environment so CMP^2. Recently I reloaded in 1.04 drivers after uninstalling old drivers 1st. I just wanted to make sure I have no redundant files in sys32 and sys32drivers.
So in sys32 I have 3 files:
1) one labeled Egosys
2) one labeled JulaASI
3) one labeled Jula pan
In System32\drivers I have 2 files
1) one labeled Jula
2) one labeled Julawdm
Is this correct or are there too many?
hi theob - that's as it should be - the Egosys.ico is just the icon.
I've been thinking about your strange metallic sound, but haven't experienced it yet. However I have a fuzzy recollection of something like that in the past - I associated it with changing the upsampling rates for cPlay while in cMP mode. Maybe that was in Vista, which is trickier. Before you throw away the Juli@, see if the problem is the same with a different soundcard, if you have one.
hi seger thanks for your response. don't have another card. I'm trying to buy another juli@ card but all retailers are out of stock, trying to buy on ebay but I'm having trouble registering on ebay. anyway I use an external clock (in my modified oppo dvd player) and sometimes I toggle between internal/external clock. maybe this is related. anyway I believe sometimes I get a slight metallic sound in normal listening but it comes/goes. when I get gross metallic overlay I simply reboot and it goes away. Going to 200 host clock control seemed to fix it but it was a little zippy at that rate. I eventually incrementally backed the clock down to 180 -175 where its good sonically but the problem (the gross metallic problem) occurs one or twice in a 24 hr period. I don't mind getting another juli@ (cics thinks it may be the card) but can't seem to find one available.
I did an experiment where I did not have one of the 2 juli@ drivers installed at all and tried to run cplay--right off the go I got the metallic sound. So I'm sure its somehow rel;ated to juli@ drivers or printed circuit board.
Does it do it without the external clock? If yes, have you tried physically removing the card and putting it back in again? (in lieu of kicking the damn thing.)
BTW, I have another way of uninstalling 1.04 drivers that's worked every time in XP so far and is much easier: I double-click on the "INFCLEAR.EXE" file in my Juli@-v1_23 folder, click the "Remove Driver" button, wait for it to finish, and then restart. It gets rid of all devices and driver files for 1.04 successfully. I use the 1.23 drivers for my kernel streaming system, but it's worth downloading even if you don't need it, just for the quick uninstall you can do on the 1.04 drivers!
I toggled back into the internal clock. I'll run this way & see. I have reloaded drivers several times in the last week. Let me better understand this infclear.exe file in juli@v_1.23. Do you still take out the card when you reload? I do it via cics within minlogon by uninstalling drivers, removing card,rebooting, installing card, go into device mgr and starting the load driver process, rebooting and go. What does infclear.exe replace?
No, I don't take out the soundcard - but for mystery problems, a bit of physical attention to the component sometimes helps (also a mystery!).
Running the infclear.exe, and clicking Remove Drivers completely removes all the drivers. It doesn't leave any driver files in Windows system32. The reason this 1.23 file works to uninstall the 1.04 drivers is that all the names of the driver files in the two versions are the same, and 1.23 was designed to work in XP, while 1.04 was actually intended for Vista.
To install new drivers, cics way is fine, and I guess it works in minlogon. I have always gone out of minlogon, started the necessary Services, and then uninstalled/installed. To install 1.04 drivers from the normal (not minlogon) desktop, usually Windows starts the process after rebooting, but if it doesn't, I use the "UnInstall_32.exe" file in the Juli@-x2v-v1_04\win32 folder and click the "Update" button, which gets the process going. As a last resort I would go to Device Manager, or use the "Add Hardware" wizard, but in fact I haven't had any problems for a long time now.
thanks. this problem is a bad one though. I sincerely believe it is a defective juli@ card so am trying to buy another.
I've got an ebay bid on another juli@ card.
Hi cicsForgive me for sounding like the proverbial "broken record" (well... locked groove would be more accurate :)) - but again I make the request to change how cPlay behaves when "INDEX 00" entries are found in CUE files.
PLEASE amend cPlay to ignore these lines (as if they were preceded by a REM statement).
I won't repeat the why's and wherefores in this post (they're documented elsewhere in the cPlay thread - just search on "INDEX 00").
Without this change, cPlay cannot faithfully play certain (validly ripped) CD-image files. Period.
With the greatest respect for all you've achieved to date... surely you don't intend the player to remove the audio between the 00 and 01 points - skipping over something the artist/producer intended us to hear - but that's exactly what happens today (under certain circumstances).
Just as gapless playback is an essential feature in cPlay, so is the ability to hear all of the program material recorded on the original CD. Please consider putting this on the 2.0b18 TO-DO list.
Thank you so much (again) for the breathtaking clarity of recent cPlay editions - and sorry for being a PITA on this issue.
Cheers,
GrantThat's not a Toy... IT'S A TOOL !!
Edits: 01/28/09
Would it sound better if UI is eliminated completely?
Hi AT...The issue I'm reporting is specific to when cPlay is used in "CUE file" mode (ie: playing WAV or FLAC referenced within the CUE).
The undesirable behaviour occurs whether cPlay is run standalone - without cMP - or from within the cMP shell.
Thanks for thinking about it.
Cheers,
GrantThat's not a Toy... IT'S A TOOL !!
Edits: 01/30/09
... I'm still getting that annoying metallic overlay but its worse and it happens more frequently. When I reloaded juli drivers my system was good for several days w/o a glitch but after getting into 2.0b17 I got one again but thats all until today. The 256 ram seems to make it worse but it only happens when I change a track within a file. If I let one file go on (in fact I had one go for 3.5 hours on repeat -- no glitch) but as soon as I get itchy trigger fingers and change tracks I corrupt it or something. Always goes away with reboot.Cics do you think now that I have a bad juli card & should just order another? Is there anything else I can try?
Edits: 01/27/09
Theo,
I heard something like this once but it never occurred again. And I am not sure it was really a metallic sound ...
It could very well be JULI@, but then again, who knows? I think you should just get another one to be sure. If the same thing happens I am sure someone would buy it from you. So even if you lost twenty dollars it would be worth knowing.
I have two JULI@'s which I find helpful when making the suggested mods. They are not really expensive and back-ups are always handy.
Just an opinion offered with sympathy (dealing with a problem myself at the moment, though I cannot afford a replacement line stage!!!!)
Best of luck,
Rick McInnis
thanks rick, think I will.
.
I press x button on bottom rt of cplay screen. If I pause a file then press file open (3rd button from bottom left of cplay screen) to access another file in explorer it seems to be ok.
Problem usually occurs when within a file I press another track that is not in sequential order. So if I'm playing track 1 then decide I want to listen to track 5, I simply press 5(double click 5) or sometimes I pause track 1 then double click 5, it loads then I get the issue. But not always it is usually after I do this repeatedly (as any audiophile who wants to listen to his favorite cuts when excited about something new like value ram or a new version of cplay) say 7 or 8 times then I get the issue. Since I 1st reloaded juli@ drivers I have reloaded 3 more times and I still get issue but this time more frequently. However if I don't select non-sequential tracks within a file I can go 3 hours and no problems.
It lists Juli@controller and one unknown device (among identified others).
When I click on unknown device it shows:
Device type: sound,video, and game controllers
Manufacturer: EGOSYS
Location: Location 1
Device Status: The drivers for this device are not installed (code 28)
To reinstall the drivers click reinstall drivers
Under Device usage it says use this device.
So apparently this is the problem but when I go through the reinstall driver I click through all the steps I get to the julawdm.sys on juli@Windows driver for X86,X64 is needed type the path...then click ok I get there was a problem installing this hardware Media
An error occured during installation ...driver not intended for this platform
So anyway I ended up with 2 juli@ panels (i deleted 1st used 2nd) and right off I get the metallic sound.
So I will go through the unistall/reinstall process all over again and see what I get.
very humbling baring my dumb errors here.
I updated juli@ drivers again and made it through the process (using 32 subset files of 1.04 drivers. is this correct?). Anyway I reboot, get my juli@ panel I reset latency and external clock( slaved to my oppo 980 at 192 khz) and I'm ok. Cplay plays and sound is wonderful (full comments coming).
But just to be sure I go through xp --> control panel ---> device mgr and check sounds and it shows one unknown device. when I click on it says Sound, video, game controllers. Mfg: unknown. Location 1, The device is not configured correctly (code 1).
What does this mean?
Now shows no unidentified devices in control panel sound subset.
We'll se how long this is ok.
Cics thanks for making me think.
Control panel still ok showing no non id'd devices.Btw all other bios settings per your latest bios posting. I'm at .86 volts for core cpu and 3 3 3 5 timing for memory.
Edits: 01/28/09
Theo,
I say this with a certain amount of confidence.
I am assuming you are loading 1.04 drivers.
I am assuming you are NOT using the supplied disk at all, simply pointing the hardware installer to DESKTOP where the drivers folder "is". (this is as it is in my case)
Did you follow segars advice on how to COMPLETELY uninstall JULI@ drivers by going into WINDOWS? If not, I will try to find the post for you.
There really is nowhere else this noise could emanate from.
If the machine is basically stable with cics's recommended settings I do not think there is anything to gain by trying others.
Let me know,
Rick McInnis
no I installed juli@ drivers the proper way and reloaded drivers from my flash drive. but 3 hours at 200 and no problems. hell I don't know Rick. It's really funny I am the only one with this problem altho Chipzahoy said he had a similar issue which went away with new 256 value ram
Edits: 01/28/09 01/28/09 01/28/09
I was talking about how you installed JULI@ 1.04.
I figured that was where there could be a conflict.
sorry rick I misunderstood (and actually stated windows when I meant juli@ drivers--I get rattled when I get a computer problem and type gobbledegook). But I did get a clean install on juli 1.04 drivers (double checked in device mgr) but still got my issue at 170 host clock. I'm back at 175 host clock and no issues since this am and the sonics are almost as good as 170.
Edits: 01/28/09
Is this not correct?
cics said something to the effect that he was not sure which MOBO was in use in your system.
Let me know,
Rick McInnis
I am using the gigabyte G31M-S2L.
...with me trying to trip it up. I think I'll slowly lower clock to where I get no issues see what I can achieve.tried 190 ...good for 2.5 hours no problem
Edits: 01/28/09
...wrt to my metallic overlay issues and I'm still trying to trip it up via changing tracks within a file. Sonics are pretty cose to 170 host clock. Does anybody remember the host clock setting for cplay 2.0b 14? Think it was 175. Had very few issues at this level.
Edits: 01/29/09
...very few issues--none in 12 consecutive hours of playback.
I continue to verify no issues in device mgr wrt juli@ drivers.
I have concluded that residual issues (metallic overlay) is due to a faulty sound card. I'm ordering another.
ll
,,
The computer was experiencing lots of metallic issues last night. No matter where I set cpu host clock control in bios, internal or external clock on juli@ panel I was getting problem every 20-30 minutes. So on a whim I set cplay at 44Khz and not only did it not get the gross problem (for at least 12 hours continous at 170 host clock control!) but it sounds substantially better. This has never happened before. 192 with -146 src always sounded better. So anyway I'm back to enjoying listening.At 192 with -146 src I was always getting a subtle hardness or ringing in the highs(I thought it was me anticipating the gross metallic problem) that I am not getting at 44. The subtle hardness has been there since version 15, never got it with 14 or before. I will admit that the hardness improved in v 16 and more so in 17 but its always there. So far 44 sounds good and no gross issues for 12 hours. This may or may not be related to the issue but is just an observation that may trigger something in somebody's head.
Going on 18 consecutive hours listening to cplay @ 44Khz --- no metallic problem.
Edits: 01/31/09
Switching tracks should work perfectly (the same parts of cPlay is used when background loading). Set cMP to Yes in settings to block the file browse function.
Try increasing CPU host frequency to 200 - this will increase CPU frequency to 1200MHz and RAM speed to 200MHz. I suspect there could be an issue with your mobo or card. I'll also try to recreate this issue on my Juli@ setup.
...I was using file browser to get to my non cue files. Especially those hi rez files downloaded from HD Tracks. How do I get Wav/Flac files by only using cmp?
does not show your music files?
Have you not told it where your files are within SETTINGS (cMP button)/FOLDERS/ADD?
not sure I understand your question. yes can find my folders for flac and pure wav (non cue wav) and I can say add them to the cmp menu but only cue files show. what am I doing wrong? I always used theopen file explorer because I could click on cue,wav or flac files and get what I needed. But with cics saying not to use this feature in a cmp^2 system I won't do it. But I don't know ghow to get my non-cue files in cmp.
theob,
cMP will not load anything other than .cue into the cMP browser. This means that you will need to create a .cue for anything you want to play in cPlay using this method i.e. your .wav/.flac files. The alternative is to enable the windows file browser in ‘cPlay settings’ and use it to select your .wav/.flac files directly. I assume this is what you have been doing to date but the windows file browser entails greater system load which, I gather, cics is suggesting you try to avoid as it may be accounting for your difficulty. The best way to create a .cue for your hi rez .flac files is to copy an existing .cue and edit it (using notepad). It is a pain I know..
Bling,
I don't have that many, I'll give it a shot.
Edits: 01/29/09
I create my cuefiles with foobar. What you need is the plugin. You load the files into foobar and with a right click choose "create cuefile" then go to the folder with the files and choose the ansi-format. Before all that you have to create an empty cuefile. I do it with text-file and just rename it to .cue rather then .txt. No editing needed. I did this with hunderts of cd's and it always worked fine. Walter
does this work with wav files? i think i tried this once with the plugin, and it created cue files with no track names, since there are no real tags in wav files -- or did i miss something?
...imposed(80) re seeking more playlist-like functioning necessary? I just realized that several cuesheets made for extended works or collections will now not work, e.g., Complete Beethoven Piano Sonatas by Annie Fisher in 9 cd collection with 102 files.
Edits: 01/26/09
which will be available in cPlay 2.0b18.
This may take a while as improving on 2.0b17's performance is very difficult. As Serge puts it "unnecessary sand being sweeped away" is a very apt description of what b17 is doing.
Refer to ReadMe.txt for important details & specifications.
Change Log (2.0b17 release):
- UI Refinements:
- Keyboard skip actions: '-' for backwards & '=' for forwards
- Cosmetics
- DSP refinements:
- Max ASIO drivers supported reduced to 3 (from 8)
- Further improvements in ASIO output efficiency!
- Other minor changes (clipping, prefetcher & streaming)
Please REMOVE previous versions before installing cPlay 2.0b17. Normal (SSE2), SSSE3 and SSSE3 B9 versions are available. Running cPlay on a CPU not supporting the required instruction set will cause cPlay to exit immediately.
For cMP² users, review these BIOS changes (underclocks FSB, CPU & disable additional USB items). Recommended Gigabyte mobo GA-G31M-S2L remains available.
I have not made a power measurement yet and I cannot rememeber where the timing recommendation came from but I tried it anyway.'
It is completely stable and sounds very natural. I have not had a chance to listen for hours yet (only 45 minutes) but I think this is worthy of further consideration. Initial impression is that voices are reproduced with greater spatial realism, forgive me for saying it but there is something happening which leaves one groping for the HOLOGRAPHIC word which is so overused and tired as to be meaningless, but ...
I have not tried any other timing settings.
All else is set per cics's recommendations.
Power - Always on (since this seems to go back and forth I thought I would say)
I will make some of my crude power measurements tonight, trying different power settings and timing settings.
I do hear some extraneous sounds at the beginning of an album, and sometimes at the beginning of a track when changing within an album, but only during the silence, I do not detect any artifacts while music is playing.
Can't wait to listen tonight.
Thanks, cics,
Rick McInnis
To tell the truth - 3-3-3-5 recommendation seems bizarre, goes totally against what I hear with my system, albeit with different memory (two of them, to be exact).
Would be great if you could compare several different settings, that differ only in increasing tRAS (5 in 3-3-3-5) all the way up to 10.
x93,
When first I saw cics recommendation for 256Mb ValueRam (3-3-3-5) I thought it was a mistake. Previously, with 512Mb OCZ low latency, I had experimented with various settings: 3-3-3-8 to 3-3-3-12 and found 3-3-3-10 to be optimum. Recently, I tried the 3-3-3-5 with the 256Mb ValueRam out of interest.. A good improvement! So my current preferrence for cMP2 is: 256Mb ValueRam and 3-3-3-5.
with my DAC line stage that is intermittent and tonight it is a problem.
Sounds great and then thunderstorms come out of my loudspeaker.
I am told it is most likely a grounding problem but with one shared power supply how can a grounding problem affect only one channel at a time?
Baffles me.
I will experiment when I can.
For the 15 minutes I was able to listen to tonight it sounded really good.
Of course, there is always the synergy thing; this timing might suit my loudspeakers? If you find these timing numbers to result in a softer sound this may explain why I find it appealing. I am using TAD 4001's for mid-range and they can always use some toning down.
I will post what I hear.
Rick McInnis
Dear cics and any other interested parties,
I connected my little HOME DEPOT power meter to the music computer this morning and this is what I found:
All measurements made within cMP/EXPLORE/DESKTOP/cpuz
with #16 and CL5 256kB VALUE RAM - timing at AUTO - 5,4,4,12 - 29 watts
with #17 and CL5 256kB VALUE RAM - timing at AUTO - 5,4,4,12 - 29 watts
with #17 and CL4 512kB VALUE RAM - timing at AUTO - 5,4,4,11 - toggles between 29 and 30 watts
with #17 and CL4 512kB VALUE RAM - timing MANUAL - 3,3,3,10 - toggles between 29 and 30 watts
(I left my CL4 256kB VALUE RAM at work so I have not been able to insert it. I do not have high expectations for it.)
with #17 and HYPER X LL 512kB - timing AUTO - 5,4,4,12 (quite surprising)
solid 29 watts reading.
with #17 and HYPER X LL 512kB - timing manual - 3,3,3,10 (seems to be stable) solid 29 watts reading.
If any of this means anything it seems there is nothing gained with the 256kB VALUE RAM as far as power usage.
I do not remember having such high readings for the HYPER X 512 kB with the older BIOS settings, in fact my memory tells me, with AUTO it was 3,3,3,8.
Any interpretations will be welcomed.
I will make measurements with the 256kB CL4 before the weekend is over.
Cordially,
Rick McInnis
the HYPER X has settled down to 30 watts.
Theo, no I have not had a chance to do that. I am still trying to fix my line stage. So all of this is objective. Can't wait to be ABLE to listen again.
I will let you know.
Bye,
Rick McInnis
Rick wrote:
> > "Any interpretations will be welcomed."
For what it’s worth, here’s my take.
Assuming that your power meter is much like mine (one of those Made-in-China plug-in thingies which is very good for the money but uncalibrated) and as the differences observed were only about two per cent of the total, I’d hesitate to read too much into your results.
You seem to have assumed, for example, that the mains voltage did not vary during your tests. I don’t see that as a problem provided you treat your readings as indicative only, in other words that “rough and ready” is as good as you’re likely to get given the test kit to hand.
A reasonable conclusion therefore is that your various configurations all consume about 30 watts and that changing RAM sizes doesn't seem to change that much. More than that is hard to justify.
The "settling down" after 30 minutes is also to be expected as components (esp electrolytic capacitors) warm up and change electrical parameters. There is no easy way to tell if these changes occur in the power supply or the motherboard or both though it surely matters.
I have done similar measurements and your results are pretty much in line with mine, which is encouraging. However, I was trying to measure how the power draw fell after disabling on-board components such as audio or LAN circuits.
There didn’t seem to be much change, one or two watts or about three per cent of the total. Changing clock speeds made a far bigger difference, from memory about thirty per cent.
Is there any chance that you could repeat your measurements using perhaps just one kind of RAM but at, say, two different clock speeds such as flat out and cMP2’s normal operating speed while playing music in the normal way?
Whatever, even if it does not greatly affect the power draw, I take it that we agree that the process of slowing things down and disabling this and that in hardware and software makes an audible difference (i.e. that cMP’s configuration routines work).
The question is why. I assumed that efforts to reduce power draw were in fact cutting down on electrical noise as clock speeds and spurious OS-related data traffic fell. I set about doing what I could with “cheap and cheerful” added smoothing and so on to see if mitigating noise electronically as well as in software would be frutiful.
As you know, list members reported that they had already or have since taken similiar steps with encouraging results.
When fmak and others comment on the unpredictability of PC audio, they make a strong point. However, I don’t think it follows that PC audio is therefore inherently unpredictable. Getting measurements of characteristics that others do not generally seem to be addressing has to be a step in the right direction.
Dear ryelands,
I offered my measurements purely as relative only to each other, not as absolutes.
I did assume a constant in power line voltage. I live in the suburbs and have found my electricity to be cleaner than most.
The last time I made the measurements I did compare cMP mode with EXPLORER mode and there was I think I remember correctly a 3 watts difference. I should make this part of the test.
I was surprised that the memory did not make as much difference as I expected. I need to audition the different memories, also. I have a strange line stage problem with is part of my DAC so I have been unable to listen when I want to. It is more when the line stage ALLOWS me to. A case of gremlins that is completely baffling.
I think with all devices you have the choice of making adjustments or having the adjustments locked in at the factory. I think the ability to change so many parameters is a large part of fmaks unpredictability.
I find my set-up to stay quite stable once a setting is made.
My impressions of v.#17 are very favorable. I have been listening with the HYPER X 512kB. Tonight I hope to try out the CL4 256kB and to make power measurements with this. I will try AUTO, 3,3,3,10 and 3,3,3,5 for timing settings and HEAR what happens.
Your post deserves more thought and I will give it this when I get home tonight!
Thanks,
Rick McInnis
Not sure if this setting will work with soundcard but would be interesting to see comparative measure.
hi rick did you do any listening tests with the various memory cards?
...but I can't seem to set my memory timings....my pc keeps resetting System Memory Multiplier to auto after I set it to 2 and my timings come out @ 5 5 5 18 (rather than 3 3 3 8). After I reset to recommended settings and continue booting I get a long delay before system beeps/boots then in cpuz I get dram speed at 400/timimgs at 5 5 5 18 . I have all other settings per cics 2.0b17 settings except for core voltage (I have at .875 volts). I have a kingston value ram coming maybe that will be easier to set. I currently have a 1 gig kingston ram
Edits: 01/24/09 01/24/09
The BIOS resets some settings to defaults if it has difficulty loading. Try RAM timings at "Auto by SPD" and repeat. If that doesn't work, try upping the core voltage - what works yesterday doesn't always work today if you are on a low setting.
Dave
thanks for your replyalready tried setting voltage control to auto (so it sets it where it wants) and tried 175/180 clock. no matter where I set DRAM setting it goes to auto and I end up at 400 mhz @ 5 5 5 18. same result. it doesn't sound bad but I heard it at 3 3 3 8 dram @ 350 and its definitely better there.
I even tried reflashing the F6x drivers for the mobo and that didn't help. Maybe I just need the 256 meg value ram.
Edits: 01/24/09
Just wait a few more days, and the 256MB Value RAM will fix everything for you!
thanks alex!
Hello, everybody. I like 2-14, 2-16 - they are detailed, but smooth. I don't like 2-15, because to me it sounds like brushed aluminium. But this version really shines: all the smoothness of 2-16 and sooo much more details, especially in the background, tighter lows, also something good happened to the mids - it all seems suddenly so clear and tonally right, especially cymbals. I always start listening to the new version of cPlay from the same 4 wavs and in the same sequence, so that initial impression won't be smudged by "new" album. This version sounds like all unnecessary sand being sweeped away.
Very impressed. Thank You, Cics for such a wonderfull saturday morning.
Serge
Hi Cics,
Thank you for your work! 2.17 provides more details during complex music playback. However, 2.17 is slightly soft and slower. 2.16 is slightly more open, bright. So... it is a difficult choice and I switch between 2.16 and 2.17.
Refer to ReadMe.txt for important details & specifications.
Change Log (2.0b16 release):
- UI Refinements:
- Only actual tracks displayed with splits shown in "()"
- Skip Forward and Backward spans multiple tracks instead of track wrap around
- DSP refinements:
- Reduce minimum RAM requirement to 160MB (allows playback in systems using only 256MB RAM which is most optimal when input media is largely 44.1k)
- Upgrade to libsamplerate-0.1.5 (SRC). A step reduction in CPU load is achieved allowing for lower CPU frequency.
- SSE2 changes done to SRC's "stereo output" further lowering CPU load and improving RAM read & write efficiency.
- Faster FLAC decoding
- Other minor changes (prefetcher & streaming)
Please REMOVE previous versions before installing cPlay 2.0b16. Normal (SSE2), SSSE3 and SSSE3 B9 versions are available. Running cPlay on a CPU not supporting the required instruction set will cause cPlay to exit immediately.
For cMP² users, review these BIOS changes (underclocks FSB, CPU & disable additional USB items). New settings are in effect to best exploit above capabilities. Recommended Gigabyte mobo GA-G31M-S2L remains available.
n.
.
cics - thoroughly enjoying v16, so light and clear. I am also enjoying the wonderful playlist capability thanks to the EOC feature, and because a friend (who had never heard of cuesheets!) who did IT at uni has made a little excel/VB program for me that automates the playlist making. It's a bit buggy at the moment, but if he has time to sort it out I'll share it. But it highlights a cPlay limitation - no Track Performer column, so you don't know who's performing on "Various Artist" albums, which is even more of an issue with constantly changing playlists. Could you let me know if you would consider adding this to cPlay - I think it would mean shrinking the font size a little more. Of course I can work around this if it is a significant sonic issue, but I need your thoughts before going further along the cuesheet editing path.
Maybe in cue sheet, edit Title text to have a short prefix added to it:
"PERFORMER: My Track Title"
That was my fall-back plan. I wonder, though, how many cPlay users have the 7" touchscreen and how many use an LCD monitor instead - not much comment on this has come out. I'm using a widescreen LCD monitor so I can do everything from the sweet spot.
I firmly believe there's very good reason why 1Gb stick of ultra low latency (spec'd at 3-3-3-10) HyperX KHX6400D2UL/1G sounded significantly better in my system than anything else from different manufacturers, ranging from 256Mb to 1Gb.
I would be EXTREMELY surprised if ValueRam sounded anywhere close to that, regardless of the size - although can't say for sure, not having tried it.
I've tried a very wide range of high quality RAM (from DDR1200 to DDR800 spec'ed at 3-3-3-8), and none, I repeat, none came close to the humble Kingston ValueRAM in terms of sound quality.
Interestingly, I also noticed that 512MB sticks almost always sound better than 1G ones with similar specs.
Getting 1 stick each of KVR533D2N4/256 (256Mb, CL4) and KVR400D2N3/512 (512 Mb, CL3).
I mean, how bad can it be at $28.34 including shipping for both?
I get a slight intermittent hiccup / pop at the beginning of tracks with this version. It's infrequent enough so that I won't abandon this version immediately but I'd be hesitant to use this version as a showcase for what this great piece of software is all about.
I haven't noticed any sonic improvement yet over the previous version (after 3 days of listening).
Chris
I get them as well - as you say probably not enough of a problem to go back - just wait for the next release!
Try increasing CPU frequency to avoid this. At 1.02GHz, there should be no dropouts when playing WAV (FLAC is more CPU intensive).
Thanks cics, I will try that suggestion. Would it be better to have only WAV files? Currently everything is stored as FLAC which I am upsampling to 24/96, if I want to go up to 24/192 would it be better to use WAV or FLAC and increase cpu frequency?
There's no difference in SQ as FLAC is decoded entirely before playback. The issue that FLAC causes is potential dropouts when cPlay does background RAM loading. Although this takes just a few seconds, FLAC decoding is insanely CPU intensive and will cause a dropout. This gets worse when output rate is at 192k as CPU headroom reduces significantly (at 96k, CPU0 load is ~40% vs ~78%).
Thanks cics - I think I will start converting my 1000 CDs from FLAC back to WAV using dbpoweramp. For the time being I will up the cpu frequency as suggested and then reset the music pc once the files are converted.
So a second bigger HDD is needed.
Thank you cics. I like b16. There was a little debate in HK whether b15 had a harsh high than b14. But I'm glad it's over now, with b16 and "back to normal".
And I can finally use 145db upsample on e7200 @ 1.2 Ghz without any pops and clicks!
Wow, what a step. So much more musical information. Details, which always seemed to be background noises are so much more present. Voices for example have elegance and modulation, even if they play a minor role. Lyrics are so easy to understand. The sound is somehow soft but dynamic. I enjoy it very much. Walter
Cics,
Thank you for this release. IMO, best so far. I can confirm improvement in SQ using 256MB Kingston ValueRam. (Previously I used 512MB OCZ low latency.)
Bling,
I initially loved 16 and still do after about 30 hours. But after a week listening to 14 with no issues I am back into a sonic metallic overlay to the music which comes on suddenly is instantly identified but goes away after reboot. But let me tell you about 16's musicality:
--- it blows away 14 (which I loved) and 15 (which I did not like)
--- it is more dynamic, more airy, more transparent than 14 by a large degree
--- it brings solo singers out into the room distinctly separate and apart from the backup singers and instruments
--- it has more hall ambiance than anything I have ever heard (source wise)
--- it reveals detail never heard before; e.g. Goldmark's rustic wedding 2nd movement there is triangle there I never heard before
--- it is magnificient (instant love affair)
That is the good news, the bad news is that I may go back to 14 if this issue persists. I really don't think it is a driver issue with juli (I had it on both .979 and 1.04 which is where I am now). It is not a pc issue, I had it on my home based amd pc and on my intel based dedicated audio pc. I don't know what it is but cics as you suggested I'm starting to think it is a defect on my juli card (it was the only thing common to both pc's) but why would it not be an issue for 4 months until I got into 2.0b15 and 2.0b16 (yes it did occur on 2.0b14 but a lot less frequent than 15 and 16).
I'm baffled, I love it but, but....
.
.
Hi theob,
I experienced the same "metallic overlay" issues as you did a while ago. My problems were a bit more severe, as I've encountered "PFN List Corrupt" and "IRQ Not Less Than or Equal" BSODs in addition to to the distortion.
These issues were magically fixed, however, when I either 1)increase memory multiplier OR 2)increase CPU Host Frequency to 175+ in BIOS, which leads me into thinking that the cause is most likely to be RAM related. My guess is that while the new cPlay 2.0b15/b16 took SQ to a whole new level, it's also alot more demanding in terms of RAM quality.
Buying better RAM "may" be the best solution to the problem without compromising sound quality. I've already placed an order on the Kingston 256MB ValueRAM that cics is currently using, and I'll let you know how everything goes.
It would really be great to get a sonic improvement and fix the problem with a 256 meg ram. I've got the kingston 1 gig ram now. Where did you go to, number wise, on system multiplier or cpu host frequency to mitigate the problem?I've just upped cpu host freq to 175. I'll report back later if this works.
Edits: 01/18/09 01/18/09 01/18/09
First I tried changing the multiplier to AUTO, which fixed the problem, but the SQ got noticeably worse. Changing the CPU Host Freq to 175 (sometimes I get away with 172) seems to solve the issue as well, and the impact on sound quality was a lot less. My Kingston 256MB Value Ram will probably arrive in a couple days, and I'll let you know whether if it's the magical solution.
thanks alot chipzahoy, 175 so far so good since last post. also I reread cics recommendation for ram for 16. If 16/44 is main source then 256 is good. and if higher source material then 512. so question to you chipz (and certainly I would appreciate cics answer) does cplay get unstable or just not work when you run 192 files with 256 m ram? or is sq bad?
Glad that it helped.
I haven't tried 256MB yet (the RAM wouldn't arrive til a couple days later), so I'm not sure. Also, I'll be away from home next week, so you might not get an answer from me for a while.
My guess is that 24/96 or 24/192 files would require multiple splits and frequent RAM loading. With 512MB RAM, 24/96 tracks over 5 minutes require one split, so 192 ones may stress 256MB RAM quite a bit.
Why don't you find out the answer by buying one as well? If cics uses the Kingston 256MB ValueRAM himself, then it can't be bad. Have faith! ;)
P.S. My CPU voltage is set to 0.875v. The lowest I've tried was 0.85v, but the system got a bit unstable.
...I'm going to inch up on clock to try to get some reasonable success at no 'metallic noise'.
I got lucky and received my RAM earlier than expected.
Everything runs smoothly, and the SQ is noticeably better than my old setup (Winchip DDR2 1200 512MB). The occasional "popping" noise while RAM loading large files has been eliminated as well, which is opposite to what I had expected.
I'm very happy with my new Kingston RAM, and I believe that you would too.
aapreciate your feedback, I think I will order it.
...I have been able to affect my problem (occurs less frequently @ .636 versus my prior setting @ .566) and the tone of the metallic overlay changes drastically @ .636. Anyway Cics do you use .636 exclusively for your testing?
I'm using Juli@ in my development computer (E6300, mini-ITX mobo). No such issues.
do you think new ram @256 meg would help as chipzahoy suggests? no I haven't reloaded my juli driver. if it is the juli soundcard I'll just buy another. I'm really reluctant to toggle minlogon to reload drivers.
You must be in XP Mode.
In the steps outlined you must manually install the drivers from Device Manager (as Windows will not prompt you on reboot). In device Manager, you should see the new device (after Juli@ is removed) highlighted as needing attention. Right-click and select Update driver... I do it this way.
...& everything is working fine. After reboot sys sounds good but in control panel under sound and audio devices it doesn't show juli in fact there are no playback devices shown (did not show before reload either). Is this normal? I went here because I wanted to set volume @ max.
Anyway thanks again for your help.
I encourage others to try this (was new to me). It is a very tidy to do it this way and allows you to try different drivers w/o the hassle of toggling minlogon.
...problem appears to have been fixed by reloading juli drivers. Cics you've done it again. But tell me what causes drivers to get corrupted? Is there some way to avoid this?
...my pc keeps resetting System Memory Multiplier to auto after I set it to 2 and my timings come out @ 5 5 5 18 (rather than 3 3 3 8). After I reset to recommended settings and continue booting I get a long delay before system boots. I double check CPUZ and it has reset System Memory Multiplier to auto. Doesn't sound good set this way. I have all other settings per cics 2.0b16 settings except for core voltage (I have at .875 volts).
Edits: 01/23/09
thanks I'll try
thanks.
one more q chipz: what cpu voltage control are you running? .9 or lower?
Edits: 01/18/09
Dear cics,
I have my old Kingston Value RAM that you recommended for AOCT, is this good enough? (DDR2 667 - KVR667D2N5/256) NEWEGG is showing one remaining available 256 - Kingston Value RAM -KVR533D2N4/256 - slower speed, lower latency - CAS 4 as opposed to 5 of what I have. Would this make any difference?
I know you are pleased to be able to accomplish this. I unfortunately did not get a chance to audition #15 due to line stage problems (it is part of my DAC). I am really excited about this one. When I get to listen to it ....
Thanks,
Rick McInnis
It's what I'm using and offers 0.86W operating power consumption!
Of course, trying other versions would be interesting.
With new BIOS changes as recommended and lower RAM, I need new power measure at AC. Previous was 24W @idle and 27W @44.1-> 192 145db SNR processing.
Dear cics,
The 256 kB I had was higher latency that what is available. I ordered this and it is to be here tomorrow.
I have some 512 kB VALUE RAM that is lower still that the 256 kB. I am going to make measurements with each of these, just to know if there is a measurable difference with my crude little meter. I will take note of latency numbers with BIOS set to AUTO.
Will post Saturday.
Unfortunately, I am still without a line stage so I have not had a chance to hear. I hope to get that sorted out Saturday, also.
Thanks,
Rick McInnis
I am having some problems with .CUE files (nothing to do directly with cPLAY) - if a track has some non-english characters (German umlaut, French grave etc) and I create a .CUE file then cPlay does not load it - the entries in the .CUE do not match the actual file names. OK - not a cPlay problem - I can get round it by setting cPLAY to allow Folder searches - but this only allows ONE track to be loaded at a time.
Does anybody know how to get round either of these problems - apart from going through 1000+ CDs and looking for all these odd characters and changing them!!
Alan
I have encounter this problem as well, but I found that the cuesheet made with Foobar matched the actual file names but that cPlay could not decode the foreign characters, ones with umlauts in my case which are actually coded as two characters. I could find no way but editing the file names to compensate. I think the limitation is indeed with cPlay which can only deal with English characters as far as I can tell. Perhaps cics can clarify.
I have encountered this problem a while ago with polish caracters and had to edit the name and tags in foobar. I think it is more a problem of the OS-language then of cPlay. Walter
The German titles played fine unedited in Foobar, meaning the OS, windows vista or xp, can and does handle the files okay.
Thanks - I thought I would have to edit the filename, I just hoped there was an easier way. I will wait a few days to see if cics has a solution - otherwise its a search and amend task.
.
Thanks for the clarification - I had already started modifying my filenames just in case that was the only solution.
After comparing , emu 1212, Xonar PCIe and Juli@
some new impressions:
till now the batteriefed Xonar delivered the best quality spdif to DAC.
After purchasing the buffalo DAC (in combination with the LC-Audio Zapfilter -a class A output stage) i tried the Juli@ with the I2S connection ( nice information from DUC985) to the buffalo DAC:
a new world of quality in every aspect.
I don't know how the xonar will perform with the I2S option .If there is someone who knows the I2S pins of the xonar i will try it.
bye
alfred
Did you compare the I2S to the SPDIF on the Juli@ and the Xonar? I ask because on the Twisted Pear forums Russ White has said that there is no difference between using SPDIF and I2S as the Buffalo generates its own clock.I'm expecting that someone will tell me that he is incorrect...
Edits: 01/11/09
hi dan631
i compared SPDIF and I2S with the Juli@ and it's day and night (sure).
And I2s on Julia is better than SPDIF with Xonar.
I2S with Xonar till now is not possible because of the lack of the connection.
With some time going on perhaps i will find the pins on the Xonar.
The problem with the I2S is the short max. lenght of ~ 20cm.I tried 100cm, it works, but it is much better with 20cm. So You have to put Your DAC to the PC( Juli@).
bye
alfred
Hi Sonics
I am considering to buy Asus Xonar D2X I have two questions
1. Does cPLay work with this card with its own drivers?
2. How you batteriefed Xonar ? Are such power supplies ready made available?
Or how to make one and is it easy to make One?
Please if you can direct me ?
Right now I am using Juli@ spdif with its Analog section removed and I occasionally get a click with cPlay only on certain albums.
Also Juli@ does not work so well with Otachan Asio plugin for Winamp.
Thank you in advanced
hi,
the wirings of I2S of Juli@ to a buffalo DAC.
???
That's not a Toy... IT'S A TOOL !!
very interesting alfred and well done. questions though: does the I2S delivery mechanism require a computer software driver? Also does the new combination of Buffalo dac and I2S significantly beat your old set up? I read between the lines that you really like it very much. Also are you into a dedicated pc/cmp shell for cplay? Also what is DUC985?
Edits: 01/06/09
I also have the Juli@ card and I have read many things about the I2S in the Juli@ quite quite some time ago.
The I2S is the interchip communication for transmitting digital signal over short distance. It does not need any extra software. A small number of dacs are equipped with I2S as it could attain better jitter performance than spdif, at least in theory, due to the separation of the clock signal and data. However, some earlier reports of the Juli@ suggested that its I2S output is of a very low voltage level and thus does not give good result with some earlier dacs that have I2S input. The buffalo is a much more recent design though.
thanks the buffalo looks like a good project for us juli owners.
I could visit Alfred some days ago and must say that it is the best system I have heard by now. Low frequences are a sensation. Incredible resolution and a to me new quality of projecting recorded material into a room. The room itself plays a minor role, because there is so much ambiance present, that it dominates the listening room. At least in Alfreds room and I know his situation quiete well and would expect this for any other place too. Funny thing is, that I just changed from juli@ to Xonar. Another point I would like to make, is about the use of batterys in a system like that. As Alfred pointed out several times, one main quality of batterys is not the clean energy but the dynamic capabilitys that only appears by using more than one battery. If you don't try it you never know, what the outcome is like. And of course, all this in combination with cplay is another world of music... Walter
The room itself plays a minor role, because there is so much ambiance present, that it dominates the listening room.
Good point. I have this experience too.
Refer to ReadMe.txt for important details & specifications.Change Log (2.0b15 release):
- Startup: implement ASIO trace log
- UI refinements:
- More status messages and cosmetics
- "Timer" Settings option added to allow for elapsed timer to be shown or bypassed. When "No" (bypassed), message displays "tr 00:00" during playback ("tr" indicates current track).
Bypassed timer gives most optimal playback as UI inteference is removed (no UI timer processing & video traffic).
- DSP refinements:
- "INDEX 00" processsed correctly in cue sheet. Diagnostics provide track information making it easier to fix cue sheet errors. After track INDEXes, a new "EOC" directive (End Of Content) can be used to stop processing content beyond time indicated, e.g. "EOC 16:08:20". This feature enables cue sheets to fully represent playlists as extraction of a single track from a multi-track content file is now possible.
cPlay limits the number of tracks per cue sheet to 80.
- Faster RAM loading (both WAV and FLAC)
- Improved prefetcher performance
- More efficient ASIO output achieved
- Other minor improvements
Please REMOVE previous versions before installing cPlay 2.0b15. Normal (SSE2), SSSE3 and SSSE3 B9 versions are available. Running cPlay on a CPU not supporting the required instruction set will cause cPlay to exit immediately.
For cMP² users, review these BIOS changes (underclocks FSB, CPU & disable additional USB items). Note: E7200 has been replaced with E7300. Same settings apply. Recommended Gigabyte mobo GA-G31M-S2L is available.
Example of a cue sheet FILE section using new EOC directive to extract one track:
FILE "My Favourite CD.wav" WAVE
TRACK 05 AUDIO
TITLE "My favourite song"
PERFORMER "My Favourite Artist"
INDEX 01 12:51:10
EOC 16:04:20
An external change was planned however due to significant changes in this release and delays from external party, this change will only be available in the next release.
Edits: 01/05/09
Hi cics
Thank you for yet ANOTHER Sonic Upgrade... cPlay v2.0b15(poorboy-SSE2 version) reveals even more detail from the usual suspects - Bravo !
I must temper this enthusiasm though, with a Bug Report: "INDEX 00" handling in CUE sheets is problematic (under XP-SP3).
When I play a FLAC file via it's corresponding CUE file - and that cue file has "INDEX 00" entries - cPlay STOPS play at the "INDEX 00", and loads the next track from the "INDEX 01" point. Yes it's done in gapless fashion... but it skips the audio between the "00" and "01" points.
The same whole-album file played as a FLAC file (no CUE involved), runs without interruption, or loss of programme material - as indeed it should - being one long stream of music.
Maybe cPlay is treating the "INDEX 00" as if it were the EOC directive ? BTW, there are no EOC commands in the CUE files I'm using.
Thanks again for your commitment to Ultimate SQ !
ps: even on my less-than-perfect hardware platform, the benefit of disabling the Timer is clearly audible... thanks for the option !
Cheers,
Grant
That's not a Toy... IT'S A TOOL !!
When I play a FLAC file via it's corresponding CUE file - and that cue file has "INDEX 00" entries - cPlay STOPS play at the "INDEX 00", and loads the next track from the "INDEX 01" point. Yes it's done in gapless fashion... but it skips the audio between the "00" and "01" points.
That's exactly how cPlay implements it. I'm not sure if I understand the bug?
Hi cicsThe "bug" is when there's continuous sound playing from Track A to Track B, and a chunk is removed - due to the INDEX-00-stop-point/ INDEX-01-start-point behaviour of cPlay.
Unfortunately - when playing single-file albums via cue sheets containing "INDEX 00" entries - cPlay v2.0b15 offers the opposite of gapless playback ("gap-full" ?).
As an example, consider a single-file rip of Dark Side of The Moon . Many of the songs crossfade into one another, giving an unbroken stream of sound. For each song, "INDEX 01" marks the start point to which the CD player will cue if we use the SKIP controls. Reading the cuesheet for this file, it's obvious the times represent the offset from time = 00:00:00 (the start of the audio file).
Observe the span of time between the "00" and "01" in these few songs (they're very short intervals):
FILE "Pink Floyd - DSoTM [Jp.Black.Triangle.CP35-3017-2.1A1.de-emphasized].flac" WAVE
--TRACK 01 AUDIO
----TITLE "Speak To Me \ Breathe In The Air"
----PERFORMER "Pink Floyd"
----FLAGS PRE
----INDEX 01 00:00:00
--TRACK 02 AUDIO
----TITLE "On The Run"
----PERFORMER "Pink Floyd"
----FLAGS PRE
----INDEX 00 03:57:14
----INDEX 01 03:57:35
--TRACK 03 AUDIO
----TITLE "Time"
----PERFORMER "Pink Floyd"
----FLAGS PRE
----INDEX 00 07:29:10
----INDEX 01 07:29:32
..TRACK 04 AUDIO
....TITLE "The Great Gig In The Sky"
....PERFORMER "Pink Floyd"
....FLAGS PRE
....INDEX 00 14:34:21
....INDEX 01 14:34:42Another (more severe) example is Mind Revolutions from the Swedish band Kaipa:
FILE "Kaipa - Mindrevolutions.flac" WAVE
..TRACK 01 AUDIO
....TITLE "The Dodger"
....PERFORMER "Kaipa"
....INDEX 01 00:00:00
..TRACK 02 AUDIO
....TITLE "Electric Leaves"
....PERFORMER "Kaipa"
....INDEX 00 08:05:38
....INDEX 01 08:09:28
..TRACK 03 AUDIO
....TITLE "Shadows Of Time"
....PERFORMER "Kaipa"
....INDEX 00 12:19:42
....INDEX 01 12:23:32
..TRACK 04 AUDIO
....TITLE "A Pair Of Sunbeams"
....PERFORMER "Kaipa"
....INDEX 00 19:10:36
....INDEX 01 19:14:26Here the "00"-to-"01" times are huge - nearly 4 seconds long (what clown set these indexes up !)... resulting in significant chunks being lost-in-playback.
For the benefit of those who don't know, "INDEX 00" was originally intended to mark the beginning of a 2-second silent gap between tracks (as per Redbook standard). Older CD players will actually count DOWN the time between "00" and "01", and then begin to count UP elapsed time of the track from the "01" mark. These days, the use of "INDEX 00" seems rather haphazard (see Kaipa above), if it's included at all. The "standard" 2-second silence has been (thankfully) replaced with "audio-in-gap" mastering (that's what Wavelab calls it...), where audio playback is allowed during the gap, and the time interval between "00" and "01" is inconsequential for CD players.
But when EAC rips a CD and includes the "INDEX 00" lines in a cuesheet, we have a problem. BTW, I checked the Kaipa file with Winamp/mp3cue, and it has funky positioning there too. mp3cue uses "INDEX 00" as the offset point (not "INDEX 01" which properly identifies the "start" of a track), so selecting a song within the cuesheet list positions you to the final seconds of the prior track. Nice.
Yes, I suppose I (and all other cPlay users) could manually edit hundreds of cue files... but that's not really much fun.
cics, if you could program cPlay to IGNORE the "INDEX 00" entries completely, I think that would fix it. And of course, the EOC directive should remain (a really helpful addition).
Sorry for the long story... I hope it makes my point more clearly.
Thanks again,
GrantThat's not a Toy... IT'S A TOOL !!
Edits: 01/08/09 01/09/09
plays great on xp pro. awaiting more time with it before I assess it sonically.
It has happened twice in 24 hours---I play cplay, it works fine, then w/o warning something gets corrupt and I start getting a mettalic overlay to the sound which is gross not subtle. It goes away with rebooting. I first thought it was me playing, changing tracks within a file too fast (because Cplay is now very responsive and fast to re-act). When this happened with 14 I just hung up the player and could not get it to respond---so I would have to reboot. Now it almost sounds like the juli driver (which is 1.04) gets corrupted. I remember when I first loaded juli drivers, and had trouble, I got a very distorted similar sound. Any ideas?
Edits: 01/06/09
Try doing a clean driver reinstall as before.
I have gone an additional 24 hours with zero problems--I am being carefull to not confuse cplay. So I'm beginning to think how I run cplay may be the cause. And I am reluctant to do the driver reinstall because I would have to get out of minlogon. Last time this created lots of problems.
Let me ask you this: if you start a file playback session--and cplay is still loading either the 1st or 2nd track-- and then (because you change your mind) you request another track during this load period by double clicking on this other track--will this trip up cplay? It did for me on 14 but I would just lock up cplay and to get out I would just re-boot. Now the same thing happens with 15 I trip up cplay but it doesn't lock up. It gets this audio distortion which, yes it's similar to before when my driver was corrupted, but it goes away upon reboot.
if you start a file playback session--and cplay is still loading either the 1st or 2nd track-- and then (because you change your mind) you request another track during this load period by double clicking on this other track--will this trip up cplay?
No. As a torture test, I play a large WAV file for which I get 4 splits. Whilst one is loading I request another, etc.. No problems.
cPlay waits for one loading to complete before launching another. Likewise if you're in the middle of a load and you quit, cPlay waits until loading is complete before exiting. Connect HDD LED which you should see light up. That's why it may appear as if it's locked. HDD LED will indicate when load is complete after which cPlay would exit.
I never shut off any led's in my pc. which one is the hdd led, is it on the granite ps or is it just the power on led next to the pwer switch or is it on the mobo?
.
The HDD led does light up ok when I change tracks. Any way to use cmp or cplay diagnostics to help pinpoint the issue? Problem re-ocurred two times this am.
More data on my problem: it occurs at a frequency of at least once every 24 hours. Problem symptom is a hard metallic sound overriding the analogue signal that is gross. Reboot fixes it. I tried going back to version 14 of 2.0 cplay and the problem is still there. I never had this problem until recently when I went to 2.0 version 15 of cplay but, alas, going back to 14 does not fix it. So my guess is that the root cause is somewhere in the hardware or something common to both 14 and 15. I'm not knowledgeable enough on IT to surmise if it's the juli card or some aspect of the software or hardware in my dedicated cmp/cplay system.
If it's the soundcard or soundcard drivers I could try to reload drivers as Cics suggests (btw I'm using 1.04 driver) but to do this I have to toggle minlogon, reload some services and go through the process of unassembling and reassembling the soundcard into the pc. Last time I did this I caused a windows lock up and had to reload windows. So before I try again I would like to try some confirmation tests but I don't know of any. But I am definitely open to suggestions. Anyone have any ideas on what I might try (that is easy to implement) to confirm it is the soundcard or drivers or should I just bite the bullet and try to reload drivers?
Additionally I had a similar problem with this same soundcard (different drivers) on my anthlon based home computer. So having said that it sounds as if the problem might be in the soundcard hardware (actually this is what cics said a couple of posts above this). So if true a new esi juli sound card might fix this but why does a reboot make the problem go away?
Anyway I appreciate any help or suggestions.
I am having DAC/line stage problems so I have not had a chance to give #15 an audition. Nonetheless, I installed it and have let it run since installation.
Today I was doing some investigation on the line stage and to do this I turned it on. (My problem is intermittent, the most INTERESTING kind of problem). What I heard was a very bizarre distortion that had nothing to do with the line stage problem, which is composed of rushing sounds and thunderclaps.
Remembering Theo's posts I rebooted the computer and the sound returned to normal. Of course, then the line stage problems presented themselves.
I have no idea what would bring this on in my case since the machine has been playing the same album for the last six or so days. In my case it has nothing to do with changing what I want to hear quickly. Since the line stage was not plugged in I have no idea when this distortion started.
I have never experienced this before. Maybe I just have gremlins everywhere in my system. That's what I get for thinking it was sounding good.
Oy veh ...
I have no idea why my linestage started acting up either.
Bye,
Rick McInnis
I reloaded 14 (didn't like 15) and the problem did occur for a day or so then no issues for a week. Now I'm into 16 (or the latest as of today) and I love it. Will report under cics posting on 16 after a day or so. I never did reload juli drivers (too much hassle) but after thinking about my problem it may have something to do with electrostatic discharges in my rug onto which my bnc cable from juli to dac lies. I noticed I could affect my problem by moving the cable around.
This version functions properly so far in Vista 32 - thanks. I had occasional unwanted noises on some albums at 145/192, same as v2.0b12 but ok at 145/96. Also works in Vista 64, but I haven't used upsampling there yet. Sounds good in XP and Vista - alfred put it well.
The EOC feature is a good innovation for making mixed playlists from tracks in different single-file albums - I tried a few and it does work. But all this editing to make our own playlists, even from multiple-file cuesheets and with the help of Foobar (which leaves most of the work to manual editing with wav files) is a large PITA. Impossible to do impromptu playlists.
I must have some conflict in my Vista system, since you seem to have no problem with this version in Vista. Any idea what it might be, anything you've done in the past to make it work better in Vista, etc?
Wow, cics seems to remember all your system details! I'm not using asio4all, so can't comment on that.
If his suggestions don't work, get back to me - it could be a matter of the details of Vista OS optimizations, which hasn't been discussed here - there are a few important differences from XP.
I still think cPlay in cMP sounds better in XP, but as I'm using Vista for wasapi players (and one other asio player) I'm glad to be able to compare with an up-to-date cPlay.
Since I usually run cMP2 in XP, I have not yet pursued full optimizing of my vista system. I would appreciate a rundown of what changes to make to optimize. You can reply here or by email: riboge@gmail.com . Thanks.
Let's start from cics optimizations - almost all work in Vista 32 same as XP. Also, Ultimate, Business, Enterprise versions are, I believe, the only ones with gpedit so the only ones that allow you to set up lock privileges for A.W.E. in cPlay (same method as with XP). I have found it usually necessary to do all the optimizations (except Autoruns) in order to reboot into cMP mode, which is not the case for XP. Take this as a start point for discussion; I hope to learn new tips and correct mistakes with inmate input: Here are the differences in my Vista setup - otherwise assume cics XP optimizations are done as well:1. Set up real Administrator account and delete others. This just makes it easier to freely use Vista as you would XP: Go to Local Security Policy (Run > secpol.msc, or Control Panel > Administrative Tools > Local Security Policy) Expand Local Policies, click on Security Options. Set "Accounts: Administrator account status" to Enabled. Set "User Account Control: Admin Approval Mode for the Built-in Administrator account" to Disabled. Now log-off, and you'll see new account named "Administrator" is available, click on it to login. Go to User Accounts and delete the other account that was set up when Vista was installed.
2. Set up AutoLogon: Run > control userpasswords2 > highlight User Name "Administrator" > untick box "Users must enter a username....." > Apply > in new box that appears just click OK > OK again to finish. Now go back to Local Security Policy > Local Policies > Security Options: Enable "Interactive logon: Do not display last user name" (I do not fully understand why, but changes 1. and 2. seemed to make it easier to get into cMP mode, as well as making Vista easier to manage)
3. Probably goes without saying, but get rid of sidebar and all gadgets.
4. Skip the "sysoc.inf" file changes: it's not there. Go to Programs and Features > "Turn Windows Features on or off": untick everything and restart.
5. Power options: Use Power Saving plan, and customize to save power in every category, allow Processor to go from 0 to 100%, and experiment with HDD spin-down time to suit yourself. (tradeoff between delays in use and less HDD noise)
6. Turn off Windows Defender.
7. Registry:
HKEY_LOCAL_MACHINE > SYSTEM > CurrentControlSet > Control > Session Manager > Memory Management > PrefetchParameters. Set "Enable Superfetch" Value to 0 (Apparently Superfetch is the main Prefetcher in Vista.) Also, when you empty the Prefetch folder, do NOT delete the "ReadyBoot" folder.
[from PeterSt at XXHE - he found they stopped some unwanted Vista intrusions:] HKLM\Software\Microsoft\Dfrag\BootOptimizeFunction: Enter value N for Enable line. Also: HKLM\Software\Microsoft\Windows\CurrentVersion\Reliability: TimeStampInterval line: set value to 0
8. Device Manager: [I do Network and System Devices all at once, with no ill effects, but take it one by one if you want to be cautious] Disable all Network Adaptors, and the System devices as per cics. The Non-Plug & Play drivers: you think they are disabled after one reboot, but they always return after the second reboot! If anyone knows how to make them stay disabled, let me know.
9. Services*: [I do these in 2 swipes, which works for me, but again, change one by one if you want to be cautious, and also to get a feel for how they change the sound.] Leave on Automatic: DCOM, Group Policy, Plug & Play, RPC, Software Licensing, Task Scheduler, User Profile Service. (Task Scheduler will not start after Windows Event Log is disabled. Software Licensing can be stopped, but you will have functional problems, like not being able to open Control Panel, etc. User Profile can be stopped, but then you can't log on normally, and go to the Default Account, where, unlike with minlogon in XP, your functionality is hopelessly limited. Again, I welcome any advice on these tricky services). Disable: Base Filtering, COM+Event, Function Discovery, Internet Connection Sharing, Network Connections, Network List, Network Locator, Network Store, Offline files, Program Compatibility, ReadyBoost, Routing and Remote, Security Accounts, Shell Hardware, SSDP, Superfetch, System Event, Window Event Log, Windows Firewall, Windows Media Center Extender, Windows Search, Windows Update. Set the rest to Manual. (Note: Do not disable WMI, as cMP will not shut down properly going from XP-mode to cMP-mode.)
10. Autoruns: Lsa Providers Tab: HKLM\SYSTEM\CurrentControlSet\Control\Lsa\Security Packages: untick all EXCEPT "kerberos" (when disabled, Windows Event Log will not start when you want it to), and "msv1_0"
11. Advanced Boot commands: You can't edit this as in XP. Instead install EasyBCD (free). Click "Advanced Options" button > DEP/NoExecute drop-down - select "Always off", and then tick box for "Enble PAE" (unless you want PAE disabled). Restart computer. Check boot commands in: HKLM\System\CurrentControlSet\Control - "SystemStartOptions". It will read: "/EXECUTE /NOEXECUTE=ALWAYSOFF /PAE /NODEBUG" (or NOPAE if you didn't tick the box). This is as close as I can get to cics XP boot optimizations.Now, if anyone knows how to disable WRP (the Vista equivalent of WFP in XP), maybe we could try minlogon, but it's a no-go for Vista otherwise. As a result, it take about 30 seconds to restart Vista into cMP mode instead of just a few seconds for XP.
I've also just started playing with Vista 64, and whatever I do, it will not allow cMP to be set as shell when rebooting to cMP-mode from XP-mode, instead it just goes back to desktop.
*For wasapi players, you should leave Windows Audio services on Automatic.
Edits: 01/08/09
something I learned, in addition to all the right stuff in your post (let me qualify that - I don't know effects of the following on cMP or cPlay, neither of which I use):- User Profile Service should be left on, but process itself (one of the instances of svchost) can be killed afterwards with no visible ill effects;
- for my purposes (no cMP, no cPlay), Session Manager (smss.exe) can also be killed;
- WASAPI:I tried 2 different players that support both WASAPI and ASIO - Foobar (via plugins) and JRiver (natively). After doing all optimizations and stripping Vista to barely functional minimum, sound quality was pretty much indistinguishable between WASAPI and ASIO, with 3 services running that are required for WASAPI (Windows Audio and two others it depends on). After getting rid of these 3 services, ASIO sounded better than with them running.
- XXHighEnd player:Under configuration where both Foobar and JRiver WASAPI worked flawlessly, XXHighEnd complained that "Engine 3 is for Vista only" (no kidding!). I have no interest in pursuing the matter.
- XMPlay player:I have trouble comprehending what is possible target demographic group of this unbelievable piece of garbage. I mean, with both ASIO and WASAPI, it sounds like it was deliberately screwed up - and no, equalizer wasn't engaged.
Of course, my software of choice sounded better to me than anything else under Vista, the same as it does under XP - just not as good as under XP.
Edits: 01/07/09 01/07/09
Thanks for the tip on User Profile in Vista.
I agree that XP, at least the way I'm using it, sounds best. I think that in a not-very-optimized setup, Vista will sound better than XP - it has a certain clarity, especially in upper range, that beats XP - that is, until you seriously optimize XP.
Re: XXHE and engine #3 - if you start WMI service before playing in Vista, it will be ok. Without WMI, it thinks it is XP! I quite like the sound of XXHE, but it is not in the same league (for me) as cMP/cPlay, especially v.2.0b15. It is interesting to keep tabs on XXHE as it is constantly changing.
XMPlay - I also did not appreciate it.
Foobar and JRiver wasapi sound good in cMP in Vista, but they have that Vista "twang" for want of a better description (XXHE has less of this for some reason) I like that I can make playlists up so easily.
Winamp is still my non-cPlay favourite. I like your otachan plugin, but prefer KS - but I don't want to start beating that dead horse of an argument.
Is there something to be gained by using VISTA?
Is it simply a challenge?
Are you using this on a dedicated music machine or is this a multi purpose machine, but then why would you use many of the optimizations if that is the case?
My curiosity has got the best of me.
Thanks,
Rick McInnis
Yes, it all starts from curiosity! I wanted to hear what the wasapi players (including XXHE) were like because they are supposed to be the "inheritors" of kernel streaming, which I have always liked. And several inmates have been stating categorically that Vista is better. I wanted to hear for myself. To compare players, I use one of 2 machines with a cMP hardware setup, each with multiple system partitions. To be sure what I'm hearing I use the best one with Gigabyte/Juli@. I have a Vista system partition on that machine and 2 XP partitions. Until cMP/cPlay allows me to make quick spur-of-the-moment playlists easily, and in order to listen to non-wav or flac files, I will use other players along with cPlay. Therefore, I wanted to try them all out!
Incidentally, now that I have Vista more or less tamed (I hope it didn't hear me say that!) I think it is better for general use than XP - you would be amazed how easily soundcard drivers and software apps are installed, the OS itself is a dream to install, most functions have been quick and reliable. And I don't think the Vista audio sound is that much worse than XP. Now I wonder if we could be lucky, and Windows 7 turns out to be better for audio than XP and Vista...
It seems Vista treats virtual drivers differently. In cMP, set Optimize to "Player" when using ASIO4ALL in Vista. You may also want to increase latency. This approach is not recommended. Non of these issues occur with XP Pro.
Try not using USB approach (noise masks music information), instead go with PCI/PCIe using your Lynx card (uncheck "SynchroLock" feature).
I had not yet tried to 'optimize' in vista. After aggressively turning off services, uninstalling anti-virus, etc, cMP2 now runs beautifully in vista using asio4all with setting still on "critical". Beautiful sound. I still cannot get past the welcome screen trying to startup in cMP mode, however.As to the usb vs soundcard thing, I find that usb at 96 via EA Offramp to my dac especially using the high end usb cable I've recently acquired is superior to using my Lynx AES card to the same dac, even when I use dual connection for 192 with the latter. This is in either XP Pro or Vista.
Edits: 01/06/09
to optimize. It works fine set to "realtime".
Cics, could you explain the differences among the three optimize settings?
Some clarification needed when doing USB / PCI comparison:
- Which cPlay version was used?
- On PCI output, no other USB connection much be engaged
- On Lynx, SynchroLock must be unchecked (and also check high resolution clock)
- RPC
- Plug'n'Play
- DCOM Process Launcher
- Group Policy Client
That's what is really impossible to turn off - everything else must go.
How disable the 4 sevice?
I want disable it,but the button is grey and do it?
I had a few more than 4 so I pared down to the 4 with the result that cMP will now start in cMP mode but the stuttering problem returned. Files will play right only if sampling is set for the rate of the file, that is, no resampling up or down will work right. I will see if I can find one or two services whose reinstatement gives me both cMP mode startup and functional resampling. Any suggestions?
My XP partition is down for a bit, but trying this version out in vista introduced delays between sections of split tracks and even delays between tracks along with stuttering if the upsampler is engaged as with previous versions. 44.1 tracks play fine if cPlay is set for that rate and 96 files do fine if the setting is 96, but if the rate is set different from the source file there is stuttering. More rapid RAM loading and proper handling of index 00:00 are not manifest, so seemingly this version is not truly vista compatible. It's frustrating not to be able to try it yet in XP, where I imagine the improvements would show up.
hi cics,
2.15 incredible subfreq.information an depth of soundstage.In simple words much more details.
Thanks
alfred
I agree with Alfred.
And, cics, thanks for the bonus of including the "Timer" option. Turning this UI function off is definitely a nice improvement in SQ!
-edward
hi cics
after 35 active audiophile years and many audiophile revelations your way to sonic improvements are due to many hours of audible luckiness.
thank You
and happy new year
alfred
Dear cics,
I have to admit several times a day I look on the thread to see if the latest, and from what you have hinted a major step in cPLAY's evolution, "version" is available. After all it has been a whole twenty-five days since the last!
From the dizzying frequency of previous updates we have become spoiled looking forward to the expected new gift with confidence that it WILL be there? I am, it seems, an ungrateful child that once accustomed to kindnesses can only find happiness with more and MORE.
Would you care to give us an expected date of arrival or would you rather retain the element of surprise?
I have taken to going to the sourceforge site thinking that maybe you have posted the new software and haven't got around to making the announcement on the thread. I am ashamed.
Maybe you think that with 2008 being such a debased year that this should be an omen of a brilliant 2009? That would be understandable!
Wishing you and your family a happy new year.
Thanks,
Rick McInnis
well, I do the same
even I don't have a dedicated cMP2
I wish you all a better sound and music from your system for 2009
specially for you cics
very much appreciated
kocsu (from Peru)
Hello cics:
I install VISTA 64 BIT run cPlay well,it run 32 bit mode.
Can you release 64 BIT cPLAY version?
THANKS!
I did not have a lot of time in the last months, so I looked after the developments, but was not able to install and x-check (cplay was so good, Iwhat could have been improved anyhow ?). I just upgraded and I am really deeply impressed.
I am running here quite a nice equipment with the RAKK-DAC with Transformer-output (no active output-stage, it goes directly into the transformer from the DAC - extremly direct sound, crystal clear). The Pre-Amp is a Soulution 720, extremly resoluting as well and this drives two Pass F4 (DIY) and those drive again two line-sources with 12 ribbons and 18 Eton 4" (those avalon uses). So, I would say overall this chain shows you nicely if something is right or wrong.
The 1B19 was nice, but it still made some "sound". The re-sampler was in my eyes changing sound, but not improving necessarily, so I stayed with the 44.1khz. The treble was a bit pro-nounced, a bit technical. In 2.14B9: The treble became iammediately smoooooth...the soundstage more realistic, everything a bit better rendered. Then: I just tried the upsampler to re-confirm my prejudices...but the opposite happend: It simply added a lot of resolution, it took aways a layer of grey...wonderful.
The only drawback: I need more CPU-Power. The 1000Mhz is not enough, I get hick-ups (Julia at 64). With 2.4Ghz it is playing fine, but what is you lower limit in CPU-power ? I read as well that CMP-options will influence the usage of CPU ? What is the best practise when you want 178k-192k-upsampling ?
That's a huge step change in SQ. Improving on B14 is going to be a big challenge!
cPlay 2.0b15 will reduce CPU load by a significant amount while doing full 192k upsampling from 44.1k. Setting Power Option to "Always On" would help a little in the mean time. As always, keep Optimize setting in cMP to "Critical" when using ASIO.
Did I mention that I have created two pure CMP-Machines ? I was so bought into CMP / Cplay that I could not resist to have one machine with all the stuff I liked best: P900 board. e2180. Juli@. Silverstone passive PSU. CMP+Cplay 1B19. This machine runs at 600mhz !!!
I have build a second machine based on the actual recommendation Gigabyte G31, e7200, same psu, Juli@ as well...On this second machine I ungraded from 1B19 to Cplay 2.14...
...today I started to do now the shoot-out of the gigants: the legacy system with the cplay I liked best against the latest machine...
CICs, on the latest machine the step from cplay 1B19 to 2.14 is very signoificant. The latest version as well has a wonderful resolution. BUt.....It still sounds "Hifi" while the old setup simply generates music...there is this feeling that you are sitting in a church or in a concert hall, while on the new machine you have the impression that you have a reproduction...hard to describe...I guess I will go an buy now some spare P900-boards...and: It will be very interesting to see how the Intel Atom will do...
N/T
I have the Biostar P4M900-M7 SE with an Intel E2160. Can you tell me how you achieved 600MHz?? Is your FSB at 100MHz? If so, how do you achieve that? My BIOS does not allow lower than 200MHz.
Thanks.
It is a bug of this board...If you reset your board to default settings and than overwrite with 100 mhz, even if it does not show, it should re-boot with 1000mhz in the strat-up and late through EIST it will reduce the multi so that you get 600mhz---that is what CPU-Z is reporting.
nevertheless...we are talking about nuances...the change of the digital cable was more shocking than the difference between the two machines. So, I guess all we decribe here is very relative to the equipment you use. CICS with his Scarlatti DAC will have for sure different results than I have, but...the RAKK-DAC with passive output has beateen the 15000$ Accuphase DC-801 in my chain...so it is not that bad at all...
I was unable to re-create the situation after BIOS update.
To tell the truth, I'm usually able to hear slightest changes in sound quality (OS changes, cables, compilation settings for ASIO plugin etc.) - but I was unable to hear any difference between "before" and "after" that BIOS update - and I specifically listened right before, and immediately after. So, I never bothered to downgrade BIOS, or anything like that.
...I can only see the vista version for download on the official page...
nt
cics, and everyone.What do you think about offline upsampling?
By upsample before play music,
will it eliminate the jitter made by realtime upsample?
Did look at this before. We have 2 problems:
- Requires much more RAM (causing more unwanted interference)
- RAM load takes way too long (5 min track requires ±2.5 min at max CPU speed)
With cPlay 2.0b15, there's significant CPU load reduction. It's fully operational but cannot be released yet (due to external change required). If upsampling 'cost' is low enough we could have a tradeoff (more RAM vs more CPU load)...
Dear cics,
Can you tell us what this will be so we can have one on hand when you release this much anticipated update?
Thanks,
Rick McInnis
Same hardware will apply.
There will be a BIOS change to fully exploit b15.
I was hoping for the best.
I have not got it configured yet. I forgot how much work is involved.
First thing I recommend is DO NOT do this with a USB-CD. I will start over tomorrow when I bring a "regular one" home from work. I got as far as starting to disable WFP but I cannot get WINDOWS disk to "boot". With as little promise as I see this is becoming a labor of love.
Well, BIOS is an interesting mix. You can disable SPREAD SPECTRUM, but you cannot set the CPU frequency, locked at 200 mHz, though you can set the multiplier. With e5200 the choices for memory multiplier are AUTO, 3.3, and +4 (whatever that means!). I seem to get the same numbers with AUTO as with 3.3. You have to force the 3,3,3,10 latency numbers.
Another strange thing is you have the FINE adjust (0.5x) for CPU frequency but I cannot tell if it is actually being used. I initially set this for 6 plus 0.5, yet in one of those BIOS screens before you get to WINDOWS it would say 6. With 6.5 x 200 and forced latency the FSB:DRAM is 3:5. I have no idea how to get this to 1:1 since I realy have no idea what it means. All I know is that cics says that 1:1 is ideal.
I suspect that substituting e7200 would allow slowing down the memory based upon what the g31 board "did" when e5200 was installed in it.
CPU voltage can be set to 0.9000 at its lowest setting.
There is no de-over voltage (or whatever that is) setting.
There is no CPU-EIST control (I hope I am getting most of this right it comes from memory and it is late and I am a bit disappointed in what this thing seems to be).
Forgive this haphazard report.
Must say that segars instructions for updating JULI@ drivers made that task easy THANKS!!!
Since I did not implement minlogon I did not take any power measurements tonight. I hope to do so tomorrow.
In addition to this my line stage's right channel tube started making incredible noises and I will have to replace this before any listening can be done. Anyone know of a good source for 6N6p's?
This MB takes forever to do anything after you turn it on. Unlike the G31 board it does not make the lights on the keyboard light up just before coming to the screen that lets you access the BIOS. After what seems an eternity the screen pops up and if you want to make changes to the BIOS you better be paying attention. There were folks on newegg who talked of how slow the board is to come alive. They were not kidding. I kept thinking with all of the restarting that something had happened and that BIOS had to be reset only to have the thing suddenly decide to make itself usable. This did make making the WINDOWS changes take FOREVER. Unlike the previous boards that seemed to get quicker with each round of changes, this thing is just as slow at the end as it was at the beginning.
Unless cics can get GIGABYTE to give us a cics BIOS version I doubt this is going to be the answer.
I will have more to report tomorrow. (!!!)
Bye,
Rick McInnis
Well, this thing certainly does not sound bad.
Of course, my comments have to be tempered by the fact that I need to go back to G31 to make a comparison.
It still makes the "pop" sound every couple of seconds which ruins listening to anything soft. I forced the memory to 3,3,3,10, even though it seems about every time I open up BIOS it has reverted itself back to automatic, so who knows, and once I am in cMP, if I use explore to use CPUZ I have to restart and who knows what is happening then ...
Friday I will leave it on AUTO and see what happens. Could the PRIORITY SET (I think that is what it is called) setting as recommended by carcass (wasn't it) could that be causing a problem with this board? Does anyone have an idea of WHAT could be causing this? It is not going away. It is/was not a breaking in noise.
Bass seems to be even more powerful. This board has many pleasing qualities I think are superior to G31, but until I go back it could be due to the fact that I missed getting to listen for a few nights and just about anything sounds good. One has to be immersed into the device and then you start hearing all of the bad stuff.
Any suggestions on the "pop" phenomena would be appreciated.
Bye,
Rick McInnis
Changing PCI latency can fix dropouts, if they are caused by some device hogging bus by the means of using very high latency setting - e.g. for many videocards it's set to max (248) for no reason whatsoever. It's obviously not the case for you.
I would speculate, if this mobo requires larger buffer for ASIO driver, it doesn't have a good chance to be better SQ-wise.
Dear carcass,
I am not sure I get what you are saying.
If you are responding to my comment about the "controversial tweak" that was a GREAT tweak, I was just wondering if that could be confusing this board.
After reading cics's post I am tending to be in agreement with you. Unless cics has something "up his sleeve" I am thinking this is a dead end.
I will try his suggestions to eliminate the periodic burps (about every two seconds) and let him know, but unless I am convinced to keep this in my "machine" to try other suggestions with the intention that this CAN be a good MB for this purpose, I suspect Saturday will be spent re-installing the G31 board.
It does seem that the suggestions given by you and cics to stop the burps would assure that the sound quality of this board is never going to approach what the G31 delivers.
I wish I could pretend the board was defective and send it back!!!
Bye,
Rick McInnis
For dropout problem ("pops"), try:
- Set Juli@ latency to 64 samples
- In Windows, Under "Power Options", set scheme to "Always On" and all else to "Never". Reboot.
- Set cPlay's DSP Buffer to Medium
These recommendations sound like back tracking to MAKE it work. Which I can see would make for an interesting academic exercise, and I am glad to try and see what happens. BUT, do you think this board has potential to be as good, or better, than the G31?
My curiosity gets the better of me.
Do you think I should try e7200?
Do you think my forced memory settings are contributing to these noises?
I will try the above and see what happens and let you know.
I will let you know tomorrow.
THANKS,
Rick McInnis
Device Manager > Show > Resources by connection > Interrupt request
If another device has same interrupt as Juli@, use other PCI slot and check. If that doesn't work then you need to disable the 'offending' device (often the network or USB).
this board has no chance of being compared to the G31 board.
It continues to make the popping sound. After making about ten changes of album it goes into SUPER popping mode and freezes up requiring it to be turned off.
It simply is not getting any better.
I (who admittedly has no idea what he is talking about) cannot see any potential for this to equal, much less better the G31. I had hopes that this might be something that furthered the ART OF COMPUTER TRANSPORTS but it seems in every way you have taught us to evaluate the goodness of a piece used in a artful computer transport to fail. It uses more power, it cannot use minimal voltage with a chip that would boot with 0.85 volts, so what it the point of even using the w5200? It will not run the memory at anything less than 5,5,5,10. It requires less advantageous latency settings (I presume) in both JULI@ and does not allow a large DSP buffer in cPLAY. I set the buffer to auto and small, neither made any difference. I did not try any other settings of latency within JULI@. I figured it was pointless.
It is not that it produces a "bad" sound, it is marred by the constant POP, which if it was a simple LP-like pop would be one thing, but it is not a quick POP it has a slow rise time and corresponding slow decay.
Of course, cics, if you want me to keep this thing in my system for things you would like to test, I will. Otherwise, for the sake of enjoying listening to music I will have to return to G31. It is very hard to enjoy using this. It was initially entertaining to evaluate it but ...
I have no idea what it needs to be useful, but it is obviously something big.
Bye,
Rick McInnis
Use DPC Latency checker tool. You should get a clean run after startup of no more than 4-5µs. Otherwise, problem is in the BIOS.
Dear cics,
This is what I found:
within XP at Desktop - 436 uS (this is a peak "blip", as all of the measurements are with a "floor" in the 6 uS range)
cMP with XP - 386 uS
Starting with cMP, and using the EXPLORE tab - 602 uS - this board HATES cMP.
Something very wrong with this board for audio. Do you think it is defective. though it seems to do what is asked?
If you think this is abnormal I will return it and ask for another one. Otherwise, it goes to the office computer!
Hope this has been of use.
Thanks,
Rick McInnis
and running the "last test" - it is a steady 9 uS, no blips at all.
Theo, I checked the BIOS all of the time. It would keep reverting back to certain settings regardless of the weather.
I was finally able to verify it was not sharing with any other device.
Either this board is defective, which is what I am telling newegg, or made for purposes other than ours.
Good riddance. One less thing to wonder about.
Bye,
Rick McInnis
this may help or not. I started getting to where my sys was sounding off. So I checked juli latency (it was at 256) I reset to 48 and started getting pops. So I double check my bios settings (had a brief power outage last night) and found host clock control not enabled and a few other things amiss. When I reset per spec all was good.
Hi Rick,
You probably know this already, but the G31 allows the user to assign Juli@ a dedicated IRQ through BIOS settings. If the Q35 BIOS offers that same option, it may help with the popping.
Hope you can make the Q35 work.
Alex
Dear cics,
Remembering something about this in THE ART OF COMPUTER TRANSPORTS I think I saw how to check this.
I see JULI@ at PCI 20 and it is the only thing listed as being at "20".
Is this correct? Everything on that list has a different number next to it so I assume nothing is being shared?
Let me know.
This means that the noises are still very annoying. Softer passages are completely obnoxious. Louder musics are not as noticeable.
What do you think?
Bye,
Rick McInnis
Dear cics,
I am afraid the instructions you gave are in shorthand and I am do not see how to do this.
There are steps missing that I would like to know but ...
I looked around at various things within the list but I could not find anything that I knew what to do about.
Let me know.
Thanks,
Rick McInnis
I will try this tomorrow.
The settings you recommended yesterday did nothing to improve the problem.
Do you think I should stick with e5200 or try out e7200? Let me know.
I know I did not check for dedicated IRQ during install so this could be it.
If the above works I will go back to "typical" recommendation and see if that is possible.
Thanks,
Rick McInnis
Up and running with all cics recommended settings (to the best of my ability with no instruction) first off - it is not using less power, as it stands, unless new BIOS settings would make a significant difference, in relative terms, this one uses about 4 more watts; 26/27 at idle in cMP and 30 while playing music.
It seems to make more pops than G31 ever did in the break-in stage.
It is not without its charms. Even with the pops it does do some things very well. It is nowhere close to being unlistenable (spell checker tells me this is not a word). If one had never heard G31 they could (I am assuming that further break-in eliminates the pops) live with it quite peaceably.
I was listening and realized I had not installed the MB drivers. It did not sound any different after wards. There seemed to be in my assumption only one driver that really only concerned the MB, the rest were media and USB drivers. I figure I do not need those.
I installed minlogon with no problems and the JULI@ drivers (1.04) went without a hitch. I installed them straight from the USB stick. Before I was under the VERY mistaken assumption I had to install the supplied dick drivers first. One can learn something everyday, can't they?
I will be able to return to G31 easily with this practice.
Well, all in all, I do not think this is going anywhere unless cics can get them to make big BIOS changes and even then, I wonder if that would make much of a difference in his theoretical ideal of minimal power usage. I do not think this MB offers any reduction in power usage at all. Of course, what do I know? I will probably insert the e7200 in just to see what happens with the BIOS settings.
First impressions of the sound (kind of like drinking really old wine and having to divine the flavors within the mold) but there is a form of improved clarity. G31 seems to resolve ambiance to the point of muddying the presentation. This board presents the ambiance but places it behind the real musical action where G31 (from memory) almost seems to place it in front of the performance (if that could possible make any sense at all). This was an impression made under less than ideal conditions so take it as nothing more than a first impression. I am leaving it running for the next 24 hours. Who knows what it may sound like tomorrow evening and the the evening after that? I might as well keep it in there awhile.
cics, do you have anything to offer? Anything you would like me to try?
Anyone else? let me know. After giving this a three day settling in I will try out e7200. I bet it does sound better and I doubt it will use up any additional power. I bet it will allow slower memory speeds. A moot point for those of us with G31 boards but this could be a decent choice for others, especially with a special BIOS for audio.
Bye,
Rick McInnis
It's kind of you to be our guinea pig with this mobo. I'm interested in the Bios profiling that this board is supposed to be capable of, unlike the G31. When you get comfortable with the important things, can you check out how (and if) this profiling works? It would allow quick changes back and forth to previously set Bios configs - useful for multi-purpose, multi-system computers.
I would be glad to experiment but you must not assume I know what I am doing!!!
I am not familiar with this. How do I do it OR where can I see how to do it?
At this point, I have everything set to as close as one can with this board to cics's recommendations.
Just let me know in more detail.
Bye,
Rick McInnis
Thanks, but Ryelands has pointed out that the G31 does Bios profiling - using F11 button from main menu, so no worries!
BIOS profiling works fine on my G31 MoBo. In the main menu, hit F11 to name and save the current profile or F12 to restore a previously-saved one. It saves hours when e.g. lowering voltages.
I find the G31 BIOS to be one of most flexible and least bug-ridden BIOS setups I've ever used - if only they were all as good.
Dave
Thanks! There it is, right under my nose!
great idea and another example of the advantage of open code
CPU & FSB are very important. North Bridge is also great to have.
Anything under GTLREF?
Power consumption at 30 watts is very good (G31 does go lower to 28 watts). These are figures at AC outlet which means cMP² is consuming even less! With B15 we should get it lower.
Dear cics,
I have no idea what you mean about the North Bridge! Explain, please.
I listened a short time this morning and the pops have disappeared. Last night it sounded like I was listening to a bad LP.
The G31 boards did not do this at all. The BIOSTAR boards sounded bad when first hooked up. This one does not sound bad but it sure made a lot of extraneous noises.
SO, the higher power consumption is OK? I am confused. This is a littel like colder weather proves the earth is getting warmer. I could not resist the analogy.
There are nowhere near the choices on the control/F1 MIT menu. No GTLREF, no Host Frequency control (set to 200), no FSB De Overvoltage Control.
There is no CPU EIST to enable, maybe it is enabled by default but this is not listed.
My CPU temperature is running a few degrees hotter. The weather here has been warmer the last few days, this could be the reason.
Tonight I will have time to give it a serious listen.
I have been in such a rush to get it up and running I did not make any written notes, going by memory, I will try to make it clear what is and is not in the BIOS.
Do you think it is worthwhile to try it with e7200?
Let me know of anything you want me to try.
Bye, and THANKS,
Rick McInnis
Hi Rick
Thanks for taking the time to investigate this mobo choice. Yikes... sounds like it won't get a cMP-recommendation with all the fuss and limitations you recount.
Sure warms my heart to gaze at the shelf where the GA-G31M-S2L is patiently waiting for me to get my thumb out...
Good luck with the Q35.
Cheers,
Grant
That's not a Toy... IT'S A TOOL !!
I was tired and frustrated when I wrote my note.
I wonder if inserting the e7200 will allow better memory settings.
Of course, if cics can get them to offer a special BIOS for this board that would change everything.
I hope to get it making music of some kind tonight.
Thanks for the thanks. It is not hard work, but it does take lots of time.
I thought I had made notes of all the important changes. I do not seem to remember the most recent kernel change decision. The one where we added \ 3gb and then I think we took it away. Was there something else added in addition to the changes from AOB? That is the only thing I do not seem to have kept a note about.
I will keep you posted.
Bye,
Rick McInnis
Of course, I could try it for myself, but thought maybe someone knows.
I am wanting to see if using linear supplies improves the sound we can get with cMP squared.
I am going to use a big battery for the 12 volts rail. I have mentioned before that I will use the ANTEC for powering relays to preserve the turn on sequence. I will have to retain this for pins 8, 9, and 16 but I figure I can lengthen the wires to allow placing the ANTEC outside of the box.
It does not seem that huge supplies are needed for these rails. The really high currents run through the 12 volts.
If anyone out there has something to add to this, I would be grateful.
Thanks,
Rick McInnis
http://www.mini-box.com/M2-ATX-HV-140w-Intelligent-Automotive-DC-DC-Power-Supply_2?sc=8&category=981
This looks like it may be ideal to power the cpu by battery. Can even be connected to the on button with the standard 2 pin connector. In-put up to 32V means a large Li-Ion power pack could also be used. Lots of possibilities
An Interesting battery Power supply / Charger:
http://www.audio-magus.com/PowerCenter_Battery_Power_Suppy_p/bps1.htm
PC XPpro TC Electronic Konnekt-8 firewire i/o, 2-Tact 2150's, Genesis 500 modified speakers, Virtual Dynamics & XLO cables
The problem with dc-dc converters is that they, too, are switchers and that really doesn't do you any good at all.
From what I can tell nothing beats a large SLA for low resistance and long term use. They are big but nothing I have seen that I can buy comes close to their performance.
I did try one of those "things" per Bernd's recommendation. Unfortunately I modified it to death or it just died all by itself. Again, an interesting aside but not a definitive answer as to whether a linear supply would be superior for a music computer.
Thanks,
Rick McInnis
Rick,
A large SLA with capacitor filter bank can indeed kill small components. Two months ago, I burned a spdif clock generator by feeding SLA with capacitor filter. The spdif clock generator works fine with SLA for weeks until I feed it with a bigger SLA and cap with lower impedance. There is a very large surge current from the SLA, and you may want to build a soft start device to limit the initial surge current.
hi,
as i mentioned earlier better than one battery are more batteries parallel to minimize ers.the damage problem is not the current of sla's but often voltage problems. after charging voltage drops slowly from 14,4 to 12,6 volt and some 12V devices are damaged even with 12,6 Volt.
bye
alfred
Dear promethk and soncis,I had not used the voltage dropping diode scheme with the picopsu and WILL with whatever I try next. I know it is important to get as close to 12.0 volts as possible.
I measured the voltage of the battery before connecting and it was 12.5 volts. Being a new battery it might have been on the shelf longer than the seller would like to admit. It did not change while connected.
Do you know of a way to limit the turn-on current from a battery?
Thanks,
Rick McInnis
Edits: 12/15/08
According to the measurements someone made on a biostar motherboard, significant power is drawn on 12V, 5V and 3.3V. Maybe you can get away with the -12V, I guess.
Hello everybody. I am back again after minlogon crash and psu + mb crash in this computer. Having to reassemble it quickly to go on working I had to use my Biostar mb that was installed in my CMP. To remain with music and to experiment a little I enabled Azalia (if I spell it right) built-in soundcard, and lo and behold! It easily catches Asio-for-all and as easily beats Creative Audigy 2 gold (platinum, diamond...). So to those undecided about a sound card, if You hesitate which one to buy, don't hesitate and try Azalia. Very meaty sound, no loss of fine detail. More wholesome mid and even more formidable bass. Try it.
Serge
P.S. Maybe Presto was right after all when he said, that built-in cards are probably capable of bit-perfect playback...
Below is screenshot from the manual I downloaded - I believe any further comments are unnecessary.Of course, if there's somebody who believes it's possible to run memory at default timings and get optimal sound quality, that's another story...
Edits: 12/12/08
this board allows more control with the "control/F1" keystroke.
If you look at the manual for the G31 board you would say the same thing.
I have one due to arrive on Monday and will let all interested know what it allows by the middle of the week.
We'll see. If it doesn't work well for audio I will put it in the office computer, which is using my spare G31 board, and keep THAT just in case something goes wrong.
Bye,
Rick McInnis
N/T
HELLO:
I want buy a soundcard,for playback music only(use foobar or cPLAY)
Please recommand a best C/P soundcard for me.THANKS!
JULI or RME9632 or....?
Which is best?analog out
9632/Lynx22/Juli?or other?
please recomand your choice!
THANKS!
Which is best?analog out
9632/Lynx22/Juli?or other?
please recomand your choice!
THANKS!
Tried several soundcards; RME, Maudio and juli@
I won't say juli@ has the best hardware, but is has excellent asio drivers and therefore sounds the best in my opinion.
Juli@ analog out & digital out is both good?
.
Dear cics,
I figure the answer will be YES but I thought I would ask anyway.
Even if I delete all MB drivers before making the change?
I have WINDOWS in a separate sector on the drive so it is not that big of a deal, though implementing MINLOGON always seems to present a challenge, along with JULI@ installation. If you did those things every day it would be easy but one seems to have to re-learn it every time.
I want it to be as fair an assessment as possible. The thought occurs, what if it doesn't sound as good and I have to do IT all again to replace G31? Oh, well, who said audio obsession was not without its miseries?
Also, what do you think of the short-cut through SAFE MODE approach to installing MINLOGON?
Thanks,
Rick McInnis
For a fair comparison, its best to do a clean install. Windows on a separate drive makes this a 1-2 hour process. Burn-in for at least 96 hours.
Never tried Safe Mode minlogon option at all. I'm used to the normal method.
Is it possible to get a power meter at AC for computer? I'm very interested in getting a reading on this hardware (Q35 and E5200). Also your E5200 goes down to 0.85000V (rather settle with 1 notch up @ 0.85625V).
Dear cics,
I will hook the above mentioned device in (it plugs into the AC outlet and and you plug the device to be measured into it) before I make the change AND afterwards!
I had used this device to see how much power various appliances were using.
Even if its measurements are not absolute we can hope that it will give us some relative comparison.
I will let you know.
It is not doing the re-install that I dread. but all of the turning off of services! Nonetheless, it is the least I can do for you. I hope I can come up with something useful to you.
I will do the traditional MINLOGON install.
Bye, and THANKS!,
Rick McInnis
With my G31 board and CPU voltage set as low as I can get this pairing to go (G31 & e7200) which is two notches above the desired 0.875:
in cMP at idle - 24 Watts
while playing music - 28 Watts
There seems to be no deviation while playing. I tried various selections to see if that made any difference - it doesn't.
Will let you know what Q35 does next week.
Bye,
Rick McInnis
"While playing music" I assume 44.1 to 192k output.
So reading at AC at 28W and assuming PSU has 80% efficiency, that makes cMP² power consumption at 22.4W!
I'm hoping to go even lower with 2.0b15.
except that I cannot get any lower with CPU voltage.
Will give the comparative numbers for Q35 Tueday.
Bye,
Rick McInnis
The legendary Juli@ has just arrived at my doorstep today. The card is still sitting in the original packaging right now, until I can figure out which driver version goes best with it. From what I've gathered by digging up the old threads, the more vocal members of our community seems to be using (or used) versions .979, 1.04, and 1.21. Are they still the choice driver versions today?
Please make me some recommendations, thank you!
Alex
*WOW* Just compared v1.04 vs. v1.23: They are really different via cplay+asio. I was so skeptical when I initially read this thread.
1. Are we to believe ESI thought v1.04 sounded "wrong" and then corrected it with v1.23 which is supposed to sound "correct"?
2. Are the engineers' ears at ESI so different than the other contributors to this thread who've made the comparison?
A big thank you to Martin, Rick, and Seger for helping out. Sorry for not replying earlier, but my "new" system had taken up almost all of my free time.
Grant, glad to know that you're enjoying Juli@+1.04 like I am. After spending the past couple days listening to my system, I'm pleasantly surprised by the Juli@'s analog output quality. The SQ is a significantly improvement over 9632 analog, and maybe even better than the various 9632+"lower-end" DAC combos.
Thanks to Grant, I still haven't the opportunity to experience the legendary digital output of Juli@. I planned on using a custom made cable, which replaces an (RCA/BNC) connector on one end with a high quality version of the female "PS/2-like" connector. But after reading Grant's incredibly informative and detailed posts, I find myself wavering between my conventional-but-reversible solution, and his exciting-but-permenant modification.
Has anyone compared between the two? How much of an improvement in SQ does the latter offer (theorefically)?
Thanks,
Alex
and convert to bnc connector in place of the RCA assuming you have access to a BNC terminated cable, which is the better way to go.
Whether you simply replace the female RCA at the the end of the break-out cable or go straight into the board, which is MUCH better, even the former improves upon the stock set-up. Not to say that is sounds BAD, not at all, just the other approaches sound even better.
Best of luck. Getting those drivers installed may test your patience.
Resign yourself to doing it wrong a couple of times so you won't be disappointed.
Bye,
Rick McInnis
Hi Rick,
Your series of daring and successful modding posts has inspired me to enhance my Juli@ using the exact same methods. I'm not known for my knowledge in electronics, however, so I may need some extra help from the expert.
Can you please offer more details on how and what you did to improve your Juli@? What type of capacitors are being used (any different from the ones you mentioned back in August?), and how many of them should I add? Anything I should watch out for during the soldering process? Which pins should I solder the caps on? A close-up photo of your Juli@ may be too much to ask for, but it would be most helpful.
Thank you again for the inspiration!
Alex
Alex,
Without the inspiration/provocation of promethk and sonics/(Albert), most notably, I would have only thought about such things. They have been the leaders in power supply improvements.
Look here:
http://www.audioasylum.com/cgi/m.pl?forum=pcaudio&n=35497&highlight=promethk&r=&session=
The photograph promethk supplied makes sense after you realize what it is showing you. There are 10 pins at the connector/header to the analogue section. Pay attention to how these are numbered.
I ended up using different pins than those suggested by pronmethk which allowed a more orderly installation of the caps. All of this is in the posts somewhere. I did not take notes so I do not have it at the ready.
I used 100uF 50V Black Gate N's because they were here and they are very good. I have not tried any other brand.
Consider adding capacitance to the power headers on the MB. I believe Alfred has posted some pictures of a very fine installation. I used his perf card approach for the 12vATX header near the CPU. I attached capacitors to the individual wires for the main header. Not that I think it is better.
If you are going to "go for it" install BNCs for the input and output of JULI@, there are numerous posts on this. Check out Grant's installation
photos. All of this is within the thread. When you see a post by promethk, sonics, or gjwAudio (Grant) read it! I know I do.
Grant certainly has put the icing on the cake with his collaborative executions of these mods with his mysterious tech he has given the codename BOB.
http://www.audioasylum.com/forums/pcaudio/messages/3/39956.html
I am about to use a LARGE 12 volts battery to take on all of the 12 volts rails. My idea is to have the regular PS power a relay so the turn on sequence can be obeyed and then have the battery connected to all of the 12 volts inputs.
If this works I want to replace the other two rails with linear supplies, again using the standard PS to actuate relays. This way I can move this out of the box and remove a source of radiation that many have speculated as being a big problem.
I hope this makes some sense.
Rick McInnis
Hi Alex
Thank you SO MUCH for asking this question... The issue had somehow escaped me, and I blindly installed the latest (v1.23) Juli@ drivers when I got the card.
Imagine my surprise to learn I was using the wrong (...or not-the-most-highly-recommended) drivers for best SQ. Stunned.
So last night I removed v1.23 and installed v1.04 (in one go, Rick)... including the fuss with Minilogon and certain services for a smooth install.
YIKES... it's like getting a new version of cPlay ! TB II was never so engaging.
Happy Listening To All...
Grant
That's not a Toy... IT'S A TOOL !!
Not me for the Juli@ mods - try sonics or Rick! Probably you are looking for a recommendation from cics, but in case he's busy: I find 1.04 is the best for asio, 1.23 is best for kernel streaming, for instance if you want to use players like Winamp and Foobar within cMP. Unfortunately, 1.23 drivers sound bad with asio.
Hi Alex,
I use the version 1.04 driver and it works just fine with XP Pro.
You should also consider removing the analog section, which I assume you won't use. 4 screws to remove and totally reversible.
If you want to go further, you can add filtration caps to the pins where the analog section used to be plugged. You can find a thread about that somewhere here, I think (not sure) it was posted by Seger a few months ago.
Next, if you use the coax output you should definitely upgrade the rca connector and the cheap cable.
More on that subject: http://www.audioasylum.com/forums/pcaudio/messages/3/39956.html
Happy listening
Martin
Is it normal that cplat or Cmp/cplay doesn't use dual core on X2 processor?
I cannot upsample higher then 96 khz, because it is to much for 1 core.
At 96khz one core is at 90% (X2@2400mhz) The other one is 0%
Is this old news?
(Couldn't find anything by using search function)
The options:
1. Player: cMP does nothing and load is spread across CPU cores
2. Realtime: use with non-ASIO players
3. Critical: use with ASIO players
Option 2 & 3 locks DSP load to a specific CPU core. Upsampling is not multi-threaded and therefore load cannot be split. Here's more info .
Release notes . Notes on SSSE3 has changed:
cPlay is available in 3 flavors: normal (SSE2), SSSE3 and SSSE3 B9. Normal requires a Pentium 4 type processor (SSE2). SSSE3 version runs on processors supporting SSSE3 instruction set. Intel (Core 2 Duo etc.) and AMD have processors supporting SSSE3. Use CPU-Z to determine if your CPU supports SSE2 or SSSE3.
Release 2.x introduces SSSE3 'B9' option that offers most aggressive optimisations. This option may not work on some setups.
Change Log (2.0b14 release):
- Minor UI refinement: "no media" message and display information (ellapsed time and volume) in default colour. "Flicker" problem fixed
- DSP refinement: SRC upsampling redesigned giving reduced RAM writes (~30% less) and much smaller L2 cache footprint
Please REMOVE previous versions before installing cPlay 2.0b14. Normal (SSE2), SSSE3 and SSSE3 B9 versions are available.
For cMP² users, review these BIOS changes (underclocks FSB, CPU & disable additional USB items). Note: E7200 has been replaced with E7300. Same settings apply. Gigabyte mobo may be difficult to find, suggested new mobo is Gigabyte's GA-Q35M-S2 .
There is a new web site selling 24/192 files (just a few albums but excerpts from a Mahler 6th demo available for free) so I am curious before I spend $ on this what is the largest file or track one can play with cplay/cmp? BTW the demo sounds great on cplay!
.
I have noticed with b13 and more with b14 that when it moves to a new track play starts on time seconds before the display changes both the highlight of the track and the time. The latter jumps from the end time of the last track to :03 or :04 of the new one. This is no real functional problem but if it can be corrected with no sq cost, it should be.
Edits: 12/09/08
Hmmm... seems like a good place to repeat this request.
As stated before, playback IS affected by this navigational artifact: when you change tracks manually - and the cuesheet contains "INDEX 00"entries - cPlay begins at the "00" point. Generally it's heard as a 2 second delay preceding your song - unless there is audible signal there - in which case you get a little tase of the finale from the previous CD track.
As an example, try it with Dark Side of The Moon -or- Abbey road . Definitely not what the artist intended.
If it doesn't affect SQ, please consider it on an upcoming "to-do" list.
Thanks again cics... this is one amazing "music appliance" !!
Cheers,
Grant
That's not a Toy... IT'S A TOOL !!
I guess I wasn't clear that I was referring to when cPlay moves itself to the first or next track not when it is done manually. I have edited to make that clear.
.
That's not a Toy... IT'S A TOOL !!
Nice evolution into a well-balanced, mature sound, the earlier thrills of highlighted trebles giving way to more subtle appreciation of inner detail and a wholeness of sound.
I have been comparing XP and Vista for a while, and at the moment prefer cMP² in XP, but other players in cMP sound good in Vista, and the way cPlay is developing makes the choice harder to make. I would hope you can look at the problem of connecting to the Vista file explorer from cMP so it can be really practical to use cMP in Vista.
Also, I have a problem with cPlay b13 and b14 in Vista with Juli@ and Gigabyte cmP². It will not start, hanging immediately on starting, possibly at the point I think it is trying to communicate with ASIO drivers, but I am not sure. The error message is "AppHangB1". I have checked the usual suspects: new downloads of cPlay, fresh installs, different Juli@ drivers, different Vista disks, and different levels of opitmization - no joy. However, I have had no problems with my older cMP² with Biostar/M-Audio card and Vista.
Some questions:
1. Do you get the cPlay window?
2. Have you tried with ASIO4ALL?
If there's no window, that means a change to WTL is needed.
The window appears with "starting" in bottom left corner, and immediately hangs - the settings window does not appear, nor the file browse window. I tried installing asio4all, but don't see how to make cPlay use it, as I can't reach cPlay settings. Changing Windows Template Library? How?
WTL added a manifest component which I removed (no relevance to XP and causes a runtime DLL to be loaded). I suspect this is causing the issue in Vista (it relates to common controls). For b15 I'll put it back but should it affect SQ, its gone.
Thanks. cPlay is at its best in XP, IMO, but Vista is interesting and it would be a shame if cPlay were effectively excluded from it. From the previews I've read, Windows 7 is Vista again with more dubious gadgets. Vista installs take only 15 minutes, and 7 aims for 10 minutes, which is very useful. Soundcard drivers and software are easier to install. Un-optimized, Vista audio sounds much better than XP, but optimized it is closer, perhaps because more optimization is possible in XP. cMP² is better in XP, I think because you have tuned it specifically for XP, but other media players sound at least as good in Vista as in XP, maybe better, and all are much improved with cMP of course. Could you perhaps think about a Vista-tuned cMP² as a future project?
HI cics,Thank you for the another excellent version. However this version is a bit different, as this version is very smooth, yet bass is tight and resolution is very high. Rhythm (timing) is great. I can hear more details than b12. Unlike b12, with B14 I can turn up the volume to quite loud without feeling annoyed by the high frequency.
By the way another comment is that I am testing offline upsampling software at the moment. The wave file is upsampled to 192Khz before feeding into cplay, resulting in further SQ enhancement especially in complex music. Just that size of the wave file increased by around 6 times. Offline upsampler frees up the CPU time so I can freely underclock to the lowest possible speed without sacrificing 24/192 capability.
Edits: 12/07/08
Hi can you please tell me which offline upsampler you use? I also want to try this with cPlay. Also how big is the final wav file?
Thanking you
Along the lines of riboge's request, I would like to see the separate artists for each track in compilation albums. Especially useful for CDs of renaissance music with so many composers I can never remember! (As I said to riboge, I have made classical cuesheets with the composer as PERFORMER)
This is the usual but very inadequate way this is dealt with, e.g., inconsistently by the online cd databases. The more interested classical listeners want to refer to the actual performer info and the composer and not to limit artistry to just one of these (the ARTIST or ALBUM ARTIST tags being the usually bastardized ones). I have since my first post on this done most of the editing to change GENRE to list composer for classical, this is a better compromise than putting composers in ARTIST, but really a complete player should use the tags correctly and include composer.
Interesting compromise. On some classical albums I've stuck artist, conductor, orchestra, etc. on the end of the title, but they can fall off the end of the cMP playlist. I'm not too concerned, because I know cics is prioritizing sound quality, and I can look up tags and other music info. on other computers. This tagging issue would become much more important if we could in future build our own playlists easily within cMP - seems a big job, but I'm no programmer.
you couldn't fix the flicker without hurting sq, which you have done anyway. It is smoother in sound but with no loss of highs or detail.
Being thus spoiled, I make a further request, namely, that in cMP some minimal search function be added and/or some way of listing by composer. I have now 205 multifile flac cuesheets for albums, mostly classical. Finding a desired work is difficult since the listing is by performers, often the conductor for orchestral works, so if you don't remember who conducted when you look for a particular Mahler symphony it can be difficult and slow. The album title is listed only via "all" which puts it second and often doesn't all make it within the window, nor is it in a fixed position in the line, as with columnar listing, so harder to spot when scrolling thru. Listing by composer in cMP would be great, but it would have to read the file tags rather than just the cuesheet to get this info, unless there is a way for composer to be included in cuesheets by programs that compose them.
So a search would probably be the easier way to go. Seems to me adding this would have no effect on sound quality since it would operate only when not playing.
In case it is useful to others, I have thought of a way to get composers into the current cMP. For classical works only I can substitute the composer for the genre in the cuesheet. Then composers will be listed along with the remaining genres via the "genre" selection. However, this means for me editing around 175 cuesheets (so far).
Frodan wrote:
"However, this means for me editing around 175 cuesheets (so far)."
1. Tag&Rename can copy text from one tag to another. It's quick and reliable.
2. I have much the same problem as you about finding files. I do the following:
When cMP is in "Start cMP" mode, click the button marked "Explorer" in its main window and launch Windows Explorer. Go to View > Folders to get a sensible layout. You cannot search but you can navigate easily enough. Click on a file to launch cPlay. Explorer is still running when you quit cPlay so you can repeat the process.
You could also try making duplicate cuesheets and filing them under a "Composer" tree even though they point to data organised differently and kept elsewhere.
Thus, Composers > Mahler > Symphonies > "Symphony No 1 - Hicksville SO, von Stonedeff.cue" points to data stored in a separate folder called "von Stonedeff".
It would take only a moment to write the cuesheets if you use Al Jordan's Recursive Cuesheet Creator (or possibly his playlist creator though I've not tried it).
My feeling is that if we want cPlay to address all our very different admin needs (a) it won't happen and (b) it could impinge on its strength as a player - we'd end up with a Foobar clone. So I try to overcome its limitations with a trick or two.
Dave
Hi everyone, have been on another planet for the past 6 months but am now catching up on all the posts, as well as updating my system to the latest versions.About the composer issue in Cue Sheets, there is a tag specifically for this in the specs called SONGWRITER. Me and my friend have been working on this issue and are transfering huge collections of music to hard disc, I've got now over 1000 CDs with the same number of cue sheets to edit and this is a big issue, not to mention managing the GENRE tags as well.
It would be nice if there was a way to support this eventually in cMP to make it a little easier to navigate a big collection of music.
I've included a part of one of our cue sheets as an example below.
Daniel Gauthier
Montreal, CanadaGigabyte GA-G31M-S2L, E7200, 2 x 1 gig HyperX LL Memory, Antec SLP-450WR PSU, WinXP sp3 on Compact Flash Drive, Jili@ Analog Out direct to Electrocompaniet Ampliwire II Amplifier, Apogée Convergence Speakers.
-----------------------------------------------------------------------------REM GENRE Symphony
REM DATE 1977
CATALOG 4988005503473
REM ORCHESTRA BPO
REM FIRM DG SHM-CD
SONGWRITER "Brahms"
PERFORMER "Karajan"
TITLE "Symphony No. 1 in C minor Op. 68"
FILE "Brahms - Karajan - Symphony No. 1 in C minor Op. 68 - 1977.wav" WAVETRACK 01 AUDIO
TITLE "I. Un poco sostenuto - Allegro"
SONGWRITER "Brahms"
ISRC DEF058702021
INDEX 01 00:00:00TRACK 02 AUDIO
TITLE "II. Andante sostenuto"
SONGWRITER "Brahms"
ISRC DEF058702022
INDEX 00 13:22:02
INDEX 01 13:26:00
Daniel Gauthier
Montreal Canada
Gigabyte GA-G31M-S2L, E7200, 2 x 1 gig HyperX LL Memory, Antec SLP-450WR PSU, WinXP sp3 on Compact Flash Drive, Jili@ Analog Out direct to Electrocompaniet Ampliwire II Amplifier, Apogée Convergence Speakers
Edits: 06/16/09
For classical music I list the composer as the PERFORMER in the cuesheet.
Nice. Smoother than 13 is probably the best way to describe. While 13 was uber open on highs it could sometimes offend (ie when my electronics not fully warmed up or source material a bit edgy). Since I love the highs I'm not sure which I like better 13 or 14. Having said that both are so good in the highs it is scary. Imaging and transparency is just crazy.
Again well done.
To many who are thinking of getting into pc audio/cplay in particular: follow the instructions from cics verbatim. It only cost me about $400 building cmp into an old pc case and periferal infrastructure. All other critical hw (cpu, obo, memory) and software settings (on bios and in windows, minlogon is a significant charecteristic) are to spec. If you do you, you will astounded by the transparency, dynamics, low level detail, extension on both ends of the frequency spectrum. Cplay/cmp is truly a masterpiece for audiophiles.
"Nice. Smoother than 13 is probably the best way to describe. While 13 was uber open on highs it could sometimes offend "
Agree. I found 13 "unfriendly" on highs.
I find the high frequencies to be in better balance with #14.
I remember the revelation from my early audio experiences that most times the component with natural treble balance can seem dull intially.
I did not get much of a chance with #13.
In fact, I detect a better balance of the upper midrange all the way up. At least it is harmonious with my TAD 4001's which are known to have a hump in the upper midrange.
As has been said, cPLAY has reached the level where it is increasingly difficult to hear what the differences are quickly. And who wants to go back? Not me. I think cics has an advantage in that he knows what he has manipulated (back to his conceptual model) and knows what to listen FOR. I trust his ears and gladly move on to the next version confident there is something worthwhile to it.
What more can be said than it is producing beautiful music and ...
THANKS,
Rick McInnis
Hello, Theob! I did all recommended optimizations, but minlogon. Is it hard to implement? Is there some easy way to do. I read that I need to download some prog and then perform some tricky changes. It scares me.
Also isn't minlogon just booting in safe mode?
Serge.
I believe seger has some great suggestions on how to do this. do a search on seger as author and follow them. cics recommends some very difficult to execute steps but seger's are easier. you have to be careful my 1st attempt resulted in a no boot situation and I had to reload windows. but it is definitely worth it. so if 1st attempt fails keep trying.edit: oh and make sure you get all other optimizations done 1st (like when using XP lite) and make the minlogon update last. because a lot of stuff doesn't work with minlogon.
Edits: 12/06/08
Hi Serge,
I believe there are quite a few posts detailing the minlogon implementation. All you need to do is start up in safe mode, rename winlogon.exe to something else, and copy minlogon.exe (renamed to winlogon.exe, of course) into the system folder, and modify your registry. I don't remember the exact values to modify, but you should be able to find it in the cMP thread.
Alex
Today and tonight it's only my music and my favourite belgian beer.
Serge.
Oh, the pleasures of solitude!
It's fun to do this without disabling WFP, but in practice I've found it easier to disable WFP - reason being you save time by not having to use safe mode each time you come in and out of minlogon - software installs, soundcard driver updates are best done from normal XP desktop, out of minlogon.
Easy Minlogon
1.Download this minlogon.cab file
You must have WinRAR or WinZip or similar extraction tool installed.
Right-click the minlogon.cab file and choose "Extract to minlogon/"
The minlogon.exe file is in the minlogon folder that appears.
2. Make a new text document and copy the following into it:
Windows Registry Editor Version 5.00
[HKEY_LOCAL_MACHINE\SOFTWARE\Microsoft\Windows NT\CurrentVersion\Winlogon]
"Config"=dword:00000017
Save it as minlogon.reg to the desktop or another convenient directory (folder)
3. Do either a. or b.:
a. Without disabling WFP: Restart and repeatedly press F8 when booting to enter Safe Mode
b. Disable WFP: 2. Download xplite Launch it, select the Windows File Protection tab, click Disabled, and Apply.
4. In C:\WINDOWS\system32 rename winlogon.exe to winlogon.exe.bak
Copy the minlogon.exe into system32 folder and rename to winlogon.exe
5. Double-click minlogon.reg to create the new dword called Config with value 00000017 which allows minlogon to work.
Restart and you have implemented minlogon.
Well I did it
1. unpacked the cab, that I have downloaded. there were 3 files one of them is also a cab with 1 file - namely, minlogon, but of different size. didn't touch the cab and didn't unpack this file
2. renamed winlogon to winlogon.exe.bak and saved a copy in a different folder
3. copied minlogon.exe from the highier level cab to system32 folder and renamed
4. created text file, typed everything correctly, = was strange (too long)
saved as minlogon.reg (should I have renamed it to winlogon.reg?)
5. Double clicked - here confusion starts. Windows asked me if I really want to enter these values or smth. Yes I said. Nothing happened. I double clicked again. same question same answer. Third time same question I answered no. Placedthe file on my desktop. Rebooted - blue screen Fatal error and so on. WHAT TO DO NOW? Can I reboot from flash drive and restore my CMP or is it dead forever
HELP!!!
Serge.
steppe - sorry to hear about the problem. For next try this should help:
1. The minlogon.exe file is a 23.8kb file, right there in the folder after unpacking the minlogon.cab file - no second extraction is needed. Please ignore the "minlogon209170.cab" and "minlogon.inf" files that are there.
2. You should copy and paste the .reg file information, to avoid typing errors. Calling it minlogon.reg is a matter of convenience for us - as long as the contents are correct you could call it anything but it must end with the extension .reg You can check in regedit to see if the entry has gone in correctly: HKEY_LOCAL_MACHINE\SOFTWARE\Microsoft\Windows NT\CurrentVersion\Winlogon and there should be a line "Config" with the value 00000017
The tiniest mistake with minlogon fries Windows, so be very careful.
One thing that has occasionally saved my ass in this regard is having a multiple-boot setup - for some (not all) mistakes, you can get into the Windows system files for the damaged system and restore them.
I may have to change the title of my shortcut to Easy but bloody dangerous Minlogon!
Thanks for explanation. It seems that I typed everything well. I am usually carefull, especially because English is not my native. I have a boot dvd with every possible OS starting from Dos 4, which I still remember. I'll try to boot from it or to install Windows to a flash drive, hope 2 gb will suffice, and then try to rename winlogon.exe.bak to... and reboot again. If it works, I'll have a rescue flash drive by my side and will try to do everything again, but this time I will create text file beforehand. Also, does it matter how many times I double click the *.reg file and say Yes?
Serge
P.S. It's late now here, so I guess I'll go and sleep over it. A pity, b14 was so fine...
Check very carefully that a new winlogon.exe does not get put in system32 folder by Windows after you rename it to winlogon.exe.bak. When you double-click minlogon.reg it should ask you to confirm once, you click yes, and then it says it the registry has been changed. That should be it. If it keeps asking, something is wrong - maybe WFP is not really off. Check with regedit that the correct line has been added with the correct value. Bon chance!
not sure what you did wrong but when I had this problem (no boot up) I tried a windows repair and that did not work. So I reloaded windows from the original disc. When you get back up all your bios settings will be ok but your Windows settings will have to be re-done.
H/W a copy of Seger's note. His method works well and is better than either the method described by cics or than my suggestion of using XPLite.
In fact, it's so easy to do that it should count as a security breach. I've tested it several times though I usually have to edit the registry manually as something I do seems to stop the reg file from working. Otherwise, it's fine.
Good luck
Dave
+++++
Minlogon without disabling WFP:
Download this minlogon.cab file from the link below.
You must have WinRAR or WinZip or similar extraction tool installed.
Right-click the minlogon.cab file and choose "Extract to minlogon/"
The minlogon.exe file is in the minlogon folder that appears.
Restart and repeatedly press F8 when booting to enter Safe Mode
In C:\WINDOWS\system32 rename winlogon.exe to winlogon.exe.bak
Copy the minlogon.exe into system32 folder and rename to winlogon.exe
Make a new text document and copy the following into it:
Windows Registry Editor Version 5.00
[HKEY_LOCAL_MACHINE\SOFTWARE\Microsoft\Windows NT\CurrentVersion\Winlogon"Config"=dword:00000017
Save it as minlogon.reg to the desktop or other convenient folder
Double-click minlogon.reg to create the new dword called Config with value 00000017 which allows minlogon to work.
Restart and you have implemented minlogon.
I've used this successfully with XP SP2 and SP3. The systems have all the cics optimizations in place before implementing minlogon.
Hello, Cics, and others. I have missed b13, so can compare to 12. It seems that new version is slightly smoother in the highs, also smallest details now sound more natural and longer, like an aftersound of acoustic gutar strings. Only one cd still remains unconquered - Robert Plant - Mighty Rearranger. On one song the voice sticks to guitar and on another very transparent highs from cymbals are somewhat grainy and "glued" together, not clean enough, but I am afraid it ia my soundcard limiting the quality. Overall sound is very good, thank you Very Much.
Serge.
P.S. I'll be having a long and musical Saturday today, thank You once again
Hi steppe,
Out of curiosity, which sound card/external DAC are you using? I had my mind set on the Juli@, but now that the new recommended mobo supports PCI-E cards, I'd like to wait and evaluate other options before selling my 9632.
Thanks!
Alex
Hi, abysstw! I use a very old Soundblaster Audigy2, Yeah, the one with DVDA capability - it is connected to active computer speakers MICROLAB smth. I live away from home now and from my main rig, so I have to make do with what I have, but my plans are as follows: Lynx 16 xlr slaved to Apogee Big Ben and to Apogee 16 DAC or Lynx Aurora 8. I use 90% redbook files, but it doesn't matter, let it be an overkill. Juli@ is unavailible here, though it is announced in price-lists, but sellers say "Don't wait, it won't come"
To tell the truth, I am not terribly disappointed. I wil get Lynx before long
Serge
To any interested:
Inspired by Bernd's battery experiment using the PICOpsu "power supply", which is a dc-dc converter supplied by whatever 12 volts source you choose, I bought an OPTIMA 75Ah battery so this would work for a few listening sessions before needing charging along with the advantage of minimal internal resistance.
I started out concerned that the battery is really a 12.6 volts source and wondered if placing a schottky diode in series before the "converter" would be a good idea. I initially thought the 12 volts lines from this thing were straight through from the source and not manipulated by the converter at all.
Testing with a multimeter showed that is was not a open circuit so I thought I would be OK. At this point I am not considering that this was a switch I was measuring and not some kind of regulator. SO, I went ahead an set it up.
As with all new devices it did not sound very good at the beginning. I told myself to be patient.
After about an hour things started getting much more interesting. There did seem to be a reduction of some kind of background noise because there was more fine detail. I began to think this was going to be worth the trouble. The board was installed "as is", standard cheap Chinese electrolytics still in place. No capacitor bank at the battery, though a bank of Panasonic FC's at the CPU pin. As I am enjoying this increasingly more interesting sound and dreaming of what might be, changing albums after a track or two to get an overview (my new favorite way, I just spin the wheel on the mouse and see what I get). I make a selection, push the GO button twice (using diagnostics) the screen shows the beginning of a new album about to play, yet it stops with only TRACK ONE displayed, and makes a pulsing sound ( about five pulses) and then nothing (no sound), screen stuck, machine will not respond. I disconnect the battery and try to reboot. Nothing happens. The irony: the album I had randomly selected was the MOTHERS OF INVENTION - FREAK OUT.
SO, I have not had a chance to investigate since this Wednesday night adventure. I hope it is just needing a BIOS reset. Is it possible that I was sending 12.6 volts to the 12 volts rail and destroyed the MB, the CPU, everything? I would appreciate any comments.
Tomorrow I will be able to find out if I have done it again. Already started looking for available boards. It seems DIRECTRON has the G31 boards available.
Has anyone tried the new semi-recommended board?
Bye,
Rick McInnis
Dear Bernd,
Instead of inserting the picoPS back into my system and making sure it still worked with the ANTEC PS, I went ahead and started modifying it. Of course, this is a stupid thing to do.
So after changing out all of those capacitors, which was an awful job, I think something went wrong with the pico since it still would not boot, either with or without the schottky in the circuit AND for some reason it required resetting ALL bios settings to get it to work with the ANTEC back in place.
I think when I rry it again I will use the same version you are using.
Do you think a 12.6 volts battery destroyed the pico. In the manual it said that this was within its tolerance. It could be a coincidence; maybe I just got a defective one?
So, reassure me, you were using this with the GIGABYTE board?
Do you know of anyone else who makes a similar product?
Are you using the one for cars? I would like to know just what is being done for the 12 volts rails. I would have to assume the "car" version is definitely doing something to set the voltage, but is the "regular" version "straight through" with just a switch in the line for the turn-on protocol? There is very little technical information about this thing available; basically NONE.
While looking at some of the similar offerings for cars, I was surprised they showed specifications for ripple ten times that of the ANTEC. They must be converting to AC and then back to DC with some kind of switching going on. Seems useless if that is the case.
Well, if it doesn't work out I have a great heater supply for my phono amp!
Bye,
Rick McInnis
So, I did not destroy anything.
Modifying the picoPSU with Panasonic FC's, installing a capacitor bank before the board, along with the schottky, just for insurance, and will listen again tonight.
Nonetheless, cics has tempted me and so the Q35 board is on the way.
Bye,
Rick McInnis
Rick 12,6V is the upper limit of the ATX-specification, the lover limit is 11.4V so i dont think you destroyed something.
Resetting the BIOS should solve the problem.
What kind of PSU are you using. The 200W-version you mentioned?
bye
Bernd
Bernd,
I figured there would be a little more margin.
I would like to know what the differences are between the two models.
I hope it is just a BIOS glitch but I cannot imagine why that would happen.
I will know more tomorrow and will let you know.
Thanks for your help,
Rick McInnis
Don't know if all BIOS settings are available.
if I knew you were going to work with it!!!
Have you, by chance, tried it out?
I hope I do not need it, but it could be a new adventure.
I put my spare G31 board in my office computer so I guess if I am not happy with it I could swap them.
Does the Q35 board allow CPU voltage adjustment?
Thanks!
Rick McInnis
E5200 & GA-Q35M-S2 may very well be an improvement when using 2.0b14.
Dear cics,
Sometime this weekend I will put the Q35 on order.
My reservations about e5200 are first and most importantly I think with the current G31 board there is no comparison in the robustness of the sound. E5200, after re-inserting e7200 sounds puny.
Then there was the aspect of not being able to set memory speed as slow as one is allowed with e7200. I have no idea how important this is in your grand schema for minimizing jitter. On the other hand the thing would "run" on voltages way lower than either of my e7200's. Of course, this means it would boot-up but cMP would not function. Not much use, but interesting. As I had relayed before it would boot at 0.850 volts. I did not go any lower since it was a moot point.
Reading the few reviews of the Q35 on the newegg site makes me think; since it seems to be getting low regard as a regular PC MB these same qualities could very well mean it is perfect for our use.
Does it having four memory slots concern you? I fear any more "stuff" than necessary. I know we will only use one. I had never seen a board with a six pin CPU PS header. Could there be any advantage in using all of the "holes". Easy enough to modify my existing cable. There could be an advantage with reducing the resistance and inductance of the connecting wires (?).
I see that the BIOS for this board does include the control/F1 for more selections. As usual they do not tell you what they are. I will let you know upon installing the board.
Let me know what you think of my concerns of e5200. I am a lazy man who prefers listening to cMP than playing with it. Though I am sure many of my posts might betray that. If it must be done, upon your word, it will be.
Bye, and thanks as always with my best wishes to you and your family,
Rick McInnis
Minimum voltage setting is 0.85V (rather go 1 or 2 notches higher). For CPU Host frequency, can you get below 200?
with e5200. Certainly no problem with e7200, either. I am not sure I am SURE of what you are asking!
Only thing was that memory setting, best I was offered was 2.5X with the e5200.
Bye,
Rick McInnis
On E5200, you may be able to go lower than 170.
Dear cics,
Does this imply that the new cPlay requires much less processing power than before? How would it affect the owners of the old combo? Should we underclock the CPU even further, or would you suggest a direct upgrade to Q35GB+E5200?
Now that the new Q35 mobo offers support for PCI-E soundcards, is the Juli@ still the cMP's card of choice? Are there any PCI-E cards that you'd like to recommend?
Thank you again for your great work. I now sincerely believe that one cannot be a true PC audiophile without cMP squared.
Cheers,
Alex
New GB Q35 mobo and E5200 has yet to be tested - this is not a recommendation.
CPU: Same settings apply as per link in 2.0b14 post. CPU processing for 145db SNR upsampling remains the same. cPlay 2.0b14 reduces RAM reads & writes (which don't consume much CPU resources).
You already have PCIE (x1) on existing mobos (including the Biostar). For the price, Juli@ remains great value and outperforms my RME HDSP9652 (x4 more expensive than Juli@). Don't have any specific recommendations.
.
Release notes . Take note of SSE2 & SSSE3 changes (SSE3 no longer works with SSSE3 versions).
Change Log (2.0b13 release):
- UI refinement: remove Windows overheads (MFC replaced with WTL)
- DSP refinement: improve RAM load process (done via DSP threads)
- Compiler optimisation done
- Fix audio glitch that occurs after play is restarted from 'stop'
Please REMOVE previous versions before installing cPlay 2.0b13. Normal (SSE2), SSSE3 and SSSE3 B9 versions are available.
For cMP² users, review these BIOS changes (underclocks FSB, CPU & disable additional USB items). Note: E7200 has been replaced with E7300. Same settings apply. Gigabyte mobo may be difficult to find, suggested new mobo is Gigabyte's GA-Q35M-S2 .
Hi cics (and Lurkers)
For the first time, I'm having trouble running cPlay. All previous versions are uninstalled, and the current install finishes as usual. Launching the program results in... nothing.
Reading over the change log, I wonder if Item 1 ( UI refinement: remove Windows overheads (MFC replaced with WTL) is in conflict with the many XP services I've disabled.
Nothing helpful is logged in the Event Viewer. Any suggestions (please...) ?
Thanks to all,
Grant
desperately wanting to listen via v2.0b13-B9 !!)
That's not a Toy... IT'S A TOOL !!
same thing happened to me
I use VISTA SP1
but 2.0B12 is ok
If this is not correct, cPlay exits immediately (i.e. won't start).
Hi cics
Thanks for your reply. Successfully installed and ran v2.0b13-SSE2... OK, fair enough.
Riddle me this: Why has every other v2.0bxx-SSE3-B9 release of cPlay run happily on exactly the same hardware/software set up ? I'm happy to get Beta13 running, just curious about what's changed, cause it isn't my platform. Is the CPU check new in B13?
BTW, my CPU specs (from Intel datasheet): Intel P4 3E GHz 1MB L2-Cache, 800FSB, PGA-478 / Product Code: BX80546PG3000ESL79L / includes SSE3 instruction set.
To Theo: thanks for the note, but a restart did nothing to help my situation.
Thank you again cics for the incredible gift of cMP... It boggles the mind to hear old favourites "for the first time" (just clobbered a friend with Brain Salad Surgery earlier this evening !)
Cheers,
Grant
That's not a Toy... IT'S A TOOL !!
I guess you overlooked the first line of his post:Take note of SSE2 & SSSE3 changes (SSE3 no longer works with SSSE3 versions).
As you stated, your CPU only does SSE3 (not the same as SSSE3).
b13 installed and plays just fine for me. BTW, nice improvement.
Edits: 12/05/08
Hi Edward
Thanks for pointing that out to me - I'm sure the subconscious mind noticed, but I've never had to make the distinction and (maybe) assumed it to be a ttypo (...see !)
"User Error" strikes again. Duh.
OK... Mystery Explained, now back to serious auditioning.
Thanks again,
Grant
That's not a Toy... IT'S A TOOL !!
same thing happened to me until I shut down and booted back up.
even though it now blinks every second. I am presuming this blinking is related to the UI changes you mention (I don't know what the initials mean), and you thought it would be less distracting greyed out. So then you greyed the volume level for symmetry. But these are important and useful to see clearly from a distance. Is there something else that can be done about the blinking instead?
The sound is incredible! It is getting really hard for me to tell if each version is better than the previous because they are all so great at this point.
Problem lies with how Windows does custom drawing. There's too much overheads and messaging going on which sadly causes jitter.
In my search for an alternative to Juli@ card I came across this card for less than 150 bucks
http://www.auzentech.com/site/products/infra_quartet.php
Not made by Auzentech but by a Korean company and distributed by Auzentech.
What is really interesting is at the price it includes word clock i/o, 192 output, ASIO 2 support, free shipping and 10 day money back thingy.
In their comparison chart they even refer to the Juli@ card.
One wonders what this puppy may sound like with a Black Lion attached or an Audiocom super clock4 or even an Antelope clock.
PC XPpro TC Electronic Konnekt-8 firewire i/o, 2-Tact 2150's, Genesis 500 modified speakers, Virtual Dynamics & XLO cables
Well, not sure if the "Multi-Function Sound Editing for Your PC Audio Studio" as stated in the Infrasonic QUARTET Audio Card is what some of us are looking for and the PCI bus couples heavily with the motherboard in terms of power lines.
How about this new pci-e 1x soundcard designed for audiophile ?
http://www.asus.com/products.aspx?l1=25&l2=150&l3=0&l4=0&model=2711&modelmenu=1
As I posted elsewhere, of 3 cards I tried (EMU, ESI, ASUS) in approx. the same price range, Juli@ was clearly the worst - and I have strong suspicion that breakout cables play significant part in that. I don't care how many samples latency can be set for, or how many people happily use it - it just didn't sound good in my setup. I still have it, BTW - didn't get around posting it on eBay yet.
Although, I would think that appropriate mods will improve all those cards with breakout cables - this Auzentech, ESI, Lynx, RME, Digigram etc.
I don't think that is much of a problem, simply replace the default breakout cable with a custom made one. That was exactly what I did with my RME 9632, and it was the difference between day and night.
Hi Uzeb
Gee, these guys seem to have checked all boxes on The Wishlist - nice card. And very brave to put something as sensitive as a microphone preamp into the horrible electronic soup of a PC enclosure (must be some Superman Kryptonite Super-Shielding going on there).
Where did you find it for < $150 CAD - my searches have not turned this up.
While a real Wordclock input is nice, all digital I/O connections must pass through the breakout cable, and the photos suggest it connects by that MOST unreliable of plugs - an HDMI-style interface.
Why on earth would you use something so fragile as that ? It relies on friction alone to maintain the connection, and the contact surface area is relatively small. I'm not the first person to complain about this connector. In a studio environment this could EASILY become dislodged at a critical moment and ruin an irreplaceable recording. True, at home in your cMP world the only disaster is interrupted playback, and the chance of the cat wandering amongst the cables is remote.
Please don't think I'm dissing the unit - it's feature set is very complete (in fact far in excess of cMP needs - 4 analogue inputs etc,). It's got digital I/O at 192 KHz, WC-IN, ASIO 2 drivers, and "affordable" pricing, so it should be on everyone's short list.
Please report back if you decide to install the Quartet.
Thanks,
Grant
That's not a Toy... IT'S A TOOL !!
"Where did you find it for < $150 CAD - my searches have not turned this up."
Actually $149.99 US direct through Azuntech; I keep forgetting our loonie is back to its place as a 3rd world currency.
Have you seen that the new PS Audio Ultra series is going to be using an HDMI plug to pass a I2S digital signal.
PC XPpro TC Electronic Konnekt-8 firewire i/o, 2-Tact 2150's, Genesis 500 modified speakers, Virtual Dynamics & XLO cables
1.despite very good results with a very good linear PS for the Xonar DXe Soundcard i tried finaly 4 sealed Panasonic batteries( 28Ah) with a highend regulator to 5Volts for the digital section of the Soundcard.
SQ is phantastic, more resolution in all frequency sections and precise soundstage and focusing.
2. After resignation for the lack of genuine ASIO-32bit drivers from ASUS now ASIO4all works with 24/192.I checked Hardware Buffers in the configurationpanel and now i can use the SC with CMP/Cplay with 192/24 .Don't ask me why.
bye
alfred
Hi Alfred
That is very interesting about using battery power... what exactly are you using for your "highend regulator" ?
Regarding ASIO4All: how can you be sure the 24/192 signal is getting through to your DAC ? Does the Xonar DXe control panel confirm this ? Just curious.
Thanks,
Grant
That's not a Toy... IT'S A TOOL !!
hi Grant,
thats the "regulator": http://www.thel-audioworld.de/module/spr/SPR-2.htm
The Control-Panel confirms 192000.(Perhaps a little miracle)
bye
alfred
Looking at possibly adding a sound card to my system to compare to fire wire. The Juli@ is a little hard to get in Canada so I was thinking of maybe the xonar PCIe card. I searched through the forum and seen there has been some issues with C-play. Does anyone know if these have been sorted out.
Do the ASIO drivers now work with Cplay and can Cplay up-sample 192 through the card. If so what kind of latencies can be achieved.
Other choices are RME 9632 which reportedly will output 192 spdif and the lynx AES-16 which is almost a grand with taxes in Canada; Not sure that makes sense to only output spdif. I think I could mod my firewire unit or a cheaper card with a nice word clock and good power supply for that kind of money.
PC XPpro TC Electronic Konnekt-8 firewire i/o, 2-Tact 2150's, Genesis 500 modified speakers, Virtual Dynamics & XLO cables
Hi Uzeb
A few years ago when I first looked for Juli@ in Canada (specifically GTA region), it was nowhere to be found. However, I'm happy to report Cosmo Music carries them - I bought mine right off the shelf two months ago. (see http://www.cosmomusic.ca/ ).
You can phone to discuss at 1.800.463.3000 - ask for Fred in the Keyboard section (...and tell him I sent you).
(DISCLAIMER: I am in no way connected to this retailer EXCEPT for being a satisfied customer)
Hope this helps...
Grant
That's not a Toy... IT'S A TOOL !!
Thanks Grant
I am in Vancouver but shipping from Ontario is a lot easier than from Florida.
I had called the distributor and they never mentioned Cosmo directed me to Florida music.
They did say they have trouble keeping them in stock; I guess all us cmp enthusiast.
Regards Uzeb
BTW I like your ending statement, I have been borrowing it for use with my wife lately when she rolls her eyes at yet another toy purchase.
PC XPpro TC Electronic Konnekt-8 firewire i/o, 2-Tact 2150's, Genesis 500 modified speakers, Virtual Dynamics & XLO cables
Hi Uzeb,
How are you powering your Konnekt-8?
There is a big gain in clarity, definition and speed by getting a very good power supply for my Konnekt-24D. I am using a big 12V power supply with some custom make filter including: murata emi filter, a couple of silmic-II and Rubycon zlh.
For the PC firewire card, I am using PCI-e 1x express card which has an isolated 5V power supply via molex. I build a simple filter on it too.
Hi promethk
I am just using the power supply that came with it the unit.
I have thought of using a Battery supply or King Rex power supply from Audio Magus. I never thought about the off board power supply possibilities with PCI-e firewire card very interesting.
Thanks
Uzeb
PC XPpro TC Electronic Konnekt-8 firewire i/o, 2-Tact 2150's, Genesis 500 modified speakers, Virtual Dynamics & XLO cables
I have been hesitant to replace my 9632 with the JULI@, mostly due to previous investments ($500 on the card, $150+ on each of the analog and digital cables) and fondness of its analog output capabilities. However, now that my cMP squared is used exclusively for digital output, I need some advice on finding a good upgrade.
What kind of difference is to be expected by switching to a JULI@, without any form of custom modding? Is it possible for the digital section to operate without installing the analog section of the card? Does the more expensive cards - such as the AES32 - offer better performance, or is JULI@ the best choice for cMP squared?
Thank you for reading through my novice questions. I really need some input for my next soundcard upgrade.
Hi abysstw
In the great scheme of things, most "studio types" would argue the RME is a "better" card... but the question is better for what ? I presume you don't need the multitrack recording/playback capabilities the 9632 is so well known for - since cMP is all about 2-channel-only.
The Juli@ is one of the few 2-channel PCI cards that offers all SRs up to 192 KHz ON S/PDIF. I've searched high and low for something with just this I/O complement for a long time - wishing to connect my ADC1 & DAC1 to the DAW at the full 192 SR. It seems the pro-audio designers are targeting users with other requirements - Firewire/USB connectivity, 24-b/96-KHz-max, 4 & 8-channel configurations, etc.
Lynx makes wonderful PCI solutions, but you're paying (handsomely) for refinements that don't apply to a cMP Playback rig (the fabulous Lynx A/D and D/A converters remain sleeping giants !). Where's the sense in that ?
Juli@ is "cheap" (no sleight to ESI intended) and does exactly what we need - and no more. Just get that digital signal out of the box ! And... you have the option of easily abandoning the analogue section to boot. Depending on the refinement of your D/A (I forget what you use) you may hear an improvement from externally clocking the Juli@ (discussed at length elsewhere). What else could a boy ask for ?
The good news is your RME is easily sold to a new owner in the home studio market.
I hope this is helpful...
Cheers,
Grant
That's not a Toy... IT'S A TOOL !!
Hi Grant,
Thank you very much for your insight. You've successfully convinced me in abandoning an expensive but unnecessary sound card upgrade.
Juli@ is no doubt a wonderful card, and probably has the highest C/P ratio among all for cMP squared playback rigs. My remaining question regards the digital output quality of Juli@ relative to RME 9632. Is it noticeably better than the RME? I think the implied answer is yes, but I want to make sure before spending $300 on the card and custom made cables. Oh, and my current in-service DAC is the Apogee Mini, if that's of any help :)
And by "you have the option of easily abandoning the analogue section", do it mean that the Juli@ can operate with just the digital section of the card without any modding?
Thanks again!
Alex
Hi Alex
Download & read the Juli@ user manual - they describe how to dismantle & reassemble the card for either -10 dB on RCA or +4 on TRS analog I/O. They DON'T mention the option of just removing the analogue portion and using Juli@ as a digital-only interface (as some hardcore cMP-ers have done)... but it works just fine.
I've never lived with an RME 9632, so I can't opine on it's S/PDIF I/O quality versus that of the Juli@. There is an element of "apples and oranges" too, since one is 192-capable and one is not - and for cMP most will agree that the 192 is a big deal (if your D/A supports it).
Since the Mini decodes 192, I'd rate that as high on your "must have" list... and the 9632 can't send 192, so what's to think about ? I will confirm that my DAC1 tells me the cMP playback is better at 192 KHz than at 96 KHz.
Regarding cables: I suggest you make one up yourself (or find a friend who's good with a soldering station). I've found that digital cables DO make a difference to the sound coming from my DAC1 (despite the "science-based" protestations of Benchmark's engineers to the contrary).
I began with Mogami, switched to Apogee's Wyde Eye... and then discovered a humble, cheap-n-cheerful video cable bested them both ! (...wanna buy some broken-in Wyde Eye !) I went so far as to terminate the wire with BNC connectors - and mod my source components where necessary (the DAC1 comes with BNC for 75 ohm coax input).
The cables come with RCA connectors - you're ready to go, out of the box. If you like what you hear, then consider "upgrading" the RCAs (to better near-75-ohm rated plugs, or go the BNC route with it's attendant fuss and benefit). Now don't laugh when you get there - just trust and order -> http://www.digitlook.com/PETRA-10-5802-DIGITAL-COMPONENT-AUDIO-CABLE-6-FT-BULK-SKU-39723_p_4843-25380.html
You'll find that these cables (the WIRE at least) does remarkably well for passing analogue audio around too. I ran a 50-ft pair between my pre- and power amps and it bested the Mogami that'd been in service for years (once I'd installed fancier RCAs).
And if you hate 'em - at these prices - you can bundle up the trash with 'em, and search for something more to your taste.
BTW, here is the solder-type BNC - very easy to work with - but hard to find.
...no, that's not the PETRA you see in the picture...
Hope this answered your questions...
Cheers,
Grant
That's not a Toy... IT'S A TOOL !!
Hi Grant,
Thank you for the advice, it has been most helpful.
I pretty much agree with everything you've said, except I'm sure that my 9632 is capable of 192k output. The diagnostics screenshot provided by cics is probably shows the RME using optical output, which only supports up to 96k. Contrary to "cMP common sense", however, playback at lower sample rates (176.4 sounds best) is sonically superior to 192k. When I shared this issue with a friend and fellow 9632 - but non-cMP - user, he told me that the SQ of the 9632/DAC1 combo is less appealing than 9632 analog out. That made me doubt the digital output quality of the 9632, which may not be as terrific as the card's reputation in the pro audio market.
Anyway, enough ranting from me. I think I'll take your advice and see how Juli@ (w/ custom cable) works out in my system. Quite a discovery, by the way! I would never have thought of connecting my sound card and MiniDAC with a video cable :P
Thanks again!
Alex
"The diagnostics screenshot provided by cics is probably shows the RME using optical output, which only supports up to 96k"
In my system, using juli@card, optical connection is superior to a "normal" digital cable. Walter
I have the Juli@ but not the RME, the Juli@ is still quite sensitive to motherboard power, as is the case of most pci-based soundcard. As you already have a RME, I doubt if an unmodded Juli@ would be a significant upgrade.
If you are willing to spend $300, there are many new options coming out, including pci-e 1x soundcard, Asus is coming out with a new two-channel pci-e soundcard which boost even higher performance than the previous pci-e soundcards reviewed here. Other choice is firewire dejittering external box with spdif output.
Yes, Juli@ can operate with analogue board removed by simply unscrewing the analogue section.
Being obedient to the letter* of the instructions given by cics, I've never had any first hand experiences running cMP squared on an AMD platform. With his wealth of knowledge and incredible diligence, cics has probably considered - maybe even tested - cMP squared on the AMD platform, and I'm sure that cics recommends Intel over AMD for very good reasons. However, after evaluating the AMD platform more closely, I do believe that it has unique advantages over Intel for the purpose of computer audio processing. First off, AMD Hypertransport 3.0 is much more flexible than Intel Host Interface, as it allows users to modify the bit stream, HT frequency, External Frequency, and voltages through BIOS. Furthermore, AMD systems allow direct communication between CPU and RAM, whereas Intel systems require such activities to go through the north bridge, therefore AMD systems should (theoretically) have noticeable lower overall latency.I understand that with its sheer processing power, low TDP, and large L2 cache, Intel is - and probably will be, in the near future - the superior platform. Still, I think the AMD platform has potential, and may yet become a great alternative, provided that AMD finally releases a respectable CPU comparable to the E7200.
Am I theoretically correct, or is it just my wild imagination? Normally, I'd do some actual work before troubling the community. Unfortunately, the only AMD PC around is my quad core gaming computer with a very powerful graphic card, so all I have right now is you guys and gals. Please share your thoughts, and help me figure out what components to buy for my next personal project. I'm looking forward to any kind of input, especially from cics and those who have first hand experience of either listening to (or building) a cMP squared on AMD. Thanks alot!
*Except that I use an externally powered wireless PS/2 mouse. Never really tested the impact on SQ, but disabling USB altogether made me feel warm and fuzzy inside.
Edits: 11/28/08
Upcoming Intel core i7 HyperThreading will be comparable to AMD Hypertransport.
According to various hardware review websites, the power consumption of a typical Intel Core i7 system ranges from 140 watts (idle) to 244 watts (full load). IIRC, the TDP of the Intel Core i7 CPU alone is 130W at full load, which is around three times of my entire cMP squared system. Intel Core i7 seems like a product line too bent on performance, which (I think) disqualifies it as a good candidate for cMP squared.
If the improved technology doesn't pass through to lower power chips, then yes, no use to us.
Release notes .
Change Log (2.0b12 release):
- Minor refinements
- Compiler optimisation done
Please REMOVE previous versions before installing cPlay 2.0b12. Normal (SSE2), SSSE3 and SSSE3 B9 versions are available. Further beta releases are not planned unless fixes are required.
For cMP² users, review these BIOS changes (underclocks FSB, CPU & disable additional USB items). Note: E7200 has been replaced with E7300. Same settings apply. Gigabyte mobo may be difficult to find, suggested new mobo is Gigabyte's GA-Q35M-S2 .
Please say you are not done. If you are then what a magnificent ride. A new improvement (and release) at a weekly pace for god knows how long. Wish we knew more about you.
Well done Mr Cics.
... and the rest of us know that master cics will never be satisfied until cMP squared is nothing short of perfection. What made you believe otherwise, Mr. Theob? :P
words like...Compiler optimisation done.... Further beta releases are not planned unless fixes are required
Theo and others,
You know that our revered cics, and his daughter, are NEVER satisfied. We should all remember to be grateful to this music loving child for she is the inspiration for this project. I thank God for her every day and wish her nothing but happiness. I am serious, every day when I use this system I think of them both with gratitude.
Too many times I have figured he has DONE IT only for another (great) leap forward to be taken. Could there be a correspondence between two little red books? Forgive my attempt at Mao humour.
I figure, as hinted at before, that at the beginning of the new year we will be receiving a gift in the form of version 3.0.
cics's research and efforts to eliminate "jitter" which he understands with much greater subtlety than almost all; he possesses an Einsteinian conceptual insight into what is happening. He continues to find ways to deliver a more coherent waveform which when,typically,discombobulated the brain finds VERY annoying. When the brain is working that hard to make music out of a typical digitally derived waveform it gets very tired and tells the rest of your body to go to bed or go work in the garden.
Thanks to cics I am not getting as much sleep as I need and my yard could use some work.
Bye,
Rick McInnis
Daughters and grand-daughters have a way of inspiring, yes. Hope you are right rick.
Sorry if you took it too seriously, I was half-joking in my previous post. The other half was me crossing fingers, hoping that "Further beta releases are not planned unless fixes are required" is just another way of saying "... be well prepared, the revolutionary cPlay 3.0 might be too much for you to handle".
Out of curiosity, should current E7200 users upgrade to E7300? Is a big improvement in SQ to be expected? Also, how does the new E7400 compare to E7300? Are there any significant difference between these three models, other than default clock speed?
Thank you in advance!
Interestingly, I have an e7200, the first one I bought, that seems to be substandard.
I base this in the fact that it will not run at CPU voltages below 1.925.
My second one, at this point, allows 0.8925 (I think that's what it is, two "clicks" above 0.875.
I cannot see the speed making any difference for our application.
If you have one that will run at low voltage, you have what you need.
Now, if cics will weigh in, because, to be honest, I am wondering the same thing!!!
Bye,
Rick McInnis
Finally built my dedicated machine using all cic’s recommended components and optimizations with the exception of the sound card which I am using a Fire wire interface. All has gone fairly well except I had a software problem caused by Gigabyte easy tune that took some sorting out.
My machine is now operational and is making wonderful music but is experiencing issues playing files. I am constantly getting ASIO has requested the player to restart messages.
This happens sometimes during play between songs. While selecting files some will play some will not. Files will not play at all using up-sampling either.
I used Cmp & Cplay with the FW unit on my general use machine for about a month with no problems so I know the FW unit is not the issue. The files I am using where either ripped on EAC or converted in foo-bar; they all played on the previous machine.
Anyone have any ideas!
PC XPpro TC Electronic Konnekt-8 firewire i/o, 2-Tact 2150's, Genesis 500 modified speakers, Virtual Dynamics & XLO cables
Hi Uzeb,
I have a similar firewire model, the TC electronics 24D pairing with E7200, my music playback is quite stable and it never request reset in normal playback. However, my system is only partially optimized.
My experience with this model is that it will request reset only when it runs out of buffer, e.g. when I multitask with some CPU intensive program or when I set the sampling rate incorrectly. So I guess you could try to increase the speed of the hardware, either FSB, memory timings and cpu multipliers.
I also tried both the tc electronics asio driver and asio4all driver.
For tc electronics driver, I could only get down barely under 200samples even at 44.1khz, I need 256 samples for 96khz.
For asio4all driver, I could get down to minimum of 64samples.
I wonder what sampling rate are you referring to when you get the 128 samples ?
I am also a bit surprised that you can go so low on this number with the native driver. I have several firewire card, including pci-e, TI chips, pci via chips and can never get anywhere near 128.
Thanks for the post
I think I solved the problem.The power scheme setting seems to have been the culprit in this case. I changed from the recommended max battery to always on and the problem stopped. There maybe other issues at play here that I will try to check out but for now I am back to enjoying music.
I tried asio4all before changing the power setting and it worked fine; like yourself I was able to go to 64 samples. Whatever problem there is with the power setting relates to the TC electronics ASIO drivers.
"I wonder what sampling rate are you referring to when you get the 128 samples ?"
128 only at 44.1khz at both power settings, 256 at 96khz (at the new power setting only up-sampling didn't work at max battery)
also another thanks to cic's for his input on this problem
UZEB
PC XPpro TC Electronic Konnekt-8 firewire i/o, 2-Tact 2150's, Genesis 500 modified speakers, Virtual Dynamics & XLO cables
Hmm.. I recall getting that message when manually changing the ASIO latency value i.e. number of samples. cPlay> cPlay Settings> ASIO settings. Perhaps you could ease the ASIO latency value and see if the problem persists. From memory, I think the default was 512 samples for ASIO4ALL.
Bling,
Thanks for the reply.
Doesn't seem to be latency related tried this already. The fire wire unit uses a custom version of ASIO 2 and can go down to 128 samples without pops and ticks.
UZEB
PC XPpro TC Electronic Konnekt-8 firewire i/o, 2-Tact 2150's, Genesis 500 modified speakers, Virtual Dynamics & XLO cables
Had a similar problem with Lynx. cPlay was changed to accommodate Lynx's ASIO driver (I could do this as I use Lynx AES16). Strangely, first time I install driver I got no playback issues but any subsequent play caused resets.
What are you using to output firewire?
Try following:
- Set cMP to 'Optimize Critical'
- Make sure you have enough CPU headroom: for E7200/E7300, minimum multiple is x8 when FSB is 170MHz and outputting 192k
- Check if WDM driver is operational - do this by playing via ASIO4ALL (set ASIO4ALL to use soundcard and you should be able to go down to 64 samples). Be sure to reinstall ASIO4ALL after doing Minlogon change
- Do a clean re-install of ASIO driver using Device Manager (may need to do same with firewire device). Make sure you have latest drivers
- Can you test using Juli@? If not, test using onboard sound hardware with ASIO4ALL
Thank you for your reply.
The TC Electronic Konnekt 8 firewire I/O worked flawless using Cmp and Cplay with my non dedicated machine using on board chip.
FW output on my Cmp machine is through a PCI card recommended by TC Electronic with a Texas instrument chip
Optimize is set to critical
I have E7200 processor with the settings as described.
I have not done the Minlogon yet.
Would rather not use ASIO4All due to the 48k limit. Is this not also compromising the sound by going through the WDM driver. Or does being able to use a lower latency make up that deference.
I am I missing something or does WDM have to be operational to use a sound card with ASIO.
I will try all these suggestions when I get home.
Thanks again especially for sharing your work with us all.
Uzeb
PC XPpro TC Electronic Konnekt-8 firewire i/o, 2-Tact 2150's, Genesis 500 modified speakers, Virtual Dynamics & XLO cables
Would rather not use ASIO4All due to the 48k limit. Is this not also compromising the sound by going through the WDM driver. Or does being able to use a lower latency make up that deference.
Most ASIO drivers are implemented as WDM. ASIO gives better control over using these drivers. ASIO4ALL is definitely worth trying.
Interesting that you didn't get this problem in your test setup - try enabling your disabled devices as well as starting services. Then incrementally disable/stop them to see which may be causing this. It may just be that Windows Audio service is needed...
Release notes .
Change Log (2.0b11 release):
- UI refinement: reduce runtime footprint
Please REMOVE previous versions before installing cPlay 2.0b11. Normal (SSE2), SSSE3 and SSSE3 B9 versions are available.
For cMP² users, review these BIOS changes (underclocks FSB, CPU & disable additional USB items). Note: E7200 has been replaced with E7300. Same settings apply. Gigabyte mobo may be difficult to find, suggested new mobo is Gigabyte's GA-Q35M-S2 .
Here we go again am I the only one that likes '11' (not talking to you cics, but you are clearly welcome to comment)? Nobody has made a sonic value judgment other than I. The highs from 192 are simply beatiful. Yes when listening during times of bad powerlines (noisy) highs can suffer but this is true of vinyl, sacd, dvda so cplay at 192 is no exception. I just can't imagine it getting better but based on the track record we uually get new version of cplay each week or so.
I have enjoyed 10 and 11 about equally.
Maybe cics is playing with our assumptions but when he says that number eleven was basically a refinement of the user interface I do assume that that is the case. I do not see how that would make much difference in the sound quality.
I have noticed that #11 has greatly ameliorated the strange sounds about twelve seconds into a REQUESTED track or, of course, whatever the first track of a selection happens to be. On my machine this would happen every time I would skip within an album or requested a new album.
I was quite pleased with the sound of #10, and do not find there to be any difference in #11, caveats above to be repeated.
For those who find it hard to believe that we can listen to "192" I wonder if the reason is, in my case, I have an upsampling DAC, a BURR BROWN SRC4192 in my processor. This device is going to upsample if the signal it receives is not already "there". There is no comparison in my set-up between the sound of letting this chip do the work versus the computer doing it. It is night and day.
Maybe if I was listening to one of the NON OVERSAMPLING, non upsampling, chips I would see what Riboge and others find problematic with the "192" as implemented within cPLAY. I have played the system to diehard analogue hyper-partisans and they can hear a beneficial difference even when the change is made without announcing what is happening.
I would be curious to know what processor Riboge is using. I am using the K&K AUDIO RAKK Dac II, powered by a Paul Hynes shunt regulator power supply, which seems to mate very harmoniously with cMP squared.
I continue to be amazed at its capabilities, this combo, though I know that what comes before always exerts the greater influence on the sound.
Audio components and our perception of sound seems to reflect the mystery of the world in microcosm. We can get close to understanding what is at work but we will never, ever, really, KNOW. And that would take most of the fun out of the pursuit.
Bye,
Rick McInnis
Rick,
Are you still having this problem? Try easing the ASIO latency value say 16 samples. cPlay> cPlay Settings> ASIO settings. This may help.
Bling,
Hi! You are not alone, I also admire 11 - in my opinion it is a real summing-up of the previous 4-5 versions, but whatever the version I SIMPLY CANNOT UNDERSTAND, why are we deprived of the possibility to load multiple wavs. I may be thinking simplistically, but the logic is like this: we can load single wavs and we load multiple wavs via a cue file, the computer doesn't give a damn where are they taken from, hence we should be able to load different wavs from different albums without any degradation of sound. For our computers its only data, the same can be said about cmp and cplay. And changing user interface requires little compared to adding an opposite phase button (with more sophisticated procedures behind the task of inversion), that most of us never use.
I am not ungrateful, no! I really open new music dimensions thanks to Sics, but this makes the fine program handicapped. Like an olympic champion sprinter with legs bound and living on a dripfeed (cuefeed).
I live away from home now and alone, so I have to do a lot of work around the house and I prefer to do it with music playing. Cmp+cplay is my single source of music, and inability to create playlists makes me sad. I made new friends here, and some are really interested in computer playback, but they constantly ask to load their favourites from different albums. Imagine how a wonderfull night easily turns with all the fussing around into a shamefull nightmare...
Now, that was said. What do You think? and others...? Am I alone?
Serge.
p.s. Also I'd like to say that Sics is doing all the work alone, while we simply and egoistically "consume". How about joining in to help him? I can design graphic part for cmp-cplay with a 256 (16) -color limited palette, and I think there are others who can help much more without adding our names to Sics'. It is surely HIS great achievement.
Edits: 11/27/08
I agree about playlists. However, you can do something similar by using foobar to make multifile, from any number of albums, cuesheets IF you add paths to the filenames afterwards. The cuesheet will cause only the first selection's album and artist (or whatever you put at the top of the cuesheet for title and performer) to display in cMP2 but will play the files selected in the sequence selected. Playing a playlist truly would involve cPlay changing the album and artist/performer displayed with each track. You can see how this might be done for flacs from their tags, but how would you do this for a list of tagless wav files? If you had a library with hierarchical folders, I guess you could get these from the path.
don't give up about the KINGSTON HyperX Ultra Low Latency (KHX6400D2UL/1G.
I just found some (unused) on ebay(germany) for 60€ each. Walter
It would make for an interesting comparison.
Would 1Gb UL be superior to 512Kb LL?
Will we ever know?
Bye,
Rick McInnis
My single 512MB Team Xtreem 800MHz 3-3-3-8 RAM module performs very well in my system. I've also heard good things being said about Winchip 1200MHz CL5 RAM kits. For those dead set on using Kingston RAM, some of my friends swear by their Kingston DDR 1150MHz CL5 kits, so they might be worth a try.
Easy, theob. I like it too and agree about the highs though I haven't taken care to do a careful comparison to the previous version. It's the holiday season. However, I cannot listen to 192 files and wonder how you can.
no disrespect meant just curious. if you go to mytek homepage (see link in one of my previous posts) you can get files that are 192.Happy Holidays!
Edit: oops just checked cplay diagnostics some of these these files may be 24/96 and/or 24/88, my mistake. But afaict(as far as I can tell) Trio is 24/192.
Thanks for pointing that out to prompt me to doublecheck.
Edits: 11/27/08 11/27/08 11/27/08
I still prefer no upsampling option when listening to cPlay 2.0b11. It sounds significantly more natural and "live"ly on my system, but loses a bit of airiness and clarity compared to 4x upsampling.
ok gang/team I'm back at 192 /internal clock. I'm not saying a good external clock won't be better but my panasonic dvd external clock is now out of the picture with 2.0b11. when I compare internal vs external clock at 48khz external still wins. but internal/192 beats all.well done cics, imaging and air are outstanding.
wow more listening makes me conclude you have spun a masterpiece in 2.0b11. More details later.
Edits: 11/23/08
Additional listening confirms initial reaction. Highs are very sensitive to input, if something is off '11' will tell you. If all is right then '11' is very open, airy and wonderful. If you have 192 files you're in for a treat. Try link below especially avalon and trio.
Interesting. I'll try out 2.0b11 at both sample rates, and see what kind of conclusions I get on my system.
That's not a Toy... IT'S A TOOL !!
.
Hi cicsI always use EAC for ripping to single-file archives. I believe it preserves the special LSB encoding when copied this way.
dBpoweramp Music Converter has a DSP feature to decode the 20-bits of HDCD and give you a 24-bit LPCM file...
...but that requires additional processing steps, time and HD space... and is perhaps contrary to the spirit of cPlay.As an example, I've got two versions of Close To The Edge by Yes (Rhino 2003 remaster and Japanese HDCD remaster). The decoded, 24-bit version sounds "cleaner". This was judged using 44.1 SR playback, so as not to confuse the issue with upsampling artifacts. And yes, I did adjust for the 6 dB difference between the two formats (...love that precise volume control !)
Wouldn't it be Grand if you could play the 16-bit/44.1/HDCD file and get the real thing coming out the other end !
Cheers - and Thank You again cics, for the incredible gift of cMP-Squared...
GrantThat's not a Toy... IT'S A TOOL !!
Edits: 11/19/08
Scaling from 20 to 24 bits (x16) is linear and will not impact SQ as no resolution is lost. Problem occurs when going other way, for example 20 to 16 bits.
Hi cics
Forgive me for not being precise in both my request and subsequent reply.
THE REQUEST: for cPlay to decode and play HDCD-encoded files directly from 16-bit rips.
As you know, HDCD promises 20-bits of resolution squeezed into the 16-bit redbook format. To expand back to 20-bits, parameters coded in the LSB instruct the hardware HDCD filter (in the player) to modify the amplitude accordingly. We've seen this approach with other compander systems (Dolby, DBX, Hi-Comm, etc.) primarily for tape storage (and analogue satellite transmission).
EAC captures this information faithfully - it's just bits to EAC. A CD-R burned from one of these rips will light up the HDCD indicator on an enabled CD player. One software player I'm aware of (from that Big Company in Redmond...) can decode the data, but it requires a physical disc to be present in the computer's optical drive.
Since the Redmond Boys bought the Pacific Microsonics technology, they've added HDCD to their media player (oops... did I say that). If you were able to incorporate such processing into cPlay, the sound of lawyers banging on the door would likely follow.
THE ALTERNATIVE:
Requires a two-step process... a) Rip with EAC to 16-bit (as usual), then b) re-process with dBpoweramp Music Converter - creating a fully decoded 20-bit signal - stored in a standard linear 24-bit PCM file (lowest 4 bits zero-padded).
24-bit files are obviously bigger on storage needs, however playing them via cPlay would avoid additional load on the processing chain (which likely yields higher SQ).
Heh-heh... I think I've thunk myself out of the possibility... what do you think cics ?
Thanks,
Grant
That's not a Toy... IT'S A TOOL !!
A great pity as 20 bits resolution is more important than sample rate. That is, a recording offering 20/44.1 is superior to 16/96 (high quality SRC upsampling makes rate less important).
Just imagine if the music industry offered all new CD releases in HDCD as standard AND Microsoft making this freely available.
Release notes .
Change Log (2.0b10 release):
- DSP refinement: 128 bit DSP processing
All critical areas changed to use advanced 128 bit processing capabilities (i.e. vectorisation & streaming) of SSE2, SSE3... processors. SRC upsampling is excluded (which only benefits from compiler optimisations).
CPU load reduces and RAM I/O efficiency improves.
- UI Refinements: Phase icon indicates 0 or 180 (previous '^' indicator is removed). Settings provides diagnostics information (available RAM, volume and phase).
Please REMOVE previous versions before installing cPlay 2.0b10. Normal (SSE2), SSSE3 and SSSE3 B9 versions are available.
For cMP² users, review these BIOS changes (underclocks FSB, CPU & disable additional USB items). Note: E7200 has been replaced with E7300. Same settings apply. Gigabyte mobo may be difficult to find, suggested new mobo is Gigabyte's GA-Q35M-S2 .
cPlay 2.0b10 is one of the Best Releases of cPlay. Inexpressible musicality.
It has turned out to be my favorite music player. Real hi-end sound. It is as if I bought a new Amplifier. Try it out to believe.
I tried Foobar2000, XMplay, AIMP, It beats them all.
Thank you very much cics to develop such a wonderful player!!!!!
Sorry for this obvious.. stupid.. question..
I've searched through the entire thread and cannot find a word about apple lossless or ALAC... While I am not exactly a long term user, I've got all my songs (quite alot of them) in ALAC formats...
May I ask, is cplay capable of ALAC format playing?
(sorry could be the wrong place to ask, as I've read mixed feelings on this, is alac inferior to flac?)
THANKS!
KK
Hi KK
You can use dBpoweramp Music Converter to transcode your ALAC library to FLAC. See -> http://www.dbpoweramp.com/
It has a batch mode (definitely useful to your purpose)... set things up and come back in a day or so to cMP-compatible Music !
Hope this helps.
Cheers,
Grant
That's not a Toy... IT'S A TOOL !!
thank you so much for your kind help!
.
I have now upgraded from 2.07b9 to 2.10b9. Thank you very much!
Very very beautiful highs, more hall sound, better dynamics. Does '...SRC upsampling is excluded (which only benefits from compiler optimisations)...' mean there is no upsampling or just -121 src? Also does it mean that -145 src is to come in a future release?
I don't want to overeact to this wonderful release until I've had more time with it but so far aces!!
Edits: 11/16/08
Absolutely the best cplay yet. Everything is better. Checked in cplay diagnostics and there is no upsampling at all, wow. This version has that '...everything feels right...' kind of sound though, even as I have associated -146 src as the only way to great imaging and highs. To me again the focus is on the transparency, the highs, the imaging, especially on big classical power pieces. Very nice. Please cics I'm not trying to pry (just understand), correct me if wrong. Is src still a potential future release?
The comment "SRC upsampling is excluded..." refers to 128bit processing and NOT as in total exclusion. Likewise FLAC decoding gains limited benefit from 128 bit processing. SRC & FLAC require changes to its source code for explicit 128 bit processing. I say explicit because FLAC and a small part in SRC benefits from automatic 128 bit translations (vectorised loops) done by the compiler (which translates source code to low-level machine instructions your E7200 understands). cPlay 2.0b10 uses explicit 128 bit instructions (SSE2) in all critical areas to fully exploit these highly advanced CPUs. By this I mean calculations are vectorised (called SIMD) and RAM read & writes are more efficient (which includes streaming). I initially expected to release this in the new year but b9 set the platform for SSE2 early! Don't be fooled by the low cost of these computer parts (CPU, mobo chipset and RAM) as they are in fact extreme high-end audio technologies.
So if you have selected -145db SNR and output is 192k, cPlay will upsample using the same SRC converter when input is not 192k.
I didn't expect a large improvement in SQ given the already excellent performance. It certainly surprised me! In my setup, HF resolution and additional detail is just remarkable. Mids gain handsomely (tube-like sweetness & emotion) and vocals are impressive! Every aspect in performance (transients, bass, imaging, low-level detail...) improves.
Dear cics,There has been so much progress with your approach, though even from the start (AOB) what this was able to do was so plainly obvious, the ability to deliver a complex waveform intact, still is the best way for me to describe what is the basis of what makes cMP great. You have added much goodness to the basic platform as time has gone on. To those of us who cannot wait to try the latest version, as with any evolving thing, there were improvements in one area and, maybe, a diminution of "goodness" in another area. Some iterations were more balanced in the hits and misses department than others.
I have enjoyed listening to all of them, except when I was without a working MB. This has been a learning experience, as trite as that phrase is. One has learned to listen more carefully and has learned what to listen FOR.
There is no question this latest one is a sort of culmination and more.
One has become so used to incremental improvements than when faced with a big improvement one tends to doubt what they are hearing. It is so easy to exaggerate differences in any component, and in our case software implementation. In my case, I have made the case obtuse by implementing two very important hardware modifications at the same time. Nonetheless, I think there is some kind of synergistic synchronicity in the appearance of this latest evolution of cics's project and being spurred on to remove the break-out cable from JULI@ and Bernd's suggestion to lower the voltage within the ANTEC EA-430. Of course, this built upon the worthwhile improvements brought about by Alfred's suggestion to add bypass capacitance to JULI@ and the MB. None of the mods, I presume, would have ANY (or much) meaning without the platform provided by cics. Inversely, these mods have brought out even more of the goodness offered by cMP squared.
All in all, at this point in time, and as we know cics never sleeps, who knows what he will discover next, but the quality of sound coming out of these unlikely devices is simply extraordinary.
Over the weekend I am hearing tiny details never heard before in recordings I have listened to hundreds of times on various CD playback systems and a variety of LP playback systems. There they are, vocal inflections I have never noticed. Quick as lightening grace notes from guitars and violins I simply never noticed before. I think to myself that I am merely WANTING this to be so. How could I have missed such details, I assure myself I just forgot they were there. And yet, this keeps happening.
I am listening to one of my favorite albums to play at blistering levels,
PIXIES - Surfer Rosa. And even here I am hearing little things Joey Santiago is doing with his tone and inflection. I hear details in Kim Deal's bass lines I never noticed before. This is NOT a subtle album and yet it abounds in subtleties never heard before. I am hearing useful musical information that was obscured before, not shuffling feet or extraneous breaths.Plenty has been said of the beauty of the high frequencies and the thunderous tuneful bass; not that there has been a lack of midrange information and musicality, it is just that there is an astonishing quality to the midrange now.
Simply amazing.
Thanks,
Rick McInnis
Edits: 11/17/08
I can only agree on what was said here. One little comment, about something that is kind of amusing but also amazing to me. If you are not native english speaking it can be sometimes hard to follow some lyrics. What amazes me is the fact, that it is so easy now to understand the meaning of what was sung and additionally I can also hear the slang of the singers, where in the past it all sounded very much the same. All that happens very naturally and makes me sometimes giggle. Good night and all the best to the US citizens for your future. Walter
You said...'It certainly surprised me! In my setup, HF resolution and additional detail is just remarkable. Mids gain handsomely (tube-like sweetness & emotion) and vocals are impressive! Every aspect in performance (transients, bass, imaging, low-level detail...) improves...'
I no longer am surprised by the sonic benefits various levels of cplay continue to acrue (I have confidence in your ability) but I'll have to admit this was a huge step up. Well done!!! I'm glad you are commenting because I was beginning to think I was the only one hearing such sonic marvels.
Where are those guys who disagreed with your theories on computer audio now?
To the doubters I say 'cPlay 2.0b10 this'. One listen to cplay 2.0b10 in a cmp shell will convince.
.
I agree with theob. Upsampling to 192Khz has always been my default option when I listen to previous versions cPlay. The 2.0b10 release, however, is completely different. Don't get me wrong, upsampling with b10 is VERY beautiful, but b10 with no SRC upsampling is one step closer to perfection.
Hi cics
I found this idea mentioned once before (buried somewhere in the posts)... Is it possible to bypass/ ignore any INDEX 00 entries when processing CUE files in cPlay ?
Listening sequentially through a whole album, it's not any trouble, but if one were to skip between tracks (or use the "REW" and "FWD" controls) it's not usually a clean transition. Either you have a silent pause preceding the sound of music, or worse, the tail of the previous track.
It would be much appreciated (at least here in my Listening Room). Thanks again cics for the incredible gift you give us with the cMP System.
Cheers,
Grant
That's not a Toy... IT'S A TOOL !!
Yes, I already commented the same thing before -http://www.audioasylum.com/forums/pcaudio/messages/3/34823.html-.
Also, I am embeding the cue sheets directly in the FLAC files. Then I have only singles FLAC files but with all tags and contents. Foobar recognizes it perfectly and show me all the tracks but cPlay only show me a single track for the whole file. It would be good for cPlay that the cue sheet could be loaded from the FLAC file -if exist- in the same manner as if it were in a separate CUE file. I believe that it is not too much work.
Regards.
Ignacio.
a few weeks ago alfred (sonics) mentioned a case with a builtin dc/dc converter which allowed him to power his whole intel-atom330 system with an external 12v psu (single rail 12v, nothing else)during my experiences powering the gigabyte mobo with such a dc/dc converter (even batterie is possible, condensator mods are much more easy to apply) i found out that SQ did increase (to me) after reducing voltage at the atx 12v output rails.
i was corious if my antec psu has an apprpriate switch to lower voltage, and is has...
one of the moments i realy enjoy in lifethere is a potentiometer inside the psu-case which allows you to regulate the 12v output to a lower level. the potentiometer can be found on the PCB of the psu (close to the case)
if interested you can test for your self by adjusting the potentiometer counterclockwise until stop. the result at my side is exactly 12v output, BIOS says 11,97v
definitely unplug power before doing this! :-)
do at your own risk...in the past my system was set to 96k playback, with lowered voltage i begin to enjoy 192k, what ever that means....
maybe i am totaly wrong, it would be nice to get some feedback..
i like it
Bernddid you understand my crude english?
this is an absolute fiendly message :-)
Edits: 11/11/08 11/11/08
Berndt: I have a 550 watt AGI power supply (I should have gotten the Antec so I could be in sync with you and Rick, but I didn't). Anyway I'm thinking maybe there is a similar potentiometer on my AGI. I looked but could not find. Did you have to take the case apart to see and access the pot on the Antec?
Edits: 11/18/08
hi theob, yes, i had to take the case apart to access the pot.
i dont know anything about the AGI psu, but can recommend the antec. i tried several psu-versions from corsair to zalman to fortron. the antec is clearly superior. i had the opportunity to measure voltage ripple/noise of the antec psu with a scope. in case of the antec voltage ripple/noise is very low (about 4mv max). the differences are clearly audible.The AGI psu may be a good choice as well. i simply dont know.
Edits: 11/23/08
Dear Bernd,
I was able to do "the mod" (wish all were so simple) this afternoon and I do hear an improvement in most areas.
Starting my machine after the adjustment was a little strange.
Since I realize that my CPU does not seem to run at the low voltages (I figure it has to be this CPU since the e5200 I was trying out would go VERY low) I set the voltage to 0.975 before making the adjustment. Even so on turn the machine would get to the WINDOWS screen and then the screen went black, which is weird since my default screen color is white.
I turned it off and this time went into BIOS. BIOS had reset a number of settings, notably CPU speed (did change voltage) and memory had reset to AUTO.
I left all alone, closed BIOS and let it proceed to WINDOWS, and this time, into cMP. Relief.
I did notice that with the memory set to AUTO, according to CPUZ my CAS readings were higher than previously (before adjusting voltage, I am assuming since I had not checked this since initial installation). After hundreds of hours of burn in is it possible for the memory to "change". I am at a loss for the proper vernacular.
I do notice, and I do not think this has anything to do with the voltage adjustment, that when a track is SELECTED versus being played in order automatically, that about ten seconds into the track there is some kind of glitch sound. If one starts the track over again, it occurs (I think) in the same spot. But, again, seemingly in the same "spot" with any track so selected. I wonder if I need to re-install cPLAY or cMP, or both?
Turned it off, again, reset BIOS to cics's specifications and all went fine.
I do not think I am dreaming when I ascribe an additional sense of spaciousness. Vocals, which I am most sensitive to (not that this makes me unique in any way, but I listen to songs more than orchestral music) have clearer textures and the tonal complexity is revealed with ease.
For those with this amazing little device this is a way to make this unassuming thing even better. Who could believe it to look at it?
Seems the rebate offer is gone (at least it is at NEWEGG). Still pretty cheap for the performance.
Thanks Bernd for your inquisitive nature and then telling us what you found. Now to try the battery power with the DC-DC converters!
Bye,
Rick McInnis
Rick, you wrote:
I do not think I am dreaming when I ascribe an additional sense of spaciousness. Vocals, which I am most sensitive to (not that this makes me unique in any way, but I listen to songs more than orchestral music) have clearer textures and the tonal complexity is revealed with ease.
Well, this describes exacly my experiences.
I am happy to hear that and that you tried the mod. I appreciate it to get some feedback here....
Intrestingly i had never any stability-issues after applying the potentiometer-mod. Glad to here, that you made it a everything is running fine.....
Bernd
Hello i'm new here and have also build the 'soundmachine' per cics reccomendation. To my surprise my Antec Earthwatts gave 12,3V (bios) and 12,23V (digi-meter). Lowering it to 11,93V by using the potentiometer clearly improved the sound!
Thanks Bernd!
I suspect your English is just fine. Though you have to give the language credit for not having words with three consonants and few compound words, though I like the thought of compound words. Any device that can add subtlety to language is good.
I would think learning to HEAR English is much more difficult than writing it. So many words that sound the same and can only be divined with context. Tricky for the unfamiliar. Trick for plenty who are familiar now that I think about it. I do not know how often this happens in Romantic and other Germanic languages.
The reason I went through my overlong introduction is that I am not sure if I got what you were suggesting we do.
Are you just minimally lowering the ATX voltage to just below 12 volts are going more drastic? Please tell me one more time.
Over the weekend I disconnected the internal fan on my ANTEC 430. The case gets slightly warm. My house is kept chilly so it might require the fan in the SE US summers. I also eliminated the IEC power cord input because I hate those things. I assembled a power line filer per Jon Risch and soldered my CARDAS power cord to its input and the output of the filter to the wires that were once connected to the IEC directly to the filter; soldered of course since I have no crimping gear.
Well, I thought it was worth the effort but who knows. It certainly cannot hurt since the filter does isolate, to some extent, the computer from the rest of my equipment.
I am always anxious to try these things so your note has made my day.
Please re-assure me how much lower you have attempted to lower the voltage. One wonders how much input voltage the MB based regulators require to behave at their best. How many different voltage regulators and different voltages are supplied by the ATX power cable? I have no idea.
Very interesting!
Thanks,
Rick McInnis
Rick, addjusting the potentiometer inside the antec case to the most left position gives you an output voltage of nearly 12v. (BIOS says 11.97v)
5v and 3.3v lines are not affected by the adjustment.
During my experimets with a dc/dc converter i was able to go down to 11.4V. The dc/dc converter has no circuits for regulating the 12V line (input=output). So i could feed the hole system with 11.4v by reducing the input voltage to a value of 11.4v. System was still stable, but SQ was loosing (less dynamic and a bit lame). To my ears 11,9v is optimal.
The goal is to power the system by batterie. I have done this already and SQ was winning big time! There is some work to do. I am searching for a circuit that to allows to plug in an external 12v PSU while the batterie is pluged in too. In "pluged in state" (psu) the batterie should be loaded AND the system should by powered by this external psu. It may be easy to accomplish this but there are some more things to consider. When running on batterie-only the voltage is dropping constantly. The PC may suffer damage if voltages goes to low and SQ is loosing below 11.9v also. (my findings). At minimum i want to have a warning LED.
Currently i am testing a new dc/dc converter that also regulates the 12v line. it is important to get a defined (regulated) 12v output, better 11.9v, so SQ is at high state all the time when system is powered by batterie only. Regulated dc/dc converters have a voltage drop. Mine is dropping voltage by 0.4v. So everything is fine when the batterie has an output above 12.4v which gives always a regulated atx-output voltage of 12v. If the batterie is down to 12v i am getting only 11.6v which is not enough.
A dropvoltage of 0.4v leads to the following "input, regulated output"-situation:
input, regulated output
-----------------------
25v, 12v (max input voltage)
.
.
15v, 12v
14v, 12v
13v, 12v
12.4v, 12v
12.3v, 11.9v (perfect)
12.2v, 11.8v (not enough)
12.1v, 11.7v (not enough)
.
.
So an regulated converter is solving the problem with high input voltages but when the batterie is drained too much things are getting worse. The solution may be a fat 12v-batterie so the voltage is not falling off too fast. The solution may be also a 18v-batterie. In case of a 18v-batterie the given batterie-capacity can be used more completely because batterie-voltage can fall to 16v or so without sideeffects.
I keep you posted, but it may take some time. I am constantly on it....
Well, now you know the hole storry. Originaly a had no intention to write it in detail, but it came out while writing....
i have done this consulting my dictionary only 10 times or so :-)
In short:
- When running the system by batterie SQ is absolutely winning.
- Reducing the voltage of the atx-12v-output has a big impact on SQ. To me SQ is definitvely increasing. My antec psu has an output of 12.4v. This value can be easily reduced by rotating the potentiometer inside the antec counterclockwise until stop. This leads to a 0.4v lower voltage, which is a lot compared to the BIOS-voltage-tweaks.
My new dc/dc converter deliveres 11.97 atx-output by default, so i think it is inside the specification.
Currently i dont care about the MB-onboard-regulators (maybe later). Interstingly PC-Health Screen in BIOS reports always exacly the 12v-atx-output voltage (12.4v with my unmodified antec psu). Has the MB 12v-regulators at all? I am no expert here...
Hm, i am wondering if i answered your initial questions :-)
thank you for new inspirations regarding the antec-fan and the Jon Risch Filter. The very good antec psu (tried many, antec IS best) is my reference when it comes to compare SQ.
have a nice day,
Bernd
Bernd,
It seems to me the trick with batteries, or any alternative power supply, is getting the voltages to "turn on" in the right sequence.
When you used the battery, I am assuming a SLA type (sealed lead acid), did you use the biggest one you could find? What rating did you try? Many good things are said about the OPTIMA RED TOP batteries.
All batteries will slowly discharge, as you say, but a REALLY big one would give you much more margin (I know this is obvious to you, I am trying to get an idea of how long you were able to listen while the battery was still in the optimum range).
I have never heard anything good said about a dc-dc converter other than they do what they are supposed to do. I would think a really good shunt regulator would be the best approach for the CPU and for the sound card. I am not capable of designing such a thing. There is someone who is and when he finds time to get more involved in this he might.
Are you regulating the battery power? Please clarify.
When you used the battery power (I am assuming for the ATX header) when did you plug it in? Before or after turning on the ANTEC? Did you disocnnect the 12 volts leads from the ANTEC for the main power header? I am not clear on whether the ATX and the regular header are interconnected in any way?
I have had good experiences with using batteries for power supplies. Nothing is perfect, but they do many things very well.
Thanks for reviving this inquiry.
Bye,
Rick McInnis
Good evening Rick, (it is 09:50pm here in germany now)
your question:
When you used the battery, I am assuming a SLA type (sealed lead acid), did you use the biggest one you could find? What rating did you try? Many good things are said about the OPTIMA RED TOP batteries.
reply:
i only have a small SLA-type battery. i feared to spend more money for a realy big one just for testing. the battery has a capacity of 9Ah. after 1 hour the pleasure is over.
your thoughts:
I have never heard anything good said about a dc-dc converter other than they do what they are supposed to do. I would think a really good shunt regulator would be the best approach for the CPU and for the sound card. I am not capable of designing such a thing. There is someone who is and when he finds time to get more involved in this he might.
my findings:
the dc/dc converter was laying around i simply used it. in the past i tested serveral PSUs and got a feeling how ripple-voltage and -noise affects SQ. i could not recall those feelings while using the dc/dc-converter. sure, ripple should be measured, i am on it..
question:
Are you regulating the battery power? Please clarify.
reply:
during my tests i used a PSU with adjustable output-voltage.
question:
When you used the battery power (I am assuming for the ATX header) when did you plug it in? Before or after turning on the ANTEC? Did you disocnnect the 12 volts leads from the ANTEC for the main power header? I am not clear on whether the ATX and the regular header are interconnected in any way?
reply:
Rick, now i understand your irritation! the dc/dc-converter is a complete tiny atx psu with a single 12v (max 25v) input and 3.3v, 5v, 11.97v output. so the hole system is driven by battery and there is no antec PSU involved.
Sorry, i was so much focused on "this a dc/dc-converter" i forgot to mention, that it is is an atx-psu also.
Bernd
Bernd,
When I read over the INTEL application sheets for CPUs I got the feeling that there was some order/protocol for powering.
From what you have said I would assume, with the battery, once the switch is in the ON position all voltages come on simultaneously?
I guess I was confused by your use of the dc converter since you only spoke of 12-12. Are you saying that with the 12 volts battery you are using several (two or three) of these converters to get the needed voltages?
I would think if you like the results you have got, so far, get a BIG battery and all should be even better. The equivalent series resistance of the OPTIMA Red Top is minuscule, a very important specification for a power supply.
Your experiment has me inspired and ready to add more batteries to my listening room.
Just looked at DIGIKEY at dc-dc converters. I am wondering if one would be OK with letting the battery supply the 12 volts needed with nothing else but a bypass capacitor and use the converters for the 5 volts and 3.3 volts. Is this, in fact, what you are saying to do?
I see there is an OPTIMA Yellow Top battery with a 75 amp/hour rating. This might be ideal.
So how long was the sound GOOD with the 9AH battery?
I hope I am not asking too much but you can tell I am intrigued by your intrepid experiments.
Thanks, again.
Bye,
Rick McInnis
Hey Rick,sorry for the delay...
I use one small dc-powered atx-psu. It is called picoPSU. This PSU is a fully compatible ATX-PSU which deliveres a combined output power of 120 watts.
The bypass capacitor (4x4700µF panasnic FC) is connected directly to the 12v input-line. The PSU looks like an ATX-header with a small PCB above the header. You could do a internetsearch for picoPSU to get some pictures. There are several different models but i hesitate to recommend one until i am realy sure that they deliver clean power and all parts are fitting together. -> I am looking for a concept thas allows an external 12V psu to charge the battery while powering the mobo (because the picoPSU is powered by this external 12V psu at the same time). running by batterie only should be done by simply unplugging the external 12v psu. I like it simple.
The output of the 9AH battery is dropping from 12.8V to 12.4V in a period of 1 hour and 30 minutes. SQ is GOOD in that periode (more natural less sharp). After that periode system is still stable but voltage is dropping rapidly. A battery with 75AH should be more than enough. A big battery has a lower inner resistance which is good for SQ as well. A lower inner resistance / higher capacity can achieved by running a few smaller batterys in parallel-mode also. (did some mailing with alfred, he mentioned this)
Bye,
Bernd
Edits: 11/18/08
Bernd,
I got the 200 watts version with the ATX12V wire attached.
Can't wait to give it a try.
Thanks again,
Rick McInnis
Bernd,
Since you have led the way you get stuck with the questions.
Do you use the ATX-12v header? I would think you would, but re-assure me.
Is a connection for this simply added to the main ATX connector? I assume the end 4 pins of the ATX connector are left empty?
It seems that most of SILENTPCREVIEW's reservations about this were based on the giant wall wart 12 volts supply, SO, a big battery should be very good.
The whole thing is easy enough to try. I will use the battery out of my car to see if it is worth going ALL THE WAY with a monster battery.
I can't wait to try this.
Thanks,
Rick McInnis
Bernd,
So you are using this PICO and instead of powering it with the "plug in the wall" 12 volts supply you connect it to the battery?
Very interesting, and SIMPLE.
So, this device will run the cMP squared system (GIGABYTE MB and JULI@)?
Please reassure me!
Thanks,
Rick McInnis
For others interested in what this thing looks like:
http://www.silentpcreview.com/article601-page1.html
exactly Rick!
the picoPSU is powering the whole cMP squared system.
the end 4 pins of the ATX connector are left empty!
Right now i am using the version picoPSU-120-WI-25V which has originaly no ATX12V wire. I used an ATX12V wire from an old ATX-PSU and soldered it to the picoPSU-120-WI-25V.
Tried to attach an image.....
The version of the picoPSU you mentioned (http://www.silentpcreview.com/article601-page1.html) has NO 12V regulators on board. If you are feeding this psu with 12.8V the output will be 12.8V.
Cant say anything about the 200W version.
Bernd
Bernd,
I keep trying to think of a way to lose some battery voltage and it occurs to me to use a SCHOTTKY diode, which would be just about the right loss.
Do you have any thoughts about this (or anyone else for that matter!).
It seems perfect at the moment, who knows what tomorrow may bring?
Bye,
Rick McInnis
hm, i tried to learn somthing about the Schottky diode.
The diode has a voltage drop of 0,4v. The diode my be usefull for lowering the input voltage but not for regulating the voltage at a constant level.
Bernd,
No intention of regulating a BIG battery, just want to lower the voltage so it comes closer to what was being supplied by the switching PS. I am assuming that a 75 Ah battery with a generous capacitor bank will be inherently regulated. I would think the load imposed upon the battery by a computer running cMP squared will be fairly constant. I repeat, I am assuming and would appreciate any dissenting comment.
I got my picoPS Friday but I am having to wait for a battery. Initial trials with just a voltmeter show negligible voltage drop using all sorts of power diodes. I guess there needs to be some real current flowing to get the drop, at least I hope so.
My reasoning was that a simple high quality diode would do less harm than a voltage regulator. I do not know the answer. Nonetheless, my thinking was most concerned with lowering the voltage. Schottky's have the lowest voltage drop, I may need to use a conventional diode to get closer to 12.0 volts, or just under that.
The three terminal regulators will consume 1.3 volts and this is too much. In addition, you do not need to regulate a large battery, there are plenty of regulators downstream (?).
The picoPS came equipped with really low quality capacitors. After making sure all is working I am going to replace these with BG's. I was a little surprised since the thing was not cheap that these were what were there. At least, with this larger board they will be easy to replace.
I ordered the version with the extra four pins 12 volts connector. It is ludicrously short. I am thinking of running a separate line to this connector from the battery's capacitor bank, anyway.
Still this seems to be the best approach to powering a music PC. Thanks for bringing it to our attention. By the end of next week I hope to be able to hear it for myself.
Bye,
Rick McInnis
I think your idea of using the diode to drop the voltage is very good. A regulator will consume much energy, requiring heat sink and etc. Normal Silicon diode with enough current rating will also do fine as switching speed and noise is a non-issue in your case as there is no ac reverse voltage.
The only thing that I may change is to save money by using low-impedance caps rather than BG. I have done quite some experiments with motherboards, and batteries powering other smaller digital boards. It seems that bulk capacitances, ie. higher uF is more important than pure sonic quality of caps. I have switched to much bigger capacitance in Rubycon ZLH, panasonic FM, chemicon, etc powering my various digital-audio devices.
Dear Promethk,
I have hoped you would put your thoughts into this project.
I have felt bad since leaving you out in one of my posts thanking all who have made useful recommendations to the improvement of the hardware. It was you who first approached improving the power supply. Your absence and my bad memory conspired against me.
One thing I have noticed with my experiments with the diode using a 10Ah 6 volts battery and nothing more than the multimeter as a load, that there is no, or not much, voltage drop. I am an electronics dilettante so I do not KNOW how the diode works in detail. I have to assume that without a real load there is not a voltage drop? Is this true?
I have ordered a 75 Ah OPTIMA battery and a few different power diodes with a variety of voltage drops to see which is best. I tend to think, per Bernd's experiments, that a slightly lower voltage might be better than hitting 12.0 volts. Have you by chance tried lowering the voltage of your power supply?
Your advice on the capacitors is taken. I have found that Ryeland's advice is true, as with any audio device there is always a best value. As much as I hate being scientific (let's get on with the listening ....) I will attempt to find a value that is best, in my case, at least.
Thanks for your advice. Your help is and has been appreciated!
Good to hear from you, again.
Bye,
Rick McInnis
Glad to work together on getting better power for music production in PC. While such work has a long history in other audio equipments like amplifier, cd player and dac, relatively little is known on how to do it right in the PC.
To get proper voltage measurement, you can use a resistor connect in series to the diode as dummy load and measure the voltage on the two ends of the resistor, any resistor from 120ohms to 1000ohms would be fine, something smaller will also work as long as you watch for heating issue.
Its true that I always have observed slightly higher 12V readings, possibly due to running the power supplies at only 10-20% of its rated capacity. As my current readings are not too far off, I am not tempted to try it yet.
You may want to try download the OCCT program which provides measurement of the 12V,5V and 3.3V, etc and provide nice curve to summarize the voltage fluctuation. Although this program relies on readings from motherboard sensors which are known to be not precise, it could still be useful to have a rough idea of what is happening in the motherboard. If you are impatient, you can set the voltage monitoring time to 6 minutes or you can stop the program anytime.
Surely a diode in series with the supply would drop the voltage by 0.7 volts irrespective of current - that's the nature of the device. (Two in series, 1.4 V and so on.)
As Capacitance (Farads) = (Period x Current) / Ripple Voltage, it is useful to have some idea of the initial ripple in your system: just adding more and more capacitance could be a waste of money and might even put a PSU at risk. (The ATX spec includes maximum values for capacitance load of, from memory, ~5,000 uF but I have little idea how how much added capacitance is safe in practice. Obviously, this does not apply to batteries.)
Whatever, I recently measured ripple on a cMP2 system running on the now-out-of-favour Biostar mobo with a fanless Nesteq PSU. On the main power connector, it was about 30 mV on each line but about 350 mV on the processor's 12V line. I was also running two 2.5" HDDs off the same PSU and ripple (measured at the drives' plugs) was as low as 3 mV.
That suggests that the 12v ATX4 plug is the place to start. As the period is not the steady-as-you-go 100 (120) Hz of rectified mains or the 0 Hz of a battery but is determined by the fluctuating power demand of a high-speed circuit, low ESR is presumably, as suggested, crtical. I don't have a 'scope so that was as far as I could go.
If you're doing experiments, it would be most useful were you to measure the effect on ripple at the ATX4 plug of different values for C. Yes, I know I could do it myself but, though I've bought some of the bits, I've not done it yet.
If better minds than mine think this is nonsense, please shout.
Best
Dave
I am quite concerned in adding caps to the pc due to the fear of the caps limit and there is not much progress in earlier. Currently I have reached up to 30000uf total and the system is very stable. I also hope to find an answer to the maximum value.
Yup, getting good measurements are the best way to go. Unfortunately, my aging digital multimeter is limited in resolution and I don't know how accurate its ac measurement is and if it gives true rms readings. The 350mV is kind of much bigger than what I imagine.
Bernd,
The image is clear as can be.
Maybe I should have waited to hear about the "wide-open" battery voltage to the MB. Of course, I should have but the excitement was overwhelming!
I can always try the one you suggest. One part of me is worried about the potential of too much voltage, the other LIKES the idea of unadulterated power (no manipulation by regulators) for the all important CPU.
It will be instructive to try them both out.
This certainly has the potential of solving the PS question. Of course, it means bringing another battery into the listening room!
I will let you know what I find.
THANKS, again,
Rick McInnis
Release notes .
Change Log (2.0b9 release):
- DSP refinement: new math routines removing need for precision (and slow) library processing
- Compiler optimisation done
Please REMOVE previous versions before installing cPlay 2.0b9. Normal (SSE2), SSSE3 and SSSE3 B9 versions are available.
For cMP² users, review these BIOS changes (underclocks FSB, CPU & disable additional USB items). Note: E7200 has been replaced with E7300. Same settings apply. Gigabyte mobo may be difficult to find, suggested new mobo is Gigabyte's GA-Q35M-S2 .
RE: > Note: E7200 has been replaced with E7300.
Does this mean the E7300 is the new preferred CPU?
I have both CPUs and want to use the better one for the job.
Thanks,
Chris
Once you become familiar with one of them, try the other and let us know if there is any difference.
I think a test of relative CPU goodness could be how low you can set the voltage and the thing still works. Not that this tells you much about the CPU itself. I was able to get really low voltage with the e5200 but as soon as the e7200 was inserted the sound was much better.
But I have found my e7200 will not allow very low voltages. So, my curiosity is engaged; how low can you take your chips?
Thanks,
Rick McInnis
does this version of cplay affect high frequencies without the double precision capability? 2.09b9 seems not as smooth as 2.09b8.
I just went back to 2.0b8 and, as they say in sports, check that last comment of mine. 2.0b8 is more elevated in the highs than 2.0b9 and so I take back the remark that maybe 8 was better than 9. When I adjusted volume, 9 was better. Sorry I must of had a bad day (moment, hour whatever) when I posted that. 9 is an evolutionary improvement. Sorry master.
:(
Hi, everybody! How many of us are using recommended Zalman HD 160XT case and how many cables did you remove? I personally, beside other, removed dual usb cable and fed the display from the back. Did you, people, remove power led cable? What other cables are safe to remove, if not using touchscreen?
Also a warning: If you use usb stick to transfer wavs from another computer, be careful and let ALL data be transfered, before unplugging the usb stick, otherwise cue files get empty and you'll get an unknown artist and unprocessable error.
Serge.
Release notes .
Change Log (2.0b8 release):
- Minimum processor supported is SSE2 (Pentium 4) which replaces the normal cPlay version (SSE). Processors not supporting SSE2 are not suitable for high quality audio
- DSP refinement
- Compiler optimisation done
Please REMOVE previous versions before installing cPlay 2.0b8. Normal (SSE2), SSSE3 and SSSE3 B9 versions are available. For cMP² users, review these BIOS changes (underclocks FSB, CPU & disable additional USB items).
What a busy young man, I was wrong I guess those bios settings really do work well after all!
:)
Hi Everyone
Today I received the sad news that the highly desirable KHX6400D2UL/512 RAM is NLA. This is the result of a long trail of web search and phone calls - many online vendors still show it as an active item - but when pressed for availability the answer is "end-of-life and discontinued due to low sales" (to quote the Kingston rep). Rats !!
Unless somebody knows of suitable (CL3) memory from other manufacturers, I suppose the "next best" choice is HyperX Low-Latency KHX6400D2LL/512 (rated as: CL4 [4-4-4-12] 2.6V | datasheet here -> http://www.valueram.com/datasheets/KHX6400D2LL_512.pdf).
Best price I've found so far is at http://www.shopblt.com/ (BLT # BL82017).
Does anyone have a better option ?
Thanks for your help,
Grant
That's not a Toy... IT'S A TOOL !!
I have been lurking here for a while and think maybe I should start contributing. I have been using cmp and cplay on my multi-purpose machine for about 2 weeks. All I can say is wow even using a machine decked out for Cad the sound is the best digital I have ever heard. I am very excited about taking the next step and building a purpose built machine based on Cic's recomedations; although sticking to my firewire i/o unit.
In searching for the prescribed HyperX memory I came across this available from NCIX computers in Canada (ncix.com) who also have US distribution.
Mushkin XP PC2-5300 512MB DDR2-667 CL3-3-3-10 240PIN DIMM Memory
Not cheap but I think this ULL stuff is becoming rare
The Mushkin web site also list a lot of other ULL memory available direct but of lesser quality. The memory listed above is no longer available through Muskin.
Anyways I thought this maybe of interest. I'm picking mine up this weekend since I live near one of the NCIX outlets in Vancouver. My new machine should be built by next week; I will post about my progress and I'm sure I have questions.
Thanks for all the info guys and mostly thanks Cic's for sharing.
PC XPpro TC Electronic Konnekt-8 firewire i/o, 2-Tact 2150's, Genesis 500 modified speakers, Virtual Dynamics & XLO cables
Well, it’s a funny thing.. I have 1GB OCZ PC-6400 which, by all reports, is v. good RAM. However, a few weeks ago, just for the hell of it, I thought to try 2x256MB ValueRAM (KVR533D2N4K2/512 PC-4200) which I had hanging around. The thing is, the ValueRAM easily improves upon the OCZ for SQ. And, it has no problem achieving 3-3-3-10 timings due, I suppose, to the cics recommended underclocking. So here I am, scratching my head and wondering why.. (Of course I suspect the reason is the low power consumption of the ValueRAM.) I have tried to get 1x512MB HyperX (as recommended) to compare with the ValueRAM & OCZ but as has been noted elsewhere that stuff is NLA. The 2x256MB ValueRAM is also NLA but it does point to the thought that the HyperX is not necessarily the grail. I say this because I would think the OCZ and HyperX are pretty much of the same family.Bling,
Edits: 11/05/08
low latency shipped by the folks who sell through AMAZON as recommended
in this comment
Posted by kocsu (A) on October 25, 2008 at 15:10:21
In Reply to: RE: HyperX 512 Low Latency Memory is available from ... posted by rickmcinnis@dogwoodfabrics.com on October 25, 2008 at 12:16:43
Hi
check again in Amazon
when I looked for the HyperX recommended by cics the first time I couldn't find the Ultra Low latency in stock, but one or two weeks ago they appeared in amazon.com
http://www.amazon.com/Kingston-HyperX-240-pin-PC2-6400-unbuffered/dp/B000RAH8JO/ref=sr_1_5?ie=UTF8&s=electronics&qid=1224972478&sr=8-5
cheers
O ordered one and was told it was shipped. Who knows?
I paid sixty eight dollars for it.
I paid forty dollars for a pair of 512 "LL"s from TIGER DIRECT.
Hope this is of use.
Bye,
Rick McInnis
PS. The LL I have replaced some low latency CRUCIAL 512's. The CRUCIAL, on AUTO, had lower latency numbers on CPUZ than the LL HYPER-X. What does it mean? I have no idea!!!
Better move fast. OR maybe not so fast ...
I was supposed to receive this by today. Did not show up.
My card has been charged.
I hope it shows up tomorrow.
I will let you know.
Bye,
Rick McInnis
PS It has been listed as discontinued on the KINGSTON site for months
Please proceed with caution ordering from the site I have mentioned.
I will post saying whether the item is received. Until then, this must be considered suspect.
Sorry,
Rick McInnis
I get a call from this RE-STOCKIT and they tell me the package was shipped from KINGSTON and they do not know what has happened.
I wrote them saying if the stuff is not in YOUR warehouse it is certainly NOT in theirs. What a ridiculous outfit. As usual with these re-sellers through AMAZON who knows how long it will take for the credit card to be credited.
I hope I did not spur anyone on to order from these folks.
IN STOCK has become another meaningless phrase in our world of debased language.
Oh, well ...
Rick McInnis
Re: “PS. The LL I have replaced some low latency CRUCIAL 512's. The CRUCIAL, on AUTO, had lower latency numbers on CPUZ than the LL HYPER-X. What does it mean? I have no idea!!!”
Rick,
I wouldn’t worry about what it means.. What is important is how the two compare in terms of SQ, (both set to 3-3-3-10 timing by manual).
Bling,
AB,
I certainly agree that quality of sound is ALL that matters. Nonetheless, it could be helpful if any sort of trend could be detected.
I found the HYPER-X LL (not UL) did have a more natural sound so, in this case, this inherent timing is not THE factor. (forgive my unfamiliarity with proper computer jargon)
I hope I actually receive the UL and more than that, I hope it does have cleaner high frequencies. Not to say I am complaining about what I perceive now.
We will never be satisfied, will we? Especially when cics has some new wrinkle for us to audition every week or so. We are getting VERY spoiled.
Bye,
Rick McInnis
To any interested:
If one is going to use 192 kHz upsampling, which is all I ever use, there is no question that the e7200 is the way to go.
I think my chip is one that requires more voltage than cics's. I guess this is not as much a potential problem now that cics is recommending more than the minimum possible voltage. This chip was manuafactured in Costa Rica. I have another one from, I think, Malaysia.
Before inserting the e7200 I had set CPU voltage to 0.875 thinking the e7200 would start with this. It did not. I had to clear the CMOS and re-set most of the BIOS setting. I set voltage to 1.0000. I will leave it there for a few days before venturing lower. As an aside it is interesting which settings revert and which stay as selected when CMOS is cleared.
BUT, the point of all of this is that the e7200 sounds much more full, robust, complete, effortless ... there is no comparison. I do not remember thinking e5200 diminished the sound when it was installed, but then I had received a working MB only a few days before so I had not the chance to re-aquaint myself. I can see I was more excited about really low voltage n the CPU than the sound. I am embarrassed (again).
So, maybe with no upsampling the e5200 could be worth a try and the slight savings. But for those going for a genuine cics cMP squared system, there is no other way than following the instructions he has given.
Bye,
Rick McInnis
Just checked the market for e7200 CPUs... I'm told it's been replaced by e7300 at the same price. Great. OK... same cost - but what does THIS mean to cMP ?
Intel data sheets look the same, except for one spec...
e7200 Bus/Core ratio: 9.5
e7300 Bus/Core ratio: 10
see -> http://processorfinder.intel.com/details.aspx?sSpec=SLAPC
and -> http://processorfinder.intel.com/details.aspx?sSpec=SLAPB
Always fun in Computerland...
Cheers,
Grant
That's not a Toy... IT'S A TOOL !!
FWIW I have the 7200 at .896 volts and it is stable and works fine.
sounds pretty cool. is the audio engine any good?
BROWSE CONTRIBUTE YOUR MASTERING TIPS AT AUDIO MASTERING TIPS AND TRICKS
I have an ACER that does a good job at being a monitor, but when the brightness is turned down it has become noisy.
Did not initially make this noise.
Is this just the way it is?
It has put me in the habit of turning it off, which can make a listening session interesting with the random selections one gets to make if you become too lazy ( or ...) to stand up and go to the monitor.
Any suggestions would be welcomed.
Thanks,
Rick McInnis
Cics,
I do not use a dedicated cmp pc and i have a lot of music without cue files. So i have to load every song..... :-(
Can you add a little future to load a complete directory recursively?
Any users of RME cards able to confirm? I tried listening for over 30 minutes, and still can't be sure. Either I'm too tired today, or it has no significant impact on SQ.
Release notes .
Change Log (2.0b7 release):
- Minimum processor supported is SSE (Pentium 3). This replaces the normal cPlay version and will work on all processors expect deep legacy models (prior to Pentium 3)
- DSP refinements: more efficient DSP & ASIO processing
Please REMOVE previous versions before installing cPlay 2.0b7. Normal (SSE), SSSE3 and SSSE3 B9 versions are available. For cMP² users, review these BIOS changes (underclocks FSB, CPU & disable additional USB items). Note: Avoid lowest CPU VID (0.85000V for E7200).
My favourite: 2.05 - it reminds me of 1.3 when the sound really started to get balanced, but now with extra clarity. The music flows effortlessly. 2.06 - similar, and reminiscent of 1.4, also excellent, but the smile reflex is stronger with 2.5 for me. 2.03 is clear and smooth with separation verging on isolation, but lacking connecting tissue. Before 2.03 my ears were hurting and I was back with 1.3! 2.07 is going downhill for me - just doesn't sing like 2.05. Can't fault the clarity and separation in any of the 2 series.
I can't at this point say I hear an improvement with newest bios spec - maybe better before, but I'm always on 192/145.
Still sometimes getting the brief judder at the beginning of first track when opening an album.
Have to say I agree with seger re new bios settings. I initially tried 2.0b7 with old bios and liked it better than new bios. (My favorable comments on 2.0b7 were with old bios settings--didn't realize there were new settings until it was mentioned by someone in this thread). 2.0b7 is very very good but probably not optimized yet.
Re: "Avoid lowest CPU VID (0.85000V for E7200)."
cics,
This is a change in strategy, is it not? Previously, the lowest possible stable cpu voltage was thought to be best. Now we are looking for a value above 0.85; one that we determine ourselves according to our own listening experience?
Bling,
According to my personal experience, 0.85V is unstable with cPlay 2.07b and the new recommended BIOS settings (8.5 x 170MHz), sometimes the system would refuse to start after being powered down for a while. While 0.85V is still doable by reverting to 8 x 170, I wouldn't recommend that if you are used to 4x upsampling like I am.
Where did you end up (i.e., stable voltage)? I'm at .896 which seems stable but may not be sonically optimum.Additionally when I tried to input 8.5 the bios display showed 8. Is this normal or said another way how can I verify 8.5?
Edits: 11/03/08
.
/
Cics,
Thank you for your work on Cplay!!
SQ is amazing!
Yep did it again! At once more detail, less bright, more midrange and body. It just sounds more right, better timbre, more dynamics. Its the same signature with every improvement (except for maybe the last where the splitting of long tracks was improved, but
yes it was a little bright). Not this time, no brightness here. Just more info especially in the midrange. My compliments to our brilliant innovator of music from computers-- Mr. Cics. Listening to Clap Hands Here Come Charlie by Ella Fitzgerald and it is wonderful. The classical pieces especially come out better with more body to instruments especially the woodwinds (which are so difficult to capture). 2.0b7 does it better than I ever heard.
Release notes .
Change Log (2.0b6 release):
- Fix track sequencing when splits occur
- Reduce chance for track splits when wav or flac file is played directly
- Replace 'Track 01' with file name when playing wav or flac directly
- Compiler optimisation
Please REMOVE previous versions before installing cPlay 2.0b6. Normal, SSSE3 and SSSE3 B9 versions are available. For cMP² users, review these BIOS changes (underclocks FSB, CPU & disable additional USB items).
Simply beautiful. Not sure which attributed more to the magic, the upgrade to 2.0b6 or finally switching to Vista, but it doesn't really matter. I've never been more satisfied with my system. I find it hard to express my gratitude with my limited vocabulary :)
Hi cics
Thank you again for adding hi-resolution file playback to the cPlay program. I just love the fact that it will tell me the truth about any project file I have on the HD. Today I discovered it does not like MONO WAV files:
The example here is a single-channel WAV file, as the property sheet informs us:
Q1: Would it be possible to have cPlay handle mono input files - sending the same data to both L & R output channels ?
The second thing I noticed (when playing a WAV or FLAC file) is the Trackname display does not give the filename, but simply "Track 01" as below:
Q2: Would it be difficult to have the actual FILENAME display in the Trackname column ? In this example I'd like to see "PZM_Room_02.wav" indicated - including the file extension. Hmmmmm... it's interesting to note the cPlay titlebar display changes the lowercase "wav" file extension (as actual filename is stored on the HD) to uppercase "WAV". Not a problem, but an observation.
Upsampling Rules in cPlay:
I'm curious about the SRC upsampling in the case of non-redbook files.
Q3: If I set Upsampler Rate for 96.0 and input a 96KHz file, is the SRC code bypassed - a direct passthru ?
Q4: If I set the Upsampler Rate for 44.1 and input the same 96KHz file, does cPaly DOWNsample - and if so, is it of the same high quality as in the upsample direction ?
I presume the same 96KHz file will be bumped up to 192KHz when I select 192.0.
Sorry about what must seem like pointless inquiry (to most of the cMP usergroup...), but knowing this stuff helps me understand what I'm actually hearing, when auditioning non-redbook music.
Thanks again for your commitment to Ultimate Quality digital playback.
Cheers,
Grant
That's not a Toy... IT'S A TOOL !!
Q1: Would it be possible to have cPlay handle mono input files - sending the same data to both L & R output channels ?
It certainly could send mono to left & right. Not a priority for now though.
Q2: Would it be difficult to have the actual FILENAME display in the Trackname column ? In this example I'd like to see "PZM_Room_02.wav" indicated - including the file extension. Hmmmmm... it's interesting to note the cPlay titlebar display changes the lowercase "wav" file extension (as actual filename is stored on the HD) to uppercase "WAV". Not a problem, but an observation.
Done in 2.0b6.
Q3: If I set Upsampler Rate for 96.0 and input a 96KHz file, is the SRC code bypassed - a direct passthru ?
SRC is completely bypassed whenever input rate = output rate.
Q4: If I set the Upsampler Rate for 44.1 and input the same 96KHz file, does cPaly DOWNsample - and if so, is it of the same high quality as in the upsample direction ?
96-> 44.1 SRC downsampling is done using same high quality converter.
That's not a Toy... IT'S A TOOL !!
Hi cics - Let me begin with another compliment on how the updated software makes music - my DAC1 never had it so good ! Once again I was rewarded with new detail & musical enjoyment (cMP 1.2 + cPlay 2.0b5). I don't know HOW you do it, but Thank You,
Thank You, Thank You !I noticed something peculiar with cPlay 2.0b5 - when loading up tracks list from the CUE sheet - the song list gets scrambled (...maybe it happend in beta 4 too).
Here's the CUE, showing five tracks within the source file:
...and here's what cPlay displays:
Notice there are five unique track numbers in column one, but they do not increment correctly. When it's playing, the sound is just glorious, but I couldn't help double-clicking on a blank entry - I experienced a program wait of about 75 seconds and then it played music again (I guess it had to do a lengthy RAM reload). BTW, the source file is 16/44.1, encoded in FLAC. Please PM me for a d/l link, should you want to test this with the actual files.I don't have an exhaustive list, but it happens with more than one album... and NOT with every album ! Here is another troublemaker:
These screen captures were done without the cMP program running - just pure cPlay. It behaves the same way under cMP too. Do the "FLAGS DCP" or "Index 00" lines cause a problem ?OK - enough mystery and evidence for now. Thanks for all the incredible work you do on this project !!
Edit UPDATE -> Additional Thoughts:
Some of the "inconsistent" track splitting I observe is expected - that's the dynamic RAM management working ! Depending on what other apps are running in the system (and therefore memory available to cPlay), the same long-ish track will be split into varying lengths. As riboge observes, the numerical track numbers (column one) should still reflect the song/track number as found in the CUE sheet, while the segments get alphabetical designations (and perhaps "b", "c", "d" etc. get column one blanked out).ALSO: NAVIGATION using " < <" and "> > " Buttons:
This may be by design or necessity, but in the 3-part "Mindrevolutions" example, when playing segment #2, repeated clicking of the " < <" button does not return you to segment #1, but rather wraps around to the end of the song - in this case the end of segment #3. I suppose this is just a constraint of the "load next part in the background" strategy.Thanks again...
GrantThat's not a Toy... IT'S A TOOL !!
Edits: 10/28/08
cics,I have cMP 1.2 + cPlay 2.0b5 B9 installed.
I would like to suggest that the cPlay display be changed to conceal track splitting. Example: in gjwAudio’s post above, the “Mindrevolutions” screen shot has track 05 split into 3 parts (each of 08:33). I would prefer, however, cPlay to conceal its “internal” splitting and manipulations and just display track 05 in the normal way. In other words, why show the splitting of track 05 into 3 parts/3 lines as it is “information” that does not assist the user in any practical way? IMO, concealing the splitting would make for a “cleaner” display. Additionally, a “problem” emerges say 10mins into track 05 when cPlay shows it is playing part 2 (length=08.33) but the “time counter” at the bottom left corner of the display will show 10.00. A consequence is that if the user wants to work out how much time remains for track 05 then he/she has to calculate 3x08:33-10:00. This "problem" would not arise if the track splitting were concealed.
Anyway, once again, thank you for your work.
Bling,
Edits: 10/29/08
I like this idea. The only problem that has occurred to me is with jumping(fast) forward and back. If a track is divided into three parts and the first is started, one is be able to forward or back in the first one only. Listing all simply as one track would mislead about this. Cics will have to say whether this can be overcome. If it can, I agree with you.
Column 1 is intended to reflect original track no. It's a minor fix and won't affect playback. I want to wait a while before releasing 2.0b6 (any more fixes???).
Yes, you may need to wait if there's not enough resources for cPlay to fully process next track (especially FLAC). You should increase CPU clock speed to reduce this impact. Background processing in cPlay is set for very low priority (to prevent SQ interference) so any other things happening on computer will delay this.
Skip for/back is designed that way. It could be changed to continue to next/prev track...
cics, is it possible to add a button in the UI to disable resampling and just play the file at the default sampling frequency?
I use a slow notebook and I set up 44.1 by default with no resampling. But when I open 88.2 or 96kHz music files, they are resampled to 44.1 when playing. That is not good.
It would be great if I select no resampling and don't need to worry about this setting.
Thanks!
Set diagnostics to Yes. When you hit Stop, diagnostics indicates whether SRC upsampling was used for each track.
What you suggesting will cause soundcard to change output rate to input rate - this is unusual.
I have not observed the scrambling you report, having opened many cuesheets in cMP2. However, I have suggested below under RE: cPlay 2.0b5 Released that the numbering of tracks should not increment for the portions of a split track. In the screen shot you show last the 06 track, while it should be 08 in cics current method of numbering, is really correctly number and the ones that follow are not.
Hi, everybody. I found these two 512mb sticks of Kingston in my old computer, installed and CPU-Z gave such results. The question is does it matter if it is DDR not DDR2. Note the latency!
You put this on the GA-G31M-S2L mobo? I thought they were different sizes.
Edits: 10/27/08
on an old computer. Hello! I brought it home from work. Used to be an unbeatable machine for graphics in its time. Microstar mb, REAL matrox millennium fanless video (try to find one), all stuff... I work in publishing and design and have many "used to be great" computers around. Some are alive. Some in spare parts but complete. Many have very good memory for their time. This one is one of the freshest, - cpu p4, northwood with core voltage 1,5, bus speed 133, core speed 3100 mHz, You have read about memory. I won't say that its hdd, that migrated to my current machine, is the fastest in the world SCSCI 320 cheetah 15k from Seagate. Just was curious if this memory will fit a spare biostar mb m900m7se, that I have freed from my accountants' machine. It was recommended as the second-best mb for cmp, and I want to share it with my non-believer friend. The problem is biostar doesn't specify if ddr memory can be used instead of ddr2, and I am not a tech professional.
Serge.
Unfortunately, I don't think DDR is going to be compatible with a DDR2 motherboard, unless it has legacy slots for DDR. I remember those low latencies, but I think the DDR2 boards are faster and more efficient, and support these excellent 45nm cpus, so better for our purposes.
Unlike most other components, the cMP^2 doesn't yet have a "definitive" choice for the main PSU. Quoted directly from cics' official cMP guide: "High quality PSU for CPU and mobo. MGE’s XG Magnum 500 is best choice offering no fan interference but they seem to have supply issues. Other alternatives? Corsair, CoolerMaster. Choose one with 700W power – multiple 12V rails offered with lower ripple noise. Also, ensure cable management is provided – allows removal of unused power cables which helps cooling and reduces RF pickup (antennae effects)."
I believe that cics recommends the XG Magnum mostly due to fan interference (or lack thereof). Assuming that fan interference can be avoided by turning it off (a friend accomplished that without a PSU meltdown), we should be able to achieve better results with cics' alternative recommendations (or other low-ripple PSUs).
For those interested in helping the community in finding the "best" PSU, please share the PSUs that you have been using, the sonics of each, and the top candidate(s) you have in mind.
I currently use the external version of the XG Magnum PSU from cics' recommendation. Can't really complain about the sonics, especially after the latest release of cMP^2. But according to several internet and real life sources, ripple tests say that the Magnum PSU isn't as impressive as the old favorite Antec EW430. The top candidate for my next PSU is the Corsair HX520. The PSU has extraordinary ripple test results, modular cables that can be easily detached, and high quality MIJ capacitors. I will probably replace my XG Magnum with the Corsair HX520 next week, and I'll give you guys a brief report when I do.
Long awaited thermaltake dual power 520W and working fine fanless. No problems. Sound is fine
Edits: 10/27/08
I'd love to test out thermaltake fanless PSU, but the low availability makes it almost impossible. I've checked with a few local dealers, and there's not one that has it stocked or is willing to order one. Oh well.
That's the PSU I'm using too - also for me the sound is better than with any other PSU I tried. In fact it was your mention of it ages ago in this forum that made me try it, so thanks, overdue! Also the price was half of the other fanless PSUs, but maybe that was a lucky fluke. But I wonder how you got it - Thermaltake has stopped making it. I'll stay away from the Zen as reports are mixed so far, and it's worrying that the XG Magnum fan can come on under light load (for some). At this point I would get the Silverstone Nightjar if I needed to - I haven't heard any negative comment on it and it's been around a long time. But the Thermaltake is doing the job for me now.
Have fun fitting that Ninja. I'm using it on my second system - Biostar with E2140 is running in 20s C with cplay 192k at 121SNR. But the mounting setup is not good. I have the Thermalright HR-01 Plus on main system and it mounts better and cools at least as well - 30s at 145SNR.
Hello, again! I came across this PSU when surfing through Thermaltake.com, after search on our Russian rambler.ru, looking for silent pc parts. At that time they (PSUs) were announced by our computer parts vendors but unavailible. I placed an order and after 4! months I got one. Now they are availible but in very small quantities here in Moscow as, I think a small batch was imported and distributed among local retailers. Their usual comment is: "Yeah, it's good, but sooo expensive..."
So I am one of those "strange" old men (I am 45) who doesn't know what to buy, and let him do it... Approximately the same story with Zalman case. So it goes.
Serge.
P.S.It is a Ninja 2 and they promise to have improved the brackets greately. Have You by chance heard of Scythe thermal grease that contains absolutely no metal particles? The sometimes sell it on Yahoo Japan.
Edits: 10/28/08
The Scythe T.I.M. sounds interesting, but could find no good reviews of it. I did a bit of web-research into thermal materials before my last build and found that Arctic Silver 5 is still as good as any of them, except in special circumstances (like extreme gaming), and it's easy to put on and off, so I'll stay with it. I would try Arctic Cooling Mx-2, which got superb test results, but as it's so thin, needs perfectly flat surfaces - must do some lapping first; it is also supposed to be the easiest to put on and off.
I had no problem fitting the Ninja to either Biostar or Gigabyte Mobo.
I used the standard grease
cheers
Martin
It's about conducting heat away, and the spring-loaded bolts that Thermalright make give a more even, snug contact, and can knock a few degrees off CPU temps - being easier to fit is just a bonus.
I have the CORSAIR VX-450 in my cMP2 machine and a HX-620 in my HTPC. Both are excellent PSUs in terms of quietness and finish. My one gripe with the VX-450 is the lack of cable management. Given that I am using granite power supplies for HDDs, USB etc. I have a large rolled up “mess” of redundant power cables pushed to one side of the case. The HX series, however, does have cable management. A must, I think. In terms of sound, I have never had the opportunity to AB the CORSAIRs against an alternative in the same set-up.
Bling,
Could not agree more with the importance of cable management, and one of the main reasons I'm replacing my current PSU. A person I know "managed" redundant cables by literally cutting them off, I guess that's one way to fix it, but not effective or safe in my opinion.
If ripple voltage matters, and it sure seems to with any other audio component, this supply I am sure has many good features but when it comes to low ripple it is out-classed by the humble EX-430.
As an aside:
If anyone decides to buy this and take advantage of the "rebate" beware, CORSAIR, in my one experience, is completely unscrupulous in their "rebate" program. Be happy with the price you pay and understand you will not get the rebate. You will not even be allowed to apply for the rebate. You will be told that the all-important form will be emailed to you but it never shows up. Caveat Emptor! I have complained to NEWEGG but with no success. I noticed that other customers have had the same problem from reading the reviews on the site.
I am going to install a power line filter into the EA-430 and bypass the IEC connector. The filter is the simple but effective Jon Risch filter. At the same time I am going to disconnect the fan and see what happens.
I suspect nothing will happen other than a smidgen more silence.
Have you personally tested the ripple of both PSUs? I own neither of those PSUs, but the reviews from spcr say that the two come very close in terms of ripple. I'll still probably go with Corsair HX520 until someone can prove otherwise, as the PSU offers good quality capacitors, easy cable management, and superior appearance. Most important of all, it can probably be connected externally using parts of my current XG Magnum External PSU.
No, I have not measured my PS and certainly there is no way I have measured this one that has been mentioned for the first time.
As my previous posts have made clear there is the possibility that the supply tested by SILENT PC was a ringer and not representative of what is actually in my possession. This, of course, is a given.
But, if tests are to be referred to, and the fact that both units were tested by the same publication and that one has a ripple voltage figure of one/tenth the other, most would have a hard time saying these were close in performance.
It is meaningless to me to beg that you buy one supply over another. It was simply my intention to point out that they do not measure the same when it comes to ripple. Maybe there are PS's with quieter fans and those that are more efficient, if that is your criteria then that is what matters most.
I believe, though I have not tried it yet, that the fan in the ANTEC can be safely unplugged considering the light duty required, and this removes any concern about noise. I cannot hear the fan from my listening seat, but every little bit of improvement adds up eventually.
Wish I knew someone who could test my example of the ANTEC to see how close it comes to SILENT PC's test results. I will look around.
Bye,
Rick McInnis
I use the Corsair HX520 and I am fully satisfied with it. There may be better choices, I didn't test other ones. I bought it at local store with my casing. It had a good review on SPCR http://www.silentpcreview.com/article692-page1.html
I left the fan on, and I can't hear it from my listening position. As for electrical noise, I don't know...
I use it only for the MOBO+Juli@. HDD and USB use alternate power sources. I also added caps to the soundcard.
cheers
Martin
Thanks for sharing! Corsair HX520 just seemed more promising to me. I'm pretty sure the fan noise is barely audible, even from a very close distance. However, the vibration and electrical interference of the fan will affect SQ negatively. There should be a noticable improvement in sonics by turning the fans off. Don't be too concerned about safety, as I turned off the PSU (Enermax Modu 625) fan on my gaming PC yesterday to confirm my friend's claim. So far, no problems have occured after almost 24 hours of (low load) operation. Nothing bad should happen, as long as you don't do anything too reckless (like playing Crysis)!
nt.
I have attempted AGAIN to implement the recommended powering of the infrared mouse with the GRANITE power supply.
I was never able to get it to work with the BIOSTAR board. Having good luck with the GIGABYTE board (finally) I though today was a good day to try again.
I have been using the recommended type of cable, using the inside the board USB headers, but letting the board power the thing. Today, I used the cable I had modified to use the remote power supply, hook it up to the same jacks and when I restart the machine I no longer have a mouse and the machine is telling me there is an unrecognized USB device.
Who has made this modification successfully and would you mind telling me if it went easily or if there was a trick to getting it to work?
I would greatly appreciate the help
Check the gigabyte mobo manual to make sure which pins on the mobo header are the right ones for the extra usb backplate port cable - the plug on my cable had the blank in the wrong corner, and when I moved it, I could get them in the right way. White farthest from backplate, green next, if you know what I mean.
Dear segars,
As always, thanks for your willingness to help.
As I said before I know the "jack" is in the header correctly. I use the same MB header when initially setting up the mouse with a standard "MB powered" cable.
I guess I just want to hear that someone ELSE has got it to work.
My machine is a dedicated cMP machine. All services are stopped, all of the instructions, per cics, have been followed.
I would have thought if the only thing to change was the powering of the cable from MOTHERBOARD to GRANITE outboard supply, it would be transparent to the machine. But, for me this is not the case!
Another question:
there are two ground wires; the true power supply ground and then that other one at the "end" of the jack/header. Do you disconnect this entirely, do you leave it in place, OR did you connect it to the "main" ground wire, also disconnected from the jack but connected to the GRANITE supply ground? I have tried all of these and there is no difference.
Just for clarification:
do you only have the green and white wires connected to the MB?
THANKS, again,
Rick McInnis
It's working for me and I took out both blacks and connected to molex connector that goes to GD, as well as reds, so just green and white connected to mobo. Computer should "see" the mouse straight away, unless there's a problem with the mouse or receiver.
I was given a mouse (Logitech) that has a "free-wheeling" scroll wheel that is great for fast scrolling through albums in cMP - heartily recommended.
Thanks,
That is what I needed to hear.
I am using a separate supply for this, the HDD has its own; I can't imagine this making any difference (?).
I am using the mouse recommended by cics, I think it is marketed as a laptop mouse, it is small and since it was what he designed the system around, it works perfectly. I assume we are using the same device. It was expensive but worth the extra money.
Just installed the new cMP and must go give it a listen.
Thanks, again,
Rick McInnis
If you keep having trouble, using a PS/2 interface wireless mouse may be a feasible alternative. That's what I did for my CAS, due to the low availability of the components in my area. Not sure how this version compares to cics' recommended setup, but I did notice an audible improvement after the switch.
something called TIGER DIRECT for forty dollars for a pair.
One must buy two, so a good excuse to talk a friend into a cMP Squared system!!!
This stuff seems to be all but unavailable everywhere else.
So for a proper cics cMP system, even though I think he is lucky enough to have ULTRA Low Latency modules, this might be your last chance. Until something better comes along, but from what I am seeing the trend seems to be going away from low latency.
Hi
check again in Amazon
when I looked for the HyperX recommended by cics the first time I couldn't find the Ultra Low latency in stock, but one or two weeks ago they appeared in amazon.com
http://www.amazon.com/Kingston-HyperX-240-pin-PC2-6400-unbuffered/dp/B000RAH8JO/ref=sr_1_5?ie=UTF8&s=electronics&qid=1224972478&sr=8-5
cheers
Hi RickThanks for the tip on HyperX L-L memory (do you know the precise Kingston part number). I'm a little confused about the "best" hardware configuration for a cMP system (...obvious I'm still in the planning stages, eh ?)
Is it true that the greatest results come from using only ONE memory module in the recommended GA-G31M-S2L mobo ? Somewhere in all these posts I thought I read 2 sticks - and dual-channel mode - was preferred.
To be sure, with v2.0b5 of cPlay, there's no need to go beyond 1GB of RAM... so maybe this suggests getting a 2GB HyperX kit (and holding one stick for a friend or rainy day). Please advise...
Cheers,
GrantThat's not a Toy... IT'S A TOOL !!
Edits: 10/25/08
From 3 March 2008:
Kingston HyperX is an excellent choice - I would recommend KHX6400D2UL/1G as its designed for lower CAS latency (this is most difficult to achieve and requires better quality parts). I use this in my reference transport.
--------------------------------------------------------------------------
From 9 June 2008:
Rather go with the recommended RAM in cMP's FAQ:
Kingston HyperX 1GB: TWIN PACK (2X512MB): KHX6400D2LLK2
-------------------------------------------------------------------
Link to AMAZON page for 512mB version of the March recommendation:
http://www.amazon.com/Kingston-HyperX-240-pin-PC2-6400-unbuffered/dp/B000RAH8JO/ref=sr_1_29?ie=UTF8&s=electronics&qid=1224975538&sr=1-29
$68.00 each piece - this is the Ultra Low Latency version which is no longer being made per KINGSTON's website.
The seller at AMAZON must have every memory KINGSTON has ever made, it goes on and on.
I got a pair of the Low Latency at the place I mentioned before for forty dollars. There is an immense difference in price!
I remember cics saying the Ultra Low was not necessary, but ... one always wonders if it gets you a little closer.
Thanks Rick for digging out the pertinent posts - and for pointing out the distinction between the LL and UL modules.
I'll look around and see what Canadian sellers have to offer - it's so much less aggravation if trans-border shipping is avoided. As to the 1GB vs 512MB choice, I suppose playing larger, high-resolution files (24/96) will require less background RAM loading with 1GB in place.
But the 512 uses less power, and therefore is potentially a "better" sounding setup... Hmmmmm - choices, choices.
Thanks again Rick for the clarification.
Cheers,
Grant
That's not a Toy... IT'S A TOOL !!
...just want to say, that I used first the LL version of the hyperX-ram. Some time later I could get hold of the ULL-version and noticed more detail and delicacy in the high frequencies. I would not like to miss this, since I know the effect. Would be interesting to know the difference between the 1G and two 500mb performance. I am using now the Gigabyte-board and the Antec-PSU. I am extremly happy with the new system and enjoy the progress of the cMP-player. Walter
Dear Walter,
Thanks for your report.
Luckily I am only using 512 mB!
cics does not recommend dual channel.
Bye,
Rick McInnis
Grant,
It would take cics an inordinate amount of time to incorporate his latest thinking into the basic cPLAY and cMP posts. He has made a few big edits (as you can see. I find checking into the forum every day to be worthwhile.
I have noticed every time he thinks that two channels might be better, he always comes back to one. As soon as cPLAY 2 appeared I counted the hours until he recommended 512mB memory and sure enough he did. He has always tried to minimize the amount of memory needed.
He is obsessed (good for us) with using as little power as possible, not from some "green" rationale but because this is what makes for the quietest environment for music to be produced from a computer.
I have found there is a more natural sound when using only one channel of RAM and I think I am sure (!) that 512 sounds better than 1gB.
I guess if you need 1gB you should use that. All of my recordings are RED BOOK so I see no need.
I am going to AMAZON right now to what if Ultra Low Latency is still available. Just have to hear for myself if there is a difference.
Release notes .
Change Log (2.0b5 release):
- Track splits performed when available RAM is low. Minimum available RAM needed is 256MB and maximum supported is 1GB. Lower available RAM translates into potentially more track splits causing background loads to occur well within a track. Ideally, this interference should be avoided.
- DSP refinements: more efficient background loading and ASIO processing
Please REMOVE previous versions before installing cPlay 2.0b5. Normal, SSSE3 and SSSE3 B9 versions are available. For cMP² users, review these BIOS changes (underclocks FSB, CPU & disable additional USB items).
A clear improvement but I do have one quibble with the numbering of tracks when automatic splitting occurs. Say track 1 is split into (1), (2) and (3). The next track is then labeled as 4, that is, its number in the newly split set of tracks rather than the original tracknumber in the file. In the recorded tracks it is track 2, which is how I think it should be listed in cMP. Ideally, the parts of a track would be designated a,b,c,... keeping numbers on the left for tracknumbers.
Works nice so far. good sound.
The "auto track splitter" works like a charm.
I applied all the new BIOS settings, and my RAM shows 3-3-3-7 at 170, FSB 1:1, in CPUZ is this normal ?
Thanks.
Martin
Martin,
Re: "[..] my RAM shows 3-3-3-7 at 170, FSB 1:1, in CPUZ is this normal ?"
It depends on your RAM. My 1GB OCZ PC6400 RAM likewise goes to 3-3-3-7 with AUTO SPD. So I set to manual to achieve the desired settings. I also agree with carcass93 (below) in that I prefer 3-3-3-10 which seems to sound more "natural" to me. (BTW I dont use juli@ card and run at 24/96.)
Bling,
I'll try it.
Thanks for the tip
Martin
.
The automatic division of tracks that are overly large works just fine. Interestingly, some tracks that were parts of 24/96 symphonic movements that I divided for use with the last version were divided further automatically. That is, these large part-movement tracks worked already but were divided even so. I guess this has to do with the 1Gb RAM loading limitation in this version; whereas I have more than 1Gb RAM and otherwise could load larger tracks.
With this version and the new smaller font version of cMP, we(you) really have a special playing system sounding superb and fully functional.
Hello, cics and everybody! I know I risk being stoned to death, but to me 2-2 to 2-4 didn't differ much soundwise. Also, they seemed harsh and grainy to me (compared to 1-17, 1-19). This version, 2-5, is a different beast absolutely. It renders music very well, adding depth to the stage and separating musicians. Better bass and highs. Wonderfully dimensional and multilayered. Much more "intelligible" I think is the best way to describe this version. Also the sounds make more sense now. Thank you very much, sir! It is a pleasure to use this player and every time installing new version I feel a thrill of good expectation. A real step forward for me this time.
Serge.
I got one of these today and installed it.
Of course, I have no idea what to expect
Within MIT in BIOS I no longer have a choice of 2.5 for a MEMORY MULTIPLIER, the closest I am offered is 2.66.
When set to this CPUZ shows 4/4/4/10, so I manually set to 3/3/3/8.
I was able to quickly bring the voltage down to 1.000.
I notice a strange hic-cup when an album is first requested. When I ask the machine to start over, there is no hic-cup. Maybe has to do with my forced memory setting.
I do think this CPU gives an even bigger holographic effect. While not being allowed identical set-up, it is hard to be sure. I was not intending to listen very long tonight yet I could not break away. Of course I had installed the latest cPLAY.
Nonetheless, I do think there is some additional goodness here, and for those coming into this now, a savings of twenty or so dollars.
Why would this CPU have different memory settings? Is it due to the smaller cache? Anyone know?
Bye,
Rick McInnis
I was confused before in my report.
Sounds quite good.
Now if I can remotely power the mouse ...
Never had to re-set BIOS, which makes me think this chip can go lower still.
I was wondering if because this chip uses less power because of the smaller L2 cache, would this mean it could benefit from, yet, lower voltages?
Just wondering if it is worth the trouble.
Should keep going lower until you either have to re-set BIOS or until the function is compromised? Of course, if sound quality got worse it would be an easy indicator.
I am sorry to be so non-intrepid. And such a wonderer ...
Rick McInnis
But one click below and cPLAY will not run. It will boot just fine but gets two seconds into the track and stops.
At .85 it is working very well.
(I think I got a really good board and, I suspect, a good version of the e5200. Absolutely no surprises.)
At this setting I get the same 0.832 in CPUZ that cics reported but I believe I remember he was using 0.875 in BIOS. Another discontinuity I cannot explain.
Letting version 2.5 run all day and will see how all sounds tonight.
As asked before:
Does anyone know why the CPU would affect my choices of memory multiplier?
As said before the lowest possible setting I am offered is 2.66. Does this have something to do with the L2 CACHE size?
Has anyone else noted that when a track is requested, especially by using the right click of the mouse, that the sound is initially a little shaky, yet, if I hit the button below the "wheel" to get it to start over again, it runs smoothly. I am not complaining because this does not bother me at all, akin to how a CD player seemed to sound better if placed in pause before the playing of the track. Is this something peculiar to my installation.
Thanks to all participants on this forum and especially thanks to cics. What will you come up with next? Do you think further refinement is possible?
Rick McInnis
wow .85 volts that must be a record! I used to get a bit of a hiccup at the beginning of a file but this new version is much better. Only got a hiccup once so far. You may be breaking new ground with the 5200. don't know about multiplier other than with my 720 I get 2.00 no problem.
I'm @ .896 volts, does .875 volts sound appreciably better IYO?
I also did some research on the E5200 on the Web last night. At the native clock speed, it consumes 2 to 3 watts less than the E7200 at 100% load. Yes, the general view is that it has lesser cache than the E7200 and so consume less power.
However, the reduction is less than expected and could as well be due to the lower FSB bus speed and the reduction of dynamic current in a partially disabled cache. That is, the E5200 is a crippled E7200 !
Whether the E5200 would go to a lower voltage could perhaps depend on the luck you have or the health of the sample of E5200, I think it could be worth a try and in general most processors will show further reduction of power.
For me, I did not "upgrade" to the E5200 from a E7200 as the reduction in power is a bit small to go through the trouble.
It's been my experience with Kingston HyperX 1Gb, that best sounding set of timings is 3-3-3-10. 9 sounds slightly worse, and 8 noticeable worse than 10.
I listened last night with this setting and certianly heard nothing objectionable.
Also, I am three clicks away from 0.875 CPU voltage.
Will listen to this awhile before forcing 3/3/3/10.
Thanks, again,
Rick McInnis
carcass,
What does the 93 signify? I hope you forgive my abbreviating your moniker.
I have not experimented at all with this, I took these values from a since edited post from cics.
I do not have the LOW LATENCY 512 kB memory, mine is the standard version of HYPERX. I installed a low latency 512kB from CRUCIAL and got better AUTO numbers than I got with the HYPERX, but still much higher than the numbers I got with the same HYPERX and the e7200 in AUTO, which gave me the expected 3/3/3/8 as per cics's post.
Does this seem weird to you? Does this have something to do with the smaller L2 cache?
I will try with letting it run in AUTO and then try the higher MANUAL setting you suggest and see if I hear the difference.
Slowly lowering the CPU voltage.
Thanks very much for your help and suggestions,
Rick McInnis
Release notes .
Change Log (2.0b4 release):
- Work-around to a Windows Kernel bug that occurs when using AWE under some conditions (AWE failure is reported in diagnostics)
- DSP refinements: reduce CPU load to accommodate gapless processing
- Gapless / seamless playback. At start of eack track, next track is loaded (in background). CPU load could increase to 100% (when upsampling and/or processing flac)
- UI refinements: remove Phase 0/180 message delay
- FLAC PCM 32 fix
Please REMOVE previous versions before installing cPlay 2.0b4. Normal, SSSE3 and SSSE3 B9 versions are available. For cMP² users, review these BIOS changes (underclocks FSB, CPU & disable additional USB items).
cPlay 2.0b4 uses 2x more RAM than previous 2.x releases for gapless playback (if there's only 1 track, RAM needed is reduced by half but still 2x more than 1.x series). Setups with 512MB RAM provides ~433MB available RAM. This is good for 16/44.1 tracks up to 10 minutes. For HiRes material (24/96) tracks up to ~5 minutes is supported with same available RAM (433MB). Adding more RAM increases track times, e.g. 1GB would support 16/44.1 tracks to ~20 minutes.
You can reduce RAM requirements by splitting long tracks in cue sheet. A 28 minute track could be split into 3 parts as follows:
FILE "C:\Music\06 Track06.wav" WAVE
TRACK 01 AUDIO
TITLE "Track 01 - part 1"
INDEX 01 00:00:00
TRACK 01 AUDIO
TITLE "Track 01 - part 2"
INDEX 01 10:00:00
TRACK 01 AUDIO
TITLE "Track 01 - part 3"
INDEX 01 20:00:00
Parts 1 & 2 is 10 minutes long with part 3 at 8 minutes (remainder). In next beta release, cPlay would do this internally! That is, split large tracks (per avail RAM) into equal parts.
I ran into the ram limitations on several of my long movement classical pieces. I will now try the cue sheet work around (seems simple enough). While playing lots of music on this latest cplay version I notice a hiccup type anomaly at the beginning of some files. Not annoying enough to complain about but if it could be eliminated that would improve performance. Is there any way of preserving 2.0b4 sonics but with ram loading to avoid the above issues. The long ram load time never bothered me but I don't want to go back to pre double precision sonic versions of cplay. Look if I am the only complainer I can live with these issues. I'm just saying nirvana for me is 2.04b sonics and ram loading (even with initial delays of 15 seconds or so).
BTW sonics do keep getting better and version 2.0b4 is no exception. Bravo!
...it doesn't start and cplay diagnostics says not enough ram. Is cplay trying to add both files (disc 1 and 2) together before playing? Disc 1 plays ok but disc 2 does not and there is only 41 minutes of music.
Unless I misunderstand you, I believe you are getting what cics described, i.e., that with 1Gb RAM you can only do up to 10min redbook files since, yes, files 1 and 2 would be loaded together at the start. If these are whole disc 20+min files, then they should not work without more RAM, if even than. This version clearly has in mind cuesheets listing each track's file separately.
how can I trick cplay into thinking disc2 is an independent disc? I tried varying the name, disc # ...nothing I tried worked.
I finally got it. I had a typo in my revised cue sheet. Thanks Riboge for hanging in there with me.
Thanks for this. XMAS has come early!
Uncompromised quality AND convenience, WOW !
Can't wait to hear the next version ! (the long songs auto-cutter...)
Cheers !
Martin
I had trouble tearing myself away from listening to express my appreciation. DSOM with track transitions undetectable with eyes shut. Even high rez symphonic movements start immediately. The sq seems improved a bit as well, though it's hard to be sure since it sounded so good already.
(The whole cMP> cPlay system would be even more perfect, for me at least, if you put out a small text version of cMP so I could see both artist and full album title on my monitor (running XP pro partition with minlogon and most recommended tweaks), or updated cMP with a small text option built in.)
Indeed, I hear an improvement in SQ (over 2.0b3) as well.
I think I now know what jitter sounds like. (Or what a reduction in jitter sounds like).
In each release of cPlay I hear less grit/grain, less smearing, and overall improvement of transparency and three-dimensionality.
2.0b4 also improves upon the "liquid" factor.
Keep up the good work!
It is an interesting phenomenon you describe reported by many: that jitter is heard mostly in retrospect(retrohear). One doesn't, it seems, know how good it can get or how bad it was until the difference is heard. I have experienced this thru several steps in system improvement.
Can't wait to try it out!
Dear cics and any others who know the answers,
While setting up a working MB (after four defectives) I noticed that many of the optimizations had mysteriously been negated which required going through and checking.
Which brings me to a few questions that might be helpful to those setting up a cMP squared system for the first time.
1. I assumed that within the Control Panel/System/Hardware/Device Manager/View/Hidden Devices that this should include the ones we are instructed to disable in THE ART OF COMPUTER TRANSPORTS. I have disabled these but wondered if this is still correct and wondered if others might be wondering the same thing.
2. When implementing carcass's REGEDIT change I was confused that there is no HKEY/SYSTEM though there in HKEY/LOCAL MACHINE/SYSTEM. Maybe this has been noted before and I missed it. Am I assuming correctly this is one and the same thing?
3. I noticed that this board has taken a few times to finally keep BIOS settings. As I have slowly lowered the CPU voltage I would notice that some things would return to standard settings. Seems to have stabilized at this point. Lowest CPU voltage I can get to is 0.936 at 0.925 it is reset BIOS time and begin again. That has happened only once and it is interesting what reverted to "stock" and what remained at cics's recommended setting. cics, how long did it take running in for your board for it o accept 0.875?
Also, I am as glad as anyone to hear that you will be able to offer this level of performance without the delay. At louder volumes it can be disconcerting, especially in some cases it seems to hic-cup. Not, that I would not live with this if this is the way it has to be. I am very pleased with the sound.
I am speculating how you can do this; will you use two sticks of 512mb and load them separately for each track? Is such a thing as that even possible?
I listened to music last night for five hours. Made up for much lost time and got familiar with version 2.
Just installed a PAUL HYNES shunt regulator for my RAKK DAC II board.
Mr. Hynes is going to assemble a cMP squared system to evaluate when he finds the time. He has the parts, but he is an LP partisan so we must be patient. If he likes what he hears he is considering a perfectionists', obtainable unobtanium, computer power supply. It won't be cheap but we will no longer wonder how much we are losing due to the power supply.
Thanks to all, but especially to cics,
Rick McInnis
Rick,Re: When implementing carcass's REGEDIT change I was confused that there is no HKEY/SYSTEM though there in HKEY/LOCAL MACHINE/SYSTEM. Maybe this has been noted before and I missed it. Am I assuming correctly this is one and the same thing?
Following is what you are after: HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Control\PriorityControl\Win32PrioritySeparation
Re: I noticed that this board has taken a few times to finally keep BIOS settings. As I have slowly lowered the CPU voltage I would notice that some things would return to standard settings. Seems to have stabilized at this point. Lowest CPU voltage I can get to is 0.936 at 0.925 it is reset BIOS time and begin again. That has happened only once and it is interesting what reverted to "stock" and what remained at cics's recommended setting. cics, how long did it take running in for your board for it o accept 0.875?
The GIGABYTE board (at least the version I have; about 3weeks old) seems to have a “failsafe” mechanism built in. If the BIOS settings prevent a successful BOOT then the POST momentarily pauses and “resets” BIOS to the last successful BOOT settings. I suspect this is what has occurred to you. A neat feature, IMO, as it saves you having to do a manual CLR-CMOS. However, the “downside” is that you may be unaware that your settings have reverted. Therefore, always a good idea to recheck things with cpuz at the least.. I am able to run at 0.9Volts reliably. 0.875 affects an occasional hiccup.
Bling,
Edits: 10/19/08
Rick,
What I thought was a "failsafe" mechanism may just be the BIOS program having a mind of its own. It seems changing the CPU Voltage upsets some of the M.I.T. settings (see theob's post below). So, if you do adjust CPU voltage then re-check your M.I.T settings also. Good luck!
Bling,
On the regedit, i figured that was the case but wanted to be sure.
In my attempts to get down to the magic 0.875 volts I have had to clear the CMOS a few times. When I move too fast it just won't do anything.
At this point, the best I can do is the notch just above 0.900 volts. I am keeping the machine playing music 24 hours a day. Maybe tomorrow it will allow 0.9 or lower.
THANKS for the help,
Rick McInnis
In answer to 3 yes I have had to double check and update bios changes when I lowered cpu core voltage. I seem to have made it to .896 with bios stable, although I'll check again tomorrow. Welcome back to music by cmp^2!!!
with the downward (voltage) spiral?
Do you set it lower and let it acclimate or did you do this dramatically?
My previous board, though of the latest version, would not go below where I am with this one, now. Maybe I will have more success with this one.
I am very glad to have a music system again. I was really starting to worry that I was unconsciously doing something very wrong to the MB's. Just a rather bizarre form of BAD luck.
Thanks,
Rick McInnis
did it a step at a time so I could acclimate to the sound. but each time I did adjust cpu voltage I had to to re-adjust stuff in M. I. T. usually the 'enable cpu host clock control' and 'dram timing'.I stopped at .896 volts cpu voltage
Edits: 10/20/08
This power supply is fanless and has very low ripples
http://www.pcper.com/article.php?aid=615&type=expert&pid=6
thats much more ripple than the antec 430. in the antec you can change the fan for an low noise slow turning fan. with this one you don't hear anything in 20cm.
just get another one.
At least in the US, the price at NEWEGG, once you get your rebate, which I am still waiting for after two months, it almost ridiculously cheap.
As I noted before, who knows if this test was representative of what gets shipped? I have no way to measure.
The fan is very quiet. I have considered just disconnecting the fan but I cannot hear it at my listening seat.
I went in and removed the unneeded wires from inside the unit which goes a long way towards cleaning up the installation. This has to reduce radiation from unneeded antennae.
Most important was taking Alfred's (sonics) advice and adding capacitance to the voltage rails. I wish I had been able to try the SIKORELS he is using.
Does anyone know of a distributor offering these capacitors in the US?
For very little money and a fair amount of effort one can have what must be close to a perfect supply for cMP squared. That is, until someone comes up with something better! And that will surely happen.
Bye,
Rick McInnis
hi rick
unfortunatly the sikorels are history. ( 20 years old) but the new and perhaps better ones are slitfoil ( combined with panasonic FM or FE ) types from mundorf.don't put more than 30000µF behind the PSU on each railbecause of the initial current flow.
bye
alfred
Alfred,
I wonder if it is the SLIT-FOILS I was thinking about.
I agree there is such a thing as TOO MUCH. I am using 14,000 uF distributed amongst the terminals.
I am about to modify JULI@ by eliminating the break-out cable.
Did you remove the output jack from the board or just go to the pcb with small wires? I would prefer to remove the thing but worry that one can do unseen damage with what I suspect is a multi layered board.
What do you recommend?
Thanks for your advice.
Rick McInnis
hi rick,
i put an 75 Ohm cable directly to the board behind the the output jack.Avoid little wires soldered to a bigger wire or a bnc out jack because of multiple soldering points with increasing resistance. yes its a multilayered board, so it would be better you dont't unsolder the jack.
i don't think the jack will do any harm. i fixed the bnc cable for security reason with a little cable-binding nylon to an open pci slot of the case.
yes 14000µF divided by 3 rails are not to much.
obsession goes on
alfred
Found it
http://www.silentpcreview.com/article684-page4.html
very impressive indeed!
hi,
some new idea to discuss:
my new project with the atom dual core( also possible with gigabyte MOBO)
takes a sata to Compactflash adapter ( http://www.mini-tft.de/xtc-neu/product_info.php/products_id/19577 ),so ican youse each CF-card as a harddisc for my little Windows( takes only 90MB).
My music is on an external SATA Harddisc with own PSU, which is connected to the PC via eSATA cable.
The ESI Juli is connected with PCI-risercard (http://www.mini-tft.de/xtc-neu/product_info.php/products_id/26987/cPath/2598_2600/pci-1x/hfx-550-riserkarte-pci---oben.html).so its possible to cut the powersupply from the MOBO and connect an exteral PSU with analogue clean power.
i asked the JULI-staff if the card uses 3,3 or 5 volt to know which cable of the risercard to cut.if someone knows please answer .
bye
alfred
Sorry for asking an elementary question, but how do you power the soundcard independently? Is it powered through the PCI-raiser card? And if so, how is it done? I can't find a connector for external power from the link you provided.
Thank you!
you have to cut the cables of the riser card.so you can do the power to the open cable that leads to the card.
not yet done, so you better wait till i have broken my juli.
Next beta release (2.0b4) will do gapless playback (with background track loading). No options needed - done under the covers.
Thank you for giving me back DAYS of my life... no need to manually create customized .CUE files to simulate gapless playback - you DID IT in a heartbeat !
I just want to add my voice to the chorus of thankful people - know that I really appreciate your hard work and get tremendous enjoyment from playing music via your clever design.
Just today I sent a friend away slack-jawed after a little audition using cMP to play familiar old favourites. Such fun. More questions to follow...
Cheers,
Grant
That's not a Toy... IT'S A TOOL !!
All praise cics!
Oh, cics, that was really a problem, all the more so in contrast with the quality of playback.
The question: In order to avoid the gap I loaded (appended) all individual wavs to create a single 630 mb wav in EAC. The cplay refused to play it saying I don't have enough (1300 mb!) memory, however cplay 1.17 and later would load everything (multiple wavs) into memory. I have 1 gb of memory. The files are 44100/16 bit at 145dB SNR. How come That cplay asked for twice amount of ram?
Serge.
sounds like it's loading twice - try RAM load in cmp on NO
Hi! I have recreated the same error without CMP. Here's what I get at the end of diagnostics. Note that the size of wav is 602 mb.
ASIO started successfully.
AUDIO source [ D:\RIP\PORCUPINE TREE\Deadwing\Halo.wav ]
Genre = Unknown
Artist = D:\RIP\PORCUPINE TREE\Deadwing\Halo.WAV
Album = WAV
FILE: D:\RIP\PORCUPINE TREE\Deadwing\Halo.WAV
Wav file size = 602MB
Rate = 16/44100 (2 channels)
Samples = 158063220 ( 59:44)
Wav data is too large!
Physical (1023MB), Available (778MB), Required (1207MB).
I SIMPLY appended all files from 1 to last one after each other in EAC and saved as a single file. No CMP was started this time, only cPlay.
Isn't it weird?
Serge.
I just figured out what cics already knew when I compared the diagnostics of a normal EAC single file rip with an "appended" wav file of the same overall size, replicating your results. The double RAM requirement is because of the need for preloading data in play, I presume. Anyway, if your aim was to get gapless playback, you have no worries with the new 2.04 cPlay. Witness the impressive inventiveness.
So 16/44.1 wav of 602MB requires almost 2x more (1207MB) available RAM.
You, sir, are a god !!!
Shall you be praised.
Seriously, I can't wait to hear that.
gratefully,
Martin
There have been claims from some CAT users about how SSD significantly improves the SQ on their systems. Since cPlay effectively bypasses the HDD during playback through RAM loading, I doubt that "upgrading" to SSD would improve a cMP^2 setup by a good margin (if at all), especially for those who use external HDDs to remove most of the vibration and interference from the computer.
Has anyone tried using SSD, as either system or data disk? What are everyone's thoughts on this?
HDD keeps spinning when idle. SSD means the end of disk noise. When it gets big and cheap enough, I'll jump for it.
Does swictching to SSD *directly* affect the music playback quality? I can easily live with the spinning noise of my HDD when idle - if that's the only difference - it's barely audible from 15cm away (Western Digital 6400AACS).
hello,
I DID use cplay 1.3 final relis. with an E-mu 0404 USB sound card, and it did work good with ASSI4ALL as well as E-MU' drivers. But since couple days I could't use EMU card with its driver (everything is OK with ASIO4ALL). I don't know how, but the driver settings have been changed for EMU driver. Now I've got this message "ASIO start FAILED. Only 32bit ASIO format supported. In addition, have no access to ASIO Settings for ASIO E-M driver in the ASIO control panel of driver. I did remove and reinstalle the USB sound card, ASSIO4ALL and cPlay but nothing to do. It doesn't work and cplay shows the same driver error. Please advice
Hi, I'm also using EMU-0404USB with cMP and cPlay 2.0b2 in a toshiba laptop (Pentium M 1.4GHz 1GB ram) and 2.0b2SSE3b9 in a Dell XPS T8300 4GB
only using EMU ASIO drivers
some workarounds for you to try
- not able to configure ASIO parameters from cPlay (buffer 2ms 32bit)
(seting cMP to load XP previously) Run foobar2K and edit the ASIO parameters from preferences-playback-output-asio virtual device, sometimes needed to delete the device and add it again
- cPlay not able to load album, message ASIO drivers not found
just return from cPlay to cMP, turn off and on EMU0404USB, load the album in cPlay again, diagnose screen will come normally
cheers
Hi,
in the meantime I fixed it with Cubase LE, anyhow many thanks for your suggestions. Did you try to use EMU0404USB with ASIO4ALL? Do you hear the difference when applying SRC and/or Rate upsampler? What's about Windows optimization and what are U most critical Windows settings?
Awaiting to read from you. Greetings
Release notes .
Change Log (2.0b3 release):
- Support for PCM 32 (WAV or FLAC) and 32 bit IEEE754 Float (WAV only)
- Fix for clipping (inadvertantly bypassed in 2.0b2) - only has effect when volume set to 0.0db
- New 'Large' buffer setting used by Auto when output rate is 48k or less. Processors with L2 cache size larger than 2MB, its worth testing Medium and Large setting (for any output rate) - new 2.x DSP has smaller cache footprint allowing for larger buffers
- RAM allocation refinement when not using AWE
- Minor DSP refinements
Please REMOVE previous versions before installing cPlay 2.0b3. Normal, SSSE3 and SSSE3 B9 versions are available. For cMP² users, review these BIOS changes (underclocks FSB, CPU & disable additional USB items).
omg you've done it again. I have Chesky Area 31 on 24/96 flac file and on 16/44 file. I am hard pressed to distinguish the 2 now on 2.0b3. On prior levels of cplay I scored the 24/96 flac file as a 100 (on a scale from 1-100)and the 16/44 file as a 94. Now they are very close. The air and the detail are phenomenal. Dynamics as always are top notch. I don't know how to describe the sonics anymore.
Well done master!!
And I forgot to mention that the delay between tracks on a continous program file, while still there, is much improved to the extent that it is not annoying to me any more.
Yes 2.0b3 track loads faster and with non-continuous tracks, gap is hardly noticed. Gap may be a problem between continuous pieces of music. Gapless/seamless playback can be done (at least in design) - not sure if its practical.
Cics,I am very appreciative of your continuing efforts with cMP2. However, I too have an issue with the delay. Simply, I prefer not to have it. We audiophiles ask for the moon, I know! I also know the “problem” could be overcome by creating a single .wav for playback. So it is not the end of the world..
I have loaded the latest cPlay; in a word, wonderful.
Bling,
GA-G31M-S2L + E7200 + XP Pro SP2 + 1GB OCZ PC6400 + cMP + cPlay 2.0b3 SSSE3 B9 upsample 145.68db SNR @ 24/96 => USB EA Freeway2 => ..
Edits: 10/16/08 10/16/08
Re: "I also know the “problem” could be overcome by creating a single .wav for playback."
Oops I goofed.. Creating a single .wav does not solve the problem. Still a one second delay between tracks.. I assume cPlay loads only one track at a time from the .wav regardless of how many tracks are in the .wav. Sorry for the misinformation. (Not the first time.)
Bling,
This version sounds wonderful. Though decreased the gap is still an issue with 24/96 files. Would the theoretically possible solution involve loading the next file into RAM while playing the current one? Or what? What would be the practical limitations?
Although this could happen over a few seconds (while next track is loading), I'd prefer having 1-2 second gap then unpredictable dropouts. Fixing dropouts would then require larger ASIO latency which is unacceptable.
as with high def classical files, e.g., a Mahler symphony with 25+ minute movements? Could there be a background loading option with the current mode standard?
Do we need this feature ?
"Dark Side of The Moon", "Tubular Bells" (I, II, III, 2003), "Sgt. Pepper's Lonely Hearts Club Band", ANY live recording with applause+audience between songs... no need to list the rest - you get the idea.
"Essential" comes to mind... don't make me reach for foobar/Winamp etc. !!
The cMP Squared platform plays Redbook audio waaaay better than the perfect-sound-forever designers could ever have hoped for - absolutely - and thanks to cics and the continuous feedback provided by the cMP Community, it's going to get even better. Whether it's better-yet-sound, or added features, the future is bright for this little project.
Without gapless playback, how can we lay claim to "best transport in the world" bragging ? Please, please, please cics - take this challenge seriously and see what magic you can conjure !
And a personal thanks from me, for adding the Hi-Rez playback feature in such quick order :) - try that with Sony !
Cheers,
Grant
That's not a Toy... IT'S A TOOL !!
Grant,
If eliminating discontinuites is the most important parameter for your musical pleasure there is no need to descend to foobar. Just go back to cMP version 1.
This is the typical audiophile dilemna: it is nothing unusual for exceptional sound quality to take a toll in convenience. One makes a choice. I for one, will gladly take superior performance every time. One advantage is no one would want to steal my equipment, too much trouble.
I do not think it is going to ruin my enjoyment of Sgt Pepper if there is a one second delay between the introduction and WITH A LITTLE HELP FROM MY FRIENDS. Yes, if I could eat my cake and have it, too, I would like that, also, but none of us have found a way to do that, yet, either.
I am gratified that there is someone like cics who is going for the BEST QAULITY HE CAN ACHIEVE, convenience be damned. I liken it to what has become of sports cars; air conditioning and cup holders, etc. (absolutely sickening), only the LOTUS ELISE delivers somehting visceral and connected. It is heroic for someone to go the purist's course. This warms my heart and pleases my ear-brain interface.
Version 2 is the best cics can do (for the moment, who knows what he will come up with next)with quality the most important goal. If in future he can find a way to make it more convenient WITHOUT COMPROMISING QUALITY he will do it. He has made it clear that this is a perfectionist's approach amd I thank God that that is his first concern.
Cordially,
Rick McInnis
extreme cMP squared partisan
Listening to "side A" of Dark Side Of The Moon, for example, with 4 sound holes, seems more like coitus interruptus than a cup holder in a sports car. Being transported with pleasure for 20, 30 or 40 minutes, is the very essence of listening to music, I say MUSIC not only sounds or worst, GEAR. CMP2 is unquestionnably the best player around, and cPlay2 takes it even further, but this gap issue is a serious drawback, IMO.
What good is the best sport car, over the second to best, if you have to pull over every 2 minute ?
That's why I think a ON/OFF switch on the single file option would please everyone and every recording, considered this could be done. Only cics can answer that question...
For now cPlay 1.4 is my personal choice.
long live CMP2
Martin
like a purist sports vehicle, either automobile or motorcycle, neither would make a good choice for long distance travel. If comfort, and an ability to remain comfortable over a long distance, is the most important parameter (analogous to continuous play) then one chooses the sedan.
It should seem obvious if both of these are possible, at this point in time, I suspect it would be part of the system. Not to say that cics will not find a way around the problem. But, I, would be disappointed if cics compromised ultimate performance for the very few records where this mars the intended presentation.
Since there are other avenues for this need why should this be important to someone trying to offer something different from what is readily available?
I am amused that one could think that cics is taking perverse pleasure in offering something that will frustrate instead of bring pleasure.
He has made his goals for this project clear. It is his intention to get the most information from the CD, or other digital source, other than SACD, as possible.
The old method for bringing a product to market was first you make something the absolute BEST you can and then try to distill that goodness into something more attuned to the general public. Of course, this is not a commercial venture. cics is doing this research for his own pleasure and, luckily for us, finds additional pleasure in offering this to others.
Maybe someone else can license the vital parts of this from cics and offer a more convenience oriented product for those who want this. But, I am surmising here, cics will continue to search for ultimate sound quality and will leave other aspects to those who find their motivation elsewhere.
And for that I am intensely grateful,
Rick McInnis
Heh-heh... very good analogy Martin - but let's leave that one alone and get back to the enjoyment of our "continuous" music tracks :)
I'm with Rick (and presumably others) on wanting the very best performance when playing back digital music (CDs, hi-rez downloads, etc.) - and appreciate the "hairshirt" aspect of getting it. The cMP/cPlayer system is the best I've ever heard, so that's it - I'm hooked.
However the gapless thing really is a spoiler... almost as troublesome as getting up to flip sides when playing records (hands-up everybody who remembers that...). In those days (which is still "today" in our house), the habit was/is to drop the needle in the lead-in groove and come back at end-of-side. No picking individual songs - and gouging the intersong gaps with clumsy cueing... just twenty minutes of joyful listening to waht the artist & producer intended (these things are sequenced and timed that way for a reason...).
My albums are stored as complete .WAV files - not split by individual tracks - for this very reason (..."don't need no stinkin random play to mess up that masterpiece"). So why not make up alternate .CUE files for the albums that fall into this category ? This way, side two of "Abbey Road" will still play as intended.
Yes, you need to don the hairshirt while creating new .CUE files, but it's a workaround that would let Rick have his Cake... AND Eat It Too ! (sorry... couldn't resist the reference).
For example... this layout:
...becomes this (note "Side Two Suite" entry):
True... there is quiet preceding "Golden Slumbers", but for my purpose, no cue point required - that's what the original .CUE file is for !
I think this is what Bling was trying to express... to cause the required "continuous" audio to load as a single chunk under cPlay 2.x. Seems to work fine for me, although maybe 2GB memory is needed for some longer selections.
I hope this helps anyone suffering gapless-anxiety. Now I can r-e-a-l-l-y enjoy TB-II with cPlay_2_0b3_SSSE3B9. Thanks again cics !!
Cheers,
Grant
That's not a Toy... IT'S A TOOL !!
...I've just listened to 4 or 5 of my favorite classical pieces and the performance is outstanding. For those interested they were (Copeland with Leonard Bernstein-music for the theatre, Antal Dorati pieces by Schuller and Bloch, Dvorak 9th by Fischer, Byron Janis playing Liszt piano concertos---2 Mercuries, one Deutche Gramaphone, one Pentatone btw) and it's like I'm hearing these pieces for the first time (at least the best I have heard them). I am astounded at the air, the hall sound, woodwinds floating like in real space. I could go on and on.
Well done, indeed!!
I would still prefer gapless playback but am happy to live with 2.0b3 as is . The sonics are spectacular in a natural, analogue, real live music sense.
and cannot wait to hear what Theo is so pleased with.
Theo, yours have been my surrogate ears during this month of involuntarily delayed gratification.
Thanks cics for your tireless exploration into the medium that should not sound as good as you are making it. Like the LP, it is funny when a technology comes to the end of its life that intrepid music lovers find a way to make it sound better than anyone ever thought it could.
Yes, we all loved/hated LPs. Compared to what early digital sounded like even 1980's LP technology sounded much better. But who would compare what is possible with the LP now with what we had to be satisfied with then? I mean those that can afford what is possible today.
Now we have the benefit of cics doing his utmost to show that digital sound can be natural and beautiful. Here at what most are saying is the twilight of REDBOOK. (I am buying more CD's than I have in years due to what I can now hear with cMP SQUARED) Best of all, with a little sweat equity almost anyone can afford to listen to beautifully rendered reproduction of music in the home.
Three (squared) cheers for cics!!!
Rick you are a poet as much as cics is a genius.
does one need special hw to get the benefit of PCM 32?
hi,
3 days before i reported from atom 330 board with e-mu 1212m SC.
Today i had the possibility to compare with the ESI Juli( which was led to a friend before).
Incredible detail, natural soundstage and coherent dynamics than never before.(Compared to gigabyte with juli , EMU 1212m and ESI Juli.)
If we don't need upsampling (seems to be a question of individual taste and perhaps DAC) this board with Juli and heavy cleaned Powersupply( cplay 2.02b and cmp) is audiophile heaven.
bye
alfred
Problem with upsampling with Atom 330 is Intel's poor choice of mobo chipset (power consumption is on the high side). So avoid this mobo if upsampling (including 121db SNR option).
All digital audio is upsampled! By this I mean outputting 44.1k from cPlay will still be upsampled down the chain. I prefer upsampling with SRC because math used calculates samples as per original analogue waveform (within defined error margin). This is done using bandlimited interpolation (based on Shannon's sampling theorem) which operates on continuous time rather than simple multiples (i.e. 44.1 to 88.2). It's amazing to see the extent of different interpolation methods used in audio. From crude linear (or ladder) rubbish to more complex methods.
hi cics,
yes the powerconsumption of the chipset in relation to the consumtion of the 330 CPU is oldfashioned.The atom itself which less cache than the E7200 has not enough horsepower for 141SRC.My dac is able to do 192/24 but i prefere nos..I hope that the atom 330 in near future is paired with a better mobo.but for all NOS fans the combination of atom with esi juliis a dreamteam.Tomorrow i give the gigabyte with esi juli and 192/24 a new change.beside my normal job i am musician in two groups with unplugged instruments and if we compare recordingsessions with 44/16 an 192/24 the last one for sure is better. but in my chain the upsampling of normal 44/16 material adds nothing which can convince me. The more laid back rests 44/16.
bye
alfredin this boa
Edits: 10/13/08
For the first time since redoing the computer per your instructions, I tried to play a HD-DVD disc last night and found that the unit will only put out 2 channels through the HDMI output but no multi-channel. Tried downloading the newest Realtek High definition audio codecs and it still only allows 2 channels output. Is there anything that could have been done with your instructions for building a CD player that would block multi-channel output?
Bill
Using Realtek may require Windows Audio service to be running (and mapped).
I would think biggest cause would be in DVD player's settings - choosing stereo over 5.1 say. There may also be a limitation of 48k output with Realtek which could be a problem with HD DVDs.
I have a cMP doing normal DVDs outputting 5.1 (Realtek SPDIF output).
Hi cics
(thank you so much for the gift of "better" digital playback !)
I wonder if you'll ever add 32-bit-FLOAT playback capability to cPlay ? I would love to be able to listen with the clarity of cMP+cPlay when evaluating mixdowns made from my DAW.
Audition, Cuebase, Nuendo, Samplitude, Wavelab, etc. all have the ability to render hi-rez output WAV files as 32-bit-FLOAT (no dithering required), and to listen in their native form through such a hi-rez chain as cMP+cPlay would be very instructive.
Maybe you can add it on the TO-DO list ? Please :)
Thanks...
Grant
That's not a Toy... IT'S A TOOL !!
Just need some test files.
PCM 32 bits (at 31 bits precision) is better than float IEEE754 32 bits (24 bits precision when using range [-1,1]). Interestingly, PCM 24 bits offers 23 bits precision. Quantisation noise would be below -150db for all these formats.
Have you compared 32 bit PCM vs Float? Reason I'm asking is PCM offers much better precision and file sizes would be the same.
Hi cics
Glad the Hi-Rez formats are on the horizon... can't wait !
Regarding the "32-bit PCM" vs "IEEE Float (0.24 float type 3)" question: No, I've not worked with full 32-bit precision files. Actually, I don't know of any DAW application that creates these.
To my knowledge, no hardware A/D converter outputs anything beyond 24-bit integer data (even then, noise floor prevents a true 24-bit dynamic range). So it would seem that 32-bit integer files would only be created by software-based operations.
Anyway, in the interest of progress, I have a sample file for you (and anyone else who cares) to play with ("IEEE Float (0.24 float type 3)" as reported by Adobe Audition in the FILE> OPEN dialog).
The selection is from a pipe organ recording I did a few years back; it's the unprocessed feed from a Calrec Soundfield stereo microphone - very true to the live event. Yes, the hum you hear is part of the event - generated by the air system for the organ pipes - it's in the air and had to be filtered in post-production. The joys of location recording.
Download the file here -> http://www.sendspace.com/file/f6rgs1
Hope this helps.
Cheers... Grant
That's not a Toy... IT'S A TOOL !!
Downloaded (2:34 32/96 float) and operational! Will release as part of 2.0b3. Adding another buffer setting allowed by 2.x.
PCM 32 requires sw operating in double precision (64 bit float). I think Sonar does this. Interim working files need to be in double precision to maintain resolution - only final rendition is done in PCM 32.
will anyone be able to make use of this on cplay/cmp?
cics: Great news... my professional workflow thanks you ! Looking forward to comparing what 24/192 does for native 32/96 sources.
Of course, maybe I DON'T want to hear... as it's a great disappointment to sweat over subtleties, only to have them obscured by conversion to Redbook standard for commercial distribution (16/44.1). Sadly, it goes with the territory.
I'm afraid we don't speak SONAR here, so I can't help you with a 32-bit PCM example.
Not sure what you mean by "Adding another buffer setting"... if it's a solution to the silence/gap we get when playing "continuous music", then THREE CHEERS again ! (just TRY listening to "Tubular Bells" ...)
Probably some tricky logic to have cPlay pre-load the "next" track just in time for a seamless transition - and yet not mess up the Forward/Back scanning and Next/Previous behaviour.
There's a worthy goal : Gapless cPlay 2.x
Thanks again cics for the quick response on my Hi-Rez request.
Cheers,
Grant
That's not a Toy... IT'S A TOOL !!
I'm looking for a good external DAC to complement my computer transport with cMP^2. My current one works fine, except that it only supports up to 24/96, and I'm eager for a 24/192 compliant one to fully realize the potential of cMP^2 on my setup.
I know that cics uses dCS Scarlatti - which probably worths more than my entire system - but that's way over my budget. Please share the information of your current external DAC with me, and/or recommend a good "cMP^2 compliant" one that's reasonably priced (preferably under 2000US, I'm still in college). Thank you in advance!
benchmark @ $1275 is good value with great performance. it accepts usb,pdif (actually bnc which is what you need for 24/192) and toslink.
The Benchmark internally resamples everything to something around 110 kHz. Is there a benefit to upsampling to 192 kHz before sending to the Benchmark when it internally resamples to a lower rate?
Thanks,
Alan
Let me put it this way: yes. I tried 24/96 through both bnc and usb versus 24/192 through bnc and 192 was way better in my system, way better. More dynamics better frequency extention both up and down, bigger soundstage.
Thank you for your recommendation!
I assume you are referring to the Benchmark DAC1, correct? I recall that in AOB computer transport 0.3, cics advised to use only non-metallic connections between the transport and external DAC to achieve galvanic isolation. Does the Toslink input of Benchmark DAC1 support S/PDIF and 24/192 signal? Being a purist, I will only consider external DACs with those features.
Toslink in general will only do 24/96. I know that cics advised to only use non metallic (which is toslink) but that limits you to 24/96. Benchmark does include s/pdif which if you use as bnc (by taking off the adapter)will accept 24/192. Bnc is best for high quality 24/192 data rates.
ADAT is used in studios and has same toslink connectors. That means high quality glass fibre cables can be used.
Challenge is in finding a DAC that has ADAT input allowing for 24/192. This would be ideal and better than USB.
I have nothing more than my intuition to tell me that galvanic isolation with transformers would have to be at least as good as all of that transducing required with tos-link. Soomehow I trust good old magnetism more than I would converting to and from flashing light.
I would think a transformer on JULI@'s output and another at the DAC's input would be best. At the least, just one at either end, along with good "wire" and BNC connectors would be hard to surpass.
One dilletante's opinion!
Bye,
Rick McInnis
did you fix your problem? I sure hope so, been a long time w/o cplay.
Yes Theo,
I am waiting for three MBs. I have to believe one of them will work.
I have been the world's repository of all bad ga-31m s2l MS's. All should be grateful for my sacrifice. Far worse than the hair shirt and continuous self-flagellation. I figure I have earned tremendous dispensation and should I die this week I will go straight to heaven, or something.
I, also, have ordered the e5200, so I now will have four processors to choose from and hopefully ONE of them will work. I am taking no chances at this point. At times I start screaming like John Lennon within COLD TURKEY (temperatures rising, fever is high). My neighbors are starting to look at me funny. I think they preferred the loud music emanating from the house to this manic, music deprived wailing. I really can't help it.
Go ahead and get the STEREOVOX. It's cheap, easily re-sold if you don't like it, has dedicated BNC connectors. I do wonder if it is a better match for transformer input DACs. I wonder if it could be a little harsh with the BENCHMARK but I am surmising.
I am plotting the addition of the AUDIONOTE transformer (for digital) to the output of the JULI@. Not a big deal, just a matter of ordering. I will let you know if I do it.
I am using Dave Davenports RAKK DAC II which uses a very modestly priced LUNDAHL transformer at its input. One could always use one of these at the output of the JULI@, for that matter. It is much cheaper than the AUDIONOTE, which uses silver wire, probably worthwhile at the frequencies involved. No question, one does need to isolate the computer from the rest of the system, even though a cics modded computer is nothing like a regular computer. Nonetheless, icing on the cake; a cake with icing made of butter, not shortening!!!
Yes, it does seem the toll taken is starting to present itself. Maybe this could be a HOUSE episode?
Bye,
Rick McInnis
well hopefully your next attempt will yield great results. just for the record last time I screwed up (trying to get out of minlogon so I could load juli 1.04 drivers) I had to reload windows. bottom line is when I started from scratch I was able to get more cmp optimzations than my initial attempt (yes I got better at it). So maybe also same for you.good luck
Edits: 10/14/08
Darn my dac can do 192 but juli can't over toslink.
ADAT can only go up to 44.1 or 48
using 2 ADAT can go up to 88.2 or 96
using 4 ADAT can go up to 176.4 or 196
most professional sound card DAC should allow that.
Mini-dac ($900) looks very interesting and can be powered by battery.
Many thanks to you guys for offering advice and great recommendations! I just read the manual of the DACs, and found out that both would support optical-in up to 192k. Both the Benchmark DAC1 and Apogee Mini-DAC have received great reviews, and one of the two will probably end up in my system.
Release notes .
Change Log (2.0b2 release):
- Important DSP refinements: reduce CPU load and improve processor cache efficiencies (vectorisation and alignment done)
- Fix skip forward and backward time displays
- Minor UI enhancements (track highlighted before loading)
Please REMOVE previous versions before installing cPlay 2.0b2. Normal, SSSE3 and SSSE3 B9 versions are available. For cMP² users, lowering CPU Host Frequency (which underclocks FSB and CPU) will help.
Dear cics,
How does enabling EIST function affect sound quality?
With my limited knowledge on computer-as-transports, I always thought that fluctuating cpu voltage and frequency (which, IIRC, is basically how EIST conserves energy) would impact SQ negatively. Of course, I may very likely be mistaken. Can you enlighten me with your reasoning for this particular setting?
Thank you!
In a cMP2 system, motherboard and processor are selected and configured for low power dissipation. This minimises the electrical noise pollution of outgoing signals - i.e. it lowers "jitter". The effect is measureable and convincing test results from an earlier (and thus less "aggressively" configured) cMP setup were provided by cics - see link below.
Although clock speeds and core voltages are set as low as possible, they must, as you suggest, perform playback without fluctuating otherwise benefits tend to be lost.
Given that the current levels (and thus the noise levels) in a computer are a function of the square of the operating voltage, a significant reduction in electrical noise can be obtained even from a modest drop in clock speeds and voltage levels. There's a boatload of material on this on the web - see e.g. silentpcreview.com.
Hope that helps,
Dave
I apologize if my question was a bit vague. My CAT has been set up in full accordance to the aggressive cMP configuration, and I do understand the importance of low clock speed and operating voltage. The part I want to find out is how EIST impacts sound quality and why. In my understanding, EIST operates by actively adjusting CPU speed and voltage according to workload. Doesn't this cause clock speeds and core voltages to fluctuate, and thereby damaging sound quality? I may very likely be wrong, but I would like to know why :)
When freq is changed, CPU stops (halt) for a short time enough to cause a dropout with low latency playback. New settings prevents this by limiting max CPU freq to just cater for desired output level. Also as Ryelands points, this lowers power consumption.
Disabling EIST is not a good idea, as CPU has to be locked to high clock speed (for 24/192), instead rather have EIST increase freq to correct level at playback time.
> In my understanding, EIST operates by actively adjusting CPU speed and voltage according to workload.
Not quite - speed stepping allows an OS to adjust CPU parameters according to load. On its own, it does nothing. As I noted in an earlier post, many reckon that Windows makes poor use of the feature and my experience is that BIOS programmers used often to screw it up some more.
However, although other schemes do behave as you suggest, by setting the power scheme to "Max Battery" as recommended (and provided nothing goes wrong), the CPU's speed is always half the maximum - it does not fluctuate. Of course, the processor needs to be fast enough to cope with the load when running at half speed but even the E1200 is that.
Dave
I never understood how the power schemes truly worked, thank you for clarifying it up for me. Now I can enable EIST without hesitation.
hi cics,
little harshness of 2.01 (specially with xonar which does more high frequency resolution) vanished without sacrificing resolution.if this was your intention it was succesful.great effort.
thanks and bye
alfred
PS: a new PC with atom dual core 330 works like a charm.
Edits: 10/11/08
Dear cics and sonics,
After my debacle with GIGABYTE MB's I would like to try this, at least, it could serve as a back-up.
One can purchase the thing for ninety dollars, certainly worth a go.
I suspect there will not be much BIOS manipulation possible. But, one wonder if its low power consumption will get one more than half-way there?
sonics, are you using the full on 24/192 upsampling?
cics, does this thing have sufficient L2 cache to do the above (if sonics is not doing this)
It sure has no more than one needs. One memory slot, one PCI slot, only two SATA headers, it seems this simplification could only be beneficial.
One wonders how noisy that little fan must be, easily remedied with a larger slow speed, I would think.
One wonders, if this thing is usable, if INTEL would be willing to modify the BIOS to cics's recommendations? (lots of wondering going on here!)
sonics, if there are any additional details you are willing to share I would greatly appreciate getting to hear them. Thanks.
Thanks to all,
Rick McInnis
PS cics, I hope all of this mess has not affected you. My thoughts have been with you.
Atom's mini-ITX form factor and minimalist design is a big plus. BIOS is limiting (but has some key things that will help) and makes it not as good as Gigabyte mobo with E7200 or E5200 (which is better) when upsampling. If you plan to bypass upsampling then Atom 330 is a brilliant choice. Atom 330 (which offers 1MB L2 cache) will allow for upsampling when using 121db SNR option.
GB offers low voltage settings that significantly reduce power. This allows for 24/192k @ 145db SNR output at just 1.36GHz with cPlay 2.0!
Alfred - Try 2.0b2 with Juli@ and upsampling.
Dear cics,
Not sure if I am reading you right ("which is better") but was that referring, in general, to the GIGABYTE/e7200 or e5200, or are you saying this e5200 is your new favorite CPU?
Of course, I am not familiar with the e5200; I do not become familiar with any of these things until they are brought to my attention through your thread, so please elucidate. Nothing wrong with saving forty dollars for the CPU for those starting from scratch.
I am willing to get one of these if this is your new recommendation.
While I have your ear would you tell me if a bad CPU could destroy an MB?
Still waiting for my next batch. And just as mystified as to why I am having so much trouble.
Thanks for the help. Cannot wait to hear cMP version 2 one of these days!
Thanks, again,
Rick McInnis
hapless dilettante
Haven't compared it to E7200 yet.
hi rick,
after one day and done the kondensatormods and an external eSATA-Harddisc,
the atom with the normaly inferior e-mu 1212 sounds as good as the gigabite with the asus xonar SC( which imho is more resolving than the July).
There are no BIOS-possibilities , but until now i don't miss anything.Perhaps the low powerconsumption and 120000µF of big Sikorels before the internal powerdivider and little low ESR-caps behind the Voltagedivider helps. I use a real pretty little case from silverstone, which is 20x15x30cm and an external powersupply which gives 12 V to the case and intern there is an DCDC converter.
I hope this helps
bye
alfredPS: i tried 192/24( only with 121.with 141 not enough horsepower) without drops.but i prefere 44-16.
Edits: 10/11/08 10/11/08
Hi Sonics,
Can I ask you how do you connect the Sikorels before the internal powerdivider?
Arthur
hi Arthur
as i said before the little silverstone case has an external switching PS which provides 12 V to an plugin on the rear side. i interrupted the cable and simply added 4 300000µF sikorels between + an -.In the case there is an DCDC converter which changes voltage to 12, 5, and 3 V for the ATX Plugin.
To each rail after the converter i added 10 low ESR 1000µF Panasonic caps.
i hope this helps.
bye
alfred
Are you using the fan on the Atom's northbridge or did you disconnect it? It looks as if it would be noisy.
Dave
hi dave
indeed he is very noisy and i disconnected the little fan and put a 90mm fan with an extra PS mechanical isolated over processor and northbridge. So in 20 cm distance you don't hear anything.
bye
alfred
Dear cics,
Do you think it is possible to have received four of these boards from two different suppliers that will not work?
I have used my original e7200, a replacement e7200 (as a control?) and my old e2140 (as control #2) and I get nothing on my monitor. I have two power supplies, one of them the capacitors added one, and I know they are working because I substituted them in my home computer (built of old music computer cast-offs).
The supply fans do come on. If I connect a fan to the MB, it comes on.
I have used various sticks of memory.
With all of the exciting things that have happened here in the last month and the simple fact I am without a music system, I am wondering if I should try another MB, but of course, that would be stupid.
I worry I am spending lots of money on returning these things.
Any idea, other than incredibly bad luck, that could account for this?
Thanks,
Rick McInnis
There is no HDD, no JULI@ connected, just trying to get a BIOS screen.
I realize this is too obvious, but have you also checked your monitor and connections?
Yeah, it is so obvious it was the last thing I checked.
But, the monitor worked fine on the other machine.
THANKS,
Rick McInnis
You've accounted for all the components, but have you tried the setup in another case? Mobo shorting somewhere, or front panel connectors bad - it's an outrageous reach, but it has to be something!
No chance of shorting.
Do you think it is possible for a bad MB to destroy a CPU?
That is the only thing I can imagine.
Of course, I could substitute the e2140 in my HOME computer and verirfy it works, but what a PAIN.
No way to verify whether my two e7200's are OK without buying another MB just to find out.
Thanks for your interest.
Rick McInnis
You must be using non-standard stand-offs as well, so look at those again. I wonder if the PSU may be better off in metal case - mobo is grounded through PSU. Worth a try to set up in a normal case. Also, interference will go in and out of a wooden case like nothing, although that's not the immediate problem.
Verifying CPU: unless you just take the computer to a prof. techie and let him figure it out, you will have to verify yourself that each component is working (or not) in order to solve this - unavoidable pain.
segars,
I know that is what is expected (transparrency to RFI) but there is no difference in noise coming from my very sensitive speakers (TAD 4001's and 2001's) with the computer on. No difference whatsover with the machine on or off.
I suspect this would not be the case with a "regular" computer, but with all of detuning of cMP squared it is certainly greatly lessened,
One of these days I am going to experiemtn with the STILLPOINTS cloth and see if that makes any difference. It certainly won't hurt.
My machine is laid out horizontally with the power supply as far away as the cables will allow. The PS is, of course, already in a metal box. It could be an improvement to place it within an additional metal box but it would still require a large enough opening for cooling and there goes your shielding.
My MB lays upon wooden rails resting on sorbothane pads.
I am going to wait for the next batch of MBs and if nothing works (YIKES!) I will jump from the Empire State Building after having one long last look at the METROPOLITAN museum. I will park my car in a no parking zone. (I do not live anywhere near NYC) Goodbye cruel world and earth hasn't been much better.
Rick McInnis
probable lunatic
Have technical area test the mobo. Also, try sourcing one from them.
Whether they would be interested in testing an MB procured from NEWEGG or elsewhere is the question.
I would have bought one from a local store if I could. Would make all of this back and forth much easier.
I will wait for the replacements (3 at this point) and hope for the best.
Thanks,
Rick McInnis
Release 2.x
The new 2.x series of cPlay represents a step change in processing performance. DSP changes results in much less RAM requirements, lower CPU load and lower RAM reads & writes. These are all important factors for high quality audio playback. Use of DLLs has been elliminated (SRC-0.1.4 upsampling is now native to cPlay).
cPlay no longer RAM loads the entire cue sheet. Instead, individual tracks are RAM loaded just before each track is played. That is, only RAM for the largest track is required thus allowing for RAM playback on systems with low available RAM. This does not apply when playing wav or flac files directly as greater RAM is needed (even more than 1.x). Cue sheets provide track information enabling track RAM loads.
With cue sheet playback, cPlay now starts much sooner. After each track, the HDD is accessed for subsequent track RAM load. There's a small delay as audio data is processed and stored into RAM. Tracks can be varied by rate or resolution, i.e. cue sheets can contain multiple content files of 16/44.1, 24/96... (wav, flac or both).
SSSE3
cPlay is available in 3 flavors: normal, SSSE3 and SSSE3 B9. Normal version runs on all processors. SSSE3 version runs on processors supporting SSE3 instruction set. Intel (Pentium 4, Core 2 Duo etc.) and AMD have processors supporting SSE3. Use CPU-Z to determine if your CPU supports SSE3 or SSSE3. SSSE3 is recommended.
Release 2.x introduces a new SSSE3 'B9' option that offers most agressive optimisations. This option may not work on some setups.
Change Log (2.0b1 release):
- New DSP design allowing for less RAM and greater efficiency. See release 2.x notes above.
- Minor UI change for phase 0/180 setting
Please REMOVE previous versions before installing cPlay 2.0b1. Normal, SSSE3 and SSSE3 B9 versions are available. For cMP² users, lowering CPU Host Frequency (which underclocks FSB and CPU) will help.
...and while it is close to the 16/44 version played back on cplay wav(192/145/2.0B9) it is better on cplay flac 24/96 downloaded from chesky's HDTracks. Doggone it!! Also with the new version of 2.0 one can play cue files bigger than 1 gig. Doggone it I thought I was done buying software. BTW if any of you other cicspack players out get a chance try some of chesky's new stuff. I just wish chesky would do cue files.More feedback on cplay2.0B9 sonics---simply spectacular in a natural sense. Yes I don't like the 1-1.5 sec delay in continous play cue files but this is minor in comparison to what 2.0B9 can provide sonically. The 24/96 version of area 31 is quicker in the highs/lows/mids but smoother. It sounds like 16/44 is just a little dull by comparison. Not by much but when has a little improvement not been venerated by the audiophile? If cplay playback of 16/44 is a 94 (on a scale of 1-100) then cplay playback of 24/96 is a 100. I'm sorry that I'll be spending more $. I cant wait to hear the RR stuff. Anybody hear it on cplay?
Edits: 10/08/08
You had to go praising another of these fantastic records! Now I'm up to 6 of them. I can recommend all the others:
Thinking about Bix
What Matters
3 Guitars
Urban Concertos
Fire Beneath My Fingers
What/who is RR?
Reference Recordings is releasing 24/176 material on disc that I believe you can convert to a pc file. Thats the good news, bad news is it costs $45.
I would say B9 timings are not a succes in my setup.
Sound gets less musical! and more harsch.
(juli@> Stello DA220MKII DAC)
This post applies to Gigabyte's GA-G31M-S2L mobo and Intel's E7200/E7300 processor. It's a refinement to cMP² Update post .cPlay 2.x (24/192 @ 145db SNR) allows for lower CPU speed. These settings are specific to cPlay 2.0b16 and onwards, for settings related to prior releases, review previous edited post. Setup ( explanation for the change is given in italics ):
Hardware & BIOS
- Use high quality RAM. Size:
- Source is largely 44.1k, use 256MB
operating power consumption drops by ~1.15W and performs much better at faster RAM timings
- Source is often more than 44.1k, use 512MB
operating power consumption drops by ~0.5W
Best results are gained using a single RAM module.
- Update Gigabyte BIOS to F6x ( Download found under "Additional files"). Refer to Gigabyte manual on how 2.
GA-G31M-S2L BIOS Setup
Each point below refers to BIOS menu item. Before applying BIOS changes, perform "Load Fail-Safe Defaults" followed by "Load Optimized Defaults".In BIOS main menu, press [Ctrl]+F1 keys to access advanced BIOS settings. Items in red are high risk and requires a low temperature CPU setup.
- Standard CMOS Features
Disable Drive A (floppy disk drive)
- Advanced BIOS Features
Disable:
- HDD S.M.A.R.T. Capability
- No-Execute Memory Protect
- CPU Enhanced Halt (C1E)
- CPU EIST Function ( Allows for lower CPU VIDs )
- CPU Thermal Monitor (TM2)
Set Init Display First to Onboard
Set Onboard VGA to Always Enable
- Integrated Peripherals
Set On-Chip SATA Mode to Non-Combined
Disable Azalia codec, Onboard LAN Boot ROM, Serial & Parallel ports
Disable Onboard H/W LAN, SMART LAN (if not using networking functionality)
Disable USB 2.0 Controller (if NOT using USB soundcard)
Disable USB Keyboard and Mouse support (not used by Windows)
Disable Legacy USB storage detect (better to replace such devices)
- Power Management Setup
Set ACPI Suspend Type to S1
Disable PME Event Wake-Up, Power On by Ring, Resume by Alarm, HPET Support
- PnP/PCS Configuration
Set PCI Latency Timer (CLK) to 128 (maximum benefit is gained when ASIO latency is below 64 samples)
- PC Health Status
Disable CPU Smart FAN Control (if no Fan is used)
Disable CPU Warning Temperature
Disbale Thermal Throttle TEMP
- MB Intelligent Tweaker (M.I.T.)
Set CPU Clock Ratio to 6 (applies to all output rates)
Set fine control to 0.0 (applies to all output rates)
Disable Spread Spectrum
Enable CPU Host Clock ControlSet CPU Host Frequency (Mhz) to 150 (best done with VID at 0.85000V or lower. if system boot is unstable increase to 160)
this gives lower FSB of 600Mhz (BIOS should reflect max cpu frequency of 900MHz). previous setting was 1.02GHz. at new 900MHz only cPlay 2.0b16+ can be used for 24/192k/145dbSNR. lower ratio locks CPU to 900MHz (see power settings in windows below). cpu load for core 0 will be ~89% for 192k output. power consumption is reduced.Set System Memory Multiplier (SPD) to 2.0 - this sets RAM speed to 300
Set DRAM Timing Selectable (SPD) to Auto
CPU-Z must reflect memory settings: 3-3-3-? @ 150MHz. last setting is RAM specific (kingston hyperx gives 7). reduced RAM reads & writes in cPlay 2.x lowers runtime power consumption. best results achieved when FSB:DRAM ratio is 1:1. although "Auto" setting yields excellent results, manual RAM timings (see previous edits for how 2) may offer improvements, e.g. using 3-3-3-5 with Kingston's 256MB Value RAM moduleSet FSB DeOverVoltage Control to -0.15V
Set CPU Voltage Control to lowest possible CPU VID
lowest setting is 0.85000 when EIST is enabled but depends largely on individual CPU manufacture (as a higher VID may be needed). see cMP² Update post for more details. make sure that EIST is disabled (see "Advanced BIOS Features" above). VID lower than 0.85000 can be achieved, e.g. 0.74375 which works in my reference setup. drop VID gradually until lowest cold start VID is achieved. although VID 0.72500 could work, cold-starts would require a CMOS reset indicating instability. power consumption with VID of 0.75000 or lower is less than 20W!Set CPU GTLREF Voltage Ratio to 0.636 (or 0.566) (both very good)
Perform Save & Exit.
Power Settings in Windows XP SP2 Professional
- goto Control Panel > Power Options, set Power Scheme to "Always On". Reboot.
System Devices
Additonal System Devices can be disabled. Goto Device Manager > Show Hidden Devices > System Devices and disable:
- Microsoft System Management BIOS Driver
- System Board
For all USB devices connected: goto Device Manager > select Device (e.g. Mouse or Keyboard) > right-click > select Properties > select Power Management Tab > Uncheck "Allow this device to bring the computer out of standby."
Allow for at least 48 hours of burn-in (this applies to existing setups as well).
Edits: 10/11/08 10/12/08 10/26/08 10/30/08 11/16/08 01/16/09 03/27/09 04/15/09 04/15/09 04/21/09
what are others using for cpu voltage for 145/192 2.0B9? I'm using 1.01 volts which seems too high. Cics are you still in high .8's?
Take it down to lowest allowed.
My other setup has 1.06250V, anything lower won't work.
...plus I am down to 11 processes in task manager and am still running 2.0B9 @ 192/145/ 128 latency plus all other 'recommended mobo settings'. Plus my power supply (new) has finally broken in. Questions: what bnc cable are you running to get 192? Do you still favor 24/96 over glass toslink versus bnc cable @ 192? I reread a lot of your jitter stuff (I'm retired with a lot of time and don't want to watch my financial accounts lately for obvious reasons so I reread your jitter analyses!) and wondering if you re run these jitter tests with new versions of cplay/pc settings or do you just do subjective listening tests? Doesn't matter to me because it(the sound) keeps getting better but I am just a curious retired engineer. I do want to replicate your play back as much as possible. I'm a little reluctant to go .875 volts because I don't want another computer crash that will force me to do another windows reload. But I want to see where my biggest tweak (short of another cplay version) might be. I'm thinking bnc cable (I use a powered Audio Alchemy I2S cable from juli out to dac in). What do you use?
thanks
BTW b9 sounds wonderful--smooth, big very nice. more feedback details to follow. its good to be back in minlogon thanks to seger
Edits: 10/04/08
I don't know that I can really fully characterize 2.0B9 yet but I don't like the 'each track load' phenomena. I have many classical/popular pieces that become ruined by the 1-1.5 second interupt between tracks (Lang Lang Live at Marinsky Theatre; Man Ray vol 3). It loses the recording artist's intent and melodic continuity with a loud tic. I can say after 24 hours of continous play at .9 volts cpu the sound is very analogue like. But with the change to 2.0B9 I also am now using juli 1.04 drivers and a bigger power supply (550 watts vs 350 watts). So a lot of changes are confounded. I really do like the sonics but by a wide margin prefer ram loading. I tried indicating yes in cmp settings for ram loading but I got the initial ram load delay and the track to track delay--worst of all worlds. It would be nice to get a choice for this.
Despite the sonic improvements in cPlay V2.0B9, I rolled back to cPlay 1.4 until someone finds a way to avoid gaps.
I cannot live with the gap between tracks, even reduced by the "ram load" in CMP. It's probably better with WAVs ripped by album, (I use FLACs ripped by tracks) but even there, a silence between tracks supposed to be uninterrupted... it reminds me of the first days of MP3, before they figured out the continuous play.
Any live album would suffer from that, let alone DSOTM, Sgt.Pepper and so many others.
Has anyone found a workaround yet ?
Thanks
Martin
Maybe this has been addressed before, but I couldn't find a reference to it. How would I play pre emphasis tracks in cplay and get them to my dac for de emphasis at that point? I have cue sheets with the pre flag, but no players seem to pass it to the dac. Any help would be appreciated (I have a DSOTM, Abbey Road, Sabbath, etc. that use these are the best masters I have for some of this stuff)
RME offers Emphasis for SPDIF output:
This needs to be manually set 'in flight'. I haven't seen similar setting in EMU, Juli@ or Lynx. Downside with RME is it won't do 192k unless using their AES card.
Soundcard's driver only accepts sample stream.
Hi Cics,
Thank you for the new version! Details of SSSEB9 is further improved by 5% compare to 1.3 SSSE3 which is significant. However, the RAMload-right-before-playback plus 145/192 feature is very sensitive even to keyboard/mouse interrupts and pops/clicks occur easily on both RME 9632 and Lynx AES16e even at max buffer. I used to do web browsing while listening to songs, but now I basically can't do anything else... So now I use the less CPU-intensive 145/96.
Just wonder would it impact SQ if the whole song is loaded into RAM before playback?
.
cics,
do you see any future possibilities to integrate cPlay with conovler engine like http://convolver.sourceforge.net/ so that multi chennel system and active system can be played?
Nice for flac users, and wav from ordinary RB CDs - quick load, only tiny delay between tracks, hardly noticable. Also solves the large-sized file problem - biggest albums load no problem now.Big plus for Vista: now I can play whole albums with just 1GB RAM installed.
(Main computer still XP and not likely to change)
Edits: 10/04/08
Hi cics,
thanks for the new version. My first impressions (after a few minutes): very detailed, yet smooth. Impressive. (B9 version)
However, I'm not quite sure about the track load feature. Playback starts very fast, that's a plus.
But, for albums like Dark Side of the Moon, ripped from DVDA in 24/96, Yikes !
A silence for over a second between tracks supposed to blend into another ... annoying, to say the least.
Do you consider a version on which this feature could be turned on or off ? Or is there a way to have 2 versions of cPlay installed at the same time ?
Thanks
I agree add a a option turned on or off load whole song file in RAM,can depends RAM size use option.
Thanks cics.
In cMP, set RAM Load to Yes. This works very well with single file cue sheets where WAV or FLAC data is available is system cache. cPlay will load directly from system cache making it much faster between tracks.
There is another reason I want pre load whole song in RAM.
I use VISTA,can do a power management let harddisk switch to standby mode if x mins no use.
Let harddisk to stanyby mode will reduce power use and RFI.
If load a track then can't do this.
With cPlay doing sequential RAM loading from the cuesheet list you have a quasi-playlist that can play a series of files from various albums or even a series of albums with a long cuesheet IF you do as I do putting a folder of cuesheets on the desktop which uses full addresses for the music files listed that reside elsewhere in a music library.
So why couldn't cPlay now easily be adapted to use actual playlists?
Is there someone who would know how to modify foo_cuesheet_creator to use the full path for each listed file rather than just the file name? I have been copying the path from Foobar file properties and doing a replace in the cusheet to accomplish this. With the full path included one can make cuesheets for files in multiple locations. With sequential file loading one can then have cuesheets of indefinite length, even ones whole library.
For single-track, CD Image file cuesheets, it's not too bad to just edit the FILE line with the full address of the target file, but for multi-track cuesheets it's a pain. Here's a trick to make it a lot easier:
1. Open album folder
2. Open cuesheet with Notepad
3. Click Edit, then click Replace
4. In the "Find what" box type FILE "
5. In the "Replace with" box type FILE " and then copy in the address of the target folder (from the address bar at the top of the opened folder), and add a \
6. Click the "Replace all" button and exit
7. Save the cuesheet, saving changes. In the case of Vista, put it in your Desktop folder. These files will play in cPlay.
Hi:
I also do it on VISTA,
(In the case of Vista, put it in your Desktop folder)
but It will some errors.
Can you post cue file?
thanks!
Did you sort out the cuesheets for Vista yet? I just noticed something that might help - When RAM Load set to YES in cMP Settings, you can't play these cuesheets and get the error message "Disk cannot be played. Check .cue file for correctness...." But if you change RAM Load to NO (as it should be with cPlay) everything works fine. I found this while switching between cPlay and other media players in cMP, the other players need RAM Load set to YES in order to work, and sometimes I forget to change.
Here's a sample (no need to post the entire cuesheet, as it follows the same pattern for all tracks):REM GENRE Jazz
PERFORMER "Thelonious Monk"
TITLE "The Composer"
FILE "E:\Music\Jazz\Thelonious Monk - The Composer\01 - 'Round Midnight.flac" WAVE
TRACK 01 AUDIO
TITLE "'Round Midnight"
INDEX 01 00:00:00
FILE "E:\Music\Jazz\Thelonious Monk - The Composer\02 - Bemsha Swing.flac" WAVE
TRACK 02 AUDIO
TITLE "Bemsha Swing"
INDEX 01 00:00:00
This is what I have been doing with the exception that I get the path from foobar file properties when I make the cuesheet in foobar by selecting the files I want to include (usually a whole album). At one point when doing a lot of them I made a macro to do all of this 'at once'.
These guys were involved in making cuesheet creator:
Brian Peyton, primary author: brian.peyton_NOSPAM_@gmail.com (2 years old address so ?)
Neil Popham: neilpopham@bigfoot.com
Haru Ayana - did the .9x updates (couldn't find an address)
Change Log (final 1.4 release):
- Faster FLAC decoding (for both 16 & 24 bit formats)
- Improved 24 bit DSP processing
- Finalise compiler optimisations
- Refine ASIO startup
Please REMOVE previous versions before installing cPlay 1.4 Final .
Both normal and SSSE3 versions are available. This is the last planned 1.x series release for cPlay. Work on 2.x has started...
Hi Cics Thanks you for such a wonderful player. My life has changed musically since implementing your player my System.
I have Esi Juli@ 48 Sample rate to a KECES SPDIF 192Khz DAC to Nad 372 Amplifier to Nubert Nuline120 Hi End Speakers from Germany.
And the sound is almosta dream if it can be expressed in words. Thanks !!
I have two questions..
1. I would like to have all the versions of cplay from the begining as I find sound on some older versions incredible and very open and natural. Yesterday I was trying out v1.b9 and it is certainly sounds better with my setup than the latest version. SSo I would like to try all your versions and find the best suiting to my system. So where can I download the older versions including beta?
2. About the latest 1.4 version I find the sound very clear and strong and each instrument comes out individually correct but somehow the together harmony and dynamics is missing. The older beta versions give me a very warm open and natural sound. Does the advanced DSP processing somehow supresse this Tube Amp kind feeling and open sound?
Your feedback is welcome. Thanks again for your time and efforts
sharesoftall Yes
B9 is doing a comeback with the new 2.x version.
Re older versions: I'll reopen B9 and B17. Others?
Changes in SQ is jitter related - Try setting DSP buffer size in cPlay 1.4.
Bass is not weaker in 1.4 - an illusion corrected by direct a-b comparisons. Reason is 1.4 is well and truly centred in midrange now, and richness extends right down to bass - good work. 145/192 best of all settings by far for me.
My initial evaluation has shown cPlay 1.4 has cured an occasional channel swap error on 24/96 FLAC files I was getting with cPlay1.1 under cMP. General SQ for 1.4 is on par with 1.1, perhaps showing better separation of instruments at the expense of a 'live' feeling - more extensive evaluation to come.
I am finding that using cPlay for 24/96 files created from foobar, using the old SRC plug-in, is delivering higher SQ than the 121.33dB 96khz upsampling option on the original WAV files via cPlay.
Could this be due to a reduced CPU loading on my old Toshiba laptop? Will v2.0 have the capability to create upsampled files for later playback to take advantage of this?
many thanks to cics for all his hard work!
So any load on CPU will affect SQ more so than when using Desktop which uses separate voltage supplies (12V1 mobo & 12V2 CPU).
Hi Cics,
When you say "24 bit DSP processing" do you mean you are manipulating the wav data during play back? If I set the volume control to max, would the data be intact?
.
I'm hearing the bass a bit weaker, maybe too weak, than 1.3 and before.
Experienced just one "hang" - this time with playlist of selected album visible. This was with Auto setting. In fact, after early success with 1.3, I had several "hangs" where the outline only of the playlist appeared, along with the unending hourglass. Using Juli@1.04, 145/192
All sounds clean and smooth, tighter for sure in SSSE3, but the treble is more open in non-SSSE3, and I still prefer it.
hi cics,
very happy with 1.4 sse3.
a little grainyness which has become obvious in the later versions ( though better resolution) vanished.Resolution of 1.4 imho a little better, but the most important change i think is smoothness.
Is the second core of the CPU important for your SSE3-cplay.Comparing one or two cores (always with SSE 3) i feel more laid back with one core.
bye and thanks
alfred
.
hi cics
i use a little windows edition for eeePC with includes minlogon ..but it only supports one processor of the dualcore 7200 processor.
with cicsplay 1.3 SSE3 i have no problems and sound is really good. do you think it will be better SQ using 2 cores ( i don't upsample) .btw.:i tried in a new pc my old e-mu 1210m card and with the new windows driver latency in and out is now the same.
bye and thanks
alfred
Hi Cics,Just would like to seek for your help... I have a small VB program that plays a playlist using cplay. It is a simple loop that for each song, starts Cplay, play the song, then exit cplay. And cplay is run in the background ("ProcessWindowStyle.Minimized") so that UI doesn't pop-up everytime the song starts (then I can do something else with my computer). However, starting cplay version 1.3, when it is run in the background it does not start playback, until I restore it back to foreground. Then Cplay starts to play. This issue does not occur in cplay 1.2. I swap between RME 9632 and Lynx AES16e and the issue persists so I don't think this is a driver issue....
Is it intended to be run only in the foreground, or would you be able to help change it back? Thanks in advance!!
-----------------------------------
FileOpen(1, "c:\playlist.m3u", OpenMode.Input)
Dim trackname As String
Dim myProcess As Process = New Process
myProcess.StartInfo.FileName = "c:\Progra~1\cicspl~1\cicsPlay.exe"
myProcess.StartInfo.WindowStyle = ProcessWindowStyle.Minimized
Do Until EOF(1)
trackname = LineInput(1)myProcess.StartInfo.Arguments = trackname
myProcess.Start()
myProcess.WaitForExit()
myProcess.Close()
Loop
FileClose()
Change Log (final 1.3 release):
- Reinstate functionality removed from 1.2.1 (ASIO channel remapping)
- Update SRC to latest version: LibSampleRate-0.1.4 (changes relate to bug fix when using low conversion rates)
- Additional compiler optimisations for better floating point precision and efficiency
- ASIO startup is done after UI
Please REMOVE previous versions before installing cPlay 1.3 Final .
Both normal and SSSE3 versions are available.
still having problems trying to launch 'an album play' from cmp. the cplay screen w/o any internals comes up but it just hangs there until I reboot.Just recently after I had reset GTL ratio to .566 I reset the cpu voltage to something below 1.125 (somewhere around .87). Anyway I just now reset cpu voltage back to something over 1.0. Do you think this may have an impact?
Leave CPU Voltage as is.Also test with 1.2.0 and let me know.
If you get the same problem for both releases, then reinstall Juli@ drivers or even better upgrade to 1.04.
Cplay 1.1 is good--no UI drop outs in 48 hours of playing. Cplay 1.2.1 lost UI on 1st attempt. So I'm back to 1.1
I'm not at home now but as I recall I did get the hang up problem with 1.2 but not 1.1. I will reset cpu voltage and go back to 1.1 and try that tonight (est usa).
cplay 1.1 and cpu voltage down to .95 and no loss of ui. I tried to trip the system up so that I would get the ui drop out and it never happened. ui would drop out when I would double/triple click an album from the cmp menu on 1.3 but does not seem to happen on 1.1.
1.3 - no bugs so far, no vanishing UI. But don't even bother to play at 145/192 settings without cMP - breaks up. 145/176 a bit iffy, all other settings ok from desktop. In full cMP 145/192 - superb.
are you running 1.04 juli drivers?
.
Hi, I was very glad with cicsplay graet sound, but now I cannt open it!
I kill process in taskmanager, reinstall many times,reboot and nothing.
Only says "cplay is already running"
I am with XP and asioforall 2.6
Please help me.
Changes in next release will prevent 'no window' situation.
I play 16/44.1 and 24/96 WAV files using cplay. It is excellent too. However, I don't like to use SRC resampling as I have a slow PC. I like the sound quality with no resampling.
With the current cplay, when I switch between different sampling rate sources, I have to manual select the right output sampling rate to avoid re-sampling. I suggest to add a button to enable straight through without using SRC.
Thank you!
Sounds like your DAC handles 24/96. It's much better to upsample 16/44.1 to this level rather than have your DAC do it (with much less precision). But you'll need CPU capability. Have you tried SRC @ 121db SNR (use much less CPU)? This would outperform any local chip based upsampling.
Yes, straight through only happens when input matches output. Will look into a 'match input rate'/'no upsampling' option.
Before i bought a juli@ i used a cheap c-media audiocard with Dogber drivers and asio4all (@48khz). The sound was nice, very Sonic. After an update to juli@ and its asio drivers, there was more detail, more clean sound, but less sonic!. Last night i'll compared some things and found out that asio4all sounds different (better in my case) then juli@ asiodrivers. The hardware buffer tab in asio4all makes the difference. When off it sounds the best, very sonic. If on the same as asio 2.0
I am curieus about other experiences with this. Plz juli@ users, try it yourself.
Do you have minlogon implemented? Windows security subsystem is a nuisance and interferes with device IO. Also, what mobo voltage settings are you using?
ASIO4ALL is very good but Juli@'s native driver should be better offering lower latency. I replaced my Juli@ with Lynx AES16XLR for Dual AES 24/192k output. Can't test.
Minlogon is new to me. I found the 'how-to' thread, but how to do it in SP3, the same way?I am using a asus Asus VM2A-HDMI (had that mobo still). Can't modify cpu voltage.
This is risky and if it fails, may require reinstall.
It works and it was very simple, just rename and copy.
Thanks for the minlogon tip. Just another step to digital music.
When I change winlogon.exe to winlogon.exe.bak in XP SP3 (or SP2) it changes back unless WFP is disabled. Can you clarify how you did it with WFP on.
hmm. I changed the name with totalcommander and copied minlogon.exe directly after that. Just rebooted and that was all. I mentioned in another thread that i installed a ripped and tweaked custom version of winxp. Maybe the file protection is turned off in that version, but i don't know how to check that.
"Just rebooted and that was all" - You changed minlogon.exe to winlogon.exe and did the registry change also, didn't you? Can you check in system32 if winlogon.exe is now 23.8KB - sometimes it reverts to original winlogon (490KB) when reboot. If it is, you have minlogon and maybe file protection is off. You can check this by running XPlite_trial , the new sp3 version, and look in WFP tab to check status.
Yes, WFP is disabled in this tweaked version of windows xp. I did the registry change before copying minlogon to winlogon. I just looked in system32. De winlogon is 24kb. When running XPlite, xp crashes!
Thanks for looking! It shows that you need to disable WFP one way or another, and that SP3 can work with it disabled. I have it going on a second system and so far it's ok.
This is a very important optimisation for Windows.
This will be useful for others using SP3.
I'll have to mention that i installed a totaly ripped and tweaked version of XPSP3 called: Windows XP Lite SP3 TweakedFast. The winlogon tweak is very good for SQ. thanks again.
And Cics; is it possible to make a guide on a simple webpage, where to mention al the tweaks to get better SQ?
Now its a labyrint to find the best tweaks ;-)
Hey there,
my hardwareconfiguration is:
GA-G31M-S2L mobo, juli@ scoundcard (driver ver. 0.979) and E7200 CPU, 1GB RAM
i am using a dedicated audio pc, there is nothing else installed, after loading XP taskmanager shows only 29MB used RAM so i think ther are less thing thas can interfere.
under cmp i am getting a blank screnn only mouse is visible. i have to restart under this condition.
under XP i can see in taskmanager that cplay is running. after killing the task i am able to load the album i tried before. after loading serveral albums the problem occurs again. cplay is not even displaying the tracklist or the diagnostics.
i played with different setting (sample rate, disabling AWE..., changeing ASIO latency) nothing helped so far.
i also reinstalled cplay 1.21
thats annoying, BUT:
cplay 1.21 gave me a great impovement in SQ.
i am very happy that i found this thread. a great adventure into listening experience. i enjoy it participating this projekt even as a passive reader.
cics, i have some more warm words in mind but i fear you mingt be overrun by those words from time to time. anyhow well done!
i hope i can give something back in the future when i feel settled in enough to post some funded findings.
Right know i am experimenting with the E5200 cpu (could not stop my from doing this :-)). very interesting it may be an option....
Thank you all!
Bernd
going to do it now...
uninstalled juli@ 0,979 drivers
installed juli@ 1.04 unified vista drivers (they are working with XP)
no GUI-Problems with cPlay 1.21 any more
no chance to compare the two driver versions regarding SQ right now.
I feel very comfortable with the 1.04 drivers. A friend of mine has same hardware/software configuration running cMP² but with juli@ drivers ver 0,979. Maybe we find some time to compare the two systems this evening...
Bernd
I would love to try with Xonar, but can't.
hi carcass,
why You don't try xonar with asio4all in cplay. Easy installation and much better than xonar with spezific asio driver in foobar or winamp.In asio4all You can limit latency at 64 .
bye
alfred
.
Hi cics! Like every other user of cMP², I'm grateful for your state-of-the-art creations :)
My current setup is almost identical to the "official" V2, except that I use a silverstone LC-17 case, GA-EG43M-S2H mobo, and an RME 9632 sound card. It closely matches (sometimes even outperform) the Ayon CD-1 in my system except for one aspect: sound stage. Playback with my computer seems to be flat and completely lack depth.
Could it be an error in my software/BIOS optimization? Maybe switching to your recommended mobo would fix the problem? Perhaps (mostly likely) buying a good DAC is the only way to bring out the sound stage in my cMP² setup?
Have you ever tried listening to cMP² without a DAC? If you think that cMP² is only meant to be used with a good DAC, can you please recommend one that's resonably priced?
Sorry for asking so many novice questions. Looking forward to your reply!
I would avoid this mobo. Rather use recommended GA-G31M-S2L mobo. Have you implemented suggestions as per update post ? Things like minlogon, priority control, PCI latency, etc. have SQ impact. Also set RME's ASIO buffer to 32 samples (lowest). I assume you have E7200 processor and good PSU .
Be sure to connect computer to separate AC circuits and use dedicated line filter with good power cord (for both AC inlets). Try not using a preamp (too much coloration).
Have you ever tried listening to cMP² without a DAC? If you think that cMP² is only meant to be used with a good DAC, can you please recommend one that's resonably priced?
Yes (as in use soundcard DAC). That's what I used in my minimalist setup (24/192 Juli@ or EMU1212 analogue outs to powered speakers with digital VC in cPlay). Sound is excellent but external DAC is better. Many DACs are available offering good value. SPDIF input is all you need - be sure it handles 24/192! Not sure if RME will output SPDIF 24/192 but Juli@ certainly does (via RCA).
Thank you for your advice.
I apologize for the bits of misinformation in my previous post. My setup uses E7200 processor and *one* (not two as you advised) XG Magnum PSU. I chose the PSU - coupled with an individually powered external HDD - to eliminate mechanical vibration and minimize heat inside the case.
I'm using a pretty good power cord for my computer (which shares a dedicated circuit and line filter with the amp), and the HDD is connected to a different circuit but uses the default AC adapter provided by the external HDD case. Is this deviation from your recommended setup detrimental to SQ? Should I power the external HDD with a good PSU instead?
I just ordered GA-G31M-S2L to replace my current mobo, which I should've done in the very beginning. Which RAM module do you recommend to go with GA-G31M-S2L? In cMP 1.0 you suggest that we use 2xKingston HyperX 512MB, but in the update post, it becomes "a single 1GB RAM module of high quality". Should I stick with Kingston HyperX? Or would any high quality (CL3? please elaborate) suffice?
Lastly, are there any conflicting optimizations (excl. BIOS) in the cMP 1.0 and Art of Building a Computer Transport 0.3 manuals? Obviously the BIOS settings between V1 and V2 are different, I just wanted to know if I've done right by applying all software optimizations possible mentioned in the documentaries.
Sorry for flooding you with so many questions, I'm just too eager to experience the nigh-perfect SQ everyone else is enjoying!
If external HD is connected via SATA then no problem. Nice PSU.
New mobo and single module 1GB Kingston HyperX 800MHz CL3 is recommended. Need about 200-300 hours burn-in.
Any changes to AOB CT 0.3 is covered in cMP installation guide (Page 6). Also, you must implemented all changes as described in update post. For example, have you done minlogon?
Yes, I have implemented all changes *possible* (quite a few optimizations had to be discarded due to suboptimal hardware), including minlogon. I have already ordered the new Gigabyte mobo, but the Kingston HyperX 800MHz CL3 is not easy to nail down in my area. If I fail to get the KHX6400D2UL, are there any good replacements from say, Transcend/OCZ/Corsair?
Once I've obtained the recommended hardware, I'll perform some listening tests and (hopefully) come up with a detailed comparison :P
As far as I can tell the KHX6400D2UL is not being made anymore and is out of stock just about everywhere; be careful: suppliers still list it and when you order may try to send you the Kingston "recommended replacement", but it won't be low latency.
So it's a good open question for everyone: is there RAM out there with 3-3-3-10 latency like the HyperX?
what segar says is true. The retailers don't seem to think there is any difference.
I never got the ultra low latency, just the low latency. The ultra low latency version has been unavailable for quite awhile.
Nonetheless, when BIOS is set up as recommended I get the "numbers" cics shows on CPUZ.
I also found that when using regular HYPER X I remember seeing the same numbers except for the "last" one in CPUZ. So, my guess is that this is not the most critical aspect of the assembly of a music computer.
I have not listened to the regular memory so I have no idea how it may affect the sound.
What "numbers" on CPU-Z are you referring to, the AOB 0.3 documentation or cics' update post? The reason you're getting the numbers is simply because - correct me if I'm wrong - underclocking the RAM frequency down to 350MHz would bring most RAM down to CL3.
Nonetheless, I believe that RAM quality is one of the critical aspects of cPlay (being a memory player), I'm just unsure whether ultra low latency RAM is so crucial to SQ.
I know the Xtreem series from the brand "TEAM" makes 3-3-3-8 RAM, but they sell it at a ridiculously high price. That's the main reason I need cics' opinion on RAM quality. If CL is critical to SQ, then I'll grit my teeth and buy it, but I'm hoping cics could point to some high quality RAM that works well with cPlay and remain reasonably priced.
I also have some other thoughts regarding RAM latency. According to Kingston official website, KHX6400D2UL operates at CL5 under normal (1.8v) voltage, and achieves CL3 at around 2.3v. However, cics advises cPlay users to leave ram voltage at 1.8v in his update post. Does this mean the Ultra Low Latency RAM will not operate at CL3 under optimal cPlay settings? If that's the case, then perhaps any good quality RAM that is CL 3, even 4 at 1.8v (and 800MHz) RAM voltage would be a good alternative to Kingston HyperX.
Of course, we would need cics or other computer experts to confirm that :)
Kingston supplies these as well. Both RAM and CPU have large impact on SQ.
RAM setting is underclocked and operates at standard DDR2 voltage (1.8V). Idea is to use highest quality RAM so low performance setting creates a low stress environment. RAM speeds at 350MHz should be a doddle and cause less electrical noise over cheaper low quality RAM offerings.
Thank you cics.
Is the operating voltage under which the claimed CL is collected significant to RAM performance? For instance, if two modules both offer CL4 at 800MHz, but one achieves CL4 at standard DDR2 voltage, while the other must overvoltage to have the same performance, then (seems to me) the former is clearly superior in quality, right? Does this heaviliy impact SQ?
Hey "cics", first of all let me just say that 1.2.1 is a very nice improvement over 1.2. The midrange definitely has more resolution and is more 3D with more front to back separation. Keep up the good work!
Second, I want to say that with another music player which I was using previously, there was also an "Invert" Phase button. I liked how that player sounded with the option checked so I decided to give Phase 180 a try with cPlay. I'm still experimenting, but I think I like what I'm hearing. It seems to bring things into sharper focus as well as better dynamics. It had been explained to me that sometimes when manipulating jitter to such low levels that acoustically the phase gets inverted (so re-inverting it to 180 brings it back). Does this sound right? Or can you give me a better explanation of what is going on here. At the moment I've been doing all my listening with headphones - do you think this is system dependent? Or has anyone else tried this with positive results?
Unhappily, many CD's are recorded out of phase for 1/2 the systems such that a compressive sound becomes a rarefacation.This dates back to the fact that Ampex and RCA in the 50's used pin 3 as the positive while the Japanese used pin 2 of an XLR.
Some people are more sensitive to this than others. There are even many CD's especially early ones where the phase alternates with each track, and some recordings with multiple mikes where some of the microphones are out of phase compARED TO OTHERS. This leads to a dulling of the sounds, poorer soundstaging and flat nonpercussive bass.
It would be wonderful if someone could add an attachment to cPlay or cMP that wsould allow a reversal of absolute polarity at some stage of the preocess.
Bill
already there. When playing cplay hit the ; button and watch the small panel next to the clpay icon (bottom left. this reverses polarity to 0 or 180 degrees.
Thanks. Will try as soon as I get home.
Bill
It had been explained to me that sometimes when manipulating jitter to such low levels that acoustically the phase gets inverted (so re-inverting it to 180 brings it back). Does this sound right?
No. Jitter distortion results in extremely small signal amplitude errors whilst phase 180 causes signal amplitude to invert! Big difference.
The difference you hear "sharper focus as well as better dynamics" is similar to mine and has to do with recording polarity rather than jitter. This setting changes for CDs as well as tracks. Phase 180 corrects recording mistakes where phase is erroneously inverted. Same can be achieved by swapping speaker terminals.
Hi,
Cplay 1.2 doesn't work at my system (Vista, Juli@) After install the settings window shows up, after pressing ok, it's gone while cplay.exe is still running as a proces. After i kill the proces and start cplay again, cplay loads only as proces, but nothing on the screen. Cplay 1.0 and 1.1 works fine.
What can be wrong?
Version 1.2.1 still doesn't work with juli@ device enabled!?
This post might help.
Xonar and Vista seems to be the problem.
Thanks again!
Just to report that the new cPlay 1.2.1 works with my Konnekt firewire interface now.
Thanks.
Any chance of upgrading to GB mobo and Intel E7200 (or E5200)? Still need those power consumption measures with this spec (low voltage settings on GB mobo is a gem).
hi cics,
thanks for looking for all little problems.
xonar asio (in windows xp) is shown now in the cplay 1.2.1 SSE3 controlpanel but don't play because xonar asio only supports max 24Bit(also in cakewalk and other audio progs) and in diagnostics 32 bits are wanted.
in vista i expect the same problem.
thanks
alfred
cPlay only supports 32 bits ASIO output format (better SQ when not catering for multiple output types). Have you tried their latest beta version which has some ASIO updates?
BTW - cPlay 1.2.1 improves on 1.2!
After switching between Auto, Small and Medium while in cMP (with cPlay stopped), the cPlay UI disappears sometimes when opening an album. Also happened outside cMP when initially testing the 1.2 cPlay - comes back after killing in Process Ex., but SQ in cMP is better w/o running Process Ex - so is this a bug or a limitation?
Such intermittent problems could be resolved with a full driver reinstall (i.e. first remove it entirely).
Even with fresh download and install, UI vanished in XP, this time on 2nd opening, and again in cMP once. Biffland's post spurred me to try 1.04 drivers for Juli@ - vanishing act once again in XP unfortunately, but nothing else yet.
Meanwhile, I've spent 2 hours comparing 0.979 and 1.04. I think there's a subtle improvement, in the same direction cPlay took from 1.0 to 1.1. Time and many more hours listening will tell. Do give it a try.
.
to me in my system (dual-boot means this only takes a minute to go from cMP in one OS to cMP in the other, so the memory is fresh).
I see from another post that you no longer use Juli@, but perhaps others reading this will test the 1.23 and 1.04 drivers and give some feedback on sound and cPlay functionality with them.
Initial cPlay trials with 1.23 drivers from desktop (XP) and with changing settings in cMP was fine, so perhaps there is a conflict with the Juli@ Vista drivers and cPlay 1.2x. That said, I've had no new probs with 1.04 as I've gone back and forth with 1.23.
I lose UI when in cmp and I try to play an album from cmp menu. if I more than double click
9triple click)with the mouse I sometimes lose UI. I ocassional lose UI when I try to play an album in cmp from the keyboard. but this occurs less than when I use the mouse. in all cases when I lose UI I need to reboot.
Problem is caused by ASIO driver. I assume you have Juli@ 0.979?
yes I have .979
Now I can't make it happen either, even by switching wickedly fast, which makes cPlay grind into action very slowly. It happened right after install when I initially ran cPlay 1.2 and 1.2.1 in XP, and a couple of times (in both 1.2 and 1.2.1) in cMP (minlogon implemented) after stopping play, changing settings, closing cPlay and reopening same album,(this is after reinstall). If it happens again, I'll download the file anew, and let you know if it still happens. For now I'm leaving it on Auto, which I am satisfied gives me the best sound in cMP anyway, and enjoying the music. With 1.1 (and 1.2) I was sort of missing something from 1.0, but not any more.
cics
i use the latest beta 1739 driver from asus xonar and i hpe asus will provide some better asio drivers.
and yes 1.2.1 is better and i am wondering how far we will go.
it remembers me on analogue times when playing with high end turntables( for ex. platine verdier and shindo stuff).and every time i thougt to be at the very end.
bye
alfred
Change Log (final 1.2 release):
- Tuning parameters now part of Settings. "DSP Buffer" at Auto (default), Small or Medium is provided. Small sets parameters as per cPlay 1.1, Medium as cPlay 1.0 and Auto selects parameters based on output rate.
Guidance: CPUs with 2MB+ L2 cache, use Auto setting. Small is ideal for high output rates (above 96k) whilst Medium is best for 96k or lower. On CPUs offering less than 2MB L2 cache, Small is recommended for all output rates.
These buffer sizes work best with low ASIO buffer or latency setting (less than 512 samples).
- Revise ASIO interface at startup. Some soundcards do NOT appear in list. A known problem with Xonar.
Please REMOVE previous versions before installing cPlay 1.2 Final .
Both normal and SSSE3 versions are available. For cMP² users, there's an important BIOS change ( CPU GTLREF Voltage Ratio ) giving improved SQ.
Hi Cics,
It is interesting to see how much imprvement you can achieve each time with new version. Can you give us some idea what was changed to achieve the effect? I mean, you can even predict the improvement of high Vs bass etc by changing the parameters!
I thought the data is bit perfect already and what you are changing is really jitter performance. But how can you predict which frequency will improve based on the software tweaks?
Would be very interesting to know!
Removed some overheads here (low level but affects SQ). I want to wrap-up 1.x series and was going through DSP for possible improvements (which I found) and address Xonar bug. Some small functionality was removed (will put back in next release if SQ not impacted).
Same bits but different sound is caused by jitter. As to predicting frequency response, that's impossible. cMP² overall design objective is to create a low stress circuit (this is fundamental to achieving ultra low jitter). Changing software (more efficient) and reducing L2 cache footprint (within reason) makes a big difference ito reducing both random and periodic jitter (see Jitter Paper ). There's jitter frequency components that must be effectively managed - I've explained it here .
HKEY_LOCAL_MACHINE\SOFTWARE\ASIO ?
Hi cics:
After I upgrade to 1.2 SSE3 version.
Woo!It sounds so excellent,very much,really!
My setting is 176.4MHz,Buffer size is small.
RME9632, 3.071 version driver.
Thanks again!
OMG, Another best sounding cplay ever!! (145db, 192, Medium DSP buffer...)
Hi Cics,
Once again thank you... I have played cplay 1.2 for a few hours now, I don't know what's magic inside the cplay 1.2, but now at 145/medium DSP/192, the music's dynamic is best I have ever heard, therefore rhythm and musicality is best among all cplay versions!
On the "bad" side, upsampling to 192 consumes a lot of cpu power and now I am forced to use 1.5Ghz. I am planning to upgrade to Lynx AES 16e and hope by doing so to improve accuracy and sound stage while enjoying the magic of 192.
Thank you again!!
Up and running 1.2. Thank you once again.
Confession/journey of an audiophile or "how good is cplay 1.2/cmp"-- originally posted in Computeraudiophile.com in an effort to spread the word.
I was listening via usb to my benchmark dac for several months and loved it. Originally I was using j river then I migrated to cplay which I absolutely loved. Could only get 24/96 out of usb so I was compelled to try new approaches to get 24/192. Since I loved cplay I built a new computer per cmp (see Audio Asylum) which wasn't as hard as I originally thought. It is basically a gigabyte mother board, intel 7200 dual processor and kingston ram (all per cics specs). I also got a juli sound card which is capable of 24/192 digital output (just what I needed). All this plus I loaded the cmp software of cics (again see Audio Asylum). This was significant because cmp software gets rid of much of the windows overhead to make for clean 800+ meg cue file playback. I hit the motherload--absolutely beautiful: dynamic, layered, transparent and very undigital. But over the last few days I tried to think what else could I do to identify bottlenecks in my system elsewhere. I had been running a dbx driverack pa electronic crossover into my martin logan monolith III speakers for ultra clean triamping. But it always concerned me that the pa took analogue in and converted it to 24/48 to perform all the crossover functions and then reconverted back to analogue. When I had a sony scd1 I tried bypassing the x-over for the highs by running the sony output directly to my high frequency amp (through the external ML passive crossover) but I didn't like it. For some reason the electronic x-over was better, so it was back to the digital crossover. But since I made so many changes (cplay/cmp)I thought it was time to challenge this prior conclusion. Since I read a comment by Elias that the Benchie could drive xlr y connectors I had my enabler for running highs directly to the panels and bass through the dbx to my bass and subwoofer amps.
What a revelation!!! I can honestly say this set up (Benchmark-> QSC amp-> Electrostatic panels of the martin logans) is so analogue like that it is scary. I know you all heard this before somebody tries a little tweak and all of a sudden instant kharma. But this pc audio is a whole different and unique (per cicplay/cmp hardware/software) approach. The results are stunning in the natural sense of transparency, non fatigue, dynamics and low level detail that I literally listen to music (rewardingly) for 5 - 7 hours a day (I am retired).
Wanted to share with all here because this approach is very cost effective. The pc parts cost me $420 and all the hardware/software for cplay/cmp specs are free. I rebuilt an old computer so I saved $ on case, cd drive, dvd drive costs. You all know how value minded the Benchie is, but you should hear it at 24/192 going straight into a high powered amp into electrostats. Finally after 40 years as an audiophile I have the sound I have wanted. Sure I can get better but we all know the feeling when we 'breakthrough' to some new level of performance and wow did I breakthrough. Please don't take this as yet another guy bragging 'I have the best system'. I know I don't in an absolute sense but I do 'for me'. I wanted to share the unbelievable performance levels one can get for very little money (by historic high end standards) so that others who share my love of / need for music can also try this approach. Yes it is difficult if you are not a computer geek but I was not and now I reflash my motherboard bios with no fear. Guys this is the real deal. It took me 6 months to build this but I am slow because I was a newbie to computers. There are a whole bunch of cplay/cmp grads on AA who are willing to help you when you run into trouble. Look I didn't like computers and resisted for a long time but now I have the front end of front ends as well as the knowledge of computer hardware/software that I picked up along the way. Whats not to like?
I share this here (in Audio Asylum) because cics does mention that once one gets to a new level of performance one may have bottlenecks downstream. I always preferred going into my digital crossover prior to my high frequency amp not by a long shot but by a noticeable margin. Not now, going directly from my dac to the high frequency amp bypasses a critical stage of src and then back again to analogue. Cplay sounds so natural and grain free this way (bypassing the digital x-over)that my prior set up in retrospect while not hard has (how can I describe)hyper highs (exaggerated but pleasing to me as a high frequency freak). The depth, natural ambience, soundstaging now is very reminiscent to my vpi/koetso analogue glory days but honestly it is way better (due to dynamics, bass and sweet sweet highs).
See link below (Computer Audio Software). I must say this is a surprise as I never believed open source would get press airtime let alone win awards!!! This takes courage - well done EnjoyTheMusic!
Great feedback (esp. the $420 bit). Performance from having full cMP² specifications implemented (including GB mobo, E7200, minlogon etc.) is exceptional. This level of spec wasn't even reviewed at EnjoyTheMusic (different mobo, processor) which would've been a significant step change.
Try latest cPlay 1.2.1 - improves on 1.2 (more liquid midrange).
N/T
Considering review used video graphics card, Intel Quad processor and so (also no Minlogon), that's a good result. At reference level (when using full specs) there's a large step up. For example, GB mobo gives excellent voltage control which helps reduce stress and lowers power consumption.
Congrats on your award, totally deserved.
You should also get the Philanthropist of the Year award if such a thing existed.
cheers.
Dear cics,
With nothing in common with what you recommend for the optimum foundation for a music computer, one wonders what he is listening to?
One has to suspect if he heard the REAL THING he might really be impressed.
I know the recognition is exciting, though I must admit I find this publication to be slightly bizarre, or should I say more bizarre than the others, it is like someone reviewing a home built version of a commercially produced device and then saying, "this is what it sounds like". The orony here is that no matter what, this would be a home built device but could you be much clearer as to what constitutes a cics cMP squared system?
I would be interested in hearing what the 6moons guy thinks of this. He strikes me as one who can hear the device without SEEING the packaging.
Maybe I am dreaming but I cannot imagine him assessing a device without it being what its designer intended.
Monday I return to the fold with a working machine.
Thanks,
Rick McInnis
PS what's with the Israel thing? Not that you have to answer that! Next notice will say you are from Finland. Bon voyage!
Isreal was a misprint which was highlighted to Bill. It's not the first time...
...could you be much clearer as to what constitutes a cics cMP squared system?
Fully specified means as per documentation and applying update post in cMP thread. For example, changing to cPlay 1.2.1 is significant as well as applying BIOS change like CPU GTLREF voltage. cMP² continues to evolve as more improvements are found (either sw tuning or new hardware). Level 1 is using Biostar mobo, level 2 is with GB mobo (current).
That's not to say that same can be done but selectively. That is, different mobo, RAM etc. - this will yield excellent results but not reference level .
Dear cics,
I was having a little fun with my sentence and maybe clarity was a casualty of my intended meaning.
I DO think you have been transparently clear as to what you think is best.
Of course, no one could try out every MB and memory, though, it would not surprise me if you have(!) but you have made it plain as to what is the SURE way to achieve a beautiful sounding digital playback system.
I will not sleep tonight if you think I thought you had been obtuse with your recommendations.
I look forward to someone trying this out AS RECOMMENDED!
Believe it or not, I have just proved to myself that I have received, yet, another defective, out of the box, GIGABYTE MB. Maybe my processor is OK after all.
Luckily for the rest of the devotees the bad ones are all being sent to me. Weird. All from Newegg.com. The one I destroyed was from ZIP ZOOM FLY, it worked until ....
Sorry for the confusion.
THANKS,
Rick McInnis
congratulations!! well deserved. just loaded 1.2.1 (initially forgot about the asio remapping) and once I loaded properly, happily running 1.2.1. feedback later
,
I'm getting a stuttering sound on playback of some classical pieces. I have 7200 so I assume I should use auto setting for dsp playback. Correct?I may have found the problem, after I load a file if I adjust volume too quickly after load bar ceases to display it causes this stutter. Seems more sensitive/touchy now.
Loss of HF details, flatter sounstage - and it's not subtle.
I understand this switch changes the proportion of memory allocated to user applications vs. system processes, increasing in favor of former, but why it can degrade sound - I'm not equipped to even take a guess (we have cics for that! :-)).
My advice is to listen carefully to your system with and without this switch.
,
If anyone has an idea about this problem, I would greatly appreciate your thoughts.
To "run-in" my GIGABYTE MB I have been leaving it running playing, in repeat mode, a selection. To save "wear & tear" on the HDD I found I could turn it off and the selected file would keep playing.
Half of the time I would turn off the machine before re-starting the HDD, the other half I would turn it on - just to see how long it took for the blue screen to appear. Sometimes it would not appear at all, the machine would just turn off. I forgot to mention that this blue screen would appear after I selected SHUT DOWN.
Well, the LAST time I turned on the HDD, with the machine running, when I turned the machine back on, it did not come on. It acted just like it did when I installed my first TWO (defective)GIGABYTE boards, nothing happened at all. I am figuring/hoping it is the MB since it is cheaper to replace than the CPU. I thought (I hoped) I had messed up something that could be cured with clearing BIOS. I have tried numerous times, shorting the pins, taking the battery out and leaving it out for twelve hours. Made no difference at all. Will not turn on, no flashing lights on the keyboard; attached to re-implement BIOS settings.
When the machine was left on I had a fan blowing through the "cooler" attached to the CPU, a very large THERMALRIGHT piece. I know there was no overheating. I had the voltage set as low as I have been able to get, .912 volts per CPUZ. All recommended settings per cics.
I have ordered a new MB.
I know one thing, I will never turn off the HDD while the computer is running again.
Any reassuring thoughts about my diagnosis? (could I have ruined the CPU?)
I got to listen to 1.1 for a few hours before this and found it to be a yet clearer view into the recording. Much more space between voices and instruments and simply more definition of the sounds these instruments are making.
I have not had a chance to hear what carcass's setting will do and it is driving me crazy!
cics is not exaggerating, his system has approached the best analogue and surpasses average LP's handily. One can enjoy CD's with no regrets.
Thanks,
Rick McInnis
Got my second one in today which I paid twenty seven dollars for next day air, and nothing.
Every one of these I have received from NEWEGG has been defective. This is my third from them. Finally got one that worked from ZIP ZOOM FLY (if you can believe that name!) that worked. And then I ruined it.
I have ordered one from another supplier.
I do not think anyone is interested in my travails. I am posting my problems as a warning about HOW I destroyed my MB and to not be surprised if an MB from NEWEGG doesn't work.
I have much catching up to do with cics's latest discoveries. Can't wait.
Bye,
Rick McInnis
Got my new MB yesterday and still NOTHING.
Figure I really made a mess.
Please do not turn your HDD on or off while your computer is running.
(Mothers tell your children, not to do what I have done ... keeps going through my head)
Ordered a new e7200 today.
With the introduction of 1.2 I had to request next day air shipping.
I cannot wait to hear cics's latest master stroke. In fact, I have brain fever in anticipation!
Bye,
Rick McInnis
PS Theo, I certainly was not the first to take cics's advice and methods seriously. I got involved last summer at the time of his update to the original ART OF COMPUTER TRANSPORTS paper. Maybe I was one of the first to assemble a machine based completely on the pieces recommended in that paper. I would not have known where to begin without that paper!
'Theo, ...... Maybe I was one of the first to assemble a machine based completely on the pieces recommended in that paper. I would not have known where to begin without that paper! '
well that deserves something akin to kudos. Sorry about your processor.
You powered HDDs from GD PS using its on/off switch?
This could cause damage to your southbridge chip on mobo (through repeated failures). Test CPU by installing your old E1200/E2140 CPU.
or would this only be needed with a different MB?
I am experiencing withdrawl without my cics MUSIC COMPUTER. I could re-insert the THETA but it is unlistenable now. THANKS ALOT, cics!!!!!
Can't wait to have my prized music machine back.
Thanks,
Rick McInnis
don't have any advice for you. I just wish you good luck for reinstalling your system. its not good to be away from 'cics' music a long time.
And I must make a comment/compliment:
It has become very important that you now possess the latest iteration of the music computer - SINCE
It is obvious you are in possession of a pair of very good and well-trained ears.
Your assessments have become an important guideline for me and now that I know we are listening to the same thing I can use them as a benchmark for what I am hearing.
It is obvious that you are capable of detecting subtle differences. Your interest in orchestral music, I feel sure, has also supplied you with much exposure to the real thing and because of that your opinion is important. My time listening to the REAL THING has become skimpy lately. I go to more electric concerts than acoustical these days.
Listening to the ASO in its bland hall and mostly tired repertoire just doesn't interest me much. I should grow up and do it anyway.
I am glad you are interested in this!
Bye,
Rick McInnis
you are too kind and too modest, weren't you the first cplay/cmp graduate? thank you.
Dear cics,
Yes, I was turning the HDD on and off with the GD power supply's switch.
I have a new board on the way. I will test the old one eventually but I suspect what you surmise is what has happened.
I will not do this little trick again. Truly false economy. When the blue screen tells you you could damage your computer I guess it is true. I, previously, thought it was overstatement. Education is never cheap.
Thanks for the advice,
Rick McInnis
Rick,
By your description, you have done nothing that I have not done and without causing damage to the mobo. Forgive the question but have you checked the basics? E.g. unplug and replug power supply harnesses to mobo; remove and refit memory module(s); etc. Are the drive(s) spinning? Can you hear the PSU functioning? I assume the standby led on the mobo is showing green..
Bling,
I have tried another power supply, no change.
I have the HDD powered separately (which led to the problem, I guess since I turned it on while the machine was running?) When I turn it on the HDD makes its whirring sound and slight vibration.
I removed and re-fitted both the memory and the CPU.
The PSU does not come on, which was the case when I installed my first GIGABYTE MB, which was defective. The power supply would not come on.
I do not have any LED's attached to the header but feel confident they will not work.
I have made a floppy copy of the BIOS file but I do not have a floppy disk drive that I can connect to the MUSIC COMPUTER. I will have to pirate one from an old machine at work tomorrow. I hope SEGARS advice will work.
Just to be on the safe side I have ordered another MB.
Thanks for the concern and advice,
Rick McInnis
Rick,
I am sorry for your trouble. I am surprised the gigabyte mobo does not have a power led.. Anyway, it should “beep” if it encounters a problem (see p79 of manual). I am assuming it does not beep else you would have mentioned it. Therefore, a good sign..
USB flash drive will work just as well as a floppy. From manual: “Extract the file and save the new BIOS file (e.g. g31ms2l.f1) to your floppy disk, USB flash drive, or hard drive. Note: The USB flash drive or hard drive must use FAT32/16/12 file system.”
Good idea to remove the sound card (if installed) or any other “add-ons” just in case it is suddenly causing a problem or conflict during power up..
Bling,
Dear Bling,
I wish I could read the manual with your understanding!!!
Can't wait to give this a try.
My appreciation,
Rick McInnis
Is there more to this than sticking a memory stick with the file in a USB port and then turning on the machine, or more precisely, attempting to turn on the machine?
I hope there is, but otherwise, nothing happens, still.
I guess I will have to be patient until the new MB arrives Wednesday.
Nonetheless, thanks for the help,
Rick McInnis
Next time you get blank screen after trying to clear CMOS (5 seconds of shorting the pins is enough) or removing the battery (a minute is enough), you may be able to save the motherboard by flashing BIOS by booting from temporarily installed floppy drive, and DOS utility - see Gigabyte pdf: DOS Utility
If that fails you might try this BIOS Recovery procedure. I haven't had to try it yet, but it looks interesting.
Thanks, again.
... has anyone tested them yet ?
Any improvement for Cmp2 over F6 ?
Thanks
Martin
HKLM\SYSTEM \CurrentControlSet\ Control\ PriorityControl\ Win32PrioritySeparationThis registry value is controlled by switching between "Programs" and "Background Services" in Control Panel--> System--> Advanced--> Performance Settings--> Advanced (see cics' document on building computer transport). In nutshell, it deals with foreground (F) vs. background (B) processes priority, i.e. length of time allocated to either, and ratio (1:1 ... 6:1).
The little twist is that System setting switches between 2 values only - but in fact there are 7 values that correspond to unique combinations. For details, see reference #2 below. We're interested only in those that give both F and B the same priority - 6:6, 18:18, 36:36 (the registry key value for the latter is 0x18, and is the setting that's toggled when you choose "Background Services").
In my system, 18:18 (i.e. registry value of 0x28) gave airier, noticeably more 3D presentation, at some expense of bass warmth - it sounds drier and less in "quantity". Setting 6:6 (0x24) was clearly inferior.
After careful listening, I decided to keep the tweak (value 0x28). Give it a try - you just may like it. There's no need to reboot, this setting takes effect immediately.
References:1. http://www.tweakhound.com/xp/xptweaks/supertweaks7.htm
2. http://blogs.msdn.com/embedded/archive/2006/03/04/543141.aspx
3. http://www.microsoft.com/technet/prodtechnol/windows2000serv/reskit/regentry/29623.mspx?mfr=true
wondering if I set this right. I set the decimal version (which was 24) to 28 but it reads in registry 0 x .0000001c (28). Is this correct?
N/T
no wonder it didn't sound as great as cics said, so I stay in hexidecimal and enter 28. Is that correct?
You know it is a wonder that 28 (entered as decimal) even worked at all but it did and it sounded ok but a lttle dark. When I put 28 in as hexidecimal it was as if somebody lit up the soundstage. It imaged ok before but not like it does now. From a cold start my electronics are a bit bright but once run for 40-50 minutes the upper mids to highs are very very good, very good see-into-ness with respect to the soundstage, now. Lots of air and dynamics are outstanding. An embarassing but learning moment for yours truly.
Rick: I bet I beat you on this (making mistakes I mean).
(I believe 0x2) if what's entered is not valid.
thanks but it actually did read 0 x .0000001c (28) in the registry. II'm wondering what I had there.
If the value that's entered is not valid, it's being ignored, and default value of 0x2 is used - it doesn't mean that invalid value gets overwritten. It's explained somewhere in Microsoft docs.
thanks again for clarifying
Such fundamental changes to OS are important and can lead to less jitter.
Have you done the cMP² update (including "/nopae")? No loss in bass performance.
I'll edit the post to include this optimisation. Brilliant!
I tried it without related power management change when you first posted it, and didn't like it, but that's to be expected.
Will try again today, this time with this tweak in question here, and power management change.
N/T
.
N/T
.
should we scrap nopae (or is it the carcass93 one) also?
Trying it and so far I like it. Yes more air at some expense of overly warm bass. Time will tell if I like it long term. Did you try the 3rd value associated with 36:36?
changing registry manually.
My observation refers to 18:18 (0x28) versus 36:36 (0x18). The other setting, 6:6 (0x24), is inferior to both.
0x28 worked for me too. My system was origninally set at 0x24, I hear more detail with the new setting (0x28). Thanks for sharing.
Change Log (final 1.1 release):
- Change tuning parameters
- Bug fixes:
- Extension in caps was not recognised
- Clipping not done for Phase 180 when not upsampling
- Cue processing error when no title exists for track
- HiRes FLAC decoding error (SSSE3 version only)
- Minor UI change
Please REMOVE previous versions before installing cPlay 1.1 Final .
Both normal and SSSE3 versions are available. For cMP² users, be sure to apply these updates .
cPlay 1.1 @ 176k or 192k gives best performance. At 96k (or less), 1.0 gives best performance. I'm going to release 1.2 that combines both aspects, i.e. tuning parameters optimises to output rate selected. cPlay 1.2 is already operational and should be released next week.
Comparing 1.1 @ 192k to 1.0 @ 96k, 1.1 is better. I'm using Lynx AES16 card to DAC (Dual AES 24/192k). I suspect Juli@ with promethk's mod would be even better. Redbook CD (SRC upsampled to 192k in 1.1) achieves analogue perfection. I always felt AA Prestige SE's tube output gave better midrange performance over Scarlatti DAC. Not anymore.
cPlay cannot detect the asio driver of a Konnekt firewire sound interface and display no asio device found. Foobar2k can detect the asio driver and play through asio correctly.
I tried both versions of cPlay, uninstalling one before the other, and also tried further to install asio4all, which only brings up asio of the motherboard realtek sound chipset.
Any take?
Xonar soundcard has similar issue.
In registry, goto:
HKEY_LOCAL_MACHINE\SOFTWARE\ASIO
Correct ASIO driver installation must have corresponding entries here. ASIO 2 provides API that accesses these reg entries. You could try to manually enter it here (not recommended). cPlay relies on ASIO 2's API.
Hi cics:
I have try 1.1 SSE3 version,after listen about 3 hours,I feel 1.0 SSE3 is better than 1.1 SSE3.
I feel 1.0 is more feeling,good sound stage.
Why?I don't know is there anybody feel same with me?
Possible... not sure yet. But for me, it resolved a very annoying bug concerning 24/96 FLAC files. Thanks cics.
On the other hand, I now get a LOT of glitches (data loss) during playback. Not sure where it comes from. I made a few changes at the same time... (yeah, I know I shouln't do that)
I installed SSSE3 1.1, I disconnected 1 RAM module, leaving one 1GB module (in the slot closest to CPU) and upgraded BIOS to F6x with RAM speed set to manual.
Anyone has an idea ?
to be continued...
Thanks
Martin
do you run cmp? before I switched to cmp I had similar issues with one gig ram and big files run at 24/192. cmp fixed that.
No,I don't install cMP.
I have try install cMP 1.0 on VISTA SP1.But there is a problem,when I want add folder,the dialog bos only show "Desktop",no more other folder.So I just use cPLAY.
Yeah, with or without cMp I get glitches, maybe a little less with cMp.
I put back the 2 RAM modules, and reset my BIOS to default values.
And of course, no more glitches... I'll go step by step, from now on !
to be continued...
thanks
Martin
My concern with the newer BIOS & Power settings is change in SQ over time. Update post now reflects RAM settings to Auto at 2x speed and power settings back to 'Max Battery'.
Dropouts may be caused by CPU clock transitions (which cause a small halt in processing that is heard in low latency setups). This happens mostly at 96k output where clock transitions between 6x and 9.5x.
Try latest update post above and let me know what happens. You may want to increase CPU host freq. Also, don't run other apps like Process Explorer and CPU-Z during playback. cPlay 1.0 is damn good and should this be the case, I'll release 1.2 with 1.0's tuning parameters!
Matched with my warmer sounding dac (tube-buffer, non-oversamp.), the 1.1 is too laid-back with some lost of high-frequency and music details c.f. 1.0. This is a setup with cPlay+cMP w/Juli@ at 96kHz, 200MHz fsb and 6.0 CPU mul. on ASUS motherboard with power filtering. I've implemented almost all of the possible bios/windows tweaks, including the minlogin.
So in my case, the 1.0 is indeed better.
Not ready to make firm judgement yet, but thought I should mention what I'm hearing so far - better balance, with centre of treble shifted nicely down in freq. range - what I've wanted since cplay began! Also silky smooth, better than 1.0. I'm using dual-boot, and set up one with 1.0 and one with 1.1 to compare, and I keep choosing 1.1. Yes, 1.0 is awesome, the prominent treble gives a stunning effect, but 1.1, or an evolution of it, may be best in the long run.
Also I think I'm hearing a new fineness of detail with carcass tweak, especially combined with 1.1, but haven't had time for a-b-ing.
My preference is for 1.1. IMO much better than 1.0 for the reasons mentioned above by seger. (I A/B the two in some back and forth testing that decided the issue for me.) I run cPlay+cMP on ASUS mobo/E2140 through USB to EA Freeway2 at 96kHz. Unlike Promethk, my DAC is not laid back which may account for the different experience to his. Thus far, the best sound I have ever had.Bling,
Idea is to offer 1.0 or 1.1 tuning parameters for 96k output.
E2140 has smaller L2 cache (1MB) and cPlay 1.1 uses a smaller cache 'footprint'.
not getting any dropouts, sonics are stunning. is it still advisable to change bios settings to older values?
Improvement you're getting is from cPlay 1.1 which favors 192k output. Those BIOS settings also favor 192k. My preference is for one set of settings - your test can help here.
jammypan - I suspect you're outputting 96k, please confirm?
Hi cics:
I have try 192k output.but must level up the lateancy to 5 ms.
why not make a set of bios / other set up values optimized for 96 and one for 192 sampling rate? It could be analogous to loading a moving coil cartridge, some like 47k ohms some like 100 or even 10 ohms. I know easy for me to say.
Sorry late to reply(But why I haven't recive a mail notice?forget it,it's not important)
I use RME-9632,So I choose 44.1 *4 =176.4,not 96 or 192.I think the orginal rate is 44.1,so I just multiple it.
I will try 192.Wait for my result!
didn't like auto and 2.0 setting. gave me fat bass, brought soundstage forward, overall sound was midrangy to me. don't take this in an absolute sense. if I never heard (manual 2.5) setting I would like it very much. but you're right I'm listening to 192.
Still believe one setting will work.
On your preferred setup (RAM SPD @ 2.5, TIMINGS @ 3-3-3-9) set following in BIOS (advanced settings, M.B. Intell...):
CPU GTLREF Voltage Ratio to 0.54 (default setting is 0.636)
Its taken a while to understand what this does.
ok I'm trying:
initial reaction: very nice, a bit mellow vs baseline, little less strong (not a bad thing). This is on Sundown by Gordon Lightfoot. I'll let it warm up; bass very nice , rolls out at you on bass guitar, very analogue like. More to be reported as I listen to different stuff.
further listening: I could live with it but bass seems a bit loose, not as dynamic as baseline. A little analogue drippy (by this I mean a bit warm/wet) sounding. I think I prefer baseline which is drier and airier and more dynamic. I can live with this but would have to adjust something else in my system to balance it out (i.e., turn down my bass amp volume).Follow up: I adjusted bass downward by 2 db (warmth region 60 to 330 hz) and its better, still not as dynamic as baseline. After listening to be my big crescendo classical music (with bass down 2 db) I miss the dynamics of the baseline. Where baseline would approach real live dynamics new setting just misses the mark.
sorry could not live with it. I went back to .636Is there a setting somewhere in between .636 and .54 that may be better for bass. I like the highs at .54 they are very smooth and marginally edgy files sound great.
Maybe its my system (or any similar balanced system) but I could not live with .54. After going back to .636 it was like my system was again 'on', dynamics, soundstage, and clarity are so good that it makes me want to listen to music for hours. Because I am retired I do listen to music for 6-7 hours a day.
.
Anyway so far so good. Time will tell. What does this do? I believe it reduces the fsb voltage further. Is that correct? Is it just another source for rfi if too high?I'll post my longer term comments in this post as edits.
Edit: After listening off and on for 8-9 hours I can say I like this setting very much. There appears to be more transparency from the midrange up, however bass is not as tight and impactful as the baseline CPU GTLREF Voltage Ratio (of .636). It is never hard or glassy which is why I like it so much. Depth is also not as great as baseline but it is a good set of tradeoffs, sonically imo. It doesn't excite me like going to ssee4 did (or whatever it was called). But it is very very good for long term listening, very analogue like.
Its used by overclockers for running high FSB clocks and/or FSB voltages. There's a non-linear relationship to CPU stability with no guiding principle to select a specific ratio (except for actually testing it).
Intel uses AGTL+ and a simplistic view of GTL is that it acts like TTL signals but operating at much lower voltages.
In cMP², fundamental design is to create an overall low stress circuit . Hence I would prefer using lowest stable CPU GTL Reference voltage.
Read a little about GTL with help of yahoo/google search. I can't avoid trying the 3rd value--above lowest, below .603 just for kicks. You were right it was an unpredictable result. I thought that it would lineraly fall between my current baseline (.603) and the .54 which I disliked immediately--not so. I'll post more on further edits. This stuff is just facinating.
Going directly into .56 I liked it right away, mids and highs are beautiful and bass is surprisingly better too! After 2 hour of listening I tried going back to .636 and it sounded harsh, etched by comparison to my new baseline .56 GTL reference voltage. It is amazing how non linear the response is at the various levels to choose from that I encourage all to try the different levels of ratios. Amazing and facinating! Nice call cics the level of transparency of cplay/cmp is way more than I ever thought I would achieve ims.
thank you
ok I'll try
Hi,
Could this be a BIOS version problem ?
1 or 2 RAM modules I get dropouts, I even installed cPlay 1.0 back and got those drops. But only on 44.1 files upsampled to 96k or 88.2. Native HiRes 96k files played ok. Can't tell at 192k my DAC doesn't take it.
So for now, it's BIOS F6x, a single 1GB module, cPlay 1.1 SSSE3 @ 145dB-96K, and all the latest tweaks except those under MB Intelligent Tweaker.
Not ideal, I admit, but at least I can listen to music without crackling :-)
Temperature doesn't seem to be an issue, it stays around 34-35C.
So maybe version 1.2 "for 96k" could help in this case.
Anyone tried BIOS F7 or F8 ?
In what order should I try to make the remaining changes in M.I.T ?
Thanks,
Martin
Hi cics:
Happy there is coming new version.I use SSE3 version before.And is there 1.1 SSE3 version?
.
Comments are to be taken in context. Cmp/cplay is still the best sound I have ever
heard from my system. Having said that here is my
feedback on new bios settings, new mobo drivers: very, very tight sound. Clean as a whistle but a little rolled in the highs imo. Once I set my high frequency amp 1 db higher on play back( and after 20-30 hours of break-in), balance was better, but still very very tight sound in both a positive and negative sense. Attack and decay seemed too tight and I had to turn up volume to get what I perceive as adequate hall sound.
Cplay 1.1 to the rescue, where 1.0 was tight on attack and decay, 1.1 is more open, relaxed. Decay now is perfect to my ears anyway. Balance is extemely good from mids to upper mids to highs. Transparency is superb! Don't know if bass is optimum (always takes me longer to get a handle on bass). Initially: overall the best yet. Just listened to several classical movements with big bass crescendo's: spot on. Another step up!
Thank you.
I've edited the cMP Update post with this change (see bottom). Only use "/nopae" in the context described.
Did you try the carcas93 tweak above? Seems to work well with "/nopae"
You should get near perfect tonal decays with midrange purity that's rich with harmonics giving analogue sound with sweetness and emotion. Tubes? Instrument separation is mind boggling - complex passages are rendered with stellar precision. Such midrange performances demands near perfect highs as is the case. Bass has excellent weight and authority.
Hi cics,
since I have trouble loadind "big" cue files in cplay, see post:
http://www.audioasylum.com/forums/pcaudio/messages/3/36161.html
would it be better to simply remove one of my two 1GB RAM and keep only 1 active as it seems to be the "new best setup" with the G31/E7200 ?
Regards
Martin
With just 2GB Windows won't allocate more than 1GB to cPlay.
Documentation suggests greater allocation is done by Windows with more installed RAM (upwards of 4GB). Adding RAM unfortunately worsens SQ.
Have been able to recreate hires flac bug in SSSE3 version. Seems like flac is incompatible with some of the new instructions.
So I'll go with only 1 stick and shorter cuesheets. Looks like the good old vinyl times are back :)
What do I lose by using "standard" cPlay over SSSE3 on this setup ?
Thanks again for your help.
Martin
best highs in the business of reproduced sound
A lot ito SQ. There's an important instruction introduced with SSE3 that cPlay benefits from. I suspect FLAC is getting issues with vectorization done in SSSE3 - plan is to test with SSE3 version.
N/T
.
I'm kind of holding off with the next move (currently P4M900/e2140), and both E5200 and E7200 satisfy my main requirement - support for SSE4.1. In fact, the only significant difference I see between two is cache size. There's no reviews/comparisons yet regarding power consumption etc.
Any thoughts?
At 192k upsampling using 145db SNR SRC, E7200's extra cache is useful, otherwise E5200 is better. Whilst both processors will do the job, the mobo is proving to be excellent.
Full setup significantly lowers power consumption and uses L2 cache as needed! Intel's EIST function not only adjusts frequency and voltage but cache as well. In XP, power scheme must be "minimal power management" to activate EIST.
Have 500 gig plus of classical music WAV and BWF 24/88 files that play beautifully on cPlay but only load as single files which is a pain with multiple movememnt symphonies.
Can one convert them in some way to cue or something else to load them in order?
Bill
I use totalcommander as filecommander for choosing songs to listen. Foobar loads the new song if choose and enter, but Cplayer has to be shut down first, before loading the new song.
Is it possible too change that?
Hi,
the simplest way I found (in fact I was helped by other members here) is to load the files in Foobar then use the plugin "foobar cuesheet creator" and save the .cue as Ansi NOT Utf in the drop down menu. This way you won't have any typing to do.
Don't hesitate if you need more advice.
Best
Martin
Here's a sample multi-file cue sheet (make sure capitals and spacing is exact):
PERFORMER "Tracy Chapman"
TITLE "Tracy Chapman"
FILE "C:\Music\01 Track01.wav" WAVE
TRACK 01 AUDIO
TITLE "Track01"
INDEX 01 00:00:00
FILE "11 - Miroirs - La vallée des cloches.wav" WAVE
TRACK 02 AUDIO
TITLE "Test 2"
INDEX 01 00:00:00
FILE "C:\Music\03 Track03.wav" WAVE
TRACK 03 AUDIO
TITLE "Track03"
INDEX 01 00:00:00
FILE "C:\Music\04 Track04.wav" WAVE
TRACK 04 AUDIO
TITLE "Track04"
INDEX 01 00:00:00
FILE "C:\Music\05 Track05.wav" WAVE
TRACK 05 AUDIO
TITLE "Track05"
INDEX 01 00:00:00
Please can you help me ?
My CMP is asking for foobar ! how can I change this for Cplay ?
I have uninstalled foobar, reinstalled Cmp and Cplay but nothing gave me a solution to my problem.
Thanks for you help
Do step 1 in attached link
http://www.audioasylum.com/cgi/m.pl?forum=pcaudio&n=35658&highlight=cics&r=&session=
- http://www.audioasylum.com/cgi/m.pl?forum=pcaudio&n=35658&highlight=cics&r=&session= (Open in New Window)
do I really need it?
Hello everybody. Been absent much too long and now feel myself lost.
Tell me please, where can I download this 145dB SRC and should I if I use only 16/44 wavs that I rip from my cds.
Also Is CMP documentation still valid as the favourite motherboard is different now. I mean bios adjustment and windows castration.
Serge.
Link below provides all changes beyond cMP and AOB Computer Transports 0.3 documentation.
Copy "cicsMemoryPlayer.pth" file from cPlay to cMP - this tells cMP to start cPlay instead of default (foobar).
As always, thank You very much for the answer and great work, that I really appreciate. The information and solutions You provide help me to enjoy my music while away from home and help to "popularize" the idea of hi-end computer playback among my friends.
Serge.
Does that solution - externally powered USB ports connected to mobo header - offer any advantage compared to powered USB hub?
I guess answer is yes, if nothing else just because all mobo ports can be disabled - but is there more to it?
If power lines are still used from mobo, then go with externally powered USB ports.
Hubs seem to activate different USB driver in Windows.
N/T
I tried searching here, but didn't find sufficient details.
How do you use external power supply for Logitech wireless mouse's USB receiver? I mean, I understand how it could be POWERED (I use external PSs for flash drive and monitor, and also linear regulated one for Asus PCI-e soundcard), but how is device connected to the system (mobo)?
There's mention of "alternate" USB ports in your posts, implying those are not mobo ports - what are these?
I'm using wireless RF remote, and would be interested in trying the same approach you did for the mouse.
Thanks in advance.
He uses the USB ports on the MB as you will see in the link.
so it's blocked. Just like e-mail sites, porn sites, some forums etc.
Dear cics,
What do YOU think of such things?
One can spend up to $1600.00 for a pair of these things before they decide on how they will power it.
Worth a try?
I see BURSON makes one for about $150.00 and based on their other offerings is probably an excellent value. Still one needs two.
Lowered the voltage again, not yet to 1.025, (will try that tonight)
stable and sounding magnificent.
Thanks,
Rick McInnis
CPU voltage to 1.00625V is possible on old mobo version.
I'm working on cPlay 1.1 with some additional BIOS changes that could help. OS jitter is a bigger deal than Juli@'s XO. There's a relationship between OS jitter and Audio jitter.
Hi cics,
I think I found a bug in cplay. Recently I had problems with FLAC 24/96 files that wouldn't play properly. I had to manually convert them to WAV before loading them in cplay, which was not convenient or logical. I accidentally loaded one in my second computer only to find that it worked. That computer has cplay 1.0 standard version.
So the bug only happens with the SSSE3 version.
I installed the older version in my Gigabyte/E7200 and it now works.
I tried again SSSE3 without success, and back to "standard"
So I assume both cplays don't decode and load files the same way ?
What is the drawback of using the "standard" version ?
Since you are currently working on version 1.1, I thought you might be interested by that issue.
I also have problems with RAM allocation. Certain cuesheets sometimes load and sometimes don't. I didn't yet figured under which circumstances.
The one I last tried to load I got the
"Memory Allocation Failed" message.
where physical RAM is 2037M, Available RAM 1831, and required RAM 1285.
in diagnostic.
I tried with AWE on or off.
Can you help me on that one ?
Best regards
Martin
I'm ecstatic you are working on something new. You are a blessing to the audiophile community.
did not take your email address home with last night
AND THEN
forgot to bring my memory sticks to work.
You will hear from me tomorrow.
Rick McInnis
Hope you don't mind me comparing you to valet, but Jeeves was no ordinary valet.
And cics you are certainly a marvel.
Can't wait for the latest.
And THANKS for steering me away from fancy clocks!
THANKS, again,
Rick McInnis
To those that are interested:Inspired by promethk's JULI@ investigations I went to work.
I noticed that not only is Pin 5 a ground so are Pins 4 & 8 which makes installing the capacitors much more convenient. Pins 2 & 4 for one cap, they are next to each other, the same for Pins 8 & 10.
Knowing I will never use the analog part of the card I bent the pins to make it easier to solder the caps. I used two Black Gate NP 100uF 50 Volts that I had. No plastic or otherwise bypassing of the BG's.
Being unforgivably unscientific I made the boot.ini change, along with the lowered voltage to the CPU, and this change, all at once. I have no idea which is the most important. I do not hesitate to say that all should be done.
The bypass is an established good electronics practice, especially in audio equipment.
There is increased information regarding image specificity. Voices have additional focus (or should we say the microphone's focus is revealed?). The illusion of width and depth is improved and not slightly.
The sound is beautiful, simply hard to believe this is being produced by a computer. Bass with power and tonality that would be impossible to experience in most rooms with a turntable unless you were on a concrete floor and using the tip top echelon of turntables. I am assuming this. I have never heard bass like this with any turntable I have owned or heard it in someone else's system.
There is a gap being closed.
Maybe we are being tricked with the upsamplers? Maybe with time I will come to the point where I find this system impossible to enjoy? Who knows? Maybe such an event is a possibility? At this point, I might be under a spell and cannot be relied upon to give honest assessments because I think this thing sounds incredibly good. Stunning it remains.
With THANKS to cics and to those intrepid investigators, sonics, segar, and promethk,
Rick McInnis
Wonderful!
Earlier, I have some concern over choosing which of the ground pins to connect the caps as I fear that there maybe multiple grounds for different paths (e.g. analogue ground and digital ground should not be mixed). Though in this digital board, I could be worrying too much?
The bass improvement is always very dramatic on each stage of refinement in my computer audio server, so much so that my smallish listening environment is too overwhelmed by the "new" quality bass contents in the music. I had to swap out my floorstanding speakers and reconnect bookshelves speakers that have been put in storage for years.
Good for you that you can consider the system done.
I am still being stuck at not able to lower the voltage of my E7200 and the new 45nm intel E5200 with 2M cache has started shipping ...
Right now I have got dozens of Sanyo solid polymer low ESR caps and I am still wondering how much am I going to put these into the computer and where are the most effective positions.
Dear promethk,
There is a misunderstanding. I certainly do not feel I am anywhere near done. I assume you took my closing of the gap comment to mean this when I meant the gap between analogue and digital musical reproduction is closing.
Every time I think cics has finished his project he comes up with something new that re-establishes what is possible. The man is never satisfied and for this I hold him in the highest regard!
I figured the pins next to the voltage rails would almost HAVE to be power grounds, and in combination with the convenience offered, I went ahead and used them. All of the ground pins show continuity with the appropriate pin of the regulator, though I certainly agree it is good practice to keep them separate.
Sorry to hear you are having trouble with the lowered voltage to the CPU. I have been running my board constantly playing a music file, and wonder if this (as suggested by cics) has something to do with getting it to work.
I keep slowly lowering my voltage.
Again, thanks for your recommendation and your help,
Bye,
Rick McInnis
congrats and well done!
Dear cics,
Wondering if there is any chance that these would improve what we are using.
Of course, I have no idea so I had to bother you.
They seem to be new this month.
Thanks,
Rick McInnis
.
I also try update my RME 9632 driver to 3.068,but after upgrade,something get wrong.It's not normal status when use cPLAY play song.So I uninstall the newest driver.
I feel 3.066 version is better than 3.067/3.068.
RME changed memory management from 3.067 onwards. Previous version caused BSOD errors.
No sonic improvement though.
Hi cics:
RME 3.069 driver comeing!,so soon!
I try it
Hello:
update again!I reinstall RME/9632 3.068 version driver,and it work good!
and I find there is a new option,but only support VISTA.
hi
some questions belonging actual developments:cics
you mentioned the lynx SC and i am thinking about to compare it with my juli and asus xonar d2x PCIe Soundcards.The new lynx 16 PCIe is very intresting. The first PCIe Slot connects directly to the northbridge and the RAM, so there are no IRQ Problems. Would you give it a try?rick,
you have done the condensatormod with the antec earth watt.
Any results in SQ ?
thanks
alfred
Dear Alfred,
That is kind of how I have to approach your question.
I think I am hearing improvements, and based on what usually happens with a judicious addition of power supply bypass capacitance there has to be some improvement.
As I said in my previous post I fear that when one EXPECTS to hear an improvement one usually does hear an improvement, either real or imagined. I think I hear additional spaciousness, which is something I think I am sensitive to and important to my enjoyment of recorded music.
I know I am glad I made the changes and figure I will go back and add 1000uF capacitors to the remaining voltage rails and utilize all of the remaining grounds. I think if bypassing is good it should be done completely since I have no idea what is happening beyond the header into the motherboard.
All caps take time to settle in. I have left the machine running while repeating an album when I am not listening. I figure there will be additional improvement as time goes by.
Of course, the biggest problem is that unless one purchases another supply it is impossible to go back and forth to see what has happened.
I am very excited about implementing promethk's idea for power supply bypass for the JULI@ card. Something else to do this weekend.
Thanks for sharing your ideas and experience,
Rick McInnis
Rick,
i had the possibility to verify the improvements.
I mentioned the harsh sound of the asus xonar in the beginning. At this time i used the antec without mods because i believed in the high quality of this powersupply.The smoother juli was in another PC (enermax-PS,) with PS-modification.
After providing 20 caps to the antec sound is exiting with the xonar soundcard.No harshness ,tremendous dynamic.
I suggest doing also the modification of the juli, but with some more caps. The xonar has its own header and you can put the caps direct before the soundcard with an easily audible improvement.
bye
alfred(
Both are very close - sorry can't make a call. Lynx does edge out RME HDSP9652.
Not keen on using PCIe as it runs off northbirdge. Northbridge chipset has much greater traffic. Also PCIe does not like high transaction volumes with small payloads like PCI. PCI slot 1 has no interrupt sharing and runs off south bridge which consumes less power and is mostly available for audio output, i.e. not busy with tasks like video and RAM traffic as nothbridge.
That's the theory... best is to test it.
Hi cics,
I have a couple of 24/96 stereo FLAC files I wanted to try in cplay, and I can't play them. They load in RAM via a .cue or individually but I only get noise like totally un-sync digital noise, not just a few cracks...
I can play those same files in Foobar with or without the SRC on, so I guess both my juli@ and my DAC can handle them.
I converted one to 16/96 and was able to play them.
I tried cplay at 121 or 145db snr and different sampling rates.
Is cplay designed to play 24 bit files ? I tried with both CPU and RAM on their normal settings (no underclock or undervolt) and with the same result.
Any idea ?
Thanks again for all the efforts you put in that project.
Martin
cPlay won't check if the decoded wav file is a compressed one. If so, you will get noise. You can check this using FLAC Front-End tool by doing a straight decode and try playing the resulting 24 bit wav file.
Hi cics,
as usual your advice was right on!
Still, I'm not sure I understand...
I converted one Flac file to Wav, with dbPoweramp- I tried both "same as source" setting and to force "24/96/2channel" before loading in cplay and both played perfectly. I tried to convert the Wav back to Flac with compression level 5, (the one I use for 44.1/16) and back was the noise.
Is it that cplay doesn't load FLAC 24/96 correctly ?
I could live with that as long as it plays Wav, but those 24/96 Wavs are huge ! But really worth it.
I guess I should leave cplay settings to 145db snr and 96k when I play those files ? Is the cplay upsampling bypassed automaticaly?
My DAC doesn't do 192.
best
Martin
FLAC decoding was tested on default FLAC settings (level 6).
cPlay decodes directly into RAM before playback. Only benefit FLAC brings is reduced HDD space.
To any interested:
This weekend I modified the ANTEC EATHWATTs supply mentioned by cics because of its exemplary low ripple voltage.
Taking Alfred's advice I added 14 1000uF Elna Silmic capacitors to the various voltage rails, the most capacitance being used on the 12 volts rails, 8 (mounted on perf board) at the four pin connector near the processor, the remainder used on various pins for the main header. To err on the safe side (?) I bypassed numerous positives and negatives. I did not bypass the electrolytics, just because I believe in the gospel of J.C. Morrison that if the cap needs a bypass you should not use it. My electronic beliefs are very eclectic, no absolutism here, my thievery is diverse!
Of course one always fears that corollary to BARANEK'S LAW (the one about how the loudspeaker you make yourself will sound excellent to you) will be in effect and one can only praise their handiwork, but I sincerely think it is an important refinement.
While fiddling about with the thing I removed the extraneous wires from inside the box which makes for a much neater installation.
I wonder how much this supply, at this power level, needs the fan? I want to disconnect it (easily done) and see if it makes any difference. From where I sit (about 18 feet from the box) I cannot hear it but it is part of cics's teachings that all motors are best eliminated and I think he is right.
My thanks to Alfred for bringing this up and his help in showing how this can be done.
AND, as always, my effusive, pandering, sycophantic, heartfelt and sincere THANKS to our brave, intelligent, and most importantly, music loving leader for pointing the way, a man with a very fine pair of ears, and a generous heart, cics.
Rick McInnis
I also did a similar thing but in a smaller scale. I only coupled 4 blackgates caps on main connectors and two caps. to to connectors to the 2.5" harddisk.
The results is also a 2 steps up 1 step down. There being better dynamics with more extended highs and lows. However, there is a tint of brightness in the voices, which prompts me to try the soundcard power mod and balanced power for the DAC. However my results could also be due to the natural voice of the blackgate or that I have not put enough caps on the 5V supply lines.
If you have bypassed the BLACK GATES with something else, I wonder if that could be your problem with the brightness. Of course, they do take a while to settle in and it is easy enought to leave the machine on with music playing for them to quickly settle. (Improve while you sleep)
J.C. Morrison, the respected tube amp designer and philosopher made the argument for NOT bypassing, and many others have come to the same conclusion. I think it is one of those cases that it does intially sound "better" but with time something begins to annoy you. I have long found that there is a brightness associated with this. I must say I do not think a BG needs bypassing, but my opinion is most unhumble so it is also easily dismissed.
(I considered using BG's for my mod, and I would have liked to have used them, I decided it might not be worth the money)
I will probably use the BG nonpolars for my JULI@ card supply bypass. I have them laying around. 100 uF 50V, maybe I will use two on each rail now that I know how to do it (thanks, again).
I think one can easily overkill with the extra capacitance. I would rather use too little than too much, especially in this situation where speed is all important, an easy two orders of magnitude more bandwidth than even the widest bandwidth audio.
I am not very scientific so it is not as if I added a capacitor to each rail separately and listened for the difference. I will probably eventually, though, place a capacitor on each of the supply rails for the 24 pin header. I have no idea what happens after the power leaves the header and enters the MB. Maybe they are ganged or maybe they go to specific areas of the board? As I said, might as well play it safe. Now that I know I can do this and it did not blow up!
Your comments and experience are appreciated.
Thanks,
Rick McInnis
Rick McInnis,
I agree very much with you that bypass could be bad and at locations like the main motherboard cords, it probably did not pay to do bypass. Thus same as you, I did not bypass the caps in the motherboards. I only have six BG, and they are much quite smaller in size than yours, ranging from 220 to 470uF. They have been breaking-in for quite some weeks. I pulled two of them out from the motherboard to do the Juli@ tweak.
Yes, figuring the optimal tweaks for the motherboard supply is indeed worthwhile, though I seems to make the wrong bet on focusing much more on the 12V where both the CPU and core components depends on. So little time and in my case, so little caps too! If someone wants to try the tweak, my current bet is more on the 5V in the motherboard, as other soundcards could also be running on the 5V.
Indeed for the Juli@ bypass, the highest speed could be best. BG non-polar sounds like a very good choice ! I could have already put them in if I had some here. I do have various metalized polypropene caps, Mundorf supreme, Solen MKP, PIO, SIO etc. However, all of them are so big such that direct mounting on the cards seems impossible. The other option is to placed them on the side via connection wires, though I am not sure if the benefit outweights the wire inductance and EMF etc.
For the Juli@ tweak, I have inserted some teflon insulation tubes into untweaked pins to insulate them, this may have misled viewer of the pic into believing that the cap's wire goes into them while in fact they are not connected to the caps but just covered by teflon tubes. The telfon tubes are needed as I have remove the plastic cover of two of the wire connectors due to the tight space.
One last thing, as there are several ground pins, I may not be using the correct ground pin for this tweak as some circuit could have more than one ground. Maybe someone could enlighten me on how to distinguish the different grounds in the Juli@.
The credits should go to cics for leading this venture into SOTA computer audio and all inmates like you, seger who shares and thus bring us forward!
Many thanks!
promethk
Dear promethk,
Your investigations have inspired many I am sure. You certainly inspired me.
No question, cics is the Sherlock Holmes of computer audio. Not only does he possess an inquisitive mind, he has much training and expertise in the medium. Where I speculate and guess, my mind operates as if this is some kind of spiritual realm, he knows what is actually at work. (I wish we could do something for him)
I want to make sure Alfred/sonics gets his due in bringing to our attention what can be done to improve computer power supplies. Of course, I know there is more to come. Cleaner switching supplies and/or conventional power supplies, there is much potential here.
This approach gets more exciting and better every day.
My thanks to all who have shared their thoughts and investigations on this, the greatest thread on AUDIO ASYLUM.
Sincerely,
Rick McInnis
hi,
there are different ways to audiophile heaven.
IMHO i think that one of the most important qualities of condensators ist low inner Resistance for a fast and powefull current delivery. You can do it by using low ESRs ( Blackgate etc.) and dividing resistance by doing many Condensators parallel.With 4 condensators you arrive at 1/8 of the resistance compared to one.
In amplifier PSUs best results i got with big capacitances (30000µF 4-5x parallel) and adding very little bypassing cond.Best very smooth results i got with styroflex 1000-3000nF,.Good results also with normal vaccum tubes, used as condensators(Kathode and anode are doing very little capacitances and deliver very fine resolution). Glimmer condensators are often harsh sounding.Bypassing with bigger mundorf cond. 0,1 - 1,0µF never brought any advatages and often result in a bright sound.
Another problem are switching PSUs in PCs. Till now i use little capacitances because i dont't know if PSU is strong enough loading the condensators to start the PC.Perhaps i try it sometimes.
bye
alfred
.
Dear cics,
The ones I could get to without fear of damaging nearby components I desoldered others I clipped as close to the board as I could.
It makes for a much cleaner installation and has to eliminate some antennae effects.
What do you think about disconnecting the fan? It seems to blow a minimal amount of air. Do you think its removal would offer any sonic advantage other than a reduction in ambient noise? (which in my case is imperceptible from my listening sofa)
I made the changes you recommended yesterday and all is very stable. I end up with a slightly lower "idle" reading on CPUZ but then I always have, for some reason.
One thing worth mentioning, when I compare my CPUZ screen with yours I see I have something showing in LEVEL 3 cache where your screen is blank. Does this mean anything?
All is sounding very fine. I feel very lucky to have this wonderful device.
Thanks!!!
Rick McInnis
I suspect CPU-Z may be corrupted or more likely, BIOS is the problem.
Reset BIOS from CMOS - see page 29 (CLR_CMOS) of GB manual. Be sure to read the notes!.
Thereafter, in BIOS do "Load Fail-Safe Defaults" followed by "Load Optimized Defaults". Only then reapply BIOS optimizations.
Hi everyone,I think it would be useful for everyone to post the results they had with the Gigabyte motherboards so we can figure out where some of the problems come from.
I am using the GA-G31M-S2L, with the E7200 cpu, clock at 180mhz, cpu voltage at 1.000, running at 121db SNR, no windows optimizations, no cMP yet, BIOS settings as recommended by cics, running the Juli@ at 192k and having no stability problems whatsoever. My CPU core temp is stable between 38C and 41C, memory shows 3-3-3-9. I can only imagine things will improve with the windows tweaks.
At the moment not looking much at the sound quality, just want to make sure the hardware is as stable as I can get it. Am also running 2x1gigs Kingston HyperX as I need the additional memory for long pieces (such as Schostakovitch's 7th Symphony which runs 90 minutes or so). XP pro is installed on a 4gig Compact Flash card in an SATA adapter, and the music is on a 2.5" SATA hard disk. Everything except the motherboard and CPU is powered by a separate external PC power supply.
My friend Andre has basically the same setup except for the external power supply, and he barly manages to get the motherboard to boot at 205-210mhz. At 215-220mhz it seems stable enough but is not good enough for running cPlay. We did some tests swapping parts between his and my setup and have determined it is definitely the motherboard at fault. At the moment we have no idea if he got a lemon or if that particular batch has problems. Another friend with the same board, cpu, and memory has no problems running at 180, but if he tries to lower the cpu voltage just a bit, the board won't boot.
So this is essentially where we are with this adventure.
Thanks cics for all the work you've put in so far, it's only fair that we give you as much input as possible. I am also curious as to just how many people are running cMP & cPlay, and on what hardware.
Frodan
Unfortunately, revision numbers are the same. Here's the old version:
New mobo has different caps around CPU. Aluminium caps (x6) next to chokes are retained but the 4 seen in pairs in top right corner are replaced with 3 Al caps.
CPU voltage for old mobo must be above 1.00000V (but not 1.0V). New mobo is stable at much lower voltages (e.g. 0.89375V).
Dear cics,
And yet I had the problem with trying low voltages.
I re-installed the older version of CPUZ and it shows nothing for Level #3 Cache so there must be something funny about my "newest" version file.
I am showing 1.014 at idle. I figure I can slowly go lower and see what happens.
The system sure is sounding good. I am going to try some bypass caps on JULI@ tomorrow.
THANKS,
Rick McInnis
Dear cics,
And all is stable. To get here I entered 1.0375.approx. I can't remember the exact number. My system always had a lower than expected idle voltage. I figure when I put in 1.000 it must have resulted in something REALLY low. Since it would not boot there was no way to see what reading CPUZ would have given. This is slightly strange, I guess?
I put the / nopae in boot.ini file.
Should I try to go any lower with the voltage?
Sounding very good. I do think I am hearing even better "attack", the leading edges are sharper and even more in time with what follows.
Thanks,
Rick McInnis
.
Dear cics,
I guess run-in time might have something to do with it?
Of course, when I was having trouble I had set voltage to 1.00000.
All is sounding extremely good.
THANKS,
Rick McInnis
Hi,
Did I miss a post somewhere ?
Is there something about a specific juli@ caps mod, other than the "sonics" mod ?
thanks
Martin
The link will take you to the discussion.
http://www.audioasylum.com/cgi/m.pl?forum=pcaudio&n=35497&highlight=promethk&r=&session=
- http://www.audioasylum.com/cgi/m.pl?forum=pcaudio&n=35497&highlight=promethk&r=&session= (Open in New Window)
I have the new version, need to check my 2 friend's boards now. Thanks
Frodan
Now that I feel somewhat comfortable going into bios what level do you prefer? Is it still around 1.00 at idle?
...ask your friend to set the memory divider to 1:1 (factor of 2) and reduce the CPU-oltage to 1 V from the beginning on, as well reduce fsb-voltage ( those setting you find when you press ctrl-f1 in the bios setup)
I'm running GA-G31M-S2L, with the E7200 cpu,clock at 200mhz, cpu voltage at .976, running at 145db SNR, with AWE, no cMP yet, BIOS settings as recommended by cics, running with asio4all @ 24/96, Kingston 1 G memory @ 3 3 3 9 and having no stability problems . Still don't have the hd running externally, still have fan on processor. I'm doing Windows optimizations step by step. After Windows optimization will do juli installation then cmp sw installation. Later and lastly will do external gd ps and fan disabling.Only issue I have (it's more of an annoyance) is that when I run cpuz it stalls out cplay (won't continue playing till I exit cplay and restart).
hi,
here are some Photos from the way of adding the condensators to the several voltage headers.I used low ESRs with a capacitance of ( added) 15000µF per voltage ( 3V, 5V, and 12V)and some little styroflex with 3900pF.
Additional i did some to the powerstrip of the xonar d2x, which sounds now a lot better than esi juli ( sorry).More substancial, more and tighter bass and higher resolution.
I use the digital out for the stereo loudspeaker and the analog out for the dual subwoofer. The analog out is also very good .bye
alfred
Edits: 08/14/08 08/16/08
- eeePC v. MINLOGON - rickmcinnis@dogwoodfabrics.com 11:37:42 08/15/08 (4)
In Reply to: RE: cPlay - the open source high-end audio player using ASIO posted by sonics on August 13, 2008 at 09:36:16
Dear Alfred,
Concerning the above: have you compared this to MINLOGON?
I would like to know.
This weekend I get to work on the added capacitance!!!
Thanks again,
Rick McInnis
- RE: eeePC v. MINLOGON - theob 05:54:08 08/26/08 (2)
In Reply to: RE: eeePC v. MINLOGON posted by rickmcinnis@dogwoodfabrics.com on August 15, 2008 at 11:37:42
rick did you implement minlogon? this is my last optimization to do and I'm a little unsure of the steps. Also since I have xp pro (not sp2 or 3) I'm not sure it is feasible.
Edits: 08/26/08
- whether SP2 has anything to do with it ... - rickmcinnis@dogwoodfabrics.com 12:56:08 08/26/08 (1)
In Reply to: RE: eeePC v. MINLOGON posted by theob on August 26, 2008 at 05:54:08
I do not know.
I get the feeling that cics thinks that SP2 is something needed.
I have also heard that sometimes when one installs SP2 into a non-SP2 set-up that problems can arise.
You must hope that cics will give his opinion on this.
I can tell you that if you do everything right MINLOGON is easy but I must add that I screwed it up the first few times, just like with JULI@, so I got lots of practice.
If you mess up you will have to re-install WINDOWS.
Did you put WINDOWS on a separate sector of your hard drive? If you did not, maybe you should start over and do this. It makes making a mistake simple. Otherwise, you end up replacing your music files when a reinstall is required.
When you are implementing MINLOGON, I discovered in my system DO NOT put anything else in the name, simply minlogon. No dot anything!!!
Just proceed carefully and all will go fine.
Another thing that might help: create your registry change files on another machine and save them on a memory stick. You are going to need to download the registry editor anyway. Use the text as supplied by cics and you are assured of it being correct.
If you are not in a hurry I will send them to you this evening. They are in my home machine. I could send you the registry editor program, also.
When doing the registry editing if you have ANY suspicion something has gone wrong, leave the program and start over. This will help eliminate the chance of re-installing WINDOWS. If I can send you the modified file, I will. But, it may not be allowed as an email since one can do much mischief with this kind of thing. I will try.
So, if you are not in a hurry, I will send this stuff to you this evening.
Bye,
Rick McInnis
PS Theo, I am impressed that you installed the BNC!!!
- RE: whether SP2 has anything to do with it ... - theob 13:44:38 08/26/08 (0)
In Reply to: RE: whether SP2 has anything to do with it ... posted by rickmcinnis@dogwoodfabrics.com on August 26, 2008 at 12:56:08
thanks rick. I'm still hesitant about minlogon. yes the mechasnics are complicated enough but because I don't have xp pro sp2 I'm wondering if its even feasible. I'm impressed you got minlogon to work (it doesn't matter how many times you tried). did you get a big sonic improvement with it? Yes if it is not too much trouble I would be happy to receive your files.
Hopefully cics can comment on this.
- RE: eeePC v. MINLOGON - sonics 03:01:25 08/26/08 (0)
In Reply to: RE: eeePC v. MINLOGON posted by rickmcinnis@dogwoodfabrics.com on August 15, 2008 at 11:37:42
hi,
this mini windows xp includes the minlogon.It takes abaout 20-24 MB inthe RAM. Whats about Your RAM-load?
bye alfred
- A few further thoughts & questions - rickmcinnis@dogwoodfabrics.com 08:50:11 08/14/08 (3)
In Reply to: RE: cPlay - the open source high-end audio player using ASIO posted by sonics on August 13, 2008 at 09:36:16
Alfred,
The one thing that I wonder about is what happens after the header/motherboard?
Are all of the 12 volts leads in parallel? One wonders if one should add capacitance to each 12 volts lead if they are being directed to specific parts of the board. Of course this question is the same for the 5 volts, 3.3 volts and the ground leads.
If they are all the same then your elegant approach is the way to go, but if this is not the case, would it be better to distribute the added capacitors among all of the leads?
I can see the sense of using multiple (flexible) leads istead of one large lead for convenience sake. Nonetheless, the way the wires are distributed within the header does not make this an obvious conlusion.
I am sure they are all connected to the same place in the power supply but fear that is not the case at the motherboard.
I would appreciate your comments.
Thanks, again,
Rick McInnis
- RE: A few further thoughts & questions - sonics 11:11:56 08/14/08 (2)
In Reply to: RE: A few further thoughts & questions posted by rickmcinnis@dogwoodfabrics.com on August 14, 2008 at 08:50:11
rick,
all voltages are parallel, but distance to the mobo matters. So i did condensators for the 3V and 5V near to the 24 Pin Haeder of the mobo and the condensators for 12V near the 4Pin header which feeds the CPU.Be aware of providing some little capacitances bypassing .
Some thoughts to the xonar S-card:Very dynamic , much information but till now a littlebit harsh in comparison to the juli. So i hesitate to recomment.The PC is still a black box, little changes create heaven ,others suicide.
greetings to all obsessives
alfred
- Thanks again - rickmcinnis@dogwoodfabrics.com 13:16:44 08/14/08 (1)
In Reply to: RE: A few further thoughts & questions posted by sonics on August 14, 2008 at 11:11:56
Dear Alfred,
Thanks for the clarification.
I look forward to hearing more about this soundcard.
Does a PCI EXPRESS card have its own power supply header? I have never seen one. I got my EA430 power supply yesterday and noticed (I had cut off the extraneous connectors to my current supply) that there is a PCI EXPRESS power wire (a 6 wires connector). I know my board has no place for this to go so am I assuming correctly that this would be attached directly to the card? This would obviously make it simple to poer the card spearately which would help me understand what you and carcass93 have been talking bout. Excuse my ignorance and thanks for all of the help!
I am grateful you are even more obsessive than me.
Thanks,
Rick McInnis
- RE: Thanks again - sonics 13:28:22 08/14/08 (0)
In Reply to: RE: Thanks again posted by rickmcinnis@dogwoodfabrics.com on August 14, 2008 at 13:16:44
rick,
xonar is fed by a floppy header, which provides 5 (red) and 12V (yellow).
So it is simple to do an amelioration. I think it is sufficiant to do the condensators to the leads ,because of the good quality of the antec earthwatt instead of an additional PS.So it keeps things simpler.
Meanwhile i try to find a way between the more laid back of the juli and the dynamics of the xonar .
bye
alfred
- Very well done, Alfred - rickmcinnis@dogwoodfabrics.com 20:08:50 08/13/08 (0)
In Reply to: RE: cPlay - the open source high-end audio player using ASIO posted by sonics on August 13, 2008 at 09:36:16
Dear Alfred,
I knew a picture would help since I had not thought of doing it the way you did it, which, of course, now that I see it is the OBVIOUS way of doing it well.
Thanks very much. Now I can get to work with confidence.
I like those BC capacitors, too. I am using ELNA SILMICS, just thought I would give them a try.
I am going to look into the soundcard you mention.
Thanks for the help and inspiration,
Rick McInnis
- RE: cPlay - the open source high-end audio player using ASIO - frennekamp@netcologne.de 14:56:50 08/13/08 (0)
In Reply to: RE: cPlay - the open source high-end audio player using ASIO posted by sonics on August 13, 2008 at 09:36:16
Great stuff, can you add some more picures...? How did you do the power-setup of the soundcard ?
- Reducing CPU temp (and power consumption!) using Gigabyte GA-G31M-S2L mobo (this is work in progress) - cics 12:16:51 08/12/08 (18)
In Reply to: RE: cPlay - the open source high-end audio player using ASIO posted by cics on May 05, 2008 at 12:31:58
Gigabyte's newest range of GA-G31M-S2L mobos have changed how voltage is applied. Fortunately, there's ample adjustment to achieve great results!
This post suggests setting voltage to Auto. Vcc reduces to 1.125V (from 1.250V).
I'm testing the following which gives lower (over 10°C) CPU temperatures (allowing for fanless CPU cooling at 24/192).
- Enter BIOS, at main menu Press [ctrl]+F1 (this gives access to advanced BIOS options)
- Select MB Intelligent Tweaker (M.I.T.)
- Enable CPU Host Clock Control
- Set CPU Host Frequency (Mhz) to 170
- Set System Memory Multiplier (SPD) to 2
- Set System Voltage Control to Manual
- Set FSB DeOverVoltage Control to -0.15V
- Set CPU Voltage Control to 1.00000V
Lower voltages result in less power consumption and less heat - this is confirmed using RealTemp utility (or BIOS PC Health Status). Last option (CPU Voltage Control) allows for even lower CPU voltages (down to 0.50V) - default is 1.125V.
I plan to try lower voltages - do not use high voltages above 1.1250V (to avoid CPU damage). It would be great if someone could do power readings of PC using above settings.
- FINALISED: CPU voltage 1.04375 works best. CPU-Z shows Vcc of 1.008V @ idle. - cics 02:50:21 08/16/08 (1)
In Reply to: RE: Reducing CPU temp (and power consumption!) using Gigabyte GA-G31M-S2L mobo (this is work in progress) posted by cics on August 12, 2008 at 12:16:51
This works for both mobo versions.When temperatures (using passive cooling) are ~60°C, do following BIOS changes:
- Main BIOS menu, press [ctrl]+f1
- Under advanced BIOS features, disable:
- CPU Enhanced Halt (C1E)
- CPU Thermal Monitor 2 (TM2)
- Under PC Health Status, disable: Thermal Throttle TEMP
You should get better transients and lower noise floor. Overall performance improves.
If CPU temperature remains high, try lowering voltage. For newer mobos, CPU voltage of 0.89375 works (further lower voltages not tested).
Edits: 08/16/08
- For new version of mobo, use CPU voltage of 0.92500V - cics 02:28:09 08/20/08 (0)
In Reply to: RE: FINALISED: CPU voltage 1.04375 works best. CPU-Z shows Vcc of 1.008V @ idle. posted by cics on August 16, 2008 at 02:50:21
This reduces temperatures and power consumption.
ONLY DO THIS IF YOU KNOW ITS A NEW VERSION MOBO OR ALTERNATIVELY YOU'RE COMFORTABLE DOING BIOS RESET FROM CMOS.
- Setup for older version of Gigabyte GA-G31M-S2L mobo - cics 13:01:19 08/13/08 (9)
In Reply to: RE: Reducing CPU temp (and power consumption!) using Gigabyte GA-G31M-S2L mobo (this is work in progress) posted by cics on August 12, 2008 at 12:16:51
Both Rick & Gagnem encountered problems using suggested setup.
On my old mobo, same problem occurs and BIOS must be reset via CMOS. This mobo prefers a higher CPU voltage setting. All other settings are fine including FSB @ -0.15V. CPU voltage needs to be higher than 1.00000V. In my setup, I get stable results using 1.0125V (which gives CPU-Z Vcc of 0.976V @ idle). Should this voltage not work, try a higher level (CPU voltage of 1.10000V should definitely work).
- Before setting bios you mention ... - theob 04:25:38 08/14/08 (7)
In Reply to: RE: Setup for older version of Gigabyte GA-G31M-S2L mobo posted by cics on August 13, 2008 at 13:01:19
...perform Load Optimized Defaults several times. Since I am a green newbie at this, what is this? Is it section 3.0 in AOB or is it a menu selection in actual bios?
- RE: Before setting bios you mention ... - Frodan 12:38:03 08/14/08 (6)
In Reply to: RE: Before setting bios you mention ... posted by theob on August 14, 2008 at 04:25:38
It is a setting in the first page when you enter the BIOS. It changes the BIOS settings to what should be ideal for that particular combination of board, cpu, and memory. Then you go into each menu and tweek the settings as per cics' instructions.
Frodan
- RE: Before setting bios you mention ... - theob 14:31:20 08/14/08 (5)
In Reply to: RE: Before setting bios you mention ... posted by Frodan on August 14, 2008 at 12:38:03
Got it thank you Frodan you are a gentleman. I've got the bios set, xp pro installed and AWE set up. Now I have loaded cicplay and cpuz into my new pc and I'm trying to set up asio4all so I can test it before I install the julia card which is still in my old pc. After I confirm through the usb output into my Benchmark dac that it works I will go back and install cmp software, minimize services, shut down all but the recommended processes and make the other Windows optimizations. Is this a reasonable approach? I want to do it slow because I am not very good at the pc thing (I am a newbie of 6months into pc audio).
Any general advice for a newbie?
- Expert status is only hours away - rickmcinnis@dogwoodfabrics.com 14:55:29 08/14/08 (4)
In Reply to: RE: Before setting bios you mention ... posted by theob on August 14, 2008 at 14:31:20
Theo,
Congratulations on the big step.
You might find ASIO4ALL even more daunting than JULII@. I have heard it can be tricky.
You are probably very good at installing JULI@ after all of your trials and tribulations. You might be surprised how easy it may turn out to be, now. Especially with segar's findings.
As you well know, making mistakes is the best way to learn. None of us like to re-do things, mainly because we want the thing to work so we can enjoy listening to music through it. Who knows you might make a brilliant mistake that will benefit us all so do not be afraid of your machine.
At the moment I am going through the learning process with implementing cics's lastest BIOS recommendations. My machine did not like them. Even though it would start WINDOWS after making the changes, when I shut the machine down and attempted to turn it back on I got nothing.
So I went through the removing the CMOS battery thing which seemed to take forever for the 'board to forget the bad settings. When I re-installed the battery it started but with a screen I have never seen before. I had my screen on the floor to make it easier to get in my homemade wooden computer box so I could not read what it said, IN TIME. I realize now it said something to the effect of if I do not tell it otherwise it will revert to my last settings which caused it to not start. So, it goes into WINDOWS, I make sure it will PLAY, which it does, and I figure I will re-do the BIOS settings. Well, when I turned it off and turned it back on I got the same thing as before. NOTHING. So I removed the battery again and this evening when I re-install and start up I will be paying CLOSE attention to what is said.
Sorry for the extended nonsense story but I do think it illustrates our common situation. I have been playing around with this for nine months and I do know what I know, but there is so much left to find out.
Don't be afraid of the machine. You will master it.
Have fun,
Rick McInnis
- RE: Expert status is only hours away - theob 04:27:30 08/15/08 (3)
In Reply to: RE: Expert status is only hours away posted by rickmcinnis@dogwoodfabrics.com on August 14, 2008 at 14:55:29
thanks, my real journey now starts
- Cics: problem setting up cplay on new computer - theob 18:47:42 08/15/08 (2)
In Reply to: RE: Expert status is only hours away posted by theob on August 15, 2008 at 04:27:30
I downloaded final version onto a flash drive from my old pc, took flashdrive to new computer,loaded it onto desktop. When I try to run it I get Processing ini file message. I get the cplay set up screen but it says problem ocurred trying to save settings. Under cplay diagnostics it says
installation path not found in registry. Reistall player. On ok settings will be activated, So I tried deleting cplay folder
(system wont let me)in order to reload so I tried uninstalling from control panel and cplay doesn't show. I need help.
i need help/advice.
- Reboot. Goto "C:\Program Files\cics Play" and run "unins000.exe" - cics 02:39:02 08/16/08 (1)
In Reply to: RE: Cics: problem setting up cplay on new computer posted by theob on August 15, 2008 at 18:47:42
That should uninstall cPlay.
Make sure to install cPlay via installer file.
- RE: Reboot. Goto "C:\Program Files\cics Play" and run "unins000.exe" - theob 05:11:28 08/16/08 (0)
In Reply to: RE: Reboot. Goto "C:\Program Files\cics Play" and run "unins000.exe" posted by cics on August 16, 2008 at 02:39:02
thank you it worked :)
- RE: Setup for older version of Gigabyte GA-G31M-S2L mobo - seger 14:00:30 08/13/08 (0)
In Reply to: RE: Setup for older version of Gigabyte GA-G31M-S2L mobo posted by cics on August 13, 2008 at 13:01:19
I am also finding the core voltage going lower than I expected. With your settings it shows just over .9v, but I raised the CPU Host Freq to 175 with RAM still on 3-3-3-8 and voltage set to +1.1 and core voltage in cpuz is 1.000. Sounds great with full-on cmp2. This is with power set for Home/Desktop in Windows. With max battery not quite right.
- This IS a work in progress - rickmcinnis@dogwoodfabrics.com 08:28:29 08/13/08 (3)
In Reply to: RE: Reducing CPU temp (and power consumption!) using Gigabyte GA-G31M-S2L mobo (this is work in progress) posted by cics on August 12, 2008 at 12:16:51
Dear cics,
I first did as the instructions specified.
My computer would not bring up WINDOWS. It would come to the first WINDOWS screen and then die.
I returned to the original configuration and all was well.
I thought I would go into this piecemeal and see how close I could get.
By the way, my CPU voltage stays at 1.104 volts in CPUZ.
I did all but the 1.0000 thing. After saving it entered WINDOWS and seemed to be fine. I then turned off the machine and restarted.
NOTHING. Nothing happens at all. It was like when I got my first GIGABYTE board.
I have removed the CMOS battery. I will try again tonight.
Is there anything else I should do?
Thanks,
Rick McInnis
- CPU voltage now reduced to 0.89375V (giving Vcc of 0.868V). Temp drops another 10°C. - cics 10:36:38 08/13/08 (2)
In Reply to: RE: This IS a work in progress posted by rickmcinnis@dogwoodfabrics.com on August 13, 2008 at 08:28:29
I have complete stability and no problems with Windows running 24/192 145db SNR. CPU-Z shows Vcc of 0.864V / 0.848V. E7200 offers VID from 0.85 to 1.3625. I'll settle with low end of voltage range.
I'll test with old version.
- Is there a way to determine which generation board one has? - rickmcinnis@dogwoodfabrics.com 12:29:46 08/13/08 (1)
In Reply to: RE: CPU voltage now reduced to 0.89375V (giving Vcc of 0.868V). Temp drops another 10°C. posted by cics on August 13, 2008 at 10:36:38
That is all at the moment.
That is quite a reduction in voltage.
How does this affect the sound? General improvement or some specific ...
Hope I can get this thing to work!!!
THANKS,
Rick McInnis
- RE: Is there a way to determine which generation board one has? - Frodan 12:39:41 08/14/08 (0)
In Reply to: RE: Is there a way to determine which generation board one has? posted by rickmcinnis@dogwoodfabrics.com on August 13, 2008 at 12:29:46
There is a place in CPUZ in the upper right hand corner which specifies the board revision, but I'm not certain what it refers to.
Frodan
- RE: Reducing CPU temp (and power consumption!) using Gigabyte GA-G31M-S2L mobo (this is work in progress) - theob 04:59:18 08/13/08 (0)
In Reply to: RE: Reducing CPU temp (and power consumption!) using Gigabyte GA-G31M-S2L mobo (this is work in progress) posted by cics on August 12, 2008 at 12:16:51
Does 1.000 volts sound better than 1.125? Still awaiting my mobo.
- RE: Reducing CPU temp (and power consumption!) using Gigabyte GA-G31M-S2L mobo (this is work in progress) - gagnem 19:00:49 08/12/08 (0)
In Reply to: RE: Reducing CPU temp (and power consumption!) using Gigabyte GA-G31M-S2L mobo (this is work in progress) posted by cics on August 12, 2008 at 12:16:51
Be careful with this one if you have an older version of the board !!!
I had to clear CMOS in order to get the PC back alive.
- A cics recommended power supply for $30.00 - rickmcinnis@dogwoodfabrics.com 13:34:32 08/11/08 (3)
In Reply to: RE: cPlay - the open source high-end audio player using ASIO posted by cics on May 05, 2008 at 12:31:58
After noticing on the cMP thread cics's recommendation of the ANTEC Earthwatss EA-430 power supply I went to my default choce for parts (newegg) and see they have it for sixty dollars with a thrity dollars rebate. The price also includes freight within the 48 states.
I immediately put one on order.
I had been wanting to try Alfred's recommendation of adding capacitance to the power supply at the connecting headers. Laziness had got the best of me and I have yet to do it. Much easier to do this with an uninstalled power supply. And what is to fear if one destroys a thrity dollars device!!! (With nonchalance I doubt I will mess it up)
Just thought there may be some interest in this.
Cordailly,
Rick McInnis
- There's a difference in GA-G31M-S2L mobo's. Try this... - cics 11:09:17 08/12/08 (0)
In Reply to: RE: A cics recommended power supply for $30.00 posted by rickmcinnis@dogwoodfabrics.com on August 11, 2008 at 13:34:32
I have 2 GA-G31M-S2L with each giving different CPU voltages. Vcc at 1.12V is preferred (as per my 1st installation), the newer version gives voltage of 1.024V at idle and with 24/192 playback, its 1.250V.
In BIOS, under 'MB Intelligent Tweaker (M.I.T.)' set 'System Voltage Control' to Auto. With this setting, I get Vcc of 1.125V at 24/192. CPU frequency setting of 170 works as well. Test this on your setup.
- RE: A cics recommended power supply for $30.00 - sonics 03:28:52 08/12/08 (1)
In Reply to: RE: A cics recommended power supply for $30.00 posted by rickmcinnis@dogwoodfabrics.com on August 11, 2008 at 13:34:32
hi,
the earthwatts 430 PSU is a real hit.It seems to be better than my enermax with condensatormodification-very smooth etc.-.
After assembling 5 cmp-PCs ( different Hard- and same software) i can contribute some impressions.
1. till now gigabyte ga 31 sla with 7200 e was the best mobo- processorcombination. (underclocked to 1200Mhz 1GB soloRAM,separate PSU for fans and harddisc and no oversampling). Oversampling not bad ,but NOS much more laid back and same amount of information.
2.antec earthwatt 430 PSU very good influence in SQ but fan is not really quiet.( no problem for me because of the distance to the pc)
I try to do the PSU tweaking but i think it will have lower influence on SQ than with hig ripple PSUs..
3.Juli ESI only digital still the most emphatic soundcard.Very rich sound ,deep and realistic soundstage and 3D-voices.Absence of harshness.
4. One PC is working with the Xonar PCIe soundcard.(5 hours, so only first impressions):only one digital out.SPDIF mutes the optical out and vice versa.So i can only use one DAC.I need a separate DAC for my subwoofer.Under normal circumstances no problem. It seems to be more dynamic in the trebles but till now not very harmonic.Perhaps it betters in some days. Tomorrow i will do a separate PSU for the soundcard via floppydisc header-a very intresting possibility.
More impressions soon.
bye
Alfred
- I'm gonna do the same - already bought linear regulated PS for my Xonar PCI-e. - carcass93 08:21:37 08/12/08 (0)
In Reply to: RE: A cics recommended power supply for $30.00 posted by sonics on August 12, 2008 at 03:28:52
Should be real easy - "+" (center) to 5V, "-" to ground on floppy connector, other two unused.Let's compare notes afterwards.
Edits: 08/12/08
- Gigabyte Motherboards - Frodan 10:23:29 08/08/08 (5)
In Reply to: RE: cPlay - the open source high-end audio player using ASIO posted by cics on May 05, 2008 at 12:31:58
As I mentioned in a previous post my friend Andre had tried the GA-EG31M-S2 with no success. Yesterday he got his GA-G31M-S2L and put it together, then updated the BIOS. He was unable to set the CPU frequency below 230mhz. I did a test on mine for the hell of it and was able to bring the clock down to 145! (Not sure how stable that would be though). His board is from a new batch as they were on back-order for a few weeks, while mine was purchased about a month ago. Any ideas?
Frodan
- RE: Gigabyte Motherboards - theob 05:08:55 08/09/08 (0)
In Reply to: RE: Gigabyte Motherboards posted by Frodan on August 08, 2008 at 10:23:29
Cics for those of us with gigabyte mobo's (GA-G31M-S2L)on back order will we be able to underclock? If you are in contact with Gigabyte can you ask why this feature seems no longer to be working? Or asked another way is there still a way to underclock with new generation of GA-G31M-S2L?
- RE: Gigabyte Motherboards - cics 02:21:55 08/09/08 (3)
In Reply to: RE: Gigabyte Motherboards posted by Frodan on August 08, 2008 at 10:23:29
Just upgraded my 2nd cMP² using new Gigabyte GA-G31M-S2L mobo. I was hoping to try the GA-EG31M-S2 but Gigabyte says this mobo can't handle underclocking. Using same Kingston 1GB RAM, I've had no problems.
Here's CPU-Z with 200 setting (lowest I set it to was 175, no problems):
Interesting to see core voltage down to 1.024V. I think this is due to power supply and may explain differences we getting.
I suggest try another RAM and PSU. Also reset BIOS from CMOS.
- RE: Gigabyte Motherboards - davegh@pacbell.net 12:22:18 08/11/08 (2)
In Reply to: RE: Gigabyte Motherboards posted by cics on August 09, 2008 at 02:21:55
cics, I am gathering the bits to build a music server, I ordered (I thought) the GA-G31M-S2L but I received the GA-EG31M-S2, I do have the 7200 processor. should I send back the GA-EG31M-S2 and get the GA-G31M-S2L? I do want to keep the power and heat down a low as I can. Also is there any penalty (or problem) with using a UPS to keep the computer from shutting down inadvertantly? I live in the Santa Cruz Mountains and we ca get some power fluctuations and drop-outs.
Thanks,
Dave
- GA-G31M-S2L is the better option as GA-EG31M-S2 can't underclock. UPS is a good idea. - cics 12:31:37 08/11/08 (1)
In Reply to: RE: Gigabyte Motherboards posted by davegh@pacbell.net on August 11, 2008 at 12:22:18
Good passive cooling is needed if not using something like Zalman's TNN-300 case. Avoid standard CPU heatsinks (it won't work). I'm burning-in this cMP² at normal voltage... temperatures are on the high side so far.
- RE: GA-G31M-S2L is the better option as GA-EG31M-S2 can't underclock. UPS is a good idea. - davegh@pacbell.net 11:51:17 08/12/08 (0)
In Reply to: RE: GA-G31M-S2L is the better option as GA-EG31M-S2 can't underclock. UPS is a good idea. posted by cics on August 11, 2008 at 12:31:37
Thanks, I've returned the GA-EG31M-S2 and ordered the other one.
Dave Hally
- BIOS optimization: carcass93 suggestion on changing PCI latency timer settings - cics 08:25:15 08/06/08 (9)
In Reply to: RE: cPlay - the open source high-end audio player using ASIO posted by cics on May 05, 2008 at 12:31:58
On Biostar mobos, the PCI Latency tool can be used to adjust timers.
Not so for Gigabyte - I suspect this is a feature of Intel's chipset. However, default latency timer can be changed in BIOS:
- Enter BIOS
- From main menu, Press Ctrl+F1 to access advanced options
- Select PnP/PCI Configurations
- Change default timer setting of 32 to 128
There's a small improvement when using cMP².
- Did this appear from nowhere? - rickmcinnis@dogwoodfabrics.com 13:35:57 08/06/08 (2)
In Reply to: RE: BIOS optimization: carcass93 suggestion on changing PCI latency timer settings posted by cics on August 06, 2008 at 08:25:15
Dear cics,
I remember seeing a post about this PCI LATENCY tool on another forum and was wondering what you thought about this. I figured you would see it eventually and would let us know.
Now it seems to have appeared out of the blue!
Are you recommending this change in BIOS? Just curious (as always).
Another question: do you have a comment on segar's post about JULI@ drivers http://www.audioasylum.com/forums/pcaudio/messages/3/35002.html
I must admit I have no idea what is being said here and I need the master to decode it.
Let us know,
THANKS,
Rick McInnis
- For Gigabyte mobo, do the BIOS change. - cics 06:33:57 08/07/08 (0)
In Reply to: RE: Did this appear from nowhere? posted by rickmcinnis@dogwoodfabrics.com on August 06, 2008 at 13:35:57
Juli@'s performance for digital output is at a high-end standard.
- RE: Did this appear from nowhere? - seger 14:25:03 08/06/08 (0)
In Reply to: RE: Did this appear from nowhere? posted by rickmcinnis@dogwoodfabrics.com on August 06, 2008 at 13:35:57
To clarify: the Juli@ electronics are wonderful, and we are all enjoying the music as a result. But the construction of the card and breakout cable is what I consider strange and not the best. For the drivers, you don't have a problem, because the 0.979 are perfect for cPlay and you won't be uninstalling and installing very much. For me, I want to be able to use kernel streaming as well as cPlay(ASIO) so I need good WDM drivers. The ones in 0.979 don't work in Windows XP, but the ones in driver package 1.21 are great. So I'm saying, why not have the ASIO drivers that work best (0.979) packaged with the WDM drivers that work best (1.21) in one set for Windows XP?
Uninstalling the 0.979 drivers properly is important if you then want to install other drivers - it doesn't work if you have files with the same name still in the Windows system folders and you want to do another installation.
- What's behavior with Gigabyte mobo - latency reverts to default? N/T - carcass93 11:37:28 08/06/08 (5)
In Reply to: RE: BIOS optimization: carcass93 suggestion on changing PCI latency timer settings posted by cics on August 06, 2008 at 08:25:15
N/T
- RE: What's behavior with Gigabyte mobo - latency reverts to default? N/T - cics 13:02:19 08/06/08 (4)
In Reply to: RE: What's behavior with Gigabyte mobo - latency reverts to default? N/T posted by carcass93 on August 06, 2008 at 11:37:28
When BIOS is changed to 128 PCI clock cycles (from default of 32) mobo accepts this setting - no override or revert to default.
Tried 64 but prefer 128. Gigabyte mobo offers only 3 options (32, 64 & 128).
I would only do this in a cMP setup as most devices are disabled - PCI latency tool shows only a few devices. Latency timers take effect only when a device is transacting. So trick is to keep traffic on all other devices (VGA, SATA & USB) to a minimum allowing Soundcard to get maximum share.
- No, I mean - what happens if you use PCI Latency tool on Gigabyte mobo? - carcass93 13:09:40 08/06/08 (3)
In Reply to: RE: What's behavior with Gigabyte mobo - latency reverts to default? N/T posted by cics on August 06, 2008 at 13:02:19
Does it at least show sound card?BTW, on different subject - did you get a chance to look at the MSI G31M-F mobo? Looks like it has all the stuff like memory timings, multiplier, voltage adjustments (not CPU voltage though).
Edits: 08/06/08
- RE: No, I mean - what happens if you use PCI Latency tool on Gigabyte mobo? - cics 07:09:12 08/07/08 (2)
In Reply to: RE: No, I mean - what happens if you use PCI Latency tool on Gigabyte mobo? posted by carcass93 on August 06, 2008 at 13:09:40
Here's my Dell PC:
Its very similar to what I get with the Gigabyte mobo and IBM tablet (all with Intel chipsets). Note soundcard latency is at 0.
Although I'm able to set 'New Latency', it fails to implement, i.e. Settings > Apply doesn't change 'Latency'. According to documentation, if Latency is 0, its not programmable (quote: "Not all devices have programmable Latency Timers. If a device has a default latency value equal to 0, then most likely it's not programmable.").
This is different on Biostar mobo where timers can be set. Although some devices show 0 latency.
MSI's G31M-F may be worth trying. I'm very pleased with Gigabyte's GA-G31M-S2L. On paper, GA-EG31M-S2 might have a slight edge but not after Gigabyte's feedback: GA-EG31M-S2 cannot be underclocked!
Other comments:
PCI latency of 128 seems ideal (even if its done in BIOS). This is how I see it: you must have ASIO buffer size of less than 64 samples (but not 64 or more). At 128 cycles, up to 512 bytes (128x32bits/8) can be transferred without timeout. Some clock cycles may be needed for overheads. 512 bytes is good for just under 64 samples (@32 bits).
You can run cMP with EMU - see here . Don't know if you tried this.
- I tried PCI Latency tool at work on IBM with 845 chipset - works fine, so may be it's chipset-specific. - carcass93 08:27:12 08/07/08 (1)
In Reply to: RE: No, I mean - what happens if you use PCI Latency tool on Gigabyte mobo? posted by cics on August 07, 2008 at 07:09:12
Can you try one thing - search the registry on your Dell for "DisableProgPCILatency" setting, and post location(s), if you find any, and values.
About EMU card - I yanked it in favor of ASUS Xonar D2X PCI-e, my impressions are below in cPlay thread. I haven't gotten around trying it with cMP yet.
- "DisableProgPCILatency" doesn't exist in Registry - I think its from older versions of the tool. - cics 02:34:51 08/09/08 (0)
In Reply to: RE: I tried PCI Latency tool at work on IBM with 845 chipset - works fine, so may be it's chipset-specific. posted by carcass93 on August 07, 2008 at 08:27:12
Version used is 3.2.
- RE: cPlay - the open source high-end audio player using ASIO - jbpezzini@hotmail.com 12:17:40 08/05/08 (1)
In Reply to: RE: cPlay - the open source high-end audio player using ASIO posted by cics on May 05, 2008 at 12:31:58
Cics, could you summarize the actual version of cPlay (update hardware description)?
- Use latest version (1.0 Final) - use SSSE3 version if your processor can support SSE3 instructions - cics 08:12:54 08/06/08 (0)
In Reply to: RE: cPlay - the open source high-end audio player using ASIO posted by jbpezzini@hotmail.com on August 05, 2008 at 12:17:40
CPU-Z utility will indicate if CPU supports SSE3.
Hardware depends on upsampling quality and rate. For best performance use these hardware specs .
- cPlay 1.0 Final (SSSE3) Released - cics 06:40:08 08/01/08 (53)
In Reply to: RE: cPlay - the open source high-end audio player using ASIO posted by cics on May 05, 2008 at 12:31:58
At the download link provided (at SourceForge), there's 2 options for 1.0 Final:
- [cPlay_1_0_Final_setup.exe] Works on all processors
- [cPlay_1_0_Final_SSSE3_setup.exe] Requires an advanced processor
This SSSE3 release is a machine specific build for processors supporting SSSE3 instructions. At a very minimum, SSE3 can be used. Intel's Core 2 Duo processors (E1xxx, E2xxx, E4xxx, E6xxx, E7xxx, etc.) support these instructions.
You cannot use this version on non SSSE3/SSE3 capable processors, otherwise cPlay will result in a (invalid instruction) crash.
cPlay benefits from additional compiler optimizations with SSSE3 and is optimized for Core 2 Duo processors.
Please REMOVE previous versions before installing cPlay 1.0 Final (SSSE3) .
- RE: cPlay 1.0 Final (SSSE3) Released - promethk 02:08:32 08/07/08 (12)
In Reply to: RE: cPlay 1.0 Final (SSSE3) Released posted by cics on August 01, 2008 at 06:40:08
Thanks again for the great contributions!Motivated by your earlier recommendations and design guidelines, I have been building my computer-audio player in the past few months with the following two sets of CPUs and motherboards:
Intel CPU E1200 and E7200
Asus mb P5K-VM (G33) and P5SD2-VM (SIS672)I brought the G33-based board in March based on some testing results showing the G33 to be the lowest power consuming board of all recent intel based chipsets. Coupling the G33 with the E1200 and Juli@ gives much superior result to my previous setup using squeezebox to dac. However, the amount of heat coming out from the G33 chipset seems higher than those from the passive heatsink of the E1200.
When the E7200 first come out, I also got one with the P5SD2 (SIS) as a "cheap bundle". Using first the E1200 and replacing the G33 based board with the SIS board gives a more relaxed and organic music.
However, my experience of the E7200 is very much different to yours. My E7200 runs very hot at 50-65C c.f. the 30C of the E1200. The VID of the E7200 is 1.2125V which is on the high side of stated range of E7200, so perhaps I got a lemon. At stock voltage and F.S.Bus speed, I couldn't say there is any improvement in audio quality c.f. the E1200. As the SIS motherboard doesn't have bios tweaking options, I tried pin modding of the processor with partial success and manage to reduce the bus speed down to 200x4 MHz.
Here are some power measurements of the different setups as I believe they do play a role in quality playback and the figures scales well with the CPU utilization. Using a watt meter between wall socket and the audio PC,
46W on a E1200 with SIS chipset, foobar with 96k SRC
46W on a E7200 with SIS chipset, cPlay with 96k, 145.68db
50W on a E7200 with SIS chipset, cPlay with 192k, 145.68db
41W on a E7200 with SIS chipset, cPlay with 44.1k/48k, 145.68db
40W on a E7200 with SIS chipset, idleThere is a Juli@ card and a single piece of 2G Kingston RAM with a 2.5" harddisk and usb keyboard/mouse. E7200 at 1.2Ghz, FSB800.
My corsair hx620 power supply has about 80% efficiency at 10% output while my components are estimated to consume thirty something watts which is only 5-6% of its rated output and its efficiency could be much lower.
While I have thought that SSSE3 optimization could further lower CPU utilization and thus lower power draw, I could not found any difference from the non-optimized one.
Wishes:
Both cPlay and cMP do not support cuesheets with unicode filenames in it, where such cuesheets can be played without problem in various versions of foobars.
Edits: 08/07/08
- E7200 should show VID of ~1.12V with temp ~25°C. - cics 07:38:04 08/07/08 (9)
In Reply to: RE: cPlay 1.0 Final (SSSE3) Released posted by promethk on August 07, 2008 at 02:08:32
Problem may well be with mobo - check CPU voltage setting. I know from other users of recommended Gigabyte mobo that setting CPU voltage to 'Auto' causes high voltages and temperatures. Mobo manufacturers seem to favor this due to OC. Rather use manual voltages.
Thanks for the power readings. At 80% efficiency, that means 37 watts power consumption for E7200 doing 96k @ 145db SNR (40 watts at 192k). You should get this lower using Granite Digital PS, alternate USB ports and 1GB RAM. Also, make sure video settings are done as per manual (you want to disable video GPU in G33).
Very interesting to see E7200 with high VID & temp is still more power efficient (E1200 does 37 watts for 96k @ 97db SNR - foobar).
- RE: E7200 should show VID of ~1.12V with temp ~25°C. - promethk 08:59:25 08/07/08 (8)
In Reply to: RE: E7200 should show VID of ~1.12V with temp ~25°C. posted by cics on August 07, 2008 at 07:38:04
I checked again the E7200 heatsink again. This time I am brave enough to use my fingers to touch the heat-pipe near the CPU directly, while expecting a burn, I feel nothing! So the E7200 sensor is malfunctioning as some users have already indicated in the internet.
I checked the Kingston data sheets for the ram modules, the 2G that I have got is their lowest power 2G model, with only 1.975W while the corresponding 1G model is 1.872W, so the difference is only 0.1 W.
On the next build, I am considering to install windows XP, to boot and operate entirely off a USB external harddisk, based on some "hack" on various configure files of windows XP that I found on the web. This will seperate the power as well as enable direct transfer of music I ripped from my other computer.
- RE: E7200 should show VID of ~1.12V with temp ~25°C. - cics 00:14:58 08/08/08 (0)
In Reply to: RE: E7200 should show VID of ~1.12V with temp ~25°C. posted by promethk on August 07, 2008 at 08:59:25
You could disable Thermal monitoring in BIOS - I do this as well as disable CIE function.
RAM, USB stuff & ROM drive (not sure if you have one) all add load which is best when bypassed. Those few watts do add up. It would be interesting to see what power measures you get with this and VID at 1.12V. Any chance of using the Gigabyte mobo?
Avoid USB mobo connection - measurements that I was doing for jitter showed 3db noise floor increase when USB component is connected.
- RE: E7200 should show VID of ~1.12V with temp ~25°C. - ninekit 19:16:21 08/07/08 (6)
In Reply to: RE: E7200 should show VID of ~1.12V with temp ~25°C. posted by promethk on August 07, 2008 at 08:59:25
Set to lowest core voltage, may I ask can you run 145db SRC @ 96k with e7200 passively cooled stock heatsink?
Thanks
- RE: E7200 should show VID of ~1.12V with temp ~25°C. - promethk 00:39:48 08/08/08 (4)
In Reply to: RE: E7200 should show VID of ~1.12V with temp ~25°C. posted by ninekit on August 07, 2008 at 19:16:21
Well for my setup, it is not possible. But noting that different E7200 could have different default VID and mine is higher than the value posted here.
My current motherboard is not flexible in voltage setting and the voltage is 1.2V. But my guess is that it is not possible because the stock intel heatsink is not very efficient, no copper and no heatpipe, very little fin for airflow. And the stock intel heatsink with fan off feels very hot. The increase in temp. will cause a rise in (dynamic) resistance, making the processor draws more power.
With a cheap and tall aftermarket heatsink with copper heatpipe for passive cooling, the situation improves a lot so that heat will no longer be problem even with 192kHz processing.
- Check your Power Options setting (it must be Max Battery and not Home Office/Desktop). - cics 23:34:37 08/10/08 (3)
In Reply to: RE: E7200 should show VID of ~1.12V with temp ~25°C. posted by promethk on August 08, 2008 at 00:39:48
Sometimes XP doesn't take the setting and must be redone. With Max Battery power scheme, VID must be lower than normal.
- Order of cmp build - theob 10:45:18 08/11/08 (2)
In Reply to: RE: Check your Power Options setting (it must be Max Battery and not Home Office/Desktop). posted by cics on August 10, 2008 at 23:34:37
Dear Cics, when one is building a cmp from 'scratch or the ground up' should one do the hardware set up in
http://www.audioasylum.com/forums/pcaudio/messages/3/33137.html
first then do the Bios and Power set up from CMP documentation 'The art of Building a computer transport' second or vice versa? Or should it be simultaneous?
- Only do BIOS setup as per post - AOB refers to Biostar mobo. - cics 12:34:45 08/11/08 (1)
In Reply to: RE: Order of cmp build posted by theob on August 11, 2008 at 10:45:18
Power setup is as per AOB.
- RE: Only do BIOS setup as per post - AOB refers to Biostar mobo. - theob 16:01:04 08/11/08 (0)
In Reply to: RE: Only do BIOS setup as per post - AOB refers to Biostar mobo. posted by cics on August 11, 2008 at 12:34:45
thanks
- Yes, 145db SRC @ 96k with e7200 passively cooled is not a problem. - cics 00:19:03 08/08/08 (0)
In Reply to: RE: E7200 should show VID of ~1.12V with temp ~25°C. posted by ninekit on August 07, 2008 at 19:16:21
You can do this for all Core 2 Duo processors as long as passive cooling is good - no need to go overboard.
My E6300 does 145db SRC @ 192k with passive cooling. Temp is ~65°C using Thermaltake heatsink. E7200 does ~25°C with Zalman case.
- cuesheets for cMP - seger 04:03:26 08/07/08 (1)
In Reply to: RE: cPlay 1.0 Final (SSSE3) Released posted by promethk on August 07, 2008 at 02:08:32
Open the cuesheet. File > Save as. In the "Encoding" drop-down, select ANSI. Save. Let it replace the file. You now have a cuesheet that will play in cMP.
- RE: cuesheets for cMP - promethk 09:11:34 08/07/08 (0)
In Reply to: RE: cuesheets for cMP posted by seger on August 07, 2008 at 04:03:26
Thanks for the help! Yes, it work!
I did try the ansi cuesheet without success with earlier version of CMP.
- Stunning - rickmcinnis@dogwoodfabrics.com 11:34:53 08/02/08 (26)
In Reply to: RE: cPlay 1.0 Final (SSSE3) Released posted by cics on August 01, 2008 at 06:40:08
Dear cics,
Just when I thought you had surely exhausted any chance of a major improvement you give us this.
I think you sell it short saying that it is limited to the high frequencies. I hear improvement all across the spectrum. The bass is more powerful and yet more sharply defined. Whereas some earlier iterations seemed to have a bit too much bass this has that abundance with control. I like this very much.
There is so little glare left that I must assume what is there is part of the recording. Much glare I thought was part of my room is no longer a hindrance with louder passages. Many things I thought were just the way it is are now found to not be so.
Listening to the EBBETS LP transfer of Beatle's records is still revealing of something missing in the CD itself. These I downloaded in FLAC from from the internet. I converted to WAV and for lack of a better way to transfer to my MUSIC COMPUTER I had to make a CD copy. STILL, these recordings display qualities unlike "regular" CD's. Maybe it is simply euphonic LP sounds (who knows what equipment was used? In fact, if someone does know I would love to be told) but they (even through all of the transmogrifications) sound very nice and especially better than ever with this latest refinement.
cics, you are quite a fellow!!!
I found an interesting thing: if one has loaded an album into cPLAY one can turn off the hard drive and it keeps on playing. I have not listened critically to a record with the HDD power on and off. One has to wonder if it would make much difference? Another motor not sending its mischief into the the MB? When one turns the drive back on the computer shuts down (blue screen and long message).
Simply stunning.
Thanks (just doesn't say it sufficiently),
Rick McInnis
- I should have said PRESSED CD's - rickmcinnis@dogwoodfabrics.com 11:33:28 08/03/08 (0)
In Reply to: RE: Stunning posted by rickmcinnis@dogwoodfabrics.com on August 02, 2008 at 11:34:53
In my comments about the EBBET'S download I should have said the above.
One has to wonder if that is the biggest obstacle facing the REDBOOK process.
I believe Todd Krieger has had much to say about this, along with many others. None quite as adamant and self-assured as Todd and I can now see how correct he is.
I wish he would try cics's approach.
- Improvements as you experienced is not limited to just HF. Bass is also exceptional. - cics 10:50:22 08/03/08 (3)
In Reply to: RE: Stunning posted by rickmcinnis@dogwoodfabrics.com on August 02, 2008 at 11:34:53
On your setup, in BIOS try setting 'Set CPU Host Frequency (Mhz)' to less than 200 (say 180 then 175). This reduces power consumption by limiting CPU clock speed. You want to get to a value that's low enough without causing instability / slow response.
- CPU Host Freuency set to 175 mHz - rickmcinnis@dogwoodfabrics.com 09:47:51 08/04/08 (2)
In Reply to: RE: Improvements as you experienced is not limited to just HF. Bass is also exceptional. posted by cics on August 03, 2008 at 10:50:22
Dear cics,
This seems to work fine at 192 kHz oversampling/-145 dB upsampler.
I noticed, intitially, some delay with the "clock" within cPLAY. The music continued without hiccups with the clock acting a little wobbly. This soon calmed down and seemed to stabilize.
I am letting the computer run continuously the next few days while repeatedly playing a record.
My voltage on CPUZ for the processor has continued to be 1.104. Nothing seems to make it change.
Thanks,
Rick McInnis
- Any preference to which host freq you prefer? - cics 08:06:47 08/06/08 (1)
In Reply to: RE: CPU Host Freuency set to 175 mHz posted by rickmcinnis@dogwoodfabrics.com on August 04, 2008 at 09:47:51
.
- Still listening - rickmcinnis@dogwoodfabrics.com 13:23:25 08/06/08 (0)
In Reply to: RE: Any preference to which host freq you prefer? posted by cics on August 06, 2008 at 08:06:47
Dear cics,
I have retained 175 mHz as a baseline and when I get enough time in I will try 180 and see if I can hear any difference.
Is there the possiblity that a higher frequency COULD sound better even if the loewr frequency is stable?
Thanks,
Rick McInnis
- RE: Stunning - theob 05:19:26 08/03/08 (20)
In Reply to: RE: Stunning posted by rickmcinnis@dogwoodfabrics.com on August 02, 2008 at 11:34:53
Of course you are right rick, I only focus on the highs because that is the unique part of the 'absolute sound' that always reveals itself to me 1st. After 2 days listening to this version of cplay all the other stuff becomes noticed by me. Its a deeper soundstage the bass is more taut, the whole spectrum has more punch. It is such a revelation that I don't understand why others don't try and report on it. It is a milestone accomplishment IMO.
BTW I still have to reboot my system every other day to get the juli drivers to work right. Also I have an old computer into which I can build cmp, I found a local store (that also helps with installation--I need this) that sells all that I need. Question to you did you take this route or did you build it from the ground up? Reason I ask is because this old computer already has xp pro on it and I just wondered if it will survive the installation of a new mobo, processor, memory.
Happy to see you are into this final cplay, it is so good its scary.
- Juli@ driver uninstall - seger 03:23:05 08/06/08 (18)
In Reply to: RE: Stunning posted by theob on August 03, 2008 at 05:19:26
Juli@ version 979 drivers don't uninstall from Windows XP neatly, and I assume this is same for 978 as the files look same (but I haven't used 978). Quickest way I've found: Go to win32 folder that is inside _drv_juli@_3264 folder and double-click on UnInstall_32.exe file. Click on Uninstall button and wait for it to complete. (This removes Juli@ devices) Don't restart yet. Go to C:\WINDOWS\system32 - find JulaASIO.dll and JulaPAN.exe and delete. Now go to C:\WINDOWS\system32\drivers - find Jula.sys and JulaWDM.sys and delete. Now restart computer.
I've used this way many times now, as I have been comparing Juli@ 0.979 and 1.21 drivers back and forth. If problems persist, try new driver download, or reinstall windows.
- RE: Juli@ .979 driver uninstall/1.04 install - theob 03:38:22 09/29/08 (17)
In Reply to: RE: Juli@ driver uninstall posted by seger on August 06, 2008 at 03:23:05
so I assume you do not take out the juli card physically from the pci slot. is that correct? then when you reboot the pc will presumably identify new hw then go into the hw install wizard of xp at which time one should have the 1.04 drivers avaiable for install. correct?
- RE: Juli@ .979 driver uninstall/1.04 install - seger 05:13:32 09/29/08 (16)
In Reply to: RE: Juli@ .979 driver uninstall/1.04 install posted by theob on September 29, 2008 at 03:38:22
Yes - no need to take soundcard itself out.
- RE: Juli@ .979 driver uninstall/1.04 install - theob 05:24:11 09/29/08 (15)
In Reply to: RE: Juli@ .979 driver uninstall/1.04 install posted by seger on September 29, 2008 at 05:13:32
thanks
- RE: Juli@ .979 driver uninstall/1.04 install - theob 06:02:51 09/29/08 (14)
In Reply to: RE: Juli@ .979 driver uninstall/1.04 install posted by theob on September 29, 2008 at 05:24:11
I tried your process to uninstall .979 and the first step of 'Goto win32 folder that is inside _drv_juli@_3264' but I cant find _drv_juli@_3264. Isn't it supposed to be in program files under Local disc (C:) ?When I installed .979 I did it from a flash drive. I'm not sure I have such a file (tried process explorer search for _drv_juli@_3264 and did not find either). I have no juli icon on my pc screen but I do have julapan.exe in Windows Task Manager. cplay/cmp does work but now I'm not sure what to do to uninstall .979 and install 1.04
I do have juliPAN in C:\WINDOWS/system32.
Edits: 09/29/08 09/29/08
- RE: Juli@ .979 driver uninstall/1.04 install - seger 07:48:41 09/29/08 (13)
In Reply to: RE: Juli@ .979 driver uninstall/1.04 install posted by theob on September 29, 2008 at 06:02:51
Next time, save the drivers in a folder in a convenient place.
For now, just go to Device Manager and uninstall the 2 Julia devices listed, then go to system32 folder and delete the Julia files (as per my post), also the 2 files in system32\drivers folder. If they are not there, go ahead and restart, and see if you can install the 1.04 drivers.
- RE: Juli@ .979 driver uninstall/1.04 install - theob 08:29:26 09/29/08 (11)
In Reply to: RE: Juli@ .979 driver uninstall/1.04 install posted by seger on September 29, 2008 at 07:48:41
I unistalled all juli .979 stuff, saved the juli 1.04 drivers into the c drive, rebooted but there is no 'Found new HW' prompt from windows. So I shut down, took the sound card out of the pci slot, and rebooted then shut down again and reinstalled the juli sound card but when I rebooted again no 'Found new HW' prompt. I tried going into control panel to force a hw install but nothing. What am I doing wrong.
- RE: Juli@ .979 driver uninstall/1.04 install - theob 09:22:10 09/29/08 (0)
In Reply to: RE: Juli@ .979 driver uninstall/1.04 install posted by theob on September 29, 2008 at 08:29:26
I found a way (in add new hardware in control panel--show all devices) to install new drivers for 1.04 but it somehow doesn't work. I can get the new devices (juliapan and juli driver) to install and show up in device list. I can even see julipan and juli driver under system 32 but I don't get a link to my dac. I tried,retried and kept getting a code 10 error trying to install julipan. I need some assistance/advice on this.
- RE: Juli@ .979 driver uninstall/1.04 install - seger 09:18:38 09/29/08 (9)
In Reply to: RE: Juli@ .979 driver uninstall/1.04 install posted by theob on September 29, 2008 at 08:29:26
To change soundcard drivers you have to be out of minlogon, and these services have to be put on automatic and started: Event Log, Cryptographic Services, Windows Audio and WMI.
Look again to see that the Julia devices in Device Manager are really gone, and the Julia files in WINDOWS\system32 and WINDOWS\system32\drivers are not there. If the old drivers are really gone, good.
Now restart again and see if the new hardware wizard pops up. If not, try Control Panel > Add hardware. Follow the steps and you should eventually end up in the familiar New Hardware wizard.
- RE: Juli@ .979 driver uninstall/1.04 install - theob 09:29:37 09/29/08 (8)
In Reply to: RE: Juli@ .979 driver uninstall/1.04 install posted by seger on September 29, 2008 at 09:18:38
I know how to get services restarted but I do not know how to get out of minlogon. Is it just a mattaer of renaming winlogon.bak to winlogon.exe and vice versa for current winlogon (rename to minlogon)?
- RE: Juli@ .979 driver uninstall/1.04 install - seger 09:44:39 09/29/08 (7)
In Reply to: RE: Juli@ .979 driver uninstall/1.04 install posted by theob on September 29, 2008 at 09:29:37
Very important to get this right:
1. Start button > Run - type regedit and click ok. Navigate to this location:
HKEY_LOCAL_MACHINE\SOFTWARE\Microsoft\Windows NT\CurrentVersion\Winlogon
2. Delete “Config” line for minlogon: (Right-click on Config and delete)
3. Rename winlogon.exe in system32 folder to minlogon.exe
4. Rename winlogon.exe.bak to winlogon.exe
5. Restart.
- RE: Juli@ .979 driver uninstall/1.04 install - theob 11:13:54 09/29/08 (5)
In Reply to: RE: Juli@ .979 driver uninstall/1.04 install posted by seger on September 29, 2008 at 09:44:39
well I thought I followed your instructions but when I restarted I got '...a problem has been detected and windows has been shut down ...start up in safe mode...' yikes I really screwed up now!
- RE: Juli@ .979 driver uninstall/1.04 install - seger 12:09:53 09/29/08 (4)
In Reply to: RE: Juli@ .979 driver uninstall/1.04 install posted by theob on September 29, 2008 at 11:13:54
May be necessary to start from scratch. But if you can get into Windows in Safe Mode, all is not yet lost - while booting repeatedly press F8 and select Safe Mode in the list that appears. If you get in you can look for the problem. I was assuming you had previously unticked the box in Folder options > View tab: "Hide extensions for known file types" - this could cause confusion when renaming files if it was ticked.
- RE: Juli@ .979 driver uninstall/1.04 install - theob 12:13:17 09/29/08 (3)
In Reply to: RE: Juli@ .979 driver uninstall/1.04 install posted by seger on September 29, 2008 at 12:09:53
could not safe start, I guess this is a windows reload, start from scratch process. Is this correct?
- RE: Juli@ .979 driver uninstall/1.04 install - seger 12:21:22 09/29/08 (2)
In Reply to: RE: Juli@ .979 driver uninstall/1.04 install posted by theob on September 29, 2008 at 12:13:17
It's a do-over. Once I forgot to delete the Config line and this happened. It may also have to do with the attempted driver upgrade while in minlogon. All system changes, software/hardware installations should be done in normal XP. My mistake was to assume you knew this - I should have mentioned it at the beginning - sorry!
- RE: Juli@ .979 driver uninstall/1.04 install - theob 13:01:09 09/29/08 (1)
In Reply to: RE: Juli@ .979 driver uninstall/1.04 install posted by seger on September 29, 2008 at 12:21:22
you have nothing to be sorry for. I did minlogon initially I knew about extensions of file names. I should have left well enough alone with .979 driver and cplay 1.1 (which did not drop UI). It's my fault buddy and part of the rite to passage, I guess. But why did it have to happen the day the stock market crashed, normally I would listen to music to make myself feel better. Well as some philosopher said what doesn't kill you makes you stronger.
- RE: Juli@ .979 driver uninstall/1.04 install - seger 14:17:23 09/29/08 (0)
In Reply to: RE: Juli@ .979 driver uninstall/1.04 install posted by theob on September 29, 2008 at 13:01:09
The rebuilds get quicker and less stressful the more you do! Good luck with the next one, and keep a list of all the steps and tweaks - it makes it easier.
- RE: Juli@ .979 driver uninstall/1.04 install - theob 11:02:58 09/29/08 (0)
In Reply to: RE: Juli@ .979 driver uninstall/1.04 install posted by seger on September 29, 2008 at 09:44:39
what about on the back side of this?
- RE: Juli@ .979 driver uninstall/1.04 install - theob 07:59:19 09/29/08 (0)
In Reply to: RE: Juli@ .979 driver uninstall/1.04 install posted by seger on September 29, 2008 at 07:48:41
ok thanks
- The old computer - rickmcinnis@dogwoodfabrics.com 11:25:56 08/03/08 (0)
In Reply to: RE: Stunning posted by theob on August 03, 2008 at 05:19:26
Theo,
cics recommends a fresh install of XP after the installation of a new MB.
Not being one overly flush with cash, I must say I think you should, also, buy the components you need to do this right.
It is simply not that expensive compared to what it would cost to, (MAYBE?)have the same level of performance. Stop trying to make the silk purse out of the sow's ear. Your old computer will do for you what it has always but you need to have a dedicated machine for listening to music with extraordinary resolution.
I think the dedicated machine should NEVER be connected to the internet, EVER.
You have heard a glimpse of what is possible. I suspect there is a considerable improvement awaiting you.
I do not know exactly what you need but there is the possibility of getting a NEWEGG account which would allow six months, no interest. Splitting what you need for this into six payments would make it a little easier. That is what I did for my initial foray into the ART OF COMPUTER TRANSPORTS. Since then I have piecemeal replaced every piece in following cics's discoveries, but I ended up with another computer for other uses, the opposite of your situation! Think how lucky you are to get in this when it has reached such a high state of refinement.
Quit dilly dallying and follow your instincts. This is worth a sacrifice.
Sincerely,
Rick McInnis
- RE: cPlay 1.0 Final (SSSE3) Released - theob 08:00:04 08/01/08 (12)
In Reply to: RE: cPlay 1.0 Final (SSSE3) Released posted by cics on August 01, 2008 at 06:40:08
Cics I just want to confirm ' sse3 is represented by athlon 64 X 2 processor which means I can try it'? Correct?
Edits: 08/01/08
- I reread the links---I'm ok to use this release-- want feedback? - theob 09:16:07 08/01/08 (9)
In Reply to: RE: cPlay 1.0 Final (SSSE3) Released posted by theob on August 01, 2008 at 08:00:04
Remember I'm running an amd athlon 64 X 2, no AME but still many of the cmp parameters (underclocking, external hard drive ps via granite digital, and as many of the windows settings I can get w/o losing email, net browsing capability).
To get quick feedback I always play some of my Mercury files (antal dorati 7 studies by paul klee, howard hanson merry mount) and these files are brutal: very dynamic but lots of air/great timbre and depth and very quick especially the xylophone and flute) and at first these files sounded a little bright but then I remembered from a cold start my electronics do sound bright. After 20 minutes of warmup I sat back and listened and OMG you've done it again --- this is what I'm talking about in home system music playback: you have taken cplay to yet another level of clarity and transparency, another f on top of triple fortissimo runs, more round and natural timbre to woodwinds and that slight delicacy to strings and the ease of the music flow is simply breathtaking. How can it get better I just don't know (yes I do know-- go the whole 9 yards of cmp but I'm practicing on an extra computer & have already installed AME feature on it and it sounds good but is limited by asio4all at 48khz sampling at this time, I will do more).
Any audiophile with any kind of desire to hear an approximation of the live unamplified music needs to hear this. Simply stunning, thank you again again and again.
- RE: I reread the links---I'm ok to use this release-- want feedback? - w.rade 09:37:47 08/02/08 (5)
In Reply to: RE: I reread the links---I'm ok to use this release-- want feedback? posted by theob on August 01, 2008 at 09:16:07
I can only admit to the given description of the audible outcom, the "cPlay 1 0 Final SSSE3" version brings. Yesterday I could visit alfred (sonics). He had installed the version and it was impressing. Alfred and me know each other for several years. He was the person, who's input triggered me to start with PC for audio. I find it extremly exiting, what happend for me with cMP in a relativ short period. Looking forward to get the Gigabyte GA-G31M-S2L after my holidays. I like encourage everybody to have a PC dedicated to music only. The return will be beyond expectation. At least it was for me. Walter
- Try Gigabyte's GA-EG31M-S2. No feedback yet but... - cics 10:59:53 08/03/08 (4)
In Reply to: RE: I reread the links---I'm ok to use this release-- want feedback? posted by w.rade on August 02, 2008 at 09:37:47
There's cleaner CPU power delivery and this mobo allows for OC to 1600 FSB which means underclocking could potentially give better results.
I plan to get one of these as well.
- RE: Try Gigabyte's GA-EG31M-S2. No feedback yet but... - theob 13:48:31 08/03/08 (1)
In Reply to: RE: Try Gigabyte's GA-EG31M-S2. No feedback yet but... posted by cics on August 03, 2008 at 10:59:53
Should I order this( Gigabyte's GA-EG31M-S2) or Gigabyte GA-G31M-S2L? Are they similar size wise?
- Not yet - it seems GA-G31M-S2L is better. - cics 08:02:14 08/06/08 (0)
In Reply to: RE: Try Gigabyte's GA-EG31M-S2. No feedback yet but... posted by theob on August 03, 2008 at 13:48:31
.
- RE: Try Gigabyte's GA-EG31M-S2. No feedback yet but... - Frodan 13:09:07 08/03/08 (1)
In Reply to: RE: Try Gigabyte's GA-EG31M-S2. No feedback yet but... posted by cics on August 03, 2008 at 10:59:53
Hi cics,I am using the GA-G31M-S2L with the E7200 in my new setup but my friend Andre is using the GA-EG31M-S2. He has not been able to lower the CPU clock below 266, any lower and the board doesn't boot, which required a BIOS reset. The same with the RAM timings, he can't reach 2.0 to get the 3-3-3-9 latencies, the best was 4-4-4-10. In addition, when we communicated with Gigabyte's tech support we were never able to get them to understand we were trying to under-clock the board and they said we should use the default settings! He has now ordered the GA-G31M-S2L and will use the other board for other kinds of tests which we will talk about later.
As I get mine optimized (After reinstalling XP for the 3rd time!), I will post more here as well.
Daniel
Frodan
Edits: 08/03/08
- Thx. I've also asked Gigatbyte to open CPU frequency setting (below 266). Let's see what happens. - cics 08:04:31 08/06/08 (0)
In Reply to: RE: Try Gigabyte's GA-EG31M-S2. No feedback yet but... posted by Frodan on August 03, 2008 at 13:09:07
.
- RE: I reread the links---I'm ok to use this release-- want feedback? - theob 17:07:12 08/01/08 (2)
In Reply to: RE: I reread the links---I'm ok to use this release-- want feedback? posted by theob on August 01, 2008 at 09:16:07
More listening and I think I can further describe the sound as getting the ictus of each note correct. Its all about the high frequencies and rise time of the leading edge of music and this version gets it right! I think for me in my system this is the biggest incremental improvement I have heard from release 1 through 20. Cics are you going to describe in layman's term what this ssse (or sse in my case) version is physically doing?
- Yes there's a transformation in HF presentation with this release - cics 01:10:46 08/02/08 (1)
In Reply to: RE: I reread the links---I'm ok to use this release-- want feedback? posted by theob on August 01, 2008 at 17:07:12
I'm still learning what it does...
The sense of realism has lifted to a new level (using standard 16/44.1)!
- RE: Yes there's a transformation in HF presentation with this release - theob 04:20:24 08/02/08 (0)
In Reply to: RE: Yes there's a transformation in HF presentation with this release posted by cics on August 02, 2008 at 01:10:46
I'm still at 176 khz sampling and -146 src and 48 latency and I was fearfull my amd may not be up to the task but it's really sounding great.
- cics - correct me if I'm wrong, but SSE3 support in AMD would require compilation with different switches. - carcass93 09:05:54 08/01/08 (1)
In Reply to: RE: cPlay 1.0 Final (SSSE3) Released posted by theob on August 01, 2008 at 08:00:04
I'm not familiar with GCC compiler, but that would be the case with Intel's.
- Some AMD processors support SSE3 natively - no need for special compile. - cics 23:30:46 08/01/08 (0)
In Reply to: RE: cics - correct me if I'm wrong, but SSE3 support in AMD would require compilation with different switches. posted by carcass93 on August 01, 2008 at 09:05:54
I'm using Intel C++ 10.1 compiler.
- Some suggestions for a cPlay 1.0 Post-Final version. - ivigueras@terra.es 04:48:05 08/01/08 (4)
In Reply to: RE: cPlay - the open source high-end audio player using ASIO posted by cics on May 05, 2008 at 12:31:58
Hello.
First all, thanks to Cics for this wonderfull player. I have been listening and comparing it in many combinations: 145 or 121db SNR, 44.1, 48, 88.2 or 96. There are very curious sonic differences between this combinations. With higher sample rates, this and other players down, don't ask me why because I can't understand it. I supose that it will be a problem from the asio driver of my EMU 0404 USB -no comments about the support from EMU-.
But, this is not the object of this message. I would like to do some suggestion for a new post-final version. I named it "post-final" because I believe that is not necesary to much development work:
1. When I open a cue file, titles and tracks are showed in the player, but when I open a flac file, only the global track is showed. I supose that it will be easy read the embebed cue -if exists- from a flac file and show it.
2. If one track in the cue file contains an "INDEX 00", this is the point where cPlay begin to play if I jump to this track. It is not correct because the player must begin to play always in the "INDEX 01" of the track, as all CD Players do, because the "INDEX 00" is usually not used for contain music but other things as silents or claps between tracks. "INDEX 00" indicates the begin of the track but not the begin of the music. This is indicate by the "INDEX 01".
3. Finally, I think that it would be nice that cPlay show the tags from the flac files.
I have more suggestions but I believe that these need a major level of development work and I understand about Cics time and work, but I would like to ask Cics if the development is closed for him or admit collaborations?. I am computer engineer. I have not many free time now but, probably, after september, I could get some time.
Thanks a lot and best regards.
Ignacio.
Regards.
Ignacio.
- RE: Some suggestions for a cPlay 1.0 Post-Final version. - cics 06:15:56 08/01/08 (3)
In Reply to: RE: Some suggestions for a cPlay 1.0 Post-Final version. posted by ivigueras@terra.es on August 01, 2008 at 04:48:05
Ignacio you have good suggestions.
I would suggest that playlists is also a much needed feature. My interest is not in the UI / ergonomics stuff. I want to spend more time on improving the sound.
So yes, I would very much welcome assistance to develop cPlay's usability. Are you registered at SourceForge?
- RE: Some suggestions for a cPlay 1.0 Post-Final version. - ivigueras@terra.es 01:52:12 08/04/08 (2)
In Reply to: RE: Some suggestions for a cPlay 1.0 Post-Final version. posted by cics on August 01, 2008 at 06:15:56
I believe that you are right: ergonomics is one thing and sound quality is another. I think that a music PC must be this two different applications: ergonomics, for classification and search, and a very good basic player, like yours, for sound quality. And while music is sound, ergonomics applications should be pasive: without consuming CPU cycles. -Of course, a basic OS is necesary-.
My point of view is that playlists should be closely linked to the content's databases. This would allow all sorts of searches and combinations of lists and, definitely, a usefull tool for classical music listeners -like me-. Of course, this requires a lot of careful design and development. Great amount of work, in sum.
I am designing and developing a lite XML database and application for my personal use. I link the database contents with the flac tags in the files. Then I open the player directly over the flac file. I was testing this with you player when I thought my suggestions.
No, I am not registered at SourceForge. Until now my applications are very rudimentary and only for personal consumption. But this last application is growing too much. As I told before, now I have not free time, but from last week of september I could help.
Regards.
Ignacio.
- RE: Finally, I have got some time to develop. - ivigueras@terra.es 07:18:24 11/13/08 (1)
In Reply to: RE: Some suggestions for a cPlay 1.0 Post-Final version. posted by ivigueras@terra.es on August 04, 2008 at 01:52:12
Well, I think that it could be a good time to work in an ergonomic dress for cPlay.
Cics, where I can begin to help?. Have you thought something about the ergonomics that you would like?
Regards.
Ignacio.
- Help is needed but not for UI... - cics 22:57:12 12/07/08 (0)
In Reply to: RE: Finally, I have got some time to develop. posted by ivigueras@terra.es on November 13, 2008 at 07:18:24
Thanks Ignacio.
I want to reduce the computational cost of SRC 145db SNR upsampling. The rewards are significant.
Some background:
Bandlimited Interpolation is a very big deal. JOS' work has laid the foundation to most accurately recalculate samples for any arbitrary time , i.e. we are able restore the original analogue waveform. Its mathematical purity makes this so.
The best expression of JOS' work is that of Erik's, i.e. SRC (Secret Rabbit Code) using the Best Sinc converter. Its single drawback is however the computational cost. Technically, a Kaiser Window (our digital anti-alias low pass filter) is applied to the ideal impulse response (sinc function). Upsampling is done though intense computations against these filter coefficients.
The idea is to create an SSE2 version of SRC's Best Sinc converter. I suspect there's significant opportunities for loop vectorisation and vectorisation in general. Source code is available here . Theoretically we could gain a ~4 fold reduction in CPU load. This would allow us to completely detune the CPU!
PS - sorry for the delay.
- cics, single or dual memory preferred? - rickmcinnis@dogwoodfabrics.com 08:08:03 07/29/08 (10)
In Reply to: RE: cPlay - the open source high-end audio player using ASIO posted by cics on May 05, 2008 at 12:31:58
Dear cics,
I have been listening to the single 1gB memory for the last few weeks and I think I like it better. It does seem to have a less "electrical" sound. Of course, I should go back and try the dual set-up but have been enjoying this SO MUCH I am happy to leave it alone.
Since you are far more intrepid than I I wonder what are your thoughts?
I have ordered some capacitors (Panasonic FC) to bypass the various voltage rails on the MB per Alfred's suggestions. One part of me thinks I should solder directly to the MB instead of to the wiring harness. Do you think this would be extremely dangerous? I have no idea what all is going on within these things; are there more than two layers of traces?
I would remove the processor but is there a danger of the voltages from the doldering tool damaging other components?
I would appreciate your opinion on both of these questions.
Hope all is well.
Thanks,
Rick McInnis
- RE: cics, single or dual memory preferred? - sonics 09:12:48 07/29/08 (9)
In Reply to: RE: cics, single or dual memory preferred? posted by rickmcinnis@dogwoodfabrics.com on July 29, 2008 at 08:08:03
hallo,
single memory in my system sounds better than dual ram.I tried it in 3 PCs.
If you want to do the condensatormodification do the cond. between the wires. It is much simpler , you cant hurt the Mobo and i see no advantage in soldering it into the mobo.. Finally i put the cond. to all three voltages and sound is much smoother.I think alsolooking for a very good PSU like cics suggesttion ( Antec )is worth considering, but there are many ways and much more PSUs to try.
If you can, do many of the Panasonics FM or FC ( for example 10x1000µF parallel to each rail).
Bye
Alfred
- RE: cics, single or dual memory preferred? - rickmcinnis@dogwoodfabrics.com 12:56:06 07/29/08 (8)
In Reply to: RE: cics, single or dual memory preferred? posted by sonics on July 29, 2008 at 09:12:48
Alfred,
Hope you enjoyed your vacation!
Thanks for your advice.
I would agree than paralleling capacitors to be a good idea but I worry that your cables could be cumbersome!
If I may get some additional information as to how you do this: I assume you are placing the capacitors as close as possible to the header?
Would you consider posting a photgraph of how you did this? Ten capacitors per "port" - I see 14 different volatage "ports" (I do not know what else to call them) on the main header. Are you putting ten capacitors in parallel on each of these?
I am thinkning about using some BLACK GATES on the 12 volt header neat the processor. Overkill. Perhaps?
Thanks in advance!!!
Rick McInnis
- Single module is better option - cics 11:24:11 07/30/08 (7)
In Reply to: RE: cics, single or dual memory preferred? posted by rickmcinnis@dogwoodfabrics.com on July 29, 2008 at 12:56:06
Alfred's implementation is cleaner. I wouldn't mess with the mobo - these things are multi-layered.
A picture would be very useful.
- BIOS refinements - rickmcinnis@dogwoodfabrics.com 13:00:50 07/30/08 (5)
In Reply to: RE: Single module is better option posted by cics on July 30, 2008 at 11:24:11
Dear cics,
Do you have any idea when GIGABYTE will make available the BIOS software you have been using?
Is this moving along?
Will it coincide with the option of diabling SPREAD SPECTRUM?
Just curious.
THANKS,
Rick McInnis
- Spread Spectrum should be available once their testing is done. They can't do another change I was looking for - cics 00:18:36 07/31/08 (4)
In Reply to: RE: BIOS refinements posted by rickmcinnis@dogwoodfabrics.com on July 30, 2008 at 13:00:50
Cannot tell when they'll make this available.
- RE: Spread Spectrum should be available once their testing is done. They can't do another change I was looking for - sonics 03:59:18 07/31/08 (3)
In Reply to: RE: Spread Spectrum should be available once their testing is done. They can't do another change I was looking for posted by cics on July 31, 2008 at 00:18:36
hi,
i ordered the antec earth watt PSU to compare with my other PSUs.Perhaps i can send some photos with the kondensatormodifikation and the new PSU.
What do you think about PCI-E Soundcards like the ASUS Xonar D2X. http://www.bit-tech.net/hardware/2008/01/13/asus_xonar_d2x_pci-express_soundcard/1.
it supports ASIO 2.0.Perhaps it works with cplay. ( i still love my juli)
bye
alfred
- ASUS Xonar D2X PCI-e impressions - carcass93 13:48:32 07/31/08 (2)
In Reply to: RE: Spread Spectrum should be available once their testing is done. They can't do another change I was looking for posted by sonics on July 31, 2008 at 03:59:18
That's what I settled on, after trying EMU-1212M (extensively) and ESI Juli@ (briefly) - I don't share enthusiasm on that one, sorry. As with previous two, I use SPDIF out to output 16/44 signal into upsampling DAC. Hardware - cics' transport (no cMP, no cPlay) based on e2140/P4M900/HyperX, software - Winamp + recompiled Otachan ASIO.Some random thoughts:
1. Additional 4-pin floppy-style power connector (meaning it doesn't feed off PCI-e slot).
2. Two apps that are put into autorun in registry can be safely deleted from there - not needed for audio playback, so nothing in memory related to the card itself. That's departure from both EMU and Juli@.
3. Level of the signal is significantly lower than with both EMU and Juli@ - not a problem, just makes comparisons a little bit difficult.
4. ASIO driver is work in progress (that is, I hope it's in progress) - it's in Beta, and there's audible one-time "pop" any time you start playback, i.e. after STOP or PAUSE.
5. I was able to drop buffer size in Otachan ASIO to 0 samples - not achievable with EMU (min. 1) or Julia (don't remember, but much higher - in 30s).
6. Stupid back-lights through analog outs, that I immediately sealed with black electrical tape. Without that, the wall behind my rack was like freaking christmas tree.
7. "Open Box" from Newegg was something like $120, plus tax and shipping. "Openness" of the box, by the way, in this case means that it's bare card - no CD, no docs, nothing. Not that you need them though, everything's downloadable.
8. Obviously, sounding superior to the other two (in my system, to my ears).
I also tried PCI version of Xonar - that one went back immediately, since when using Otachan ASIO, instead of signal there was just some kind of digital noise. I guess ASIO driver for that one is in even earlier stages than PCI-e version...
Edits: 07/31/08 07/31/08
- RE: ASUS Xonar D2X PCI-e impressions - sonics 02:39:09 08/01/08 (1)
In Reply to: RE: ASUS Xonar D2X PCI-e impressions posted by carcass93 on July 31, 2008 at 13:48:32
thanks for your information.
I tried also emu 1212 and juli and juli was superior used with cmp and cplay.
I wonder if xonar with exteral clean powersupply via floppy cable would gain sonic performance- and will work with cplay-?
So i have to spend the money and try it.
bye
alfred
- Exteral power supply - possible will be beneficial. My experiment.... - carcass93 07:16:58 08/01/08 (0)
In Reply to: RE: ASUS Xonar D2X PCI-e impressions posted by sonics on August 01, 2008 at 02:39:09
with CRAPPY external power supply (switching, plugged into the same line as the one that feeds hard drive) yielded significant performance degradation - that tells me that the other way around is also possible.
- Alfred, would you try to show us how you have ... - rickmcinnis@dogwoodfabrics.com 12:57:56 07/30/08 (0)
In Reply to: RE: Single module is better option posted by cics on July 30, 2008 at 11:24:11
implemented these capacitors?
I know I could simply do it, but I have the intuition you have evolved a technique for this and we would greatly benefit if you could SHOW us how you did it!
I went to the link you provided me and I see the diagrams for the MB voltage cable header. Is this what you intended for me to see? I have been using the diagram within the GIGABYTE ownder's manual.
THANKS cics for the advice against direct soldering to the MB. To be honest I wanted someone to tell me ABSOLUTELY NOT since it would be a rather delicate task.
I fear placing the caps within the wiring harness will be rather clumsy but that is nothing compared to destroying the MB!!!
I know I had the impression that the music sounded better with the single memory but I do have good reason to doubt myself. Thanks for the reassurance, cics!
Thanks to both of you,
Rick McInnis
- Finally got .978 drivers to work & sw/hw questions - theob 05:20:19 07/28/08 (0)
In Reply to: RE: cPlay - the open source high-end audio player using ASIO posted by cics on May 05, 2008 at 12:31:58
But other users of juli should be aware that juli will not survive a standby power down, you have to reboot. At least that is what the esi user community forum says. So I am not worried about that aspect of juli. But the .978 drivers do seem more stable on my system. 4 days of no gross distortion and no loss of julapan.exe. Sound @176/48 latency is excellent.
Btw I found the major source of tics/pops: 2 residual processes from Logitech camera lvcomser/lvcomserv (or something like that) plus indexing on my hard drive. When I chose no indexing and uninstalled logitech I am virtually pop free.Question to cics: I got an old computer from my daughter that has xp pro and an intel celeron 667 mhz cpu, socket 370 fc pga, 256 ram. Can I use to build up to cmp? Since this computer has xp pro can I transfer it to my computer (d/l to a flash drive and upload to my pc?) I need xp pro to get awe capability. Its not sp2 though. I know 667 mhz is slow but couldn't I reset to higher speeds by overclocking?
--------------------------------------------------------------------------------
RE: Step 13: You must select the folder within
Edits: 07/28/08
- RE: cPlay - the open source high-end audio player using ASIO - riboge 11:20:57 07/17/08 (5)
In Reply to: RE: cPlay - the open source high-end audio player using ASIO posted by cics on May 05, 2008 at 12:31:58
I have been following the development of cPlay along and using it on a couple different systems. There are limitations I just don't understand not being deeply competent in computers or programming:
1. Most of all, why if you have 3-4GB or RAM can you not get at least 2GB of RAM loading using cPlay? I get that AWE seems to impose an additional limitation of RAM available for this to 1GB, but even without it I have never gotten more than about 1.4GB cuelist to load. This may be sufficient for most CDs (though not multi-CD items) but it rules out most classical recordings at 24/96. Surely if users want to upsample to 96 they would want to avail themselves of 96 recordings, which would have to be potentially even better. In this case you have files that RAM load and play without processing while playing, for instance, as some have expressed interest in. Why is this not easily achieved having gotten this far? Is it just that cics feels no need to pursue this or some intrinsic limitation?
2. Why must the cPlay window and its font be fixed in size, which is quite suboptimal for anyone not using a small touchscreen? Again, is it a real problem to add an adjustment option or just that cics isn't willing to do it? And if the latter, why is no one else who is able willing to either?
3. The same goes for loading a series of files from a list to play in sequence. With 3GB of RAM one should be able to load a first 1GB of RAM to start playing and then while it is playing a second 1GB RAM loaded to play when the first is finished at which point the first 1GB starts loading the third file and so on. This would make virtually all recordings playable as well as a playlist of different recordings. What is the limitation? Why has no one taken this on?
- Available RAM, UI & Playlists - cics 23:17:11 07/19/08 (4)
In Reply to: RE: cPlay - the open source high-end audio player using ASIO posted by riboge on July 17, 2008 at 11:20:57
why if you have 3-4GB or RAM can you not get at least 2GB of RAM loading using cPlay?
This is controlled by Windows where more RAM is provided depending on installed capacity. Unfortunately, you want to limit RAM installed.
Why must the cPlay window and its font be fixed in size
UI can be changed. This needs development work. cPlay uses a simple construct. I need other developers to take this forward.
The same goes for loading a series of files from a list to play in sequence.
As above - playlist can be developed too.
- RE: Available RAM, UI & Playlists - riboge 11:09:30 07/20/08 (0)
In Reply to: RE: Available RAM, UI & Playlists posted by cics on July 19, 2008 at 23:17:11
Windows controls how much RAM is available for RAM loading via cPlay, you say, as a function of total RAM available. Is this not alterable or circumventable by any possible change of settings, bios, or whatever? Seems implausible to me, but then I am not informed enough to be a good judge.
Merging with a player with good GUI does seem to be a promising idea, if possible. What is so great about Album Player?
- RE: Available RAM, UI & Playlists - frennekamp@netcologne.de 06:01:34 07/20/08 (2)
In Reply to: RE: Available RAM, UI & Playlists posted by cics on July 19, 2008 at 23:17:11
Actually I had contact with the devloper of album player and he was very interested to integrate the technology of cplay into album player - has he contact you ? This would give us the player with one of the best GUIs with the best sound...if you allow that
- Great idea. I'll lookout for his email (nothing yet). - cics 21:21:42 07/20/08 (1)
In Reply to: RE: Available RAM, UI & Playlists posted by frennekamp@netcologne.de on July 20, 2008 at 06:01:34
.
- RE: Great idea. I'll lookout for his email (nothing yet). - frennekamp@netcologne.de 15:54:42 07/25/08 (0)
In Reply to: RE: Great idea. I'll lookout for his email (nothing yet). posted by cics on July 20, 2008 at 21:21:42
Its:
Peter van der Burg (AlbumPlayer Support)
support@albumplayer.com
BEst Regards
Frank
- RE: cPlay - the open source high-end audio player using ASIO - w.rade 20:47:11 07/14/08 (7)
In Reply to: RE: cPlay - the open source high-end audio player using ASIO posted by cics on May 05, 2008 at 12:31:58
Hello everybody. Since a few years I am using PC for listening to music. Some time ago I build a PC with a mobile cpu and the AOpen i855GMEm-LFS motherboard. I am just starting to use cmp and can only say "it's absolutetly amazing". Now with cmp and cplayer I encounter a strange problem. I have no problem, opening wav-files. The first days I could sometimes get cuefiles started, sometimes not. Now cplay crashes everytime, I try to open a cuefile, but it's not because of the size (I have 1 GB). There is no fault reported in the log, but I loose connection with the emu-soundcard. The player processes wav-files after the crash but there is nothing happening on the emu patch-mix and of course nothing to hear. Only after a restart of the PC I can at least listen to wav. Maybe someone can relate to it? By the way. I am from germany and happy to see, that there are more people heading for this way. I am just a music lover and thrilled by what I read in this cmp-thread. Thanks to cics for his approach and commitment. Walter
- Don't suspend SVCHOST and LSASS (in CMP) with EMU cards - cics 02:37:23 07/15/08 (6)
In Reply to: RE: cPlay - the open source high-end audio player using ASIO posted by w.rade on July 14, 2008 at 20:47:11
When this happens, in Process Explorer you'll see cPlay.exe - just kill it to return to cMP.
Unfortunately, EMU drivers come with a great deal of bloat causing indefinite waits / locks. cPlay tries to accommodate this with timer delays - this is only required for EMU and no other soundcard.
For example, go to Device Manager > View > Show Hidden Devices > Non Plug and Play devices. You should see about 15 additional devices from EMU / Creative. This is very bad design and here's the bad news: to remove it needs a new windows install.
My suggestion is to rather go with another soundcard but this would need a Windows reinstall.
- RE: Don't suspend SVCHOST and LSASS (in CMP) with EMU cards - w.rade 16:32:12 07/15/08 (3)
In Reply to: RE: Don't suspend SVCHOST and LSASS (in CMP) with EMU cards posted by cics on July 15, 2008 at 02:37:23
"When this happens, in Process Explorer you'll see cPlay.exe - just kill it to return to cMP" = this is exactly what I did. I can return to cMP but loose the sound. I better try the ESI card. I have heard it already and seems to be a excellent soundcard. Today I tried a new installation of XP and set the bios to default but it didn't help. Since I have a large collection of music (4TB), I am looking for a solution like I am using right now with two usb-cases holding each up to 4 harddrives wich I can connect easaly to my system. USB seems not to be the preferred way, so I wonder if something similar existeds on SATA. Does someone have an opinion about it?
Walter
Edits: 07/15/08
- RE: Don't suspend SVCHOST and LSASS (in CMP) with EMU cards - w.rade 02:42:40 07/26/08 (2)
In Reply to: RE: Don't suspend SVCHOST and LSASS (in CMP) with EMU cards posted by w.rade on July 15, 2008 at 16:32:12
just want to say, that I got the juli@- card and finally can load cue-sheets. Thanks for the recommendation. Another point, where I took a decision, was to solve my 4TB-problem. I started now, to convert them to flac, since they get played out of ram. Before cMP I found flac not as natural sounding when processed by the cpu. So far I feel greatful about the performance of this system. To be true. I was once a strong believer in the analog music reproduction and had for years no reason, to change for digital. When I did, there was still a remeberence of a quality, which I missed. Now it seems, I am at last on the winning side. Thank you so far for
the refind input. Walter
- Great feedback - yes this is very analogue. On 4TB (or less), have you tried eSata? - cics 02:56:06 07/26/08 (1)
In Reply to: RE: Don't suspend SVCHOST and LSASS (in CMP) with EMU cards posted by w.rade on July 26, 2008 at 02:42:40
Try avoiding mobo's USB ports.
- RE: Great feedback - yes this is very analogue. On 4TB (or less), have you tried eSata? - w.rade 13:15:33 07/27/08 (0)
In Reply to: RE: Great feedback - yes this is very analogue. On 4TB (or less), have you tried eSata? posted by cics on July 26, 2008 at 02:56:06
My Mobo provides only 2 SATA-ports. If I don't want to downsize my collection or switch harddisks I would need a mobo with 3 to four ports. But I will try the eSATA anyways. Maybe it leads me to the point, to accept for a while the compromize to have limited access to my soundfiles. There are boards with up to 5 SATA-ports available. The problem will be, that most of them have no graphic on board.
- Agree about EMU software - couple of points though. - carcass93 07:48:39 07/15/08 (1)
In Reply to: RE: Don't suspend SVCHOST and LSASS (in CMP) with EMU cards posted by cics on July 15, 2008 at 02:37:23
More than half of these devices are actually not needed for audio and can be disabled. Also, if the card is removed and software uninstalled, they still remain there, but there's no need to re-install Windows - you just need to uninstall them one-by-one manually.
- Didn't try that approach but its good to know of a work around without doing reinstall. Thx. - cics 00:20:52 07/16/08 (0)
In Reply to: RE: Agree about EMU software - couple of points though. posted by carcass93 on July 15, 2008 at 07:48:39
.
- Warning: Anybody thinking AMD - don't buy Biostar A690G board. - carcass93 10:48:15 07/14/08 (0)
In Reply to: RE: cPlay - the open source high-end audio player using ASIO posted by cics on May 05, 2008 at 12:31:58
My current setup is cics' transport (no cMP, no cPlay) using e2140 and Biostar P4M900 mobo, with Winamp + Otachan ASIO (recompiled to take advantage of SSSE3) as player. Just for fun, decided to check out what AMD setup can do and bought 4050e CPU and Biostar A690G mobo.
A690G WOULD BE an excellent board - supports multiplier and core voltage adjustments (I dropped it to 5x and 0.75V respectively, and system is stable), and all kinds of memory timings. There's only one big BUT(t):
Damn thing is unable to retain some of memory timings, namely Tcas, Tras, Trc. You'll change them, and it will still keep the defaults, which by the way are ridiculous 5, 18, 26 respectively. Funny enough, it retains all other memory timings. Needless to say, updating BIOS didn't help.
So, A690G goes back, and I will probably try A740G in place of it, which has different BIOS (AMI). One thing I noticed about AMD setup is how much cooler it is compared to e2140/P4M900, including memory (the same one stick of Kingston HyperX 1 Gb). On the other hand, AMD is in inherent disadvantage in that it only supports SSE3 (no SSSE3, no SSE4) - it really made audible difference, when I compared different ASIO builds.
- This gigabyte board is a pain... - frennekamp@netcologne.de 11:12:01 07/13/08 (4)
In Reply to: RE: cPlay - the open source high-end audio player using ASIO posted by cics on May 05, 2008 at 12:31:58
...so, I followed your advice an bought the e7200 and the gigabyte board...set the bios to the settings you specified...
I have a OCZ 2GB XTC 4-4-4-12 800 mhz DDR2. The Gigabyte runs it even at 200mhz / 400 mhz at only 6-6-6-18 at 2T.
The P4M900 allowed 3-3-3-9 1T, which was very audible.
I found the secret Bios-Setting of the Gigabyte which is press ctrl-F1 when you are in the main setup-page. I got the memeory running at 5-5-5-15, but after that the bios somply resets.
As well, get below 200mhz FSB is impossible. And even this setting is not always possible, the board resets quickly to 266mhz, you donÄt really have control.
Soundwise...too early to judge. It is playing now for 4 hours. I will definitely compare it to the 21800 / P4m900 at 600 mhz.
as well, when you are in advanced mode (ctrl-F1), it show which interrupts are shared in PCI1 and PCI2 and no PCI-Port seems to have an exclusive interrupt...not sure, how this can be set up in the right way though...
I asked myself which board I should try next, where I can really get the latency from my memory and use the 45nm e7200 and habe spread spectrum disabled and can heavily underclock the e7200...any hints ? Gigabyte seem not to be ideal.
- RE: This gigabyte board is a pain... - w.rade 13:56:35 07/27/08 (0)
In Reply to: RE: This gigabyte board is a pain... posted by frennekamp@netcologne.de on July 13, 2008 at 11:12:01
what are the experiences with this GA-EG31M-S2 so far? Which cpu seems to be preferebal? This board is interesting for me since it provides 4 sSATA connectors.
- RE: This gigabyte board is a pain... - cics 01:21:55 07/14/08 (2)
In Reply to: RE: This gigabyte board is a pain... posted by frennekamp@netcologne.de on July 13, 2008 at 11:12:01
At 96k output, FSB of ~170MHz should be very stable (RAM @ 3-3-3-9).
At 192k output, leave FSB at 200MHz.
Can you try another RAM?
You may look into Biostar's G31 offering using AMD BIOS. Gigabyte's EG31 mobo may be interesting.
- RE: This gigabyte board is a pain... - appleteapot 07:07:16 07/14/08 (1)
In Reply to: RE: This gigabyte board is a pain... posted by cics on July 14, 2008 at 01:21:55
Similar experience here... I "migrated" from Asus Commando+E6320+HyperX 2x1G DDR2-800+Asus G7300 silent to E7200+G31+HyperX combo. I used to run Commando at 600Mhz and it was very stable. Now with E7200/GA-G31M-S2L/HyperX I used the same RAM and HDD as before. Asus Commando was P965 SLI-ready mobo designed for overclock, while G31 is a All-in-one MATX board.
(1) The motherboard wouldn't boot the at first time. My HyperX DDR2-800 RAM pair requires 1.9V, and I had to use a cheap RAM to boot up the motherboard, change RAM Voltage to 1.9 before I insert HyperX.
(2) The motherboard cannot go below 1.2GHz. I tried 6x-9x multiplier and base speed cannot go below 200Mhz
(3) The sound stage of E7200/GA-G31M-S2L at 1.2Ghz is not as good as E6320/Commando at 600Mhz, but is MUCH MUCH better than Commando at 1.2Ghz.
(4) E7200 is incredibly cool at 27 degree. However, the heat is on G31 chip and RAM and I still need a fan on the case. Even so, G31 is much cooler than Commando.
(5) I don't know whether it's related but one of my 3 HDD's crashed last night. I need to fix it later.
(6) the heat generated by Corsair HX 620 PSU are the same for both E7200/G31 or E6320/Commando - cool and quiet.
- Allow burn-in for about 200-300 hours at default FSB. Then try underclocking. - cics 02:43:22 07/15/08 (0)
In Reply to: RE: This gigabyte board is a pain... posted by appleteapot on July 14, 2008 at 07:07:16
Also try Gigabyte support.
I wanted some BIOS changes and I've received an experimental BIOS release (F6x) which allows for Spread Spectrum to be disabled. There's one more BIOS change I need... didn't expect this great service.
- RE: cPlay - the open source high-end audio player using ASIO - Doc Gaw 09:29:11 07/12/08 (1)
In Reply to: RE: cPlay - the open source high-end audio player using ASIO posted by cics on May 05, 2008 at 12:31:58
Got the Juli card up and running and it does indeed sound better than the M-Audio firewire unit previously used. Running to Lexicon 12B and Integra 9.8 pre-pro's and both sound great decoding a 24/176 stream from Juli card.
Am using Vista which doesn't have k-mixer. Would it be possible to produce cPlay or some similar program with RAM playback which could use the motherboard's SPDIF output at 24/192 or 176 without the necessity for card or ASIO, for external D/A conversion by pre-pro? Seems to me this might further drop noise and jitter as only the motherboard would be needed.
- Mobo SPDIF output can be used via ASIO4ALL - cics 01:29:55 07/14/08 (0)
In Reply to: RE: cPlay - the open source high-end audio player using ASIO posted by Doc Gaw on July 12, 2008 at 09:29:11
Juli@ is excellent value and comes with quality drivers.
There's no need for anything special to use mobo output which is a PCI device (built-in soundcard). I disable this in favor of other PCI cards.
ASIO4ALL is a wrapper over non ASIO devices and works very well. I can't get it to perform beyond 48k even with mobo sound hw supporting 192k.
1st prize is getting Realtek to provide native ASIO drivers. This will be better.
- cPlay 1.0 Final Released - cics 02:46:15 07/12/08 (57)
In Reply to: RE: cPlay - the open source high-end audio player using ASIO posted by cics on May 05, 2008 at 12:31:58
Change Log (final release):
- Change labels (b19 to final)
- Fix for Lynx soundcards
Please REMOVE previous versions before installing cPlay 1.0 Final .
cPlay was tested using Lynx AES16-XLR. Firmware revision 25 (update23) and driver versions 14e & 16 (current). Stable performance achieved (no dropouts) doing 24/192 Dual AES output to Scarlatti DAC. Latency set to lowest of 32 samples. In cMP², Lynx soundcard may perform better with SynchroLock™ feature disabled (when using internal clock).
- I have 2 of each subsection of juli files in my c drive - theob 06:19:38 07/18/08 (0)
In Reply to: RE: cPlay 1.0 Final Released posted by cics on July 12, 2008 at 02:46:15
Is this the source of my problem? I'll try to delete redundant files from my memory.It seems I am still not w/o corrupted drivers. Let me ask you a question about the 18 step process you outlined. When I download 'the compressed drivers' from the esi site I save the compressed files right from the start in c:\juli@, then I unzip by right clicking on the folder and extracting. Is this correct? Also do you think because my hard drive is always 'on' with the gd outboard ps that this would make a difference?
I'm getting good at all the 18 steps except I get no success apparently.More data on this issue. After a bit of idle time my system drops julapan.exe. Its not on the tool bar nor is it in task manager. When this happens I get noise. But if I go to the c drive and find julapan.exe and run it I'm ok. Does that suggest anything? It's almost as if the system remembers the crash and tries to get rid of this driver whenever it can. Make any sense?
Edits: 07/18/08
- Is this LYNX card the better solution? - rickmcinnis@dogwoodfabrics.com 08:16:07 07/15/08 (3)
In Reply to: RE: cPlay 1.0 Final Released posted by cics on July 12, 2008 at 02:46:15
Dear cics,
I looked up this card and see that it is much more expensive than our friend JULI@. Is it worth the extra money?
My DAC would accept an XLR input so that is not a concern.
I know you are still letting it settle in, BUT, are you hearing something better?
I would like to know.
I will recheck my CPU voltage tonight.
THANKS, as always,
Rick McInnis
- RE: Is this LYNX card the better solution? - cics 00:19:10 07/16/08 (2)
In Reply to: RE: Is this LYNX card the better solution? posted by rickmcinnis@dogwoodfabrics.com on July 15, 2008 at 08:16:07
Scarlatti DAC can only do 24/192 via dual AES links which is the only reason I got this card. The other option was to get an RME card - since I have one already, I thought its better to try something new.
Will it better the Juli@? I doubt it as Juli@ offers excellent low latency playback at 48 samples (difference between 32 and 48 becomes less important with such a high sampling rate). Had my DACs handled 24/192 through coax I would be using Jul@. Unfortunately I can't do a direct comparison.
There's more performance to come from Gigabyte mobo - I have an experimental BIOS version that allows for Spread Spectrum to be disabled. Juli@ improves on this.
- Are you actually getting 48 samples latency playback with the Gigabyte board/Juli@ now? - seger 10:44:12 07/18/08 (1)
In Reply to: RE: Is this LYNX card the better solution? posted by cics on July 16, 2008 at 00:19:10
Previously you said the board forced you to use 128 samples, so how have you overcome that? I ask because the low latency is critical for kernel streaming, which I still want to use some of the time.
- With BIOS update (F6) no glitches at all at "48" - rickmcinnis@dogwoodfabrics.com 12:22:23 07/18/08 (0)
In Reply to: RE: Are you actually getting 48 samples latency playback with the Gigabyte board/Juli@ now? posted by seger on July 18, 2008 at 10:44:12
With the "original" BIOS it was a mess, but with the update, I have never had a single hiccup.
No longer a concern.
Rick McInnis
- RE: cPlay 1.0 Final Released - n_vincent 17:28:32 07/14/08 (4)
In Reply to: RE: cPlay 1.0 Final Released posted by cics on July 12, 2008 at 02:46:15
CICS,
I'm currently running my cmp setup with a rme hdsp32 and thinking of going to a lynx aes (or rme hdsp aes) for finding my dcs delius in dual aes mode. Does the improvement of going from rme hdsp 32 at 96 khz to lynx aes at 192 in dual aes mode will be very significative ?
Why did you not go for the lynx aes (pci bus) and not the pci express one ?
Nicolas from Paris
- RE: cPlay 1.0 Final Released - riboge 19:42:04 07/14/08 (3)
In Reply to: RE: cPlay 1.0 Final Released posted by n_vincent on July 14, 2008 at 17:28:32
I don't believe from emails with him cics has had much experience with the Lynx yet. I have had more. Dual aes comprising 192 does sound audibly better than single sending 96. Cics has in mind having dual aes of Lynx comprising 384 by doubling 192 output. I have my doubts that this will make much of an improvement with current sources starting at 44.1 or 96.
The Lynx pci express card is brand new so few have experience with it, but I would assume it would be comparable to the pci card. It uses the same drivers, etc.
- Using Lynx AES16-XLR Dual AES output at 192k - cics 02:58:57 07/15/08 (2)
In Reply to: RE: cPlay 1.0 Final Released posted by riboge on July 14, 2008 at 19:42:04
Card is still burning-in. I'm also waiting for a new Gigabyte mobo before installing in reference system. Not going to 384k at all - 192k is perfect.
Initial testing shows promise using headphones from Scarlatti outputs. Galvanic isolation is lost but 192k at 145db SNR is a big benefit. Lynx h/w and drivers seems very robust and can handle 192k output at 32 samples latency - the best I've seen! I'm currently running this 24x7 without any problems.
As to PCIe - not sure how this would handle low latencies as PCIe is better used with large payloads and fewer transactions. Still new and need to see more reviews.
- RE: Using Lynx AES16-XLR Dual AES output at 192k - riboge 13:25:55 07/15/08 (1)
In Reply to: RE: Using Lynx AES16-XLR Dual AES output at 192k posted by cics on July 15, 2008 at 02:58:57
I suppose you realize that if cPlay is set to 192 and you use dual output from your Lynx aes, this results in your DAC seeing 384 combining its two inputs. So do you mean you set cPlay to 96 and use dual output or set cPlay to 192 with single output?
- RE: Using Lynx AES16-XLR Dual AES output at 192k - n_vincent 00:03:00 07/16/08 (0)
In Reply to: RE: Using Lynx AES16-XLR Dual AES output at 192k posted by riboge on July 15, 2008 at 13:25:55
192 k is achieve with dual aes link 96 for each link 96x2
384 k is achieve with quad aes link 96 for each link 96x4
No consumer dac do that, a Dcs pro dac can do that in quad link
- Is there any advantage to installing this w/o Lynx card? - rickmcinnis@dogwoodfabrics.com 09:31:53 07/14/08 (46)
In Reply to: RE: cPlay 1.0 Final Released posted by cics on July 12, 2008 at 02:46:15
Dear cics,
From what I see there is most likely no reason to do this unless one is using the Lynx card? Or should it be done anyway?
I do think I like the single memory better. I do need to go back and forth to make an assertion, but at this point there seems to be added naturalness, less electrical sound. Of course, I worry that I WANT it to sound better; an audio placebo. I guess as long as I can convince myself of this I might as well enjoy it until ...
My CPU voltage is lower than what you show on your cpuz "photos". I figure this is good? I am using MAX BATTERY power
THANKS,
Rick McInnis
- RE: Is there any advantage to installing this w/o Lynx card? - cics 03:08:05 07/15/08 (45)
In Reply to: RE: Is there any advantage to installing this w/o Lynx card? posted by rickmcinnis@dogwoodfabrics.com on July 14, 2008 at 09:31:53
Final release fixes an issue with Lynx drivers. No sonic benefit.
What voltage are you getting with 24/192 output? It should be higher then 1.120V (closer to 1.184V).
- RE: Is there any advantage to installing this w/o Lynx card? - theob 03:20:17 07/15/08 (44)
In Reply to: RE: Is there any advantage to installing this w/o Lynx card? posted by cics on July 15, 2008 at 03:08:05
I installed a juli card uccessfully yesterday and played the system for 4 hours. This morning cplay boots up loads a file and diagnostics seem ok but it doesn't play. It just sits there w/o moving the clock and no sound.
- RE: Is there any advantage to installing this w/o Lynx card? - cics 03:27:59 07/15/08 (43)
In Reply to: RE: Is there any advantage to installing this w/o Lynx card? posted by theob on July 15, 2008 at 03:20:17
Also test with foobar to see if soundcard is fully operational.
What happens when you hit pause/play button?
- RE: Is there any advantage to installing this w/o Lynx card? - theob 03:38:48 07/15/08 (42)
In Reply to: RE: Is there any advantage to installing this w/o Lynx card? posted by cics on July 15, 2008 at 03:27:59
After I reboot, the system asks me if I want to run a file (for juli), it must be a driver or something because after I run it loads everything is back to normal.
- RE: Is there any advantage to installing this w/o Lynx card? - theob 03:43:39 07/15/08 (41)
In Reply to: RE: Is there any advantage to installing this w/o Lynx card? posted by theob on July 15, 2008 at 03:38:48
Could it be possible that my .979 driver is only on the desktop and not where its supposed to be (where is that btw?). Sould the driver be placed in another folder?
- Try reinstalling drivers per Rick's instructions. Then reboot to ensure you get no other system requests. - cics 03:55:34 07/15/08 (40)
In Reply to: RE: Is there any advantage to installing this w/o Lynx card? posted by theob on July 15, 2008 at 03:43:39
Juli@'s 0.979 driver should work 100% on XP SP2 Pro.
Are you using SRC@145db SNR?
- RE: Try reinstalling drivers per Rick's instructions. Then reboot to ensure you get no other system requests. - theob 04:00:51 07/15/08 (39)
In Reply to: RE: Try reinstalling drivers per Rick's instructions. Then reboot to ensure you get no other system requests. posted by cics on July 15, 2008 at 03:55:34
I am running -146 db src at 176 khz. Interesting for 16/44 files (remember I have cool & quiet which varies the processor speed by load) I run 1800 mhz versus 2300 mhz with usb feed at 24/96 settings. Sounds great!! Cant I just move the driver to the right file? I went through an emotional experience fooling with the re-install yesterday, would like to avoid. Also I have sp2 home edition not pro, didn't take that step yet.
Edits: 07/15/08
- RE: Try reinstalling drivers per Rick's instructions. Then reboot to ensure you get no other system requests. - theob 04:21:29 07/15/08 (38)
In Reply to: RE: Try reinstalling drivers per Rick's instructions. Then reboot to ensure you get no other system requests. posted by theob on July 15, 2008 at 04:00:51
Reinstalled .979 per rick and rebooted. Upon reboot it asked me to run the juli panel driver. somehow that didn't get installed properly. I'll try again this time reinstalling the panel file from the disk. Sound right?
- RE: Try reinstalling drivers per Rick's instructions. Then reboot to ensure you get no other system requests. - theob 04:30:30 07/15/08 (37)
In Reply to: RE: Try reinstalling drivers per Rick's instructions. Then reboot to ensure you get no other system requests. posted by theob on July 15, 2008 at 04:21:29
I reinstalled the juli panel file, rebooted and it still asks me if I want tun run julipan.exe at startup. why is that? and what happens if I say no?
Edits: 07/15/08
- RE: Try reinstalling drivers per Rick's instructions. Then reboot to ensure you get no other system requests. - theob 04:34:39 07/15/08 (36)
In Reply to: RE: Try reinstalling drivers per Rick's instructions. Then reboot to ensure you get no other system requests. posted by theob on July 15, 2008 at 04:30:30
just noticed I have 2 julipan .exe running on task manager. So I delete one and I should be ok but why does it keep prompting me to run julipan.exe?
- RE: Try reinstalling drivers per Rick's instructions. Then reboot to ensure you get no other system requests. - theob 06:17:58 07/15/08 (35)
In Reply to: RE: Try reinstalling drivers per Rick's instructions. Then reboot to ensure you get no other system requests. posted by theob on July 15, 2008 at 04:34:39
After system standby kicks in I lose (apparently) julipan.exec again. So I rebooted and it asks me if I want to run it again. I have a propsed workaround to set my control panel power to never go into standby or shut off monitor or hard drives but I would like to eventually fix this. Any ideas?
When I was doing initial system setup off the juli disc I ran into a hard system shut down, could I have corrupted something then? After windows asks me if I want to run julipan.exe it mentions that d/ling files from the internet can be a problem. Did I go into some kind of system protect because of this system crash and maybe windows is being cautious? I'm just speculating now I have no idea whats going on, I am obviously not a computer guy but since I can get it to work I'll just limp along this way until I come up with some explanation.
- It is best to start over - rickmcinnis@dogwoodfabrics.com 07:58:35 07/15/08 (34)
In Reply to: RE: Try reinstalling drivers per Rick's instructions. Then reboot to ensure you get no other system requests. posted by theob on July 15, 2008 at 06:17:58
Theo,
Some of what you are experiencing sounds familiar, but other things, I do not think I have experienced the same.
I have found the more times you go through the install process the better you get at it. I think on my fifth install I finally REALLY got it. I had never worked with computers before being inspired by cics to do this. I screwed things up plenty of times in the process of learning what was happening.
From what you are saying the first thing I would check is that cPLAY is set for channels 3,4. But the fact that the "timer" sits still does sound like something bigger than that.
You should uninstall everything.
I think if you start completely over you will be better off.
I have had two ESI buttons in my tray before. One of them will go away eventually. At the moment, from my last install (and probably my cleanest) I have had only one.
For peace of mind in knowing that all is correct, just start over completely. It will go much better the second/third time.
Look in DEVICE MANAGER, too, to make sure all vestiges of JULI@ are gone before starting over.
Be patient and deliberate and all will go well!
Rick McInnis
- Cplay played with very high distortion-I have a problem - theob 04:19:44 07/16/08 (31)
In Reply to: RE: It is best to start over posted by rickmcinnis@dogwoodfabrics.com on July 15, 2008 at 07:58:35
This morning when I tried playing cplay it played with very high distortion superimposed on the high frequencies(sounded like a bell superimposed on music). So I figured it was time to try again to reinstall juli audio and juli panel drivers.
So I tried for my 5th time to unistall then reinstall drivers. This time it installed ok but after boot up it asks if I want to run the julipam file which they label as a risk with the message 'open file warning'. This was always here but I never paid attention to it before. But I said yes and things seem to be temporarily fine, cplays ok now.Rick did you have julipan.setpan file on your desk top(should I delete it)? I feel this maybe related to my problem. Also once you installed .979 did you delete the place where you temporarily stored the driver for it?
I feel this is not right yet but am looking for some suggestions on how to fix. I still have two julipan files rumnning in task manager.
Tried a sixth time this time deleted my jula file on desktop and went back to d/l the .979 file again and replace on my desktop. After I uninstalled drivers and rebooted it wouldn't reboot. So I shut it down and turned on again(this is after I deleted all original .979 driver files from my desktop). Upon turn on I still got asked if I want to run julipan with an open file warning again, I said no this time and proceeded with disc install. I got the control panel installed and came to the audio portion of the install and 'install wizard' asks me to choose the path (which shows up as the right path in the box )but when I click ok it just sits there. It says it cant find the driver on my desktop but its clearly there, (did you unzip the driver after you put on your memory stick?)
I need help.
Edits: 07/16/08 07/16/08 07/16/08
- I hope I have not led to your problems - rickmcinnis@dogwoodfabrics.com 08:10:55 07/16/08 (12)
In Reply to: RE: Cplay played with very high distortion-I have a problem posted by theob on July 16, 2008 at 04:19:44
I did not remember that unzupping was needed and did not bring this up.
I gave an outline of instruction based on you knowing the basic stuff.
When a file needs to be unzipped it usually lets you know. If you got .979 installed your machine must have unzipped them, I would have to presume (DANGEROUS!)
Since your machine is so different than the one (and only one) I am familiar with there must be something that is making a difference.
After cics gave you his list of instructions, which were more thoughtful than mine, I do not know what to tell you to do.
I guess if I was facing what you are I would do a fresh install of WINDOWS, which is my sure cure for everything, but my machine is a dedicated machine and this is not a big deal, especially with WINDOWS and program files on a small segment of the HDD.
Sorry you are having problems.
Rick McInnis
- RE: I hope I have not led to your problems - theob 08:44:35 07/16/08 (9)
In Reply to: RE: I hope I have not led to your problems posted by rickmcinnis@dogwoodfabrics.com on July 16, 2008 at 08:10:55
not your fault rick, you've been very helpfull. its just very frustrating for me when I don't know some of the basics and I just try stuff until it works. this one is my fault
so are you saying you didn't unzip the juli drivers until it asked you to?
Edits: 07/16/08
- I don't hink you could use the drivers until they are unzipped - rickmcinnis@dogwoodfabrics.com 10:20:52 07/16/08 (8)
In Reply to: RE: I hope I have not led to your problems posted by theob on July 16, 2008 at 08:44:35
the fact that you were able to install the .979 drivers must mean they were unzipped.
I have an idea, though.
Delete everything JULI@ and then when you re-install do not use the CD at all. Both times let both sets of drivers install with the computer finding the drivers. They are still in there, I bet.
I had a problem once with the thing continuing to ask for some component. Then, for some readon, I just let the computer do the installation by itself (not pointing to anything) and it installed the thing.
Do not have the CD in the drawer.
Let me know what happens,
Rick McInnis
PS this is how we learn to do this stuff!
- RE: I don't hink you could use the drivers until they are unzipped - theob 10:49:57 07/16/08 (7)
In Reply to: RE: I don't hink you could use the drivers until they are unzipped posted by rickmcinnis@dogwoodfabrics.com on July 16, 2008 at 10:20:52
yes I tried this I went through all of cics steps (unzipped by right clicking and extracting) and got to step 13 where the installation wizard asks you to choose a specific place to find the drivers (and I specified as per cics recommendation) but the wizard would not let me choose that c drive location. So I need some other way around getting the control panel and audio drivers installed. very frustrating, windows apparently remembers that it had a problem last time or something. I'm listening to 24/48 through my usb into the Benchmark and I hate it I want juli back.
oh well
- RE: I don't hink you could use the drivers until they are unzipped - rickmcinnis@dogwoodfabrics.com 14:27:19 07/16/08 (6)
In Reply to: RE: I don't hink you could use the drivers until they are unzipped posted by theob on July 16, 2008 at 10:49:57
When you get to the point where you say it won't let you go further are you "clicking" items in the "box"?
I wish I knew computer jargon better, but I don't. But, when you click on BROWSE and then select WHERE YOU HAVE THE DRIVERS, you have to click on an item in the box below, I think you go through two "levels" before you find what the computer wants.
Of course, the fact that you got it once is very mystifying and would have to mean you DID the above.
But, I thought I would say the above in hope it might help.
I think you need to go ahead and get the parts for a dedicated box and let this machine do what its main jobs have been. I think you are missing much of what this sistem can do. Use this as an excuse!!!
Bye,
Rick McInnis
- RE: I don't hink you could use the drivers until they are unzipped - theob 17:31:30 07/16/08 (5)
In Reply to: RE: I don't hink you could use the drivers until they are unzipped posted by rickmcinnis@dogwoodfabrics.com on July 16, 2008 at 14:27:19
I got through the install wizard but I'm stuck with being asked to run julapan.exe and losing the driver after a power stand by initiation. We'll see if cplay gets distorted tomorrow am like it today. then I'll try another 18 step re-install.
- RE: I don't hink you could use the drivers until they are unzipped - theob 03:37:50 07/17/08 (4)
In Reply to: RE: I don't hink you could use the drivers until they are unzipped posted by theob on July 16, 2008 at 17:31:30
cplay doing well this am
Edits: 07/17/08
- Cplay sounding great this am :) nt - theob 06:42:29 07/17/08 (3)
In Reply to: RE: I don't hink you could use the drivers until they are unzipped posted by theob on July 17, 2008 at 03:37:50
.
- RE: Cplay sounding great this am :) nt - theob 07:30:49 07/17/08 (2)
In Reply to: RE: Cplay sounding great this am :) nt posted by theob on July 17, 2008 at 06:42:29
Dear cics could my problem with juli (losing the pan.exe after power stand-by initiation) be due to the the fact that the hard drive is never powered off because I'm using the granite digital outboard ps? All the times I rebooted the hd was never off.
Edits: 07/17/08
- Pinging rickminnis - theob 17:06:31 07/17/08 (1)
In Reply to: RE: Cplay sounding great this am :) nt posted by theob on July 17, 2008 at 07:30:49
Rick I replaced the rca connector on the breakout cable with a female bnc. How sensitive is the solder joint to the sq? I'm a little disappointed. There seems to be a dirty overlay on the sound. Appreciate your assistance on this. It made this sound before I modded it but I thought that it sounded this way because it was not bnc. I may have a bad break out cable.As before when I go in power standby I can't run cplay until I reboot.
Edits: 07/17/08 07/18/08
- Don't bother rick it wasn't the cable - theob 05:23:57 07/18/08 (0)
In Reply to: RE: Pinging rickminnis posted by theob on July 17, 2008 at 17:06:31
I went through the 18 step re-install authored by cics and the warni ng about running julapan.exe keeps coming up at step 8 but I said no this time, the rest of the install went ok. I'm listening now, sounds good no subtle crackles. But I wondering why even after I unistall drivers (during the 18 step process) that I keep getting this request. Maybe I have to reinstall every other day.As before when I go in power standby I can't run cplay until I reboot.
Edits: 07/18/08
- RE: I hope I have not led to your problems - Ryelands 08:43:57 07/16/08 (1)
In Reply to: RE: I hope I have not led to your problems posted by rickmcinnis@dogwoodfabrics.com on July 16, 2008 at 08:10:55
If I asked really, really nicely, could the next person commenting on or asking about cMP and/or cPlay possibly be so daring as to start a new thread? For folk who follow the discussion, it's, well, getting more like Tristram Shandy every day.
Best
Dave
- at least this is just two volumes - rickmcinnis@dogwoodfabrics.com 10:30:10 07/16/08 (0)
In Reply to: RE: I hope I have not led to your problems posted by Ryelands on July 16, 2008 at 08:43:57
It is getting long but this way everything is one place. If it got spread around it would be more difficult to keep up.
One of us should volunteer to condense all of the important stuff into a document that cics could give him imprimatur. He is probably working on something like this already. Note, the revision of his first cPLAY thread which contains much information not in the original post.
Maybe we all should delete posts that are not doing much good for the general public?
cics, do you think we should do this with our extraneous posts?
Rick McInnis
- RE: Cplay played with very high distortion-I have a problem - theob 06:39:41 07/16/08 (17)
In Reply to: RE: Cplay played with very high distortion-I have a problem posted by theob on July 16, 2008 at 04:19:44
Guys I need help. Tried reinstalling juli using just disc (with 1.21 drivers). This process sent me into a system crash and I neede to restore my system to a level from several days ago. I'm gun shy about trying to reinstall juli again but I think I need some help.
- RE: Cplay played with very high distortion-I have a problem - cics 06:54:34 07/16/08 (16)
In Reply to: RE: Cplay played with very high distortion-I have a problem posted by theob on July 16, 2008 at 06:39:41
There must only be one instance of JulaPan.exe!
Do the following:
- Download 0.979 Juli@ driver
- Unzip to folder c:\juli@
- In Device Manager, Uninstall Juli@ Audio device
- In Device Manager, Uninstall Juli@ Controller device
- If you have more Juli@ devices, uninstall them
- Shutdown (as in power down)
- Remove Juli@ card
- Start computer and verify no Juli@ stuff happens
- Shutdown
- Install Juli@ card
- Start system, in a few moments, you should get a prompt on new hw found
- Select Advanced
- Select Specific Location and let the installer know where to find the driver sw, i.e. c:\juli@
- Juli@ Controller gets installed first
- Installer prompts again for driver. Do same steps.
- Juli@ Audio device is installed
- Reboot computer
- Juli@ is now installed and ready for use. Whenever explorer is killed, Juli@'s config tool (JulaPan.exe) disappears from task bar. This is normal (its a Windows bug). You will still see JulaPan.exe in task manager.
- RE: Cplay played with very high distortion-I have a problem - theob 10:02:38 07/16/08 (14)
In Reply to: RE: Cplay played with very high distortion-I have a problem posted by cics on July 16, 2008 at 06:54:34
(cics pls answer ONLY when convenient)anbody else can answer at any time
I got all the way through step 12 (Select Advanced on the install wizard)
when I go advanced and browse I do indeed find c:\juli@ in my c drive but I am not allowed to select it. The ok button gets disengaged or faded.
Is there another way around this?
Edits: 07/16/08
- Step 13: You must select the folder within c:\juli@ which would activate OK button. - cics 11:26:47 07/16/08 (13)
In Reply to: RE: Cplay played with very high distortion-I have a problem posted by theob on July 16, 2008 at 10:02:38
.
- RE: Step 13: You must select the folder within c:\juli@ which would activate OK button. - theob 09:11:49 07/18/08 (10)
In Reply to: RE: Step 13: You must select the folder within c:\juli@ which would activate OK button. posted by cics on July 16, 2008 at 11:26:47
Dear Cics: it seems I am still not w/o corrupted drivers. Let me ask you a question about the 18 step process you outlined. When I download 'the compressed drivers' from the esi site I save the compressed files right from the start in c:\juli@, then I unzip by right clicking on the folder and extracting. Is this correct? Also do you think because my hard drive is always 'on' with the gd outboard ps that this would make a difference?I'm getting good at all the 18 steps except I get no success apparently.
More data on this issue. After a bit of idle time my system drops julapan.exe. Its not on the tool bar nor is it in task manager. When this happens I get noise. But if I got to the c drive and find julapan.exe and run it I'm ok. Does that suggest anything? It's almost as if the system remembers the crash and tries to get rid of this driver whenever it can. Make any sense?
Edits: 07/18/08
- The size of julapan.exe keeps getting smaller in task manager over time - theob 08:23:02 07/19/08 (9)
In Reply to: RE: Step 13: You must select the folder within c:\juli@ which would activate OK button. posted by theob on July 18, 2008 at 09:11:49
It starts out at a little over 4200K then over time it gets smaller to 1800K then to around 600K. This could take 1 to sometimes 8 hours.
Rick or cics is this normal? When it gets smaller I seem to have more noise (I have to run at 64 latency to minimize it).When julapan gets below 4200k I end it in task manager go to c:\juli and open the file again to run it and it then comes back to original size at 4200K. I can't believe this is normal.I did reinstall drivers exactly per the 18 step process for the 2nd time.
Appreciate any ideas.
I'm running 176Khz and -145 src. Going all the way down to 44Khz eliminates noise (like small annoying tics) but the reason I got juli was to run 176 or 192. I can run 192 with -121 src but there is some ticking noise).
Edits: 07/19/08
- Redo steps but use Juli@ version 0.978 dirvers. Version 0.979 is known not to work on some XP versions. - cics 23:09:44 07/19/08 (8)
In Reply to: RE: The size of julapan.exe keeps getting smaller in task manager over time posted by theob on July 19, 2008 at 08:23:02
.
- Could this problem have been so simple that until now it ... - theob 05:36:41 07/20/08 (7)
In Reply to: RE: Redo steps but use Juli@ version 0.978 dirvers. Version 0.979 is known not to work on some XP versions. posted by cics on July 19, 2008 at 23:09:44
...it has eluded me. I de configured jui as a recoder and midi playback system (in my control panel) and the crackles are virtually gone. I read some early cics posts and one intrigued me. Something that stated juli's mixer was terrrible, so I figured why not try it.
Well I listened for 2 hours yesterday afternoon at -145 src, 24/176, 48 latency and my system was perfect. Came in late last night tried it for 20 minutes, perfect. Just fired up the system this am, listening at low levels for 15 minutes and again crackles are none to be heard.
Let me tell you about yesterday's listening session. There are moments in my life as an audiophile whereby things all seem to gell into what I call 'a religous experience'. Especially dificult classical music, there is the telarc sampler, several mercuries (Howard Hanson, Percy Grainger, George Bizet) and one file of Allison Krause (something Country in the title). Allison Krause and Union Station were in the room. Frederick Fennel only sounded this good through vnyl before. The sound is utterly the best I have ever heard in my system. Yesterday was the grandest of religious experiences.
I did not want to post until I really confirmed through Monday but since you so graciouly posted yet another suggestion I decided to tell you once again how uterly beatiful this play back system is(I know I'm not done yet) and to thank you for all your work into this.
At my settings in cplay the imaging, sound staging is stunning, the bass and dynamics are awesome. I swore to myself if I ever get through this week I would attempt no more moves for awhile. I think I'll just listen for a bit before I venture out again. Rick has said it, a friend of mine has said it....when you go through problems like this you really go through a fast learning cycle. What a ride and learning experience this week has been!!! Whew!!!
- .978 juli drivers helped me - theob 11:34:37 07/21/08 (6)
In Reply to: RE: Could this problem have been so simple that until now it ... posted by theob on July 20, 2008 at 05:36:41
Dear cics followed your advice and d/l .978 juli drivers. More stable plus I no longer get the warning/question about running julapan.exe. I believe my system is now stable and while I have not entirely eliminated the tics they can be controlled by not using the pc for anything but playback. I even pull my usb keyboard input from the pc and only have my mouse connected. When I get 'what I call a quiet a/c line' the performance is outstanding. But a question: what is the cause for the tics other than using the pc for other functions? It seems that latency setting does not effect the tics. I can minimize by running no src and 44Khz but thats no fun. I know the ultimate solution is cmp but these tics sound like IRQ phenomena and with a pci connection and no sharing of IRQ I thought juli was immune to this. Is that correct?
These questions are in the form of system improvement rather than problem solving. I can now get stable/highly enjoyable listening sessions if I don't use the pc. I'm running -145 db src at 176khz with 48 latency setting on juli configuration. Cool $ Quiet gives me 1.8 Ghz processor speed (better than 2.3 ghz at -145 db at 96khz)
- RE: .978 juli drivers helped me - cics 07:09:21 07/23/08 (5)
In Reply to: RE: .978 juli drivers helped me posted by theob on July 21, 2008 at 11:34:37
I can now get stable/highly enjoyable listening sessions if I don't use the pc. I'm running -145 db src at 176khz with 48 latency setting on juli configuration. Cool $ Quiet gives me 1.8 Ghz processor speed (better than 2.3 ghz at -145 db at 96khz)
That's the trick - you want the finest upsampling quality with just the right amount of processing power. E7200 is a better choice together with cMP.
When using PC (internet, mail etc) and playing music with such an aggressive setup (48 samples latency, SRC@145db SNR & 176/192 output) you will get dropouts (as critical PC resources are now being shared across more activities).
- RE: .978 juli drivers helped me - theob 07:53:21 07/23/08 (4)
In Reply to: RE: .978 juli drivers helped me posted by cics on July 23, 2008 at 07:09:21
You are right as long as I do nothing other than play cplay I'm ok.But, sigh, bad news today after 2 days of beautiful music the sound playback sounded overlayed with a bell (very obnoxious distortion). I rebooted and it was ok. Is Julapan.exe supposed to vary in size with time? Starts out at 4000K + and diminishes to 600K in task manager. I guess its ok if I have to reboot every two days but something still is not right. Also I don't get the system asking me if I want to run julapan.exe(which is good news I guess). But when I d/l from esi website the .978 drivers I ended up getting .979 but I figured since it works ,ok, and it did for 2 days. Should I go for .974 drivers per your 18 step re-install?
Edits: 07/23/08
- RE: .978 juli drivers helped me - theob 10:49:09 07/23/08 (3)
In Reply to: RE: .978 juli drivers helped me posted by theob on July 23, 2008 at 07:53:21
I d/l drivers again and this time got .978 (last time I think I did not delete all .979 juli files from my sys)but system asked several questions that may or may not be related to my issue:
1) during d/l of juli driver my system asked me if I wanted to d/l an update of java (I said no) then proceeded with the juli driver d/l
2) after reboot system asked me if I wanted to run julapan.exe (I said yes)this was like before. I'll accept this if .978 works
- The real .978 juli drivers do help me / how is celeron 667 cpu - theob 09:43:22 07/26/08 (2)
In Reply to: RE: .978 juli drivers helped me posted by theob on July 23, 2008 at 10:49:09
But other users of juli should be aware that juli will not survive a standby power down, you have to reboot. At least that is what the esi user community forum says. So I am not worried about that aspect of juli. But the .978 drivers do seem more stable on my system.
Question to cics: I got an old computer from my daughter that has xp and an intel celeron 667 mhz cpu, socket 370 fc pga, 256 ram. can I use to build up to cmp?
- Its best to use a dual core processor - cMP² exploits both CPU cores. - cics 11:18:08 07/30/08 (1)
In Reply to: RE: The real .978 juli drivers do help me / how is celeron 667 cpu posted by theob on July 26, 2008 at 09:43:22
Quality of mobo is also very important and I suggest going with recommended components.
You can use this gear as a practice run / benchmark.
- RE: Its best to use a dual core processor - cMP² exploits both CPU cores. - theob 11:50:52 07/30/08 (0)
In Reply to: RE: Its best to use a dual core processor - cMP² exploits both CPU cores. posted by cics on July 30, 2008 at 11:18:08
thanks, yes it might be good to practice
- RE: Step 13: You must select the folder within c:\juli@ which would activate OK button. - theob 12:07:46 07/16/08 (1)
In Reply to: RE: Step 13: You must select the folder within c:\juli@ which would activate OK button. posted by cics on July 16, 2008 at 11:26:47
good news and bad news. I can now get through the installation. but after reboot system asks me if I want to run julipan.exe again (identified as a warning again) but I say yes. cplay plays wonderfully. After forcing system into standby I lose something that prevents cplay from playing. when I reboot its back. This is where I was yesterday and after 12 hours of no 'stand by' idle it played with distortion.
we'll see after some idle non 'system stand by' time.
- RE: Step 13: You must select the folder within c:\juli@ which would activate OK button. - theob 12:25:54 07/16/08 (0)
In Reply to: RE: Step 13: You must select the folder within c:\juli@ which would activate OK button. posted by theob on July 16, 2008 at 12:07:46
but there is no dual julipan running in task manager, so thats additional good news.
- RE: Cplay played with very high distortion-I have a problem - theob 07:25:41 07/16/08 (0)
In Reply to: RE: Cplay played with very high distortion-I have a problem posted by cics on July 16, 2008 at 06:54:34
thanks I'll try again.
- RE: It is best to start over - theob 09:48:02 07/15/08 (1)
In Reply to: RE: It is best to start over posted by rickmcinnis@dogwoodfabrics.com on July 15, 2008 at 07:58:35
Rick I am mapped in cplay to juli 3,4 and it works great. In fact I am overwhelmed at how good it sounds. Its just that everytime I go into standby I lose my juli panel. Then I have to reboot. But resetting power options on my control panel prevents it from ever going into stand bye which is ok by me because its like leaving my dbx crossover on all the time---it sounds better. I will try reinstalling after I listen for a while.
- RE: It is best to start over - theob 14:09:14 07/15/08 (0)
In Reply to: RE: It is best to start over posted by theob on July 15, 2008 at 09:48:02
I tried going to the control panel and updating drivers for both juli audio and juli panel and the result was no better drivers were found(and yes I put the disc back in for updating the juli panel drivers). I already have 0.979 so no problem there.
I'm perplexed as to why going to a system 'stand by' loses something such that I need to reboot to get cplay to work. Only thing I can come up with that is different during bootup is that all the processes I end (to free up ram) are back in the task manager. So I double checked each one and the only ones that may be suspect are sprtsvc and dlatrlw which I cant get a good handle on in process.library.com Anybody know anything about these processes?
- Cics:I'm getting a Juli card -- you said to remove analogue board - theob 09:34:50 07/11/08 (14)
In Reply to: RE: cPlay - the open source high-end audio player using ASIO posted by cics on May 05, 2008 at 12:31:58
Can I do this right from the start before installing and loading juli drivers? I will also only use spdif out at 192 to my Benchmark. Was this the juli mod you said was required to run 24/192 out or is there something else too?
Also can you confirm thet the analogue board is the one that detaches and can be turned around for either rca out or balanced out?
- My installation - rickmcinnis@dogwoodfabrics.com 13:14:20 07/11/08 (13)
In Reply to: RE: Cics:I'm getting a Juli card -- you said to remove analogue board posted by theob on July 11, 2008 at 09:34:50
You should try installing it w/o the analog part which is the removable part, otherwise you would have nothing to plug in.
I replaced the RCA for a BNC connecctor to be sure the connector was not a filter. I never tried it with RCA's since my cables were BNC, which for this application is the superior connector.
Be sure to download the new .979 drivers and have them within a USB memory or on your DESKTOP.
I let the supplied CD install the control panel, which is always the first thing installed with this device. Let machine do this automatically as recommended by WINDOWS.
After this is installed, remove the CD. I have found that if you install the second set of drivers on the CD, it becomes slow and strange to remove them in favor of the new (.979) driver.
Usually you will be prompted for the second set of installation (the language is awkward but I cannot think fo a better way to put it), instead of doing the RECOMMENDED thing (letting WINDOWS find the file to install) select the option of pointing the computer to where the driver is. Using the BROWSE button, keep clicking till you get what the machine wants (as you can tell I do not know proper computer terminology but I hope you get what I am saying).
For me, with the driver on the USB stick, you first click on the USB drive, you click JULI@ (or whatever it is you named this when you downloaded it) I am going from memory now and finding my memory is not photographic, SO once you have selected THE FILE AS YOU NAMED IT just keep clicking the choices given under what you first selected until the machine allows you to go further. Then click the OPEN button and it installs the thing and then you are done.
The control panel will come up and in the right corner it will verify that you have the .979 driver version. Under CONFIGRATION, select LATENCY and select 48. You are done with this part.
Now, go into cPLAY SETTINGS, I have found on my application that I must select channels 3,4 to get sound instead of the default 1,2. Make any other pertinent choices here and then close. When you open cMP you should begin to experience musical pleasure that will continue to improve as everything settles in.
I hope this made sense. I was trying to save cics the trouble. Of course, it was cics who led me through the installation.
Bye,
Rick McInnis
- RE: My installation - theob 15:17:29 07/11/08 (12)
In Reply to: RE: My installation posted by rickmcinnis@dogwoodfabrics.com on July 11, 2008 at 13:14:20
Let me ask some clarifying f/u questions.
1) Install only the digital hardware into my physical pc (i.e., without the removable analogue board). Correct?
2) Convert the rca connector on the juli board with a bnc adapter. Then use a bnc cable from the juli board to my dac. Correct?
3)Download the new .979 drivers (from where?) and put them somewhere accesible (on desktop). Correct?
4) Load the juli cd and only install the control panel (first item) but continue the installation by pointing to the new drivers via browse. Correct?
5) Then I click something to open (is this part of the install process?) and the control panel opens. Correct?
6)Verify .979 then select latency.
7)This one has me confused 'go into cplay or juli to select channels 3,4'?
8) Begin the breakin process.
Does this sound right and could you please clarify 3), 5), and7)?
Appreciate your help.
- RE: My installation - rickmcinnis@dogwoodfabrics.com 08:44:51 07/14/08 (11)
In Reply to: RE: My installation posted by theob on July 11, 2008 at 15:17:29
Let me ask some clarifying f/u questions.
1) Install only the digital hardware into my physical pc (i.e., without the removable analogue board). Correct?
There is no need for the analogue board to be attached to install JULI@.
2) Convert the rca connector on the juli board with a bnc adapter. Then use a bnc cable from the juli board to my dac. Correct?
I assume you are handy with a soldering iron and wire strippers. Get a female BNC with SOLDERABLE ends, not crimp. Look at DIGIKEY. I am using an AMPHENOL product. Get 75 ohms impedance. Correspondingly, use a digital cable with BNC's. I like the relatively inexpensive STEREOVOX which comes with BNCs and RCA adaptors, if you insist on using them. But, for these frequencies BNCs are what works best. I hope you have a BNC input jack on your DAC. Of course, cics prefers TOSLINK. I have no way of trying this out, so it is BNCs for me.
3)Download the new .979 drivers (from where?) and put them somewhere accesible (on desktop). Correct?
http://download.esi-audio.com/?w=esi&p=6&g=1&l=en
choose the first item on the list.
DESKTOP or USB memory stick, whichever you prefer.
4) Load the juli cd and only install the control panel (first item) but continue the installation by pointing to the new drivers via browse. Correct?
As written. Most likely the computer will recognize there is something new at turn on. Have the CD in place. Do not "search the internet for latest drivers" - let it do the automatic install, as recommended, for the "control panel".
5) Then I click something to open (is this part of the install process?) and the control panel opens. Correct?
The control panel will not appear until the whole porcess has been done.
After installing the drivers for the control panel, you should be prompted to install the second part of JULI@. Remove the CD before selecting anything. I am assuming you will have the .979 drivers on a USB stick or on your DESKTOP. Then start by not having the computer look for latest drivers on the internet. THEN, when it prompts for where to look, select YOU picking where to look, and then using the BROWSE button select where you have placed the drivers. When you get to the right place, you will then be allowed to proceed. This will make sense when you actually do this. I am going from memory, but this will make sense when you get to this point.
6)Verify .979 then select latency.
from CONFIGURATION, choose 48. Then save, just to be sure. Name the file: theob!
7)This one has me confused 'go into cplay or juli to select channels 3,4'?
cPLAY, either outside of cMP or within. I do it before going into cMP. But, I am sure one can do it either way. Of course, within cMP you must select a music file, which then will engage cPLAY. From there, either way, you hit the cPLAY button on the bottom right hand corner to make this selection.
8) Begin the breakin process.
Yes, just leave it playing music at any volume you choose. That way when you get a chance to listen seriously it will be sounding better. I think it improves steadily for the first week and then it is as good as it will be.
- RE: My installation - theob 09:23:45 07/14/08 (10)
In Reply to: RE: My installation posted by rickmcinnis@dogwoodfabrics.com on July 14, 2008 at 08:44:51
thank you, very precise and clear.
- RE: My installation - theob 12:40:23 07/14/08 (9)
In Reply to: RE: My installation posted by theob on July 14, 2008 at 09:23:45
Got everything installed including hardware, panel and drivers; reset my audio windows panel, reset cplay but somehow I got the 1.21 version of the juli driver installed. Question how do I replace this with the .979 driver? I do have the .979 driver on my desktop but I guess I blew it during installation. I did not modify my break out cable so I can only get 176 out to the Benchmark. But right out of the shoot it sounds good. But I do want the other driver. Appreciate any help.
Edits: 07/14/08 07/14/08
- RE: My installation - rickmcinnis@dogwoodfabrics.com 13:15:06 07/14/08 (8)
In Reply to: RE: My installation posted by theob on July 14, 2008 at 12:40:23
You must have installed both sets of drivers from the CD. That is the only way you could have done this.
Go to SOUNDS & AUDIO DEVICES within CONTROL PANEL - then to HARDWARE, then to "the other" JULI@ thing (NOT control panel) I am not at home so I caanot look at the exact thing. Go to the PROPERTIES for that device and then to DRIVERS, then to UPDATE DRIVERS. It should ask you the same questions as during the intial install. This time make sure you do not have the CD in the drawer. No way you can screw up if it is not there.
An alternative would be to delete those drivers (from the "place" mentioned above) and then go to ADD HARDWARE in CONTROL PANEL and go through its process. It will attempt to find the hardware, but most likely will not, so you answer a few questions, pick MULTIMEDIA SOND CONTROLLER when you get asked that, that is if it does not find JULI@, and go from there. Be sure to carefully read what it asks you before picking a choice and you should get it installed eventually.
I prefer the second method. It works better for me.
There is always the choice of deleting both sets of drivers in SOUNDS & AUDIO DEVICES/HARDWARE/DRIVERS and starting over completely.
It took me a few attempts before I got it right, too.
Hope this is of use.
Rick McInnis
- RE: My installation - theob 13:54:10 07/14/08 (7)
In Reply to: RE: My installation posted by rickmcinnis@dogwoodfabrics.com on July 14, 2008 at 13:15:06
thanks
- Bummer I had a crash, lost juli control panel and juli driver 1.2 - theob 14:22:55 07/14/08 (6)
In Reply to: RE: My installation posted by theob on July 14, 2008 at 13:54:10
So now I've got the hardware installed but nothing works. How do I reinstall w/o taking the hardware out?
- RE: Bummer I had a crash, lost juli control panel and juli driver 1.2 - theob 14:51:22 07/14/08 (5)
In Reply to: RE: Bummer I had a crash, lost juli control panel and juli driver 1.2 posted by theob on July 14, 2008 at 14:22:55
Got it back it up. But I don't see the control panel for juli. Also when I gto the computer control panel I don't see juli at all, but its working fine @ 176 sampling rate, -145 db src. And it sounds great. But now I'm confused why I don't see juli opn the task bar or in my control panel.
- RE: Bummer I had a crash, lost juli control panel and juli driver 1.2 - rickmcinnis@dogwoodfabrics.com 14:53:53 07/14/08 (4)
In Reply to: RE: Bummer I had a crash, lost juli control panel and juli driver 1.2 posted by theob on July 14, 2008 at 14:51:22
Theo,
It will do that.
It should show up on a re-start.
Let me know,
Rick McInnis
- RE: Bummer I had a crash, lost juli control panel and juli driver 1.2 - theob 15:03:12 07/14/08 (3)
In Reply to: RE: Bummer I had a crash, lost juli control panel and juli driver 1.2 posted by rickmcinnis@dogwoodfabrics.com on July 14, 2008 at 14:53:53
yes it came back up plus I got .979 working. Whew!!!! Thanks for all your help.
- RE: Bummer I had a crash, lost juli control panel and juli driver 1.2 - fr338 19:45:20 07/14/08 (2)
In Reply to: RE: Bummer I had a crash, lost juli control panel and juli driver 1.2 posted by theob on July 14, 2008 at 15:03:12
theob,
So is it better than USB connection ?
- RE: Bummer I had a crash, lost juli control panel and juli driver 1.2 - theob 03:55:40 07/15/08 (1)
In Reply to: RE: Bummer I had a crash, lost juli control panel and juli driver 1.2 posted by fr338 on July 14, 2008 at 19:45:20
loaded question because with juli I run at 176 khz and it sounds magnificent better than I ever imagined, Added bonus is I get a non-shared irq (no pops and crackles). with usb I ran cplay at 24/96 but cics tells me that asio4all was d/s the system from 96 to 48. So I really don't have an apples to apples comparison for you.
Bottomline, in my system, way better--richer, more air, more speed.
- RE: Bummer I had a crash, lost juli control panel and juli driver 1.2 - fr338 19:01:33 07/15/08 (0)
In Reply to: RE: Bummer I had a crash, lost juli control panel and juli driver 1.2 posted by theob on July 15, 2008 at 03:55:40
That's awesome.
- USB ASIO questions - fr338 07:37:32 07/09/08 (11)
In Reply to: RE: cPlay - the open source high-end audio player using ASIO posted by cics on May 05, 2008 at 12:31:58
I am still struggling to see how to use my USB device with SRC.( this may seem kind of silly to others but I always think laptop is the quietest PC) Cics once mentioned ASIO4ALL is 16bits & 44.1KHz only. I believe so since this is shown on the setting in ASIO4ALL and I cannot alter it. But can I use USB ASIO driver instead ? I installed USB ASIO 2.6.1 but Cics player rejects to operate saying it need ASIO 32 bits. I asked USB ASIO help desk and Markus response to me USB ASIO is 32 bits.
When using foobar with the same drivers above I am able to use SRC. In SRC mode foobar has something missing in the 16/44.1 cics. Cics is very clean, but foobar seems to have more relax presentation using 24/96 src.
- RE: USB ASIO questions - theob 07:55:56 07/09/08 (10)
In Reply to: RE: USB ASIO questions posted by fr338 on July 09, 2008 at 07:37:32
I use asio4all with cplay & my benchmark and set the asio4all panel offline to 24/96 with -145 db src and it works.
i tried d/l ing the asio usb driver and I continue to get a 'fail to install' message and I tried all 3 xp drivers shown on the web site.
So go into your offline asio4all panel (should be on desktop) and reset. Stop your cplay and restart. If you choose to see diagnostics you will see if its set at 24/96 or 16/44.
What operating system do you have to get asio usb driver to work? I have xp sp2 and could not get any of the 3 xp drivers to install.
- RE: USB ASIO questions - fr338 08:22:46 07/09/08 (9)
In Reply to: RE: USB ASIO questions posted by theob on July 09, 2008 at 07:55:56
I am unable to set ASIO4ALL. The version number is 2.8. Say, currently since maybe I have installed USB ASIO , it says "AudioDevice on USB Bus 2.6.1 -next indented line - In: 2x44.1kHz, 16bits - next line - Out: 2x44.1kHz,16 bits
Do you have a different UI than this one that can change setting ?
I am using winxp pro tablet version. I suppose the same as a normal pro except adding touch panel driver...
- RE: USB ASIO questions - theob 08:30:40 07/09/08 (8)
In Reply to: RE: USB ASIO questions posted by fr338 on July 09, 2008 at 08:22:46
only thing I can surmise about your issue is that your pc won't take greater than 16/44 on asio-usb or that you are not using the right asio panel . Unistall asio4all and try again you may get the asio-usb panel. In fact I may try that to install 2.8 but markus said it should not be an issue either way.
- RE: USB ASIO questions - fr338 08:36:55 07/09/08 (7)
In Reply to: RE: USB ASIO questions posted by theob on July 09, 2008 at 08:30:40
I use a USB-SPDIF converter that has a spec of 16/44.1(or 48k) only. Would that be the reason ?
But again why I use the same ASIO4ALL with the same 16/44.1 setting can run with foobar SRC ?
- RE: USB ASIO questions - fr338 23:51:50 07/10/08 (6)
In Reply to: RE: USB ASIO questions posted by fr338 on July 09, 2008 at 08:36:55
theob,
Do you think my problem is because I am using winxp pro ? Are you using winxp home ?
- RE: USB ASIO questions - theob 04:17:52 07/11/08 (5)
In Reply to: RE: USB ASIO questions posted by fr338 on July 10, 2008 at 23:51:50
I honestly don't think that is the answer. I am using xp home editio n. I'm stll upset I can't get usb-asio. Markus never answered my last question regarding why none of the xp versions could be instaslled on my pc.
- RE: USB ASIO questions - fr338 09:02:32 07/11/08 (4)
In Reply to: RE: USB ASIO questions posted by theob on July 11, 2008 at 04:17:52
theob,
Do you use both USB and PCI audio card digital output to your benchmark ? Can you share whether USB excel PCI or the other way round ?
- RE: USB ASIO questions - theob 09:41:33 07/11/08 (3)
In Reply to: RE: USB ASIO questions posted by fr338 on July 11, 2008 at 09:02:32
I will be able to tell you next week after I get my juli card how juli at 192 sounds. Last week I tried a Prodigy 7.1 lt but could get only 24/96 out of spdif. But comparing usb versus prodigy at 24/96 through an Audio Alchemy powered i2s cable, the usb cable sounded better.
Edits: 07/11/08
- RE: USB ASIO questions - fr338 09:45:11 07/11/08 (2)
In Reply to: RE: USB ASIO questions posted by theob on July 11, 2008 at 09:41:33
I look forward to your comparison. I am struggling if I should build a desktop with Juli@ to get the full capability of SRC.
- RE: USB ASIO questions - theob 11:07:53 07/11/08 (1)
In Reply to: RE: USB ASIO questions posted by fr338 on July 11, 2008 at 09:45:11
what dac do you have?
- RE: USB ASIO questions - fr338 17:02:22 07/11/08 (0)
In Reply to: RE: USB ASIO questions posted by theob on July 11, 2008 at 11:07:53
I use a USB to spdif converter and then a NOS DAC from Promitheus.
- The End of History? - rickmcinnis@dogwoodfabrics.com 08:33:02 07/07/08 (18)
In Reply to: RE: cPlay - the open source high-end audio player using ASIO posted by cics on May 05, 2008 at 12:31:58
Dear cics,
To paraphrase the title of Fukuyama's book ((which, too, was a bit (no pun intended) premature)), now that I have the GIGABYTE/e7200 set up complete with all recommendations made, and allowing it to run for about twelve hours, I listened seriously last night for about two hours.
I wonder how much more information can be revealed? The sound seemed to flow effortlessly from the speakers. Words sung were much more intelligible than ever before, spaciousness did not sound contrived, no black or grey background, no "color" to the ground at all. Sounds came from where they were intended (I have to believe). It gave the impression that this was DONE. Which is kind of sad, part of the fun of this had been the increments of increasing quality. Can there really be any additonal significant steps?
I am using the JULI@ card at 192kHz, -145 db version of SRC with no glitches at all.
I figure at this point trying to deal with radiation from the power supply and from the processor and motherboard are worth pursuing. I have been looking at the ERS cloth from STILLPOINTS as something worth implementing.
Not that I am hearing RFI interference but we know it is there.
The e7200 stays cool, even with the GOLDEN ORB perched atop the chip weighted down by a small choke. The hardware does not clear the capacitors near the processot. I have a new THERMALRIGHT heat sink due today. Seems to be an updated version of the one recommended in the cMP MANUAL. I cannot imagine it ever heating up at all with this thing.
The best part of a significant upgrade such as this is that you get to really enjoy listening to all of your favorite recordings,again,since you are hearing new details, etc.
For those who are getting the GIGABYTE board, I found using their software that allows BIOS software changes within WINDOWS to be foolproof, since I was able to do it!
If you are using an USB CD-ROM you will probably find as I did that the MB does not recognize it when it matters for doing your WINDOWS installation. As per cics's recommendations: use an internal drive at the IDE connector. For the loading of WINDOWS this does a much better job. MUCH quicker and then you can remove it when you are done.
see:http://www.audioasylum.com/forums/pcaudio/messages/3/33792.html
So, yet again, my sincerest thanks to cics for his enlightened selfishness. While doing something to increase his pleasure while listening to music in the home, he has also shared his knowledge with all who will pay attention. I wish I could offer something half as great in return. At this point all I have to offer is gratitude.
Rick McInnis
- In your cMP² can you try single 1GB RAM stick option - cics 08:32:48 07/09/08 (12)
In Reply to: RE: The End of History? posted by rickmcinnis@dogwoodfabrics.com on July 07, 2008 at 08:33:02
I wonder how much more information can be revealed?
I'm testing a few options and my preference is for the single RAM module option. Although this halves memory performance, there's one less device in the equation...
Try it on your setup using previous 1GB RAM stick.
BTW - just hooked up Juli@ soundcard (24/96 SRC@145db SNR) and its excellent!
- RE: In your cMP² can you try single 1GB RAM stick option - sonics 09:31:31 07/10/08 (0)
In Reply to: RE: In your cMP² can you try single 1GB RAM stick option posted by cics on July 09, 2008 at 08:32:48
hi cics
i agree . i tried several ram solutions and 1GB was always better than dual ram.
Looking forward two offline upsampling because all online sessions with 145 SRC upsamling are less natural.
bye
alfred
- Of course I will & battery MB musings - rickmcinnis@dogwoodfabrics.com 08:23:10 07/10/08 (10)
In Reply to: RE: In your cMP² can you try single 1GB RAM stick option posted by cics on July 09, 2008 at 08:32:48
Dear cics,
THOUGH, as much trouble as I went through to get the 2 512mB sticks I would almost guess you are testing ME!!! My loyalty has been proved since I am listening to the 2 x 512 set-up. Easy as can be to change out the memory. and always fun to see what happens.
Not tonight, but over the weekend I will give it a serious listen.
I feel like saying PERFECT SOUND FOREVER (as long as your HDD is backed up).
It is, up to now, a pleasure to listen through this. Maybe in a month's time some flaws will become apparent, but they obviously will not be grotesque ones. It is a great system, cics.
What do you think of Alfred's interest in battery power? I looked around yesterday at what would be required. My first impression, after looking at INTEL's application notes, is that the turn on sequence is very complicated and important for the "life" of the processor. I guess one could use the regular power supply to power relays to turn on batteries as the sequence requires?
If one was to pursue this: what is the most important place for quiet power? The processor, the memory, the sound card? In what order would you place importance?
Obviously, the easiest rail to swap is the 12 volts rail. Is the processor powered by the four pin connector, only?
What do you think of doncheng's idea of storing the upsampled music within memory; letting the processor do very little during playback? I do not know how to calculate such things but I would guess the 24/192 upsampled "file" would exceed the 4 gB limit of our MB. So would this be possible with the current BEST set-up?
I would enjoy hearing your opinions if you have the time.
Congratulations, and extravagant thanks for the effort and its execution,
Rick McInnis
- Dual vs Single Channel; RAM loading upsampled WAV; Voltages - cics 20:47:49 07/10/08 (5)
In Reply to: RE: Of course I will & battery MB musings posted by rickmcinnis@dogwoodfabrics.com on July 10, 2008 at 08:23:10
There's advantage from using dual channel memory and could reduce jitter by a very small amount. Unfortunately (like most things) there's a tradeoff: another electrical device.
Arthur's suggestion: Same applies to RAM loading upsampled WAV data as significantly more RAM is needed. Of course, one approach in cPlay would be to RAM load track at a time with each track upsampled into RAM and not during playback. This could be accommodated with just 1GB RAM. This needs a playlist GUI tool that would marshal one track to the next... That would be very nice indeed! Any volunteers to do the coding?
Voltage rails: I would start at 12V1 which unfortunately carries 12V, 5V & 3.3V lines of which 5V & 12V are critical. PCI slots offer 3.3 or 5 volt supply - you can tell this by where the notch is: front (close to mobo edge) is 3.3V and rear is 5V (Gigabyte & Biostar mobos). Soundcards are designed to operate with either voltage.
Alfred is correct, 12V2 (4 pin 12V connector) is used to power CPU.
- BUT do you think this is worth pursuing? - rickmcinnis@dogwoodfabrics.com 08:52:33 07/11/08 (4)
In Reply to: RE: Dual vs Single Channel; RAM loading upsampled WAV; Voltages posted by cics on July 10, 2008 at 20:47:49
Dear cics,
In the great scheme of approaching "perfection in music reproduction using a PC" have you considered what would be the pecking order of importance?
In your opinion, what are the most important things to be concerned with and almost as importantly, which concern you the least?
If someone out here was to lend a hand to your method, where would it be most helpful? You have already specified one, with the loading of one upsampled song at a time into the memory. Where would you place that?
If for fun we wanted to create some "one song files" to experiment with this approach is there an easy way to implement this? I guess we should see what is gained before going too far with it!
In this tweaker's world where we have succeeded in making digital playback every bit as tweaky as analogue, why not? For the absolute best recordings and favorite songs even a convoluted approach could be worthwhile. I think of the many 12" 45's I have that sounded much better than the LP even though there was limited playing time. Who cares for a few minutes of something better?
I fear that if playback was truly perfected we would not listen any more. There is a challenge to make it better (or is it just- different) that drives us for better or for worse. Usually both at the same time.
C'est le guerre!
Thanks as always to cics,
Rick McInnis
- RE: BUT do you think this is worth pursuing? - riboge 09:27:06 07/11/08 (3)
In Reply to: RE: BUT do you think this is worth pursuing? posted by rickmcinnis@dogwoodfabrics.com on July 11, 2008 at 08:52:33
Yes, of course it is. Any method of ram-loading a sequence of files from a list(playlist)would greatly enhance the usefulness and applicability (e.g., hd downloaded music) of cPlay and cMP as well as making loading of processed files possible. The other high priority re GUI would be to have an option to resize the window and/or font. The latter would seem to me as a non-programmer to be a simple thing for anyone with the know how.
Could you explain the request for how to make single song files? Single flac files are already playable in cPlay and foobar can make a cue file for a single file.
- Files already upsampled into the memory - rickmcinnis@dogwoodfabrics.com 13:27:20 07/11/08 (2)
In Reply to: RE: BUT do you think this is worth pursuing? posted by riboge on July 11, 2008 at 09:27:06
As cics said in his reply, that if the ALREADY upsampled files (which would allow the processor to "idle" during playback) were loaded one song/movement at a time this could be done with the 1gB of memory.
Analogous to how cPLAY converts a FLAC file to WAV before sending it to the memory as opposed to converting it while the file is in use. This way a 24/192 file would be played back from memory while the porcessor loafs with little to do. Seems intriguing to me. I'd like to hear what a difference this would make.
Especially since most of cics's recommendation are based upon keeping the processor as unoccupied as possible, this is an obvious step; another evolution of the cics method of musical digital playback!
Bye,
Rick McInnis
- There's an elegant design for this but unfortunately it's not feasable - cics 01:32:19 07/14/08 (1)
In Reply to: RE: Files already upsampled into the memory posted by rickmcinnis@dogwoodfabrics.com on July 11, 2008 at 13:27:20
Problem is a 5 minute track requires some 30+ seconds of pre-processing!
- RE: There's an elegant design for this but unfortunately it's not feasable - sonics 02:43:57 07/14/08 (0)
In Reply to: RE: There's an elegant design for this but unfortunately it's not feasable posted by cics on July 14, 2008 at 01:32:19
hello cics,
i upsampled a 16/44 to 96/24 with wavelab( Steinberg). Time is about 10 sec, for a dual core at 2Ghz for a 5 Min song. Sound is not better on my system, but soundstage is not loosing quality with 96/24 offline.
bye
alfred
- RE: Of course I will & battery MB musings - sonics 09:28:07 07/10/08 (3)
In Reply to: RE: Of course I will & battery MB musings posted by rickmcinnis@dogwoodfabrics.com on July 10, 2008 at 08:23:10
Hallo,
I think battery power is a littlebit to much for all the rails.My Hifi setup with 12 Batteries is complicated enough.
I have very good results with switching PS and a big block of for example siemens sicorels with low ESR. Switching PSs with my amps are better than conventional PSs !?. So we can try perhaps to do some good kondensators parallel two the 12 V rail. The 4 Pin connector is an additional support for the processor by a power consuption over 60 Watts.So perhaps all 12V lines have to be buffered. Its a littlebit trial and error.
bye and thanks
alfred
- Alfred, where is a good place to start? - rickmcinnis@dogwoodfabrics.com 15:15:45 07/10/08 (2)
In Reply to: RE: Of course I will & battery MB musings posted by sonics on July 10, 2008 at 09:28:07
Alfred,
It seems you have made some modifications to the MB.
Please be specific where you find improvements to be useful. Let me know which rails have benefited from the added, and in ther case of SICORILS better capacitance. I would like to like to give it a try but I have no idea where or how to start.
I have a second GIGABYTE board I would like to implement your suggestions. I would be glad to mirror what you have done and offer my impressions if that is of interest to you.
Thanks,
Rick McInnis
- RE: Alfred, where is a good place to start? - sonics 01:52:02 07/12/08 (1)
In Reply to: RE: Alfred, where is a good place to start? posted by rickmcinnis@dogwoodfabrics.com on July 10, 2008 at 15:15:45
hello ric
You are absolutly right: obsession is feed by the last 5 % and nothing is more boring than perfection.
but!!! i did it. there are now 6000µF (6x 1000µF of panasonic FM or Fc low ESRs)
and some styroflex cons with 3000 to 4000nF parallel to the 12+Volt at the 4 Pin connector for the CPU. All 12+ connectors are from the same source (0 Ohm messured between all 12V cables) so you can do the condensators elsewhere and you can try bigger capacitances.
Sound is smoother( Female vocals at high frequencies).
So my steps in impact on sound in order of importance:
1.cics cmp and cplay!!!at 44100Mhz
2.juli esi (after trying E-mu 1212 and usb to spdif)
3.external power for sata harddisc and fans
4.condensatormods for 12V+
4.unplugging monitor and usb sticks during playback.
here is a link to PS cables:
http://www.elektronik-kompendium.de/sites/com/0601151.htm
i think it is worth trying it, .
bye
alfred
- Have you measured AC Ripple across all lines? - cics 02:52:24 07/26/08 (0)
In Reply to: RE: Alfred, where is a good place to start? posted by sonics on July 12, 2008 at 01:52:02
Take a look at this 430W PSU (Antec Earth Watts) manufactured by Seasonic. For all 3 voltage lines its the best ever ripple voltage I've seen to date.
There's also the Zalman ZM600 with slightly better 5V line. 12V line is not as good but this is at 150w load - way more then cMP² requirement.
- RE: The End of History? - sonics 09:21:28 07/07/08 (2)
In Reply to: RE: The End of History? posted by rickmcinnis@dogwoodfabrics.com on July 07, 2008 at 08:33:02
Hallo
i am german , so forgive for my poor english.
I read your post and i am very near to your listening experience.My listening time with very expensive stuff comes to a lucky end.
I worked much time with digital amps and there the PS with very low inner resistance is very important for SQ . So i do
power supply with many paralleled sealed batteries and big capacitors and for the richness of sound some little glimmer and styroflex capacitors also parallel.So the next step will be a PC-powersupply with this attempt in the power-lines.The big work will be the search for the lines ( 12,5, 3, Volts ) which are the most imortant,. If there are some result i will post it.
cheers
alfred
- I look (?) forward to hearing (!) what you come up with - rickmcinnis@dogwoodfabrics.com 10:58:42 07/07/08 (1)
In Reply to: RE: The End of History? posted by sonics on July 07, 2008 at 09:21:28
Dear Alfred,
All of us audio-obsessesed know whow important the power supply is to all audio components.
I, too, use batteries for the above 500 hz amps in my system. It was the only way I could achieve quiet. Even there, the ESR makes a big difference. If you are not using SLA batteries of LARGE size you are not hearing what batteries can do.
Maybe we do not have to go with all batteries for a computer supply? The twelve volts rail seems to do most of the work, and with the GRANITE outboard supplies for the USB, it might not be such a big deal.
I think the sound card uses the twelve volts rail, also. Is this correct?
I look forward to hearing what you come up with.
As we all say over here, I wish my German was as good as your English.
Your intentions came through clearly.
Bye,
Rick McInnis
- RE: I look (?) forward to hearing (!) what you come up with - sonics 13:52:15 07/07/08 (0)
In Reply to: RE: I look (?) forward to hearing (!) what you come up with posted by rickmcinnis@dogwoodfabrics.com on July 07, 2008 at 10:58:42
Dear Rick,
the world of hifi is sometimes pretty small.In the very first beginning in sla experience i tried also big batteries , they have smaler
esr than little ones, but 4, 8 or even 12 batteries paralleled are blowing up SQ because of devided ESRs.(I am using sla powerd amps above 80 Hz and i gave away high- highend valve amps for the battery powered amps )
Soundcard unfortunately is using 3,3 , 5, and 12 Volt in the PCI-rail but i agree, that it would be the the 12+- V line to give a first try.
bye
alfred
- RE: The End of History? - theob 08:51:02 07/07/08 (0)
In Reply to: RE: The End of History? posted by rickmcinnis@dogwoodfabrics.com on July 07, 2008 at 08:33:02
Congratulations I am envious of you. I have some of the cmp ideas worked into my dell. But I know I want to join you some day in going the whole nine yards.
Happy listening!!
- RE: The End of History? - artcheng@netvigator.com 08:38:04 07/07/08 (0)
In Reply to: RE: The End of History? posted by rickmcinnis@dogwoodfabrics.com on July 07, 2008 at 08:33:02
I suppose if the upsampling can be done while loading the wav to RAM, that will save a huge chuck of CPU resources and may make further improvement
- RE: cPlay - the open source high-end audio player using ASIO - sonics 02:36:59 07/07/08 (1)
In Reply to: RE: cPlay - the open source high-end audio player using ASIO posted by cics on May 05, 2008 at 12:31:58
Hallo cics,( Sorry posted before in the wrong chapter)
at first many thanks for your player. A big footprint in pc-audio history.
I am working for 3 years with PC-audioand i used foobar ,
different soundcards: -1212m e-mu etc.
diffeent USB to SPDif converter:
Trend, and hoerwege usb to spdif).
The PC only as a transporter with an external DAC, because all solutions for DA in the PC are lacking quality.
Best results till now:
AMD 3800 underclocked to 800Mhz. 1GB DDR1 with 200FSB.
SATA Drive with external PS.
Windows stripped down to 70MB in the RAM.
AND !!the ESI Juli latency 48 and upsampling to 88,2.
This system is blowing away every CD-Playeer i Know ( High End).
So next steps are a look at Intels Atom Processor on a car-motherboard with 12 V powered by sealed batteries ( 4x12 V to achieve little inner resistance.)
Are there any thoughts at the atom- processor -solution
And!! Lokking for cplay 20 21 22 and so on .
Cheers
alfred
- Intel's Atom Processor - cics 05:43:21 07/07/08 (0)
In Reply to: RE: cPlay - the open source high-end audio player using ASIO posted by sonics on July 07, 2008 at 02:36:59
Great feedback. Thx.
See this discussion on the Diamondville Platform. Rather wait for dual core version which would allow for most optimum playback using cMP.
I suspect these processors will only do SRC@121db SNR. Definitely worth a try.
- Need help using cPlay with Lynx ASIO - kwlam 13:41:04 07/05/08 (9)
In Reply to: RE: cPlay - the open source high-end audio player using ASIO posted by cics on May 05, 2008 at 12:31:58
Hi cics,
Thank you for building this really nice stuff. I have setup a system with cMP + foobar2000 with a Lynx L22 card. The improvement is significantly better than the original AOBCT and could easily beat most of the high end CD players.
I also installed cPlay but couldn't get it to work with the Lynx ASIO driver. I tried cPlay with and without cMP and got the same error messages. There are 3 messages; the last message indicates a cPlay restart is required but restarting cPlay didn't help. Is there a way to allow cPlay to start playing the track without asking for a restart? Here are the messages....
Message #1:
cPlay Diagnostics
cicsPlay 1.0b18, licensed under GNU GPL
ASIO Technology by Steinberg Media Technologies GmbH
ASIO is a trademark and software of Steinberg Media Technologies GmbH
ASIO Driver Initializing...
Name (ASIO Lynx)
Player ASIO version (2)
Driver ASIO version (2)
Message (No ASIO Driver Error)
Channels (inputs: 16, outputs: 16)
Buffer details (min: 32, max: 1024, preferred: 32, granularity: -1)
Sample rate ( 44100)
Output Ready? Supported
Preparing buffers... successful
Latencies (input: 98, output: 104)
Hardware supports additional optimization - enabled
ASIO Driver initialized.
DSP starting...
DSP Buffer (6144 samples or 139ms/ 7.2/ 1378.1) allocation successful
ASIO Driver starting...
Input 1 [ Lynx L22 Record 01 Left ] Active (0) type (32bits)
Input 2 [ Lynx L22 Record 01 Right ] Active (0) type (32bits)
Input 3 [ Lynx L22 Record 02 Left ] Active (0) type (32bits)
Input 4 [ Lynx L22 Record 02 Right ] Active (0) type (32bits)
Input 5 [ Lynx L22 Record 03 Left ] Active (0) type (32bits)
Input 6 [ Lynx L22 Record 03 Right ] Active (0) type (32bits)
Input 7 [ Lynx L22 Record 04 Left ] Active (0) type (32bits)
Input 8 [ Lynx L22 Record 04 Right ] Active (0) type (32bits)
Input 9 [ Lynx L22 Record 05 Left ] Active (0) type (32bits)
Input 10 [ Lynx L22 Record 05 Right ] Active (0) type (32bits)
Input 11 [ Lynx L22 Record 06 Left ] Active (0) type (32bits)
Input 12 [ Lynx L22 Record 06 Right ] Active (0) type (32bits)
Input 13 [ Lynx L22 Record 07 Left ] Active (0) type (32bits)
Input 14 [ Lynx L22 Record 07 Right ] Active (0) type (32bits)
Input 15 [ Lynx L22 Record 08 Left ] Active (0) type (32bits)
Input 16 [ Lynx L22 Record 08 Right ] Active (0) type (32bits)
Output 1 [ Lynx L22 Play 01 Left ] Active (1) type (32bits) > Left Channel
Output 2 [ Lynx L22 Play 01 Right ] Active (1) type (32bits) > Right Channel
Output 3 [ Lynx L22 Play 02 Left ] Active (0) type (32bits)
Output 4 [ Lynx L22 Play 02 Right ] Active (0) type (32bits)
Output 5 [ Lynx L22 Play 03 Left ] Active (0) type (32bits)
Output 6 [ Lynx L22 Play 03 Right ] Active (0) type (32bits)
Output 7 [ Lynx L22 Play 04 Left ] Active (0) type (32bits)
Output 8 [ Lynx L22 Play 04 Right ] Active (0) type (32bits)
Output 9 [ Lynx L22 Play 05 Left ] Active (0) type (32bits)
Output 10 [ Lynx L22 Play 05 Right ] Active (0) type (32bits)
Output 11 [ Lynx L22 Play 06 Left ] Active (0) type (32bits)
Output 12 [ Lynx L22 Play 06 Right ] Active (0) type (32bits)
Output 13 [ Lynx L22 Play 07 Left ] Active (0) type (32bits)
Output 14 [ Lynx L22 Play 07 Right ] Active (0) type (32bits)
Output 15 [ Lynx L22 Play 08 Left ] Active (0) type (32bits)
Output 16 [ Lynx L22 Play 08 Right ] Active (0) type (32bits)
ASIO started successfully.
AUDIO source [ D:\CD\ITouchTheStar\Track01.wav ]
Genre = Unknown
Artist = D:\CD\ITouchTheStar\Track01.wav
Album = WAV
FILE: D:\CD\ITouchTheStar\Track01.wav
Wav file size = 46MB
Rate = 16/44100 (2 channels)
Samples = 12183360 ( 4:36)
RAM allocation (47MB) successful
DSP Started.
Play starts after RAM load (available 1810MB, system 2038MB).
Volume (-30.0db, phase 0)
SRC (1, 44100)
SRC upsampling will be bypassed.
-------------------
Message #2:
cPlay
Player has stopped or did NOT start successfully. This may occur if:
1. Settings has not been done
2. No .cue, .wav or .flac was supplied
3. An external ASIO event occurred
New sample rate or latency
Driver requested resync or reset
4. ASIO did NOT start successfully
5. DSP did NOT start successfully
6. Content file is invalid, too large or you don't have enough RAM
7. No ASIO dirvers installed
Restart player and/or correct settings.
OK
--------------------
Message #3:
cPlay Diagnostics
ASIO buffer settings (Rate or Latency) have changed. Please restart player.
RAM Loading...
File 'D:\CD\ITouchTheStar\Track01.WAV'
OK
- Some experience to share - alfred-kraut@arcor.de 10:08:58 07/06/08 (0)
In Reply to: RE: Need help using cPlay with Lynx ASIO posted by kwlam on July 05, 2008 at 13:41:04
Hallo cics,
at first many thanks for your player. A big footprint in pc-audio history.
I am working for 3 years with PC-audioand i used foobar ,
different soundcards: -1212m e-mu etc.
diffeent USB to SPDif converter:
Trend, and hoerwege usb to spdif).
The PC only as a transporter with an external DAC, because all solutions for DA in the PC are lacking quality.
Best results till now:
AMD 3800 underclocked to 800Mhz. 1GB DDR1 with 200FSB.
SATA Drive with external PS.
Windows stripped down to 70MB in the RAM.
AND !!the ESI Juli latency 48 and upsampling to 88,2.
This System is blowing away every CD-Playeer i Know ( High End).
So next steps are a look at Intels Atom Processor on a car mobo with 12 V powered by Sealed Batteries ( 4x12 V to achieve little inner resistance.)
And!! Lokking for cplay 20 21 22 and so on .
Cheers
alfred
- RE: Need help using cPlay with Lynx ASIO - cics 17:52:53 07/05/08 (7)
In Reply to: RE: Need help using cPlay with Lynx ASIO posted by kwlam on July 05, 2008 at 13:41:04
Problem is in Message #3.
Your Lynx card is changing critical settings (which must be done at ASIO initialization). cPlay is gettting this message at start of play which it then stops and gives message #3. I've seen this same issue on Vista. It's much better to use XP SP2 Pro. Can you test using another soundcard?
- RE: Need help using cPlay with Lynx ASIO - kwlam 11:30:31 07/07/08 (0)
In Reply to: RE: Need help using cPlay with Lynx ASIO posted by cics on July 05, 2008 at 17:52:53
Hi cics, I'm able to use cPlay with RME Fireface ASIO. cPlay sounds very nice, I had briefly compared it with foobar 9.5.3 on my XP machine that is without any optimization. cPlay is clearly a much better player....Thanks
- RE: Need help using cPlay with Lynx ASIO - riboge 16:09:47 07/06/08 (2)
In Reply to: RE: Need help using cPlay with Lynx ASIO posted by cics on July 05, 2008 at 17:52:53
From my own experimenting Lynx asio will not work under Vista or XP pro sp3 running any version of cPlay after b7. It does work with b7, so you could try that though sq advances were made subsequently.
For general info: EA freeway 2 and offramp turbo 2 will run with Asio4all in cPlay or cMP/cPlay. This is so although its maker says it doesn't work with ASIO and indeed it does not work with creative asio or lynx asio, in my experience.
- I'll try and get a Lynx card and test - cics 23:42:59 07/06/08 (1)
In Reply to: RE: Need help using cPlay with Lynx ASIO posted by riboge on July 06, 2008 at 16:09:47
.
- Lynx L22 working with cPlay 1.0_final - kwlam 14:57:11 07/15/08 (0)
In Reply to: RE: I'll try and get a Lynx card and test posted by cics on July 06, 2008 at 23:42:59
Hi cics, I just tried the latest cPlay 1.0_final. It works with the Lynx L22. Thanks.
- RE: Need help using cPlay with Lynx ASIO - kwlam 20:03:56 07/05/08 (2)
In Reply to: RE: Need help using cPlay with Lynx ASIO posted by cics on July 05, 2008 at 17:52:53
Thanks for the reply, I'm using XP Home SP3 with a MSI G33M motherboard.
I will try RME Fireface 800 and report the result later.
- Use AWE with cPlay - cics 00:09:11 07/06/08 (1)
In Reply to: RE: Need help using cPlay with Lynx ASIO posted by kwlam on July 05, 2008 at 20:03:56
AWE gives better results. Setup of AWE (lock privilege) is explained here .
Any chance of using XP Pro?
- RE: Use AWE with cPlay - kwlam 14:40:07 07/06/08 (0)
In Reply to: RE: Use AWE with cPlay posted by cics on July 06, 2008 at 00:09:11
If I can get cPlay to work with Lynx or RME Fireface, I would also try AWE. Other than AWE, is there any other reasons for using XP Pro? Thanks!
- Nice bug in P4M900 gives me 600Mhz...which sounds best - frennekamp@netcologne.de 10:52:20 07/05/08 (3)
In Reply to: RE: cPlay - the open source high-end audio player using ASIO posted by cics on May 05, 2008 at 12:31:58
I dpn'T know if anyone else noticed that: ON the P4M900, when it is completely untouched, it gives the CPU 100 Mhz as reference instead of 200 Mhz. CPU-ID reports really this 600 Mhz when no heavy hit is there. This sounds in my setp best. I figured out, that you can get rid of this bug if you increase in the setup to 201 mhz, save and afterwards go back to 200 Mhz, and than you have for a e2180 as well 1200 as the orgininally intended lowest seed. But 600 mhz sounds better. Clearer.
Cics, have you compared that with the G31 / e7200 ? How low can the Gigbyte G31 board bring a CPU ? And would the E8200 be better or the e7200 ?
- CPU Underclocking - cics 17:44:35 07/05/08 (2)
In Reply to: RE: Nice bug in P4M900 gives me 600Mhz...which sounds best posted by frennekamp@netcologne.de on July 05, 2008 at 10:52:20
That's a nice bug! E1200 with SRC @ 121db SNR would do very well. Benefit of lowering CPU clock is lower power consumption.
I haven't done a direct G31/E7200 vs P4M900/E1200 comparison but can confirm that G31/E7200 using SRC@121db SNR is very similar - I couldn't tell any differences from memory. Using SRC@145db SNR is a different matter. Sound improves. This is best done using E7200 which at 96k output, I can underclock to 966MHz. I prefer not going lower to maintain some headroom on CPU 0 (load is ~78%).
Avoid E8200 - E7200 offers lowest power consumption (lower than E1200).
- OK, my P4M900 burned one PCI-Slot - so stay with that model or go to G31 if no upsampling is wanted ? - frennekamp@netcologne.de 10:48:53 07/06/08 (1)
In Reply to: RE: CPU Underclocking posted by cics on July 05, 2008 at 17:44:35
I somehow smoked the PCI-Slot of that P4M900...so need to buy a new motherboard.
I have no yet compared the new SRC 145 db upsampler completely, but I still kike no upsampling in the PC best. I have an integrated upsampler in the RAKK-DAC which anyhow does upsampling to 192khz with any input signal...so, I guess we compare now simply upsamplers and effecs on data transport / jitter. I like the upsampling done in the DAC best currently.
So, if I don't need the muscle of the e7200: Would you recomend thatI go to e7200 at 966 Mhz with the G31 board you suggest or buy another P4M900 ?
THX for your advice.
- RE: OK, my P4M900 burned one PCI-Slot - so stay with that model or go to G31 if no upsampling is wanted ? - cics 23:40:14 07/06/08 (0)
In Reply to: RE: OK, my P4M900 burned one PCI-Slot - so stay with that model or go to G31 if no upsampling is wanted ? posted by frennekamp@netcologne.de on July 06, 2008 at 10:48:53
Go with Gigabyte/E7200 - it betters E1200.
My preference is for upsampling in computer and not DAC. You want to do very little math in DAC. Also, DACs don't specify type of upsampler being used - its subject to constraints (processing power) and often result in inferior methods.
cMP 2 into dCS Scarlatti DAC at 24/96 is superior to non-upsampled 44.1 output (I get digital artifacts like grainy / forward highs). I want to now send 192k using dual AES.
- F6 has replaced F5d - rickmcinnis@dogwoodfabrics.com 09:19:06 07/05/08 (0)
In Reply to: RE: cPlay - the open source high-end audio player using ASIO posted by cics on May 05, 2008 at 12:31:58
Dear cics,
When I went to get the recommended F5d I see there is a new iteration.
I could not get the machine to recognize QFLASH with the USB memory so I used the @BIOS software on the supplied CD(I think that is right) and installed within WINDOWS which was as simple as could be. After using it I deleted it.
Connecting an IDE CDROM to the GIGABYTE board was the trick, it was recognized and the combination resulted in the fastest WINDOWS installation I have experienced.
The first part (with the blue screen) blazed by, it always seemed to take forever before.
I checked the forum before going upstairs to install cPLAY #19 and to listen to a proper implementation. As always it will take some time to setlle in.
THANKS,
Rick McInnis
- RE: cPlay - the open source high-end audio player using ASIO - Doc Gaw 06:13:12 07/05/08 (1)
In Reply to: RE: cPlay - the open source high-end audio player using ASIO posted by cics on May 05, 2008 at 12:31:58
Cics: Are you still using the Juli@soundcard or have you found something better for balanced analog or SPDIF output?.
- Soundcards used - cics 07:00:08 07/05/08 (0)
In Reply to: RE: cPlay - the open source high-end audio player using ASIO posted by Doc Gaw on July 05, 2008 at 06:13:12
For reference setup outputting 24/96 SPDIF I use RME HDSP9652 or Juli@ (without analogue PCB). Both are very good with Juli@ having a slight edge over RME. Juli@ also has advantage of 24/192k SPDIF output.
Since implementing new hardware specs, I haven't tried the Juli@ - will be doing it soon. Previous setup had both cards installed as I got dedicated IRQs and this made it easier to switch DACs (Juli@ to AA Prestige SE & RME to dCS Scarlatti). Not so with Gigabyte which offers only one PCI slot with dedicated IRQ.
My 2nd setup uses EMU1212M which offers better balanced analogue outs than Juli@. Juli@'s analogue outs are average quality.
I plan to get the Lynx AES-XLR for dual AES 192k output (not soon though).
- RE: cPlay - the open source high-end audio player using ASIO - gagnem 09:35:55 07/04/08 (2)
In Reply to: RE: cPlay - the open source high-end audio player using ASIO posted by cics on May 05, 2008 at 12:31:58
Hi cics, I finally build : AOBCT+cMP+cPlay, with the G31/E7200 HW.
Even if some hardware and software tweaks still need to be done, the results are stunning.
I just have a few questions.
1. in BIOS setup, under M.I.T. "Enable CPU Host Clock Control" and "Set System MM (SPD)" both keep reseting themselves every time I reboot the machine. I'm sure I pressed F10 to save. I did it a couple of times, and all the other settings are ok. I use F5D bios version. Is this normal ?
Has someone else reported this ?
2. In AOBCT, step 14 (kernel changes) do you recommand item d. (/pcilock) with this hardware ? I don't see the "PNP OS Installed" option.
3. Almost all my tracks are in FLAC one file per track. I still can't figure a way to simply create a .cue file that works with Cmp, cPlay.
If I understand well, cmp converts FLAC files to WAV during RAM loading, so I won't gain (except loading speed) to convert them all to WAV.
Is there a workaround other than loading the tracks one by one, which is pretty far from being ideal.
Thanks for all the time you put in this project.
It is VERY appreciated.
- RE: cPlay - the open source high-end audio player using ASIO - seger 15:27:37 07/05/08 (0)
In Reply to: RE: cPlay - the open source high-end audio player using ASIO posted by gagnem on July 04, 2008 at 09:35:55
Foobar cuesheet creator plugin works great with flac files. Details are here
- RE: Questions - cics 10:24:37 07/04/08 (0)
In Reply to: RE: cPlay - the open source high-end audio player using ASIO posted by gagnem on July 04, 2008 at 09:35:55
1. in BIOS setup, under M.I.T. "Enable CPU Host Clock Control" and "Set System MM (SPD)" both keep reseting themselves every time I reboot the machine. I'm sure I pressed F10 to save. I did it a couple of times, and all the other settings are ok. I use F5D bios version. Is this normal ?
F5d is correct BIOS. Gigabyte mobo does resets when it can't boot. Try setting higher host frequency - start with default 266MHz and work downwards. Set RAM SPD to x2 always.
2. In AOBCT, step 14 (kernel changes) do you recommand item d. (/pcilock) with this hardware ? I don't see the "PNP OS Installed" option.
Yes - I get no problems. There's more BIOS options when you press Ctrl+F1 and IRQs are set for Auto.
3. Almost all my tracks are in FLAC one file per track. I still can't figure a way to simply create a .cue file that works with Cmp, cPlay.
If I understand well, cmp converts FLAC files to WAV during RAM loading, so I won't gain (except loading speed) to convert them all to WAV.
Is there a workaround other than loading the tracks one by one, which is pretty far from being ideal.
Contact Seger - he has found a way to build cue sheets for multiple FLAC files - see here . cPlay decodes FLAC during RAM Load - so sonically both FLAC and WAV sound the same.
- Just listened to src upsampler in clpay and ... Oh My God - theob 07:21:01 07/03/08 (20)
In Reply to: RE: cPlay - the open source high-end audio player using ASIO posted by cics on May 05, 2008 at 12:31:58
I never tried before because I didn't think it would work (running dell amd antlon dual core with underclocking/under volting; asio4all ; Benchmark). Anyway with undeclocking/undervolting upsampling to 24/96 sounds very very good no more upsample subtle hardness but bass and dynamics were a little lacking vs 16/44 setting. But src is another story--best of all worlds. When I had my wife listen to it she asked me what I bought, what new hardware did I get. When I told her nothing its the Secret Rabbit Code in cplay (which I already had but never tried)she said '...yea right!'
The imaging is spectacular not overdone just very natural. The whole soundstage is more coherent. The dynamics/bass never fail to impress. This reminds me of my younger days listening to vinyl off my vpi/ET II/ moving coil cartridge.
Thank you Cics you can never be thanked enough.
I just might go all the way and build a cmp from ground up now that I have seen/heard a glimpse of the 'top of the mountain'.
- RE: Just listened to src upsampler in clpay and ... Oh My God - appleteapot 09:17:38 07/03/08 (19)
In Reply to: RE: Just listened to src upsampler in clpay and ... Oh My God posted by theob on July 03, 2008 at 07:21:01
Gee, I totally agree with you...Using the maximum configuration, AOBCT+CMP+CPlay, mobo underclock to 600Mhz, disable LAN, 1394 and everything else, I think this is poor man's Weiss Jason+Medea
Edits: 07/03/08 07/03/08 07/03/08
- RE: Just listened to src upsampler in clpay and ... Oh My God - cics 10:07:01 07/03/08 (18)
In Reply to: RE: Just listened to src upsampler in clpay and ... Oh My God posted by appleteapot on July 03, 2008 at 09:17:38
Use new hw specifications for new build. This gives 145db SNR performance.
Do you have FSB at 600MHz?
- This is what I have - theob 09:06:37 07/05/08 (4)
In Reply to: RE: Just listened to src upsampler in clpay and ... Oh My God posted by cics on July 03, 2008 at 10:07:01
CPU-Z 1.46 report file
Processor(s)
Number of processors 1
Number of cores 2 per processor
Number of threads 2 per processor
Name AMD Athlon 64 X2 4400+
Code Name Brisbane
Specification AMD Athlon(tm) 64 X2 Dual Core Processor 4400+
Package Socket AM2 (940)
Family/Model/Stepping F.B.1
Extended Family/Model F.6B
Brand ID 4
Core Stepping BH-G1
Technology 65 nm
Core Speed 2004.3 MHz
Multiplier x Bus speed 10.0 x 200.4 MHz
HT Link speed 1002.2 MHz
Stock frequency 2300 MHz
Instruction sets MMX (+), 3DNow! (+), SSE, SSE2, SSE3, x86-64
L1 Data cache (per processor) 2 x 64 KBytes, 2-way set associative, 64-byte line size
L1 Instruction cache (per processor) 2 x 64 KBytes, 2-way set associative, 64-byte line size
L2 cache (per processor) 2 x 512 KBytes, 16-way set associative, 64-byte line size
Chipset & Memory
Northbridge NVIDIA GeForce 6150 rev. A2
Southbridge NVIDIA nForce 410/430 MCP rev. A3
Memory Type DDR2
Memory Size 1024 MBytes
Memory Frequency 334.1 MHz (CPU/6)
CAS# Latency (tCL) 5.0 clocks
RAS# to CAS# (tRCD) 5 clocks
RAS# Precharge (tRP) 5 clocks
Cycle Time (tRAS) 15 clocks
Bank Cycle Time (tRC) 21 clocks
Command Rate (CR) 1T
System
System Manufacturer Dell Inc
System Name Dimension E521
System S/N 38462D1
Mainboard Vendor Dell Inc
Mainboard Model
BIOS Vendor Dell Inc
BIOS Version 1.1.6
BIOS Date 04/07/2007
Memory SPD
Module 1 DDR2, PC2-5300 (333 MHz), 512 MBytes, Hyundai Electronics
Module 2 DDR2, PC2-5300 (333 MHz), 512 MBytes, Hyundai Electronics
Software
Windows Version Microsoft Windows XP Home Edition Service Pack 2 (Build 2600)
DirectX Version 9.0c
- RE: This is what I have - cics 00:56:39 07/06/08 (3)
In Reply to: RE: This is what I have posted by theob on July 05, 2008 at 09:06:37
I'm not familiar with AMD / NVidia.
From what you showing, outputting 24/96 with this setup using SRC@145db SNR will work. Much better results can be gained using a more power efficient setup - by this I mean your current setup is using high clock speeds (CPU @ 2GHz, RAM @ 334MHz). Also, L2 cache is not ideal at 512KB per processor - limited to 1MB and not shared but split: 512KB is allocated per CPU core makes it less ideal. Intel Core 2 Duo (Conroe and up) implements a shared L2 cache.
- RE: This is what I have - theob 08:48:50 07/09/08 (1)
In Reply to: RE: This is what I have posted by cics on July 06, 2008 at 00:56:39
ok here are my follow up questions.Since I have an amd AMD Athlon(tm) 64 X2 Dual Core Processor 4400+ cpu with Socket AM2 (940)can I really install a Intel Core 2 Duo (Conroe and up) w/o replacing the mobo?
Can I install one of your recommended rams and will it work with my amd if I'm stuck with it? Here is my current ram again DDR2, PC2-5300 (333 MHz), 512 MBytes, Hyundai Electronics. Or is this ok?
I'm looking for near term solutions with current pc since I cannot buy all the cmp hardware at this time. But cmp is my longterm solution.
Edits: 07/09/08
- RE: This is what I have - cics 10:02:10 07/09/08 (0)
In Reply to: RE: This is what I have posted by theob on July 09, 2008 at 08:48:50
can I really install a Intel Core 2 Duo (Conroe and up) w/o replacing the mobo?
No.
Can I install one of your recommended rams and will it work with my amd if I'm stuck with it? Here is my current ram again DDR2, PC2-5300 (333 MHz), 512 MBytes, Hyundai Electronics. Or is this ok?
Yes, use recommended Kingston RAM. I'm not familiar with Hyundai - I suspect these are not designed for overclocking. Rather keep this RAM and start with new mobo & CPU. You can then build up from there as all other components like DDR2 RAM and HDDs can be used.
- RE: This is what I have - theob 03:00:58 07/06/08 (0)
In Reply to: RE: This is what I have posted by cics on July 06, 2008 at 00:56:39
If I keep the processor is there anything else I can do in bios to improve this? (Recognizing I am into Cool & Quiet which altough variable in speed/voltage does underclock/undrrvolt somewhat.)
- RE: Just listened to src upsampler in clpay and ... Oh My God - appleteapot 09:40:28 07/04/08 (10)
In Reply to: RE: Just listened to src upsampler in clpay and ... Oh My God posted by cics on July 03, 2008 at 10:07:01
I have Bus speed at 100Mhz, multiplier x6, FSB 400, Core speed 600Mhz (all at lowest possible). My motherboard is Asus Commando which is very nosiy
- re: BUS at 100MHz, FSB at 400MHz, CPU at 600MHz - cics 10:11:52 07/04/08 (9)
In Reply to: RE: Just listened to src upsampler in clpay and ... Oh My God posted by appleteapot on July 04, 2008 at 09:40:28
On Gigabyte mobo I can go down to 161MHz bus speed with RAM at 322 and CPU at 966 - anything lower becomes unstable. This gives improvements and mobo is dead quiet. Vcc (CPU voltage) stays constant at 1.120V during 96k 145db SNR playback (with EIST at x6, i.e. CPU constant at 966MHz) and RAM is set for 3-3-3-9.
For 192k output, a higher bus speed would be needed.
- RE: re: BUS at 100MHz, FSB at 400MHz, CPU at 600MHz - appleteapot 11:29:31 07/04/08 (8)
In Reply to: RE: re: BUS at 100MHz, FSB at 400MHz, CPU at 600MHz posted by cics on July 04, 2008 at 10:11:52
I guess you may try set RAM back to 5-5-5-15 and 100Mhz bus (or lowest) to get 600 Mhz (or lowest) core speed. Focus (sound stage) is at max level while high frequency noise is at lowest, so cplay can "squeeze" further the finest details out from the wave file. Just that at 600Mhz you can only upsample to 88.2KhzFor CPU, would 45nm 8000 series be better than 7000 series? Because there are more L2 cache in 8000 series.
For SRC145 at 192Khz, my current 6320+Commando need around 1.2-1.5Ghz in order to run cplay smoothly, but setting speed at such speed will sacrifice 1/3 of the sound stage :( The Commando mobo is definitely not audiophile-friendly.
Do you know whether Asus G31 motherboards are any good? I just "feel" that Asus BIOS are more stable than Gigabyte.
Edits: 07/04/08
- RE: re: BUS at 100MHz, FSB at 400MHz, CPU at 600MHz - cics 02:55:13 07/05/08 (7)
In Reply to: RE: re: BUS at 100MHz, FSB at 400MHz, CPU at 600MHz posted by appleteapot on July 04, 2008 at 11:29:31
5-5-5-15 allows for 160MHz Bus. I've settled with 3-3-3-9 and 161MHz Bus. This allows for 24/96k output without any Vcc changes plus some headroom.
E7200 is best option due to lowest power consumption. L2 cache at 3MB is more than whats needed.
Had a look at ASUS G31 offerings and their BIOS settings seems restrictive. Gigabyte's EG31 or G31 offers greater flexibility - just make sure you do the BIOS update. I've asked Gigabyte to include 2 additional BIOS settings (Spread Spectrum & L2 Cache ECC checking) - disabling these would be a further improvement. Gigabyte responded positively to this request...
- RE: re: BUS at 100MHz, FSB at 400MHz, CPU at 600MHz - appleteapot 06:37:49 07/06/08 (1)
In Reply to: RE: re: BUS at 100MHz, FSB at 400MHz, CPU at 600MHz posted by cics on July 05, 2008 at 02:55:13
Hi Cics,I re-check 600Mhz vs 900Mhz again and they do sound different... At 600Mhz/ upsample 88.2, the playback has better sound stage, finer details and is more airy. On the other hand, although I can upsample to 96Khz at 900Mhz, the airy feel loss about 15% and I have to set it back to 600Mhz after repeated comparisons. Highly recommend to see if 600Mhz is achievable...
Edits: 07/06/08 07/06/08
- Interesting tradeoff between power consumption and level of Upsampling - cics 06:13:00 07/07/08 (0)
In Reply to: RE: re: BUS at 100MHz, FSB at 400MHz, CPU at 600MHz posted by appleteapot on July 06, 2008 at 06:37:49
Seems in your case using E6320 processor, 600MHz & 88.2 is better than 900MHz & 96. That additional benefit to 96k is lost due to more power (causing interference).
Not sure if you'd get same results using E7200 processor - lowest stable underclocking using latest BIOS (F6) is 990MHz (FSB @ 165MHz).
Overall, I would do 192k at higher power consumption using E7200 as benefit of 192k with SRC@145db SNR is significant.
- RE: re: BUS at 100MHz, FSB at 400MHz, CPU at 600MHz - artcheng@netvigator.com 03:03:24 07/05/08 (4)
In Reply to: RE: re: BUS at 100MHz, FSB at 400MHz, CPU at 600MHz posted by cics on July 05, 2008 at 02:55:13
Cic,
Seems that G33 is slightly more power efficient?
http://forums.microsoft.com/WindowsHomeServer/ShowPost.aspx?PostID=3189161&SiteID=50
- G33 vs G31 power consumption - cics 03:34:14 07/05/08 (3)
In Reply to: RE: re: BUS at 100MHz, FSB at 400MHz, CPU at 600MHz posted by artcheng@netvigator.com on July 05, 2008 at 03:03:24
See this review which tests both G33 & G31 chipsets using ECS mobo. Look at power consumption without the graphics card:
Both are close but in cMP, video in northbridge is disabled. Load testing has bias towards graphics whilst both chipsets offer same RAM performance . It's worth testing.
- RE: G33 vs G31 power consumption - artcheng@netvigator.com 03:56:13 07/10/08 (2)
In Reply to: RE: G33 vs G31 power consumption posted by cics on July 05, 2008 at 03:34:14
Seems like the G43 chip set is even more power saving.....
- RE: G33 vs G31 power consumption - appleteapot 09:44:28 07/10/08 (1)
In Reply to: RE: G33 vs G31 power consumption posted by artcheng@netvigator.com on July 10, 2008 at 03:56:13
Hi ackcheng,
where did you get the G43 power consumption information? I am still thinking of buying a full size board and fill all the empty slots with NO-PCI.
On the other hand, I just bought GA-G31M-S2L and 7200 this evening. However, they come DOA... I need to replace them tomorrow. I may get a Asus instead.
- RE: G33 vs G31 power consumption - artcheng@netvigator.com 17:58:03 07/10/08 (0)
In Reply to: RE: G33 vs G31 power consumption posted by appleteapot on July 10, 2008 at 09:44:28
No direct comparison yet. Only report on the general use of 65nm processing that makes the G43 and g45 chipset less power hungry. I am planning to buy G33 chipset at I cannot find G31 chipset that allows me to use 4G of ram
- RE: Just listened to src upsampler in clpay and ... Oh My God - theob 12:41:25 07/03/08 (1)
In Reply to: RE: Just listened to src upsampler in clpay and ... Oh My God posted by cics on July 03, 2008 at 10:07:01
Which one is fsb on cpuz?
- Interesting, with src running, processor and memory speeds go back up - theob 13:25:11 07/03/08 (0)
In Reply to: RE: Just listened to src upsampler in clpay and ... Oh My God posted by theob on July 03, 2008 at 12:41:25
...so I am not getting underclocking with src as I do with -121 db upsampler. But -146 db upsampler still sounds better. But
I know now if I go complete cmp
(all the way) sonic performance will be better.
- I implented AMD's cool & quiet routine to underclock/undervolt - theob 07:18:16 07/02/08 (2)
In Reply to: RE: cPlay - the open source high-end audio player using ASIO posted by cics on May 05, 2008 at 12:31:58
and it seems to work. Verfied through cpuz. However AMD's cool & quiet user guide say I also can verify by going to my computer properties and see if the cpu processor is now showing around 800 mhz (which it does not). But cpuz does. Is cpuz the preferred authority here?
BTW it does sound better.
- Go with CPU Z - it shows correct clock speed and voltage. - cics 10:24:48 07/02/08 (1)
In Reply to: RE: I implented AMD's cool & quiet routine to underclock/undervolt posted by theob on July 02, 2008 at 07:18:16
.
- RE: Go with CPU Z - it shows correct clock speed and voltage. - theob 10:29:41 07/02/08 (0)
In Reply to: RE: Go with CPU Z - it shows correct clock speed and voltage. posted by cics on July 02, 2008 at 10:24:48
thanks
- Want to under clock and under volt on amd anthlon 64 x 2 processor 4400... - theob 19:05:31 06/30/08 (0)
In Reply to: RE: cPlay - the open source high-end audio player using ASIO posted by cics on May 05, 2008 at 12:31:58
...but are hesitant to go into bios and reset a whole bunch of stuff (like I was)? I have a recomendation use Cool&Quiet technology from AMD. Bottom line is you select Cool & Quiet routine from the control panel and download drivers from amd and you're there. If you call for properties on my computer you can verify that C&Q is working by confirming that clock speed is now around 800 Mhz (mine went from 2.3ghz to 990 mhz. I believe stuff is also undervolted also. All of this was originally for a 'green power consumption brand tag' for amd. Yes it is dynamic but it works. And it also helps audiofiles using cplay. I found a significant sonic performance improvement. Thanks again cics and frodan. There is more than 1 way to skin a cat.See link
- http://www.amd.com/us-en/Processors/ProductInformation/0,,30_118_9485_9487%5e10272,00.html (Open in New Window)
Edits: 06/30/08
- RE: cPlay - the open source high-end audio player using ASIO - Frodan 00:58:33 06/29/08 (2)
In Reply to: RE: cPlay - the open source high-end audio player using ASIO posted by cics on May 05, 2008 at 12:31:58
Did a first test yesterday using an old P4 1.4ghz, doing all the optimizations that were possible, including powering the hard disk from a separate supply and turning off everything that wasn't needed. The computer was sitting in the kitchen with wires through a hole punched in the wall so I didn't have to listen to the CPU fan.
At first listen the amount of detail was simply amazing. Despite a problem tweeter in the left speaker, the sense of actually being there was uncanny. Vangelis' Mythodea CD being a very recent recording with all kinds of instruments and vocals was particularly outstanding.
I have a CD of a 60 piece orchestra playing film music for which I was in and around the orchestra when it was recorded, and this is the first time from any sound system I can recognize where all the instruments are positioned and they actually sound as I remembered.
Now am setting up the first real version using a donated Biostar P4M890 motherboard with a Celeron D1200 processor, and 2 standard sticks of Kingston 1 gbyte RAM. 2 Low Latency sticks are on the way and I will upgrade to the Gigabyte motherboard and E7200 CPU as soon as funds permit it.
I have generated a new copy of the XP Cd, putting back a few things I had removed on the previous version. I am still using nLite to generate them because it allows me to remove all the excess stuff from Windows and considerably speeds up the install process. I have also received my Juli@ sound card and can't wait to hear the results.
The only problem I have had so far is using the Raid driver that came with the motherboard, it loads properly after hitting F6 in the setup, but when Windows starts copying files to the hard disk, it can't find the driver file (viamraid.sys), somehow Windows looses the reference to it.
Now that I have something worth writing about I will post regular feedback here.
Frodan
- That's amazing feedback, thx. RE: RAID on VIA... - cics 11:38:30 06/29/08 (1)
In Reply to: RE: cPlay - the open source high-end audio player using ASIO posted by Frodan on June 29, 2008 at 00:58:33
Using RAID won't give sound benefit, just faster RAM load times as playback now occurs directly from RAM.
For RAID installation, you must copy VIA's XP driver files onto a floppy. Windows installation then installs drivers from Floppy Drive. Also, keep floppy in place throughout installation as Windows accesses it.
- RE: That's amazing feedback, thx. RE: RAID on VIA... - Frodan 13:20:01 06/29/08 (0)
In Reply to: RE: That's amazing feedback, thx. RE: RAID on VIA... posted by cics on June 29, 2008 at 11:38:30
I've had to ditch the Raid as Windows refused to copy the driver from the floppy even though it had loaded it after I had pressed F6. Anyways, the Gigabyte doesn't have Raid and we don't need it anymore so I'm not installing it.
BTW Almost done with this version, can't wait to listen to it.
Frodan
- RE: cPlay - the open source high-end audio player using ASIO - seger 09:19:55 06/27/08 (4)
In Reply to: RE: cPlay - the open source high-end audio player using ASIO posted by cics on May 05, 2008 at 12:31:58
Just listening to Buxtehude harpsichord album in cMP/cPlay v19, 24/192, 121SNR, AWE - breathtaking.
Some difficulty with 2nd system and emu0202USB. Diagnostics saying "ASIO start FAILED. Only 32 bit ASIO format supported." Then system/cPlay lock. Change emu to 32 bits and it's ok, and sounds wonderful. Why only 32 bits?
With the emu I have to use foobar to get at asio settings - remove the emu asio, then add again, then configure, and close foobar without playing, or the CPU maxes out and crash! But the analog output of this annoying device sounds so good, I'm keeping it.
Btw, winamp/otachan asio is playing with 16, 24, or 32 bits.
- cPlay only supports pure 32 bit ASIO - cics 23:44:30 06/27/08 (3)
In Reply to: RE: cPlay - the open source high-end audio player using ASIO posted by seger on June 27, 2008 at 09:19:55
ASIO standard defines many input and output formats. Trying to accommodate all would compromise quality as data conversion is done in most important area of player (giving more jitter distortion). There's a way to potentially avoid this but you end up with a clumsy unmaintainable solution.
For EMU / Creative cards, never set svchost and lsass to suspend (in cMP). Your system will go into indefinite wait. ASIO settings also depend on these to be operational.
New hardware specifications allows for SNR145db option - this is pure analogue stuff and will better what you're getting!
- RE: cPlay only supports pure 32 bit ASIO - fedespan 10:37:35 04/09/09 (2)
In Reply to: RE: cPlay only supports pure 32 bit ASIO posted by cics on June 27, 2008 at 23:44:30
I can use cplay with a Novation Speedio but with my new 0404 sub that's all I can get. I already tried ALL (reinstalling etc.). Could you help? Cplay sounds great on the other card...
------------------------------------------------------------
cPlay 2.0b20 (SSE2), licensed under GNU GPL
ASIO Technology by Steinberg Media Technologies GmbH
ASIO is a trademark and software of Steinberg Media Technologies GmbH
SRC (libsamplerate-0.1.5) licensed under GNU GPL
(see 'note *1' in specifications located in readme.txt for code changes done)
ASIO Driver Initializing...
Name (ASIO E-MU 0404 | USB)
Player ASIO version (2)
Driver ASIO version (2)
Message (No ASIO Driver Error)
Channels (inputs: 4, outputs: 4)
Buffer details (min: 88, max: 22000, preferred: 440, granularity: 44)
Sample rate ( 44100.0)
Output Ready? Supported
Preparing buffers... successful
Latencies (input: 523, output: 656)
ASIO Driver initialized.
DSP starting...
DSP Buffer (14080 samples or 319.3ms/ 3.1hz/ 100.2hz) allocation successful
ASIO Driver starting...
Input 1 [ Analog IN A ] Active (0) type (NOT SUPPORTED, type=17)
Input 2 [ Analog IN B ] Active (0) type (NOT SUPPORTED, type=17)
Input 3 [ SPDIF IN L ] Active (0) type (NOT SUPPORTED, type=17)
Input 4 [ SPDIF IN R ] Active (0) type (NOT SUPPORTED, type=17)
Output 1 [ Analog OUT ] Active (1) type (NOT SUPPORTED, type=17) > Left Channel
Output 2 [ Analog OUT ] Active (1) type (NOT SUPPORTED, type=17) > Right Channel
Output 3 [ SPDIF OUT ] Active (0) type (NOT SUPPORTED, type=17)
Output 4 [ SPDIF OUT ] Active (0) type (NOT SUPPORTED, type=17)
ASIO start FAILED. Only 32 bit ASIO format supported.
- RE: cPlay only supports pure 32 bit ASIO - cics 11:33:23 04/09/09 (1)
In Reply to: RE: cPlay only supports pure 32 bit ASIO posted by fedespan on April 09, 2009 at 10:37:35
Use Foobar to access EMU's ASIO Settings. You should be able to set its bit depth to 32. Your current setting is 24 bits (type 17).
EMU needs to allow for this change outside of ASIO runtime settings.
- RE: cPlay only supports pure 32 bit ASIO - fedespan 12:36:15 04/09/09 (0)
In Reply to: RE: cPlay only supports pure 32 bit ASIO posted by cics on April 09, 2009 at 11:33:23
that's exactly what I've done several times; when I reopen control panel I find it correctly set at 32, but cplay still sees it at 24.
Now I solved it! Looking with regedit, I found 2 instances of ASIO control panel; foobar acted on the other one, and cisplay was set at 24. I manually changed the value to 32, now it works :) Thank for your hint, I figured out what was happening when you explained the meaning of the 17.
- A bunch of questions (just clarifications--just love cicsplay) - theob 09:15:20 06/23/08 (5)
In Reply to: RE: cPlay - the open source high-end audio player using ASIO posted by cics on May 05, 2008 at 12:31:58
I am running xp -> asio4all-> Benchmark with cplay set at 16/44. What happens when I run 24/96 flac files? Is it downconverting to 16/44 or is it playing as 24/96? When I set cplay to 24/96 I hear a difference but not sure whether I prefer either setting (bass is better at 16/44). So if I want to get Reference Recording capability at 24/176 (I currently get good 24/96 flac performance) I was thinking of an es juli card to get 24/176 digital out through toslink or spdif. This way I could get asio 2.0 and use a toslink at 25 feet, site my Benchmark right next to my front end--get the galvanic isolation (I miss the opticis even though it didn't work for me ultimately)-- and use short xlr cables to my dbx x-over. Is too far fetched?
- RE: A bunch of questions (just clarifications--just love cicsplay) - cics 10:57:09 06/23/08 (4)
In Reply to: RE: A bunch of questions (just clarifications--just love cicsplay) posted by theob on June 23, 2008 at 09:15:20
What happens when I run 24/96 flac files? Is it downconverting to 16/44...
With cPlay set to output 44.1k, all material will be converted to 44.1 - in your case, 24/96 flac will be converted to 24/44.1. cPlay's diagnostic indicates if SRC upsampling (or in your case downsampling) is done.
...Is too far fetched?
No. Toslink won't go beyond 24/96. Using RCA/BNC (mod required) will go to 24/192. If upsampling to 192k - use the new hardware specifications which allows for SRC@145db SNR.
- What happens when I set cplay to 96 and play 24/94 wav or flac? - theob 07:00:51 06/25/08 (0)
In Reply to: RE: A bunch of questions (just clarifications--just love cicsplay) posted by cics on June 23, 2008 at 10:57:09
Am I getting true 96 playback from asio4all? It seems to work and sounds very good. The asio4all control panel also says up to ' 96k ' sampling rate'.
Edits: 06/25/08
- W/O getting a soundcard like es juli how does one get asio... - theob 06:23:20 06/24/08 (1)
In Reply to: RE: A bunch of questions (just clarifications--just love cicsplay) posted by cics on June 23, 2008 at 10:57:09
with 192 capabilities?
- You must use an ASIO compatible card for 192k ASIO output. - cics 09:37:57 06/24/08 (0)
In Reply to: RE: W/O getting a soundcard like es juli how does one get asio... posted by theob on June 24, 2008 at 06:23:20
.
- Thanks appreciate your feedback. nt - theob 12:11:23 06/23/08 (0)
In Reply to: RE: A bunch of questions (just clarifications--just love cicsplay) posted by cics on June 23, 2008 at 10:57:09
.
- Using SRC @ 145db SNR - cics 08:35:07 06/22/08 (32)
In Reply to: RE: cPlay - the open source high-end audio player using ASIO posted by cics on May 05, 2008 at 12:31:58
Since b17, the option to use above ultra high quality upsampler is possible. This however requires much more CPU processing ability. Solution is to switch to Intel's latest 45nm processor technology, E7200. Recommended north bridge chipset in support of this CPU is Intel's G31. This combination is implemented using Gigabyte's GA-G31M-S2L mobo.
cMP 2 (cMP + cPlay) using SRC @ 145db SNR betters previous best setup (Biostar P4M900, E1200, cMP, cPlay 1.0b19). E7200's suitability for audio is superior to previous best (E1200) irrespective of CPU load requirements.
Overall Setup
- PCI card must be installed in slot 1 (closest to CPU) to ensure dedicated interrupt. Slot 2 forces interrupt sharing - avoid this!
- Use high quality RAM - a minimum of 1GB is recommended. My reference setup using Zalman's TNN-300 case uses Kingston's HyperX 1GB twin pack (2x512MB: KHX6400D2UL). Make sure to use both RAM slots as Dual channel RAM acces is enabled. cMP setup gives ~920MB available RAM.
- Expect CPU temperatures ~25 o C with fanless (heat sink only) cooling and CPU fully under-clocked & under-volted. This is room temperature stuff and reflects how much superior 45nm technology is.
- CPU load at 96k is ~60% with CPU fully under clocked (1.2GHz) and FSB at 800Mhz. That is, same as E1200.
- CPU load at 192k will be higher together with CPU clock increasing to 1.9GHz.
GA-G31M-S2L BIOS Setup
Each point below refers to BIOS menu item. Before applying BIOS changes, perform Load Optimized Defaults.
System installation can be done before or after these changes.
- Standard CMOS Features
Disable Drive A (floppy disk drive)
- Advanced BIOS Features
Disable HDD S.M.A.R.T. Capability, No-Execute Memory Protect, CPU Thermal Monitor (TM2).
Make sure CPU EIST Function is Enabled
Set Init Display First to Onboard
Set Onbaord VGA to Always Enable
- Integrated Peripherals
Set On-Chip SATA Mode to Non-Combined
Disable Azalia codec, Onboard LAN Boot ROM, Serial & Parallel ports
Disable Onboard H/W LAN, SMART LAN (if not using networking functionality)
Disable USB 2.0 Controller (if NOT using USB soundcard)
- Power Management Setup
Set ACPI Suspend Type to S1
Disable PME Event Wake-Up, Power On by Ring, Resume by Alarm, HPET Support
- PnP/PCS Configuration
defaults used
- PC Health Status
Disable CPU Smart FAN Control (if no Fan is used)
Set CPU Warning Temperature to 60 o C/140 o F
- MB Intelligent Tweaker (M.I.T.)
Set CPU Clock Ratio to 9
Set fine control to 0.5
Enable CPU Host Clock Control
Set CPU Host Frequency (Mhz) to 200 - this sets FSB to 800Mhz (BIOS should reflect max CPU frequency of 1.9GHz)
Set System Memory Multiplier (SPD) to 2 - this sets RAM speed to 400
Make sure System Voltage Control is Manual and remaining voltages are Normal. Ignore flashing voltage warning.
Perform Save & Exit.
Using latest CPU-Z (1.45) utility, you should get:
CPU at 1.2Ghz, FSB @ 800MHz and Vcc at 1.120V. Only version 1.45 or later will recognize the E7200.
RAM settings.
Allow for at least 24 hours of burn-in.
Latency
Unfortunately, Intel's G31 / ICH7 (North / South chipset) prefers slightly larger payloads. This may be due to its superior bandwidth at 2GBps (that's 16G bits per second!) - VIA chipset offers 1GBps. Extreme low latencies give rise to minor but audible dropouts. Lowest stable latency using RME's HDSP 9652 is 128 samples. This may improve as the mobo burns-in.
Fortunately with cPlay, sound quality is not affected.
- Up and running, but a new problem - rickmcinnis@dogwoodfabrics.com 08:00:32 07/03/08 (3)
In Reply to: RE: Using SRC @ 145db SNR posted by cics on June 22, 2008 at 08:35:07
Dear cics,
As always your guidance is needed.
Got the GIGABYTE/e7200 working last night.
My question is: how important is it to do a WINDOWS re-install? It seems to be doing fine without doing this.
BUT:
I cannot get the MB (I guess) to recognize that I want to use the WINDOWS installation disk. At least in my case, where before the BIOSTAR board would prompt you if you wanted to boot from the DISK, I do not see where this comes up at all!
I set my boot priority to be the USB CD-ROM drive. I know it recognizes the drive since I used it to install the GIGABYTE MB drivers.
As usual, I am missing something. I have looked at the manual to no avail. I do not doubt it is there. What am I doing wrong? I want to install minlogon and disable SFC, and if a new install would be better, I need to go ahead and do that.
What happens when one accesses the WINDOWS disk from MY COMPUTER when you are already in WINDOWS? If I install from here would it retain my HDD partition?
Sorry for all of the questions.
I listened for awhile last night but had set JULI@ latency at 48 which resulted in those blips. Inbetween those blips I could hear much promise. More AIR than ever and the seetest high frequencies, yet. Thuderous, but fully controlled bass. A giant step in resolution and listening pleasure.
Cannot wait to hear it with the BIOS upgrade. Thanks for making it easy for me.
THANKS,
Rick McInnis
- RE: Up and running, but a new problem - cics 09:35:37 07/03/08 (1)
In Reply to: RE: Up and running, but a new problem posted by rickmcinnis@dogwoodfabrics.com on July 03, 2008 at 08:00:32
I prefer doing a reinstall of Windows for new mobos with different chipsets. Otherwise, you'll get unpredictable results.
Try connecting ROM drive using standard IDE connector. Also, in BIOS set SATA Mode to Auto. You can always go back to the USB connection after Windows installation.
- RE: Up and running, but a new problem - rickmcinnis@dogwoodfabrics.com 09:54:05 07/03/08 (0)
In Reply to: RE: Up and running, but a new problem posted by cics on July 03, 2008 at 09:35:37
Dear cics,
Thanks,
I do not have an IDE cd-rom but it is certainly easy enough to get one.
I will give this a try. I will pirate one from one of the machines at work. Don't tell a soul ...
I will let you know.
THANKS,
Rick McInnis
- more proofreading needed - rickmcinnis@dogwoodfabrics.com 08:04:41 07/03/08 (0)
In Reply to: RE: Up and running, but a new problem posted by rickmcinnis@dogwoodfabrics.com on July 03, 2008 at 08:00:32
should have said sweetest high frequenices and THUNDEROUS bass.
Thuderous bass is more like what I got from the THETA.
- RE: Using SRC @ 145db SNR - ninekit 01:32:11 07/02/08 (6)
In Reply to: RE: Using SRC @ 145db SNR posted by cics on June 22, 2008 at 08:35:07
http://www.driverheaven.net/motherboards-networking-misc-forum/158626-gigabyte-p35-dpc-problem-possible-solution.html
DPC latency problem seems common on Gigabyte mobo, I've found the above link, hope it can solve your extreme low latency problem.
Apparently its an BIOS issue.
- Latency problem seems to be solved - cics 10:33:51 07/02/08 (5)
In Reply to: RE: Using SRC @ 145db SNR posted by ninekit on July 02, 2008 at 01:32:11
Interesting read and yes with latest beta Gigabyte BIOS (F5d) , low latency playback is stable. Will test for a while.
DPC latencies show large improvements after BIOS upgrade.
Great post! Thank you.
- RE: Latency problem seems to be solved - ninekit 17:55:19 07/02/08 (2)
In Reply to: RE: Latency problem seems to be solved posted by cics on July 02, 2008 at 10:33:51
Happy to share, just like what you are doing with cMP and cPlay.
May I ask if I just have 512M of RAM, are there any chance for me to enjoy cPlay?
Will it play a .cue pointing to a .flac file small enough to fit my available memory?
- cPlay will work with 512MB RAM. FLAC files are decoded during RAM Load... - cics 00:28:05 07/03/08 (1)
In Reply to: RE: Latency problem seems to be solved posted by ninekit on July 02, 2008 at 17:55:19
thereby eliminating this CPU overhead during playback.
With 512MB RAM, available RAM would be ~420MB, cPlay's diagnostics will show how RAM is needed to load decoded FLAC file (i.e. WAV file). If you have FLAC files by track then playing individual files won't be a problem.
- RE: cPlay will work with 512MB RAM. FLAC files are decoded during RAM Load... - ninekit 00:52:08 07/03/08 (0)
In Reply to: RE: cPlay will work with 512MB RAM. FLAC files are decoded during RAM Load... posted by cics on July 03, 2008 at 00:28:05
This is a very smart way of handling available resources.
I will go check out how to construct .cue pointing to multiple .flac files then.
Thanks
- How does one install this? - rickmcinnis@dogwoodfabrics.com 13:07:37 07/02/08 (1)
In Reply to: RE: Latency problem seems to be solved posted by cics on July 02, 2008 at 10:33:51
Dear cics,
Is this installed like the BIOS drivers? Do I make a CD or load it onto a USB MEMORY stick and , ... everything is obvious, or is there something tricky about it?
I am supposed to get my MB today. I hope my luck has changed and am able to get this running tonight.
THANKS,
Rick McInnis
- Update from BIOS menu using Q-Flash. Instructions are in the manual. - cics 00:21:39 07/03/08 (0)
In Reply to: RE: How does one install this? posted by rickmcinnis@dogwoodfabrics.com on July 02, 2008 at 13:07:37
Copy BIOS file (g31ms2l.f5d, 512KB) to a USB stick then access Q-Flash.
- Any opion aboutthe Gigabyte GA-G33M-S2L ? - frennekamp@netcologne.de 15:05:15 06/29/08 (2)
In Reply to: RE: Using SRC @ 145db SNR posted by cics on June 22, 2008 at 08:35:07
...it has the ICH9-chipset on board...?
- Better to use G31 chipset as it uses a few watts less power - cics 00:56:42 06/30/08 (1)
In Reply to: RE: Any opion aboutthe Gigabyte GA-G33M-S2L ? posted by frennekamp@netcologne.de on June 29, 2008 at 15:05:15
Reducing power consumption is very important - every bit helps.
Intel has done a good job with these chipsets. G31 runs cool - compare this to Qxxx range and you'd get heat stains on your chassis!
- RE: Better to use G31 chipset as it uses a few watts less power - Scuba 05:02:10 02/08/09 (0)
In Reply to: RE: Better to use G31 chipset as it uses a few watts less power posted by cics on June 30, 2008 at 00:56:42
Hello cics,
I am building a dedicated PC for audio playback.
What do you think about Gigabyte Q35M-S2 ? The TDP for Q35 northbridge is just 13W. For the G31, TDP is 15.5 W . I also think that the GA-Q35M-S2 has a superior power supply layout. Are there any inconvenients for the Gigabyte Q35M-S2 ?
About the processor: the E5200 is OK ? E5200- 2.50Ghz, FSB 800Mhz, 2MB Cache, 45nm ;E7200- 2.53Ghz, FSB 1066Mhz, 3MB Cache, 45nm .
Thank you.
- Disable Intel System Devices - cics 09:15:50 06/28/08 (3)
In Reply to: RE: Using SRC @ 145db SNR posted by cics on June 22, 2008 at 08:35:07
System Devices
Specific to Intel's G31 chipset, you can disable these devices:
When SMBus Controller is disabled, System Restart will not work - you can only perform a System Shutdown (and power-up).
- Also disable 'Numeric data processor' - this is a legacy device and can be done on all mobos. - cics 17:54:22 07/05/08 (2)
In Reply to: RE: Disable Intel System Devices posted by cics on June 28, 2008 at 09:15:50
.
- RE: Also disable 'Numeric data processor' - this is a legacy device and can be done on all mobos. - texastea006@yahoo.com 07:26:46 07/29/08 (0)
In Reply to: RE: Also disable 'Numeric data processor' - this is a legacy device and can be done on all mobos. posted by cics on July 05, 2008 at 17:54:22
Thanks for these. I always enjoy disabling what my system doesn't need.
-Mike
- Also disable 'Microcode Update Device' - another legacy device. - cics 02:39:04 07/26/08 (0)
In Reply to: RE: Also disable 'Numeric data processor' - this is a legacy device and can be done on all mobos. posted by cics on July 05, 2008 at 17:54:22
.
- RE: Using SRC @ 145db SNR - gagnem66 10:18:20 06/26/08 (1)
In Reply to: RE: Using SRC @ 145db SNR posted by cics on June 22, 2008 at 08:35:07
Hi cics, thanks for your answer, I will go with the Gigabyte/7200.
I have already bought my RAM : KINGSTON HYPERX 800MHZ DDR2 NO-ECC 2048MO LL (CL4). They came as a 2-pack.
Now do you think it would be best to use only 1 module (less power and noise) or 2 modules (benefit from dual-channel) ?
Thanks again
- Use both RAM modules (for dual-channel) - cics 11:24:08 06/26/08 (0)
In Reply to: RE: Using SRC @ 145db SNR posted by gagnem66 on June 26, 2008 at 10:18:20
Intel chipsets perform best with dual-access.
- Is Low Latency Memory Still Necessary? - rickmcinnis@dogwoodfabrics.com 13:30:56 06/23/08 (3)
In Reply to: RE: Using SRC @ 145db SNR posted by cics on June 22, 2008 at 08:35:07
Dear cics,
With the comments made about the GIGABYTE MB: is low latency in the memory still a consideration?
CONSIDERING that the HYPERX low latency memory seems to no longer be available, I am looking for guidance as to what is next best.
THANKS,
Rick McInnis
- RE: Is Low Latency Memory Still Necessary? - cics 09:54:47 06/24/08 (2)
In Reply to: RE: Is Low Latency Memory Still Necessary? posted by rickmcinnis@dogwoodfabrics.com on June 23, 2008 at 13:30:56
Both low latency and RAM clock rating is important. Although RAM is implemented at lowest speed of 200MHz (with standard latencies), better quality RAM is an important factor for best sound quality.
Try RAM from Corsair, Mushkin or OCZ - all excellent choices. You may need to go for the 2x1GB option.
For Kingston HyperX, try this seller - they seem to have the 1GB twin pack in stock.
- BUT, it is not the Ultra Low Latency - rickmcinnis@dogwoodfabrics.com 10:42:50 06/24/08 (1)
In Reply to: RE: Is Low Latency Memory Still Necessary? posted by cics on June 24, 2008 at 09:54:47
Dear cics,
Thanks for the link. I thought I had checked them but I guess I did not.
The real problem is finding your recommended ULL series from HYPERX. On their website they say it is no longer available.
I went through pages of GOOGLE trying to find someone who has it still in stock. I could find 1gB singles, but nowhere could I find "kits" of any capacity.
What does one lose with this slighly higher latency?
THANKS,
Rick McInnis
- ULL series is not a critical factor. Version as per link also does low latency - its an excellent choice. - cics 12:14:13 06/24/08 (0)
In Reply to: RE: BUT, it is not the Ultra Low Latency posted by rickmcinnis@dogwoodfabrics.com on June 24, 2008 at 10:42:50
.
- RE: Using SRC @ 145db SNR - seger 10:34:59 06/23/08 (4)
In Reply to: RE: Using SRC @ 145db SNR posted by cics on June 22, 2008 at 08:35:07
In my cMP based HTPC if I went for the gigabyte mobo and E7200 do you think my Thermaltake "Dual Power" 520/380W PSU, which I only use fanless so 380W max, will be sufficient for the best SRC upsampler in cPlay?
http://www.thermaltake.com/product/Power/PurePower/w0107_108/w0107w0108.asp
Does the fan in your Coolermaster get on your nerves at all? You once said you would get the MGE Magnum 500W fanless. I never could find it in Europe at all, but I just saw it for sale in the US for $38.99, special promotion:
http://www.clubit.com/product_detail.cfm?itemno=A6224501
No good for me in UK, unfortunately.
- RE: Using SRC @ 145db SNR - cics 12:00:16 06/23/08 (3)
In Reply to: RE: Using SRC @ 145db SNR posted by seger on June 23, 2008 at 10:34:59
In my cMP based HTPC if I went for the gigabyte mobo and E7200 do you think my Thermaltake "Dual Power" 520/380W PSU, which I only use fanless so 380W max, will be sufficient for the best SRC upsampler in cPlay?
Yes. My reference setup uses the Zalman TNN-300. PSU is fanless at 350watts and system has been running 7x24 since Fri 21h00. Gigabyte/E7200 seems to consume even less power doing 24/96 upsampling at 145db SNR. Unfortunately, I don't have a power measuring device.
Does the fan in your Coolermaster get on your nerves at all? You once said you would get the MGE Magnum 500W fanless. I never could find it in Europe at all, but I just saw it for sale in the US for $38.99, special promotion:
I use it in my second cMP which outputs analogue directly to powered speakers. Very quiet fan and its not bothersome. MGE was very good (with a handy power gauge). This post points to the best PSU I've seen to date.
- RE: Using SRC @ 145db SNR - seger 13:20:24 06/23/08 (2)
In Reply to: RE: Using SRC @ 145db SNR posted by cics on June 23, 2008 at 12:00:16
I considered the Fortron, but was put off by the venting - more interference getting into the PC? Also in its first SPCR test it blew up on the bench! And the very high price. The specs are certainly good though.
- The tested unit is no longer available - rickmcinnis@dogwoodfabrics.com 13:36:26 06/23/08 (1)
In Reply to: RE: Using SRC @ 145db SNR posted by seger on June 23, 2008 at 13:20:24
I see NEWEGG has a new fanless 400W model available.
It has received good reviews from those who have bought it. It seems to hol itself together.
http://www.newegg.com/Product/Product.aspx?Item=N82E16817104035
Can handle both 120 amd 240 V input.
Looks ineresting.
Rick McInnis
- RE: The tested unit is no longer available - frennekamp@netcologne.de 15:19:22 06/28/08 (0)
In Reply to: RE: The tested unit is no longer available posted by rickmcinnis@dogwoodfabrics.com on June 23, 2008 at 13:36:26
I can suggest silverstone. Dead-quiet.
I had as well fortron, but these somethime introduce all kind of noise even though they are fanless.
- RE: Using SRC @ 145db SNR - gagnem66 10:23:42 06/23/08 (1)
In Reply to: RE: Using SRC @ 145db SNR posted by cics on June 22, 2008 at 08:35:07
Hi Cics,
I'm about to build a system, and getting a bit confused on which mobo to use. Since my FLAC files don't all have a .cue file (and I have lots of them) I'll still have to use Foobar for a while. Is the Gigabyte board a good choice for me even if it lacks RAID ? or should I stick with a P4M900 / E1200? In short, is a Sata drive on the standard IDE that bad for audio ?
And finally, Is it okay to use CMP/Cplay with CDs ripped to multiple FLACs and cue sheet ? That way I won't have to rip again, only create .cue files.
Thanks
- RE: Using SRC @ 145db SNR - cics 11:48:27 06/23/08 (0)
In Reply to: RE: Using SRC @ 145db SNR posted by gagnem66 on June 23, 2008 at 10:23:42
I'm about to build a system, and getting a bit confused on which mobo to use.
Use the Gigabyte GA-G31M-S2L as suggested. I now have 60+ hours on this setup and its amazing.
Since my FLAC files don't all have a .cue file (and I have lots of them) I'll still have to use Foobar for a while.
You can play flac or wav files directly in cPlay without a cue file. Either right-click file and select Open with... or from cPlay, set the file browser to flac or wav (it defaults to cue).
Is the Gigabyte board a good choice for me even if it lacks RAID ? or should I stick with a P4M900 / E1200? In short, is a Sata drive on the standard IDE that bad for audio ?
Gigabyte/E7200 is better. RAID has no sound quality benefit since cMP (and cPlay improves this). Both cMP and cPlay avoid disk traffic during playback hence no IDE interference.
Is it okay to use CMP/Cplay with CDs ripped to multiple FLACs and cue sheet ? That way I won't have to rip again, only create .cue files.
Yes. This would be using the multi-file cue standard, i.e. a single cue sheet which contains many FILE directives (typically one per TRACK).
- New recommended MB on the way ... and - rickmcinnis@dogwoodfabrics.com 11:13:34 06/22/08 (0)
In Reply to: RE: Using SRC @ 145db SNR posted by cics on June 22, 2008 at 08:35:07
Dear cics,
The memory you recommend, the ultra low latency HYPERX, is no longer available anywhere from my searches.
I just this morning realized this was a very specific recommendation. I had no idea there were so many versions of this available.
I was wondering what your next choice would be?
Please advise at your leisure.
In the interim I will use the 2 x 512 I got, it does specify low latency but is half the speed. What will be the likely affect of this?
Will report as soon as the MB arrives.
THANKS,
Rick McInnis
- Any reason why GIGABYTE board will not accept JULI@? - rickmcinnis@dogwoodfabrics.com 19:18:50 06/20/08 (11)
In Reply to: RE: cPlay - the open source high-end audio player using ASIO posted by cics on May 05, 2008 at 12:31:58
Dear cics,
I finally got the opportunity to install the new board this evening and as soon as I insert JULI@ nothing will happen. I take JULI@ out and all is normal.
Could there be some kind of conflict?
I( have installed it in digital only format. It does seem hard to believe that it would be recognized with the analog part attached but I will try that tomorrow. As usual, I am hoping you know what to do! On re-reading I want to re-emphasize, the monitor screen remains black and the little LED's on the board keep flashing. I wish I knew what these are signifying.
Very disappointing.
The BIOS is quite different. I have done what I think is right but look forward to your guidance with this.
THANKS!!!
Rick McInnis
- On Gigabyte mobo, use PCI slot 1 (closest to CPU) - cics 02:12:52 06/21/08 (10)
In Reply to: RE: Any reason why GIGABYTE board will not accept JULI@? posted by rickmcinnis@dogwoodfabrics.com on June 20, 2008 at 19:18:50
This ensures soundcard gets dedicated interrupt.
Also, you MUST install Juli@ before implementing minlogon (which should be done last).
- by the way ... - rickmcinnis@dogwoodfabrics.com 09:17:24 06/21/08 (8)
In Reply to: RE: On Gigabyte mobo, use PCI slot 1 (closest to CPU) posted by cics on June 21, 2008 at 02:12:52
Dear cics,
Have you started using this MB yet? Just wonddered if you had got one yet.
Also, I am nowhere near MINLOGON. I always make that the last thing I do. I have attempted to go my own way only to find I should have stuck with your instructions. I follow the papers as directed now.
At this point, to repeat, with JULI@ "installed" I get nothing but a blank screen, so no need to get anywhere close to implementing MINLOGON.
- Reset BIOS using 'Load Optimized Defaults' - cics 12:06:39 06/21/08 (7)
In Reply to: RE: by the way ... posted by rickmcinnis@dogwoodfabrics.com on June 21, 2008 at 09:17:24
Then reboot without soundcard.
After Windows starts successfully, only then install soundcard.
I have E7200 and the other Gigabyte mobo (GA-G31M-S2L) - currently burning-in. Newer mobo is not yet available. What I can say now is the E7200 processor is simply outstanding (low power, low heat, fast)! Getting the right mobo is my next step.
- Something is wrong - rickmcinnis@dogwoodfabrics.com 13:23:42 06/21/08 (6)
In Reply to: RE: Reset BIOS using 'Load Optimized Defaults' posted by cics on June 21, 2008 at 12:06:39
Dear cics,
When I could not get ANYTHING to happen (by this I mean the monitor does not display anything, those 4 LED's on the MB near the processor just flash continuously) so I removed the MB battery for 15 minutes (I was doing other things) replaced it, nothing has changed.
I removed the processor and made sure it was absolutely centered in the socket. I removed and replaced the memory (tried both the HYPER X 1gB and the "regular" 2 x 512 mB) nothing gets the machine to do anything.
Last night I had got about half way through "optimizations" before trying again I had originally installed the JULI@ card on the new board and found then that nothing would happen. I then took it out and got BIOS to come up. I then installed WINDOWS (even though without re-installation cMP came right up, though I could not use it since there was no JULI@).
After doing a few pages of AOCT instructions I installed the card again (so many opportunities with all of the re-starts). Nothing, just like before, would not go to BIOS screen, monitor screen is blank, the on/off switch is orange, not green as it would be if it was receiving something. I remove JULI@ and I get the same thing, nothing. And it has stayed this way since.
Do you think the MB has a problem or the processor? I have a very slow fan connected to the CPU FAN header, it comes on. Those four LED's come on and just flash. They say something about CPU phase. Does the CPU have anything to do with the board displaying the BIOS stuff?
I figure something has failed. What do you think?
To repeat, before this final crack up, the machine turned on and off, at least, eight times with no glitches.
Very disappointing.
I dread having to deal with returning a dead whatever it is. I hope you can tell me where I am going wrong.
Thanks,
Rick McInnis
- Try another CPU (E2140). If same problem then mobo is faulty. - cics 14:39:24 06/21/08 (5)
In Reply to: RE: Something is wrong posted by rickmcinnis@dogwoodfabrics.com on June 21, 2008 at 13:23:42
This could also be a BIOS issue - try contacting Gigabyte support.
From what you explained, the mobo is the problem.
- I read your mind - rickmcinnis@dogwoodfabrics.com 15:45:15 06/21/08 (4)
In Reply to: RE: Try another CPU (E2140). If same problem then mobo is faulty. posted by cics on June 21, 2008 at 14:39:24
Dear cics,
I came downstairs to tell you I had substituted the e1200 (I hope that is just as valid as the e2140, which I am using in the computer I am using to write this note).
Nothing changes, I have to believe it is definitely the MB.
I do not understand what you mean by a BIOS issue.
Do you think the MB is OK just some kind of software problem? So, it is not a matter of just getting another board from NEWEGG?
I am sorry to be so pathetic but if I need to contact GIGABYTE what do I tell them? Please give me specifics, you know I am a computer dilettante!
Do I tell them there is a conflict with JULI@? Or just tell them what has happened?
AS ALWAYS, thanks for your patience and help,
Rick McInnis
- That's a faulty mobo - NewEgg should replace it. Get Gigabuyte GA-G31M-S2L mobo instead - cics 06:47:39 06/22/08 (3)
In Reply to: RE: I read your mind posted by rickmcinnis@dogwoodfabrics.com on June 21, 2008 at 15:45:15
I'll do a separate post on how to setup the GA-G31 mobo. After some 30+ hours of burn-in, sound quality is very good and betters previous (E1200, Biostar mobo, 24/96@121SNR).
New setup allows for 24/96@145 SNR and does so with less power consumption! E7200 remains underclocked (same as E1200, ie. FSB 800MHz & CPU 1200MHz) but with lower Vcc @ 1.120V. This is amazing considering the heavy CPU performance needed for SRC 245db SNR upsampler.
- New MB is faulty also - rickmcinnis@dogwoodfabrics.com 19:14:21 06/25/08 (2)
In Reply to: RE: That's a faulty mobo - NewEgg should replace it. Get Gigabuyte GA-G31M-S2L mobo instead posted by cics on June 22, 2008 at 06:47:39
Dear cics,
I am starting to think I will never listen to music again.
Got the recommended MB today. Installed it AND NOTHING at all.
Changed out memory. Did nothing.
Installed e1200. Still nothing.
Reinstalled e1200 into BIOSTAR board, comes right up.
Are these GIGABYTE boards EXTREMELY fragile? I have handled both boards carefully and yet I have had two consecutive bad boards. I am very sad about this.
NEWEGG will still charge me a re-stocking fee on a BAD board, so this time I will have to wait for a replacement. I would prefer to go ahead and order again. I fear once I finally get the thing running again I will have lost all reference points. Though I have no fear it will sound very good.
Has anyone else had MB problems like this?
Can you think of something I am doing wrong? Up to now all has performed without a problem. I would think my assembly technique has only improved by this point. Oh, well.
AS always,
THANKS!!!
Rick McInnis
- Sounds like the power switch connector is incorrectly installed - cics 00:57:22 06/26/08 (1)
In Reply to: RE: New MB is faulty also posted by rickmcinnis@dogwoodfabrics.com on June 25, 2008 at 19:14:21
Only connect the power switch wire - leave out HDD LED, Power LED, Reset connectors.
Make sure to connect to correct pins.
- I know the MB is getting power (unfortunately) - rickmcinnis@dogwoodfabrics.com 07:50:43 06/26/08 (0)
In Reply to: RE: Sounds like the power switch connector is incorrectly installed posted by cics on June 26, 2008 at 00:57:22
Dear cics,
I have a very low speed fan I use in the summer. I know it is not ideal (one of these days I will power it separately) but at the moment I figure it tells me that power is getting to the board. I keep my house at 80 degrees, and sometimes I will leave the computer on.
When I "turn it on" this fan starts spinning. But, that is all that happens.
My other board has these surface mount LEDs that would flash. I have no idea if this board has those, but there is nothing flashing when this MB is given power.
I looked on the NEWEGG site and it seems there are others who have received this board and found theirs defective, also.
I wish there was something I was doing wrong but I think it is just a bad board. Very frustrating.
Thanks,
Rick McInnis
- I had tried both slots - rickmcinnis@dogwoodfabrics.com 09:12:23 06/21/08 (0)
In Reply to: RE: On Gigabyte mobo, use PCI slot 1 (closest to CPU) posted by cics on June 21, 2008 at 02:12:52
Dear cics,
This has been mystifying.
I had left it last night in frustration, now even without JULI@ inserted it is doing the same thing, which is nothing.
I will try booting with WINDOWS disk.
Before inserting JULI@ initially I had made some BIOS changes. I very well could have done something that is causing my problem.
I had tried both slots, and this made no difference though I do realize you are recommending the near slot for other reasons.
I will re-set BIOS, insert JULI@ into the CORRECT slot and see what happens!
As you well know, I will keep you posted and pestered with my need for assistance.
I am very anxious to hear what this can do.
THANKS!
Rick McInnis
- cPlay 1.0b19 Released - cics 08:11:44 06/19/08 (2)
In Reply to: RE: cPlay - the open source high-end audio player using ASIO posted by cics on May 05, 2008 at 12:31:58
Change Log (final beta release):
- Tuning changes
Please REMOVE previous versions before installing cPlay 1.0b19 .
- RE: cPlay 1.0b19 Released - appleteapot 00:06:17 06/23/08 (1)
In Reply to: RE: cPlay 1.0b19 Released posted by cics on June 19, 2008 at 08:11:44
Hi cics,
After 1 day of listening, the embedded CMP takes cplay resolution to a new level, however, bass seems a little bit of booming and not as tight as B17.
I am not sure whether it is caused by improvement of efficiency or just my computer's issue? I use a work horse XP to test b19, w/o AOBCT.
-------------------------------------
cicsPlay 1.0b19, licensed under GNU GPL
ASIO Technology by Steinberg Media Technologies GmbH
ASIO is a trademark and software of Steinberg Media Technologies GmbH
ASIO Driver Initializing...
Name (hdspwdm)
Player ASIO version (2)
Driver ASIO version (1)
Message (No ASIO Driver Error)
Channels (inputs: 8, outputs: 8)
Buffer details (min: 64, max: 64, preferred: 64, granularity: 0)
Sample rate ( 96000)
Output Ready? Supported
Preparing buffers... successful
Latencies (input: 109, output: 222)
Hardware supports additional optimization - enabled
ASIO Driver initialized.
DSP starting...
DSP Buffer (12288 samples or 128ms/ 7.8/ 1500.0) allocation successful
ASIO Driver starting...
Input 1 [ HDSP9632 Analog 1 ] Active (1) type (32bits)
Input 2 [ HDSP9632 Analog 2 ] Active (1) type (32bits)
Input 3 [ HDSP9632 SPDIF L ] Active (1) type (32bits)
Input 4 [ HDSP9632 SPDIF R ] Active (1) type (32bits)
Input 5 [ HDSP9632 ADAT 1 ] Active (1) type (32bits)
Input 6 [ HDSP9632 ADAT 2 ] Active (1) type (32bits)
Input 7 [ HDSP9632 ADAT 3 ] Active (1) type (32bits)
Input 8 [ HDSP9632 ADAT 4 ] Active (1) type (32bits)
Output 1 [ HDSP9632 Analog 1 ] Active (1) type (32bits)
Output 2 [ HDSP9632 Analog 2 ] Active (1) type (32bits)
Output 3 [ HDSP9632 SPDIF L ] Active (1) type (32bits) > Left Channel
Output 4 [ HDSP9632 SPDIF R ] Active (1) type (32bits) > Right Channel
Output 5 [ HDSP9632 ADAT 1 ] Active (1) type (32bits)
Output 6 [ HDSP9632 ADAT 2 ] Active (1) type (32bits)
Output 7 [ HDSP9632 ADAT 3 ] Active (1) type (32bits)
Output 8 [ HDSP9632 ADAT 4 ] Active (1) type (32bits)
ASIO started successfully.
AUDIO source [ G:\Master\Music Clips\Wave\Porco Rosso\Porco Rosso Soundtrack-15-Adoria no Umi he [To the Adriatic Sea].wav ]
Genre = Unknown
Artist = G:\Master\Music Clips\Wave\Porco Rosso\Porco Rosso Soundtrack-15-Adoria no Umi he [To the Adriatic Sea].wav
Album = WAV
FILE: G:\Master\Music Clips\Wave\Porco Rosso\Porco Rosso Soundtrack-15-Adoria no Umi he [To the Adriatic Sea].wav
Wav file size = 18MB
Rate = 16/44100 (2 channels)
Samples = 4859820 ( 1:50)
AWE sucessful
RAM allocation (19MB) successful
DSP Started.
Play starts after RAM load (available 1003MB, system 2047MB).
Volume ( 0.0db, phase 0)
SRC (0, 96000)
- B19's performance - cics 11:32:07 06/23/08 (0)
In Reply to: RE: cPlay 1.0b19 Released posted by appleteapot on June 23, 2008 at 00:06:17
B19 is a new milestone! In my setup there's almost no jitter artifacts (unnatural highs, sloppy/boomy bass, masking/veiling, artificial bloom).
B18 uses a much smaller L2 cache 'footprint' which is exploited in b19. Changes in b19 goes back to my original design goals - I guess it works now given all the efficiencies gained.
Every transport has a jitter profile (with the worst being CD spinners). With computers, the 'default' environment is a source of jitter - cMP effectively addresses this and more. In a cMP setup, the audio player becomes the largest jitter source. This jitter is induced by digital logic. B19 excels here.
I would recommend doing as much of cMP's implementation in your setup. Using cMP's software (which runs on any XP computer) with Optimize set to Critical is very important.
Also, set your volume to -0.5db and not 0.0db as this allows for interpolation to occur where sound is at 0dbFS. Otherwise, clipping is done. There's far too many recordings with sound peaks regularly at 0dbFS or worse, clipped.
- cPlay 1.0b18 Released - cics 23:31:07 06/17/08 (0)
In Reply to: RE: cPlay - the open source high-end audio player using ASIO posted by cics on May 05, 2008 at 12:31:58
Change Log (final beta release):
- Settings caters for all options (Upsampler, AWE & cMP)
- Tuning changes due to SRC at 145db SNR and b17 efficiencies
Please REMOVE previous versions before installing cPlay 1.0b18 .
Details and setup of AWE is available in cPlay 1.0b17's release notes . There's no need to manually change .ini file with b18.
- Can you give us the magic of b9 back ? - frennekamp@netcologne.de 12:04:46 06/16/08 (7)
In Reply to: RE: cPlay - the open source high-end audio player using ASIO posted by cics on May 05, 2008 at 12:31:58
I have checked it back and forth...i checked this with b12 and b17: B9 is the best, most magical version so far in my eyes. It adds a kind of 3D-holographic effect and a "it's simply there-clarity" and micro-dynamics, which are lost in later releases.
I do not use any upsamling in the PC, this as well destroys micro-dynamics and makes everything more flat, less 3D. My DAC (RAKK-DAC with passive-Output-Stage) does the upsampling (but could as well accept up to 192khz as inpput).
My system is quiet high-resolution (Soulution 720, Pass F4 in bridged mode with cost-no-object-DIY-components in there and a Line-Source with Ribbons and Eton-Chassis)...and I highly appreciate your effort, this is the byfar best sounding player, compared it again iriver (ver high resolution, but a bit mechanical sounding; Foobar with KS, a bit less mechancal, but still not as smooth and 3D, Album-PLayer: I love the touch-screen-GUI, wonderful; but I don't get the Asio-Option to work, so it sounds low-resoluted).
- And to think I thought #13 was the magical one! - rickmcinnis@dogwoodfabrics.com 12:51:15 06/16/08 (6)
In Reply to: RE: Can you give us the magic of b9 back ? posted by frennekamp@netcologne.de on June 16, 2008 at 12:04:46
But, after many more hours of listening and perhaps becoming acclimated I am satisfied that cics ear's have made the correct choice.
Of ocurse, taking a systems approach it could well be that #9's strenghts (and weaknesses) make for the mythical synergy with your set-up which certainly sounds formindable.
I. too, am using the RAKK DAC #2, in my case, but with Davenport's tube line stage instead of the transformer set up. I am using Edgar horns so my set up prefers a more laid back source, need TAD 4001's no help with transients, some might say they CREATE transients.
I, at first, missed something about #13, which may have been akin to #9, there was a greater sense of dimension, but I have come to believe this was not what is on the disk, in fact, there seemed to be a bit of a sparkle in the background, like disk noise on LP's exagerating a sense of space, in my opinion.
I would say stick with #17 for a week and then go back and see what you think.
On second thought you did not specify how faithfully you have followed the protocols for your music computer's et-up. I bet, without dismantling all of the unneeded services could effect how well the upsampler works and sounds. I am using the recommended JULI@ card and upsampling to 192K before sending to the RAKK DAC. I have been dazzled by the sound I am getting. I would have never thought I would be listening to this quality of digital sound. Figured that quality was way over my head, money wise.
Excuse me if my suspicions are wrong, but if you have not followed the instructions to the letter you are not hearing what cics has created.
Cordially,
Rick McInnis
Grateful listener
- RE: And to think I thought #13 was the magical one! - frennekamp@netcologne.de 15:35:12 06/16/08 (5)
In Reply to: RE: And to think I thought #13 was the magical one! posted by rickmcinnis@dogwoodfabrics.com on June 16, 2008 at 12:51:15
Yup, you are wrong...this is a PC build and configured exavtly using Cics advise...as well with the Juli-Card.
Any yes, you ra re right, I am talking about this extra dimension, which makes it magical (B9)...in my setup, it is extremely natural and simply helps your brain to accept this realismn of sound more easily. The newer versions sound more like a reproduction to me...but I am glad, that I am not the only one realizing that there is this difference.
- RE: And to think I thought #13 was the magical one! - appleteapot 18:54:05 06/16/08 (4)
In Reply to: RE: And to think I thought #13 was the magical one! posted by frennekamp@netcologne.de on June 16, 2008 at 15:35:12
I think if CICS can let us change the .ini file or UI to customize the different settings between b9 to b17 then it would be great. I look forward to study the source code once the final version has released, and maybe we can help comment as well.Personally my best experience come from cics AOBCT 0.3 + cMP 1.0 + cplay b17. When I use RME 9632 -> Benchmark -> Beyer DT990 Headphone, the difference between b17 and b13 can be distinguished, and B17 has the best resolution among all versions. Good resolution what I think a player's responsible for. For alternating the "taste" of the sound I think it is best to allow us fine tune it ourselves. Frankly speaking for playback of male voice, I prefer b13/b9, but for symphonies or complex instruments b17 is still the best.
And I just guess that if your back end equipment has such high resolution, it might be worthwhile to use a "good" sound card, e.g. RME or Lynx or above. Working at 96Khz-192Khz is CPU intensive and alot of electric noise is generated that will impact sound quality.
Edits: 06/16/08 06/16/08 06/16/08
- RE: And to think I thought #13 was the magical one! - artcheng@netvigator.com 20:11:19 06/17/08 (1)
In Reply to: RE: And to think I thought #13 was the magical one! posted by appleteapot on June 16, 2008 at 18:54:05
One way to solve the upsampling CPU load is to do the upsampling when loading the wav file to RAM disc. But the amount of RAM has to be sufficient to hold the upsampled file.
- RE: And to think I thought #13 was the magical one! - appleteapot 21:40:53 06/17/08 (0)
In Reply to: RE: And to think I thought #13 was the magical one! posted by artcheng@netvigator.com on June 17, 2008 at 20:11:19
Hi ackcheng,In the latest version B17, AWE is used to lock the RAM, and latest version of SRC is used. When I run b17, SRC at 96Khz, CPU usage is at about 40-44%. There is no clicks and pops. In comparison, old pre b17 SRC at 96Khz consumes about 15% CPU. My PC is currently running at 900Mhz, RME hardware latency at 64.
------------------------AWE sucessful
RAM allocation (44MB) successful
DSP Started.
Play starts after RAM load (available 1441MB, system 2047MB).
Volume ( 0.0db, phase 0)
SRC (0, 96000)
Edits: 06/17/08
- Excuse my incorrect assumptions - rickmcinnis@dogwoodfabrics.com 19:54:28 06/16/08 (1)
In Reply to: RE: And to think I thought #13 was the magical one! posted by appleteapot on June 16, 2008 at 18:54:05
Dear Frennekamp,
I am pleased to hear you have followed the program. Please excuse my jumping to conclusions.
I, too, to repeat myself, thought I was missing something with #17. Maybe my speakers weaknesses are similar to cics's and we are both hearing systems that work better with #17. I must admit it just as easily could be that I am so enamored with cics and his project that when he says THIS IS BETTER I believe him and hear it as such. I like to think I am an unbiased listener but I do wonder about my ability to be SURE of what I am hearing at times.
I find myself getting so wrapped up in the performance that I forget that I am supposed to listening for differences and I end up simply enjoying music. Something to be said for that, too.
With these comments I defer to both your and APPLETEAPOT's assessments.
cics is simply one of the most generous men I have ever had the pleasure of knowing; though only within this very formal situation. He has been extremely helpful. He has important responsibilities and yet, luckily for us, has a love for music as his preferred leisure activity, along with the beautiful little girl whose picture has accompanied the AOCT instructions and the software. I would suspect he would attempt to supply what it is you would like to help him further assess how to finish cPLAY.
In the meantime I will gladly follow where he leads me with much gratitude. I trust those ears.
Thanks to all who have taken this project seriously and have made constructive criticism and most importantly to cics,
Rick McInnis
- I am completely with you - frennekamp@netcologne.de 06:43:18 06/17/08 (0)
In Reply to: RE: Excuse my incorrect assumptions posted by rickmcinnis@dogwoodfabrics.com on June 16, 2008 at 19:54:28
I am completely you...what Cics does for us is absolutely WONDERFUL. I was really thinking to get the big DCS-GEar, but I wanted really a modern system with all the convenience possible. I am as well totally impressed how small DACs like the RAKK can sound like big gear (mine beat an Accuphase DC-801, which I own as well) based on CICS's developments.
It would be indeed wonderfull if we can have a switch between the two versions, because I really hope that Cics continues to deveop the player and I want to follow the development path, especially when he starts to improves the GUI...meant as a wish, not a complaint.
The ideal player would offer a GUI like Album-PLayer fitting to a touch-screen.
- cPlay 1.0b17 Released - cics 10:38:25 06/10/08 (40)
In Reply to: RE: cPlay - the open source high-end audio player using ASIO posted by cics on May 05, 2008 at 12:31:58
Change Log (final beta release):
- Implement further efficiencies in DSP
- Physical RAM allocation method used (for WAV RAM LOAD). Activated when 'AWE' flag is in cicsPlay.ini file. This advanced technique requires LOCK privilege setting! AWE stands for Address Windowing Extensions - allocation occurs directly (system available RAM drops immediately). Windows may NOT allocate all 'Available RAM' - in which case, cPlay reverts to standard approach
- Advanced DSP buffer management implemented
Please REMOVE previous versions before installing cPlay 1.0b17 .
New buffer management favors lowest stable ASIO latency settings. If latency is too high a warning is given in Diagnostics (just before DSP buffer allocation).
Using AWE
To enable AWE feature in cPlay, you must add the line "AWE" into the "C:\Program Files\cics Play\cicsPlay.ini" file. Next time cPlay is used, RAM will be allocated using AWE. Example:
ASIO=ASIO4ALL v2
ASIOCHANNEL=0
SRC=1
RATE= 48000.000
VOLUME=-14.0
PHASE=0
PLAYBACK=S
DIAGNOSTICS=Yes
FILTER=1
AWE
Diagnostics will show whether AWE was successful at RAM allocation. Process Explorer or Task Manager will not show RAM allocated to cPlay's working storage, instead you'll see the reduction in Available RAM.
Make sure you have LOCK privileges set on your computer. Windows will not offer all "Available RAM" for AWE - in which case cPlay reverts to standard allocation. Adding additional RAM will prevent this. AWE offers the potential to load up to 4GB of RAM - not tested.
How 2 set LOCK privilege
Perform these steps exactly:
- Start Menu > Run > enter "mmc" > OK button
- File > Add/Remove Snap-in > Add button
- Select "Group Policy Object Editor" > Add button
- Finish button
- Close button
- OK button
- Local Computer Policy > Computer Configuration > Windows Settings > Security Settings > Local Policies > select "User Rights Assignment"
- Double-click "Lock pages in memory"
- Add User or Group button
- Object Types... button
- check Groups > uncheck Built-in security principals & Users > OK button
- Enter "Administrators" under "Enter the object names to select" > OK button
- Apply button > OK button
- File > Save
- File > Exit
- Reboot
Lock privilege is now enabled. If this is not done, cPlay will fail on AWE and revert to standard RAM allocation.
SRC @145.68db SNR
In above example cicsPlay.ini file, setting 'SRC=1' to 'SRC=0' informs cPlay to use the CPU intensive 145.68db SNR upsampler. Diagnostics will reflect 'SRC(0, rate)' to show converter being used. A word of caution: do NOT attempt to use this on lightweight processors (anything less than E4xxx Core 2 Duo processor). A minimum of 2MB L2 cache is required. For example, the E2140 (1MB L2 cache) and Pentium 4 3GHz (1MB L2 cache) can only accomplish 44.1-> 48 - anything beyond this results in a locked/slow computer. There's a tradeoff here: more CPU power means more electrical inteference (and increased power consumption) thus reducing benefits of superior upsampling.
E6300 (2MB L2 cache, 1.86GHz) processor just manages to cope with 192k upsampling where CPU load is at ~93% (individual CPU core). There's no chance for undervolting / underclocking. RAM must operate at 266MHz (make sure FSB operates at 1067MHz). cMP using this setup operates without CPU fan giving CPU temperatures of ~65 celcius - this is adequate (system halts at 85 celcius). In cMP, no dropouts or clicks occur when Optimize is set to Critical. Upsampling to 96k may allow for undervolting / underclocking. Intel's new E7200 (based on 45nm technology) may be an excellent choice offering low power consumption and brutal performance.
Only try this with 1.0b17 due to its advanced processing methods, CPU efficiency and reduced Windows Kernel overheads.
Keyboard Actions
- '[' / ']' for Previous / Next
- ';' for phase 0/180
- '.' for stop
- shift+Home for volume up
- shift+End for volume down
- Enter to play highlighted track
- Alt+F4 or Esc for exit
- RE: cPlay 1.0b17 Released - audioAl 06:01:39 06/14/08 (0)
In Reply to: RE: cPlay 1.0b17 Released posted by cics on June 10, 2008 at 10:38:25
Great Info,Thanks
Now I Understand
- RE: cPlay 1.0b17 Released - appleteapot 08:23:13 06/11/08 (3)
In Reply to: RE: cPlay 1.0b17 Released posted by cics on June 10, 2008 at 10:38:25
Hi Cics,
I find b17 has better base extension and resolution compare to b16. And you are correct that AWB consumes too much CPU. I have 2G Ram, E6320 on Asus Commando, u/c to 840MHz (140x6). This is a working PC without "Art of Building Computer Transport 0.3" XP optimization. In order to run AWE and Upsample to 192Khz, I need at least 1.5GHz. However, 1.5Ghz has negative impact to sound stage.
Now I stay at b17, 840Mhz and upsample to 96Khz. I don't know why but I "feel" that I need to compare b13 and b17 again later.
- B17's performance - cics 12:36:38 06/11/08 (2)
In Reply to: RE: cPlay 1.0b17 Released posted by appleteapot on June 11, 2008 at 08:23:13
Yes, overall resolution improves. Setting phase 0/180 is very important - polarity errors in recordings is surprisingly common.
In my reference setup, b17 using SRC@121db SNR, 96k and AWE gives:
- more detail (which I didn't think was possible!)
- cleanest (tight and articulate) bass so far
- mids have more definition and dynamics
- cleaner highs
- vocals have amazing presence
- excellent tonal decay ☺
In my second setup I have SRC@145db SNR, AWE and 192k (analogue outputs directly to powered speakers). All above comes through and slightly more engaging and natural midrange.
I'm still learning B17's performance using tracks I'm familiar with. To hear new detail together with overall naturalness is hugely enjoyable. Nothing I've tried in the past could do this.
- RE: B17's performance - appleteapot 16:15:45 06/11/08 (1)
In Reply to: RE: B17's performance posted by cics on June 11, 2008 at 12:36:38
Hi Cics,
You are right again... I just re-install b13 and I hear the difference immediately. B17 excels in clarity, and in b13 the punch of music seems being chopped.
However the CPU consumption is still a consideration. With AWE, cplay takes up > 50% CPU time from time to time and the occasional clicks/pops return :( I suspect that CPU at 6MB cache will be able the compensate the reduced clock speed, i.e. use 8400@lower speed rather than 7200@higher speed.
Lower clock speed is critical at my motherboard (maybe Commando motherboard is too noisy) and I can easily hear the difference between 840Mhz, 1.2Ghz and 1.5Ghz
- With cMP's sw (optimize critcal), you'll get no clicks or dropouts. - cics 07:42:46 06/12/08 (0)
In Reply to: RE: B17's performance posted by appleteapot on June 11, 2008 at 16:15:45
You can run cMP on your setup in XP mode.
One of my CPU cores is consistently at ~93% at 192k output and not a single click when using cMP! Without cMP, there's no chance.
- RE: cPlay 1.0b17 Released - theob 02:55:54 06/11/08 (9)
In Reply to: RE: cPlay 1.0b17 Released posted by cics on June 10, 2008 at 10:38:25
I don't have the option for selecting the 'Group Policy Object Editor'. There are other selections includingnet framework 1.1 configuration
activex control
certificates
component services
computer management
device manager
disc fragmenter
disc management
event viewer
folder
indexing service
ip security monitor
ip security policy management
link to webb address
local users and groups
performance logs and alerts
removable storage management
services
shared folder
wmi controlShould I select any of these?
Edits: 06/11/08 06/11/08 06/11/08 06/11/08
- What version of windows is this? - cics 07:46:40 06/11/08 (8)
In Reply to: RE: cPlay 1.0b17 Released posted by theob on June 11, 2008 at 02:55:54
On XP Pro SP2, the option is available.
Try "local users and groups".
- RE: What version of windows is this? - theob 08:08:37 06/11/08 (7)
In Reply to: RE: What version of windows is this? posted by cics on June 11, 2008 at 07:46:40
home edition version 2002 xp service pack 2
Edits: 06/11/08
- RE: What version of windows is this? - theob 10:17:26 06/11/08 (6)
In Reply to: RE: What version of windows is this? posted by theob on June 11, 2008 at 08:08:37
under local users and groups there is a 'allow computer to be changed from command line' should I check this?
- Just checked XP Pro: Local User & Groups won't set LOCK privilege. - cics 11:44:55 06/11/08 (5)
In Reply to: RE: What version of windows is this? posted by theob on June 11, 2008 at 10:17:26
It seems "Group Policy Object Editor" is the only way to get LOCK privilege. No problems with XP Pro SP2.
B17's other improvements will work without AWE, so its not critical. I prefer AWE as it removes some kernel overheads.
- RE: Just checked XP Pro: Local User & Groups won't set LOCK privilege. - theob 07:29:10 06/12/08 (3)
In Reply to: RE: Just checked XP Pro: Local User & Groups won't set LOCK privilege. posted by cics on June 11, 2008 at 11:44:55
If I install xp sp 3 will that give me the capability to get Lock Privileges?
- Should do. Make sure its XP SP3 Professional. - cics 07:35:37 06/12/08 (2)
In Reply to: RE: Just checked XP Pro: Local User & Groups won't set LOCK privilege. posted by theob on June 12, 2008 at 07:29:10
.
- RE: Should do. Make sure its XP SP3 Professional. - theob 09:07:48 06/12/08 (1)
In Reply to: RE: Should do. Make sure its XP SP3 Professional. posted by cics on June 12, 2008 at 07:35:37
is this it?
Windows XP Service Pack 3 Network Installation Package for IT Professionals and Developers
- RE: Should do. Make sure its XP SP3 Professional. - cics 16:38:31 06/12/08 (0)
In Reply to: RE: Should do. Make sure its XP SP3 Professional. posted by theob on June 12, 2008 at 09:07:48
No. Newegg offers XP Pro SP3 for $140. I use SP2 but have received good feedback from SP3 users.
- RE: Just checked XP Pro: Local User & Groups won't set LOCK privilege. - theob 11:51:03 06/11/08 (0)
In Reply to: RE: Just checked XP Pro: Local User & Groups won't set LOCK privilege. posted by cics on June 11, 2008 at 11:44:55
thanks for checking!
- If 2gB of memory is needed ... - rickmcinnis@dogwoodfabrics.com 12:54:40 06/10/08 (0)
In Reply to: RE: cPlay 1.0b17 Released posted by cics on June 10, 2008 at 10:38:25
Dear cics,
Sould we get a pair of 1gB cards or one 2gB card?
Thanks (yet) again,
Rick McInnis
- Since I am incapable of reading between the lines!!! - rickmcinnis@dogwoodfabrics.com 12:43:14 06/10/08 (21)
In Reply to: RE: cPlay 1.0b17 Released posted by cics on June 10, 2008 at 10:38:25
Dear cics,
So what is the recommended set-up?
Should I order the e7200 since I would like to continue upsampling to 192k with the JULI@? Do I need to get another 1G memory card in addition?
This all seems quite exciting but slightly puzzling to my dilletante's ears.
Please give us direction!
Thanks, as always,
Rick McInnis
yr obd svt
- E7200 is ideal when using SRC@145db SNR, no need for additional RAM - cics 02:08:44 06/11/08 (20)
In Reply to: RE: Since I am incapable of reading between the lines!!! posted by rickmcinnis@dogwoodfabrics.com on June 10, 2008 at 12:43:14
I need confirmation from Biostar whether E7200 is supported on the P4-M900 SE mobo. There may be a need for a BIOS update. One of my cMP's is running the E6300 perfectly (at FSB 1067) and I'm hoping there's no need for BIOS update.
ITO recommendation: E7200 is like the E1200 but on steroids. That is, for roughly the same amount of CPU electrical interference you get significantly more computational power! This is perfect for the ultra high quality upsampler. Look at this E7200 review (especially the last sentence). I plan to get the E7200 for my reference setup.
- E7200 is NOT supported on Biostar P4M900 mobos. Need another VIA based mobo. Anyone with suggestions? - cics 02:41:02 06/11/08 (19)
In Reply to: RE: E7200 is ideal when using SRC@145db SNR, no need for additional RAM posted by cics on June 11, 2008 at 02:08:44
Just received feedback from Biostar tech support - they cannot support the E7200 45nm processors.
- what does 45nm mean? & other comments - rickmcinnis@dogwoodfabrics.com 08:06:20 06/11/08 (2)
In Reply to: RE: E7200 is NOT supported on Biostar P4M900 mobos. Need another VIA based mobo. Anyone with suggestions? posted by cics on June 11, 2008 at 02:41:02
Dear cics,
Installed #17 and the WINDOWS refinements. Intially had some trouble getting the JULI@ to retain the lowest latency setting. Eventually it did but I have no idea why?
This one has solid lateral imaging, but I think I am detecting less front to back than with the best of the previous iterations, that airiness does not seem to be here in full measure. Of course, it is/was so euphonic it very well could be an artifact as opposed to true information retrieval. What do you think?
Had no problem with AWS (a first!). Your diagnostics are starting to make sense to me now. They are very helpful once one realizes their importance.
SO, what does 45 nm mean? Since I doubt I will be of much help in finding a suitable MB for the E7200 it is purely to satisfy curiosity. I know you like the VIA chipsets, I presume since you like the ease of implementing RAID; might this have to be dispensed with to use the E7200?
What does one look for for the best choice in MB? Maybe someone out here can be of help.
Along with the extraordinary SNR which I would think would allow low level details to be revealed, is there anything else to be gained with the "other" upsampler?
Thanks, at least one thousand times and then once more just to be safe,
Rick McInnis
- RE: what does 45nm mean? & other comments - cics 12:00:01 06/11/08 (1)
In Reply to: RE: what does 45nm mean? & other comments posted by rickmcinnis@dogwoodfabrics.com on June 11, 2008 at 08:06:20
Intel 45nm technology is a useful site for understanding the importance of 45nm. VIA chipsets have very low power consumption. This is important.
There should be no issues whatsoever with Juli@ at 48 samples (lowest latency) - the problem you had is strange!
Upsampling: At 121db SNR, SRC is brilliant. At 145db SNR, SRC is achieving extreme levels of perfection. My experience suggests a better midrange (in my 2nd setup).
- What do you think of these MB's - rickmcinnis@dogwoodfabrics.com 15:04:47 06/11/08 (0)
In Reply to: RE: what does 45nm mean? & other comments posted by cics on June 11, 2008 at 12:00:01
Dear cics,
Looking around for compatible MBs with VIA chipset I find:
ASUS P5VD2-VM SE LGA 775 VIA P4M900 Micro ATX Intel Motherboard
http://www.newegg.com/Product/Product.aspx?Item=N82E16813131199
It's cheap! I see the chip review article was using an ASUS MB.
Trying to help (ineptly, I'm sure)
Thanks,
Rick McInnis
- RE: E7200 is NOT supported on Biostar P4M900 mobos. Need another VIA based mobo. Anyone with suggestions? - artcheng@netvigator.com 04:02:04 06/11/08 (15)
In Reply to: RE: E7200 is NOT supported on Biostar P4M900 mobos. Need another VIA based mobo. Anyone with suggestions? posted by cics on June 11, 2008 at 02:41:02
When you found a suitable MB, please let us know too!
- Mobos of interest - cics 03:21:39 06/12/08 (14)
In Reply to: RE: E7200 is NOT supported on Biostar P4M900 mobos. Need another VIA based mobo. Anyone with suggestions? posted by artcheng@netvigator.com on June 11, 2008 at 04:02:04
Northbridge chipset to go for is Intel's G31 (offering lower power consumption and plenty performance).
Gigabyte GA-EG31M-S2 may even better Biostar P4M900-M7 SE mobo. Another alternative may be Gigabyte GA-G31M-S2L - nice reviews at Newegg.
GA-EG31M-S2 on paper seems better than GA-G31M-S2L offering better power supply to CPU. Your thoughts?
- Should I give it a try? - rickmcinnis@dogwoodfabrics.com 12:17:14 06/12/08 (13)
In Reply to: RE: Mobos of interest posted by cics on June 12, 2008 at 03:21:39
The MB is not terribly expensive, easily a justifiable expense in the cause of better music reproduction.
I guess I want reassurance that the e7200 is the way to go. No need for the e8200 which was suggested by another enthusiast?
I am ready to give this a try.
Awaiting your instructions!
Thanks,
Rick McInnis
- YES with E7200 - cics 16:45:30 06/12/08 (12)
In Reply to: RE: Should I give it a try? posted by rickmcinnis@dogwoodfabrics.com on June 12, 2008 at 12:17:14
BIOS will have slight differences.
A new Windows installation is necessary.
For SRC@145db SNR and 192k output, this setup would support underclocking and undervolting. It's also a very good idea to add a second RAM module - to enable dual channel access, i.e. get another exact RAM module as you currently have (giving 2GB RAM). Allow for good burn-in time. What PSU are you using?
- Have you decided which MB you like better? - rickmcinnis@dogwoodfabrics.com 09:18:24 06/13/08 (11)
In Reply to: RE: YES with E7200 posted by cics on June 12, 2008 at 16:45:30
Dear cics,
The slightly cheaper one you seemed to like better is the one (naturally) I would order unless you have firmed up your recommendation.
Is there something about the GIGABYTE you like over the other companies also using the INTEL G31? Just wanting to be sure, especially when an install is required. Though I've got so used to that now, I should be able to do it without the instructions, but I still need the instructions!
The more I listen to #17 its superiority comes through. I do believe what I thought I was missing was a pleasant distortion. #17 is beautiful.
I am using the power supply I asked you about, the ENERMAX MODU82+ EMD625AWT 625W ATX12V Ver.2.3. It is very quiet and I like the idea of a strong supply loafing along. Whether it sounds any better than the "thermaltake", I did not do any back and forth. Excuse my lack of scientific analysis for this. They claimed they were using premium parts in the thing and I took the bait. I also like not having to cut off cables. If you do not need them you can leave them off.
I still cannot get my onboard USB (w/ Granite supply) to work. I have made sure, ten times, that all is connected correctly and it still will not work. What do you do with the "fifth" wire? I wonder if this is where I go wrong. Mine is not connected to anything.
I am using the TDK filters, though I am sure it would be better to do it RIGHT.
I will get a pair of 1gB memory cards and use the existing one in my regular computer. Did you say that these are matched when bought in pairs or am I dreaming (again).
Your thoughts are NEEDED and appreciated!
Thanks!!!
Rick McInnis
- RE: Have you decided which MB you like better? - cics 10:09:34 06/13/08 (10)
In Reply to: RE: Have you decided which MB you like better? posted by rickmcinnis@dogwoodfabrics.com on June 13, 2008 at 09:18:24
Gigabyte GA-EG31M-S2 would be my choice as it offers improved power supply to CPU and there's a new power management feature that may prove useful.
Your 625watt PSU presents another option which would work with your current setup. You could get an E4300 and run it standard with RAM set to 200MHz. This is very similar to an E6300 without some features. I have the E6300 do 192k with RAM at 200Mhz - no problems. This is easier - just do "Load Optimized Defaults" in BIOS and reapply all changes.
On GD USB port, try having HID service running, check PSU is connected (use a USB powered device to check), check data connections.
Yes RAM is best purchased in matched pairs.
I have my 2nd cMP in the reference setup (using 700watt PSU and underclocked E6300 doing 96k at 145db SNR). Everything as noted in "B17's performance" with added midrange beauty - absolutely stunning!
- Those processors do not seem to be available - rickmcinnis@dogwoodfabrics.com 11:08:38 06/13/08 (9)
In Reply to: RE: Have you decided which MB you like better? posted by cics on June 13, 2008 at 10:09:34
Dear cics,
I looked on NEWEGG and the processors recommended seem to be unavailable.
The GIGABYTE board, the e7200 and the matched RAM, if this results in the best possible implementation of cPLAY and CMP, I will go with it. I was not complaining about the re-install. I want to go with what you think is the best approach.
Now, I am confused (again). Please tell me what you think is best. If you do, indeed, prefer those other processors advise where I could find them, please.
Thanks,
Rick McInnis
- Best: Gigabyte mobo+E7200+matched 1GB RAM (2x512MB) - cics 22:41:48 06/13/08 (8)
In Reply to: RE: Those processors do not seem to be available posted by rickmcinnis@dogwoodfabrics.com on June 13, 2008 at 11:08:38
That's what I'm planning to implement.
- RE: Best: Gigabyte mobo+E7200+matched 1GB RAM (2x512MB) - Frodan 19:24:32 06/17/08 (3)
In Reply to: RE: Best: Gigabyte mobo+E7200+matched 1GB RAM (2x512MB) posted by cics on June 13, 2008 at 22:41:48
Hello cics,I am following with interest and getting ready to test a Biostar M890 with an E2140 processor.
I seem to recall that you preferred using the Biostar MB because of the on board RAID which allowed us to not use the IDE subsystem in Windows XP. Is there something that has changed since then as the Gigabyte has no onboard RAID?
Frodan
Edits: 06/17/08 06/17/08
- RE: Best: Gigabyte mobo+E7200+matched 1GB RAM (2x512MB) - cics 23:20:13 06/17/08 (2)
In Reply to: RE: Best: Gigabyte mobo+E7200+matched 1GB RAM (2x512MB) posted by Frodan on June 17, 2008 at 19:24:32
Biostar (rather use P4M900 and not P4M890) cannot run E7200 processor. Gigabyte mobo recommended is very new but offers similar features to the Biostar. E7200 is essential for highest quality SRC upsampling (145db SNR).
- RE: Best: Gigabyte mobo+E7200+matched 1GB RAM (2x512MB) - Frodan 12:33:48 06/18/08 (1)
In Reply to: RE: Best: Gigabyte mobo+E7200+matched 1GB RAM (2x512MB) posted by cics on June 17, 2008 at 23:20:13
The Gigabyte GA-EG31M-S2 does not have an onboard RAID controller. Wouldn't a board like the GA-VT890P using the same VIA P4M900/8237S chipset as the Biostar suit better? Another board might be the GA-G33M-DS2R which uses the Intel G33 chipset and has onboard RAID as well.
Has the requirement to use RAID to disable the IDE drivers in WinXP been eliminated or are you thinking of not using the onboard SATA ports at all and using either a seperate PCI-E SATA RAID controller or some other form of storage, such as an outboard bootable USB or network based drive?
I was working with the Biostar because I happened to have access to one cheap but will now go to a more recent option as I do need it to do HD Video as well as sound. (Using a multiboot config for optimised versions of XP). I have yet to obtain a CPU however and the E7200 sounds like an excellent solution for low wattage yet quite powerful CPU.
Frodan
- RE: Best: Gigabyte mobo+E7200+matched 1GB RAM (2x512MB) - cics 08:06:30 06/19/08 (0)
In Reply to: RE: Best: Gigabyte mobo+E7200+matched 1GB RAM (2x512MB) posted by Frodan on June 18, 2008 at 12:33:48
With RAM playback using either cMP's method of loading into System Cache or even better through cPlay, benefits of using RAID (no IDE) is done away. The only remaining benefit of RAID is reduced RAM load time as sequential disk reads is much faster. With SSDs this benefit will also be done away with as SSD's offer significantly faster read speeds. SATA is used.
Avoid G33 chipset as it uses slightly more power.
- Intel e8190 & RAM latency - rickmcinnis@dogwoodfabrics.com 09:40:44 06/16/08 (3)
In Reply to: RE: Best: Gigabyte mobo+E7200+matched 1GB RAM (2x512MB) posted by cics on June 13, 2008 at 22:41:48
Dear cics,
Since you got us started looking at new implementations it is easy to get side-tracked.
Just for fun as I was looking at the INTEL stuff I saw something called a e8190 which seems to be an e7200 with higher buss speed and more L2 cache. Now, of course, I have no idea if this is good or bad and my curiosity has got the best of me, so I am asking. It is also a 65W processor. Just wondered if this could be a, yet, better choice?
It does not seem to be available at the moment which could make it all a moot point. (stab at wry humour)
I assume one should get the RAM with the lowest latency? How great a disadvantage of getting 2 - 1gB cards? It seems to have jumped back and forth, just to be on the sage side, especially if one were to load some 96K 24 bit files into the system.
Just want to be sure.
THANKS,
Rick McInnis
- Got the recommended stuff today - rickmcinnis@dogwoodfabrics.com 07:51:13 06/19/08 (2)
In Reply to: RE: Intel e8190 & RAM latency posted by rickmcinnis@dogwoodfabrics.com on June 16, 2008 at 09:40:44
Dear cics,
Got the recommended MB and processor today.
I hope I got the right memory, KINGSTON 512Mx2|KST KVR400D2N3K2/1G R. This had the lowest latency though I could have had higher latency with more speed. Was my choice correct?
Will get it working ASAP. I guess I will not be able to resist starting off with #18.
Thanks,
Rick McInnis
- RE: Got the recommended stuff today - cics 08:23:18 06/19/08 (1)
In Reply to: RE: Got the recommended stuff today posted by rickmcinnis@dogwoodfabrics.com on June 19, 2008 at 07:51:13
Rather go with the recommended RAM in cMP's FAQ:
Kingston HyperX 1GB: TWIN PACK (2X512MB): KHX6400D2LLK2
Compare both B18 & B19.
Note that some BIOS changes done on Biostar cannot be done using the G31 chipset as used on the Gigabyte mobo. Also with the G31 chipset, you should see ~50MB more Avail RAM (more efficient VGA RAM allocation).
- HYPERX seems to be in short supply - rickmcinnis@dogwoodfabrics.com 09:35:33 06/19/08 (0)
In Reply to: RE: Got the recommended stuff today posted by cics on June 19, 2008 at 08:23:18
Dear cics,
It turns out I could not have got that if I had looked at the cMP FAQ (which I should have, of course. I figure my repeat questions might be of use to others; is that excuse acceptable?) This seems to be out of stock at NEWEGG. I tried another shop and they, too, were out of stock. Until I can get the specified package I will get to know what I have.
When the HYPERX becomes available again I will see if there is a difference between the two qualities of memory.
I will start with #18 (it will take all of my willpower to not go with your latest version). Sunday evening I will make the change to #19. I will start with SRC=0.
I am very excited about hearing what this can do.
I will let you know Monday morning.
THANKS,
Rick McInnis
- Cics I was with you to 16 but I don't understand 17 - theob 12:05:54 06/10/08 (1)
In Reply to: RE: cPlay 1.0b17 Released posted by cics on June 10, 2008 at 10:38:25
So if I follow all of the steps I can run more than 1 wav file at one time? Is that correct? But how do I implement that? But to use src I need 2-4 gigs of ram, is that right? If src is not used is there still a sonic advantage in using 17? Can you give specific steps on how to run several wav files?Let me be clear:
How does one insert a line (AWE) into the cicsplay program? (Honestly I do not know how to do that.)
I want the src badly. I am a believer in your approach to audio playback. I only have 1 gig of memory and I have asio4all. If I have to buy 4 gigs of memory and buy asio usb I will do so to get src and hirez capability. I'm just confused on the steps I have to do on my side. Please sir be more specific and give us a 'how to implement version 17 for dummies'.
Edits: 06/10/08 06/10/08
- RE: Cics I was with you to 16 but I don't understand 17 - cics 02:33:30 06/11/08 (0)
In Reply to: RE: Cics I was with you to 16 but I don't understand 17 posted by theob on June 10, 2008 at 12:05:54
Multi-file cue sheets is supported since b15. If you have cue sheets that have more than one FILE directive then this is for you otherwise don't worry about it. For my setup I use single file cue sheets only.
No additional RAM is needed with b17 - the point on adding more RAM only applies to AWE. (Yes, to use it the line AWE must be added in the file). Windows doesn't allocate all Avail RAM (as you see in task manager) and if you want to always use AWE then adding more RAM helps. cPlay will always revert to standard RAM allocation when AWE fails. In my setup with 1 GB RAM (and avail RAM is ~830MB) I have just 8 CDs that won't do AWE.
AWE is entirely optional (but recommended). For this to work, you must follow the steps outlined (to set LOCK privilege).
How does one insert a line (AWE) into the cicsplay program?
Locate the file (cicsPlay.ini) using File Explorer. Double-click it. Notepad is the file editor - go to end of last line, hit Enter, then type AWE. Save & Exit.
...I have asio4all. If I have to buy 4 gigs...
With ASIO4ALL you can only upsample to 48k (which wouldn't need lots of CPU power). You don't need to buy more RAM if you have sufficient avail RAM (~820MB or more). The comment on 4GB is related to AWE feature.
- RE: cPlay - the open source high-end audio player using ASIO - Doc Gaw 12:05:35 06/08/08 (1)
In Reply to: RE: cPlay - the open source high-end audio player using ASIO posted by cics on May 05, 2008 at 12:31:58
Possibly a dumb question; Can cPlay be made to play not using ASIO with Vista which doesn't use the K-mixer?
- Yes - use ASIO4ALL - cics 10:13:50 06/10/08 (0)
In Reply to: RE: cPlay - the open source high-end audio player using ASIO posted by Doc Gaw on June 08, 2008 at 12:05:35
It emulates an ASIO driver but sends audio via Kernel Streaming. ASIO4ALL is an excellent choice but limited to 48k.
- cPlay 1.0b16 Released - cics 23:43:34 06/07/08 (6)
In Reply to: RE: cPlay - the open source high-end audio player using ASIO posted by cics on May 05, 2008 at 12:31:58
Change Log (final beta release):
- Revert back to heap RAM allocation (best)
- Fix for exit playback mode (wait until Settings is done before cPlay exits)
Please REMOVE previous versions before installing cPlay 1.0b16 .
RAM load is limited to ~1GB (Windows limitation). Looking into accessing physical RAM directly...
For setups using analogue VC, set cPlay's volume to -0.5db rather than 0.0db. This allows SRC upsampler to interpolate original sample points at 0dbFS. Clipping is avoided.
On my reference setup: Juli@ feeds AA Prestige SE. Juli@ improves slightly over RME HDSP9652. Scarlatti DAC has tighter bass whilst AA delivers a superior midrange (tubes?). Before cPlay, the Scarlatti DAC enjoyed a definite advantage over AA - not anymore!
- RE: cPlay 1.0b16 Released - Giulio 21:48:00 06/09/08 (0)
In Reply to: RE: cPlay 1.0b16 Released posted by cics on June 07, 2008 at 23:43:34
Hi Cics,
as the previous poster says excellent transparency. Much more detail. Compared to other players, of which Album Player is the one that sounds better to my ears, Cplay sounds to me to be more laid back (possibly too much) as if slowing the music a tiny bit.
Support for multiple-file cue sheets is the icing on the cake!
Thanks for the excellent work.
Best
Giulio
- RE: latest version error when opening most cue files - aljordan 19:31:56 06/09/08 (3)
In Reply to: RE: cPlay 1.0b16 Released posted by cics on June 07, 2008 at 23:43:34
Hi,
I am getting errors opening many cue files in the latest release. The output from the diagnostics window is at the bottom of the note. The application is complaining about not being able to allocate memory, even though there is plenty RAM available. I tested loading a flac file fcrom the cue sheet and there are no problems doing so. Any suggestions?
Thanks,
Alan
cicsPlay 1.0b16, licensed under GNU GPL
ASIO Technology by Steinberg Media Technologies GmbH
ASIO is a trademark and software of Steinberg Media Technologies GmbH
ASIO Driver Initializing...
Name (ASIO E-MU 0404 | USB)
Player ASIO version (2)
Driver ASIO version (2)
Message (No ASIO Driver Error)
Channels (inputs: 4, outputs: 4)
Buffer details (min: 88, max: 22000, preferred: 440, granularity: 44)
Sample rate ( 44100)
Output Ready? Supported
Preparing buffers... successful
Latencies (input: 523, output: 656)
ASIO Driver initialized.
DSP starting...
DSP Buffer (21120 samples or 479ms/ 2.1/ 100.2) allocation successful
ASIO Driver starting...
Input 1 [ Analog IN A ] Active (0) type (32bits)
Input 2 [ Analog IN B ] Active (0) type (32bits)
Input 3 [ SPDIF IN L ] Active (0) type (32bits)
Input 4 [ SPDIF IN R ] Active (0) type (32bits)
Output 1 [ Analog OUT ] Active (1) type (32bits) > Left Channel
Output 2 [ Analog OUT ] Active (1) type (32bits) > Right Channel
Output 3 [ SPDIF OUT ] Active (0) type (32bits)
Output 4 [ SPDIF OUT ] Active (0) type (32bits)
ASIO started successfully.
AUDIO source [ C:\Cues\Beth Orton - Trailer Park.cue ]
Genre = Unknown
Artist = Beth Orton
Album = Trailer Park
FILE: Z:\Beth Orton\Trailer Park\01 - She Cries Your Name.flac
FLAC file size = 27MB (decoded during RAM load, RAM for 'Wav data' is allocated)
Wav data size = 48MB
Rate = 16/44100 (2 channels)
Samples = 12630240 ( 4:46)
FILE: Z:\Beth Orton\Trailer Park\02 - Tangent.flac
FLAC file size = 45MB (decoded during RAM load, RAM for 'Wav data' is allocated)
Wav data size = 75MB
Rate = 16/44100 (2 channels)
Samples = 19796196 ( 7:28)
FILE: Z:\Beth Orton\Trailer Park\03 - Don't Need a Reason.flac
FLAC file size = 30MB (decoded during RAM load, RAM for 'Wav data' is allocated)
Wav data size = 51MB
Rate = 16/44100 (2 channels)
Samples = 13440504 ( 5: 4)
FILE: Z:\Beth Orton\Trailer Park\04 - Live As You Dream.flac
FLAC file size = 19MB (decoded during RAM load, RAM for 'Wav data' is allocated)
Wav data size = 30MB
Rate = 16/44100 (2 channels)
Samples = 7896840 ( 2:59)
FILE: Z:\Beth Orton\Trailer Park\05 - Sugar Boy.flac
FLAC file size = 25MB (decoded during RAM load, RAM for 'Wav data' is allocated)
Wav data size = 43MB
Rate = 16/44100 (2 channels)
Samples = 11513040 ( 4:21)
FILE: Z:\Beth Orton\Trailer Park\06 - Touch Me with Your Love.flac
FLAC file size = 39MB (decoded during RAM load, RAM for 'Wav data' is allocated)
Wav data size = 75MB
Rate = 16/44100 (2 channels)
Samples = 19730340 ( 7:27)
FILE: Z:\Beth Orton\Trailer Park\07 - Whenever.flac
FLAC file size = 22MB (decoded during RAM load, RAM for 'Wav data' is allocated)
Wav data size = 39MB
Rate = 16/44100 (2 channels)
Samples = 10310580 ( 3:53)
FILE: Z:\Beth Orton\Trailer Park\08 - How Far.flac
FLAC file size = 27MB (decoded during RAM load, RAM for 'Wav data' is allocated)
Wav data size = 44MB
Rate = 16/44100 (2 channels)
Samples = 11786460 ( 4:27)
FILE: Z:\Beth Orton\Trailer Park\09 - Someone's Daughter.flac
FLAC file size = 28MB (decoded during RAM load, RAM for 'Wav data' is allocated)
Wav data size = 43MB
Rate = 16/44100 (2 channels)
Samples = 11307240 ( 4:16)
FILE: Z:\Beth Orton\Trailer Park\10 - I Wish I Never Saw the Sunshine.flac
FLAC file size = 25MB (decoded during RAM load, RAM for 'Wav data' is allocated)
Wav data size = 47MB
Rate = 16/44100 (2 channels)
Samples = 12471480 ( 4:42)
FILE: Z:\Beth Orton\Trailer Park\11 - Galaxy of Emptiness.flac
FLAC file size = 57MB (decoded during RAM load, RAM for 'Wav data' is allocated)
Wav data size = 102MB
Rate = 16/44100 (2 channels)
Samples = 26945100 ( 10:11)
WINDOWS memory allocation failed!
Physical (2038MB), Available (1631MB), Required (603MB), Max supported (2048MB).
- I'm able to RAM load via multi-file cue sheet to ~1GB using b16. Windows fails thereafter. - cics 10:10:53 06/10/08 (2)
In Reply to: RE: latest version error when opening most cue files posted by aljordan on June 09, 2008 at 19:31:56
The error message in your diagnostics indicates that Windows failed to allocate the RAM cPlay is requesting. Try:
Setting Windows to 'System Cache'.
If that fails, set '/3GB' option in boot.ini file of Windows OS (take great care here). This option is only suggested for dedicated audio computers.
B17 will offer another way to RAM load (XP or later). What OS are you using - recommended is XP Pro SP2.
- RE: I'm able to RAM load via multi-file cue sheet to ~1GB using b16. Windows fails thereafter. - aljordan 17:18:51 06/10/08 (1)
In Reply to: RE: I'm able to RAM load via multi-file cue sheet to ~1GB using b16. Windows fails thereafter. posted by cics on June 10, 2008 at 10:10:53
Hi Cics,I am using Windows XP Professional SP2. Setting to use system cache fixed the issue for some of the cue files, but not all. I'll look into trying soe of the other options.
Thanks,
Alan
Edits: 06/10/08
- RE: I'm able to RAM load via multi-file cue sheet to ~1GB using b16. Windows fails thereafter. - cics 12:02:53 06/11/08 (0)
In Reply to: RE: I'm able to RAM load via multi-file cue sheet to ~1GB using b16. Windows fails thereafter. posted by aljordan on June 10, 2008 at 17:18:51
Using System Cache is important. Windows won't give you the full Available RAM with AWE.You should get close to 1GB allocation.
- RE: cPlay 1.0b16 Released - sharesoftall@googlemail.com 05:14:10 06/08/08 (0)
In Reply to: RE: cPlay 1.0b16 Released posted by cics on June 07, 2008 at 23:43:34
Excellent sound. Even from my Edirol UA-1EX usb sound card sound is superb. I am hearing details in CDs I never knew existed. I am using analogue outs of my sound card. I just built my Media PC and planning to upgrade to a DAC
But I am completely new in DAC area.
I am already happy as the sound is much much better than my stand alone NAD C542 CD player
Thank you cics for developing such a exceptional player
sharesoftall Yes
- cPlay 1.0b15 Released - cics 10:40:13 06/06/08 (9)
In Reply to: RE: cPlay - the open source high-end audio player using ASIO posted by cics on May 05, 2008 at 12:31:58
Change Log (final beta release):
- Tuning parameters reverts to b13 (best)
- Cater for multi-file cue sheets
- (Minor) DSP efficiency implementation
- (Minor) Prevent explorer from displaying cPlay instead of file type
- Non-heap RAM allocation implementation
Please REMOVE previous versions before installing cPlay 1.0b15 .
Multi-file cue sheets look like this:
PERFORMER "Tracy Chapman"
TITLE "Tracy Chapman"
FILE "C:\Music\01 Track01.wav" WAVE
TRACK 01 AUDIO
TITLE "Track01"
PERFORMER "Tracy Chapman"
INDEX 01 00:00:00
FILE "Pink Floyd - The Wall 1.flac" WAVE
TRACK 02 AUDIO
TITLE "Track02"
INDEX 01 00:00:00
FILE "Track03.wav" WAVE
TRACK 03 AUDIO
TITLE "Track03"
INDEX 01 00:00:00
As can be seen, there's lots of flexibility here! I'm not planning for more beta releases unless there's bugs. cPlay 1.0b15 will remain for a while to allow for greater test coverage.
- RE: for those of you who have discs ripped to multiple files... - aljordan 21:07:00 06/07/08 (3)
In Reply to: RE: cPlay 1.0b15 Released posted by cics on June 06, 2008 at 10:40:13
For those of you who have CDs ripped to multiple files instead of a single large file, and would like cue sheet support for the CD as described in cics message above, please read on.
I am currently creating a little Java application that will recursively parse my music library and create a single cue sheet for all songs found in each directory. For example, my music library has a directory structure of:
Library Root\Artist\Album\Songs
The application will create a cue sheet at each directory where songs are found, referencing all of the songs in that directory. Currently it is just a command line application, but if anyone is interested I can make a user interface for it and make it available to others.
Alan
- RE: for those of you who have discs ripped to multiple files... - theob 05:34:18 06/08/08 (1)
In Reply to: RE: for those of you who have discs ripped to multiple files... posted by aljordan on June 07, 2008 at 21:07:00
Either I would appreciatethe knowledge to use it. But I mean very very simple step by step knowledge.
- RE: for those of you who have discs ripped to multiple files... - aljordan 08:10:57 06/08/08 (0)
In Reply to: RE: for those of you who have discs ripped to multiple files... posted by theob on June 08, 2008 at 05:34:18
Hi,
Send me an email telling me what the directory structure of your music library looks like.
Alan
- Just make sure songs don't exceed 1GB ram. - cics 21:37:00 06/07/08 (0)
In Reply to: RE: for those of you who have discs ripped to multiple files... posted by aljordan on June 07, 2008 at 21:07:00
Nice tool.
- RE: cPlay 1.0b15 Released - aljordan 13:12:10 06/07/08 (0)
In Reply to: RE: cPlay 1.0b15 Released posted by cics on June 06, 2008 at 10:40:13
Hi cics,
Thank you for implementing multiple file cue sheet support.
Alan
- Magnificent - rickmcinnis@dogwoodfabrics.com 11:55:55 06/07/08 (2)
In Reply to: RE: cPlay 1.0b15 Released posted by cics on June 06, 2008 at 10:40:13
Dear cics,
It is as stated.
So far, completely stable with that dimension/air quality of #13.
There is tonality to the bass that either is a new refinement or I am just now picking up on it, per the usual bromide, I am hearing things not heard before.
I heard a certain pontiff speaking of up-samplers and a resulting "big, dumb sound" which I am at a loss to understand from my experience. It would seem a BIG dumb sound would be preferable to a small, dumb sound? The sound of #15 is BIG and quite sophisticated. It is funny that one of the persistent complaints of digital vs. analog sound was its relative lack of scale and dimension. Which leaves me wondering just what it is the pontiff is after? I guess when one's goal is to derive the best sound possible from ITUNES one has a very different point of reference? Sorry for all of the questions but I am still trying to understand that bizarre comment! (made in another thread within this forum)
There are always many ways to achieve greatness and one is cautioned at being too sure theirs is the only way. "Jitter" is kind of like sin, there are many types, many sources, etc., but we seem to look foolish when we decide which is the worst form.
All my experience with cics's approach to digital playback tells me I am hearing an intact waveform (for lack of a better phrase) or at least the most intact I have heard. It would seem that "jitter" is the phenomenon that disembodies the musical waveform that makes digital playback the annoyance it becomes when listening for more than an hour. The brain becomes exhausted trying to put the pieces into place and says, ENOUGH.
I am sure there is much progress left to be made, but cics is on to something very important. Dismiss his approach if your marketing plan must or if your ego cannot stand that this, relatively, simple approach makes your kilodollar investment seem silly, but for the music lover this is an incredible gift that will bring much pleasure.
In addition, what you learn about computers in the process is an interesting side benefit.
THANKS,
Rick McInnis
unabashed fan
- RE: Magnificent - theob 18:15:51 06/07/08 (1)
In Reply to: RE: Magnificent posted by rickmcinnis@dogwoodfabrics.com on June 07, 2008 at 11:55:55
Nobody has said it better, you made music out of words. Nice post and yes I agree.
- B16 is in the works... - cics 21:38:32 06/07/08 (0)
In Reply to: RE: Magnificent posted by theob on June 07, 2008 at 18:15:51
Yes B15 has excellent bass, but b13 offers better midrange. B16 addresses this and a bug fix.
- RE: cPlay 1.0b15 Released - theob 11:43:59 06/06/08 (0)
In Reply to: RE: cPlay 1.0b15 Released posted by cics on June 06, 2008 at 10:40:13
Good call (ie., going back to 13) while I never posted I did prefer this as well. Still having trouble with pops while using my opticis cable. By passing the opticis makes the problem go away, go figure. I'm not sure just how to configure multiple file cue sheets. Do you do in EAC or elsewhere?
- cPlay 1.0b14 Released - cics 23:08:22 06/02/08 (10)
In Reply to: RE: cPlay - the open source high-end audio player using ASIO posted by cics on May 05, 2008 at 12:31:58
Change Log (pre-final release):
- Minor tuning change
- Retain phase setting between cPlay sessions
- Retrieve volume to 0.5 db level
- Support for up to 384k sampling rate (done via ini file)
Betters 1.0b13!
Please REMOVE previous versions before installing cPlay 1.0b14 .
- RE: cPlay 1.0b14 Released - theob 05:05:54 06/03/08 (9)
In Reply to: RE: cPlay 1.0b14 Released posted by cics on June 02, 2008 at 23:08:22
I have been having trouble presumably with asio4all. Any thoughts of using direct sounf?
- Try reinstalling ASIO4ALL - cics 11:15:56 06/03/08 (8)
In Reply to: RE: cPlay 1.0b14 Released posted by theob on June 03, 2008 at 05:05:54
Make sure to remove it before reinstallation.
DS is not a good choice.
- RE: Try reinstalling ASIO4ALL - fr338 03:03:08 06/04/08 (4)
In Reply to: RE: Try reinstalling ASIO4ALL posted by cics on June 03, 2008 at 11:15:56
I am using USB-> spdif converter to dac. I have two questions.
1. When using foobar, SRC is functional up to 176K( although the converter can take 48KHz hardware-wise). With cics player, I can play strictly on 44.1K or 48K only. What is the reason for this ?
2. Can cics play multiple wav and flac files from different ripped discs like the playlist function in foobar ?
- Reponse to questions - cics 11:30:07 06/04/08 (3)
In Reply to: RE: Try reinstalling ASIO4ALL posted by fr338 on June 04, 2008 at 03:03:08
When using foobar, SRC is functional up to 176K( although the converter can take 48KHz hardware-wise). With cics player, I can play strictly on 44.1K or 48K only. What is the reason for this ?
Sample rate supported is determined by your ASIO driver. If you're using ASIO4ALL, max rate is 48k.
Can cics play multiple wav and flac files from different ripped discs like the playlist function in foobar ?
B15 will allow for muti-file cue sheets (playlist). That means you can mix wav and flac files (as long as they have same bit depth and rate) into a single cue sheet and have cPlay play it from RAM. Currently only single file cue sheets can be played.
UI for dynamically building playlists is not supported - I hope someone has the time to extend cPlay in this direction after final release.
- RE: Reponse to questions - fr338 22:22:52 06/04/08 (0)
In Reply to: RE: Reponse to questions posted by cics on June 04, 2008 at 11:30:07
I use ASIO4ALL too during running foobar.
My setup : laptop -> usb_to_spdif converter -> audio DAC
The converter is using TI chip, that suppose to be able to accept up to 48K only. But in reality foobar with SRC dll can get to 176K rate, which seems to be the Intel core solo 1.2G limit.
- RE: Reponse to questions - appleteapot 17:37:41 06/04/08 (0)
In Reply to: RE: Reponse to questions posted by cics on June 04, 2008 at 11:30:07
Hi Cics,
Thanks again for adding the cue sheet feature. However, since many of us are using foobar m3u playlist, I would then prefer create a program to feed in m3u list. Could you please keep the convenient features like command line and auto exit features in all subsequent versions? So that we can continue work on m3u playlist for cplay.
I am still heading to replace foobar with cicsplay.
I will test b14 later on as I am sooooo happy with b13 :)
Thanks!
- RE: Reponse to questions - theob 12:11:27 06/04/08 (0)
In Reply to: RE: Reponse to questions posted by cics on June 04, 2008 at 11:30:07
Does asio usb support other than 44 and 48 khz?
- RE: Try reinstalling ASIO4ALL - theob 12:29:57 06/03/08 (1)
In Reply to: RE: Try reinstalling ASIO4ALL posted by cics on June 03, 2008 at 11:15:56
I tried reinstalling, problem persists and shutdown my playback. Perplexed!
- RE: Try reinstalling ASIO4ALL - theob 13:24:25 06/03/08 (0)
In Reply to: RE: Try reinstalling ASIO4ALL posted by theob on June 03, 2008 at 12:29:57
I took out my opticis cable, I'm running usb from dell to benchmark. It fixed the issue. I'll let it run for a few days to see if the pops come back.
- RE: Try reinstalling ASIO4ALL - theob 12:20:54 06/03/08 (0)
In Reply to: RE: Try reinstalling ASIO4ALL posted by cics on June 03, 2008 at 11:15:56
ok thanks
- cPlay 1.0b13 released - cics 04:13:18 05/31/08 (11)
In Reply to: RE: cPlay - the open source high-end audio player using ASIO posted by cics on May 05, 2008 at 12:31:58
Change Log:
- Minor tuning change
- Prevent ASIO driver from issuing Resyncs
Please REMOVE previous releases before installing cPlay1.0b13 .
- RE: cPlay 1.0b13 released - seger 13:11:48 06/02/08 (2)
In Reply to: RE: cPlay 1.0b13 released posted by cics on May 31, 2008 at 04:13:18
Stunning (cPlay v.13). Excellent definition, especially in bass, and better in midrange since I last tried it (version 3 or 4), and thanks for making the titles smaller (why not do the same for cMP – it looks gigantic now and I can only see the first 2 or 3 words of the titles!). cPlay is comfortable to listen to now, better balance, and still sounds best with 192K SRC, which I find puzzling, since in Foobar I usually don’t like it – not with ASIO and especially not with KS. Also I now have superb sound out of my second (old) PC with its E-mu 0202 USB which struggles with Foobar ASIO and won’t accept KS at all.
You might reconsider the full-screen display that now occurs when cPlay is used in cMP – Process Explorer is obscured, which could be a problem if there’s a system hang (although it has functioned fine, with no problems so far), plus it would be good to see CPU usage and other details.
Do you think you could make cPlay accept multiple-file cuesheets at some point?
- B13 vs B14 - cics 11:27:40 06/03/08 (1)
In Reply to: RE: cPlay 1.0b13 released posted by seger on June 02, 2008 at 13:11:48
Do you think you could make cPlay accept multiple-file cuesheets at some point?
I think this is needed before final release...
Comparing b13 to b14 is interesting. B13 has the edge over b14 in the midrange whilst b14 offers slightly more 'air'. After a lot of back and forth comparisons I'm more inclined to go with b13's tuning parameters. Check out b14 and let me know.
- RE: B13 vs B14 - seger 15:16:23 06/03/08 (0)
In Reply to: RE: B13 vs B14 posted by cics on June 03, 2008 at 11:27:40
Good news about the multiple-file cuesheets.
Re 13 vs 14: Maybe the highs are toned down a tiny bit in 14 – that’s good, although I don’t hear what I would call airiness. I think the overall balance in 13 is more pleasing, so I’m leaning that way after a couple of hours of back-and-forthing.
How about making the default about -10dB instead of -30? (sore clicking finger!)
- Instability like #9? - rickmcinnis@dogwoodfabrics.com 08:43:58 06/02/08 (3)
In Reply to: RE: cPlay 1.0b13 released posted by cics on May 31, 2008 at 04:13:18
Dear cics,
I have found #13 to sound very good. Of course, straightening out my cMP problem (start cMP) could have made a difference.
As an aside, what is the difference between the computer srarting cMP on booting versus "going into" cMP from XP? Are more processes running in the latter mode?
The bass is almost too powerful. I think I need to reduce the level on my subwoofers (below 75 hz) about a dB and see if that cures my complaint.
Have you noticed an instability like #9's? #12 was stable, but #13 will just STOPS, then the machine re-boots, goes into cMP (as it should) and is ready to go again. Slightly disconcerting when listening at realistic volume levels. Silence where, before, there was a room of sound. I went through this three times last night.
Do you think this is something peculiar to my situation?
Otherwise, I think it sounds wonderful, and I mean that literally.
THANKS,
Rick McInnis
- RE: Instability like #9? - cics 11:19:39 06/03/08 (2)
In Reply to: RE: Instability like #9? posted by rickmcinnis@dogwoodfabrics.com on June 02, 2008 at 08:43:58
...what is the difference between the computer srarting cMP on booting versus "going into" cMP from XP? Are more processes running in the latter mode?
cMP mode doesn't start task bar processes that interfere with SQ.
Reboot may be related to Windows - check if you have enough free disk space (at least 200MB). I have had b13 running for 12+ hours without any issues.
Is this a new install of Windows?
- plenty of space on drive - rickmcinnis@dogwoodfabrics.com 08:34:27 06/04/08 (0)
In Reply to: RE: Instability like #9? posted by cics on June 03, 2008 at 11:19:39
Dear cics,
I have 1.6 gb of space left in the sector where WINDOWS is installed.
I did notice I had inadvertently left a memory stick in one of the USB ports. I did not notice this until after listening for the evening. Could this make a difference?
It seems it went awhile before crashing (and re-booting itself) but after the first time, the second occurrence was in about half the time and the third was almost immediate, so I went to bed.
I have not tried #14 yet.
Maybe there was an error in the installation of #13?
From the way you speak of these two versions I will go back and forth between them to see what I think.
THANKS,
Rick McInnis
- RE: Instability like #9? - rickmcinnis@dogwoodfabrics.com 13:20:59 06/03/08 (0)
In Reply to: RE: Instability like #9? posted by cics on June 03, 2008 at 11:19:39
Dear cics,
Yes, it is a new install.
When I did this one I did it EXACTLY as specified. No liberties!
I will check to see how much free space there is.
When I created the sector for WINDOWS I believe I made it larger than you suggested since I was using a 750 gB drive.
Before when the system would crash it would not come back up like it does now.
I did not listen last night so I will see if it continues to happen.
I had left the machine on before the listening session. It would idle just fine (or at least I thought it was, now that I think about it, it could have gone through its cycle and I not even know (?))
I will try #14 tomorrow.
Thanks for the help and advice,
Rick McInnis
- RE: cPlay 1.0b13 released - appleteapot 20:13:46 05/31/08 (1)
In Reply to: RE: cPlay 1.0b13 released posted by cics on May 31, 2008 at 04:13:18
Hi Cics,
I just tried to playback a 24/96 wav file (purchased from linn.com) but got the following message. The file could be played using Foobar without issue. Could you please help check whether there is any setting change I need to do? Thanks in advance for your help!
cicsPlay 1.0b13, licensed under GNU GPL
ASIO Technology by Steinberg Media Technologies GmbH
ASIO is a trademark and software of Steinberg Media Technologies GmbH
ASIO Driver Initializing...
Name (hdspwdm)
Player ASIO version (2)
Driver ASIO version (1)
Message (No ASIO Driver Error)
Channels (inputs: 8, outputs: 8)
Buffer details (min: 256, max: 256, preferred: 256, granularity: 0)
Sample rate ( 96000)
Output Ready? Supported
Preparing buffers... successful
Latencies (input: 301, output: 414)
Hardware supports additional optimization - enabled
ASIO Driver initialized.
DSP starting...
DSP Buffer (36864 samples or 384ms/ 2.6/ 375.0) allocation successful
ASIO Driver starting...
Input 1 [ HDSP9632 Analog 1 ] Active (1) type (32bits)
Input 2 [ HDSP9632 Analog 2 ] Active (1) type (32bits)
Input 3 [ HDSP9632 SPDIF L ] Active (1) type (32bits)
Input 4 [ HDSP9632 SPDIF R ] Active (1) type (32bits)
Input 5 [ HDSP9632 ADAT 1 ] Active (1) type (32bits)
Input 6 [ HDSP9632 ADAT 2 ] Active (1) type (32bits)
Input 7 [ HDSP9632 ADAT 3 ] Active (1) type (32bits)
Input 8 [ HDSP9632 ADAT 4 ] Active (1) type (32bits)
Output 1 [ HDSP9632 Analog 1 ] Active (1) type (32bits)
Output 2 [ HDSP9632 Analog 2 ] Active (1) type (32bits)
Output 3 [ HDSP9632 SPDIF L ] Active (1) type (32bits) > Left Channel
Output 4 [ HDSP9632 SPDIF R ] Active (1) type (32bits) > Right Channel
Output 5 [ HDSP9632 ADAT 1 ] Active (1) type (32bits)
Output 6 [ HDSP9632 ADAT 2 ] Active (1) type (32bits)
Output 7 [ HDSP9632 ADAT 3 ] Active (1) type (32bits)
Output 8 [ HDSP9632 ADAT 4 ] Active (1) type (32bits)
ASIO started successfully.
Processing CD [G:\Master\Music Clips\Wave\10 - Adagio in E K 261.wav]...
Genre = Unknown
Artist = G:\Master\Music Clips\Wave\10 - Adagio in E K 261.wav
Album = WAV
Content file = G:\Master\Music Clips\Wave\10 - Adagio in E K 261.WAV
Wav file size = 231MB
Wav file is NOT supported.
Only 16 or 24 bit stereo (at any sample rate) is supported.
- RE: cPlay 1.0b13 released - cics 09:01:19 06/01/08 (0)
In Reply to: RE: cPlay 1.0b13 released posted by appleteapot on May 31, 2008 at 20:13:46
It's either not a stereo WAV file (not the case) or it uses a different format type. cPlay only accepts format type 1 (no compression). The wav file you have is compressed - this is rare. There's way too many compression formats and catering for them is a project on its own.
Unrelated: Try setting latency in your RME card to 32 samples - 256 is the default.
- RE: cPlay 1.0b13 released - appleteapot 11:46:29 05/31/08 (1)
In Reply to: RE: cPlay 1.0b13 released posted by cics on May 31, 2008 at 04:13:18
Hi cics,
Thanks! Personally I like 13 over 12. I feel the bass is more solid.
After a few days of reading VB documentation on-line, I have gathered bits and pieces from the tutorial sites and come up with the following very stupid codes which will do the playlist job... with lots of hardcode. Before running the program, I have to create a m3u playlist using Foobar and put it in C:\. There is no way to break the program during playback unless using Task Manager, and the user has to close the cicsplay.exe in order to skip track :) Forgive me as this is my very first VB program...
The codes have been tested in Visual Studio 2008. However, per MSDN, myProcess.WaitForExit() cannot use the "start" command so I cannot use realtime mode as in cicsplay.bat.
----------------------------------------
Public Class Form1
Private Sub Button1_Click(ByVal sender As System.Object, ByVal e As System.EventArgs) Handles Button1.Click
FileOpen(1, "C:\cics.m3u", OpenMode.Input)
Dim trackname As String
Dim myProcess As Process = New Process
myProcess.StartInfo.FileName = "C:\Progra~1\cicspl~1\cicsPlay.exe"
myProcess.StartInfo.WindowStyle = ProcessWindowStyle.Minimized
Do Until EOF(1)
trackname = "C:\" & LineInput(1)
myProcess.StartInfo.Arguments = trackname
myProcess.Start()
myProcess.WaitForExit()
myProcess.Close()
Loop
FileClose()
End Sub
End Class
- B13's performance - cics 09:09:37 06/01/08 (0)
In Reply to: RE: cPlay 1.0b13 released posted by appleteapot on May 31, 2008 at 11:46:29
B13 improves handsomely on b12. It's a small but very important (jitter-related) change that was made in b13. Same samples, better sound! Not sure if b13 can be improved.
Use playback mode 'Exit' in cPlay - cPlay exits at end of track. No need to kill process in task manager.
- cPlay 1.0b12 released - cics 08:44:07 05/29/08 (5)
In Reply to: RE: cPlay - the open source high-end audio player using ASIO posted by cics on May 05, 2008 at 12:31:58
Change Log:
- Increased time delays to support EMU/Creative cards
- Low-level optimization applied in ASIO interface
- Minor improvement in DSP performance
- Use new tuning parameters for all soundcards
This release should work with most soundcards including EMU / Creative.
Midrange performance improved whilst retaining sweet highs and tight bass.
Please REMOVE previous releases before installing cPlay 1.0b12 .
- cPLAY and START cMP - rickmcinnis@dogwoodfabrics.com 13:24:25 05/30/08 (1)
In Reply to: RE: cPlay 1.0b12 released posted by cics on May 29, 2008 at 08:44:07
Dear cics,
Since installing cPLAY I have noticed that I cannot engage START cMP. When I request it, it goes through the motions as it would have before, yet when I restart my computer it does NOT automatically go into cMP.
Initially I thought this was what was supposed to happen. I should have asked earlier; dangerous assumptions at work.
Wednesday I asked for START cMP before listening, but did not restart before I listened. I turned the computer off for the night. Yesterday, I turned it on all excited about #12, it went throught the typical BIOSTARR screen then to the WINDOWS screeen with the blue balls and then to a very blank desktop and would do nothing else. I tried all of the obvious things (well, obvious to me, which is a greatly miniting factor!) to no effect.
I restarted and tried to get SAFE MODE, this was this morning before work (it hit me at breakfast that this should be tried), I was not able to engage SAFE MODE and did not have time to try again. I will try again this evening.
As usual, I beg your patience for advice. Does anything come to mind that might explain what I have done?
The main question is: with cPLAY, does cMP behave the same as it did with FOOBAR?
I cannot wait to hear #12's midrange and micro dynamics!!!!
Thanks,
Rick McInnis
- RE: cPLAY and START cMP - cics 23:17:38 05/30/08 (0)
In Reply to: RE: cPLAY and START cMP posted by rickmcinnis@dogwoodfabrics.com on May 30, 2008 at 13:24:25
...with cPLAY, does cMP behave the same as it did with FOOBAR?
Yes.
Start cMP sets Windows to use cMP as the shell and not Explorer. You must not make any hardware changes when in cMP mode - make sure you switch to XP mode before doing hw changes.
Suggest you go back to XP Mode, reboot and test. Then change to cMP mode and reboot.
- Won't work with vista! Others have this problem? - Riboge 06:09:24 05/30/08 (0)
In Reply to: RE: cPlay 1.0b12 released posted by cics on May 29, 2008 at 08:44:07
On my Dell XPS 410 running Vista with a Lynx aes16 card, b12 as b10 and b11 won't work. Each time one attempts to play a file a message comes up saying the asio settings have changed and cPlay must be restarted. This happens whether or not there have been any changes. I happen also to have a Creative X-Fi card and don't have the same problem using it. Cics tells me all those using XP giving feedback do not have this problem. Any other vista users with a lynx card out there please report your experience with b12.
- RE: cPlay 1.0b12 released - appleteapot 09:24:38 05/29/08 (1)
In Reply to: RE: cPlay 1.0b12 released posted by cics on May 29, 2008 at 08:44:07
As always thank you very much for yet another great update. As always, I can only feel there is a difference between each cPlay versions, as the changes are subtle (maybe my system resolution is not high enough) but I can always find the audible difference between cplay and foobar. Before cplay, I never realize that player can contribute such a great difference.
Currently I feel that b12 === b9 > b8 > > > b10 > b7 > b6 .... b1 > foobar 9.4.4
Keep going for refinements! (And sorry for begging for the playlist feature every couple days)
- B12's performance - cics 20:06:46 05/29/08 (0)
In Reply to: RE: cPlay 1.0b12 released posted by appleteapot on May 29, 2008 at 09:24:38
I've never experienced such a pure midrange until b12. You should get near perfect micro-dynamics and very articulate bass.
It's to do with jitter - that's why audio players sound different yet send the exact same bits to the soundcard. Correct upsampling is the other contributor.
Once the source code is available, I hope someone could add playlists.
- cPlay 1.0b11 released - cics 15:42:04 05/27/08 (1)
In Reply to: RE: cPlay - the open source high-end audio player using ASIO posted by cics on May 05, 2008 at 12:31:58
Change Log (b10, b10a & b11):
- UI messages in accordance with license agreements
- Playback mode 'Exit' added. At end of wav data, cPlay terminates
- Use less aggressive compiler optimizations. cPlay 1.0b9 can be unstable on some Windows installations (depends on soundcard used)
- Reduce Windows kernel overheads
- DSP and Skin components created
- Fix to cater for some soundcards
- Added small time delay at start & exit for better stability
- Fix for string operations applied
- Use default compiler settings for non-critical components
Your feedback on this release is very important for final release (inclusive of source code).
Please remove previous version before installing cPlay 1.0b11 .
- initial impressions - rickmcinnis@dogwoodfabrics.com 08:07:17 05/29/08 (0)
In Reply to: RE: cPlay 1.0b11 released posted by cics on May 27, 2008 at 15:42:04
Dear cics,
Listened to #11 last night.
My instability problem is gone.
BUT, am I detecting the loss of the airiness that was what I (thought) I was perceiving with #9?
I know instability gets old quickly, but I did miss the dimension #9 (seemed) to have. Since I have not gone back and forth (and for all I know I no longer have #9 at my disposal).
Is this one of those either/or things? Like so much in out lives, compromise is ineveitable and I would agree with your choice of ease of use versus this subtle effect of goodness.
Nonetheless, listening to version 11 is a pleasure.
Thanks very much for it,
Rick McInnis
- cPlay 1.0b10 released - cics 11:08:02 05/27/08 (0)
In Reply to: RE: cPlay - the open source high-end audio player using ASIO posted by cics on May 05, 2008 at 12:31:58
Change Log:
- UI messages in accordance with license agreements
- Playback mode 'Exit' added. At end of wav data, cPlay terminates
- Use less aggressive compiler optimizations. cPlay 1.0b9 can be unstable on some Windows installations (depends on soundcard used)
- Reduce Windows kernel overheads
- DSP and Skin components created
cPlay 1.0b10 (and 1.0b9) is NOT recommended with Creative/EMU soundcards.
I plan to do one or two more releases before final release which would include source code. Your feedback on this release would be very useful in getting the final release out sooner.
appleteapot - Your suggestion for 'Exit' mode is excellent and has been implemented.
Please remove previous version before installing cPlay 1.0b10 .
- A proposition to sics - steppe 08:44:02 05/25/08 (1)
In Reply to: RE: cPlay - the open source high-end audio player using ASIO posted by cics on May 05, 2008 at 12:31:58
Hello, Sics. Thanks again for the player and transport.
Here is my proposition, however amateurish and nonprofessional.
In the Windows task manager under process menu we can assign priority to the process from realtime to..., However, there is another line just below which can be translated into English as Assign correspondence(sorry, my XP is Russian) it opens a dialogue which allows to assign the process to both or CPU 0 or CPU1. Is it possible and can it be beneficial to assign all remaining tasks(processes) to CPU 0 and leave cpu 1 to music, so as to "physically" separate them?
Sorry if it sounds foolish, I am not a computer professional, just an eager user.
Serge.
P.S.I tried to do it but my windows denies access, and I am admin. Maybe You can do it, if it's worth it.
Edits: 05/25/08
- Thread optimizations done in cMP and some in cPlay - cics 04:25:44 05/27/08 (0)
In Reply to: RE: A proposition to sics posted by steppe on May 25, 2008 at 08:44:02
Setting of thread priorities is done in cPlay.
The optimization you're referring to (affinities) and more is done in cMP. Use Optimize 'Critical' setting. It's best with ASIO.
I'm not planning to add these affinity optimizations in cPlay itself. You should be able to manually set affinities (using process explorer) on Windows.
- Somebody please add facility to play a sequence of files - Riboge 07:59:25 05/23/08 (4)
In Reply to: RE: cPlay - the open source high-end audio player using ASIO posted by cics on May 05, 2008 at 12:31:58
Though I don't know how, I can't imagine it is all that hard to program the re-loading to RAM of the next file as the one playing vacates it. Even combining files from a previously ripped or downloaded album when these are 24/96 doesn't work since the resulting single file is too large. I find a 1.4G WAV (or flac that converts to WAV this large)will not load with cPlay even when its diagnostic shows available RAM to be larger. Most of us have a lot of albums in multi-file form and prefer it that way. cPlay and cMP are of very little practical use limited to one RAM load as large as Windows will allow. Solving this would make these promising players actually usable day-to-day.
- RE: Somebody please add facility to play a sequence of files - appleteapot 10:28:29 05/23/08 (3)
In Reply to: RE: Somebody please add facility to play a sequence of files posted by Riboge on May 23, 2008 at 07:59:25
I just picked up "VB for Dummies" and working on it... :( But I don't know how it can be done as cicsplay.exe does not have a return code upon finished playback.
- Will be releasing skin component of cPlay (in VC 6.0) - cics 12:56:14 05/24/08 (2)
In Reply to: RE: Somebody please add facility to play a sequence of files posted by appleteapot on May 23, 2008 at 10:28:29
This would be easiest for doing a really nice UI with flexible playlists etc...
There are a few small but important things I want to test/change in the DSP and thereafter package the skin. Should be done in a few days.
- Perhaps cics should start a new thread... - Tony Lauck 16:21:09 05/24/08 (1)
In Reply to: RE: Will be releasing skin component of cPlay (in VC 6.0) posted by cics on May 24, 2008 at 12:56:14
We're now onto page 2 of the Computer Audio Asylum and I almost missed that last post.
Tony Lauck
"Perception, inference and authority are the valid sources of knowledge" - P.R. Sarkar
- Of two minds - rickmcinnis@dogwoodfabrics.com 13:25:12 05/27/08 (0)
In Reply to: RE: Perhaps cics should start a new thread... posted by Tony Lauck on May 24, 2008 at 16:21:09
I wonder if cics should add posts to keep the messages eaily seen by newcomers and "fans" alike but on the other hand keeping all of the information together does have its advantages.
I finally figured out that simply bookmarking the thread made it easy to keep up.
Now, a day does not go by without checking into the cMP and cPLAY threads.
Excuse my enthusiasm but these ARE the most exciting threads in all of the AUDIO ASYLUM.
Can't wait to try out #10.
THANKS immensely cics.
Never ceases to amaze me there is someone like cics with his infinite curiosity and determination to make a great sounding digital playback device and he GIVES IT AWAY. An audio philanthropist!
Rick McInnis
- Has anyone noticed a high pitched whine? - Dawnrazor 20:15:57 05/22/08 (1)
In Reply to: RE: cPlay - the open source high-end audio player using ASIO posted by cics on May 05, 2008 at 12:31:58
It doesnt come through the speakers or anything, but I can hear it from my PC which in my non audio rig sits a foot or so from me.
When cplay is up and running it is there. When it is not it is gone. WHen a song is playing, it is like a high pitched slow machine gun.
I bet I wouldnt hear it in my main system as I sit further away. Something is different though about this player...Foobar doesnt produce this sound!
- I have a moan with CPU bound programs. - Tony Lauck 13:30:18 05/23/08 (0)
In Reply to: RE: Has anyone noticed a high pitched whine? posted by Dawnrazor on May 22, 2008 at 20:15:57
I haven't heard anything with foobar or cPlay, but when I am doing big number crunching, e.g. batch sample rate conversion of large files using Sound Forge, I can hear my computer moaning (perhaps the power supply filter capacitors).
Tony Lauck
"Perception, inference and authority are the valid sources of knowledge" - P.R. Sarkar
- Some Sonic Impressions and some ideas for the future - frennekamp@netcologne.de 12:16:37 05/22/08 (0)
In Reply to: RE: cPlay - the open source high-end audio player using ASIO posted by cics on May 05, 2008 at 12:31:58
I am listening with the CMP you defined basically with a Juli@-Card with 64 samples latency. The DAC is a RAKK MK2 which can directly accept up to 192khz input. It uses a transformer-output stage and this little monster has just completely beaten up the big Accuphase DC801, which I own as well. Much better resolution, much more natural timbre, nothing that messes up the sound. I guess I need to try the Scarlatti next as well.So...I played a bit around:
- CPlay against Foobar: They sound different, foobar seems to have more "attack", plays more in the foreground, cplay more laid back.
- Cplay started through CMP - a different league. Much more natural, more dynamics. So, I continued with this combo.
- CPLAY/CMP with 176khz or 192 upsampled in Cplay: Very smooth trebles, nothing is harsh, very soft, nice....but as well a slide dust or haze on the whole imaging, little los in micro-dynamics
- same but with 44.1khz: Full attack, trebles are brighter, dynamics are brutal, I guess that is how NAIM soundsBAsed on the review of a german newspaper which reviewed the ripping sound of different software, I ripped the same Opus 3 CD with EAC and itunes....yes, itunes !!! Try it please ! It creates wav-files and play those back with cmp/cplay ! I found them to soound much more 3D, brutal dynamics, the EAC rips sound a bit flat compared to those...
...soooooo....if you find the same true and can reproduce this effect, I guess one of the big wishes I would have is a more itunes-compatible way to play back files from the folder structure of itunes. The first simple solution could be, that you allow Cplay to select a complete folder and it plays simply all files in a folder, what ever it finds in the folder...
THX a lot for this marvelous piece of software !
Edits: 05/22/08
- cPlay 1.0b9 released - cics 11:37:44 05/21/08 (1)
In Reply to: RE: cPlay - the open source high-end audio player using ASIO posted by cics on May 05, 2008 at 12:31:58
Change Log:
- Minor UI enhancements (remembers last file type used e.g. .flac when browsing)
- Important refinements - need for machine specific builds removed, e.g. SSE3. Significant efficiencies achieved by not using default windows compiler
Still evaluating this new c/c++ compiler... Impressive!
Please remove previous release before installing cPlay 1.0b9 .
- RE: cPlay 1.0b9 released - Dawnrazor 19:57:32 05/22/08 (0)
In Reply to: RE: cPlay 1.0b9 released posted by cics on May 21, 2008 at 11:37:44
I guess the only way to play a song is to use the folder on the bottom??
- cPlay 1.0b8 released - cics 13:00:37 05/18/08 (4)
In Reply to: RE: cPlay - the open source high-end audio player using ASIO posted by cics on May 05, 2008 at 12:31:58
Change Log:
- Minor UI enhancements
- Improvements in critical areas of playback algorithm
I'm done with changing critical areas! Both wav and flac audio files benefit from these enhancements. Further improvements can only be achieved by creating machine specific exe files - this is cumbersome.
Please remove previous release before installing cPlay 1.0b8 .
- RE: cPlay 1.0b8 released - appleteapot 09:28:52 05/21/08 (1)
In Reply to: RE: cPlay 1.0b8 released posted by cics on May 18, 2008 at 13:00:37
Thank you very much for a such a great player!
Since playlist is not a priority item, is it possible to add a command line option to feed in sound file, so that we can write our own playlist?
Thanks!
- RE: cPlay 1.0b8 released - cics 11:24:24 05/21/08 (0)
In Reply to: RE: cPlay 1.0b8 released posted by appleteapot on May 21, 2008 at 09:28:52
Yes:
cicsPlay.exe [myCue.cue]
OR
cicsPlay.exe [myWAV.wav]
OR
cicsPlay.exe [myFLAC.flac]
- cPlay 1.0b8 released--initial sonic impressions - theob 08:16:25 05/19/08 (0)
In Reply to: RE: cPlay 1.0b8 released posted by cics on May 18, 2008 at 13:00:37
Hard to be certain since I made some other improvements in playback chain(dell e521/xp sp2-> cplay-> asio-> benchmark usb...) but it sounds like more directional changes similar to the last iteration. More dynamic, more extension top and bottom, better soundstage. So keep it coming and thank you again for all your tireless work.
- RE: cPlay 1.0b8 released - aljordan 15:30:29 05/18/08 (0)
In Reply to: RE: cPlay 1.0b8 released posted by cics on May 18, 2008 at 13:00:37
Hi cics,
Thank you for this effort. I finally tried CPlay once you started supporting FLAC files.
I have a couple of feature requests that you might consider as you have time.
1) Would it be possible to have the "File Type" selection default to "All Files", or the last chosen file type? I realize that the software is based around CUE sheets, but for those of us who rip to separate files, cue sheets are superfluous.
2) Related to the request above, for those of us who rip a CD to separate files, it would be nice to be able to open more than one file at a time. I haven't found a CUE sheet format that works with your software that points to more than one file.
Thanks again for your effort.
Alan
- ok I've got Granite Digitals coming for hdd... - theob 07:18:43 05/18/08 (2)
In Reply to: RE: cPlay - the open source high-end audio player using ASIO posted by cics on May 05, 2008 at 12:31:58
...but what should I do about my fan? Can one of the gd's power it?
- oops gd's don't fit my sata 5 pin power supply - theob 13:28:17 05/21/08 (1)
In Reply to: RE: ok I've got Granite Digitals coming for hdd... posted by theob on May 18, 2008 at 07:18:43
Are there 5 pin power supply granite digitals?
- Fixed it was just using wrong connector of the 2 on the gd - theob 07:24:04 05/23/08 (0)
In Reply to: RE: oops gd's don't fit my sata 5 pin power supply posted by theob on May 21, 2008 at 13:28:17
Externalizing the power supply for hd brings a nice lack of haze in the lower registers. For example listening to Allison Krausse I never realized how very light her voice really is. On classical there is more depth, lack of boom, more definition in the bass and yes Edward an im-rovement in low level detail across all genres. At first wasn't sure I liked it so I took it out and sounded so wrong, artificially heavy, knarly (best word I could come up with).
Thanks cics and Edward.
- Concerning memory - steppe 15:25:16 05/17/08 (1)
In Reply to: RE: cPlay - the open source high-end audio player using ASIO posted by cics on May 05, 2008 at 12:31:58
Dear Cics!
I read in your paper on cmp memory player that it is important with BIOSTAR mb (I managed to get P4 M900-M7 SE)to have only one memory bank full. To do it I have bought a 2gb Transcend DDR2 667 chip. Is this requirement still valid with your new player, or shall I go get another 2 gb chip?
ALSO, I and others, I suppose, would really appreciate the possibility to load multiple wavs in a player, just like a playlist from different folders.
Thank You for Your work.
Serge.
- RE: Concerning memory - cics 01:10:17 05/18/08 (0)
In Reply to: RE: Concerning memory posted by steppe on May 17, 2008 at 15:25:16
See this post regarding playlists.
On memory, addition of another module increases power consumption - on good quality RAM this amount is small ~1.5watts operating power (see data sheet). I would suggest adding the RAM and test. Others have reported improvements when reducing modules.
Am considering increasing cPlay's RAM load capacity to 4GB (XP's limit) - this would allow playback of hi-res 24/96 material to ~120 minutes! With flac, files would be much smaller. RAM load will take longer.
- RE: cPlay - the open source high-end audio player using ASIO - Doc Gaw 04:21:49 05/17/08 (2)
In Reply to: RE: cPlay - the open source high-end audio player using ASIO posted by cics on May 05, 2008 at 12:31:58
When playing anything other than cue files, can only load one file at a time. Have may classical recordings with movements over several files. How can one load several files for smooth operation.
- RE: cPlay - the open source high-end audio player using ASIO - cics 00:51:59 05/18/08 (1)
In Reply to: RE: cPlay - the open source high-end audio player using ASIO posted by Doc Gaw on May 17, 2008 at 04:21:49
Yes its a limitation for now.
A playlist component is needed which would enable selection of any material which cPlay uses to RAM load... This UI stuff is not yet a priority - more tuning work to follow.
- RE: cPlay - the open source high-end audio player using ASIO - appleteapot 08:57:14 05/22/08 (0)
In Reply to: RE: cPlay - the open source high-end audio player using ASIO posted by cics on May 18, 2008 at 00:51:59
Currently I associate wav extension to cPlay.bat, and within the cplay.bat I just kill the cicsplay process regardless and run it again. So in effect I just need to double click a wave file every time I want to play a song without any need to first close the running cicsplay.exe process.
Using the same method is it possible create a batch file to process a playlist (say m3u) ? For each song in the playlist, you only need to force the cicsplay to exit when it finishes playback. Then for next song you just repeat the process: kill the process (just in case) and re-run the cicsplay.exe. There may be a slight long wait between songs, but given the program's efficiency I think it will just be a short wait.
Content of cplay.bat:
echo off
start /b /wait g:\process.exe -k cicsPlay.exe
start /b /realtime G:\Progra~1\cicspl~1\cicsPlay.exe %1
exit
Would it be easier to implement? Sorry that we keep asking for the playlist feature, as I believe most of us are looking forward to use this wonderful player to replace Foobar completely.
Thanks again for an excellent wave playback program!
- cPlay 1.0b7 released (supports FLAC) - cics 02:29:11 05/17/08 (0)
In Reply to: RE: cPlay - the open source high-end audio player using ASIO posted by cics on May 05, 2008 at 12:31:58
Change Log:
- FLAC (1.2.1) supported
- Minor fixes (first pause and stop playback option)
- Minor refinements (improved random playback order)
FLAC 16 and 24 bit stereo audio files is supported. cPlay decodes FLAC file during RAM loading - you should see significant CPU load at this time. This means that cPlay incurs no additional CPU overheads during playback!
You must have sufficient RAM for the decoded FLAC file (WAV data), otherwise RAM load cannot be done. There's no sonic difference in playing FLAC or WAV files since the decoded FLAC (recreates WAV data perfectly). FLAC endoing and decoding tests show NO differences in WAV comparisons (using EAC).
Feedback from FLAC users is needed.
Please remove previous version before installing cPlay 1.0b7 .
- I guess itunes would be the perfect UI... - frennekamp@netcologne.de 15:08:23 05/16/08 (0)
In Reply to: RE: cPlay - the open source high-end audio player using ASIO posted by cics on May 05, 2008 at 12:31:58
....I know, I know...this is not a priority...I was nking: I would be wonderful to use the playback engine of cplay in itunes...for earlier versions that had been a plug-in for itunes which used the foobar-engine for playback. The same for the new version and with cplay...that would be simply a dream.
- Does cPLAY "start" CRYTOGRAPHIC SERVICES - rickmcinnis@dogwoodfabrics.com 11:45:52 05/16/08 (4)
In Reply to: RE: cPlay - the open source high-end audio player using ASIO posted by cics on May 05, 2008 at 12:31:58
Dear cics,
I had asked easlier as an aside.
When I stop CS cPLAY no longer can find my music files.
Is this correct or does it point to some kind of problem?
Please let me know,
Thanks,
Rick McInnis
- RE: Does cPLAY "start" CRYTOGRAPHIC SERVICES - cics 02:23:06 05/17/08 (2)
In Reply to: RE: Does cPLAY "start" CRYTOGRAPHIC SERVICES posted by rickmcinnis@dogwoodfabrics.com on May 16, 2008 at 11:45:52
Sounds like you using a USB HDD which requires CS.
- No, internal HDD, only - rickmcinnis@dogwoodfabrics.com 09:10:31 05/17/08 (1)
In Reply to: RE: Does cPLAY "start" CRYTOGRAPHIC SERVICES posted by cics on May 17, 2008 at 02:23:06
Dear cics,
Only USB device is the mouse.
By the way, I cannot get the system to recognize the new mouse.
Do I need to exit minlogon to get this to work?
Please advise.
THANKS,
Rick McInnis
- RE: No, internal HDD, only - cics 00:45:37 05/18/08 (0)
In Reply to: RE: No, internal HDD, only posted by rickmcinnis@dogwoodfabrics.com on May 17, 2008 at 09:10:31
Make sure CS is running, then install mouse, test & reboot.
No need to undo minlogon.
Once stable, set CS to manual and make sure you use same USB port.
- CRYPTOGRPAHIC, of course! - rickmcinnis@dogwoodfabrics.com 14:11:29 05/16/08 (0)
In Reply to: RE: Does cPLAY "start" CRYTOGRAPHIC SERVICES posted by rickmcinnis@dogwoodfabrics.com on May 16, 2008 at 11:45:52
somewhat embarrassing
- Real puzzle for me - theob 05:50:03 05/16/08 (2)
In Reply to: RE: cPlay - the open source high-end audio player using ASIO posted by cics on May 05, 2008 at 12:31:58
Yesterday I revised some of my settings to ineffect reduce graphics capability and reset some services to manual from automatic. All the actions were intended to free up some ram so I could play larger eac cue files. Before this action I was limited to about 47 minute cue files but after I was running up tp 65 minute cue files. I very pleased and began to re-rip many of my > 47 minute cd's into cue files. At the end of the day I re-booted the computer. This morning much to my horror I was back to a limit of 47 minute cue files and could not play my re-ripped files. Can you give me somw idea of what happened and how to get back to 65 minute play back? I know getting more ram is the ultimate solution but if I played 65 minute files yesterday why not today?
- RE: Real puzzle for me - Dawnrazor 14:26:09 05/16/08 (0)
In Reply to: RE: Real puzzle for me posted by theob on May 16, 2008 at 05:50:03
T,
I am guessing a bit based on your post, but it sounds like you went into the taks manager and shut down some processes. When you rebooted they all just reloaded.
I think you have to change what loads and what doesn't load.
- RE: Real puzzle for me - racerxnet 08:03:18 05/16/08 (0)
In Reply to: RE: Real puzzle for me posted by theob on May 16, 2008 at 05:50:03
You want to play for more than an hour EH! Quit toying around with this player and use Foobar, Jrivers Media, or another ASIO based player. Then you can play music all day. Sooner or later the placebo effect will wear off.
Resistance is futile!
Just having fun with you Theob as you appear to be a fanboy of this very basic software.
Have fun.
- cPlay 1.0b6 released - cics 06:50:29 05/13/08 (4)
In Reply to: RE: cPlay - the open source high-end audio player using ASIO posted by cics on May 05, 2008 at 12:31:58
Change Log:
- Runtime optimizations applied: playback algorithm exactly as 1.0b5 with 1.0b6 implemented differently at compiler level
- Some UI enhancements
This is an interesting (and unplanned) release. Apart from minor UI enhancements (RAM loading done separately from UI), the playback algorithm is exactly the same as 1.0b5. However, compiler options are used to generate greater efficiency. There's a difference!
Your feedback in comparing 1.0b5 vs 1.0b6 would be most useful.
As before, remove previous version before installing cPlay 1.0b6 .
- My impressions of #6 - rickmcinnis@dogwoodfabrics.com 07:45:20 05/14/08 (0)
In Reply to: RE: cPlay 1.0b6 released posted by cics on May 13, 2008 at 06:50:29
Dear cics,
I have not gone to the trouble of going back and forth to make my assessment. I know this should be done, and I probably would do it if I had doubts about what I am experiencing.
I agree with you, it sounds magnificent. I know I am not missing anything from what I was hearing with #5. Of course, we are not dealing with night and day differences. Though that would be neat! But, this process has taken digital sound to such heights as incremental gains are all that can be expected. I am delighted byt he sound. My favorite criterion is how late I stay up listening. I have not been getting enough sleep the last couple of weeks.
I would like to point out to the others that this appoach, cics's approach, is a system. One can cut and paste, pick and choose, but without doing what he has described, borne of much fiddleing and tweaking, one does know what this system is capable of.
In the world of super expensive transports to build a dedicated computer for this purpose is a BARGAIN. One does not fret when their $5000.00 transport cannot send emails. Build this as suggested and experience digital MUSIC reproduction. No, it is not analogue but in many ways this betters analogue. In my most unhumble opinion this has allowed digital music reproduction and analogue music reproduction to stand side by side with their own individual strengths. To really learn what is in a recording one needs to hear it in both formats. Each casts the music so that different aspects of a recording are revealed.
I have noticed that when using cPLAY that CRYPTOGRAPHIC SERVICES are started even though I have them set to MANUAL per ART OF COMPUTER TRANSPORTS paper. If I force it to close, cPLAY cannot access my music files. Is this correct?
cics, thanks for the remote mouse advice. I will get one of those on order today.
Thanks cics for this extraordinary device!
Rick McInnis
- cPlay 1.0b6 released-- initial impressions - theob 08:03:20 05/13/08 (2)
In Reply to: RE: cPlay 1.0b6 released posted by cics on May 13, 2008 at 06:50:29
Seems more open on top/wider deeper soundstage/more dynamic. At first I thought it was minor but the more I listened (4 albums-Queen Latifah Dana Owens album, Lang Lang playing Beethoven; Lincoln Portrait, Soundtrack Big Bad Love) the more I liked it and thought it was more than a subtle or small improvement.
Keep em comin' (improvements that is).
- RE: cPlay 1.0b6 released-- initial impressions - cics 03:13:01 05/14/08 (1)
In Reply to: RE: cPlay 1.0b6 released-- initial impressions posted by theob on May 13, 2008 at 08:03:20
In my reference setup , cPlay 1.0b6 is magnificent. I want no compromise in the midrange and this version does it very well (without loosing any of the strengths that prior cPlay versions offered).
- Don't disagree it's just that... - theob 06:25:51 05/14/08 (0)
In Reply to: RE: cPlay 1.0b6 released-- initial impressions posted by cics on May 14, 2008 at 03:13:01
... I describe/notice different aspects of the sound. Definitely better, keep up the great work.Btw I freed up some ram by some settings/process deletions. My question is I typically have over 50 processes running. I know you mentioned that in cmp you got it down to about 11. I use my computer for email/internet/pictures/music. What processes can I disable to maintain these applications? I was hoping for something like no more than 40. Is that achievable given my needs? How can I understand what each process does so I can make a value judgement to keep/disable?
Edits: 05/14/08
- cPlay 1.0b5 released - cics 14:05:47 05/11/08 (12)
In Reply to: RE: cPlay - the open source high-end audio player using ASIO posted by cics on May 05, 2008 at 12:31:58
Change Log:
- SRC does not clip output when out of range - implemented in cPlay. This issue affects some recordings only! It's best to use digital VC in such cases (as no clipping is required)
- More accurate available RAM calculation
- [cMP users only] Added feature to disable file browising. Use this to DISABLE file browse button. This prevents accidental indefinite waits in cMP when lsass and/or svchost is suspended. Just add line "CMP" to "cicsPlay.ini" file
- Minor UI stuff
- Restrict to running only one instance of cPlay. Trying to run cPlay more than once will result in an error message
Make sure to remove previous release before installing cPlay 1.0b5 .
Next release will take longer unless a bug fix is needed.
- SRC does not clip output when out of range ... - rickmcinnis@dogwoodfabrics.com 07:26:00 05/13/08 (10)
In Reply to: RE: cPlay 1.0b5 released posted by cics on May 11, 2008 at 14:05:47
Dear cics,
I suspect I am not the only one who does not understand what this means.
Is there an advantage to using the inherent volume control? If that is so, what would be the best average amount of attenuation? I had not intended to use the included attenuator. One assumes that "0" is best but that does not mean the assumption is correct.
Also, what would be better:
a. a remote control mouse
b. a mouse connected with a long USB cable
c. a mouse connected with a long PS2 cable
d. a keyboard connected with a long PS2 cable. I have found that my set up will not recognize an USB keyboard
By better I mean which would introduce less garbage back into the computer.
The remote control mouse would be connected as per cMP instructions with separate power supply.
What do you think?
Thanks,
Rick McInnis
PS I had re-installed the previous version of cPLAY and therefore was not getting the advantage of the phase switch.
When I got this "corrected" all worked very well. This has added much
additional pleasure to listening. Especially as I have discovered that most recordings sound best with the 180 position!
- Use USB mouse which is powered from alternate Granite Digital PS. - cics 04:18:45 05/14/08 (3)
In Reply to: RE: SRC does not clip output when out of range ... posted by rickmcinnis@dogwoodfabrics.com on May 13, 2008 at 07:26:00
This way you can use a wireless one without any compromise.
- Should one use the LOGITECH software? - rickmcinnis@dogwoodfabrics.com 10:44:34 05/16/08 (2)
In Reply to: RE: Use USB mouse which is powered from alternate Granite Digital PS. posted by cics on May 14, 2008 at 04:18:45
Dear cics,
I got my NANO mouse today.
Should one use the disk supplied?
Thanks,
Rick McInnis
- I see the answer in cMP installation manual - rickmcinnis@dogwoodfabrics.com 11:43:25 05/16/08 (1)
In Reply to: RE: Should one use the LOGITECH software? posted by rickmcinnis@dogwoodfabrics.com on May 16, 2008 at 10:44:34
Sorry,
- rusbes@gmail.com cMP installation manual - rusbes@gmail.com 02:34:02 04/22/12 (0)
In Reply to: RE: I see the answer in cMP installation manual posted by rickmcinnis@dogwoodfabrics.com on May 16, 2008 at 11:43:25
Why PCI Latency Timer (CLK) include a 128 if the sound at umenâ prior to installation of the 128 Latency Timer is well below was if I am not mistaken-36. Can 128 is needed only for sound Julia? And what better 128 or less important?
- RE: SRC does not clip output when out of range ... - Tony Lauck 09:00:01 05/13/08 (5)
In Reply to: RE: SRC does not clip output when out of range ... posted by rickmcinnis@dogwoodfabrics.com on May 13, 2008 at 07:26:00
"I suspect I am not the only one who does not understand what this means."
When using an upsampling sample rate converter it is possible for an interpolated value to be greater than either adjacent sample in the source file. (Imagine a sine wave that peaks exactly at the interpolated point.) So even if all the samples in a WAV file are below the maximum allowable value that can be stored in a file or sent to a DAC it is possible for some of the upsampled values to be above the limit. The result will be clipping distortion unless something is done about it.
What the SRC used in cPLAY is doing is to allow over valued samples. It won't be possible to send these to the DAC unchanged—they will have to be clipped if no digital volume control is used. However, if a digital volume control is used then all the samples can be reduced to avoid the necessity of clipping.
Perhaps this explanation is still confusing. However, if you use an audio editor that allows you to look at sample values and do SRC conversions you will be able to see as well as hear these operations, making it easier to understand what is going on.
Tony Lauck
"Perception, inference and authority are the valid sources of knowledge" - P.R. Sarkar
- Do you recommend a level of attenuation? - rickmcinnis@dogwoodfabrics.com 09:37:28 05/13/08 (4)
In Reply to: RE: SRC does not clip output when out of range ... posted by Tony Lauck on May 13, 2008 at 09:00:01
Dear Mr. Lauck,
Your explanation is not confusing. You confirmed what I thought I was reading but wanted to be sure.
SO, should one keep cPLAY, at say -5dB, I picked this number out of the ether. I would appreciate your, and cics's, recommendation.
Of course, it makes one think this should be automatic (pre-set level of attenuation), but with the great disaparity in relative colume levels with CD's, I can see that might not be a good idea. I have some CD's that are extremely dim in level and I assumne automatic attenuation would do them no good and then there are many recent CD's (the Paul McCartney MEMORY ALMOST FULL as an example) that is so unbelievably "loud" that I suspect it is one that would send the upsampler into this clipping mode.
So, as I try to answer my own question, should one use cPLAY's attenuator (mainly) to level out individual CD's?
Thanks for the help,
Rick McInnis
- RE: Do you recommend a level of attenuation? - Tony Lauck 18:21:15 05/13/08 (3)
In Reply to: RE: Do you recommend a level of attenuation? posted by rickmcinnis@dogwoodfabrics.com on May 13, 2008 at 09:37:28
The optimum amount of digital attenuation will be a tradeoff between reducing the resolution of the DAC (too much attenuation) and clipping when playing transients with lots of high frequencies (too little attenuation). The tradeoff will depend on details of the SRC and characteristics of each recording.
If I had a 16 bit DAC, I probably wouldn't be doing upsampling in my computer or using any digital attenuation. I would want to keep all the bits I had. However, even without doing any computer upsampling, it is possible that the DAC itself might be doing its own upsampling and if poorly designed might run into problems with digital overload. So if I heard distortion at loud peaks in either case I would try perhaps 6 db of digital attenuation to see if it fixes the problem on a particular recording.
I have a 24 bit DAC which I run at 96kHz, upsampling with foobar2000 or cPlay when necessary. (Listening tests convinced me that doing this upsampling resulted in better sound. Whether or not this will help you will depend on the design of your own DAC.) I have fixed my analog volume control at a level whereby the loudest possible recording will not clip my speakers without any digital attenuation. I do all my playback volume adjustment with a digital volume control. Normally it is set around -14 db, and seldom goes out of the range -18 to -6 db. Some over compressed recordings (the ones that might cause digital overload) are so loud that they require more attenuation, but these recordings don't sound any good anyhow.
Tony Lauck
"Perception, inference and authority are the valid sources of knowledge" - P.R. Sarkar
- RE: Do you recommend a level of attenuation? - cics 03:40:42 05/14/08 (2)
In Reply to: RE: Do you recommend a level of attenuation? posted by Tony Lauck on May 13, 2008 at 18:21:15
The only thing I would add with cPlay, is to set volume to -0.5 db to avoid clipping. Anything more is not needed.
cPlay does 64bit double precision calculations. There are many recordings/tracks that are clipped already for which SRC when interpolating, will create sample values out of range. For 32bit ASIO, setting volume to just -0.5 db would remove any upsampler related clipping - really bad recordings would need more than -0.5. At -6db, signal levels halve in amplitude!
- Worst case would be closer to -3dB - Slider 10:57:39 05/16/08 (1)
In Reply to: RE: Do you recommend a level of attenuation? posted by cics on May 14, 2008 at 03:40:42
Though -2dB is probably sufficient. That way you don't lose any resolution and assure the intersample peaks won't be distorted. Some hardware digital filters take care of this automatically when oversampling and interpolating, but I don't know about the software upsamplers.
- More Like 8 to 10 db, actually - Tony Lauck 13:47:46 05/23/08 (0)
In Reply to: RE: Worst case would be closer to -3dB posted by Slider on May 16, 2008 at 10:57:39
Using SRC (best sinc) from foobar2000 to upsample from 48 kHz to 96 Khz, I determined that the absolute worst case was 8.1 db. When I repeated this with Sony SoundForge, the value was 10.2 db. However, I don't suggest that these values are realistic except for people listening to 24 kHz square waves spliced together 180 degrees out of phase. I assume that values will be different for other software or upsampling parameters.I haven't tested yet with the newer version of SRC that is used by cics in cPlay.
Tony Lauck
"Perception, inference and authority are the valid sources of knowledge" - P.R. Sarkar
Edits: 05/23/08
- Dowloaded cPlay 1.0b5 and can now get 500 megs ... - theob 02:59:21 05/12/08 (0)
In Reply to: RE: cPlay 1.0b5 released posted by cics on May 11, 2008 at 14:05:47
... to work as a one cue file. Progress thanks!! Is there any more opportunity to get more ram to become available on the programming side ?Otherwise I need another 1 gig ram chip.
Edits: 05/12/08
- cPlay 1.0b4 released - cics 02:08:37 05/11/08 (1)
In Reply to: RE: cPlay - the open source high-end audio player using ASIO posted by cics on May 05, 2008 at 12:31:58
Change Log:
- Tuning parameters same for all soundcards
- Additional efficiencies implemented in critical areas yielding improvements and slight reduction in CPU load :-)
- Phase 180 '^' indicator between track and ellapsed time display (for Phase 0, nothing shown). Useful to confirm playback polarity
- UI enhancements:
- You can now double-click desktop icon and cPlay automatically brings up file selection window. In cPlay, you can select new material:
New settings (if done) will take effect and polarity is reset (if phase 180 was active, on requesting new material, phase is reset to 0), volume is maintained
*** WARNING ***: this feature must not be used with cMP. If svchost & lsass is suspended during playback, cPlay and your system will lock indefinitely
- Volume saved on exit. Settings no longer require volume. On intial startup (after installation), volume is -30.0db thereafter retrieved at 1db level, i.e. 0.5 bit is ignored
- Reduce font size
- Extend .wav file support for more strange structures (or chuncks)
Make sure to remove any previous release of cPlay before installing 1.0b4 .
- Everything new working as reported--nice! nt - theob 06:15:07 05/11/08 (0)
In Reply to: RE: cPlay 1.0b4 released posted by cics on May 11, 2008 at 02:08:37
Thank you
- RE: cPlay - the open source high-end audio player using ASIO - racerxnet 08:00:17 05/10/08 (0)
In Reply to: RE: cPlay - the open source high-end audio player using ASIO posted by cics on May 05, 2008 at 12:31:58
Phase inversion has been on the Bryston Pre Amps for some time. It has a subtle effect at higher volumes, but nothing dramatic. Otachan has several ASIO files for foobar to use, so it might be beneficial to try several for sonic benefits.
Foobar .083 with the Otachan .047 ASIO driver and Otachan sampler has a more natural presentation over the Cic's player. Foobar being more holographic and better seperation. The snap of a snare more pronounced, fingers sliding across the guitar strings more discernable.
Hope the next version is more user friendly than the current. I'll try all as they come out. The code should be open source so that others with more experience can add to it. Similar to the Foobar source code.
- Sonics (especially the highs) are darn near perfect - theob 07:30:33 05/10/08 (0)
In Reply to: RE: cPlay - the open source high-end audio player using ASIO posted by cics on May 05, 2008 at 12:31:58
The high frequencies are very natural, sounding very coherent with respect to the midrange. Lots of hifi playback hardware/software separates the highs from the mids which lends itself to edginess, hardness, etc etc. Not with cplay. There is a mellowness to it but not unnatural mellowness. Mellow like in real.
More and more appreciated as I listen further.
Thanks again!! Btw the phase switch works well.
- cPlay early impressions - seger 02:05:30 05/10/08 (4)
In Reply to: RE: cPlay - the open source high-end audio player using ASIO posted by cics on May 05, 2008 at 12:31:58
I’ve been comparing cPlay, Foobar, and Winamp, running from Windows, from cics realtime bat files, and in cMP. System optimized more or less according to cics, with M-Audio AP192 soundcard, latency at minimum 64 samples ( <1ms). Winamp with KS (out_ks363), buffers at minimums, no resampling. (Winamp with the otachan asio plugin is not bad, but lost out to KS with me long ago.) Foobar is with ASIO – I have never got Foobar KS to work with this soundcard.
First, music delivery:
It is easier to compare the two ASIO players – Foobar and cPlay have the same quality of sound, the ASIO effect, I suppose, and are really very similar. But cPlay is an advancement, and I find myself choosing it over Foobar every time. It has the very clean ASIO sound, but slightly less “lean” and hollow, a bit rounder, more vibrant. I’m talking about a very small increment here, but this is an achievement. cPlay is a more definite winner with me in 192K resampling over Foobar. I usually prefer Foobar without resampling, except with certain CDs that seem to need it – it is smoother with resampling but a bit sterile. cPlay gets the smoothness with less of the “sterility”, so that I find myself leaving it on all the time.
Winamp with KS has a different quality. Combined with the M-Audio card, there seems to be a synergy that keeps surprising me. This is a rawer sound, immediate, but not as clean as the ASIO players, with or without resampling. It is also more natural, sonorous, and emotionally moving for me. It is centered more in mid-range, while Foobar and cPlay seem to accentuate the treble. Using 192k resampling, cPlay gets closer to the emotional draw of Winamp than Foobar, but isn’t quite there for me, except for certain CDs that need “taming” (e.g., too heavy on bass, poor production quality) Mostly I find myself gravitating back toward Winamp. Has anybody else tried it with this sound card?
Played with the cics bat files I find Winamp and Foobar perceptibly improved compared to playing from Windows, and cPlay very slightly better as well. The big improvement comes when playing any of them in cMP – the effect is the same for all of them – exhilarating.
Secondly, the interface:
Foobar – fantastic, as you all know, so easy to customize and use. Winamp – ugly compared to Foobar, but you can streamline it if you work through all the menus and submenus. Very easy to use, except for an annoying glitch in the KS setup – the output buffer resets to maximum every time Winamp closes. It is worth resetting every time – I can hear the difference. cPlay – awkward that it only works from context menu, and playlists have to be reloaded each time it is used. Nothing in the UI can be customized except resampling.
While they all sound much better in cMP, the cumbersome interface of cMP almost ruins the enjoyment. I understand cics is prioritizing sonic improvements for now – fair enough.
- Please say some more about setting up .bat files - theob 09:58:17 05/14/08 (3)
In Reply to: RE: cPlay early impressions posted by seger on May 10, 2008 at 02:05:30
Because you were so helpful with cue files and ram replacement and since you obviously like the bat files can you tell me what to do (bat files for dummies steps if you will)to implement thisd
- RE: Please say some more about setting up .bat files - seger 13:15:09 05/14/08 (2)
In Reply to: RE: Please say some more about setting up .bat files posted by theob on May 14, 2008 at 09:58:17
You just want to play cPlay in real time, right? Cics has already put a .bat file in cPlay: right-click on the cue file for the music you want to play, select “Open with”, then “Choose program”, then click on “Browse” and navigate to C:drive/Program Files/ cics Play. In the cics Play folder is the cicsplay.bat file. Double click on that and you will go back to the “Open with” menu, and cicsplay.bat will be in the “Recommended Programs” list, highlighted. Click on the OK button and cPlay will appear and the desktop will disappear. You should notice a slight improvement in sound quality. Next time you want to run cPlay in real time just right-click the cue file/Open With and select cicsplay.bat from the menu. If you tick the box next to “Always use the selected program to open this kind of file”, then OK – then next time you want to play a .cue file with cicsplay.bat, just double-click the .cue file.
If you want even more from the music, do all of the optimizations in cics’ v03 and cMP pdf documents and play your music in cMP – stunning.
- RE: Please say some more about setting up .bat files - theob 17:33:02 05/14/08 (1)
In Reply to: RE: Please say some more about setting up .bat files posted by seger on May 14, 2008 at 13:15:09
You just want to play cPlay in real time, right? Cics has already put a .bat file in cPlay: right-click on the cue file for the music you want to play, select “Open with”,I get this far and then all I see is Cics play in the recommended programs. I don't see a choose program. I do see browse but all that it shows is cics play and when I click on that it just pops the boot up file for playing whatever I chose??? Then it plays the file.
What am I doing wrong?
Edits: 05/14/08
- RE: Please say some more about setting up .bat files - seger 00:31:36 05/15/08 (0)
In Reply to: RE: Please say some more about setting up .bat files posted by theob on May 14, 2008 at 17:33:02
With v1.0b6 you can just double click the bat file and then choose your music. I don't hear any real difference though. The bat files for foobar made a big difference. I tried adapting those for cPlay, just out of curiosity, but couldn't get it to work. Comment from cics is needed here!
- RE: cPlay - the open source high-end audio player using ASIO - fmak 05:49:20 05/09/08 (2)
In Reply to: RE: cPlay - the open source high-end audio player using ASIO posted by cics on May 05, 2008 at 12:31:58
I have tried 1.03b on 2 systems
EMu 0404 usb - works correctly but some 24/96 files won't open (see 2L.no download)
LynxAES16 - opening page incorrectly identifies card as L22 and analog/digital options incorrect - but plays fine. Won't open 2L 24/96 files.
Sound; very clean, hf very clear and incisive, lf less than Foobar or Audition 1.5 and subjectlively seems to roll off early. I have twin large subwoofers giving firm bass where it is there.
In my system severly limited by ram avaialble as I have some very large hirez files.
- RE: cPlay - the open source high-end audio player using ASIO - cics 05:59:51 05/09/08 (1)
In Reply to: RE: cPlay - the open source high-end audio player using ASIO posted by fmak on May 09, 2008 at 05:49:20
Thanks for the feedback. Your high-res files are 24/96. How much RAM do you have installed?
I still have some strict wav file validations in cPlay which I'll remove in b4. The identification information and channels you see is supplied by Lynx's driver. Only Lynx can change this.
- RE: cPlay - the open source high-end audio player using ASIO - fmak 06:39:41 05/09/08 (0)
In Reply to: RE: cPlay - the open source high-end audio player using ASIO posted by cics on May 09, 2008 at 05:59:51
No, 24/176.4 and 192k. Some are 3 GB from continously ripping 192k DVDVs.
I have 2 GB ram.
- RE: cPlay - the open source high-end audio player using ASIO - Dawnrazor 19:35:25 05/08/08 (6)
In Reply to: RE: cPlay - the open source high-end audio player using ASIO posted by cics on May 05, 2008 at 12:31:58
Hey Cics,
Thanks for doing this!!
I haven't yet put this on my main hifi rig, so I can't exactly comment on the sound yet.
Being a volume user, I can tell you that having to click on it repeatedly is not a good thing. COuld you make it ramp?
Also, it kind of works with Mp3toys. I know you didn't design it to do so, but if you could make a few changes, it might work with mp3 toys and give yourself a way to address the critics that have issue with the interface.
What happens is that mp3 toys will send a song and the cplayer will play it (if you press play...if the playback could be automatic then it would work). But when another song is sent from Mp3toys, then another instance of Cplayer is opened up, and the ASIO gets screwed up. If you would allow only one instance than I think it would work fine.
I'll let you know when I get a chance to listen critically.
- Try using cPlay's keyboard mappings - cics 06:19:51 05/09/08 (5)
In Reply to: RE: cPlay - the open source high-end audio player using ASIO posted by Dawnrazor on May 08, 2008 at 19:35:25
cPlay keys are:
- '[' / ']' for Previous / Next
- ';' for phase 0/180
- '.' for stop
- shift+Home for volume up
- shift+End for volume down
- Enter to play highlighted track
- Alt+F4 or Esc for exit
I'm not familiar with Mp3Toys but it should offer keyboard mappings. In my setup, I rarely ever click - just use mouse wheel button: push+UP/Down gives volume Up/Down (with cMP). My other setup, I use the 7" touch screen - it's addictive!UI changes will follow later - although I plan to keep it minimal. I can't handle all that visual noise & clutter.
cPlay is designed to support multiple instances - allowing for output from multiple cards. Yes, on same soundcard you'll get an error. Could you exit cPlay then revert to Mp3Toys and play something else? cPlay starts playback immediately (just make sure you set Diagnostics to No in settings).
Edits: 05/09/08
- I must be missing something here. - Dawnrazor 13:37:55 05/09/08 (4)
In Reply to: RE: Try using cPlay's keyboard mappings posted by cics on May 09, 2008 at 06:19:51
I put this on my main rig. I don't have the cue file issues I was having and the player starts right up.But here is the issue and it is the same one with mp3toys:
When I am listening to an album and i go into explorer and select another cue sheet, it opens ANOTHER instance of cplay. That 2nd instance will never work with the Lynx single client ASIO drivers....because the first instance is using them! SO I have to exit the first and then select the 2nd cue file.
How are you supposed to change cue files without exiting every time?? Am I missing something.
Now if it would just work with one instance then the current songs would get replaced, then it would work with the way I use it and with Mp3toys which sends wave files to cplay, much like me selecting them in explorer.
Also, my "keyboard" on my wireless touch screen doesn't have a shift lock button. And since I have to exit the player everytime, I have to hit the volume at least 32 times to get back to 0 ( I am counting the times I have to hit the shift key as well). Could you just make it remember its volume when it exits or give a 0 option?
ALso, I don't think my 1ghz PC is fast enough for 192. It had all kinds of artifacts. 96 works ok
I'll hopefully get some time for critical listening and let you know.
Thanks.
Edits: 05/09/08
- RE: I must be missing something here. - cics 15:51:51 05/09/08 (3)
In Reply to: RE: I must be missing something here. posted by Dawnrazor on May 09, 2008 at 13:37:55
Ok, for each play request, you must restart cPlay - that is after each play, you must exit cPlay before playing another. If you don't exit cPlay, then you'll have another cPlay instance running (which would give an error as ASIO driver is not available).
For VC, choose 'None' in cPlay's settings - this will give you 0.0db volume. Saving volume is something I'll look into (I've had a similar request).
- RE: I must be missing something here. - Dawnrazor 16:36:49 05/09/08 (2)
In Reply to: RE: I must be missing something here. posted by cics on May 09, 2008 at 15:51:51
THanks C,
THis is one of those times I wished I was wrong.
Ok, for each play request, you must restart cPlay - that is after each play, you must exit cPlay before playing another. If you don't exit cPlay, then you'll have another cPlay instance running (which would give an error as ASIO driver is not available).
That doesn't strike you as odd?? I certainly cant think of any other player that operates like that.
What would be the problem with just erasing the existing cue??
- RE: I must be missing something here. - cics 16:53:33 05/09/08 (1)
In Reply to: RE: I must be missing something here. posted by Dawnrazor on May 09, 2008 at 16:36:49
It's a deliberate design choice to allow for Windows optimizations (lsass suspension) when using cMP. Lsass (Windows security subsystem) does create audible interference (in my reference setup) and foobar's design will not allow for lsass to be suspended.
I could make cPlay start up once only and thereafter process play requests - I'll give it some thought as to how I can have both options user selectable.
- RE: I must be missing something here. - Dawnrazor 21:21:18 05/09/08 (0)
In Reply to: RE: I must be missing something here. posted by cics on May 09, 2008 at 16:53:33
Well, if you can do it as user selectable, that would be great. A very big thanks if you can accomplish this.
- RE: cPlay - the open source high-end audio player using ASIO - racerxnet 16:37:02 05/08/08 (1)
In Reply to: RE: cPlay - the open source high-end audio player using ASIO posted by cics on May 05, 2008 at 12:31:58
Loaded the player, tried a few songs and was not impressed. No drag and drop, no file directory within player to load songs, ASIO playback was no better than Foobar 083. Ochtan sampler or SRC @ 96000.
Had audio playback issues where as foobar did not.
All in all a (good effort) at best so far. Minimal approach with minimal results compared to foobar. The user interface needs a lot of work.
- What soundcard are you using? - cics 05:51:24 05/09/08 (0)
In Reply to: RE: cPlay - the open source high-end audio player using ASIO posted by racerxnet on May 08, 2008 at 16:37:02
This is very important. In cPlay's design, it caters for 2 broad soundcard categories (at driver level) with each having slightly different parameters. Your feedback will help identify which parameters are working better. It would also help if you could post the initial diagnostic window.
If you have an EMU/Creative soundcard, there may be a chance for playback issues - test with ASIO4ALL to see if same occurs.
UI disappoints (for now/some) until tuning stage and flac is done... cPlay is designed to complement cMP.
- Is there a way to decrease font size ? - Mutliblitz 12:07:23 05/08/08 (1)
In Reply to: RE: cPlay - the open source high-end audio player using ASIO posted by cics on May 05, 2008 at 12:31:58
...I have here a 12" touch-screen and the font is too big for me and I would love to see all tracks on one page with no need to scroll...
- Not yet... will consider later. - cics 13:22:38 05/08/08 (0)
In Reply to: RE: Is there a way to decrease font size ? posted by Mutliblitz on May 08, 2008 at 12:07:23
UI is a very low priority for now. Next on the list is flac.
- Ok you have convinced me to try... - theob 10:05:43 05/08/08 (12)
In Reply to: RE: cPlay - the open source high-end audio player using ASIO posted by cics on May 05, 2008 at 12:31:58
...I currently use j river with asio4all. Should I just download cplay and will that bring asio et all along with it? Another question: I have utilized some of your ideas in CMP but not all, will cplay still be better than where I'm at? Also in the User Manual:
*************************************************************************
Main screen offers essentials only. Large buttons and fonts makes for excellent touch screen friendliness and operating from a distance. cPlay's desktop icon is used for initial setup. Thereafter, play audio as follows:
Select .cue or .wav file from File Explorer (double-click on .cue starts cPlay)
Select 'Open With' (right-click on file to be played)
Select cicsPlay
Playing audio with RealTime priority:
Select .cue or .wav file from File Explorer
Select 'Open With' (right-click on file to be played)
Locate and select 'cicsPlay.bat' where cPlay was installed (default is 'c:\program files\cics Play'). Note: If you did NOT use the default install folder, edit cicsPlay.bat and correct the path. Hint: use command 'dir *.* /x' to identify folder's DOS shortname.
**************************************************************************
What does '...Select .cue or .wav file from File Explorer...' mean? Can't I just go to my music, select the files and then open with cplay?
Also what does
***********************************************************************
'...Locate and select 'cicsPlay.bat' where cPlay was installed (default is 'c:\program files\cics Play'). Note: If you did NOT use the default install folder, edit cicsPlay.bat and correct the path. Hint: use command 'dir *.* /x' to identify folder's DOS shortname...'
***********************************************************************
mean? Just say a little more about this to clarify. I'm not sure what to do here.
- RE: Ok you have convinced me to try... - cics 13:31:15 05/08/08 (11)
In Reply to: RE: Ok you have convinced me to try... posted by theob on May 08, 2008 at 10:05:43
...I currently use j river with asio4all. Should I just download cplay and will that bring asio et all along with it? Another question: I have utilized some of your ideas in CMP but not all, will cplay still be better than where I'm at?
Yes - cPlay will work with your current ASIO4ALL setup.
Yes - cPlay will work with whatever optimizations you've done from cMP.
What does '...Select .cue or .wav file from File Explorer...' mean? Can't I just go to my music, select the files and then open with cplay?
Yes - it means exactly that.
'...Locate and select 'cicsPlay.bat' where cPlay was installed (default is 'c:\program files\cics Play'). Note: If you did NOT use the default install folder, edit cicsPlay.bat and correct the path. Hint: use command 'dir *.* /x' to identify folder's DOS shortname...'
Use cicsPlay.bat if you want to run cPlay in RealTime priority (i.e. Windows gives it preferential treatment). If you're using cMP's operating sofware, this is done for you.
- 1st attempt unsuccessful - theob 17:37:07 05/08/08 (10)
In Reply to: RE: Ok you have convinced me to try... posted by cics on May 08, 2008 at 13:31:15
Here are my diagnostics
ASIO Driver initialized.
DSP starting...
DSP Buffer (10240 samples or 232ms/ 4.3/ 172.3) allocation successful
ASIO Driver starting...
Input 1 [ Not Connected 1 ] Active (0) type (32bits)
Input 2 [ Not Connected 2 ] Active (0) type (32bits)
Output 1 [ Not Connected 1 ] Active (1) type (32bits) > Left Channel
Output 2 [ Not Connected 2 ] Active (1) type (32bits) > Right Channel
ASIO started successfully.
Processing CD [C:\Documents and Settings\ted betley\My Documents\My Music\Aaron Copland\Lincoln Portrait\01 Lincoln Portrait.wav]...
Wav file size = 155MB
Wav file structure is NOT supported.
Genre = Unknown
Artist = C:\Documents and Settings\ted betley\My Documents\My Music\Aaron Copland\Lincoln Portrait\01 Lincoln Portrait.wav
Album = WAV
Wav file = C:\Documents and Settings\ted betley\My Documents\My Music\Aaron Copland\Lincoln Portrait\01 Lincoln Portrait.wav
- RE: 2nd attempt unsuccessful - theob 19:24:47 05/08/08 (9)
In Reply to: RE: 1st attempt unsuccessful posted by theob on May 08, 2008 at 17:37:07
Here are diagnostics
SIO Driver Initializing...
Name (ASIO4ALL v2)
Player ASIO version (2)
Driver ASIO version (2)
Message (No ASIO Driver Error)
Channels (inputs: 2, outputs: 2)
Buffer details (min: 64, max: 2048, preferred: 512, granularity: 8)
Sample rate ( 44100)
Output Ready? Not supported
Preparing buffers... successful
Latencies (input: 0, output: 764)
ASIO Driver initialized.
DSP starting...
DSP Buffer (10240 samples or 232ms/ 4.3/ 86.1) allocation successful
ASIO Driver starting...
Input 1 [ Not Connected 1 ] Active (0) type (32bits)
Input 2 [ Not Connected 2 ] Active (0) type (32bits)
Output 1 [ Benchmark 1.0 1 ] Active (1) type (32bits) > Left Channel
Output 2 [ Benchmark 1.0 2 ] Active (1) type (32bits) > Right Channel
ASIO started successfully.
Processing CD [C:\Documents and Settings\ted betley\My Documents\My Music\Aaron Copland\Lincoln Portrait\01 Lincoln Portrait.wav]...
Wav file size = 155MB
Wav file structure is NOT supported.
Genre = Unknown
Artist = C:\Documents and Settings\ted betley\My Documents\My Music\Aaron Copland\Lincoln Portrait\01 Lincoln Portrait.wav
Album = WAV
Wav file = C:\Documents and Settings\ted betley\My Documents\My Music\Aaron Copland\Lincoln Portrait\01 Lincoln Portrait.wav
I got out puts to map to my benchmark but cannot get inputs from asio4all into cicsplay. What am I doing wrong?
- Good progress! cPlay and ASIO is setup correctly - cics 05:41:48 05/09/08 (8)
In Reply to: RE: 2nd attempt unsuccessful posted by theob on May 08, 2008 at 19:24:47
The important message is "Wav file structure is NOT supported." - cPlay does strict validation on wav files (which I'll loosen later).
Try another wav file - preferably one you ripped using EAC - it's wav file will work.
- RE: Good progress! cPlay and ASIO is setup correctly - theob 05:58:08 05/09/08 (7)
In Reply to: RE: Good progress! cPlay and ASIO is setup correctly posted by cics on May 09, 2008 at 05:41:48
Thank you for your response. Unfortunately all my files were ripped with wmp11. I'll get eac and try that (I've been wanting to anyway, this gives me a good reason).
- RE: Good progress! cPlay and ASIO is setup correctly - theob 06:37:46 05/09/08 (6)
In Reply to: RE: Good progress! cPlay and ASIO is setup correctly posted by theob on May 09, 2008 at 05:58:08
Just got EAC downloaded (still trying to learn how to optimize eac rip) and successfully ran Cicsplay versus j river (both with eac rip). Conclusion I have a new player---Cicsplay is much clearer, more extended in highs and lows, much more dynamic. I'm loving it. All of this sonic excellence and a polarity switch to boot!
Thank you so much!!!!
To the extent that you will allow wav files ripped through other than EAC in future realeases will certainly be appreciated. I have over 120 gigs like that.Now I have to learn how to best run EAC.
- Listened to 1st classical piece thru Cicsplay - theob 09:19:31 05/09/08 (5)
In Reply to: RE: Good progress! cPlay and ASIO is setup correctly posted by theob on May 09, 2008 at 06:37:46
Imaging and dynamics are thrilling. I can't overstate how good this sounds. This is a real breakthrough!!!
But I need eac counseling. I'm not getting automatic cd data lookup in EAC so as a result I'm getting the files ripped and placed in weird places. Do I have to register for freedb or because I was previously using wmp11 am I being blocked or do I have the wrong setup in tools or whatever (probably)?
Where can I get help for eac?
- Should I still set preferences to cache? - theob 12:21:52 05/09/08 (4)
In Reply to: RE: Listened to 1st classical piece thru Cicsplay posted by theob on May 09, 2008 at 09:19:31
Or is cicsplay immune to cache/hard drive?
- RE: Should I still set preferences to cache? - cics 15:59:06 05/09/08 (3)
In Reply to: RE: Should I still set preferences to cache? posted by theob on May 09, 2008 at 12:21:52
Keep your cMP settings that you've made - there's no need to change anything. I run 2 cMP's (in separate systems) each fully configured with cPlay.
I can't help you on those EAC features (never used them).
That's great feedback. Version 1.0b4 will have wider wav file acceptability and should work with wmp rips.
- RE: Should I still set preferences to cache? - theob 17:00:05 05/09/08 (2)
In Reply to: RE: Should I still set preferences to cache? posted by cics on May 09, 2008 at 15:59:06
I had a hard down today. Got a system error and my usb signal went dead. Cplay was still playing but because of the system error I was not getting any signal to my dac. Microsoft (automated message) said that it was due to a new device (but there wasn't any) or new software which I do have (Cplay) so I removed it from my computer and went back to j river until I could hear from you. J river is working ok so my usb signal is back.
Any ideas?
- RE: Should I still set preferences to cache? - theob 17:29:59 05/09/08 (1)
In Reply to: RE: Should I still set preferences to cache? posted by theob on May 09, 2008 at 17:00:05
here is my error message
Follow these steps to solve the problem with a device driver
You received this message because a device driver installed on your computer caused Windows to stop unexpectedly. This type of error is referred to as a "stop error." A stop error requires you to restart your computer.
Troubleshooting
--------------------------------------------------------------------------------
Depending on which situation is applicable to you, do one of the following:
If this problem occurred after you installed a new hardware device on your computer, the problem might be caused by the device driver. Use the Dell Driver Reset Tool or uninstall the driver.
How do I disable or uninstall a device driver?
Click Start, and then click Control Panel. If you are using Classic View, click Switch to Category View.
Click Performance and Maintenance, and then click System.
Click the Hardware tab, and then click Device Manager.
Click the plus sign (+) next to the faulting device. You should now see the device listed.
Right-click the device, and then click Disable or Uninstall.
If this problem occurred after you added new software, the software might have added a driver that caused the problem. Try removing the software.
How do I remove a program?
Click Start, click Control Panel, and then click Add or Remove Programs.
Click Change or Remove Programs, click the program you want to remove, and then click Change/Remove or Remove.
If you don't know the specific driver or software, try performing a System Restore.
Go online to check for updated drivers on the Microsoft Update website:
Microsoft Update
For information about your support options, go online to the Support.Dell.Com website.
I got it again while running j river so I fairly certain its not cplay but what the heck is it any of you experts?
- I think I solved my 'stop' problem...associated with eac... - theob 06:02:22 05/10/08 (0)
In Reply to: RE: Should I still set preferences to cache? posted by theob on May 09, 2008 at 17:29:59
...but how can I play a whole album in Cicsplay with one mouse click? I'm setting up cues in eac but I keep having to play each track separately.
- Should we delete FOOBAR? - rickmcinnis@dogwoodfabrics.com 09:23:03 05/08/08 (5)
In Reply to: RE: cPlay - the open source high-end audio player using ASIO posted by cics on May 05, 2008 at 12:31:58
Dear cics,
Is there any reason to retain FOOBAR now?
Please advise,
Rick McInnis
- RE: Should we delete FOOBAR? - racerxnet 16:41:30 05/08/08 (3)
In Reply to: RE: Should we delete FOOBAR? posted by rickmcinnis@dogwoodfabrics.com on May 08, 2008 at 09:23:03
Foobar will not interfere with the player. If you do not like tagged files for an album list and want to play with the very basics of a player then delete foobar.
This thing needs some serious polishing to come close to Foobar basics. Sound quality is subjective and I felt there was no audible gains to be had running ASIO to my system.
- Hey Racer... - Presto 18:41:08 05/08/08 (2)
In Reply to: RE: Should we delete FOOBAR? posted by racerxnet on May 08, 2008 at 16:41:30
"Sound quality is subjective and I felt there was no audible gains to be had running ASIO to my system."
When you said this, what were you comparing to ASIO? I'm not hunting for a debate - I'm collecting data on what people think about different methods.
Myself, I find kernel streaming and ASIO to sound pretty much the same. Directsound v2.0 sounds pretty good too.
Then again, I have my system set up so that no other app can ever stream ANY audio while my player is in use. (I select "no sounds" for system sounds, and use the onboard sound codec for all default devices, and check the "use only default devices" box).
Cheers,
Presto
- RE: Hey Racer... - racerxnet 19:02:00 05/08/08 (1)
In Reply to: RE: Hey Racer... posted by Presto on May 08, 2008 at 18:41:08
Hey Presto,
I tried all different plugins for Foobar and winamp. SRC, KS, otachan 047 ASIO, otachan sample rate converter. I am not sure I hear dramatic differences on my 2 systems. But then again, my hearing is shot from 25 years in the printing industry without much protection. Just thought I would try the player for the heck of it and compare.
As a programmer I wonder if he can let the user select a playlist and allow the files to be dumped after use and load as ram space permits. This eliminates the ram size limitation and one can cue as many songs as one likes. This can be done as I can program in many languages.
My system for testing is 4 Bryston 7BSST amps, Bryston BP26 pre, Infinity RS1B, Wavelenght brick, or Tri Vista 21 DAC.
Foobar 083 on XP
C2 duo E6320
- RE: Hey Racer... - Riboge 12:10:49 05/14/08 (0)
In Reply to: RE: Hey Racer... posted by racerxnet on May 08, 2008 at 19:02:00
If you know how to program what you describe about loading from a playlist then dumping from ram to make room for more on the list, etc, which I surely don't, I entreat you to work on this to enhance this open source player for all. The creator of the player tells me he is busy developing the flac capability and would welcome someone else developing the multi-file playback capability.
- Other players necessary until we go to 64 bit PCs - Tony Lauck 09:43:38 05/08/08 (0)
In Reply to: RE: Should we delete FOOBAR? posted by rickmcinnis@dogwoodfabrics.com on May 08, 2008 at 09:23:03
cPlay is a memory player. As such, it can only play music that fits in RAM. Even if I were to expand my RAM memory to the maximum addressable by my 32 bit processor I would still not have enough memory to hold some of my hi-res albums. Until our PCs and cPlay go to 64 bit operating systems other players will still be necessary at least for some music.
For example, the Linn studio master St. Matthew Passion occupies 4.76 GB after FLAC files has been expanded to WAV (24 bits at 88.2 kHz). Many of the 101 tracks play continuously track to track, making the player's job harder. Foobar2000 plays the separate files correctly. cPlay does not, because it has to load a WAV file into memory before it can play it.
Tony Lauck
"Perception, inference and authority are the valid sources of knowledge" - P.R. Sarkar
- cPlay 1.0b3 released - cics 12:39:49 05/07/08 (3)
In Reply to: RE: cPlay - the open source high-end audio player using ASIO posted by cics on May 05, 2008 at 12:31:58
Change Log:
- Phase 0/180 setting using keyboard ';' key or mouse button 4 (cMP). Previous use for ';' was Stop. Phase defaults to 0 on start. Stop key now set to '.'
- Playback order option added: 'Repeat', 'Stop' or 'Random'. Changes in this option take immediate effect. That is, if set to random, cPlay selects tracks randomly (including next / previous)
- Minor refinements
- Fix for NCH tone files
- Fix for ASIO output format support message
Presto - NCH now works and you should be able to do sweep tests
Rick / Tony - Can you test Phase 0 / 180?
Fred - I suspect your problem is also related to the NCH fix. Let me know if 1.0b3 fixes it
If you have a prior version installed - remove it before installing 1.0b3. Note in 'Add or Remove Programs' previous versions had same 'cPlay 1.0b1' name - its a labeling error.
- RE: cPlay 1.0b3 released - Multiblitz 11:59:59 05/08/08 (1)
In Reply to: RE: cPlay 1.0b3 released posted by cics on May 07, 2008 at 12:39:49
Ok, I have just installed it...not listened yet to it, but what is very promising:
Your player is doing real multithreadening, while foobar / SRC in foobar is not.
Do you remenber my comment, that SRC sound poorer than no upsampling in foobar ?
A theory one could have is: I am using a e2180 and I have power-mamgt on...so it really runs with no oversampling at 1200mhz, with oversampling it switches to higher frequencies !
...in foobar. with your player it stays at 1200 mhz even if I upsample to 192khz. So, tommorow I will do my listening and report back...
- RE: cPlay 1.0b3 released - Presto 18:43:06 05/08/08 (0)
In Reply to: RE: cPlay 1.0b3 released posted by Multiblitz on May 08, 2008 at 11:59:59
"Do you remenber my comment, that SRC sound poorer than no upsampling in foobar ?"
I have always felt this way. I always prefer Foobar with no SRC - even the blessed secret rabbit! ;)
Cheers,
Presto
- Right on, Cics! - Presto 15:09:46 05/07/08 (0)
In Reply to: RE: cPlay 1.0b3 released posted by cics on May 07, 2008 at 12:39:49
Cics:
Hey thanks. No biggies on the NCH thing - I just used a sweep from a different sweep generator. But hey - it's a nice perk for those who bought and paid for NCH.
Can't WAIT to get home and hit that ";" button during playback and see if we can hear some absolute polarity changes! ;)
Thanks again! Nice work!
Cheers,
Presto
- RE: cPlay - the open source high-end audio player using ASIO - Pedja 18:56:19 05/06/08 (12)
In Reply to: RE: cPlay - the open source high-end audio player using ASIO posted by cics on May 05, 2008 at 12:31:58
Cics,
I have read the PDFs you've been putting together before and admittedly I've never liked them at all: way too many arbitrary statements.
But this software is interesting effort. Putting its interface aside (frankly, it is below any criterion anyhow) this will be very welcome contribution which will help understanding how particular use of computer resources may change the things soundwise.
And, comparing to the usual Foobar2000, it sounds... mellow.
(And before anyone comes with such an explanation, let me warn you on time: it has nothing to do with jitter. Yes, I checked it.)
Keep up the good work.
Pedja
- "...on time: it has nothing to do with jitter." If not Jitter than what? - cics 04:51:16 05/07/08 (11)
In Reply to: RE: cPlay - the open source high-end audio player using ASIO posted by Pedja on May 06, 2008 at 18:56:19
.
- RE: "...on time: it has nothing to do with jitter." If not Jitter than what? - Pedja 02:19:53 05/08/08 (10)
In Reply to: RE: "...on time: it has nothing to do with jitter." If not Jitter than what? posted by cics on May 07, 2008 at 04:51:16
That's something we have to find out yet. Apparently the way the software engages computer resources does implicates particular subjective sonic effects. And to my knowledge your software is unique, placing the whole file into memory.
As for the jitter, I've never found software makes any difference in this regard, unless one screws up something, of course.
Different PCs may and do have different jitter performance but it is hardware related.
Pedja
- RE: "...on time: it has nothing to do with jitter." If not Jitter than what? - drrd 15:32:26 05/08/08 (1)
In Reply to: RE: "...on time: it has nothing to do with jitter." If not Jitter than what? posted by Pedja on May 08, 2008 at 02:19:53
foobar can play from RAM (buffer whole files), check the advanced settings.
And this is absolutely about jitter. If you record a file played via cplay asio and the same via foobar asio you will get exactly the same bit perfect results. if they sound different it isn't because of altered bits it's because of jitter.
- RE: "...on time: it has nothing to do with jitter." If not Jitter than what? - Pedja 15:46:41 05/08/08 (0)
In Reply to: RE: "...on time: it has nothing to do with jitter." If not Jitter than what? posted by drrd on May 08, 2008 at 15:32:26
Great. I am waiting forward to the jitter measurements. I expect them to be different than those I made.
- RE: "...on time: it has nothing to do with jitter." If not Jitter than what? - cics 14:38:57 05/08/08 (5)
In Reply to: RE: "...on time: it has nothing to do with jitter." If not Jitter than what? posted by Pedja on May 08, 2008 at 02:19:53
That's something we have to find out yet. Apparently the way the software engages computer resources does implicates particular subjective sonic effects
Software is an abstract level (esp when using high level languages) and gives the perception that its all logic without physical consequences. As you get lower (some stuff in C, assembler and firmware), software becomes very physical (i.e. lots of electrical signals - data signals, clocks, etc). Eventually you get to this conclusion: Digital is Analogue ! That is, there's no such thing as digital physics, just physics.
This distortion (i.e. lots of electrical signals which depend explicitly on the software thats running) from a pure audio viewpoint is jitter . This can be measured as the soundcard's XO is destabilized. Jitter defines such destabilization as either random or deterministic - software impacts both.
- RE: "...on time: it has nothing to do with jitter." If not Jitter than what? - Pedja 15:16:10 05/08/08 (4)
In Reply to: RE: "...on time: it has nothing to do with jitter." If not Jitter than what? posted by cics on May 08, 2008 at 14:38:57
Digital is digital, and digital is not analog, though some problems associated to the digital electronics are indeed purely analog in nature, and jitter is exactly such a problem.
If you read my post better you could see that I assume the different players sound different, assuming also the reason behind this is exactly the different way they control/use computer resources.
Yet, the jitter performance is not a matter of taste or personal opinion - the jitter can be and is normally measured. So if you claim the thing X has different jitter performance than thing Y, you should post some measurement to support that, shouldn't you? Especially when making sweeping statement like "This distortion (i.e. lots of electrical signals which depend explicitly on the software thats running) from a pure audio viewpoint is jitter. This can be measured as the soundcard's XO is destabilized ." (bold is mine) And actually, if you read my post better, you might read that your software and Foobar2000, and any other any correct player for that matter, have the same jitter performance and I've said this because I've actually measured them . I could post the plots but there was no reason to do this: they are the same.
Pedja
- Jitter Measurements - cics 05:33:19 05/09/08 (3)
In Reply to: RE: "...on time: it has nothing to do with jitter." If not Jitter than what? posted by Pedja on May 08, 2008 at 15:16:10
See this thread: Jitter Research, Analysis & Measurement which I posted a while back. After reading the main post read this follow-up .
What you'll see is how the jitter measurement (Jpp) differs significantly. The only change in the entire equation is software with all else the same (i.e. hardware, audio chain and mains supply). The important thing here is that all software matters across all levels (firmware/BIOS, operating system, drivers, audio player etc.).
Digital is digital, and digital is not analog, ...
That's a bad assumption.
...have the same jitter performance and I've said this because I've actually measured them . I could post the plots but there was no reason to do this: they are the same.
If you don't mind, post one of the 2 plots. Also, the level of optimizations applied on the computer transport you measured.
- RE: Jitter Measurements - Pedja 14:06:50 05/09/08 (2)
In Reply to: RE: Jitter Measurements posted by cics on May 09, 2008 at 05:33:19
> What you'll see is how the jitter measurement (Jpp) differs significantly.
> The only change in the entire equation is software with all else the
> same (i.e. hardware, audio chain and mains supply). The important thing
> here is that all software matters across all levels
> (firmware/BIOS, operating system, drivers, audio player etc.).
Well, important thing ...... the above is simply incorrect statement. As long as software is any good, it doesn’t change a squat about jitter. I didn’t say it is not possible to produce such artifacts by software, I said that one has to screw up something to do that, and this apparently was the case. Everything you have posted was the case where your software screwed something up in this regard. Yet you claim that “all software matters across all levels”.
As for the plots you’ve posted... they generally look too good to be true for PCI based S/PDIF source, and this particular source is most probably driven by PLL clock and with performance claimed by manufacturer “ < 1 ns”. Having a PCI S/PDIF source with shown performance would be much more important news than anything else we are now talking about. So, my question is what the PLL corner frequency of Scarlatti DAC is?
Also, for the record... when making the jitter measurements of analog output, it is important to use the frequency that is a divisor of sampling frequency, since only this way you avoid quantization noise, and only that way you comply with proposed standard. Your FFT windowing was also apparently inappropriate. Or, if it was, what was the cause of the skirt around 7kHz?
> That's a bad assumption.
No, it is not assumption at all, it is rather a matter of fact. One has to learn to tell one from the other though. In that regard, the protocols, data processing and alike belong the digital, whereas voltage, current and alike belong the analog world. Connection between these two worlds is possible by so called sampling and that’s where the jitter we are talking about belongs. Anyhow, the “digital is analog”... no way.
> If you don't mind, post one of the 2 plots. Also, the level of optimizations applied on the computer transport you measured.
I actually posted plenty of plots showing performance of PC based audio systems during a couple of last years to the diyhifi.org, and recently you may find a sort of summary on my own site. I am not familiar with the term “the level of optimizations” though.
Pedja
- RE: Jitter Measurements - cics 16:39:43 05/09/08 (1)
In Reply to: RE: Jitter Measurements posted by Pedja on May 09, 2008 at 14:06:50
The changes that yielded those results had nothing todo with 'screwed up' software. In the follow-up post you'll understand that I replaced the Windows default shell (Explorer) with cMP and as explained in the pdf documents Windows Explorer is not good for audio playback. Also, Juli@'s mixer process was removed. These are all valid software changes.
The fact those measurements reflect quite clearly jitter improvements is unquestionable. I was able to repeat them across different input frequencies AND over different days.
dCS don't provide corner frequency details - perhaps you could ask them? As to RME's spec'd 1ns jitter - its a guide . This is discussed in the thread.
Any analogue frequency will be afflicted with jitter distortion. The choice of recording sampling frequency and bit depth (at 24/96) makes quantization noise an unimportant factor (at 24 bits this noise is below -150db). The plots (in dbrms) are excellent and given Scarlatti DAC's cost I would expect them to be so!
You not getting the point on "digital is analog" - what I'm saying is digital processing has real physical consequences by way of voltages, currents etc. and as a result, its application (i.e. the software & firmware) needs similar due care taken as for analogue. Things like efficiency, low power consumption, EMI/RFI, etc. all matters - this is analogue.
You claimed you have measured the 2 (which are the same). Why not present it? Looking at other plots is not the point. As to the level of optimization, I'm referring to specific details of the computer that was measured (esp those items that are changed from its default setting).
- RE: Jitter Measurements - Pedja 17:51:24 05/09/08 (0)
In Reply to: RE: Jitter Measurements posted by cics on May 09, 2008 at 16:39:43
What did you want to achieve performing the measurements over different days? You could perform them over different years and they'd be the same flawed – and that because your setup is simply not good for this purpose. Firstly, the S/PDIF output you used most probably had several orders of magnitude higher jitter than anything shown in your graphs. On the other side, the jitter shown in the graphs is low most probably because the D/A converter you used had strong jitter attenuation. You can not perform measurements you wanted to perform using a setup like that. I'm not affiliated with dCS so I have no better chances to get their answer than you do. And it is you who should be interested in this rather than me, since it was you who had to know what you were actually using to perform measurements.
As for the difference you have found, you have to check again things to see what actually happened. For start I'd guess that it had something with mixer, and yes, I'd normally assume that any mixer had to be removed from the process anyhow.
I didn't claim that I measured two, I did said that I measured jitter performance of your s/w versus jitter performance of Foobar2000, and I said that the performance was the same. In fact I measured about any software I could come across. I was told that there will be night and day differences in jitter, this software, that software, this setting, that setting, this ASIO, that ASIO, not ASIO, etc. Nothing, and I stress, nothing of this, once you set the PC to output data directly and with no resampling, had any influence on given PC jitter performance. Particular jitter performance is simply associated to the hardware used, and differences in jitter performance between different hardware are relatively easily visible. As for the rest, including the graphs, please refer to the article I directed you to in my previous post.
Pedja
- RE: "...on time: it has nothing to do with jitter." If not Jitter than what? - multiblitz 10:56:32 05/08/08 (1)
In Reply to: RE: "...on time: it has nothing to do with jitter." If not Jitter than what? posted by Pedja on May 08, 2008 at 02:19:53
...so this means we need to compare the sound of motherboards ?
- RE: "...on time: it has nothing to do with jitter." If not Jitter than what? - Pedja 14:43:30 05/08/08 (0)
In Reply to: RE: "...on time: it has nothing to do with jitter." If not Jitter than what? posted by multiblitz on May 08, 2008 at 10:56:32
Yes. And not only the sound but also entirely technical performance as jitter and noise levels.
- cPlay 1.0b2 released - cics 21:48:20 05/05/08 (17)
In Reply to: RE: cPlay - the open source high-end audio player using ASIO posted by cics on May 05, 2008 at 12:31:58
Kiep - thanks for the feedback. Also received good feedback via email. Please try version 1.0b2 (at same link provided or below).
If you have 1.0b1 installed - remove it before installing 1.0b2
- Won't work for me - JamesB 21:31:57 05/06/08 (3)
In Reply to: RE: cPlay 1.0b2 released posted by cics on May 05, 2008 at 21:48:20
Hi,
I thought I'd give this a try, but I can't get it to play with a M-Audio Transit driver. When clicking on a WAV file, there is a diagnostic message that is displayed really quickly before the program exits, I had to hit 'print screen' pretty smartly to capture it. It might not be working because I'm using Windows 2000, or it might be related to the errors shown in the diagnostic message...
-- James :)
- Good catch! Your soundcard is using 16 bits in 32 bit words - cics 04:25:45 05/07/08 (2)
In Reply to: RE: Won't work for me posted by JamesB on May 06, 2008 at 21:31:57
cPlay only supports pure 32 bit ASIO soundcards. ASIO standard defines a great many input and output formats. Trying to accommodate all would compromise quality. So I decided to stick with the most common standard - 32 bits.
Never thought I would see that! A workaround would be to use ASIO4ALL which would then KS to M-Audio. ASIO4ALL should also give you better latency to 64 samples. Never tested cPlay on Windows 2000 - there may be issues here (but unlikely).
- RE: Good catch! Your soundcard is using 16 bits in 32 bit words - JamesB 03:27:10 05/08/08 (1)
In Reply to: RE: Good catch! Your soundcard is using 16 bits in 32 bit words posted by cics on May 07, 2008 at 04:25:45
Yep, works fine with ASIO4ALL. Sounds pretty good actually but I don't like having to click on the files to play them. Do you know of any software that can provide a front end for a music library but invoke cPlay as the player?
Thanks, and good work...
-- James :)
- Use cMP's operating software - cics 13:50:18 05/08/08 (0)
In Reply to: RE: Good catch! Your soundcard is using 16 bits in 32 bit words posted by JamesB on May 08, 2008 at 03:27:10
Links are provided in the main post. Make sure you follow the simple instructions in cMP's installation manual. Also see this useful post by Seger on how to create cue files from wav files.
cMP's operating software works on any machine with all sorts of configurations. Setup is straight forward and it presents your music library (using cue files) by Genre/Artist/All. You can navigate remotely using a mouse. More importantly, cMP optimizes the Windows environment for audio: things like thread optimizations and the Windows system footprint is reduced.
In cPlay's installation folder you'll see 'cicsMemoryPlayer.pth' file - use this (copy to cMP's installation folder) to have cMP launch cPlay instead of foobar (cMP default).
- Question about sweep files - Presto 11:23:01 05/06/08 (2)
In Reply to: RE: cPlay 1.0b2 released posted by cics on May 05, 2008 at 21:48:20
Cics:
I wanted to play some sweep files to do some checking on the output of the player... but for some reason a sweep generated by NCH tone generator results in a "can't play" error (something to do with data block not found or information not found). The sweep files play in all other players I have.
Anyways, have you encountered this? *Or*, is there a reason why this is?
Just wondering...
Cool concept! Will be following this one for sure! Nice work! Thanks!
Cheers,
Presto
- Its cPlay... - cics 13:23:31 05/06/08 (1)
In Reply to: RE: Question about sweep files posted by Presto on May 06, 2008 at 11:23:01
I'll look at what NCH is generating - there's a few different WAV structures in use. cPlay does strict validations here.
- RE: Its cPlay... - Presto 16:52:58 05/06/08 (0)
In Reply to: RE: Its cPlay... posted by cics on May 06, 2008 at 13:23:31
Hey - I have other sweep file generators I can try...
I will try those as well!
Cheers,
Presto
- Few other things to add - Kiep 08:51:15 05/06/08 (2)
In Reply to: RE: cPlay 1.0b2 released posted by cics on May 05, 2008 at 21:48:20
File Browse/Select from the player (or drag'n'drop if it is not too hard to do)
Scroll for volume
Playback progress bar
Other than that it sounds really good. I actually like the 96k upsampling on it.
- RE: Few other things to add - cics 13:31:19 05/06/08 (1)
In Reply to: RE: Few other things to add posted by Kiep on May 06, 2008 at 08:51:15
For 1.0b3 I have:
- Phase 0/180 (can change this remotely using mouse button 4 with cMP, otherwise use keyboard ';' key)
- Playback order (repeat, stop, random)
- Minor refinements (slightly less UI thread interference)
It's the last point above that makes things like progress bars a bad idea. File selection and Volume Scroll is a possibility - nice-to-have for now. Next is FLAC.
- I think "elapsed / total length" display is enough myself (nt) - Presto 15:39:57 05/07/08 (0)
In Reply to: RE: Few other things to add posted by cics on May 06, 2008 at 13:31:19
nt
- Any chance of a polarity switch? - rickmcinnis@dogwoodfabrics.com 08:28:42 05/06/08 (1)
In Reply to: RE: cPlay 1.0b2 released posted by cics on May 05, 2008 at 21:48:20
Dear cics,
I could not wait to get home and install these (and now I see a new version is already available).
As one settles in you could hear much finer resolution of the high frequencies. I wonder if this is the improved signal to noise ratio at work. All that was good before seems to have remained with worthwhile improvements elsewhere.
Could not turn it off until one-thirty (damn it!)
I remember when I asked you before about polarity inversion you said it would have to be done within FOOBAR. Well, now that you have removed FOOBAR (or I am assuming such) would it be possible for YOU to introduce this into cPLAY? Once one gets used to listening for the effect you miss not being able to effect it. It would be icing on the cAKE!
THANKS VERY MUCH for your efforts!!! I thoroughly enjoyed listening to music last night. I can't wait to see if I can hear a differnence with the latest cPLAY.
Rick McInnis
- It's coming in 1.0b3 - cics 13:41:33 05/06/08 (0)
In Reply to: RE: Any chance of a polarity switch? posted by rickmcinnis@dogwoodfabrics.com on May 06, 2008 at 08:28:42
See here on 1.0b3
Juli@ in your setup would work very well - notice CPU load at 192k is much lower than before (~45% vs ~70%). You should underclock/undervolt that E1200 processor - no need for extra voltage at 192k! Also try setting RAM to 200MHz, disable '1T CMD Support' and keep DRAM Timing to 'Auto by SPD'.
- RE: cPlay 1.0b2 released - fmak 04:24:57 05/06/08 (2)
In Reply to: RE: cPlay 1.0b2 released posted by cics on May 05, 2008 at 21:48:20
It would be good if the Player can support 32 bit files. I cannot play any of the downloaded or converted (using Audition) hirez files.
Perhaps you will also consider a drag and drop facilty to enable entire folders in .wav to be loaded and played.
- Do you have a link on those 32 bit files? - cics 13:47:36 05/06/08 (1)
In Reply to: RE: cPlay 1.0b2 released posted by fmak on May 06, 2008 at 04:24:57
cPlay is limited to 2GB RAM load (Windows limitation). Even so, it could play in excess of 40 minutes (assuming 96k). Definitely worth looking into.
Drag n drop stuff is nice-to-have for now.
- RE: Do you have a link on those 32 bit files? - fmak 21:59:25 05/06/08 (0)
In Reply to: RE: Do you have a link on those 32 bit files? posted by cics on May 06, 2008 at 13:47:36
Actually it seems to be the 0404usb with 32 bit file structure. I opened a 24/96 .wav file and got the following fault report
ASIO Driver Initializing...
Name (ASIO E-MU 0404 | USB)
Player ASIO version (2)
Driver ASIO version (2)
Message (No ASIO Driver Error)
Channels (inputs: 4, outputs: 4)
Buffer details (min: 192, max: 48000, preferred: 960, granularity: 96)
Sample rate ( 96000)
Output Ready? Supported
Preparing buffers... successful
Latencies (input: 1096, output: 1384)
ASIO Driver initialized.
DSP starting...
DSP Buffer (30720 samples or 320ms/ 3.1/ 100.0) allocation successful
ASIO Driver starting...
Input 1 [ Analog IN A ] Active (0) type (32bits)
Input 2 [ Analog IN B ] Active (0) type (32bits)
Input 3 [ SPDIF IN L ] Active (0) type (32bits)
Input 4 [ SPDIF IN R ] Active (0) type (32bits)
Output 1 [ Analog OUT ] Active (1) type (32bits) > Left Channel
Output 2 [ Analog OUT ] Active (1) type (32bits) > Right Channel
Output 3 [ SPDIF OUT ] Active (0) type (32bits)
Output 4 [ SPDIF OUT ] Active (0) type (32bits)
ASIO started successfully.
Processing CD [D:\Music\96k2LBeethoven\2L49SACD96k.wav]...
Wav file size = 328MB
Wav file structure is NOT supported.
Genre = Unknown
Artist = D:\Music\96k2LBeethoven\2L49SACD96k.wav
Album = WAV
Wav file = D:\Music\96k2LBeethoven\2L49SACD96k.wav
You can get hirez files from 2L.no, Linn and DVD-V 192k rips.
- RE: cPlay 1.0b2 released - artcheng@netvigator.com 03:02:14 05/06/08 (1)
In Reply to: RE: cPlay 1.0b2 released posted by cics on May 05, 2008 at 21:48:20
Cics,
Thanks once again for your wonderful program. Seems it comes with a volumne control as well. Can you share with us which bit depth is the volyme control work in?
- VC is 64 bits double precision (real numbers) - cics 13:51:26 05/06/08 (0)
In Reply to: RE: cPlay 1.0b2 released posted by artcheng@netvigator.com on May 06, 2008 at 03:02:14
In my setup, I can't hear any loss of resolution at low volume - there's amazing ambient information even at -30db.
- Sounds different than Foobar 0.83 - Kiep 14:55:50 05/05/08 (6)
In Reply to: RE: cPlay - the open source high-end audio player using ASIO posted by cics on May 05, 2008 at 12:31:58
On a non-cmP desktop 2Ghz, 2GB with SB X-Fi using ASIO:
More forward, wider images, vocals seem to be a bit fuzzier than glossy clean Foobar. Tighter bass, good timing, pretty good overall
My 2c.
- RE: Sounds different than Foobar 0.83 - Presto 11:19:22 05/06/08 (5)
In Reply to: RE: Sounds different than Foobar 0.83 posted by Kiep on May 05, 2008 at 14:55:50
Kiep:
At first when I a/b'd (sighted/not blind) these two, I thought they were pretty darned close if not the same. Then I picked some REALLY familiar test tracks and strange things started happening.
I could swear the Cplayer has a differenting sounding top-end. It is very detailed, and very succinct - not harsh or edgdy, just very well defined. Then when I go back to Foobar, I can't tell whether there is *less* detail or just a slight droop in the top end - but there is a difference. I want to see if I can pick it out blind...
I have a hard time believing this is all in my head, but I would *really* love to start measuring soon to see if it is or not. F/R is measurable, but high-end detail is another matter.
Strangely, I cannot play back .wav sweep files on cplay to do bit-comarisons between it and Foobar. I will talk to Cics about this issue.
I do like the idea of a minimalist ASIO player with switchable built in SRC!! I am *not* interesting in *cool* looking GUI's over SQ!
Cheers,
Presto
- RE: Sounds different than Foobar 0.83 - Riboge 11:27:03 05/07/08 (4)
In Reply to: RE: Sounds different than Foobar 0.83 posted by Presto on May 06, 2008 at 11:19:22
I hear what seems a quite similar difference between the two and would also be curious whether this is heard by others and whether it is a quality or just a fr difference in the high end.
- RE: Sounds different than Foobar 0.83 - Presto 12:50:38 05/07/08 (3)
In Reply to: RE: Sounds different than Foobar 0.83 posted by Riboge on May 07, 2008 at 11:27:03
Riboge:
Yes, I agree that something seems quite different, especially on the high end and I cannot tell if it is *different* or if there is just some sort of level difference. I've done a fair amount of l-padding on tweeter and have worked with zobels, etc. to know what a level difference "sounds like", and for some reason I *feel* there is more to this perceived difference than just a hf lift or shelf. The high-end details just seem to be more succinct (more defined, clear cut) with cplayer. It could be "new toy bias factor" too! Gotta watch out for this! When I get a new toy, the "hope" it will be a GOOD SOUNDING new toy can influence my judgement. I mean, do we ever try something new hoping it will SUCK? Not really!
But whatever is hitting the DACS on my PCI card is seemingly not the same coming from Foobar 0.8.3 and Cplayer. It makes one wish to have the test equipment to get eye patterns from each signal before they hit the DACS.
Could playing 100% from memory be a reason?
This is a facinating new toy for us to experiment with! :o)
Cheers,
Presto
- RE: Sounds different than Foobar 0.83 - Kiepu 08:27:56 05/08/08 (2)
In Reply to: RE: Sounds different than Foobar 0.83 posted by Presto on May 07, 2008 at 12:50:38
I think this software deserves as much attention as Foobar.
The thing I pay attention to is timing which imo is what either makes or breaks the digital playback. The cPlay is quite good in this regard and I would say probably a bit better than Foobar (in my short experience).There are areas where things can get improved, focus is one of them, I still hear some haze on vocals, not sure if this is my setup (headphones) or something else. As someone pointed out there is still some mellowness which I don't think is part of the recordings.
Exciting tool/toy nevertheless and it will be interesting to see how it develops.
Edits: 05/08/08
- RE: Sounds different than Foobar 0.83 - audioAl 06:05:37 05/31/08 (1)
In Reply to: RE: Sounds different than Foobar 0.83 posted by Kiepu on May 08, 2008 at 08:27:56
Hello:I built an AMD 64 duo-core AM2 processor based ATX Xion case with the Micro Star International Mainboard, it has the K9N Neo V2/V3 with the MS-7369 mainboard and Realtek ALC888 Audio. It has S/PDIF output. Can I use my HT Receiver with just a cable Optical ??
Now I Understand
- RE: Sounds different than Foobar 0.83 - theNorskie 06:12:50 01/21/09 (0)
In Reply to: RE: Sounds different than Foobar 0.83 posted by audioAl on May 31, 2008 at 06:05:37
I installed cPlay today. I tried changing Asio settings from 32 to 16 for some reason. After that I haven't been able to get it to work. I first tried to restart the player, but was not able to open Asio settings anymore (fated out). I then tried uninstalling and reinstaling cPlayer, but with same result. I've also tried to reinstall the driver for my EMU 04040 USB soundcard withoud any success.
Help!
Edits: 01/21/09
FAQ Post a Message!
Forgot Password?
Moniker (Username): Password (Optional): Remember my Moniker & Password (What's this?)  Eat Me E-Mail (Optional): Subject:
Message: (Posts are subject to Content Rules) Optional Link URL: Optional Link Title: Optional Image URL: Upload Image: E-mail Replies: Automagically notify you when someone responds.
To view your new posting or follow-up, click on the RELOAD or REFRESH button on your browser.
This post is made possible by the generous support of people like you and our sponsors: