Computer Audio Asylum

Music servers and other computer based digital audio technologies.

Return to Computer Audio Asylum

Message Sort: Post Order or Asylum Reverse Threaded

Page: [ 1 ] . . . [ 29 ] [ 30 ] [ 31 ] [ 32 ] [ 33 ] [ 34 ] [ 35 ]

cMP - the open source high-end Memory Player

41.183.0.21
Posted on December 30, 2007 at 05:42:01
cics
Audiophile

Posts: 1320
Joined: November 9, 2006
cMP

The open source high-end Memory Player

December 2007

Music delivery is perfected. cMP (cics Memory Player) delivers memory playback (without RAMDisk or other utilities) thereby eliminating disk interference during playback.

In creating cMP, its critical objective was the implementation of fundamental designs that:

  1. Remove disk traffic interference during playback without having the inconvenience of manually loading files into a virtual RAM drive, etc..
  2. Ensure time critical sample delivery to the external DAC. Streaming of samples from precision upsampled (to 24/96) sound buffers to the soundcard must be fast, lean and mean! There's no room for any bottlenecks or overheads here, otherwise we risk increasing jitter and compromise Bit Transparency.
  3. Provide better power supply and remove the need for a ferrite core on the mouse connection.
  4. Provide flexibility to choose any player (Foobar2000, Winamp etc.).
  5. Further minimize the Windows footprint to that more optimal than XP's Embedded SP2 operating system as used in large scale commercial applications. This provides the foundation for the 2nd design goal.
  6. Offer a CD/DVD-ROM drive without impacting sound quality.
  7. Offer proper remote usability as locating that mouse cursor from a distance is cumbersome.
  8. Contain costs but also allow for a complete elegant one box solution.

Such designs in theory should render a stunning sonic experience. Its implementation however was a massive challenge. Did this theory meet the sonic expectations? Yes as evidenced in cMP's bit perfect performance and ultra low jitter.

How does it sound? The most descriptive word that comes to mind is emotional. cMP's sonic purity is truly profound. It grips those musical lines across any genre with remarkable precision. Soundstage improves, there's better layering and even more ambient information. Bass is tighter, with natural vocals & instruments.

FRONT



BACK



More pictures and screenshots are available from the cMP documentation link provided below.

FEEDBACK

Pre-release testing by music lovers located all over the world yielded further insight and perfection of cMP.

  • A music lover based in the US with limited or no computer training:

    This has been a great way to learn about computers in addition to making something very useful for enjoying music.

    Compared to my ###### (read expensive high-end traditional CD transport), even with some tweaking and learning left to do, this sounds much better. Dynamic, yet relaxed (effortless?).

    I started listening last night at 7:00 and had to force myself to go to bed at twelve. Amazing spaciousness. It is truly getting close to GREAT analog and is much better than average analog already. It is simply amazing.

    I found your instructions easy to follow ONCE I understood what I was being told to do. To repeat, I have never done anything like this before, so your instructions were clear without resorting to "hand holding". I think the process was easier than you warned. I kept saying to myself, "that wasn't so bad!"

    After cMP:

    My only audio friends are very analogue-centric. Sometime in January I am going to have them over to audition the cMP. I think they are going to be amazed. Of course, I am confident that JULI@ is going to be the icing on the cake. If this takes the whole thing another step closer I think they will be grateful that one can enjoy listening to CD's as much as LP's. Of course, in many ways your approach does things even LP's cannot do. It is becoming a toss-up, but the best kind; one can enjoy either one equally well!

  • A music lover based in Europe with more than 2000 CDs:

    the quality of sound reproduction is quite amazing. My motivation for building was that I am not wealthy and could never dream of, let alone buy, a high-end CD player. Even this project was considerably more than I could properly afford and I began it with some trepidation. But it's been well worth it.

    After cMP:

    I spent several hours last night testing the latest cMP. I did several A/B runs with cMP on the one hand and Foobar launched via old batch files and reading direct from disk on the other. There is no doubt in my mind that the sound using cMP is a significant improvement even with my modest setup.

    The quality is quite stunning, significantly better than previous. (My partner agrees with this - she's no audio junkie but she is an informed music lover.)

  • A music lover based in Australia:

    I have been wanting to relay to you my own experience with your methods and give you some feedback . So far, I have found your approach produces very good results so I must convey to you my thanks and appreciation. At some point soon I will formalise my thoughts and send them to you or post them. I believe what you have done has become an excellent resource/asset to people interested in PC Audio.


FEATURES

  1. Disk traffic interference during playback is eliminated. Advanced RAM playback is based on available physical RAM (as reported by Task Manager or Process Explorer). This means up to 4GB (XP's limit) can be used allowing for 24/192 RAM playback. Recommendation is to use 1GB RAM (which yields available physical RAM of ~830MB - enough for any CD at 16/44.1). cMP achieves memory playback via the system cache and not through a simulated virtual drive. This approach removes Windows disk I/O overheads (irrespective of whether disk is physical or virtual) hence, it's more optimal.

  2. Extensive Windows & Player (Foobar2000 or other) optimizations yielding much lower jitter and bit perfect delivery. System level optimizations are done at runtime (which cannot be achieved using .bat files and utilities like Process Explorer). cMP is designed to work with 32 CPU cores!

  3. Open architecture. cMP allows for any player to be used: Foobar2000, XXHighEnd, Winamp etc. (player just needs to handle .cue files like foobar2000 otherwise play entire .wav). Any ripper software can be used (as long as it conforms to .cue single file standard like those created by EAC). Additional flexibility is provided, for example use cMP to drive your HTPC playing any movie from any genre (a .cue file for each .iso file is needed).

  4. All .cue files are processed by cMP which means the player focuses only on playing the CD. In Foobar2000, the album list control component is not needed. In techno speak this is called separation of concerns.

  5. Power delivery to the mobo is improved by further reducing interference. The need for a ferrite core is removed.

  6. Full function remote control using a wireless mouse is achieved. Your entire library can be navigated: play any CD, eject it, jump to any track or change volume. This is achieved using just ONE mouse button: the Wheel! Use the mouse as a normal one (left-click) or hold it in your hand (like a remote control) and wheel away using your thumb. Of course other buttons are used for more convenience. There's no need to find that tiny mouse cursor and left click! (You can still do this if needed).

  7. Your entire library that you so diligently ripped is fully compatible with cMP. No proprietary / lock-in stuff here which forces you to re-rip. Just point cMP to your music folders containing .cue and .wav files (or flac or other). cMP supports any content file as long as cue files are defined for them. For flac content files, the known EAC bug of adding an additional .flac suffix to the content file's name must be removed (otherwise your cue's content file directive will not be found).

  8. CD/DVD-ROM drive is implemented without compromising playback quality. For the purist, this can be unplugged without opening computer or rebooting.

Bottom line: cMP is a (one box) highly advanced memory player that's easy to setup and use. cMP's core strength lies in newly developed operating system software components (built in c/c++) that uses XP SP 2 as a foundation. Windows Explorer is done away with while cMP takes over and more. Bit perfect delivery is achieved yielding an ultra high resolution of more than 23.5 bits (a limitation in measuring instruments prevents measuring cMP's perfect resolution). DAC's have yet to achieve this resolution (as a minimum of 141db SNR is needed)! For cMP's bit perfect measurements, see (Bit Perfect Measurement & Analysis ).


GETTING STARTED

Visit cicsMemoryPlayer.com where you'll find detailed easy to follow instructions and much more.

Ensuring clean AC power delivery to equipment is very important. For DAC, pre-amp and amplifiers, use a dedicated AC circuit. For source equipment, use another AC circuit as this is where cMP's main PSU should be powered from. Use another less important AC circuit for powering cMP's 2nd dirty power inlet. Use of quality power cords and line conditioners is recommended.


Special thanks to those that provided such brilliant feedback and suggestions during pre-release testing. cMP's software is at version 1.0b as more feedback is desired. Your input, insights, criticisms, experiences and suggestions all help.

Thank you.

 

Hide full thread outline!
    ...
Thanks very much - had no idea it could be so simple!, posted on November 15, 2011 at 09:22:49
rickmcinnis@dogwoodfabrics.com
Audiophile

Posts: 3040
Location: Atlanta
Joined: December 15, 2003
 Now to see if Brian Lowe can make a 1.3 volts regulator!

I have lost my love of batteries. Too much trouble. I should be ashamed. But if there is no other way I will give them a try.

Sorry for my barrage of posts.

THANKS,

 

I will need your blessings of good luck!, posted on November 15, 2011 at 09:28:12
rickmcinnis@dogwoodfabrics.com
Audiophile

Posts: 3040
Location: Atlanta
Joined: December 15, 2003
 Must say I am a littel disappointed about not getting the benefit of lowering the 5 volts rail's current requirement. Oh, well ... you did say there was the more important benefit of better sound quality.

I am a relatively experienced hobbyist so I think, with your instruction, I can do this without making too much of a mess. I sincerely appreciate your recommendations and guidance.

I will post when I have actually done it!

I am assuming the chips to remove are obvious? I will extrapolate from your pictures of the H55 variant from a year ago.

Thanks very much,

 

RE: Please suggest hardware for my 2nd CMP (sorry for the extra posts..I think I hit reload too many times), posted on November 15, 2011 at 17:53:33
Mihaylov
Audiophile

Posts: 355
Location: Moscow
Joined: March 11, 2010
 Hi Mike! I think that H10.ODD FANLESS CHASSIS or H5.S FANLESS CHASSIS are good choice. Both cases are compatible with mini-ITX and mATX. If the flexible riser card will not work (which is not necessarily otherwise such a decision would have not been used and not offer by manufacturer), you simply buy Juli - it 's digital part must fit in the case.


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

 

RE: Please suggest hardware for my 2nd CMP (sorry for the extra posts..I think I hit reload too many times), posted on November 15, 2011 at 20:06:22
whodat06
Audiophile

Posts: 58
Location: gulf coast
Joined: March 22, 2006
 That doesn't give me much confidence. Sounds like a risk. I'm still in search of a case I guess. Thanks for the input.

 

Removing the video chips will make noticable improvement in SQ, posted on November 16, 2011 at 08:18:08
jackwong96
Audiophile

Posts: 219
Joined: April 20, 2010
 but risky!

Elimination of video interference makes a difference, I've tested it thoroughly on several boards.

Reduction in current consumption will ease the power supply, improvement would largely depend on the ability of PS.

If you want to test how it sound like when the video chips are removed, read the link and implement the "display off".

I've tried the "display off" with the new S2V board where the VGA to DVI converter was removed and it made a difference. However, you can achieve much better improvement with the 2 video chips removed.

 

RE: Removing the video chips will make noticable improvement in SQ, posted on November 16, 2011 at 13:03:49
theob
Audiophile

Posts: 3180
Location: ann arbor michigan
Joined: November 4, 2000
 You have given me an inexpensive way to get into the new intel hw. Thanks.

I still am interested ih hearing the full story of the .01 ohm resistor.

 

RE: Please suggest hardware for my 2nd CMP (sorry for the extra posts..I think I hit reload too many times), posted on November 16, 2011 at 20:28:21
whodat06
Audiophile

Posts: 58
Location: gulf coast
Joined: March 22, 2006
 H3 is the only model in stock. I asked them to notify me when they are in stock of the H10 and H5, be. There is only a $10 difference in price between those two models, but I guess I'd lean towards the larger one because I might fit more inside and it has an optical drive slot which I could use to rip discs with in the future. The extra footprint isn't ideal, but it should be okay. Pity because not having a case will set me back or I'll be forced to buy something else in the interim.
Just realized I have to buy DDR3 ram because DDR2 apparently isn't compatible. and now a Juli@ card too. The cost is adding up...

 

The Story of a 0.01 Ohm Resistor, posted on November 17, 2011 at 09:31:21
jackwong96
Audiophile

Posts: 219
Joined: April 20, 2010
 Try this: connect a 0.01 ohm resistor to the tweeters of your speakers and try to determine which setting is better. I bet they sound differently but they are no better than the other.

OK. Try it if you have time or read the below story.


I’ll try to cut it short.

Many years ago whilst in the course of seeking musical fidelity, I met a guy who built amps and speakers for living. He looked like ET as he has got a big head and skinny body. Here and after I called him X. X was a graduate from an electronic institution who has more than ten years of experience in the field of HiFi. He was willing to share his knowledge with me. At that time, I had some knowledge on audio DIY and refrained from buying commercial audio gears, because I could built something good with much less cost. I also built speakers and that‘s what X interested me.

He told me how to accurately adjust the tweeter/mid-range & tweeter level of a pair of speakers so that they would sound really good. The bass unit was connected directly whilst the tweeter level attenuated using resistor ladder network. That sound usual to me, what surprised me was that he was tuning the tweeter level using some 0.01 ohm resistors(and some others bigger in value) in a passive two way speaker system.

He explained that there was a perfect matching point when the tweeters were attenuated to reach the exact same level as the woofer, and that sweet spot could be missed so easily. He tuned the resistance of the network up and down (with solder) and listened to the changes. I wasn’t sure whether he was exaggerating or trying to fool me, but everything he told me was useful, true and confirmed in textbook. I however had no opportunity to witness the successful tuning and audition the sonic results.

I was not sure about X’s 0.01 ohm setting methodology. And I told myself should that be true then all commercial speakers would be inaccurate in attenuation setting and thus far from good sounding, because they all used 5-10% tolerance resistors. No, I told myself that can’t be the case.

However, X’s methodology kept popping up as I was not satisfied with the speakers I built. About 2 years later, I decided to give it a try when setting a pair of 2-way speakers.

At first, I had no success at all despite my attempts for two months. At some points sub-sonic emerged and I remembered X had told me that it was near but not close to the spot. I went on varying the value of the resistor network using a chain of 0.0.1 and 0.03 ohm resistors(of course there were others with bigger values), moving up and down and at the same time maintaining an overall resistance of 8 ohm. It was annoying because each and every time I had to solder both legs of the ladder, and for both speakers.

My ears were fatigue after listening for 2 or more hours almost everyday for more than two month, because I need to compare the sonic difference each and every time I changed the values. Nevertheless, faith had driven me to keep on trying despite so many unsuccessful attempts.

I attempted further for almost three months, and eventually I hit the spot. It was so distinguishably good at that exact spot that moving 0.01 ohm upward or downward would totally ruin the balance. It’s just that sweet spot and nowhere else. And I had to agree with X.

I demonstrated to some audio friends the difference between the exact tuning spot and off 0.01 ohm or further, all of them could tell where the right point was. The differences were so significant that friends who had no interest in HiFi could judge.

During the rest of DIY life, I attested X’s speaking setting methodology on five more occasions when building speakers for myself and friends. Once I spent more than one years tuning a pair of 12" 3-way Dynaudio speakers. That was really difficult because the mid and tweeter had to be matched first and then bundled to move up and down the resistor network. Matching the mid and high alone was not easy when they were nowhere match to others in level and there were a lot of resistor soldering works(4 sets)to do. The last attempt was done about three years ago with a pair of Morel 6” 2-way speakers. The owner auditioned the changes in SQ (and unsuccessful attempts) during the two month tuning period, and he finally got a pair of sweet sounding speakers.

One interesting thing about this setting method is that you can’t possibly use other standard value resistors in place of the original chain of resistor and in order to maintain accuracy the connecting wires of the tweeter will also have to moved intact into the cabinet.


I’m not expecting anyone to believe in what I said. I myself didn't believe it either when first heard about it. It is just my experience.

 

RE: The Story of a 0.01 Ohm Resistor, posted on November 17, 2011 at 10:40:54
theob
Audiophile

Posts: 3180
Location: ann arbor michigan
Joined: November 4, 2000
 Thank you very much ...very interesting...as always.

 

Another bit of inspiration - , posted on November 17, 2011 at 11:44:34
rickmcinnis@dogwoodfabrics.com
Audiophile

Posts: 3040
Location: Atlanta
Joined: December 15, 2003
 I wonder if I have the patience to do this ... but i have no doubt that is absolutely true.

What order of crossover were you tuning? One could see the higher the order the more importance this tuning would have. My guess is that the first order would be less critical but would still need a critical balance for best performance.

All is compromised with our humble hobby - with this critical tuning requiring all of those resistors and requisite solder joints there is a concern that something ELSE is lost. Different ears have different sensitivities; we can't have it all ...

I am now thinking: how to use this method?

Thanks,

 

RE: Please suggest hardware for my 2nd CMP (sorry for the extra posts..I think I hit reload too many times), posted on November 17, 2011 at 15:16:24
Mihaylov
Audiophile

Posts: 355
Location: Moscow
Joined: March 11, 2010
 =I asked them to notify me when they are in stock of the H10 and H5, be.=
- I did as well. I like a small case because it's perfect for my network configuration of cMP2 but need a place inside case for a linear power supply (the case itself will serve as an excellent heat sink).

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

 

RE: Another bit of inspiration - , posted on November 20, 2011 at 05:34:44
jackwong96
Audiophile

Posts: 219
Joined: April 20, 2010
 Not only patience. In order to succeed, apart from basic knowledge you will also need a keen interest in DIY and most importantly faith.

I used a mixed combination from 1st order to 3rd order in different projects.

".....My guess is that the first order would be less critical but would still need a critical balance for best performance."

The latter half is right. My experience was that the tuning was irrespective of whatever order/combination of crossover network used, the order determine the steepness of the slope at crossover points, but it has nothing to do with leveling the speaker units. They all need critical balance.

".....tuning requiring all of those resistors and requisite solder joints there is a concern that something ELSE is lost."

I think I gained everything I wanted to hear. Whether the solder joints and resistors will cause a lost in sonic quality, you can easily determine that by replacing any resistor in your audio gear (which is in the signal path) by ten resistors of one tenth the value.

The way I see it, music we hear has in fact gone through many resistors, capacitor and transistors/tubes, they are present in the circuit as they are required to do a job. I do not see more resistors a devil. Good soldering skills and use of quality components will ensure good sound.

If you're interested. I can dig out my stuff.

 

Remove chips not needed for I2S connectivity, posted on November 20, 2011 at 05:44:18
jackwong96
Audiophile

Posts: 219
Joined: April 20, 2010


Hi,

Great job! You are going as crazy as me!

Regarding remove of chips the 74HC125 need not be removed. Alternatively, just lift pin 14 Vcc (the pin above the "D" of DGND1). Check its connection to the reg.

http://www.datasheetcatalog.org/datasheet/SGSThomsonMicroelectronics/mXrytxt.pdf

74HC125 is driving the Dig trans and optical output. Without power supply, the trans will no longer work, hence no need to remove. The optical output will not be working too, but it will still draw a small amount of current.

 

I would like to think that I've been inspired by you!, posted on November 20, 2011 at 15:30:59
GStew
Audiophile

Posts: 633
Location: NE Mississippi
Joined: September 21, 2001
 Jack,

In any case I did just what you suggested for the 74HC125. BTW, the comments made (partly by you) in the thread about a year ago on a tube SPDIF buffer thread here helped me understand how that part of the Juli@ card works... thanks!

I also removed the TOSLINK connector (by un-soldering it) and the digital input optocoupler (by just twisting it off with pliers, since it is an easily-obtained part if I wanted to undo the change).

Immediate results (no break-in time for something like this!) seemed like a slight increase in HF clarity, but was not large. I have higher hopes that un-powering the AK4114 will produce a larger improvement, but I still need to confirm that it's removal won't stop the card from functioning in I2S mode. I have an older, somewhat sproratic Juli@ that I'm setting up to test this and I hope to do that test sometime this week.

I am still getting a little of what I identify as the ceramic cap signature. They've been running in for about a week (maybe 150 hours total on-time)... If it persists over the next several days, I'll try disconnecting them. If that removes what I'm hearing, I'll try different caps there (small organic polymers next). Got any suggestions for a good low-inductance local digital circuit bypass?

A related question here... back when you first posted here about your extreme mobo mods, you'd suggested that you were looking at simplifying the process by just cutting or disconnecting power pins instead of doing full chip removals. Now in your latest posts on removing chips, cutting power pins is not mentioned. Is this is a bad option for eliminating chips on a motherboard? Not that I am afraid of pulling chips, just lazy!

TIA!

Greg in Mississippi
Everything matters!

 

Agree, cool-looking cases..., posted on November 20, 2011 at 19:29:20
GStew
Audiophile

Posts: 633
Location: NE Mississippi
Joined: September 21, 2001
 But one might be able to make-do with some of the DIY amplifier cases available nowadays.

Instead of heatpipes, has anyone played with smaller, amplifier-styled custom-cut heatsinks like the ones Peter Daniel has shown occasionally on his threads on DIYAudio.com?

With an under-clocked, under-voltaged system something like this might work ok?

Greg in Mississippi

















Everything matters!

 

PS/2 Mice & other recent interesting HW comments..., posted on November 20, 2011 at 19:59:29
GStew
Audiophile

Posts: 633
Location: NE Mississippi
Joined: September 21, 2001
 There have been a few comments about alternative HW in recent cMP/cPlay threads. So of course, I have some questions:

1. PS/2 mice. I really like this idea, especially if I can get entirely away from any USB processing in my system. But I really like the remote-ability of the wireless mice. Does anyone know of or has used a wireless PS/2 mouse or an adapter that will allow the use of a USB wireless mouse via a PS/2 adapter? I have USB-PS/2 adapters which will allow a wired USB mouse to work into a PS/2 port, but they don't seem to work with my wireless mice.

2. Transcend IDE Flash Module 1GB and similar. Mihaylov, I saw you mentioned this as a better alternative to the standard SATA SSD. What makes it better? I would have thought sticking with only SATA would have been better than a mix of IDE & SATA. Perhaps relatedly, FMAK posted about an IDE DOM (Disk on Module) drive for a small music server he put together here: http://www.audioasylum.com/cgi/t.mpl?f=pcaudio&m=98415 . Anyone know anything about these?

3. Non-Native PCI Mobos. Ok, as Rick McInnis, Jack Wong & others have pointed out, the latest SandyBridge chipset appropriate to our purposes does not support PCI natively. Some of the boards using this chipset do appear to have non-native PCI slots. Also, Bibo01 wrote 'Q67 (also Q65) chipset microATX motherboards do have native PCI'. Has anyone used a Mobo without native PCI support with a PCI soundcard for their interface and can report on the sound? As Jack Wong said, these boards may be significant steps up from today's boards... maybe the suboptimal PCI processing will be swamped by the inherent SQ improvement? Or has anyone seen and/or used any of these Q67/Q65 boards that support PCI natively?

4. SSD Under-volting. Douwe01nl wrote a wonderfully intrigueing post with several HW tweak opportunities. One was undervolting his SSD. Anyone else tried this? Except for being focused on Juli@ mods right now, this would be an easy one for me to try as I have linear supplies with variable regulators on my SSD & HDD. But hopefully others have tried it, but just haven't reported.

More on Douwe01nl's thread.

TIA!

Greg in Mississippi
Everything matters!

 

RE: SSD @ 2,6V, posted on November 20, 2011 at 20:18:31
GStew
Audiophile

Posts: 633
Location: NE Mississippi
Joined: September 21, 2001
 Douwe,

Sorry for not responding earlier. Great post! Thanks for letting us know some of the tricks you've learned... and also the results of some of the tests you've done. Do you have any more comparisons of various configurations you'd be willing to share?

Also post pictures. They are, as people say, worth at least 1000 words each!

Now some questions...

1st - on undervolting the SSD, did you open the SSD to determine that the 3.3v pins are not connected? It sounds like you currently just feed 2.6v to the 5v pins. Is this correct? Have you tried it at any other voltages... does it sound better as you go lower and it won't work any lower than 2.6v or something else?

2nd - on changing out the CCFL inverter on your monitor... why would that make a difference in the sound? Did you go from a wall-voltage (assume 240v) inverter to a battery-powered one? Also, is the inverter just a higher-voltage supply or does it have 'startup' logic? Could it be replaced with an appropriate linear supply? Also, have you tried any of the LED supplies as shown on the website you referenced?

3rd - Why does using a battery-sourced set of inverters improve the sound over just plugging the power supplies you have into the wall? Just curious... I've been resisting going battery for awhile, but your experience (along with JackWong's) are causing me to weaken.

TIA!

Greg in Mississippi

Everything matters!

 

RE: PS/2 Mice & other recent interesting HW comments..., posted on November 20, 2011 at 20:29:12
GStew
Audiophile

Posts: 633
Location: NE Mississippi
Joined: September 21, 2001
 What about the Q67 chipset boards. Gigabyte currently lists one:

http://www.gigabyte.com/products/product-page.aspx?pid=3824#sp

Not sure if it's available yet, Newegg doesn't list it.

Any thoughts on it? Are other Q67/Q65 boards with native PCI coming soon?

Greg in Mississippi


Everything matters!

 

Thanks, posted on November 21, 2011 at 08:52:54
jackwong96
Audiophile

Posts: 219
Joined: April 20, 2010
 Greg,


I don't think AK4114 can be removed, it should be responsible for I2S output. See p.29.
http://www.asahi-kasei.co.jp/akm/en/product/ak4114/ak4114_f04e.pdf

Ceramic cap never sound good to me. Oscon caps do a better job than ceramic, Siemens Sikorel is even better. I also like Philips or BC but they are no longer available. Panasonic caps (those 105 *C, series M or FM, body in blue and word in gold colour)do a decent job although they aren't as good as the above or Black Gate.

For GA-H55M-UD2H, the following can be removed by cutting of their legs off and removing legs by solder,those smaller ones like video chips are too small to be cut off (I've tried):

- IDE controller next to the P24 connectors. However, the regulator below the roll of 3 SATA connector must be removed by desoldering.

- the sound chip at the corner

Better SQ improvement can only be done by removing the video chips however.

I just found it more neat and tidy using hot air gun to remove chips. Now I can heat up the chip directly with a temperature controlled hot air gun (just bought) and pick up a chip with a sharp nose plier. Tried on small chips like the video one on unused boards with full success without tamper with any other components.


Good luck

Jack




 

You are welcome... and I'm not sure the I2S..., posted on November 21, 2011 at 10:26:39
GStew
Audiophile

Posts: 633
Location: NE Mississippi
Joined: September 21, 2001
 Comes from the AK4114 in this case. Envy 1724 also has I2S output (sorry, I don't have a link for the datasheet, not sure it's available openly on the net, but if you send me an email I can send it).

My notes from my original Juli@ follies on the sources of the I2S lines are as follows:

J3 Pin 1 -> 1724 PSDOUT[0] ( I2S data out) -> AK4358 SDTI1 (I2S data in)

J3 Pin 5 -> 1724 PSYNC ( I2S Word Clock ) -> AK4358 LRCK (I2S Left/Right Word clock)

J3 Pin 7 -> 1724 PBCLK ( I2S bit clock out ) -> AK4358 BICK ( I2S bit clock in )

J3 Pin 9 - > Xilinx on the digital board -> AK4358 MCLK ( I2S Master Clock in )

J3 Pin 11 - > Xilinx & AK4114 on the digital board - > AK4358 PCN (Reset)

I suspect the AK4114 provides I2S from a SPDIF input to the 1724 which then passes it onto the AK4358 DAC. But as I'm killing the SPDIF input for my purposes, I'm hoping that doesn't matter.

That doesn't mean that the AK4114 doesn't provide the I2S to the Envy 1724, but with the clocks directly connected to the 1724, I suspect it is the one that provides the I2S.

But I2S out still may not work without the AK4114 available. We'll see!

On the caps, I hope to try Oscons there a little later in the week UNLESS the ceramics mellow out a bit more. I'll let you know on that too. Still, even with the ceramics, the sonics are quite nice... just not as sweet in the highs as I'd like under all conditions.

And thanks on the GA-H55M-UD2H chip removal info. See my questions about the Q67-based board in another thread... maybe the last native PCI option?

Thanks again!

Greg in Mississippi


Everything matters!

 

Not only cool-looking cases..., posted on November 21, 2011 at 11:49:55
Mihaylov
Audiophile

Posts: 355
Location: Moscow
Joined: March 11, 2010
 but also this massive thick aluminum case can serve as a heat sink for power components (rectifiers, diodes) of linear PS.

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

 

B65, Q65 and Q67 are the chipsets for business class systemboards., posted on November 21, 2011 at 12:21:29
Mihaylov
Audiophile

Posts: 355
Location: Moscow
Joined: March 11, 2010
 Therefore the BIOS settings might be very simplified w/o downclocking and downvoltage features. See for example the GA-Q67M-D2H-B3 manual. But there are the systemboards with suitable BIOS settings from other manufacturers for example Asus P8Q67-M DO.


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

 

About IDE DOM (Disk on Module), posted on November 21, 2011 at 12:39:42
Mihaylov
Audiophile

Posts: 355
Location: Moscow
Joined: March 11, 2010
 DOM is better because it's volume is smaller. The more volume SSD has more powerful processor. I want to use 256 or even 128 MB DOM.


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 tried some of those boards but never listened to them, posted on November 21, 2011 at 14:14:43
rickmcinnis@dogwoodfabrics.com
Audiophile

Posts: 3040
Location: Atlanta
Joined: December 15, 2003
 Not the one Mr. Wong mentioned though.

The one's I tried I could not get JULI@ to install. Something would be weird so I put the MB back in the box and back to NEWEGG. Tried three fo them.

Under-volting the SSD? That just makes no sense to me at all. Certainly would not want to give it any more than the median 5 volts but I think I refuse to believe that could make a sonic difference. I do love being proved wrong, though!

I hope that board Mr. Wong has tried will work 'cause I do not know what to do with that processor!!! Thinking about getting one as an excuse to do find the minimal nLite install that will allow AWE.

You need to do another picture of the JULI@ board with some explanation of what is what. That is more stuff attached to JULI@ than I would have thought possible!

 

I was being philosophic ..., posted on November 21, 2011 at 14:20:24
rickmcinnis@dogwoodfabrics.com
Audiophile

Posts: 3040
Location: Atlanta
Joined: December 15, 2003
 there is no question it is a miracle that what we hear can sound as good as it does when you consider what it has gone through to get to us.

I think there is a ready audience here to hear anything you have to offer Mr. Wong. You are an inspirational tickerer and experimenter (my favorite kind of person) and your work is of forward thinking.

So, the answer is, PLEASE -

 

I agree with Rick, posted on November 21, 2011 at 15:25:48
theob
Audiophile

Posts: 3180
Location: ann arbor michigan
Joined: November 4, 2000
 We used to have a tv commercial a few years ago about a certain brokerage house that traded securities. The commercial went something like '...when E F Hutton speaks everyone listens...' and here in AA Pc audio when Jack Wong speaks everyone listens ... and tries to learn. At least I do.

 

RE: Not only cool-looking cases..., posted on November 21, 2011 at 22:39:25
whodat06
Audiophile

Posts: 58
Location: gulf coast
Joined: March 22, 2006
 I like the heat sink idea, however, I ordered an Antec NSK2480 case so I would have something to work in for now.
I need to buy RAM next, and assume I want to shoot for the lowest latencies as it appears was done here with kingston hyperx, and not valueram:
5-5-5-11
http://www.audioasylum.com/forums/pcaudio/messages/7/72016.html
another looks lower here:
5-3-3-9-24-1T
http://www.audioasylum.com/forums/pcaudio/messages/7/74155.html
I could not find:Corsair CM3X160C9DHX using an internet search.

Since I'm doing video also, I figure I'll have to keep using another 1GB stick as before,even after loading xp. I don't see hyperx in 1GB sizes but I'm assume getting 2GB would sound worse since people seem to like 512KB, which I don't see at all in DDR3.
Is this what I should get?
Kingston KVR1333D3N9/1G 1GB 1333MHz DDR3 Non-ECC CL9 DIMM?

 

H61M-S2H works fine with Julia@, posted on November 22, 2011 at 07:47:06
jackwong96
Audiophile

Posts: 219
Joined: April 20, 2010
 and there was no degrade in SQ when compared to USB.

Note that this board has Rev. 1.2 and 1.3. The 1.3 version has one choke and some components removed around the CPU, don't buy.

My suggestion for selecting a MB, apart from other requirements, is to look at the CPU power circuit. 6 or more chokes around the CPU is preferred.

 

RE: Non-Native PCI Mobos, posted on November 22, 2011 at 08:16:31
jackwong96
Audiophile

Posts: 219
Joined: April 20, 2010
 If you must change your MB, it seems your choice is very limited and the H61 MB should be a good choice. It advantages are non-presence of unnecessary/unwanted peripheral (only 2 SATA and 2 USB controllers), hence less current consumption. I will listen to Julia@ on the H61-S2h and H55-UD2H respectively and compare with my USB converter to confirm if there is any degrade of SQ.

The H55M-UD2H is a fine MB. I had no intention to change it until I tried the H61 MB, but I'm looking at H67MA-UD2H-B3, a discontinued top-of-the-line product with an 8 phase (12 chokes) CPU power circuitry. It should sound much much better than its elementary counterparts.

I like the latest Sandy Bridge chipset/CPU because the current consumption is reasonable, hence quality power supply is made more affordable.

Regarding the H55M MB and i3 530/540, I just found some second-hand H55M-USB3 in China, which was a top model with low Rds(on) mosfets, similar to the H61/67 series. Since the power circuit is superior to UD2H, it may sound better. The MB is being shipped. I'll try it when I get it.

I guess our CMP machines are moving on as new CPUs/Chipsets roll out and products discontinue. I'm looking forward to the 22mn CPU technology coming next year. I guess we may be able to power the 5V rail with 500ma of current.

 

Much obliged! , posted on November 22, 2011 at 08:47:59
jackwong96
Audiophile

Posts: 219
Joined: April 20, 2010
 Rick and Theob, I do hope the skills will not be forgotten and be useful to someone who needs to diying speakers. I'll dig out my hardware(I kept 30 years of electronic waste material and reuse them as necessary) and refresh myself on the procedure. Need to do some calculation, make a simple set up for demonstration and photo taking. Wait one week for the necessary works to be done, and in the meantime I am still playing around with the new H61 MBs.

 

RE: H61M-S2H works fine with Julia@, posted on November 22, 2011 at 09:55:58
Mihaylov
Audiophile

Posts: 355
Location: Moscow
Joined: March 11, 2010
 Thank you, Jack! Good news.

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

 

How to Power the Memory - Linear Regulator, posted on November 22, 2011 at 10:04:21
jackwong96
Audiophile

Posts: 219
Joined: April 20, 2010
 Just for fun whilst playing around with the H61 MB. The ram was powered by a linear Reg. LT1083, and the power taken from the 3.3V rail of the motherboard.

The MB is H61M-S2V. Its front end is short, therefore, the reg. can be soldered directly to the board (the choke powering the ram needs to be removed). Output pin to the output of choke, Adj pin to Gnd of Cap and Input soldered by a wire to P24 3.3V. Yes, the ram is drawing more than 2.1A of current, the heat sink in the photo is minimum size.

To improve quality of the power source, a choke (those found around the CPU) can be added in series to both the input and output of the reg. respectively. In addition, quality caps should be added in parallel to the both input and output pins. This would greatly reduce switching noise of the 3.3V rail and moderately reduce dissipation of the reg.
Under this config, the reg. output 1.25V which is sufficient for most ram to be used in H61 boards (my Kingston ValueRam will start up at 1.1V). For H55M-UD2H, 2 resistors should be added to adjust the voltage of reg. to 1.3V.


The 3.3V rail has a low usage. Most ATX PS can supply the additional 2.1A for the ram w/o problem. Just try it out as an experiment to hear the difference.






As an aside, have you ever think of a hybrid ATX to linear reg. config. i.e. The ATX 12V powering a 5V reg.(e.g. LT series) for the MB P24, the ATX 5V powering the 3.3V rail of the P24. And more, the P24 3.3V powering a 1.3V reg. for the ram! Then add supplies for the P4 12V and the HDD. With this, you are done with linear Power supply with a bonus of linear PS for the ram. Of course, filtering chokes and caps should be added as mentioned above. This config should be easier to build than conventional linear supply and is cheaper too. You only need to deal with PCBs, probably 2, 4 regulators with heat sink, 8 resistors, some caps and chokes. Just cut off the red, orange and yellow wires, leaving all wire on the P24 plug intact. The multiple Gnd wires need not be cut, just connect the Gnd of regulators to the Gnd of P24 in the MB. No issue of powering sequence or any start up problem whatsoever.

If anyone is interested in building this Hybrid Reg., I can explain in details.


 

RE: How to Power the Memory - Linear Regulator, posted on November 22, 2011 at 10:33:19
theob
Audiophile

Posts: 3180
Location: ann arbor michigan
Joined: November 4, 2000
 Is this easily adaptable to a G31-es2l type mobo? I think this is very cool.

 

RE: How to Power the Memory - Linear Regulator, posted on November 22, 2011 at 10:40:06
jackwong96
Audiophile

Posts: 219
Joined: April 20, 2010
 Yes.

 

RE: Not only cool-looking cases..., posted on November 22, 2011 at 10:48:13
Mihaylov
Audiophile

Posts: 355
Location: Moscow
Joined: March 11, 2010
 =Is this what I should get?
Kingston KVR1333D3N9/1G 1GB 1333MHz DDR3 Non-ECC CL9 DIMM?=
Kingston ValueRAM KVR1066D3N7/1G DDR-III CL7 is better because it have lower timings (CL7): 7-7-7 and it should be work with timings 5-5-5-11 (9 or even 7) in 800 MHz.

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

 

RE: How to Power the Memory - Linear Regulator, posted on November 22, 2011 at 11:12:28
Mihaylov
Audiophile

Posts: 355
Location: Moscow
Joined: March 11, 2010
 Hi Jack!
As you think whether it is necessary to install such linear regulator for memory if the full linear PSU for P24 and P4 is used? In other words the linear power supply for memory gives a SQ enhancement in this case (with full linear psu)? And how to make it for H55M-UD2H?


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

 

PS/2 Wireless Mice Solved...., posted on November 22, 2011 at 19:18:16
GStew
Audiophile

Posts: 633
Location: NE Mississippi
Joined: September 21, 2001
 I found my collection of various USB->PS/2 adapters. I had remembered them all being the same type... WRONG! Only two were the same... and two others DID allow a USB mouse dongle to work into the PS/2 port. Not sure why some work and some don't, maybe one pin is not connected in some, but I got two now!

That explains why I did see some USB/PS/2 wireless mice sold.

Greg in Mississippi
Everything matters!

 

RE: H61M-S2H works fine with Julia@, posted on November 22, 2011 at 19:21:27
GStew
Audiophile

Posts: 633
Location: NE Mississippi
Joined: September 21, 2001
 Agree, very good news.

Also interested in what you find with the 2nd-hand board.

Just to be clear, do the H61M-S2H have native PCI via the chipset or a bridge?

TIA!

Greg in Mississippi
Everything matters!

 

RE: B65, Q65 and Q67 are the chipsets for business class systemboards., posted on November 22, 2011 at 19:40:47
GStew
Audiophile

Posts: 633
Location: NE Mississippi
Joined: September 21, 2001
 Bummer.

Thanks for the info!

Greg in Mississippi



Everything matters!

 

RE: H61M-S2H works fine with Julia@, posted on November 22, 2011 at 19:41:12
Mihaylov
Audiophile

Posts: 355
Location: Moscow
Joined: March 11, 2010
 H61M-S2H don't have native PCI.

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

 

How to Power the Memory - Part 1 Linear Regulator etc , posted on November 22, 2011 at 22:51:17
jackwong96
Audiophile

Posts: 219
Joined: April 20, 2010
 Hi Serge and fellow members of this forum,

When we start to tamper with the PWM circuit on the MB, we begin to scratch its heart and soul. We are fully aware of the improvement brought about by implementation of full linear PS in place of the switching PS in ATX format.

Would a linear power supply for memory give a SQ enhancement, even if a full linear PSU for P24 and P4 is used? Perhaps I should rephrase the question in technical terms: "Would replacement of the on-board single phase switching power supply by a linear power supply enhance SQ?" I would say most definitely. Our full linear PS implementation has proven this. Nevertheless, improvement in SQ will largely depend on the quality of the linear power supply used. My experience is that whilst powering the MB with SLA batteries + 3-terminal reg., I then powered the memory with batteries resulting in good SQ improvement. Replaced by a LT1083 also improved SQ but not to the extent of battery. When powering the 5V rail with batteries I was in heaven. As Greg said, "Everything matters!" All minor improvement work add up. Don't know how it will sound when powering both 5V rail and memory by batteries at the same time.


With regard to implementation of memory power supply on H55/H61 series MB, as well as others, I have the following options in mind, in order of superiority:

1. NiMh battery (suitable for H61 MB)
2. Shunt regulator buffered by NiMh battery (suitable for H55/57/61/67 MB)
3. Shunt regulator alone (suitable for all MB)
4. Series regulator made of discrete components (suitable for all MB)
5. 3 terminal series regulator (suitable for all MB)
6. Onboard switching regulator buffered by NiMh battery (suitable for H55/57/61/67 MB)

So far, I have tried options 1, 5 and 6 on MBs. I have also tried 2, 3, 4 and 5 on other audio circuits. Hence I can tell the difference.

Application of the above options will largely depend on how desperate you wish to improve SQ, the ease of construction and MB model/electrical characteristics.

For H61 series MB, memory can run as low as 1.1V. Therefore, I will definitely go for NiMh battery(1.25V). However, there are some variations, e.g. whether to use 2-3 D size Nimh batteries or 1 Nimh battery backup by a large LifePo battery (2 more variations here in terms of backup battery connection - whether to use resistor or shunt regulator for voltage drop, the latter would be a combination of option 1 & 2 and may be a balanced approach in terms of maintaining purity of power supply and stability of voltage).

For H55M-UD2H, pure battery is not feasible, as memory won't work below 1.3V or so. Therefore, other options should be used. For G31 MB, the DDR2 memory runs at 1.8V and is therefore not suitable for battery operation.

It will be a long story. Hence, I will divide my writing into parts and start with the easiest options.

Option 6 - NiMh battery buffer for H55/57/61/67 MB

You only need a relay e.g. 12V, a diode, four wires and a C or D size Nimh battery in a holder with connecting terminals.

(a) No need to remove the choke powering memory
(b) Fix the battery/holder and relay somewhere on the MB near to the memory using double side adhesive tap.
(c)Connect the coil of relay to 12V and ground of P24 on MB, a diode 1N4001-4007 or similar in parallel to the coil(cathode towards P24 12V).
(d) Connect the negative of battery holder to P24 ground of MB.
(e) Connect positive of battery holder to the moving arm (COM) of the relay.
(f) Connect NO pin of relay to the output point of the choke




Upon MB startup, the battery will be connected, by the relay, in parallel to the on-board memory power supply. It is simple work which can be done in half an hour. The modification is safe, will improve SQ but won't hurt the MB, just be cautious about polarity. In addition, Ram voltage should not be set to 1.45V or above, use the lowest voltage possible to run the ram. The battery acts like a capacitor under this config. and I found this buffer better sounding than bypass caps. Of course, quality caps should also be added.

Option 1 - NiMh batteries for H61/67 MB

The connection is similar to option 6 above but the choke should be removed. More batteries connected in parallel are needed for longer playing time.

A stock battery charger can be put/stick inside the case. Negative terminal of charger to Gnd of MB, Positive connected to the NC pin of relay. When MB is power off, it will charge the batteries provided that it is always connected to the AC mains.


Option 5 - 3 Terminal Linear Regulator for all MB

I'll take some photos to illustrate the construction. Please feel free to ask any question. See you in part 2.

 

RE: How to Power the Memory - Part 1 Linear Regulator etc , posted on November 23, 2011 at 06:02:32
theob
Audiophile

Posts: 3180
Location: ann arbor michigan
Joined: November 4, 2000
 I never worked with relays so I am unfamiliar with them. I assume you get a 12 v relay is that correct? What current capacity? Is it obvious what terminals are NO, COM and NC? Where is coil output? What value of voltage should batteries be? Is it critical? I think the onboard voltage regulator is still operating so battery voltage is not too critical, correct? How do you remove the choke cut or desolder? Which component is the choke on G31-e2sl?

 

RE: How to Power the Memory - Part 1 Linear Regulator etc , posted on November 23, 2011 at 07:13:25
Mihaylov
Audiophile

Posts: 355
Location: Moscow
Joined: March 11, 2010
 Thanks Jack for the detailed answer. It is very interesting. We wait for continuation.

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

 

RE: How to Power the Memory - Part 1 Linear Regulator etc , posted on November 23, 2011 at 08:42:12
jackwong96
Audiophile

Posts: 219
Joined: April 20, 2010
 Hi Theob,

Unfortunately G31-es2l uses DDR2 ram which runs at about 1.8V, at present no battery can deliver such voltage. To enjoy listening to battery-powered memory, you will have to wait until you change to H61/67 chipset MB or newer 22nm CPU next year. My apology for my unclear presentation. I've made relevant amendment.

In your case, a 3-terminal linear regulator set to about 1.8V is the way to go. I've built an attachment board with 3-terminal regulator on H55-UD2H. I also have a G31 MB and will built a general board for use in all MB including yours. Will present tomorrow.

 

Remote mice and other subjects, posted on November 23, 2011 at 08:42:38
rickmcinnis@dogwoodfabrics.com
Audiophile

Posts: 3040
Location: Atlanta
Joined: December 15, 2003
 Glad to hear you found a solution.

My experience with the remote mice seemed to involve instability. Maybe it was the USB connection and the PS/2 will be more stable but ever since I quit using a remote mouse I have never had the machine do anything strange.

With the remote mouse there were times the machine would stop and require clearing the BIOS. Who knows why? I figured I could use the exercise of getting off of the couch and having to go to the screen to use the mouse in the front of the room. I did like being able to switch polarity from my seat, though. I was using the keyboard for this before we found that cMP sounds better without the keyboard.

To use an PS/2 keyboard would require a new build and one wonders if Serge found the keyboard driver to be more troublesome than the mouse drivers? One of these days ...

The idea of the IDE "hard drive" is intriguing but unless someone says it sounds MUCH better than the SATA SSD I think we might lose as much as we gain. I like being able to easily switch SSDs back and forth between my music computer and my main computer for making changes. I guess this could be useful once all of this OS tweaking is done. I assume we could copy our final configuration to the IDE FLASH mounted in another machine?

 

Thank YOU, Sir ..., posted on November 23, 2011 at 08:48:53
rickmcinnis@dogwoodfabrics.com
Audiophile

Posts: 3040
Location: Atlanta
Joined: December 15, 2003
 We will wait patiently.

I am in the midst of trying to find good balance between the drivers of my system. I am using one of Dave Slagle's autoformers for attenuation but can see no reason I cannot do fine tuning after that with your method to find the critical balance.

Thanks, again!

 

RE: How to Power the Memory - Part 1 Linear Regulator etc , posted on November 23, 2011 at 09:08:55
theob
Audiophile

Posts: 3180
Location: ann arbor michigan
Joined: November 4, 2000
 Thanks Jack...take your time I appreciate it.

 

Project for the long weekend - WHICH battery to use?, posted on November 23, 2011 at 10:15:25
rickmcinnis@dogwoodfabrics.com
Audiophile

Posts: 3040
Location: Atlanta
Joined: December 15, 2003
 I keep going in circles trying to find the best battery for this.

I will start with a battery and then segue into a regulator when Brian Lowe finishes work on his low voltage regulator.

Mr. Wong, can you give me a recommendation? I figure the most capacity I can find - I have seen low discharge batteries of 10,000 mAH - is that what to use?

Thanks,

 

H55M-USB3 experiences and mobo question, posted on November 23, 2011 at 12:07:00
Douwe01nl
Audiophile

Posts: 71
Location: Amsterdam
Joined: September 29, 2011



Hi Jack, great to read your very interesting posts on MB's!

I changed to H55M-USB3 about a year ago, and chose this MB because of the cpu-power circuit. I must say i was a bit disappointed in SQ. At that time i moved from a medieval Asus-AMD-K7 (Athlon thunderbird) which runned @800 MHz. This board had a few interesting features, like no onboard sound, no onboard lan, adjustable VGA signal strength, very detailed BIOS. I bypassed all caps with Sikorel/Wima combo's (see photo, as per your directions, thanks a lot!), which had great influence on SQ. Of course this board couldn't do CPlay and i did SoX-upsampling by hand. This was all aimed on learning and trying to squeeze everything possible out of this 1999-MB. As a preparation for building my "real" CMP2.

The change to H55M-USB3 gave a lot nice improvements (RAMspeed, headless-opportunity, Cplay, upsampling, lower powerconsumption / battery-opportunity).
But i had expected more in terms of SQ. SQ was only a little better than the Asus/K7 board.
Other experience:
- changing CPU-speed on the h55m makes a lot of difference (900 lots better than 1200),
- playing with two cores it seems to sound tiny better than 1 core; just a little more depth and headroom. How could that be possible? larger L2 cache?

I think the H55M-USB3 has good potential, but quiet a few needless noisy & powerconsuming chips have to be removed: IDE-chip, usb3-NEC, hdmi, dvi, sound, fan regulators. I think these chips overrule other improvements i did, for example on Juli@: just subtle improvements when linear psu-ing 5v and LifePo-ing 3,3v. Not the 'quantum leap' that other inmates describe. (MB has hybrid psu: linear+pico)

Since chip-removal is out of my league, and i don't need 192 KHz upsampling (88.2 is enough; my modded Benchmark-DAC1 upsamles to 176) i have a few questions:
1.
For going headless / 88.2 upsampling would it be possible / better to move to Atom?
Possible advantages:
- Asrock Atom D525 board, underclockable to 900 Mhz; fanless when underclocked
- simple design and few features (elegant)
- power consumption and -regulation is easier to control
- battery-power would be more simple
- removal of the two big regulators near CPU and battery-feeding (or replace the 78xx regs by lowdrop regulators)??
- not so many caps to bypass with Sikorel-Wima-combo
Possible disadvantage could be the small L2+L3 (1M+0M)




2.
Or should i stick to the h55m and try to cut the power-pins of the mentioned chips (and leave them on the board)?

Thanks in advance!

Douwe

 

RE: SSD @ 2,6V, posted on November 23, 2011 at 14:07:34
Douwe01nl
Audiophile

Posts: 71
Location: Amsterdam
Joined: September 29, 2011









Hi Greg,
First the SSD: i fried it. by too often pulling it out an putting back in while mobo is running. Rick's warning came too late. So forget this "nice soundpop-no-ssd-trick". Indeed there's no ssd anymore...

1. Undervolting SSD
This is how i worked. I solderd wires to the 3,3v pins of a spare plug. SSD didn't work. So i connected the 5v line and just lowered voltage until it stops working: at 2,6v it could start. LifePo's can deliver nice clean 3,3v. it works. The lower the better sq.
Cut the 12v line. This will be regulated down to 5v, as you can see on the foto (backside ssd) (link to site at bottom of post). On front- and backside foto's i think i see 3,3v is not connected to rest of the board. I think the clock needs the 2,6v. Controller, cache and ram could do with 1,3 - 1,5v i think (but do not know).
Compared to Mihaylovs setup: a lot more ram-chips and a cache chip to feed and control. I will try 1-2GB ssd in future.

2. CCFL inverter
Why does it affect sq? This inverter takes 12v from the monitor, takes it up to 800v and then throws the garbage on the shared ground. Very dirty. Monitor-ground is connected to CMP-ground.
On the pictures you can see i just pulled the inverter-plug. Be sure before you do this, to set all colours to 100%. This minimizes regulation-noise.
One of the two ccfl's is connected to a "dedicated" inverter (€6,-). The other is not connected at all, one is enough. Brightness and contrast regulation not possible anymore (same for colours). It helped in my setup. There is no 'start-up'logic, just 12vto 800v. I don't know if you could replace it with a linear supply, i wouldn't (too dangerous for me).
Monitor-psu is now: 13,5v SLA -> inverter to 220v -> Dell smps 12v. I'm planning to go headless. If not i'll try the 13,5v sla directly to monitor. Hope this will work and not kill monitor. What would you think?
I didn't try the leds: more expensive and less result.

3. Battery-sourced inverters
At one time i ran into a cheap 2nd hand set of sla's and sinus-inverters. So i just gave it a try. It turned out well. Of course the inverters produce some ripple, but fridge-washingmachine-computers-etc-ripple is eliminated. BTW: i also tried cheap Aldi/Lidl blockwave-inverters. They produced better sq for the digital parts (mobo, cpu, ssd), due to lack of sinus-filter. Analogue parts (DAC) need of course sinus-inverter.
Note that i'm a battery-freak, and i like the thought of running CMP2 on solar-power...
If you aren't, wait another 1-2 years, and run the whole setup on LifePo's.

please send some feedback if you tried the ssd and/or ccfl mods.

Douwe

 

RE: Project for the long weekend - WHICH battery to use?, posted on November 23, 2011 at 18:29:01
jackwong96
Audiophile

Posts: 219
Joined: April 20, 2010
 Hi Rick,

I recommend Powerex D Size NiMh battery. Once, I did a HiMh batteries comparison for use in my DAC. Sanyo(poor self discharge rate) and GP sounded poor. Energizer and Uniross were good sounding. Powerex was the best during my tests, with the lowest internal resistance hence good frequency response & overall SQ, very low self discharge rate.

http://www.mahaenergy.com/store/viewItem.asp?idProduct=432
http://www.mahaenergy.com/store/viewItem.asp?idProduct=342

I have not tried the IMEDION and don’t know whether it would be better than the other.


For h55m-ud2h running battery powered memory, 2 batteries in parallel will offer longer listening time, use a ram that will boot up at the lowest voltage.

When dealing with the battery holder and wiring, try to make the wire in direct contact with the battery terminals. Used good quality wires, preferably AWG 16# -14# (1.3mm or more in diameter). If you can’t or don’t want to stick the battery holder onto the mb, put it in front of the P24 socket.

Wish you a successful trial and nice weekend.

 

RE: H55M-USB3 experiences and mobo question, posted on November 23, 2011 at 19:24:52
jackwong96
Audiophile

Posts: 219
Joined: April 20, 2010
 Hi Douwe,

I’m pleased that my posts on MB’s are of some use to you guys.

On the choices of MB, I strongly recommend you stay away from Atoms CPU/MB. Please see the posts below.

http://www.audioasylum.com/forums/pcaudio/messages/7/74567.html
http://www.audioasylum.com/forums/pcaudio/messages/7/74547.html


The 1st photo you posted is an Asus MB, not Gigabyte. I’m not sure the bypass caps are soldered to the right place as I couldn’t see the top of MB. Just a hint, all chokes, in particular those around the CPU and Ram need bypass caps.

If you wish to change your CPU/MB, I recommend GA-H61M-S2P-B3 and Celeron G440, a low cost solution.

This MB doesn’t have many unnecessary devices, no DVI and HDMI chips, just sound and lan chips, 2 fan connectors. The bios is tweakable just like the S2V.

http://download.gigabyte.asia/FileList/Manual/mb_manual_ga-h61m-s2p-b3_e.pdf




One more good thing about the Gigabyte H61M MB is that, unlike H55M series there are SMD chokes filtering the powerline before the lan and sound. As such, disabling the chip is easy, just desolder the chokes. You can cut off the legs of the mosfets next to the fan connectors.

I will collect my 2nd hand H55M-USB3 after work today. May be worthwhile to wait for my report on this MB.

 

RE: H55M-USB3 experiences and mobo question, posted on November 24, 2011 at 00:33:09
Ryelands
Audiophile

Posts: 1867
Location: Scotland
Joined: January 9, 2009
 On the choices of MB, I strongly recommend you stay away from Atoms CPU/MB. Please see the posts below.

I cannot comment on the GA-H61M-S2P-B3 but neither of the "tests" you cite strongly to recommend staying away from Atom MoBos is meaningful as the test conditions are not described. Atom boards have known limitations in our scenario; it is easy to get them to perform badly, not so easy to get them to perform well.

The D510MO would in any case not be the Atom board of choice (though I've used one a dual-boot setup).

Dave

 

Briefly tested H55M-USB3, posted on November 24, 2011 at 09:23:22
jackwong96
Audiophile

Posts: 219
Joined: April 20, 2010
 Collected a 2nd hand H55M-USB3 in the evening. Couldn't wait to listen to it. I also borrowed an unmodified H55M-UD2H from a friend for comparison.

An Antec Earth Watt ATX PS, i3/530, SSD and USB DAC were used. First of all, listened to unmodified UD2H MB to get an idea how it would sound with stand ATX PS. Bios setting was optimized. Without a linear PS and MB modifications, UD2H was a bit dull and boring, things were hidden a bit in the background. Background noise was a bit high.

Bios setting for the USB3 was similar to UD2H, but since I'm not familiar with this MB I set the CPU and Graphic voltage 0.03V higher. When the USB3 fired up, things were a bit different. I must say that with the same PS, this MB has better rejection to poor quality of switching PS, probably due better power circuit. Everything is just better and it was more enjoyable listening to it than the UD2H MB. After a brief comparison, my initial impression was that the USB3 MB was better sounding than the UD2H MB. The major difference between the two MB seems like a better PS was used on the USB3.


 

Deep-Digital Circuit Bypassing..., posted on November 24, 2011 at 21:39:00
GStew
Audiophile

Posts: 633
Location: NE Mississippi
Joined: September 21, 2001
 Jack,

So I've run the Juli@ digital card with the additional ceramic cap bypasses in for a couple of weeks... and while it's improved, I'm still not thrilled by it. There's a bit of 'confusion' to it and a very slight sense of 'harshness' in the highs on some recordings.

Very curious about this as these are all deep-digital processing chips... the controlling CPU, the PCI/Audio controller, and some helper chips. Looking at a lot of the commercial, even high-end commercial DACs & digital playback units (SACD/DVD/CD/SD players, etc.) out there, they generally had some number of ceramic caps as PS bypasses. And the Juli@ already has all of those ceramic 'BCxxx' caps.

Darned... it makes a LOT of sense to me to use very low-inductance caps mounted in a way to keep that inductance low around these deep-digital processing chips. And the ceramics seemed to fit the bill AND be the industry standard.

So my next experiment will be to do another Juli@ digital that's identical to this one except these bypasses will use some small Oscons... I have some 33uf/16v that will work well here and are small enough to fit without a lot of hassle.

Continuing my research, I went looking for situations where people used caps other than ceramics for digital chip bypasses... and found these:

http://www.diyaudio.com/forums/digital-source/186245-dac-2496-ak4393-dac-kit-cs8416-ak4393-5532-a-5.html#post2622112 ("PPS film (again if you can get them) bypassing on the digital electrolytic's and those massive power filter caps")

http://www.diyaudio.com/forums/digital-source/142562-microsd-memory-card-transport-project-post2594576.html?highlight=pps+capacitor#post2594576 (for putting non-SMD PPS caps on the regulator outputs on their SD Card Player)

http://www.diyaudio.com/forums/digital-line-level/79452-building-ultimate-nos-dac-using-tda1541a-81.html (he also has some interesting regulators that filter both the rail and the ground... very curious about these!)

And looking closer at the AckoDAC & the RAKKDAC, I now suspect they both also use some SMD poly caps around their digital circuits, at least on the digital side of the DAC.

Of course, a number of them also use some tantalum caps there too (as does the Jul@ around the AK4114 digital input-output transceiver and the ADC & DAC & output stages on the analog card). I'm not ready to go there yet, but I don't have a problem with PPS SMD caps.

Now I'll also take the one with the ceramics, pull them off, and put on 1uf PPS caps (which I just happen to have a small supply of here).

I also found some people who swear by using BG NX caps as local bypasses, at least around DACs. I agree that's a good bet, but most people won't be able to duplicate that and I'm trying to come up with mods that can be duplicated with easily-available components, so I'm not planning to try my limited supply of those no-unobtainable caps on my Juli@ digital section.

I still owe some replies on this thread to Uncle Leon on his latest capacitor work and to fork. I hope to get to them this week.

Later!

Greg in Mississippi

P.S. Another interesting poly cap trick is discussed in this thread: http://www.diyaudio.com/forums/digital-line-level/196474-ess9018-try-new-try-more-3.html where Bunpei replaces the exotic TP shunt regulators for the Buffalo-II's AVCC with simpler & more pedestrian LT1763s (Actually a Toshiba TAR5SB which is a near equivalent) & a 100uf/800 polyprop cap from CDE on the output & likes it a LOT in comparison!


Everything matters!

 

More interesting HW stuffs PLUS Questions for Mihaylov & Rick McInnis..., posted on November 24, 2011 at 22:06:27
GStew
Audiophile

Posts: 633
Location: NE Mississippi
Joined: September 21, 2001
 First, I had mentioned the EXA-U21 USB interface in my "Further Juli@ Follies" thread as what is likely THE hot USB->I2S interface for computer audio for a DIY'er. But there's another available that looks to also be very good and less expensive:

http://www.diyaudio.com/forums/digital-source/188902-xmos-based-asynchronous-usb-i2s-interface.html

and another that might also be a good bet:

http://www.diyaudio.com/forums/digital-source/185761-open-source-usb-interface-audio-widget.html

Then there is the K&K I2S interface I mentioned that uses PS Audio's HDMI-cable-based standard. Again, the coolest thing about it is that they make a sender that is designed to mount on the Juli@ digital section. Well there's now another option:

http://www.diyaudio.com/forums/twisted-pear/201106-introducing-bit-teleporter.html

It does provide some level of isolation between the source and receiver and there is no ground connection between them. This means it might work better than K&K's when used with a DAC other than K&K's RAKK DAC.

And the Twisted Pear people mention in that thread that they are also working on a USB->I2S interface that will likely compete in quality with the EXA-U21, so that will be another option in the near future.

Finally, a couple of questions...

Mihaylov, you have info on Juli@ clock upgrades on your website. It looks like you've done at least two upgrades, both of them changing just one of the clocks on the Juli@. Were the Juli@ cards still able to work for all of the standard sampling rates( 44/48/88/96/176/192) after the clock upgrade?

And for Rick McInnis, I remember you have a RAKK DAC. Did you upgrade to the latest MkIII version? If so, I have a couple of other questions for you.

Later!

Greg in Mississippi
Everything matters!

 

RE: PS/2 Mice & other recent interesting HW comments..., posted on November 24, 2011 at 23:15:28
audiodan@tiscali.it
Audiophile

Posts: 108
Location: Cagliari - Sardinia
Joined: July 7, 2009
 Hi all
I think I've been first in use of 2100T + H67MA UD2H B3 (now the same mobo is product with newer Z68 chipset)
http://db.audioasylum.com/mhtml/m.html?forum=pcaudio&n=89089&highlight=audiodan@tiscali.it&r=&search_url=%2Fcgi%2Fsearch.mpl%3Fsearchtext%3D%26b%3DAND%26topic%3D%26topics_only%3DN%26author%3Dig%26date1%3D%26date2%3D%26slowmessage%3D%26sort%3Dscore%26sortOrder%3DDESC%26forum%3Dhug
and I can confirm his wide sonic superiority respect to my previous H55. I'm thinking that linear PSU couldn't be necessary now.
We all must approach the new HW problems in the future developement of cMP2 because PCI is dead forever but mainly XP will be no further supported in brief by new components. I suppose that new asincronous USB sound card, if supplied by good ASIO driver, could solves many problems about good and easier future cMP2 systems, but which OS will can use?

 

I have had the III since it came out, posted on November 26, 2011 at 07:47:01
rickmcinnis@dogwoodfabrics.com
Audiophile

Posts: 3040
Location: Atlanta
Joined: December 15, 2003
 and I am pleased with it though I have never compared it to anything else for reasons given numerous times. It makes different recordings sound different which is how I figure a component is working well. No glaze of unwanted continuity spread over every thing you listen to.

I think the I2S adapter that Dave made for the music server company is not for me. He even said I was better off with what I have. My main concern is the power supply. I think we can do better on our own.

I remain completely unconvinced by USB. I am sure it has got better than when I first began using it but it was such a revelation moving to JUL@ and that has not stood still either with what we have all learned.

Not that I would mind of it turns out to be the killer format. But I will wait for the ole preponderance of eveidence before making that change.

Have you had a chance to insert one into your system or are you where I am - next to impossible to insert; which im my case there is an addition to that: I could, but the set-up would be so different as to make the comparison worthless?

 

How to Power the Memory - Part 2.1 : Three Terminal Linear Regulator , posted on November 26, 2011 at 08:30:20
jackwong96
Audiophile

Posts: 219
Joined: April 20, 2010
 How to make a 3-Terminal Regulator for H55M-UD2

1. Remove the choke as per photo.

2. Cut a small piece of PCB that will fill the gap.




3. File with a filing tool to ensure good fitting.






4. Prepare two resistors with strong leads that will pass through the two holes of the choke. One of the leads is for support purpose and there is no actual connection to the circuit.




5. Put the heatsink on the PCB to mark position. Do not put it too close to the P24 socket.



6. Drill holes for LT1083 and chokes




7. Place component on PCB. (I forgot to put resistor to adjust the regulator to 1.3V). Anyway, you will see the placement of components and connections more clearly in the next part.




8. Soldering. The resistor lead near the middle of the PCB is only soldered to the Mobo to provide support. There is no connection to the circuit in the PCB.




9. Connect In to 3.3V and Ground to P24.




Please see Part 2.2 for the circuit diagram. In part 2.2, an external supply will be built.

 

How to Power the Memory - Part 2.2 : Three Terminal Linear Regulator , posted on November 26, 2011 at 10:32:30
jackwong96
Audiophile

Posts: 219
Joined: April 20, 2010
 In Part 2.1, I built a regulator board attached to H55M-UD2H to power the memory, but I never listened to it singing because I used batteries.

In this Part, an external regulator board is built for general use with all Mobos. A GA-G31M-ES2C Mobo is used for the test. For your interest about the listening test, it is at the bottom of this page and I recommend you read about the results first.

Construction

1. Prepare components, mark position/holes with a pencil and dill holes.



2. Circuit diagram and components layout




0.1uf cap should be added to all caps. You can also add the cap that I crossed out for even better filtering effects.









3. Remove the choke from the Mobo. May need to use solder iron to press out the choke from the solder side.



4. Connect wires from P24 3.3V and Ground, as well as from the choke output.





5. To power the DDR2 ram, tune the VR to maximum resistance, i.e. 50 ohm combined.

6. Connect the wire to the regulator board. If you're not sure whether the regulator will work, connect it to a power source e.g. a 6V battery or other power supply and check the output.




Listening test.

SSD powered by batteries, Antec Earthwatt 430, battery powering USB isolator and stock Musiland 01USD, battery driven AD1865 DAC, 45-tube power amp. CPU running at 960Mhz, forgot what CPU was used, must be a low-end one. Bios optimized.


The Mobo with DDR2 Ram booted up when the regulator was about 1.53V, with power taken from P24 3.3V. I miscalculated the resistor values(this set of resistors, i.e. 200/50 ohm, will output 1.53V not 1.8V), but never mind, it booted without problem.

Few day ago, I listened to this Mobo while doing some tests and I wasn't pleased with its SQ at all. Last time I assessed some Mobos, I graded this Mobo slighting above my Atom Mobo and the H55-UD2H was much better sounding.

With the implementation of linear regulator for memory, it sounded so much better than I could image, it is way out of expectation. Running the ram at lower voltage (1.5V) has some bearing on SQ(standard is 1.8V).



I further reduced the voltage to 1.3V whilst CMP was running. Power off
and then power on again. It booted up without problem at 1.3V. Current was about 2.3A.



At 1.3V, I was getting better music. Some noticeable improvement, minor details was clearer and background noise was lower. Under this set up with standard switching PS, no separate P24 12V and stock USB converter, I would rate the Mobo (modified with bypass caps) sounding very close to stock H55M-UD2H or perhaps better in some aspects.

I strongly recommend building a linear regulator for the memory of your Mobos. Theob, you should try. It is not complicated at all but it really makes a difference and you will be paid off. I believe the memory power circuit is more important than the rest in the mobo e.g. CPU core and graphic, it is because wave files are loaded to the memory for playing and it is the source of what we will be hearing. Therefore, a noise free environment is paramount.

Boards like H67M-UD2H-B3 also has a single phase switching power circuit, albeit using low RDS(on) Mosfet, hence will sound better. Nevertheless, it is still a switching regulator with that particular sound signature. The 2 chokes before and after the regulator help filter out much of the switching noise and listening was enjoyable, even when no linear regulator was used in the front end. Powering the regulator by P24 3.3V is convenient but not mandatory.

In Part 3, I will skip Option 4-"Series regulator made of discrete components" as I never have liked it. Instead, I'll go to Option 3 to build a sub-voltage shunt regulator(my favour), to power the memory. The design goal is an adjustable shunt regulator capable of delivering voltage as low as 0.6V to 1.3V and with more than 2A of current delivery capability. I've sourced the necessary components and is waiting for its arrival.




 

Thanks Jack!, posted on November 26, 2011 at 11:07:34
Mihaylov
Audiophile

Posts: 355
Location: Moscow
Joined: March 11, 2010
 This is very very interesting and useful. Continue please.

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

 

RE: More interesting HW stuffs PLUS Questions for Mihaylov & Rick McInnis..., posted on November 26, 2011 at 11:25:38
Mihaylov
Audiophile

Posts: 355
Location: Moscow
Joined: March 11, 2010
 =Were the Juli@ cards still able to work for all of the standard sampling rates( 44/48/88/96/176/192) after the clock upgrade?=
- Yes, Greg. But the sampling rates 44.1/88.2/196.4 use internal stock clock resonator on the board of Juli and sampling rates 48/96/192 use external clock (I use LCAudio clock as world clock for Juli and CA DAC Magic due to that LCAudio clock have two outputs). My CA DAC Magic upsample any input signal to 96 KHz so I use only 96 output sampling in cPlay because I think that SoX in cPlay do upsampling better than DAC Magic.

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

 

RE: How to Power the Memory - Part 2.1 : Three Terminal Linear Regulator , posted on November 26, 2011 at 11:52:09
theob
Audiophile

Posts: 3180
Location: ann arbor michigan
Joined: November 4, 2000
 None of your pictures came through on my pc. Hope its not a problem on my side.

 

RE: How to Power the Memory - Part 2.1 : Three Terminal Linear Regulator , posted on November 26, 2011 at 12:17:39
jackwong96
Audiophile

Posts: 219
Joined: April 20, 2010
 Rectified.

 

RE: How to Power the Memory - Part 2.1 : Three Terminal Linear Regulator , posted on November 26, 2011 at 13:12:14
theob
Audiophile

Posts: 3180
Location: ann arbor michigan
Joined: November 4, 2000
 Thanks Jack...looks fine now. Remember I have the G31-E2sl mobo. I remember a aseries of posts between you and Gene wherein you helped Gene do this but I had a difficult time following Gene's pictures/markups...didn't appear clear on my pc. But after I figure out what I have to do ... I would like to do this mod.

 

RE: More interesting HW stuffs PLUS Questions for Mihaylov & Rick McInnis..., posted on November 26, 2011 at 13:14:11
jackwong96
Audiophile

Posts: 219
Joined: April 20, 2010
 Greg,

Just want to let you know that stock Musiland 01USD sound slightly better than stock Julia@, when both were powered by the Mobo. If you like the SQ of Julia@, I think you would like Musiland 01USD too, and it has many rooms for modification. You have not considered Musiland probably because it has no I2S output, but I can tell you that I2S can be taken from the chip but it would involve very careful soldering work. I've done it on a Musiland 01US.

Stock Musiland may not sound best, but the chip it uses is similar to Hiface. Mine sounds very good after modification but I have nothing to compare to apart from Juila@ which remains unmodified. I bought it in China @USD55. It is a good choice for modification.

 

G31-E2SL is similar to G31ES2C. Same modification in Part2.2 applies. , posted on November 26, 2011 at 13:23:15
jackwong96
Audiophile

Posts: 219
Joined: April 20, 2010
 .

 

RE: cMP - the open source high-end Memory Player, posted on November 27, 2011 at 02:39:46
twlai
Audiophile

Posts: 25
Joined: August 6, 2011
 Hi,

I have recently purchased a pack of LiFePo4 battery which provides 3.3V*4=13.2V. What should be the ohm (R) and watt (w) of the resistor connected in series with the line feeding the CPU (P4) at 12V?

Regards,

Peter

 

LiFePo4 Battery from 13.2V to 12V (or 11.8V), posted on November 27, 2011 at 02:42:16
twlai
Audiophile

Posts: 25
Joined: August 6, 2011
 Hi,

I have recently purchased a pack of LiFePo4 battery which provides 3.3V*4=13.2V. What should be the ohm (R) and watt (w) of the resistor connected in series with the line feeding the CPU (P4) at 12V?

Regards,

Peter

 

RE: LiFePo4 Battery from 13.2V to 12V (or 11.8V), posted on November 27, 2011 at 03:50:12
jackwong96
Audiophile

Posts: 219
Joined: April 20, 2010
 Hi,

What ohm (R) of the resistor depend on what mobo you're using and whether you're powering the P24 12V or P4 12V. For typical 500ma on P4 12V, a 2 ohm 5W resistor will do the job. Or you may plug the battery in directly to the P4 12V and leave it there. Yes, the P4 12V is shut off when the Mobo is powered off, hence no current running and no need for a switch for power on.


Or may be a good idea to use low dropout 3-terminal regulator to stabilize voltage at 11.5V(within 5% Tolerence of 12V).

 

RE: LiFePo4 Battery from 13.2V to 12V (or 11.8V), posted on November 27, 2011 at 04:01:03
twlai
Audiophile

Posts: 25
Joined: August 6, 2011
 Any 3-pin LDO regulator suggested?

I also want to use a capacitor bank (http://www.ebay.com.hk/itm/220899802413?ssPageName=STRK:MEWAX:IT&_trksid=p3984.m1423.l2649#ht_1522wt_901) and the diodes will be replaced by 2 wires.

Therefore:-

LiFePo4 battery 13.2V --> capacitor bank --> LDO regulator (Belleson?) --> CPU (P4)
LiFePo4 battery 13.2V --> capacitor bank --> LDO regulator (Belleson?) --> pico-PSU --> MoBo (P24)

I will report whether SQ will improve as compared to Linear PSU.

P.S. my LiFePo4 battery is bought in China.

 

RE: LiFePo4 Battery from 13.2V to 12V (or 11.8V), posted on November 27, 2011 at 04:20:14
twlai
Audiophile

Posts: 25
Joined: August 6, 2011
 Hi Jack,

I also live in Hong Kong and also a civil servant!!! What a coincidence!!!

May I e-mail you asking more questions about Computer Audio?

 

RE: LiFePo4 Battery from 13.2V to 12V (or 11.8V), posted on November 27, 2011 at 04:40:54
jackwong96
Audiophile

Posts: 219
Joined: April 20, 2010
 LiFePo4 has very low internal resistance, in mohm, so only very high speed small cap can help them further. The caps on the ebay link do not appear to be high grade ones, I don't think they'll help. LiFePo4 is alread the best choice.

If you want to improve SQ, invest on the 5V rail which is in the audio chain and thus most critical. P4 12V powers the CPU core and graphic etc and has a secondary role. The CPU core is somewhat like the engine of a car but the ram and I/O circut are like a ECU (the brain) which programs and direct how the engine will run.

LDO ICs with 3A or more current delivery capability, such as LT1084, 1085 & 1086 can do the job. You may try using high quality regulators, but I don't know whether they will improve SQ. Please give it a try, as I want to know the result.

 

RE: LiFePo4 Battery from 13.2V to 12V (or 11.8V), posted on November 27, 2011 at 04:47:40
jackwong96
Audiophile

Posts: 219
Joined: April 20, 2010
 Yes. No problem. For general questions about CPLAY/CMP/Computer do share in this forum so that everyone will benefit. We can go out for tea and have a discussion, you might have seen me before!

 

RE: How to Power the Memory - Part 2.2 : Update , posted on November 27, 2011 at 04:56:35
jackwong96
Audiophile

Posts: 219
Joined: April 20, 2010
 Further to my tests with 1.3V, the DDR2 800 1G Ram will boot up at 1.25V. The sound is mellow at this voltage and the Mobo sound like a tube amp.

I also tested the regulator board on the H61 Mobo and there was good improvement, but not as drastic as the G31 Mobo. The stock H61 Mobo sounded thinner than the G31 but is more analytical. After the regulator is used, H61 Mobo sounded thicker and more balanced.

 

RE: cMP - the open source high-end Memory Player, posted on November 27, 2011 at 05:26:14
twlai
Audiophile

Posts: 25
Joined: August 6, 2011
 Small cap!? You are right as core audio technology also say so: http://www.coreaudiotechnology.com/blog/?p=24.

I should use Oscon cap. What value (voltage and farads) would you suggest?

 

RE: LiFePo4 Battery from 13.2V to 12V (or 11.8V), posted on November 27, 2011 at 05:32:43
twlai
Audiophile

Posts: 25
Joined: August 6, 2011
 Small cap!? You are right as core audio also suggests so: http://www.coreaudiotechnology.com/blog/?p=24.

Therefore, I should use Oscon. What value (Voltage and Farads) should I use?

Is this one OK: http://www.ebay.com.hk/itm/68Pcs-220uF-63V-korea-cap-Power-supply-board-kit-sc-/200619648046?pt=LH_DefaultDomain_0&hash=item2eb5dce42e#ht_2354wt_901? I can replace all cap with Oscon!!!

 

RE: LiFePo4 Battery from 13.2V to 12V (or 11.8V), posted on November 27, 2011 at 08:00:42
jackwong96
Audiophile

Posts: 219
Joined: April 20, 2010
 Oscon is not good enough, I recommend Siemens Sikorel. 100-220uf 16V will do the job, bypassed by 0.1uf MKP cap.

I recommend buying from Taobao.com rather than Ebay, cheaper.


The source.
http://search.taobao.com/search?q=sikorel&searcy_type=item&s_from=newHeader&source=store&ssid=s5-e&search=y

 

RE: LiFePo4 Battery from 13.2V to 12V (or 11.8V), posted on November 27, 2011 at 14:42:15
twlai
Audiophile

Posts: 25
Joined: August 6, 2011
 Thank you, Jack.

I understand that the cap should be connected in parallel. How many pieces do I need?

How should I connect the bypass cap? Should I also connect it in parallel but closest to the load?

Regards

 

RE: LiFePo4 Battery from 13.2V to 12V (or 11.8V), posted on November 27, 2011 at 17:44:07
jackwong96
Audiophile

Posts: 219
Joined: April 20, 2010
 Battery supply is quite pure and has no ripple. Hence big caps are not necessary. They have noise however. The small caps are for absorbing battery noise and enhancing high frequency response. Therefore, quality rather than quantity matters. 100-220uf + 0.1uf will be fine. I used 22uf 63V Sikorel, just happen to have such value and they work fine.

One suggestion is to connect the caps to the bottom of the Mobo, and in this, beneath the P4 12V socket.

See link.
http://www.audioasylum.com/forums/pcaudio/messages/7/75904.html

 

RE: LiFePo4 Battery from 13.2V to 12V (or 11.8V), posted on November 28, 2011 at 04:24:02
twlai
Audiophile

Posts: 25
Joined: August 6, 2011
Thank you, Jack.

I now decide to connect in the following way:

LiFePo4 (13.2V) --> 1.5R, 5W resistor in series (later change to Caddock) --> 220uF 16V cap (later change to Simens SIKOREL125) in parallel --> 0.1uF cap (later change to MCap) in parallel --> CPU (P4)

I try to avoid using any IC in the circuit as far as possible.

I have seen your mod on the mobo. Wow, great great work!!! I also wish to have my mobo mod like yours!!! I now use gigabyte 890GPA-UDH3 which is a AMD system as AMD has a lower SMI latency value.

Thank you for the advice again.

 

Thanks, again., posted on November 28, 2011 at 08:45:35
rickmcinnis@dogwoodfabrics.com
Audiophile

Posts: 3040
Location: Atlanta
Joined: December 15, 2003
 Did not get to do it this weekend but there is always the next!

 

cMP - Minlogon, posted on December 2, 2011 at 00:05:09
audiofanatic
 Reading thru various posts about implementing minlogon I still do not understand why/if minlogon should be applied after all registry and other changes? Does it make a difference when I apply minlogon first and then do all the changes?

 

RE: cMP - Minlogon, posted on December 2, 2011 at 00:44:27
Ryelands
Audiophile

Posts: 1867
Location: Scotland
Joined: January 9, 2009
 Does it make a difference when I apply minlogon first and then do all the changes?

Don't see why - I use an imagefile of a "vanilla" cMP2 install when building a new system to save doing the same old c**p every time. All the basic "optimisations" have been done including the minlogon and AWE changes but no drivers are loaded. For a new box, I copy the imagefile to the boot partition, boot up and load the chipset, LAN and DAC drivers. It works fine.

HTH

 

Can cMP decode DVDs?, posted on December 2, 2011 at 08:53:18
Beetlemania
Audiophile

Posts: 1217
Location: Utah
Joined: November 1, 2003
 I've tried searching this unweildy thread without luck. I need a PC mostly as a music server but plan to occasionally use it to watch music DVDs or just plain DVD movies (no HD such as purple-ray). Can a fully configured (fanless) cMP do that?

 

No., posted on December 2, 2011 at 09:04:09
Mihaylov
Audiophile

Posts: 355
Location: Moscow
Joined: March 11, 2010
 Otherwise it is not cMP2 ;)

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

 

One of the first things I do, posted on December 2, 2011 at 10:00:45
rickmcinnis@dogwoodfabrics.com
Audiophile

Posts: 3040
Location: Atlanta
Joined: December 15, 2003
 sometimes I will install chipset drivers first.

Do it early on since some of the optimizations will have to be repeated if you implement MINLOGON after doing the optimizations.

Be sure to make that registry change before re-starting! I have forgot to do it a few times.

 

AK4114 required for Juli@ functioning..., posted on December 2, 2011 at 16:58:14
GStew
Audiophile

Posts: 633
Location: NE Mississippi
Joined: September 21, 2001
 I tried un-powering the AK4114 digital audio interface chip on the Juli@ last weekend by lifting the power pins. After this was done, that Juli@ no longer functioned correctly. It caused a requested resource error.

I guess it doesn't matter where the I2S originates... Juli@ needs the AK4114.

OTOH, the rest of the findings last weekend were VERY positive. Will post an update later this weekend.

Greg in Mississippi
Everything matters!

 

RE: AK4114 required for Juli@ functioning..., posted on December 4, 2011 at 09:07:42
jackwong96
Audiophile

Posts: 219
Joined: April 20, 2010
 Can I find a PCI card similar but better than Juila@?

 

I'm not sure what that'd be... , posted on December 4, 2011 at 21:14:46
GStew
Audiophile

Posts: 633
Location: NE Mississippi
Joined: September 21, 2001
 Or even if there is one... especially if you are willing to modify the card.

My opinion... Few if any sound cards are designed to provide a high-quality SPDIF & I2S feed for use in a dedicated, high-end, tweaky music like ours. The Juli@ uses older, but still good chips. No other card I know of provides such easy access to the I2S feeds and removal of 90% of the stuff we don't need by just lifting off the analog portion. And it takes well to modification... and is inexpensive enough to do so with impunity.

I'm interested in hearing of alternatives. But I think that most of the development today is around USB-connected interfaces... and the EXAU21, the Luckit.biz XMOS-based unit, and the upcoming Twisted Pear unit are the best bets at this time in this realm... but it will change at time marches.

Greg in Mississippi

P.S. Listening to the Juli@ digital section with mostly Oscon additional filtering caps instead of large-value ceramics. Interesting contrast... Both are good and each has it's strong points. I'm going to delay my next report a few days to give this card (it was unused) and it's mods time to break in a bit and also swap back in the ceramic-bypassed card with a couple of tweaks to try and reduce it's negatives while keeping it's strong points. Couple of teaser photos attached...









Everything matters!

 

RE: How to Power the Memory - Part 2.2 : Three Terminal Linear Regulator , posted on December 8, 2011 at 12:40:04
theob
Audiophile

Posts: 3180
Location: ann arbor michigan
Joined: November 4, 2000
 Ok Jack I just may take a crack at this. I'm thinking of taking a 3.3v feed from a 2nd antec I use to run my cpu. I assume there would not be a problem with ground loops but prefer to hear your opinion on this. Also would wire length from circuit to the G31 mobo be critical? I have run out of realestate in my pc. Also to remove the g31 choke should I also use a heat gun?

I have a couple of spare mobo's and another cpu so this would be a nice project wherein I can take my time do it right and have my current mobo/cpu as a backup.

One thing regarding the heat sink do I need one that big or is that just a piece you had handy?

Also what size chokes in terms of millhenries and power? What wattage on resistors? I take it that in to the regulator is lead from left of 2RO in your picture and out to the board is the the lead (or hole) on the right of 2RO? If you are booting at lower than 1.8v (1.3-1.5v) does this make a battery feasible?

 

To Dawnrazor: My letters to you are perceived as spam again, posted on December 10, 2011 at 14:33:28
Mihaylov
Audiophile

Posts: 355
Location: Moscow
Joined: March 11, 2010
 Hi!
If the power supply of Lynx 2b is organised also as at Lynx 22 namely: analogue power for the two chips CS4396 (+5 V) and for the two chips AK5394 (+5 V) is coming from the regulator LM340S that gets a feed from system board’s power bus + 12 V, the necessary current consumption makes: 20х2=40 mA (for the two CS4396, see datasheet) + 165х2=330 mA (for the two AK5394, see datasheet) = 370 мА, i.e. 500 mA should be enough.


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

 

Nettop as an alternative, posted on December 11, 2011 at 01:19:28
Eunegis
Audiophile

Posts: 49
Joined: August 26, 2010
 Hi all,

I happend to plug an ordinary Win7 Laptop with foobar and no optimizations at all into my chain in front of the DAC (an Audio-gd NFB-2, connected through Audio-gd's Digital Interface). And I found it to sound surprisingly good in comparison with my fully optimized cMP²-system according to cics' recommendations - actually, a difference was hard to find...

That rose my curiosity, so now I've ordered a little Nettop, a small SSD and a USB-HDD with external power supply. I'll put a fully stripped XP on the Nettop, and then we'll see...

It seems to me that tweaking the software comprehensively is far more important than the hardware, except one little detail: connecting the DAC trough asychronous USB connection (NOT synchronous) is absolutely crucial.
Example: I'm listening through earphones (Bayer DT880) on my workstation-PC without any tweaks. I'm using a TeraDak 3.1D "luxury version" for it with the earphones connected directly to its line out. That one, too, sounds surprizingly clean and nice, whereas it sounds absolutely horrible and totally unlistenable if I connect a DAC (e. g. Audio-gd FUN) through its simple synchronous USB.
The TeraDak 3.1D provides asyncronous USB connection, plus it is said to reclock the signal (whatever that means for that cheapish little unit). However, the difference to a regular synchronous USB-DAC is striking.
By the way: Audio-gd's Digital Interface provides asynchronous USB as well. The fact that it is bus powered (external PSU available) doesn't make a relevant difference to me (the power seems to be refined significantly inside).
Just in case anybody wants a comprehensive explanation Play_Mate is a "powerful" source...

To cut a long story short: just for the curiosity, for learning effects and maybe to pull newbies on the cMP²-side does anyone want to share his/her experience in using Nettops/Laptops as a more convenient to handle replacement for a classical cMP²-computer?

Cheers,
Eunegis

 

Easy ferrite tweak using granite digital power supplies, posted on December 11, 2011 at 06:44:06
theob
Audiophile

Posts: 3180
Location: ann arbor michigan
Joined: November 4, 2000
 Mark here is an interesting/easy way to test ferrites (for those that haven't tried ferrites)and a nice way to use those unused granite digital power supplies. Take one of these gd's solder in some fast caps (small value oscons for example) into the pata conector (2 inners are grounds 2 outers are + 5v and 12 volt supplies). Now put some Radio Shack ferrites before and after the caps (to get a pi filter). Then put some ferrites on the ac feed to the gd. Now try the gd on a hdd in your cmp...real nice sq pop.

I replaced one of my antec fed hdd's with this setup and it is better. The antec is fully pi filtered.

 

RE: Nettop as an alternative, posted on December 12, 2011 at 14:20:26
Douwe01nl
Audiophile

Posts: 71
Location: Amsterdam
Joined: September 29, 2011
 Hi Eunegis,
About 1,5 year ago i tried cmp on an acer Aspire One ZG5 (Atom N270 and 16GB SSD inside). This was the chain:
nettop - hiface (small clock batch / LifePo4-psu) - Benchmark DAC1 (modded tentlabs clock) - Hexateq class D amp - PMC OB1 speaker
It sounded surprisingly good, but not as good as my 'classical' cmp2 setup.
Problem is (amogst others) the fan inside the nettop. There are ways to stop the fan, but it runs too hot then. Don't know about the current available nettops.

In my current CMP-machine i run Win7 in a dualboot. Win7 has great drivers, i can imagine it sounds almost as good on a nettop/laptop as cmp2.
I ordered an Atom mini-itx board (asrock ad525pv3), just to try. This board has detailed bios, and cpu can run @900MHz. I will report the results.

I think newbies could try cmp (as a dualboot together with Win7), and then decide: go cmp2 the 'classical' way (best sq, spend loads of time on tweaking), or stick to Win7 on nettop or htpc (best convenience).

Douwe

 

RE: Nettop as an alternative, posted on December 12, 2011 at 15:08:41
Eunegis
Audiophile

Posts: 49
Joined: August 26, 2010
 Hi Douwe,

thanks for your reply!
The nettop question is an interesting topic in general, I believe. And most likely relevant to many. Your statement proves that in my eyes and is exactly related ro what I'm thinking about.

I also figuered out that the fan must be the crucial part in a nettop. Besides the BIOS probably, but since nettops already run in powersaving mode by nature tweaking on that side might not be too effective, eventually.

I was wondering about several things:
- cics claims cPlay's resampling to be a very intensive calculating process. But since the Core i3 is strangled to the maximum it should work with almost any nettop CPU, shouldn't it?
- will a nettop's CPU really become too hot? I'll try it with the Lenovo S205 (equipped with an AMD E350) I'll receive soon. I hope I'll find some software to switch of the fan. I'll report on that.
- in case of effective fan control and using an SSD plus an external USB-HDD that itself is powered externally a nettop seems to be very close by nature to a desirable cMP²-system according to cics' principles, it seems to me. Some notebooks and nettops almost never spin uo the fan, so a fairly cool, fanless design seems to be generally possible. It will depend very much on the calculating power needed by cPlay's resampler since the rest of the system is suffocated anyway. WIN7 might need a lot more calculating power by itself...

More opinions please! Thanks in advance.

Eunegis

 

RE: Nettop as an alternative, posted on December 13, 2011 at 14:26:49
Douwe01nl
Audiophile

Posts: 71
Location: Amsterdam
Joined: September 29, 2011
 Laptop/notebook will never get higher than second best, SQ-wise. The Lenovo S205 is a laptop, and very difficult to tweak hardware wise (fan control by software is bad for sq and cause vibrations, hdmi-port is noisy like JackWong stated). Sorry about my remark on nettop-cpu running hot: i meant netbook-cpu. I think nettops

CMP and best SQ is all about tweaking software AND hardware AND BIOS and lots of time (and brings lots of fun, imo).

Allmost all, if not all, nettops have very unflexible bios (maybe except for FitPC?), so you should choose very carefully, read manuals etc.
otoh some inmates have very good resultswith atom or FitPC, like Ryelands:

http://www.audioasylum.com/forums/pcaudio/messages/9/97956.html

Like cics states in the guide atoms can upsample to 96, but you need core i3 SSE4 to upsample to 192KHz.

So wait for the Lenovo, try cmp, try tweaking software. If you have fun in tweaking and want more, do the next step to core i3 (or wait for my reports on Atom@900MHz, hopefully somwhere in january). Take time to listen and enjoy music.

cheers!
Douwe


 

RE: To Dawnrazor: My letters to you are perceived as spam again, posted on December 13, 2011 at 14:48:25
Dawnrazor
Audiophile

Posts: 12589
Location: N. California
Joined: April 9, 2004
 Thanks for following up here.

I think this might be the BEST way for me to upgrade to a lynx Aurora! :)

Who knows when I'll get around to it though. Busy busy busy.

Thanks again for your help!

Cut to razor sounding violins

 

RE: Nettop as an alternative, posted on December 14, 2011 at 01:09:56
Eunegis
Audiophile

Posts: 49
Joined: August 26, 2010
 Hi Douwe,

I hear what you are saying! Nevertheless: I'll try it.

I was wondering what role a netbooks batteries are playing in that game. If I leave the netbook plugged into the electric outlet while its battery remains installed, will the battery behave like a kind of power filter? It should in my imagination, since it is charged even while using the netbook, so electric power will not bypass the battery. Right?

Eunegis

 

SQ net+battery was better than battery-only - nt., posted on December 14, 2011 at 02:34:56
Douwe01nl
Audiophile

Posts: 71
Location: Amsterdam
Joined: September 29, 2011

 

Page: [ 1 ] . . . [ 29 ] [ 30 ] [ 31 ] [ 32 ] [ 33 ] [ 34 ] [ 35 ]

Page processed in 0.284 seconds.

[ Top Top of Page ] [ Contact Us ] [ Support/Wish List ] [ Copyright © 2024, Audio Asylum, all rights reserved ]
[ General ] [ Speakers ] [ Tubes ] [ Vinyl ] [ Digital ] [ Hi-Rez ] [ Video Asylum ] [ Cables ] [ Tweaks/DIY ] [ Music ] [ Films ]