|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
85.147.82.48
In Reply to: RE: cMP - the open source high-end Memory Player posted by cics on December 30, 2007 at 05:42:01
Yesterday I added ‘/nopae’ to the boot.ini file
It made a significant (!) improvement in SQ.
The improvement in SQ is in the same range as putting the extra caps on the Antec Earthwatts 430 mobo power supply leads P4 and P20/24.
Since I overlooked Cics posting on the adding of ‘nopae’ too the boot.ini file and since there also weren’t any reply’s on that post, I guess many others will also have overlooked it either.
This brings me too a contribution I could do to this project. Although I have a degree in engineering, I don’t have any knowledge of IT, electronics, PC’s and software. So I can’t make a any contributions in these area’s, but I could make a contribution in the information / manual area.
I could update Cics manuals and information on the AOB computertransport, cMP and Cplay. Since the project already runs for 2 years the information has scattered over many forumposts. That’s how I missed the /nopea information.
So Cics if you think it’s a good contribution to the project too update the information in your manuals please e-mail me.
My job can be very bussy at times, so I can’t comply too real hard deadlines. But I think it would be nice if al relevant information which is now somewhat scattered in the forum, is put together once again in a new updated manual. This way I’m not only taking from this project, but I i'm also bringing too this nice project.
ESI juli@ or LynxL22 dig i/o -> Lavry Black DA10 -> Mogami Gold -> Klein & Hummel O300
Follow Ups:
Thanks for discovering this again.
I suspect /nopea & /nopae has /nop in common (and its unique) and that's why you got same results - just an educated guess as I don't think there's a sophisticated parser at work.
/nopae works - initial few hours (at least 48 hours), HF is more forward but this settles down. A lot has changed since I last tested this.
My interest in using "/nopae" was to force the kernel into 32 bit addressing (and not 36+), i.e. less address lines. There were some issues I had but will retest.
New documentation is planned and I will get to it once cPlay 2.0 is finalised. It's looking very good with the help of other inmates. Will be great if you could assist in proof reading the material.
Hi Cics,
Thankx for reply. So my offer in helping too update the manual is already done by other inmates. I would be glad to do some proof ready. Let me know.
About the ‘/nopae’ swith voor the kernel.
This information from microsoft confused me.
http://msdn.microsoft.com/en-us/library/ms791539.aspx
It says:
“On Windows XP with SP2, when you disable DEP by using /noexecute=alwaysoff, Windows disables both DEP and PAE. This is the equivalent of using /noexecute=alwaysoff /nopae. “
Since /noexecute=alwaysoff is already in the cMP2 boot.ini file, the extra addition of ‘/nopae’ shouldn’t be necessary.
But I added '/nopae' anyhow.
I was surprised by the significant change in sound.
The difference perceived by me, is definitely not a ‘placebo-effect’ that is fooling me.
The difference in sound is just too big/significant for that.
I also label the difference as being: better.
Since the Microsoft information is confusimg, I searched for other inmates who really tried it. But it looks like Theob is the only one who really tried it.
What your opinion on the change in sound?
Do you label it as being beter ?
ESI juli@ or LynxL22 dig i/o -> Lavry Black DA10 -> Mogami Gold -> Klein & Hummel O300
Hi cics / hfavandepas
I have just added '/nopae' to my limited cMP2 rig and have achieved similar sonic results (more 3dimensional image and solid bass).
My old laptop rig consists of the following:
CPU Intel Celeron 650MHz (SSE instruction set)
RAM 192MB (max)
XP Pro SP1
cMP 1.2
cPlay 2.04 (due to my limited RAM)
regards
monaco
Hi Monaco,
Thank you for trying and feedback.
ESI juli@ or LynxL22 dig i/o -> Lavry Black DA10 -> Mogami Gold -> Klein & Hummel O300
cics added but then took away this nopae recommendation for other technical reasons.
Hi Theo,
This is exactly why I offered Cics too collect and update the scattered information on this project.
First there was this post on 2008-08-29:
Don't use "/nopae" optimisation (5.25) Posted by cics on 2008-08-29, 10:16:42
But 1,5 weeks later there was this one on 2008-09-08:
Further improvement is gained with "/nopae" optimisation (3.48)
Posted by cics on 2008-09-08, 05:00:19 (196.11.134.77)
I've edited the cMP Update post with this change (see bottom). Only use "/nopae" in the context described.
So I think many missed this post, because I see very few reply’s on this post.
That’s very strange because the difference in sound is significant.
I expected at least so see discussions on whether it is an improvement or an dis-improvement. But not even that. Nothing. No comments on SQ.
While all other (very little) tweaks are extensively discussed,
hardly anyone discusses this one on SQ.
But changes in sound (in my setup) are significant.
Sow I thought this post must be overlooked by many.
Too make a statement: in my setup it sounds not different, I think it sounds beter.
In what way beter? In the same way better, as the extra caps did on the Antec 430.
ESI juli@ or LynxL22 dig i/o -> Lavry Black DA10 -> Mogami Gold -> Klein & Hummel O300
I cannot find the update post that mentions \nopae. Could you please clarify what the circumstances are in which to use this.
Search for ‘nopae’ with the search function, and you wil find this post:
Further improvement is gained with "/nopae" optimisation (9.89)
Posted by cics on 2008-09-08, 05:00:19 (196.11.134.77)
I've edited the cMP Update post with this change (see bottom). Only use "/nopae" in the context described. .......
http://db.audioasylum.com/cgi/m.mpl?forum=pcaudio&n=36517&highlight=nopae&r=
From here you go too:
cMP update:
http://www.audioasylum.com/forums/pcaudio/messages/3/36343.html
Here you find the context described.
This is why I offered Cics toe help rewriting the documentation.
I think is have become too much scattered during time in al kinds of posts.
ESI juli@ or LynxL22 dig i/o -> Lavry Black DA10 -> Mogami Gold -> Klein & Hummel O300
I had seen those already, but it remains unclear. There is no mention of nopae in the Update. The update describes a system with Gigabyte mb, a specific E7200 cpu, and many settings including to turn Eist on in the bios (which someone reported was needed for /nopae to be helpful). Is the context the mb and cpu or is it also each and every instruction in the update or only some of them? It remains unclear if all these are necessary and on what basis all other mb's and cpus are ruled out. Couldn't it be that other combinations would benefit from /nopae, especially if they allow turning Eist on?
Writing a manual that makes things truly clear is not an easy thing. I can help if you like.
Hi Riboge,
Thankx for your response.
Why not just try it? And let your ears decide.
I agry it’s unclear.
But in The Netherlands (where I live) we have a saying: ‘Proberen gaat boven studeren’.
Which means something like: ‘trying goes above (endless) studying’
And that’s what I did: I just tried it.
Risks of trying are low.
What worse things could happen if you try ?
Adding /nopea gave a significant difference in sound.
( /noexecute=alwaysoff /fastdetect /nopea /timeres=9800 /nodebug /pcilock /3GB )
I use the word ‘different’, because different isn’t always ‘beter’.
But to my ears it is significantly beter.
In what way beter?
In the same category of ‘beter’ than the ‘cap-mod’ on the P4-powerline and the P20/24 powerlines from the PSU to the mobo I did last weekend.
(with cheap caps, no expensive ‘audiophile’ caps)
I know sometimes it can be difficult to decide if a difference is also ‘better’
And also if one isn’t fooled by ‘the placebo-effect’
But what other options are there than just to try?
So let me know what your ears are telling you.
:-)
(system: mobo Gigabyte: ga-g31m-s2l bios: F9 Processor: intel E7200 RAM: 2 x 1Gb ram Kingston low latency KHX6400D2LL/1G1GB 800MHz DDR2 Non-ECC CL4.
I don’t use the EIST function with corresponding XP power-settings, since I can’t hear any sound differences when its ‘on’ or ‘off’. But I do see an extra 1 or 2 µs latency in dpclat when EIST + XP powersettings are enabled. So for the time being it stay’s off)
ESI juli@ or LynxL22 dig i/o -> Lavry Black DA10 -> Mogami Gold -> Klein & Hummel O300
Hi hfavandepas
Great topic - The Lost Tweak !!
I mean no disrespect to anyone here, just a little pointer to my fellow cMP Comrades... Don't Be Lazy and cut-n-paste the entire parameter string from this post into your BOOT.INI file (...like I did).
You'll find the switch-in-question suffers a typo - "/nopEA" should be "/nopAE" (my uppercase, to indicate the trouble).
"Caveat" indeed !
Cheers,
Grant
That's not a Toy... IT'S A TOOL !!
Still being somewhat ashamed (understatement) because of the typo i made (/nopea),
I again looked up the microsoft help-pages. These pages make the confusion complete
(links see below)
So just keep try-ing and give feedback on what your ears are telling you.
(though keep in mind the possibility of a placebo-effect)
Links to MS help pages:
http://support.microsoft.com/kb/900524 says:
Disable PAE mode in Windows XP with SP2 and later versions of Windows XP
Add the following switches to the Windows XP Boot.ini file:
/noexecute=alwaysoff /NOPAE
For example, the Boot.ini file may appear as follows:
[boot loader]
timeout=30
default=multi(0)disk(0)rdisk(0)partition(1)\WINDOWS
[operating systems]
multi(0)disk(0)rdisk(0)partition(1)\WINDOWS="Microsoft Windows XP" /fastdetect /noexecute=alwaysoff /NOPAE
http://msdn.microsoft.com/en-us/library/ms791539.aspx says:
DEP and PAE
On Windows XP with SP2, , when you disable DEP by using /noexecute=alwaysoff, Windows disables both DEP and PAE. This is the equivalent of using /noexecute=alwaysoff /nopae.
ESI juli@ or LynxL22 dig i/o -> Lavry Black DA10 -> Mogami Gold -> Klein & Hummel O300
Hi Grant,
It's even more confusing too me now....
I ran to my cMP PC to see if the typo (/nopea) was in the boot.ini
And it was !!
:-(
Like I mentioned earlier in a post too Cics:
Microsoft writes about /nopae:
“On Windows XP with SP2, , when you disable DEP by using /noexecute=alwaysoff, Windows disables both DEP and PAE. This is the equivalent of using /noexecute=alwaysoff /nopae.”
See: http://msdn.microsoft.com/en-us/library/ms791539.aspx
But what is happening in my PC, with this typo (/nopea) in my boot.ini ?
I definitely hear a significant difference in sound quality (better too my ears)
Does the typo disable ‘alwaysoff’?
I’m always very much aware of the possibility of the placebo-effect.
That’s why I always ask others to try too.
For now let’s wait what Cics has too say
ESI juli@ or LynxL22 dig i/o -> Lavry Black DA10 -> Mogami Gold -> Klein & Hummel O300
Hi hfavandepas
Thanks for stirring up attention to this adjustment - in my setup, it makes the sound "better"... and that's another little step forward for the cMP Miracle.
As to what the MS documentation says... maybe the guy who wrote the doc hasn't talked to the guy who wrote the code... cause there's no placebo going on in my room - the change is real... and really welcome !
It's like getting new cPlay update from cics... just a little bit better every time.
Happy Listening All !
Cheers,
Grant
That's not a Toy... IT'S A TOOL !!
Thank you for bringing up the /nopae optimization. The SQ improvement is indeed too significant to be dismiss as a placebo. Soundstaging improved noticeably, and bass seems more defined.
Hi Clare t,
Cics brought it up. Not me. The honor goes too Cics.
But as Cics wasn’t sure about it, probably nobody really tried it.
There was hardly any response too Cics /nopae post.
But in my setup I think it’s better (not just different).
Thankx for trying and feedback.
Hope others will try it too and post there feed back
ESI juli@ or LynxL22 dig i/o -> Lavry Black DA10 -> Mogami Gold -> Klein & Hummel O300
.
I have now tried it with my Lenovo thinkpad. It may be a slight improvement or the same, hard to be sure, certainly not worse.
My point was your edit about 'only in this context' implies a danger in doing it in a different one, e.g., mine. Perhaps it should say it is only known to help with these components and it is unknown whether there are problems with others.
‘Caveat’ ??? Hmm wonder what that means…..
I’m not a native English speaker. So you got me googl-ing on ‘caveat’
Now I understand what you mean.
The following piece of text was not mine. It’s the search-result of ‘nopae’. It’s Cics text.
If you do a search by yourself on ‘nopae’ you will see:
Further improvement is gained with "/nopae" optimisation (9.89)
Posted by cics on 2008-09-08, 05:00:19 (196.11.134.77)
I've edited the cMP Update post with this change (see bottom). Only use "/nopae" in the context described. .......
ESI juli@ or LynxL22 dig i/o -> Lavry Black DA10 -> Mogami Gold -> Klein & Hummel O300
Sorry, I guess I was even more confused than I realized. I read your post to mean you made the edit. I guess my comment was meant for cics. And I still haven't gotten over studying Latin in school. Caveat is Latin for Beware or Be Warned. Anyway, you got me to try nopae, so something good came of it.
Hi Riboge,
No problem, can happen.
At least you tried /nopae
ESI juli@ or LynxL22 dig i/o -> Lavry Black DA10 -> Mogami Gold -> Klein & Hummel O300
what can I say, I tried it, it does sound better. thank you.
Hi Theo,
Thank you for trying and your feedback.
I hope others wil try it too and give there feedback on this kernel optimasation.
Thankx
ESI juli@ or LynxL22 dig i/o -> Lavry Black DA10 -> Mogami Gold -> Klein & Hummel O300
Post a Followup:
FAQ |
Post a Message! |
Forgot Password? |
|
||||||||||||||
|
This post is made possible by the generous support of people like you and our sponsors: