|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
174.68.103.254
In Reply to: RE: Fishy secrets part 4 or real AWE posted by steppe on January 14, 2012 at 17:18:11
"1000 is for 4096 hex" is hard to parse. Do you mean 1000 in hex equals 4096 decimal, or 1000=4096 hex? You gave a screenshot in which you used 10000 hex which is 65536 decimal. What is the meaning of that number? One might suppose it means 65.536 Kb, but that seems all wrong here. 65MB would also not seem suitable.
In any case, using 10000 as you did, I experienced none of the softness theo reports and perhaps some benefit in liveliness and detail. This is obscured by my making other changes concurrently in refining my try out of mini-xp. This is really sounding wonderful, better I think with just tweaks and fishies than xp home with all tweaks, fishies and steppes too. It is the smallest installation I have since I have not gone the xp lite route (?yet).
Follow Ups:
Is widely known as one of the most obscure tweaks.
it started probably with NT and Windows 2000 REALLY benefits from it, it's not only attested by me, as I use in my cmp a w2k KERNEL, but by many people on the net. It is used for securing ram solely for i/o operations. BTW, I think the values are in bytes, so that 1000 stands for 4 kbyte - the standard page size in w2k and xp and 7 for x86 processors. In w2k retail registry there was, in session manager, an additional entry, but it was removed from w2k sp1 and onward. Many people tried to detect the effect of this tweak on xp machines, they attest that the xp kernel understands this command but... and then all the info is deleted by MS. I am now using the value 20000 - the effect is even better, and the mouse movement is smooth like upon polished glass covered with oil, if I describe it not too dramatic. Try for yourself, IT WON"T HARM. As for additional critical worker threads, you cannot add more than 16!
Serge.
Hi Serge
You've been holding out on us - using a different "ntoskrnl.exe" ! Which version is it ? I ask because ever since Win2K-RTM, all Windows kernels have not recognized the IoPageLockLimit parameter. This can be (almost conclusively) verified by running the Sysinternals utility "strings.exe" and searching for the keyname in the text output file.
** see -> http://technet.microsoft.com/en-us/sysinternals/bb897439 to download strings.exe **
This string search technique is referenced here -> http://www.wxpnews.com/archives/wxpnews-039-20020820.htm.
Read the section called Debunking the "Speed Up your Windows XP Box with IRQ Prioritization" Myth, (about half way down the page). Interestingly, Jamie mentions the "IoPageLockLimit Hack" in bullet point three:
The same "strings" output will show that another long-written-about registry hack, IoPageLockLimit, is also missing in action. This one at least used to be there, in Win2K RTM and earlier; in Win2K SP1 they kept the mechanism but the limit was then spec'd by a new value, IoPageLockPercentage; in Win2K SP2 and later they dropped the whole thing.
There is also a thread on MSFN Forum discussing this issue - a concise summary is found in the last post -> http://www.msfn.org/board/topic/25684-registry-myths-1-iopagelocklimit/page868768#entry868768.
Sooooo... I wonder which Win2K edition your kernel file came from ? This might explain the reports of other cMP-Slimmers, with regard to using IoPageLockLimit.
I must hasten to add: THANK YOU Serge, for your tireless efforts to squeeze the most out of the cMP^2 concept!
Cheers,
Grant
That's not a Toy... IT'S A TOOL !!
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: