how-to block ads
|
|
Share Topic
 |
 |
|
|
|
| reply to Pontiac Freak
Re: If you go beyond 128 Mbs RAM, do this! Pontiac Freak, this tweak works well in most cases. It depends on how the system is setup to manage the page file. These three conditions are crucial: 1)Permanent swapfile, 2)(Large) paging buffers enabled, 3)>=128 MB installed RAM. The later the version you have of Windows, the less you get away with when tweaking memory management. WinME is practically "off limits" for tweakers - Win3.1 lets you do a lot of things to it - Win95 lets you do a little less - Win98 lets you alter some simple things like conv. memory settings, but dislikes the tampering with system core components if not done very precise. You can however get away with a lot of tweaks if you use them in correct combinations with other tweaks, and in the right context. You can't just alter one thing and expect Windows to adapt every affected aspect of the system to it. In this case, when you set the value 'ConservativeSwapfileUsage=1' what you do is you turn off some parts of the VMM and VFAT, namely those parts that separates Win95's VM handling from Win98's(It should be called "Win95SwapfileUsage" to be self-explanatory). What you loose by doing this is 'asynchronous operations', which easily explained means that a pagefile data transaction must wait in line for its predecessor to finish. This is no good of course, so if you use this tweak, and want to continue using asynchronous pagefile operations, you also have to manually tell the memory manager to use pagebuffers by putting in the line 'PageBuffers=X' where 'X' is a value between 0 and 32. Win98 uses 4K buffers as default, but use X==32 for maximum performance. Now 32K buffers only work in 32-bit disk operations, therefore make sure you use this at all times by also putting in the lines '32BitDiskAccess=on' and '32BitFileAccess=on'.
So, here's what it should look like now:
[386Enh]
ConservativeSwapfileUsage=1
PageBuffers=32
32BitDiskAccess=on
32BitFi leAccess=on
PagingDrive=X:
MinPagingFileSize=Z
MaxPagingFileSize=Z
The last three lines is to use a fixed swap file. Replace the X and Z's with appropriate values('X:' is which drive you want the swapfile to reside on, and the Z's are the Min and Max sizes of it). Min and Max is the same value. For more info about fixed swap file, read this.
See the correlations between these various tweaks: 'ConservativeSwapfileUsage' requires 'PageBuffers', and 'PageBuffers' requires '32-bit disk access' and a 'permanent swapfile'.
For a good compilation of 386enh tweaks and other System.ini tweaks, go here. | |
| Ping-Pong:
As usual, a fine, informative analysis about a MS problem that you've offered us. Many thanks. Note that the first link in my initial post briefly, concisely, explains the problem you've set out in your own crisp effective manner.
To save time for those looking at the thread this far, I'll reproduce the relevant paragraphs dealing from the link with this: (Note that the relevant paragraph that refers VFAT etc in your memo is set off by .....)
" . . . Based on benchmark testing by MS-MVPs Mike Burgess, Alex Nichol, and other MS-MVP colleagues, I have concluded that this tweak [ConservativeSwapfileUsage=1 in the system.ini] can be of real value in many Win98/ME systems with 128 MB of RAM or more (and on some 64 MB systems). If your machine has sufficient RAM that, for your particular use of your computer, there is relatively little swap file usage, then this tweak will probably improve system performance by reducing the overhead in swap file maintenance activities that you do not really need. [NOTE: In contrast to behavior in Win98, testing of this by Mike Burgess in Windows Millennium Edition showed no significant performance difference with or without it. However, because it does affect swap file run-away growth on some computers -- such as mine! -- I am presently recommending this on Windows ME.]
"The default value of ConservativeSwapfileUsage was 1 in Win95, and 0 in Win98 and WinME. Therefore, this recommended tweak reverts a portion of the Win98/ME improved memory management system back to the way Win95 did things. (That's what sounded like such a bad idea at first, because Win98 or ME is generally far superior to Win95 with regard to memory management.) A valid question, though, is whether it is a good (that is, useful) idea for systems that have grown past the RAM size originally contemplated as typical when Win98 was created. For many computers, the answer seems to be: Yes, it is.
"According to MS Knowledge Base article Q223294,
"'..........Windows 98 added a new feature, PageFile_Call_Async_Manager, that allows the Memory Manager to asynchronously write out page file (swap file) buffers during periods of time when VFAT file system activity is not busy. This feature can affect the behavior of VxDs [virtual drivers] that monitor and/or otherwise intercept PageFile VxD functions............'
"(See the article itself for more technical background on the rather clever way this function operates.)
"This feature is part of improvements made initially in Windows 98 that lend the appearance of faster performance to a Win98/ME computer because many of the small time-consuming activities are deferred to moments when you are not actually asking the computer to do anything. The tweak described above disables this feature. As MS-MVP Jeff Richards summarized it, "It's more conservative in scheduling non-critical swap-file activity."
"Microsoft advises that this change may be "at some cost in overall system performance." Furthermore, in theory it would seem that this would make the computer seem slower. But only experimentation on a given system (and by a particular user on that system) can determine what the subjective performance difference (pro or con) would be. Based on experience to date, I would say it is probable that this tweak will have a positive effect on systems with an unusually large amount of RAM, and very little swap file activity. It is unlikely that it would have a positive effect on other systems, and might have a negative impact on their performance.
"Since the tweak is so easy to undo -- just remove, or comment out the line added to SYSTEM.INI -- "power users" will likely want to experiment with this and give more feedback to the general community."
So Ping Pong: quite possibly your tweaks deal with the problems that might crop up on some systems. Later today, then---when I get some more time (back to my work now)---I'll try myself to experiment with your added tweaks.
You get my vote for sure anyway.
Michael
[text was edited by author 2001-04-20 15:57:41]
[text was edited by author 2001-04-20 15:59:37] | |
wozzeck
join:2000-09-11
Saint Louis, MO | I don't see that section in my system.ini ? what's the deal?
Win 98 | |
| reply to PingPong
Ping Pong: Bad Results, Alas! Ping Pong:
Well, with a little time now, I tried adding the six extra lines to ConservativeSwapfileUsage=1 you suggested. The result? A much slower system, resulting in Windows 98 accessing the physical swapdrive not long after 85Mbs of RAM!
I then went to a site, using google---»www.63rd.co.uk/System%20Memory%2···k.htm---
which had a much clearer explanation of asynchronous operations than Microsoft's contorted gobbledegook (do the people who write these sites for MS ever have a non-technical person read them as a check?).
What I found was that it's far better to use synchronous logic---the normal function, as you no doubt know, for controlling digital code on PCs (tick,tock; tick,tock)---as opposed to asynchronous. According to the UK site, synchronous logic is the normal windows default (98):
"Right click on My Computer, select Properties. Select the Performance tab. Now hit the File system button. Go to the Troubleshooting tab. Some of these definitions were obtained via the MS Knowledge Base, although only the important ones are covered here.
Disable new file sharing & locking semantics. This setting controls file-locking mechanisms in Windows. Tick (Disabled) this setting if you are currently experiencing problems with some programs, although this should be seen as a last resort, your system will perform optimally with this setting Unticked (Enabled).
Disable synchronous buffer commits. This setting manages the function calls to the File-Commit API to return immediately without checking to see if the data was correctly written to the drive. By default, Windows uses synchronous buffer commits. You can change this setting to enable asynchronous buffer commits for programs that may need this functionality."
So . . . well as a non-technical person, certainly as someone not as savvy as you are, with your MS diploma and far greater knowledge, all I can say is this: I'm not sure how losing asynchronous operations would be bad, assuming the conservativeswapfileusage=1 does this (as the MS base indicates it will, along with you yourself), since even Windows prefers a synchronous logic. And simultaneously adding all the other lines you mention---save PageBuffers=32 (already in my system.ini)---does anything but slow down the use of RAM.
Any clarification would help. As I mentioned, I could be wrong about all this save one thing: the use of RAM is far faster than the physical swapfile on the hard drive. And the use of a fixed swapfile seems to interfere with Windows 98 own management of the swapfile, which seems to anticipate---through its concerns for backward and lateral compatibility of data---what size it prefers the swapfile to be.
Michael
PS: Another site says that asynchronous operations are mainly for transferring data over SCSI cables, and even then is a poor fall-back as opposed to using the synchronous logic:
"Disable Synchronous Transfers: There are 2 methods of transferring data over SCSI cables, Synchronous & Asynchronous mode. Asynchronous mode transfers use an interlocked handshake method where a device (adapter or drive) cannot do the next data transfer operation until it receives positive acknowledgment that the other device received the last data transfer properly. Synchronous transfer mode permits the sending device to send multiple data packets without receiving acknowledgment that the receiver actually received every data packet sent. As a result, more data can be sent/received before acknowledgement is required. You should only tick this if you are experiencing problems with a hard drive(s), e.g. overlapped requests, as performance is slower when Synchronous Transfers is Disabled. Leave it unticked. "
»www.3dspotlight.com/tweaks/memor···nt.shtml | |
|
