Creating a custom Page File for Virtual Memory

[DJ-phYre]

I have been having some problems with my page file, so I turned it off recently... well I am going to create a custom entry instead of letting the system handle it...

It recommends 1534 MB.... I have 1024 MB of RAM right now

What should I put in:

Initial Size (MB): ________
Max Size (MB): ________

 

[Aussie983]

with a gig of ram im suprised u need a pagefile

 

[DJ-phYre]

with a gig of ram im suprised u need a pagefile

Well there are some apps that give me all these warnings about NO WINDOWS PAGE FILE... just getting a bit annoying

 

[LeeJend]

Always make the pagefile fixed size. It fragments horribly otherwise and causes all your new installs and data to fragment.

Size - Do you really want Windows pawing through 1.5 gig of HD files everytime it needs to go looking for things? 1.5x memory made sense when memory was 128M or 256m but with 1 gig keep it small. Start at 768 mB and see if you need more. If so, delete the pagefile, defrag the disk and create a new larger one of fixed size.

Also:
-Defrag before you create any new pagefile.
-Keep the pagefile on the same HD as your operating system otherwise you run into spinup time delays.
-Keep the pagefile on a small partition where you do not store your data.

The above is what I do and I haven't had any pagefile relaetd problems.

 

[Xie]

A search would have found this post .

 

[dreamliner77]

Time for perris to chime in...

He'll know the answer.

 

[Perris Calderon]

leejend, you must NEVER make the pagefile a fixed size..., your information is incorrect

there is NO FRAGMENTATION WHATSOEVER OF THE PAGEFILE, it doesn't ever happen on a healthy drive...ever.

in addition, there is NO performance hit whatsoever with a pagefile that's too big, and there are HUGE performance hits with a pagefile that is too small...bigger is ALWAYS BETTER then smaller when it comes to the pagefile...

you need AT LEAST AS MUCH PAGEFILE AS YOU HAVE MEMORY IN USE...this is allready benchmarked and documented

a fixed pagefile WILL slow some users down, a smaller pagefile then the amount of memory installed WILL slow some users down...those ideas will benefit no person...please read the article I posted in the thread that xie posted

 

[ming]

also, there is NO FRAGMENTATION WHATSOEVER OF THE PAGEFILE, it doesn't ever happen on a healthy drive...ever.

I'm not sure that I agree with your statement here. My machine has a healthy HD, and when running Diskeeper defrag on boot, I've found fragmentation of the pagefile before. Okay, I can't argue that there was major fragments, but a few fragments still counts.

 

[GoNz0]

leejend, you must NEVER make the pagefile a fixed size..., your information is incorrect

there is NO FRAGMENTATION WHATSOEVER OF THE PAGEFILE, it doesn't ever happen on a healthy drive...ever.

in addition, there is NO performance hit whatsoever with a pagefile that's too big, and there are HUGE performance hits with a pagefile that is too small...bigger is ALWAYS BETTER then smaller when it comes to the pagefile...

you need AT LEAST AS MUCH PAGEFILE AS YOU HAVE MEMORY IN USE...this is allready benchmarked and documented

a fixed pagefile WILL slow some users down, a smaller pagefile then the amount of memory installed WILL slow some users down...those ideas will benefit no person...please read the article I posted in the thread that xie posted

unless your a gamer and it decides to change the size of the PF while your playing a CPU intensive game, happend a few times, that was your theory out the window for my situation, fixed works best for me

 

[Perris Calderon]

unless your a gamer and it decides to change the size of the PF while your playing a CPU intensive game, happend a few times, that was your theory out the window for my situation, fixed works best for me

the reason it decided to change the size of your pagefile while you were gaming is that it was too small

if the commit xharge nnever reached the commit limit it couldn't expand

 

[GoNz0]

not sure perris m8, but the games i play like battlefield 1942 desertcombat mod swallow the 1gb of ram i have, i used to get problems from as little as 10 minutes into a round, the game would freeze, it's not something windows was prepared to sort out as the problem continued for weeks, until i fixed my PF at 1.5 gb, not one lock up since, mabey if windows played ball and resized once or twice, but it couldn'e make it's mind up, so i made it's mind up for it.


[dealer edit] sorry, heh, I hit edit instead of quote]

dealer edit mistake, this post is now back to the original[/dealer edit]

 

[GoNz0]

not sure perris m8, but the games i play like battlefield 1942 desertcombat mod swallow the 1gb of ram i have, i used to get problems from as little as 10 minutes into a round, the game would freeze, it's not something windows was prepared to sort out as the problem continued for weeks, until i fixed my PF at 1.5 gb, not one lock up since, mabey if windows played ball and resized once or twice, but it couldn'e make it's mind up, so i made it's mind up for it.

*edit* a 2 second freeze up while your flying an F14 doesnt go down well



*edit* doh hit quote not edit

 

[Perris Calderon]

I'm not sure that I agree with your statement here. My machine has a healthy HD, and when running Diskeeper defrag on boot, I've found fragmentation of the pagefile before. Okay, I can't argue that there was major fragments, but a few fragments still counts.

the pagefile will fragment in these situations;

if you adjust the minimum size, or if you change the amount of the memory and the os changes the initial size

ALSO

diskeeper will putt your pagefile on a differant area on your drive, and can fragment the pf also

use the system internals free program to monitor your pagefile extents, and you will see ONCE IT IS CONTIGUOUS it remains contiguous even after expansion, ( once you reboot)
all of this is explained in the paper that's posted in the thread xie sites, you guys should read that

once it's contiguous, it remains contiguous forever on a healthy drive

if expansion is invoked, the expanded extent is removed on reboot, and thre original pagefile HAS to be in the original condition

the issue is that you've mmade the initial minimum too small, and your operating system is expanding the pagefile

that's the point, you must not have a pagefile so small the os wants to expand it, when it does expand, the fragmentation IS ONLY FOR THE EXPNSION EPISODE and you have a pagefile that the initial minimum needs to be increased, which of course will ahve to be defragemnted ONE TIME and then it will always be defragmented

 

[Perris Calderon]

not sure perris m8, but the games i play like battlefield 1942 desertcombat mod swallow the 1gb of ram i have, i used to get problems from as little as 10 minutes into a round, the game would freeze, it's not something windows was prepared to sort out as the problem continued for weeks, until i fixed my PF at 1.5 gb, not one lock up since, mabey if windows played ball and resized once or twice, but it couldn'e make it's mind up, so i made it's mind up for it. ]

well, try two gig initial minimum then, with expansion to 4 gigs

the pagefile WILL NOT expand untill your commit charge reaches the commit limit...that's the ONLY mechanism that makes th pf expand, nothing else

if you were expanding, your os was requesting more virtual memory then the os was prepared to deliver

when that happens, if the os can't get the virtual memory from the pagefile, it gets the memory by unloading an exe or dll

most gamers that have the problem you just pointed out get the problem in the first place BECAUSE they listened to the missguided advice that a small pagefile is somehow better for performance then a big pagefile, and lowered it from the default

I'm guessing you lowered your initial minimum below the defalt, which has been the case everytime someone has brought their box to me with that problem, but if not, then the 1.5 is too small also, and the initial minimum needs to be increased

 

[GoNz0]

no perris, it was set to auto, why start with a 2gb minimum going up to 4 gb?

 

[Perris Calderon]

if your pf expanded using this game, then the commit charge reached the 1.5 limit, and therefore expanded it

your initial minimum needs to be so large that the os never wants to ewxpand it

 

[Perris Calderon]

that's the point, the more memory intensive programs get, which was bound to happen with the amount of memory available, the more virtual memory they request.

the initial minimum needs to be larger then the os requests, when it's fixed, if the ois requests more virtual memory, since it can't get it from the pagefile it will unload a dll or exe.

this has to happen or it will freeze

 

[Perris Calderon]

does this game have a dem?

I want to download and get this expansion you're talking about.

 

[GoNz0]

well this is a typical load situation, battlefield vietnam loaded, gb of ram used ( the usual 17000 k left, PF jumped to 684mb

can you point out any gains from fixed at 1.5 to your 2gb minimum, wont 2gb min, cause performance issues as its so damn big ?

 

[GoNz0]

oh sorry battlefield vietnam is a demo less game, its not an expansion. if you got battlefield 1942 you could download the beta of desertcombat, but desertcombat will not play on the battlefield 1942 demo.