Microsoft themselves have contradicting information pertaining to the virtual memory ange paging/swap file sizes and how it's running.
worse yet is the fact that there is very little information that is detailed or strictly stating anything about the windows 7 vista and xp memory systems and subsystems itself ... if your wonder what i'm talking about in terms of a subsystem... it's the other activities that opperate within the main paging/swap file or virtual memory system itself. I'm not really entirely sure how that isn't understood. If you want to information about it and take the time to compare and contrast between the different versions.. also taking into consideration of cacheing and superfetch not to mention how vista/7 handle free/unused memory.... have fun looking through the whitepages or some of the more heavily detailed ..... details....
I've looked over some of it.. and short of taking several pages to explain it, most of which won't make much sense to many.. it's enough that what we've already discussed here in short is confusing to many.....
I'm not preaching.. i'm providing alternative or outside of the box thinking and suggestion. I emplore people to actually "try" or at least "review" alternative rather than parroting the myths or articles or does and don'ts others have said.
Worse yet is when some of these articles contradict themselves within the same paragraph in a way. Majority of them won't clearly state the pro/cons of either situation short of naming a few things off in the case where memory is limited or running out and of course the fallout that could occur when hit which REALLY isn't actually a huge problem.... but then go on to say how yes it's a performance booster followed by no it's a performance reducer... without CLEARLY outlining each individual senario. Many of them don't even give you any kind of usage information at all, period.. zilch.. and just conclude don't do it... bad...
bad enough 99% of them still referrence material from 1993 windows 3.1 and don't dive to much beyond windows xp.
You've got technical experts in the specific field that don't even agree with each other on it either.
So what do you do?
What is microsoft to do?
I'll try an enlighten a bit.. that's all i'm doing.. in the end it's the users choice. Without trying to use self defeating arguements against either.
It's really not that complicated. You do whatever you like, but your choices are these:
A: You leave windows to continue managing your page file even when a good chunk of the experts all agree that even in windows 7, allowing it to manage it isn't a problem, but it is nowhere near "efficient"... specially with any amount of physical ram above 2gb installed prior to having windows installed (resulting in hugely oversized page file). This ia FACT, anyone with 8gb or 16gb or more of ram is going to see a rediculious amount of the page files "minimum" allocated size consuming a huge chunk of physical disk space.
B: You more efficiently manage the page file, setting a minimum and initial or even just setting a minimum value that is more adequate and efficient than the windows managed version. This is what i recommend MOST For the general public if they think they might have issues (unlikely) or just want a balance between both methods. An initial size of no less than 800mb and no larger than 2048mb. It is a fair and balanced and more efficiently setup page file. Enjoy the combination of both worlds essentially. Setting a maximum size to the same value as the initial is fine, and isn't a terrible idea, but a little common sense goes a long way, obviously if your in memory instensive situation that consumes tons and tons of ram and likely page file space, doesn't make sense to cut this down and crash your program that decided to max it out and then fail to allocate any more. These cases are very rare today.....
C: Disalbe the Page file and give it a try.... If you have basically an excess of 4gb of ram, you in most cases aren't going to have a problem.... due to most programs still being limited to 32bit address limits, even if you experience one of the "cons" that they said would plague a user without a page file, frankly i would view it as a pro, Memory leak, it'll kill itself when it tries to exceed or runs out of memory to allocate for itself usually resulting in that program crashing. Alternatively with a page file .. specially with the giagantic page files, while it "should" dump still, it's weirdly able to sometimes just keep pumping the page file up because windows keeps offloading it somewhere to keep the system ram free.
either way.. the ONLY time performance is degraded on a system with the page file disabled only occurs when there isn't enough physical ram. This is coming from about 7 different articles that are the most recent i could find that actually involves windows x64 7/vista.
So everything will be fine and dandy... so long as you got plenty of ram.. and lets face it, someone running lets say 8gb of ram... isn't likely EVER going to run into a problem running 32bit game or program and hit a snag. It's not likely to occur for a number of years yet as even the few 64bit programs available today still don't jump beyond 4gb.
Heres a quote from someone that tested the whole page file system in x64 and later in another article purposely crashed the program by pumping up a page file to over a 1tb just to see how things would work.
Some feel having no paging file results in better performance, but in general, having a paging file means Windows can write pages on the modified list (which represent pages that aren’t being accessed actively but have not been saved to disk) out to the paging file, thus making that memory available for more useful purposes (processes or file cache). So while there may be some workloads that perform better with no paging file, in general having one will mean more usable memory being available to the system (never mind that Windows won’t be able to write kernel crash dumps without a paging file sized large enough to hold them).
Essentially what he/she is saying is what i have been saying...
If you got lots of ram... and aren't running low.... there is potentially a performance improvement, if you happen to BSOD at anytime WHICH won't be due to lacking a pagefile.. you just don't get a memory dump.. which doesn't matter as most people don't ever use/read/view it anyways.
Another referrence to part of the article..
You’ll notice that the default configuration is for Windows to automatically manage the page file size. When that option is set on Windows XP and Server 2003, Windows creates a single paging file that’s minimum size is 1.5 times RAM if RAM is less than 1GB, and RAM if it's greater than 1GB, and that has a maximum size that's three times RAM. On Windows Vista and Server 2008, the minimum is intended to be large enough to hold a kernel-memory crash dump and is RAM plus 300MB or 1GB, whichever is larger. The maximum is either three times the size of RAM or 4GB, whichever is larger. That explains why the peak commit on my 8GB 64-bit system that’s visible in one of the screenshots is 32GB. I guess whoever wrote that code got their guidance from one of those magazines I mentioned!
This basically states what i've stated before.... Microsoft's managed pagefile setup is terribly setup.
Microsoft took the easy out, they just put in a generic whatever works system that covers all the bases and removes any possibility of lacking in page file size under even the most sevear and totally unrealistic circumstances... It's essenitally overkill... Waste.... Overly redundant and unnecessary for most.
As far as posting links to articles... that's pretty futile as like i said.... it's a total circular linkfest.... where one article links and discusses another, which jumps to a post, which than addresses a microsoft article, which then gets countered with another microsoft article that contradicts the first microsoft one, which then continues onto several repeated and discussed, debated, name calling ensues... stupidity is brought to the surface, and everyone does what they wants anyways and thinks the other guy is a total -_-ing idiot.
sadly......
So in the end.. what can we say?
Make up your own mind, do what you want. You haven't anything to worry about "trying" whatever settings you wish, play with it.... Need a BSOD Dump crash report? enable it and do whatever you did to bsod so that you can review it.... but like i said... even i rarely ever view the bsod dumps because the bsod itself will indicate what went wrong.. no need for dumps floating all over the place.
Now before i start repeating myself again and potentially getting interupted by a customer resulting in my lost train of thought.. going to finally hit post