Pagefile.sys

[medeiom]

Well all I can say is that it works for me. I certainly respect your opinions and this was an interesting debate. I hope Windows 7 users will take it upon themselves to test out what really works.

 

[Digerati]

This is a technical discussion. It's really not a matter of opinion. We are not choosing between a red one or a blue one where opinions matter.

There is just no empirical evidence to suggest disabling the PF is beneficial. Even your own observations don't suggest any benefits - only that your computer did not break when you disabled it.

Curious - have you tried enabling it to see what happens?

 

[medeiom]

response

I agree this is a technical discussion and I will end this debate with my last comments here. In the past, I did have Pagefile enabled on my SSD and I did experience a "minor" glitch in regards to performance.

Now, a lot has changed over the months in terms of firmware on my SSD, updates, etc,. But I will enable it and see whether or not Pagefile is at all beneficial to me with regards to my specs. I willing to bet it won't be necessary, but I'll give it another whirl.

Thank you.

My specs..
ThinkPad T400
256GB Crucial SSD (firmware 0007 )
8GB Samsung DDR3 RAM
Win7 Ultimate 64bit
CPU Intel core 2 Duo 3.06

 

[Digerati]

But I will enable it and see whether or not Pagefile is at all beneficial to me with regards to my specs. I willing to bet it won't be necessary, but I'll give it another whirl.

Sadly, you are still totally missing the point. Instead of seeing if enabling the PF is beneficial to you, you need to ask yourself, what are the benefits to disabling it? What is it you think you know that Microsoft doesn't? It is their product that creates and uses the PF. And we saw where your information was incorrect with your claims about SSD makers.

What makes you more qualified than the 100s of PhDs, computer scientists, and master programmers that Microsoft has on the Windows development staff?

Do yourself a favor. Find someone, anyone, who supports your position. I have asked repeatedly for you show us some supporting evidence - you have provided none.

If you can't find any, that should tell you something.

I note too that Windows is not the only OS that uses a PF. So does Linux - even with lots of RAM installed.

 

[pallesenw]

What makes you more qualified than the 100s of PhDs, computer scientists, and master programmers that Microsoft has on the Windows development staff?

How can they know what his needs are?

 

[Digerati]

How can they know what his needs are?

How does a doctor know what your needs are? They don't. They know what most humans need because they went to years of school. They can determine what your specific needs are because they are FORMALLY TRAINED diagnosticians. And they use specialized tests using specialized equipment with the results analyzed by more formally trained specialists.

And MOST IMPORTANTLY - they prescribe and offer medical advise based on YEARS of empirical testing and retesting - and established published guidelines - not guesses.

But also, your point is EXACTLY one I made earlier. How is it proper or logical to recommend or advise others to disable their PFs when we don't know their needs? It is not. Therefore, we, as helpers, should not be giving advise that is not supported by any evidence it does good.

"Because it did not break my machine" is NO reason to tell others to disable the PF.

Also, they don't know his needs. That is EXACTLY why the default is for Windows to analyze and manage the PF size so the PF can be adjusted on the fly as needed, based on the demands THAT SPECIFIC machine is currently experiencing.

How does he know what his needs are? Does he have inside knowledge how every program he downloads and installs will use memory? We already know some software expects to find a PF, whether it uses it or not. 3rd party software makers don't know what his needs are either, therefore, their only option is to program to the defaults too - and that is an enabled PF.

 

[pallesenw]

How does he know what his needs are?

Well, he can start by checking his peak counter in task manager. You like to refer to people like Mark Russinovich. He goes to great lengths to tell people how to check what their needs are.

We already know some software expects to find a PF, whether it uses it or not

How many can that be? And how many have an option to disable that check?

 

[Digerati]

Well, he can start by checking his peak counter in task manager. You like to refer to people like Mark Russinovich. He goes to great lengths to tell people how to check what their needs are.

Yes, and Mark tells how to adjust it - NO WHERE does he, or anyone else say to disable it - unless you can show us a link to a white paper, research paper, KB article or some official document (not a forum poster) that shows where "disabling" the Page File completely is beneficial. Can you? Because I'm pretty darn good at using Google and I have found zero, absolutely no evidence that says disabling the page file improves performance, improves stability, or improves anything. Yet there is plenty of evidence proving having a page file does have benefits.

How many can that be? And how many have an option to disable that check?

How many can that be? How should I know? Examples have been given, but does it matter? If it is a program a poster needs, he needs it.

As for how many have an option to disable it? How should I know that either? But more importantly, how does medeiom know? Or anyone else? Those unknowns alone should be reason enough to leave the defaults as they are. There are 1000s and 1000s of programs out there, some new, many old. Unless otherwise documented, all we can assume is they were programmed with Windows default settings in mind.

@pallesenw - note this discussion is not about tweaking the PF for optimal performance. It is about whether or not it is beneficial to disable it completely, or not. And so far, no one has presented any evidence, or even a plausible explanation justifying disabling the PF. None.

 

[pallesenw]

A program as such never needs a page file. What it needs it available memory. If you somehow have a program that malfunctions without a page file, well yes, then you have a need for it.

If I have 16 GB physical ram and never ever commit more than 3-4 GB, well then the need is not that big.

The need to cache as much data as possible is less when you have a very fast SSD.

 

[Digerati]

A program as such never needs a page file. What it needs it available memory. If you somehow have a program that malfunctions without a page file, well yes, then you have a need for it.

If I have 16 GB physical ram and never ever commit more than 3-4 GB, well then the need is not that big.

The need to cache as much data as possible is less when you have a very fast SSD.

That has absolutely NOTHING to do with this controversy. I ask again, can you show us ANY, just one little shred of evidence, a paper or KB article that reports disabling the PF is beneficial?

 

[Digerati]

A program as such never needs a page file.

How can you make that assumption? Do you know exactly how all the 1000s and 1000s of programs Windows supports uses or needs memory? I think not.

"Need" does not matter anyway. What matters is how the program is coded, and more importantly how Windows manages it, AND all the other tasks and processes running.

 

[pallesenw]

How can you make that assumption? Do you know exactly how all the 1000s and 1000s of programs Windows supports uses or needs memory? I think not.

"Need" does not matter anyway. What matters is how the program is coded, and more importantly how Windows manages it, AND all the other tasks and processes running.

It is very simple. You cannot tell Windows to allocate memory in the page file. That is how I know. You can request memory to always be in physical memory, but that is different.

 

[Digerati]

Who cares?

Have you been following this thread?

14 SevenGurus have come in to this thread to say do not delete the page file, providing MANY links to reliable sources like BlackViper (who mentioned several major programs that require PFs), LifeHacker, Microsoft, Mark Russinovich, SSD makers, and more - all supporting the same position.

NOT ONE recognized expert in this thread, and apparently, across the Internet too, suggests disabling the Page File. NOT ONE!

You cannot step into the middle of bunch of experts and tell them they are all wrong without something to support your stand. You've provided nothing. I have been pleading since February in Post #50, and countless times since for some shred of evidence - a link to ANY source to support your position. And you haven't done it. Because you can't. There just is none.

Clearly, pallesenw, since you have now taken the ball and are arguing the point, I will ask for the umpteenth time, show us just one shred of empirical evidence from a reliable source that suggests disabling the PF provides any benefit.

If you cannot do that, then step up and admit it. Don't just keep saying you are right, because contrary to what seekermeister said in Post #56 there is a consensus among the SevenForums experts in this thread and across the Internet; there is no benefit to disabling the page file.

There may be some advantage to moving it, or making it a fixed size, but those are different issues.

 

[pallesenw]

Who cares?

You argued that millions of applications cannot work without a page file, because they are programmed to use it. I think you should care yourself to know that this is not the case.

Other than that, I am just trying to tell you that a page file is about need.

 

[pallesenw]

A guru said in this thread:

"MS recommends... as do I...

Amount of Ram x 1.5 = Min
Amount of Ram x 3 = Max"

Lets use Mark Russinovich again. In one of his webcasts he actually mocks people that says that (including a Microsoft employee)

 

[Belgarionbg]

I was convinced from one of my colleagues that with mine 6GB ram I can afford to minimize the size of pagefile to 16mb. He said "mate, you will have no problems".
And he was right, for about 2 hours after change. Then I decided to play some games and guess what - the framerate has dropped, actually in almost every game I saw strange freezing for about 1 second in every ~15s. It was impossible to play normal. It took me about 20min. of checking everything just to realize that the culprit was pagefile.
So I rised it to 1GB - almost same sh**, but 3GB do the job. /these games are almost every nfs from nfshp2 till last one/
Whats the conclusion? Pagefile is needed. And its better to be hosted in hdd other than system for good performance.

 

[pallesenw]

Belgarionbg, there might be something else wrong... anyways, it doesn't change a bit about what I said. If your commit limit needs to be bigger then obviously you need to make the page file bigger (or install more physical ram).

 

[Cr00zng]

I've used my Vista 64-bit with 4 GBs memory, Windows 7 now has 8 GBs memory, and both had/has the page file disabled. In both cases, the main purpose of the PC is web editing with Adobe Web Premium, MS Office 20xx, and playing games such as Black-Ops, BattleField, etc. Certainly, there are other "stuff" on my PC, but these are the main ones.

Not even with Vista, much less with Windows 7, did I notice any performance hit. Nor did I notice any frame rate drop, but then again; does it really matter even if it did, when your machine hits close to a 100? Especially, when it is an online game...

As such, while I understand the "experts" advise against disabling the page file, I don't see the need for it and in my case set to disabled.

 

[medeiom]

pagefile

Thank you pallesenw for seeing my point that users should experience themselves on whether or not they need a Pagefile with a huge amount of RAM.

I can tell you that Microsoft will eventually come up with such a way that Pagefile will no longer be needed as PC's come equipped with 8, 12 16 Gigs of Ram or more. Technology will get us to that point!

This is related to ONLY 64bit users...

"In general, performance concerns related to pagefile access are much more effectively dealt with by adding more physical memory"...

Paging - Wikipedia, the free encyclopedia

The way I see it, the more Ram you have..and I mean more than 6 or 8GB on 64bit, the CHOICE is yours. You the Windows 7 or Vista user, have the freedom and the choice to disable Pagefile IF you can accurately determine that your system is NOT USING more than 80% or Ram. Love those words...Choice and Freedom to do so as you wish!

 

[Digerati]

You argued that millions of applications cannot work without a page file

I did not! Don't go making stuff up or twist my words around. I said "some", and other SevenGurus provided examples.

I can tell you that Microsoft will eventually come up with such a way that Pagefile will no longer be needed as PC's come equipped with 8, 12 16 Gigs of Ram or more.

Oh? Got a link to support that wild claim?

@medeiom and pallesen - Why do you continue to refuse to provide ANY, just one supporting link to your claims?

You guys are wasting everyones' time. Seriously. Provide some proof. 14 recognized experts on this forum had the professional courtesy to NOT expect you to automatically believe us. So we provided multiple links to substantiate what we claimed.

So have the same professional courtesy with us, and backup your claims with some documented evidence.