Do you play games with pagefile on or off?


  

125 members have voted

  1. 1. Do you play games with pagefile on or off?



Recommended Posts

I know some games/programs don't run well with pagefile off, but the ones that don't I just turn off my pagefile because the game loads faster and it's much more smoother (Ex: Starcraft 2). Just curious if you guys do this also

And I also use a hardware profile to keep all the services off when gaming :) . Forgot to mention, I have no problems at all on Starcraft 2 with the pagefile off

TL;DR version: Let Windows handle your memory/pagefile settings. The people at MS have spent a lot more hours thinking about these issues than most of us sysadmins.

Many people seem to assume that Windows pushes data into the pagefile on demand. EG: something wants a lot of memory, and there is not enough RAM to fill the need, so Windows begins madly writing data from RAM to disk at this last minute, so that it can free up RAM for the new demands.

This is incorrect. There's more going on under the hood. Generally speaking, Windows maintains a backing store, meaning that it wants to see everything that's in memory also on the disk somewhere. Now, when something comes along and demands a lot of memory, Windows can clear RAM very quickly, because that data is already on disk, ready to be paged back into RAM if it is called for. So it can be said that much of what's in pagefile is also in RAM; the data was preemptively placed in pagefile to speed up new memory allocation demands.

Describing the specific mechanisms involved would take many pages (see chapter 7 of Windows Internals, and note that a new edition will soon be available), but there are a few nice things to note. First, much of what's in RAM is intrinsically already on the disk - program code fetched from an executable file or a DLL for example. So this doesn't need to be written to the pagefile; Windows can simply keep track of where the bits were originally fetched from. Second, Windows keeps track of which data in RAM is most frequently used, and so clears from RAM that data which has gone longest without being accessed.

Removing pagefile entirely can cause more disk thrashing. Imagine a simple scenario where some app launches and demands 80% of existing RAM. This would force current executable code out of RAM - possibly even OS code. Now every time those other apps - or the OS itself (!!) need access to that data, the OS must page them in from backing store on disk, leading to much thrashing. Because without pagefile to serve as backing store for transient data, the only things that can be paged are executables and DLLs which had inherent backing stores to start with.

There are of course many resource/utilization scenarios. It is not impossible that you have one of the scenarios under which there would be no adverse effects from removing pagefile, but these are the minority. In most cases, removing or reducing pagefile will lead to reduced performance under peak-resource-utilization scenarios.

http://serverfault.com/questions/23621/any-benefit-or-detriment-from-removing-a-pagefile-on-an-8gb-ram-machine

The paging file is needed in some scenerios, as Fred Derf's post points out.

Why not create a ram drive and point your pagefile to it? As long as you have enough memory, this would guarantee the best of both worlds.

You'd be improving the performance of the pagefile but hurting the performance of your memory, and cutting the amount of memory available making the pagefile needed more often.

That sounds silly.

Why not create a ram drive and point your pagefile to it? As long as you have enough memory, this would guarantee the best of both worlds.

This is the worst idea in history. All you will be doing is making less memory available and hurting performance. It does nothing to improve anything. You might as well turn the page file off in that scenario. Doing so would be preferable, as Windows doesn't actually need it for anything (apart from saving crash dumps, which are optional anyway). It's only a safety net. It's technically possible for an application to allocate memory in the page file directly and so these would fail unless they were programmed with a fallback, but I know of no applications or games that do this.

He doesn't state his configuration, so I'm going to assume it's a modern Windows 7 setup. Windows actually does not go and randomly page all the time. If you aren't low on memory, it does not page to disk. Can the OP actually provide data from the resource/performance monitor to illustrate the problem he's having?

Windows will still swap memory to disk even with the pagefile disabled. It'll secretly use temporary files instead. Don't discount the placebo effect for the personal anecdotal evidence that you may hear about a performance boost.

It isn't uncommon for half of the test subjects on fake medication (i.e. sugar pills) to identify that the "medication" have improved their condition. That's the placebo effect in action. There's this whole mind over matter thing going on when you believe that something will work.

When it comes to pagefiles, however, the experts (the ones that actually understand Windows' memory management) will tell you to keep it on while the amateur hobbyists will tell you to disable it for a performance boost. It's your choice though.

Windows will still swap memory to disk even with the pagefile disabled. It'll secretly use temporary files instead.

There is no such thing as a secret page file. This is a myth that I think originated from the task manager in older versions of Windows, which incorrectly labeled certain graphs "page file usage" when that is not what they were. If you disable all page files, it really disables them.

The rest is right, which is why I suggest that he actually measures it.

Turning the pagefile off for more performance is one of the bad ideas which will never die out, similar to the TCP/IP patch some people keep suggesting for better download speeds since XP SP2.

Just to provide an authoritative source for my last claim. From the 5th edition of Windows Internals, page 781:

qxjo11.png

If you look at the source of the memory manager, you can find various optimizations for when there's no page file. Just to show that Windows is designed with this in mind.

n507f7.png

Messing with the page file is pointless.

+1

Any proven performance gains from turning it off?

Its not if there is any proven performance or not, its that (nowadays) it is pointless

There is no such thing as a secret page file. This is a myth that I think originated from the task manager in older versions of Windows, which incorrectly labeled certain graphs "page file usage" when that is not what they were. If you disable all page files, it really disables them.

The rest is right, which is why I suggest that he actually measures it.

Well here is my question then:

If you open a program that uses 1GB of RAM and you have 512MB of RAM and your page file is disabled, what happens?

Turning off the pagefile is a bad idea for sure, but wouldn't manually setting both the min and max value to the same amount prevent the pagefile from getting fragmented?

Well here is my question then:

If you open a program that uses 1GB of RAM and you have 512MB of RAM and your page file is disabled, what happens?

Windows will complain that it is low on memory and ask you to close one of more applications.

This topic is now closed to further replies.
  • Recently Browsing   0 members

    • No registered users viewing this page.
  • Posts

    • Samsung messages is still alive and well. They reversed that and decided to our more effort into it.
    • I think every American should have a course in a 'dry sense of humour' at school; and perhaps 'using sarcasm in jest' oh, and also 'the use or irony in humour'.
    • What they really need to do is automatically spell "loses" and "lose" correctly since nobody seems to know how to anymore. Then they blame spellcheck/autocomplete and don't realize autocomplete is just filling in the word for them automatically so it was misspelled in the first place.
    • If someone chooses to continue using SB and therefore goes through the manual intervention in the thread, afterwards the BSOD problem is gone. Whether they then re-enable the task doesn't matter, they're done, though on such machines it might pay to keep it disabled in case the next update (if there is a next) causes the same problem. OTOH, if someone disables SB in the BIOS, the problem is also gone. Incidentally, I noticed that this task exists even on machines that don't support SB. It's just installed across the board...and runs. Doing what on such machines is a little hazy.
    • qBittorrent 5.2.2 by Razvan Serea The qBittorrent project aims to provide a Free Software alternative to µtorrent. qBittorrent is an advanced and multi-platform BitTorrent client with a nice user interface as well as a Web UI for remote control and an integrated search engine. qBittorrent aims to meet the needs of most users while using as little CPU and memory as possible. qBittorrent is a truly Open Source project, and as such, anyone can and should contribute to it. qBittorrent features: Polished µTorrent-like User Interface Well-integrated and extensible Search Engine Simultaneous search in most famous BitTorrent search sites Per-category-specific search requests (e.g. Books, Music, Movies) All Bittorrent extensions DHT, Peer Exchange, Full encryption, Magnet/BitComet URIs, ... Remote control through a Web user interface Nearly identical to the regular UI, all in Ajax Advanced control over trackers, peers and torrents Torrents queueing and prioritizing Torrent content selection and prioritizing UPnP / NAT-PMP port forwarding support Available in ~25 languages (Unicode support) Torrent creation tool Advanced RSS support with download filters (inc. regex) Bandwidth scheduler IP Filtering (eMule and PeerGuardian compatible) IPv6 compliant Available on most platforms: Linux, Mac OS X, Windows, OS/2, FreeBSD qBittorrent 5.2.2 changelog: FEATURE: Use D-Bus to show file in file managers (Chocobo1) #24340 BUGFIX: Fix friendlyUnitCompact precision calculation (vafada) #24323 BUGFIX: Remove all top-level folders (glassez) #24333 BUGFIX: Use proper API for checking exit status (Chocobo1) #24349 BUGFIX: Delete stale lockfile when hostname mismatch (TurboTheTurtle, glassez) #24363 BUGFIX: Fix wrong removal procedure of watched folder paths (Chocobo1) #24413 BUGFIX: Don't reannounce before interface changes are applied (glassez) #24447 BUGFIX: Use Latin script for Bosnian locale name (Andy Ye) #24342 WEBUI: Fix performance of global checkbox toggling (tehcneko) #24316 WEBUI: Fix Safari transfer list header misalignment (Piccirello) #24377 WEBUI: Fix error when submitting magnet before metadata loads (Piccirello) #24378 WEBUI: Use correct row id when updating Rss Downloader feed selection (Chocobo1) #24402 WEBUI: Use SameSite=Lax for session cookie to fix cross-site login (Piccirello) #24422 WEBUI: Bring back properties panel expand/collapse button (vafada) #24430 WEBAPI: Only use X-Forwarded-Host header when reverse proxy support is enabled (Chocobo1) #24457 RSSS: Fix "RSS Smart Episode Filter" RegEx (nathanon-akk, glassez) #24398 RSS: Fix previously matched episode format (glassez) #24452 WINDOWS: Fix Python fallback search path (TurboTheTurtle) #24325 WINDOWS: NSIS: Allow to install x64 binary on ARM64 (Chocobo1) #24358 Download: qBittorrent 5.2.2 | 41.1 MB (Open Source) Download: qBittorrent 64-bit installer (qt6) | 43.6 MB Links: qBittorrent Home page | Screenshot Get alerted to all of our Software updates on Twitter at @NeowinSoftware
  • Recent Achievements

    • Veteran
      branfont went up a rank
      Veteran
    • Reacting Well
      Almohandis earned a badge
      Reacting Well
    • First Post
      Cosminus earned a badge
      First Post
    • One Year In
      ThatGuyOnline earned a badge
      One Year In
    • Week One Done
      Jeroen Wilms earned a badge
      Week One Done
  • Popular Contributors

    1. 1
      +primortal
      498
    2. 2
      +Edouard
      185
    3. 3
      PsYcHoKiLLa
      126
    4. 4
      Steven P.
      87
    5. 5
      neufuse
      72
  • Tell a friend

    Love Neowin? Tell a friend!