Performance-oriented Windows tweaking

[imported_goku]

I have to disagree with the OP on the fact that disabling themes does not improve performance. I know for a fact that it does improve performance while disabling certain services can be questioned. When you disabled the themes+ go into advanced, settings for performance, and un tick all that fade, shadows and animate crap, You will see a performance improvement, especially on older systems. Most of these tweaks usually improve the speed of navigating through the system and not necessarily when running a game I mean you don't exactly use the "fade and desktop shadows" when in the middle of quake..

 

[JackMDS]

Originally posted by: goku
I have to disagree with the OP on the fact that disabling themes does not improve performance. I know for a fact that it does improve performance while disabling certain services can be questioned. When you disabled the themes+ go into advanced, settings for performance, and un tick all that fade, shadows and animate crap, You will see a performance improvement, especially on older systems. Most of these tweaks usually improve the speed of navigating through the system and not necessarily when running a game I mean you don't exactly use the "fade and desktop shadows" when in the middle of quake..

You do not need to disable the related service. All of the above can be dealt with by not using Wall Papers, and disabling the effects in the Folders, and Start menus.

:sun:

 

[xtknight]

What is the difference between virtual memory and the page file? And what is paging that occurs when no page file is set?

 

[gsellis]

This is off topic. I will send you a PM

 

[Fresh Daemon]

I have to disagree with the OP on the fact that disabling themes does not improve performance. I know for a fact that it does improve performance while disabling certain services can be questioned.

When I installed XP on System I, Windows automatically disabled all but three of the visual enhancements. System II ran fully enabled and I did not notice any decrease in speed. Based on the specs of these systems I would say that if you cannot run all of XP's visual enhancements without suffering noticeable slowdowns then the computer is simply not fast enough, and you would be better to stick with 2K.


As to the validity of the pagefile results, to those who don't believe me, maybe they'll believe Anand Lal Shimpi. He recently reviewed Gigabyte's i-RAM and tested it as a pagefile drive here.

Some excerpts that reflect his impressions:

There was no real tangible performance difference between putting more memory in the system and using the hard disk for the pagefile or putting less memory in the system and using the i-RAM for the pagefile.

...for the most part, our system was slower when we had less memory and stored the swapfile in it than when we had more memory and less swap file.

For most people, you're much better off just tossing more memory in your system.

Bear in mind that the i-RAM is a solid-state device that tests as having 3-6.75x the I/O performance of a WD Raptor, and the results of putting the pagefile on it were so underwhelming, how big do you possibly think the results of putting the pagefile on another hard drive, or on another partition of the same hard drive, could be?

If the i-RAM is not really a good replacement for more memory, there is no possible way that pagefile tweaking could be.



I've appended this information to the original post as well.

 

Originally posted by: KoolDrew
Actually, RAID-0 would be bad for the pagefile. This is because each disk has to seek to their portion of the data, thus increasing the average seek time.

Uhh, if they're working at once, any seek time would be negated by the reading speed.

If you're acting like it's so much damn slower, then wouldn't the best thing be disabling the paging file? Then that would eliminate what you say is several "million times slower".

No, it wouldn't because you would still be paging to disk. All you would be doing is forcing all "private" virtual memory to stay in RAM and only allowing code and mapped files to be paged. Even if some of the "private" stuff has not been touched for hours and will not again, it will have to stay in RAM. This will cause more paging of code, for a given workload and RAM size.

The bottom line is, you should never disable the pagefile.

Uhh, ok. So certain things will be "hard-paged" anyway, but so what? That doesn't give any reason for not disabling the pagefile. There's always going to be private data that will cause hard disk usage, or more of it. And since I don't keep a ton of junk open all the time (like most of you, from what I gather), I don't want all of the extra hard disk usage slowing down my important programs.

Bottom line is, no one has given any good reasons why someone like myself should have a pagefile set.

Originally posted by: n0cmonkey
Tisk tisk. Do you need to show your roots so well? Lighten up. Show me proof you turned off Virtual Memory and not the pagefile. The screen shot showed that you had turned off the pagefile in the Virtual Memory section. So, you were wrong. It's ok, we all make mistakes. :milk;

Please, seriously, provide a good test to prove your points. You can provide the results, or we can do it. It's not a big deal to me. A lot of what you say makes sense, in theory, but doesn't seem to work out in practice. :light:

It would also help if you calmed down and didn't act like the trashy white drunk guy sleeping on the bench outside of a trailer park in the middle of Georgia.

How can you be so god damned stupid? THE CONTEXT WAS THE OPTIONS FOR VITUAL MEMORY, NOT VIRTUAL MEMORY ITSELF AS A SEPARATE TERM. LEARN TO FUCKING READ LIKE A NON-RETARD.

And I think you might need to see a counseler and work out whatever issues you have with your dad.

Originally posted by: gsellis
That is because you have no clue what you are talking about. You don't even know which is the system drive and which is the boot drive. And you think you can give advice or disagree with tuning?

Bye-bye. You will be banned for forum violations.

I doubt anyone cares about the backwards Microsoft terminology. Your point is ostentatious and irrelevant.

Originally posted by: Fresh Daemon
Since PhyuckHughe's time here will be very short I see no need to respond. In the meantime I have repaired the screenshot links. I don't seem to be able to find the shots from System II right now but they will follow later.

Respond? Well, you certainly should be answering for your inconclusive test results that are worth nothing, but your failure to show any concern is truly pathetic. You're spreading half-assed advice, just like BV is. You're no better than him, and at least he's trying to do something worthwhile, instead of trying to "debunk" something without proof and suggesting that the best thing to do is just upgrade.

 

Originally posted by: Fresh Daemon

I have to disagree with the OP on the fact that disabling themes does not improve performance. I know for a fact that it does improve performance while disabling certain services can be questioned.

When I installed XP on System I, Windows automatically disabled all but three of the visual enhancements. System II ran fully enabled and I did not notice any decrease in speed.

Well compared to your terrible testing and apparent ignorance as far as what's happening in your computer, I would say that you're not exactly the best person to be judging that. And the dude already said that there wouldn't be much of a performance difference except in the tiny places that they're used.

Based on the specs of these systems I would say that if you cannot run all of XP's visual enhancements without suffering noticeable slowdowns then the computer is simply not fast enough, and you would be better to stick with 2K.

Absoeffinglutely terrible. Again, the "upgrade instead of getting to most out of what you have" mentality. You just don't see the small differences all of these things are making. The requirements for XP aren't that much higher than 2000, so what you say here only applies to a tiny margin, and therefore is mostly utter bullcrap.

As to the validity of the pagefile results, to those who don't believe me, maybe they'll believe Anand Lal Shimpi. He recently reviewed Gigabyte's i-RAM and tested it as a pagefile drive here.

Wow, umm, it even says that it's something that's hard to test, and they didn't notice any difference. Really, I would rather go off my own results than what someone is just telling me, consiering it's subjective.

Some excerpts that reflect his impressions:

There was no real tangible performance difference between putting more memory in the system and using the hard disk for the pagefile or putting less memory in the system and using the i-RAM for the pagefile.

...for the most part, our system was slower when we had less memory and stored the swapfile in it than when we had more memory and less swap file.

For most people, you're much better off just tossing more memory in your system.

So? What does the "i-RAM" have to do with the topic? The only other thing I see being mentioned here is that it's better to have more RAM, which is a no-brainer.

Bear in mind that the i-RAM is a solid-state device that tests as having 3-6.75x the I/O performance of a WD Raptor, and the results of putting the pagefile on it were so underwhelming, how big do you possibly think the results of putting the pagefile on another hard drive, or on another partition of the same hard drive, could be?

Depends on the use. Wow, one test says one thing. Big whoop. I'm sure if I had a drive that was 5 times faster than what I have, I would notice a difference. Nevertheless, there would still be a difference no matter how big the change is. And if what I used the computer for was heavily dependant on swapfile usage, it would be just that more worthwhile to get the settings optimized for my system.

By the way, mentioning that the i-RAM supposedly has "3-6.75x the I/O performance" is really cheap and doesn't mean squat. Maybe you should use all of the test results next time you make a comparison (or maybe just read them), instead of pointing out one factor that doesn't really mean anything.

If the i-RAM is not really a good replacement for more memory, there is no possible way that pagefile tweaking could be.

Uhh, that doesn't even fit into the discussion. But wait a second... from what you say here, wouldn't it be better if I had a lot of RAM and disable the paging file, as opposed to any other configuration with a paging file?

 

[KoolDrew]

If the i-RAM is not really a good replacement for more memory, there is no possible way that pagefile tweaking could be.

I agree that messing with the pagefile is not a good replacement for more memory, but moving the pagefile to another physical drive is still a good idea.

 

[Fresh Daemon]

I agree that messing with the pagefile is not a good replacement for more memory, but moving the pagefile to another physical drive is still a good idea.

It sounds like it would help, but I think that it for it to be noticeable there would have to be a lot of disk I/O activity at the same time as the paging (that is, I/O activity unrelated to the paging). Otherwise, you're just moving the bottleneck elsewhere. However, I think it has to be said again that if something like the i-RAM is so underwhelming, any Winchester drive (even some 15,000rpm SCSI monster) is going to be even more so. It's definitely not worth investing in another hard drive!

 

[gsellis]

Originally posted by: Fresh Daemon

I agree that messing with the pagefile is not a good replacement for more memory, but moving the pagefile to another physical drive is still a good idea.

It sounds like it would help, but I think that it for it to be noticeable there would have to be a lot of disk I/O activity at the same time as the paging (that is, I/O activity unrelated to the paging). Otherwise, you're just moving the bottleneck elsewhere. However, I think it has to be said again that if something like the i-RAM is so underwhelming, any Winchester drive (even some 15,000rpm SCSI monster) is going to be even more so. It's definitely not worth investing in another hard drive!

If you move the whole pagefile from the boot drive, you also lose the advantage of getting mini-dumps. If you want to move the pagefile, leave a 'token' pagefile on the boot drive (16MB?). The core OS components that a crash is going to analysis are stored in it, but only if it is on the boot drive (not the system drive).

And for the record, the system drive is where POST looks for boot information. It is usually C:. The boot drive is where %windir% is. This comes into play with multi-boot installations or alternate media starts.

 

[F U]

Originally posted by: KoolDrew
Actually, RAID-0 would be bad for the pagefile. This is because each disk has to seek to their portion of the data, thus increasing the average seek time.

Uhh, if they're working at once, any seek time would be negated by the reading speed.

If you're acting like it's so much damn slower, then wouldn't the best thing be disabling the paging file? Then that would eliminate what you say is several "million times slower".

No, it wouldn't because you would still be paging to disk. All you would be doing is forcing all "private" virtual memory to stay in RAM and only allowing code and mapped files to be paged. Even if some of the "private" stuff has not been touched for hours and will not again, it will have to stay in RAM. This will cause more paging of code, for a given workload and RAM size.

The bottom line is, you should never disable the pagefile.

Uhh, ok. So certain things will be "hard-paged" anyway, but so what? That doesn't give any reason for not disabling the pagefile. There's always going to be private data that will cause hard disk usage, or more of it. And since I don't keep a ton of junk open all the time (like most of you, from what I gather), I don't want all of the extra hard disk usage slowing down my important programs.

Bottom line is, no one has given any good reasons why someone like myself should have a pagefile set.

Originally posted by: n0cmonkey
Tisk tisk. Do you need to show your roots so well? Lighten up. Show me proof you turned off Virtual Memory and not the pagefile. The screen shot showed that you had turned off the pagefile in the Virtual Memory section. So, you were wrong. It's ok, we all make mistakes. :milk;

Please, seriously, provide a good test to prove your points. You can provide the results, or we can do it. It's not a big deal to me. A lot of what you say makes sense, in theory, but doesn't seem to work out in practice. :light:

It would also help if you calmed down and didn't act like the trashy white drunk guy sleeping on the bench outside of a trailer park in the middle of Georgia.

How can you be so god damned stupid? THE CONTEXT WAS THE OPTIONS FOR VITUAL MEMORY, NOT VIRTUAL MEMORY ITSELF AS A SEPARATE TERM. LEARN TO FUCKING READ LIKE A NON-RETARD.

And I think you might need to see a counseler and work out whatever issues you have with your dad.

Originally posted by: gsellis
That is because you have no clue what you are talking about. You don't even know which is the system drive and which is the boot drive. And you think you can give advice or disagree with tuning?

Bye-bye. You will be banned for forum violations.

I doubt anyone cares about the backwards Microsoft terminology. Your point is ostentatious and irrelevant.

Originally posted by: Fresh Daemon
Since PhyuckHughe's time here will be very short I see no need to respond. In the meantime I have repaired the screenshot links. I don't seem to be able to find the shots from System II right now but they will follow later.

Respond? Well, you certainly should be answering for your inconclusive test results that are worth nothing, but your failure to show any concern is truly pathetic. You're spreading half-assed advice, just like BV is. You're no better than him, and at least he's trying to do something worthwhile, instead of trying to "debunk" something without proof and suggesting that the best thing to do is just upgrade.

 

[F U]

Originally posted by: Fresh Daemon

I have to disagree with the OP on the fact that disabling themes does not improve performance. I know for a fact that it does improve performance while disabling certain services can be questioned.

When I installed XP on System I, Windows automatically disabled all but three of the visual enhancements. System II ran fully enabled and I did not notice any decrease in speed.

Well compared to your terrible testing and apparent ignorance as far as what's happening in your computer, I would say that you're not exactly the best person to be judging that. And the dude already said that there wouldn't be much of a performance difference except in the tiny places that they're used.

Based on the specs of these systems I would say that if you cannot run all of XP's visual enhancements without suffering noticeable slowdowns then the computer is simply not fast enough, and you would be better to stick with 2K.

Absoeffinglutely terrible. Again, the "upgrade instead of getting to most out of what you have" mentality. You just don't see the small differences all of these things are making. The requirements for XP aren't that much higher than 2000, so what you say here only applies to a tiny margin, and therefore is mostly utter bullcrap.

As to the validity of the pagefile results, to those who don't believe me, maybe they'll believe Anand Lal Shimpi. He recently reviewed Gigabyte's i-RAM and tested it as a pagefile drive here.

Wow, umm, it even says that it's something that's hard to test, and they didn't notice any difference. Really, I would rather go off my own results than what someone is just telling me, consiering it's subjective.

Some excerpts that reflect his impressions:

There was no real tangible performance difference between putting more memory in the system and using the hard disk for the pagefile or putting less memory in the system and using the i-RAM for the pagefile.

...for the most part, our system was slower when we had less memory and stored the swapfile in it than when we had more memory and less swap file.

For most people, you're much better off just tossing more memory in your system.

So? What does the "i-RAM" have to do with the topic? The only other thing I see being mentioned here is that it's better to have more RAM, which is a no-brainer.

Bear in mind that the i-RAM is a solid-state device that tests as having 3-6.75x the I/O performance of a WD Raptor, and the results of putting the pagefile on it were so underwhelming, how big do you possibly think the results of putting the pagefile on another hard drive, or on another partition of the same hard drive, could be?

Depends on the use. Wow, one test says one thing. Big whoop. I'm sure if I had a drive that was 5 times faster than what I have, I would notice a difference. Nevertheless, there would still be a difference no matter how big the change is. And if what I used the computer for was heavily dependant on swapfile usage, it would be just that more worthwhile to get the settings optimized for my system.

By the way, mentioning that the i-RAM supposedly has "3-6.75x the I/O performance" is really cheap and doesn't mean squat. Maybe you should use all of the test results next time you make a comparison (or maybe just read them), instead of pointing out one factor that doesn't really mean anything.

If the i-RAM is not really a good replacement for more memory, there is no possible way that pagefile tweaking could be.

Uhh, that doesn't even fit into the discussion. But wait a second... from what you say here, wouldn't it be better if I had a lot of RAM and disable the paging file, as opposed to any other configuration with a paging file?

Originally posted by: gsellis
If you move the whole pagefile from the boot drive, you also lose the advantage of getting mini-dumps. If you want to move the pagefile, leave a 'token' pagefile on the boot drive (16MB?). The core OS components that a crash is going to analysis are stored in it, but only if it is on the boot drive (not the system drive).

Really? Care to provide any sources with credence to back up that claim?

And for the record, the system drive is where POST looks for boot information. It is usually C:. The boot drive is where %windir% is. This comes into play with multi-boot installations or alternate media starts.

Microsoft terminology. Doesn't apply to everywhere.

 

[Sunner]

You know, it's against the forum rules to start multiple accounts when you get banned.

 

[Fresh Daemon]

Bottom line is, no one has given any good reasons why someone like myself should have a pagefile set.

How about Black Viper? He reports that without a pagefile he gets crashes, mysterious problems, programs refusing to load, games running without sound, and more.

That doesn't sound like fun to me.

 

[F U]

Originally posted by: Fresh Daemon

Bottom line is, no one has given any good reasons why someone like myself should have a pagefile set.

How about Black Viper? He reports that without a pagefile he gets crashes, mysterious problems, programs refusing to load, games running without sound, and more.

That doesn't sound like fun to me.

Well if I had any of those problem, that would certainly be a reason! But the thing is, I've never had ANY problems with it. The most I've seen it affect is when Photoshop loads and it gives a warning, saying that not having a pagefile doesn't affect performance and it isn't recommended. Well, the thing about that is that I actually did notice a difference, and I didn't have to wait up to several seconds to switch back to it.

This just makes me think that Black Viper, and you since you seem to agree with him based on personal experience (although I think you act like that for many things you don't really know), only had problems with it because of some user error.

 

[Fresh Daemon]

This just makes me think that Black Viper, and you since you seem to agree with him based on personal experience... only had problems with it because of some user error.

So, when BV's findings disagree with your ideas, it's user error - he's an idiot. And when they coincide, he's an intellectual giant. It does not occur to you that if he makes a mistake somewhere, he could make a mistake anywhere. It also does not occur to you that to accept some of his claims without evidence or proof is fallacious when you actually concede that his other findings are erroneous.

That's called cognitive dissonance.

 

[n0cmonkey]

F U, what testing procedure would you impliment? How would you test BV's changes? What changes would you make that you would test?

 

[MikeyIs4Dcats]

Originally posted by: n0cmonkey
F U, what testing procedure would you impliment? How would you test BV's changes? What changes would you make that you would test?

he can't hear you...

 

[n0cmonkey]

Originally posted by: MikeyIs4Dcats

Originally posted by: n0cmonkey
F U, what testing procedure would you impliment? How would you test BV's changes? What changes would you make that you would test?

he can't hear you...

He can respond with his fifth account.

 

[okb]

Goody...

 

[Fresh Daemon]

Update: Added a new section on a modern computer. The results were just as underwhelming. Services tweaking is useless, it's official.

 

[gsellis]

Originally posted by: Fresh Daemon
Update: Added a new section on a modern computer. The results were just as underwhelming. Services tweaking is useless, it's official.

Thanks.

 

[Nnyan]

I did my own testing on several systems based on BV's (and others like tweakxp.com) guides. While they were small I did see increases of memory and available cpu cycles. This did lead to some increased performance (the increased performance it was nothing to crow about but at least on my systems they were there). If I recall correctly I did 5 runs of each bench and monitored using Free Meter and Serious Samurize.

What I did notice was a decent decrease in boot times and shut down times. I also left the computers running for 5 days and at the end of that time period I had a noticeable (again nothing huge) gain in available memory.

I never did get around to quantifying this but while my systems never "seemed" faster they seem to have fewer incidents of "lag" or being bogged down.

 

[Fresh Daemon]

While they were small I did see increases of memory and available cpu cycles.

Granted, the available memory is increased. What I've seen in all these tests is that there isn't any increase in performance because of this. The reasons for this have been gone over in this thread, but basically, Windows will page out an unused service so although it appears that more RAM is freed up, you aren't actually gaining anything since Windows would have freed the RAM anyway.

This is why the benchmarks before and after tweaking, even when all available RAM is used up, don't show any real difference. All that these tweaks do is remove the flexibility from the system, for instance, if you ever needed one of those services, Windows could load it into memory and it'd work. OTOH, if you'd disabled it, then whatever app needed that service just wouldn't run.

What I did notice was a decent decrease in boot times and shut down times.

That may well be true, but I don't find that to be any useful metric for computer performance anyway. Macs still sell despite the fact that MacOS X takes forever to start compared to XP. So does Linux. Most people here (and most computer users in general) boot and shut down their computers once a day, if that.

 

[n0cmonkey]

Originally posted by: Fresh Daemon
Most people here (and most computer users in general) boot and shut down their computers once a day, if that.

Wait, these things turn off?