Performance-oriented Windows tweaking debunked

Fresh Daemon · Sep 4, 2005

"Something is bunk, then it's de-bunked." -- Jerry Seinfeld

Black Viper (amongst others) runs a Windows tweaks page offers many changes and customizations that users can make to MS Windows in the name of greater performance. BV recommends that many active-by-default Windows services be set to manual activation or disabled altogether, to save on memory useage and CPU cycles.

I have taken two systems with clean installs of Windows XP and reviewed these tweaks. Both are lower-end systems that should theoretically benefit most from this treatment, neither has the prefix giga in either its CPU speed or RAM capacity.

Part I: Low-end system

First up is the PII-300 system, in fact a PII-233 overclocked to 300MHz, with 160MB of RAM and an nVidia TNT2 PCI graphics card. This system is not really capable of being a useable XP box. We'll see if BV can help make it into one.

First I measure RAM utilization in Task Manager, then I run some components of PCMark 2004 since this system lacks the hardware capability to run portions of the test. As peak commit charge while running this suite is about twice the physical RAM, if performance can be improved it should show here. Each benchmark was run three times and the results averaged, with a reboot in-between to prevent caching skewing the results.

I also ran 3DMark 2001SE in the same average-of-three manner, although not all tests were completed since, again, the computer isnt capable. Then I ran Quake II timedemos, best-of-three FPS on demo1.dm2. 3DMark and Quake II were run at 640x480x16 to avoid a possible graphics card bottleneck.

Before the results, a note on usability. The "Power User" level is a misnomer since it actually disables many services a power user might want or need, such as the IMAPI CD-burning service, the logical disk manager, the Themes (the pretty interface, which I happen to prefer to the mid-90s look), and the Windows firewall. This also poses a security risk. Since I was not running a 3rd-party firewall (didnt want to waste the RAM), once I had gone to the Power User tweak level the computer was vulnerable. I simply unplugged the Ethernet cable from this point on, although it wasnt necessary at the Barebones level because a computer thus configured isnt actually capable of using any kind of network. Dont imagine that the services BV disables to achieve his Power User and Barebones configurations are unnecessary or useless for the desktop user, theyre not, and memory savings here definitely come at the expense of functionality.

RAM useage

This shows RAM used on first booting the system and consulting Task Manager. The Manager does consume a little overhead, but its the same across the board. I also show the saved RAM, the percentage improvement, and the value of that saving, calculated at about 8.7 cents per MB (Newegg offers 1GB of Corsair value RAM for $89.75).

Fresh installation: 90.6MB
Safe tweak level: 85.4MB (5.2MB saved or a 5.74% improvement, 7.5% more available, value: 45 cents)
Power User tweak level: 66.1MB (24.5MB saved or a 27.04% improvement, 35.3% more available, value: $2.13)
Barebones tweak level: 57.2MB (33.4MB saved or a 36.87% improvement, 48.1% more available, value: ($2.91)

This does look good as a percentage. However, bear in mind that on a system with 2GB of RAM the absolute saving will be practically identical. On the 160MB box, getting 33.4MB back amounts to 20% of total physical RAM, but with 2GB, its a piddling 1.63%. At current prices, just buy more RAM, especially considering the massive loss in functionality to achieve that 1.63%.

PCMark results (File Compression & File Decompression in MB/s)

These types of operations tend to show performance increases with better latency and speed in benchmarks. Lets see how the different configurations fare, along with percentile improvement over an untweaked system.

Fresh installation: 0.494/3.036
Safe tweak level: 0.482/3.025 (-2.43%/-0.36%)
Power User tweak level: 0.474/2.990 (-4.22%/-1.54%)
Barebones tweak level: 0.486/3.011 (-1.65%/-0.83%)

The tweaks are an abject failure here, yielding an extremely slight but across-the-board decrease in performance. Its not for certain that it is the tweaks which have made the performance worse since the changes are so small and inconsistent, however, we can certainly say that they have made absolutely no improvements.

PCMark results (File Encryption & Virus Scanning in MB/s)

Fairly standard desktop tasks. Lets see how the tweaks can help performance here.

Fresh installation: 3.521/181.629
Safe tweak level: 3.487/187.717 (-0.98%/+3.35%)
Power User tweak level: 3.520/184.549 (-0.03%/+1.61%)
Barebones tweak level: 3.486/179.930 (-1%/-0.94%)

Again, we see absolutely infinitesimal changes, but not overly favourable to tweaking and unsupportive of the idea that freeing up more RAM automatically increases performance - the configuration with the most free memory was the worst performer.

PCMark results (Grammar Check in KB/s)

Another pretty standard task.

Fresh installation: 0.395
Safe tweak level: 0.385 (-2.6%)
Power User tweak level: 0.383 (-3.13%)
Barebones tweak level: 0.381 (-3.67%)

More infinitesimal numbers that definitely wouldnt be noticed by the end-user, however, here we actually have a trend: more tweaking makes performance progressively worse, although not in proportion to the RAM saved, so again the results don't really support any conclusion other than that services tweaking does not have a positive effect on performance.

PCMark results (Image Processing in MPixels/s)

Image processing tasks are supposed to like lots of RAM. Lets see how freeing up a little more helps.

Fresh installation: 1.615
Safe tweak level: 1.616 (+0.06%)
Power User tweak level: 1.570 (-2.87%)
Barebones tweak level: 1.596 (-1.19%)

Again, nothing noticeable. Tweaking proved to be a complete waste of time here as well.

3DMark results (Car Chase, Dragothic, Lobby)

Here are the recorded framerates, along with the average % change in performance (sum of the % changes/3).

Fresh installation: 4.0/3.1/10.1
SafeSafe tweak level: 3.9/2.9/10.0 (-3.49%)
Power User tweak level: 3.9/3.0/10.2 (-1.63%)
Barebones tweak level: 3.9/3.0/10.0 (-2.3%)

Again, the tweaks seem to have a negative effect, but not a consistent one. The only clear messages here are that tweaking does not help, and also that this system seriously sucks at 3D graphics.

Quake II FPS

An old game, but within the capabilities of this system.

Fresh installation: 25.4
Safe tweak level: 25.7 (+1.18%)
Power User tweak level: 25.57 (+0.67%)
Barebones tweak level: 25.47 (+0.28%)

Although this may look like a triumph for tweaking, absolutely nobody, ever, is going to notice 0.3fps (the biggest increase). Nor are the results consistent. In fact, these kinds of results could very easily be caused by something else in the system and not tweaking at all.

The conclusion at this point is obvious: tweaking doesn't help real-world performance at all, at least, not in gaming or any of the common desktop tasks tested here. In fact, it seems to be an overall net loss. Don't bother, if you need more RAM, buy some.

Part II: Mid-range system

OK, so this system isn't exactly mid-range, but it can run XP comfortably. It's a PIII-667 with 640MB of RAM and a GeForce DDR card. It's a brand-new installation of XP SP2, untouched except for the 71.89 nVidia drivers and the installation of the required benchmarking suites and programs.

RAM useage

Fresh installation: 93.7MB
Safe level: 86.7MB (7MB saved, 7.5% reduction or 1.2% more available RAM, value: 61 cents)
Power user level: 70.0MB (13.7MB saved, 14.6% reduction or 2.5% more available, value: $1.19)
Bare-bones level: 61.7MB (22MB saved, 23.5% reduction or 4% more available, value: $1.91)

The absolute and percentile RAM savings here are even less than on the low-end system. The PIII box uses more memory because it has more attached hardware, but even so, it seems the more RAM you have (and the more peripheral hardware), the less these tweaks are worth.

Before I get into the benchmarks, I ran 5 tests on each one (with a reboot in between each) on the same system configured in the exact same way. This is to give you some idea of the margins of normal variation. They are listed as min/max/avg/% variable. Scores after tweaking within these rough boundaries cannot be ascribed to the tweaks, but rather to the anomalous behaviour of a highly complex system such as a modern computer.

PCMark 2004: 956/1023/987 -3.24/+3.65%. The PCMark scores show the greatest variation.
3DMark 2001SE: 2375/2414/2395 -0.84/+0.79%. 3DMark is a lot more consistent.
Quake III: 64.3/64.5/64.4 -0.16/+0.16%. This is the least variable of the benchmarks. A Quake III benchmark run should vary by more than 0.1-0.2 FPS before being ascribed to outside influence.

PCMark 2004

Fresh installation: 1019
Safe level: 1041 (+2.16%)
Power user level: 1044 (+2.45%)
Bare-bones level: Did not complete (system too crippled at this point), but the individual results are similar.

Tweaking appears to offer a performance increase here, but it is so slight that it falls well within the boundaries of normal variation. There is no conclusive proof that tweaking has helped performance at all.

3DMark 2001 SE

Fresh installation: 2363
Safe level: 2379 (+0.68%)
Power user level: 2414 (+2.16%)
Bare-bones level: 2409 (+1.95%)

The Power-user and Bare-bones configurations appear to offer some performance advantage outside of the normal tolerances. However, the inconsistency of these results suggests that it's not due to saving RAM and CPU cycles, or else the bare-bones system should have performed better than the power-user one, not worse.

Quake III Arena 1.32, OCAU's slayer demo

Fresh installation: 64.4
Safe level: 64.7 (+0.47%)
Power user level: 64.2 (-0.31%)
Bare-bones level: 63.9 (-0.78%)

Again, inconsistent and unnoticeable results. As before, nobody will ever notice 0.3fps, and the more services are disabled, the worse the performance gets, not the better.

In conclusion to Part II, the idea that services tweaking can actually produce useful and noticeable performance gains is proven wrong again. In no way are these tweaks worth the time or the sacrifice in functionality.

Part III: "Swap files", Themes and other miscellany

Now we come to BV's non-services recommendations.

First off, he discusses "swap files" (by which he means page files). His recommendations, from the worst performance to the best:

1) The Default: A dynamic swap file on the same partition and physical hard drive (usually C as Windows.
2) A dynamic swap file on a separate partition, but on the same physical hard drive as Windows.
3) A static swap file on a separate partition, but on the same physical hard drive as Windows.
4) A dynamic swap file on a separate hard drive (and preferably, controller) from Windows and frequently accessed data.
5) A static swap file on a separate hard drive (and preferably, controller) from Windows and frequently accessed data.
6) No swap file at all. Some software may fail. You also need "much" memory to do this. Greater than 512 MB, but I recommend 2 GB.

He has actually run some benchmarks himself, and has noticed no difference in FPS. However, his methodology isn't that good. With the load he places on the system he can't even be certain that the pagefile is even being used, and if it's not being used, it's performance won't impact FPS at all.

I went back to System 1, because it's much easier to waste 160MB of RAM than 640MB. On bootup, I loaded Task Manager, 2 instances of Internet Explorer (Google home page), Windows Movie Maker (no file loaded), Outlook Express, and 4 instances of Paint, each with a 1152x864 24-bit BMP file loaded.

Total load is about 185MB, so we know for sure that we're hitting the pagefile. This is borne out in benchmarks. The same system without any extra load gets around 25FPS in a Quake II timedemo, at exactly the same settings with this load, it gets around 20.

On to the benchmarks. Three runs of each were run and the results averaged.

Test 1: Default (dynamic, Windows-managed pagefile on system partition

Quake II: 20.5FPS
3DMark 2001 SE: 279
PCMark 2002: 739 CPU, 416 Memory, 194 HDD

Test 2: Dynamic, Windows-managed pagefile on a different partition of the same disk

Quake II: 20.3 (-1%)
3DMark 2001 SE: 279 (0%)
PCMark 2002: 738/416/193 (-0.1%/0%/-0.5%)

Distinctly underwhelming. Nothing that could be perceived.

Test 3: Static pagefile (1000MB, as recommended by BV) on a different partition of the same disk

Quake II: 20.1 (-2%)
3DMark 2001 SE: 279 (0%)
PCMark 2002: 737/421/189 (-0.3%/+1.2%/-2.6%)

Absolutely nothing noticeable again.

Test 4: Dynamic pagefile on a different hard drive

Quake II: 20.2 (-1.5%)
3DMark 2001 SE: 279 (0%)
PCMark 2002: 737/409/195 (-0.3%/-1.7%/+0.5%)

On this supposedly higher-performance configuration, the memory gain and HDD loss of the last configuration are both gone. This would seem to indicate that we're seeing variations within the benchmark's margins of error rather than actual performance differences. Anyway, there's still nothing noticeable.

Test 5: Static pagefile (1000MB) on a different drive, using a different controller channel

Quake II: 20.1 (-2%)
3DMark 2001 SE: 279 (0%)
PCMark 2002: 737/421/189 (-0.3%/+1.2%/-2.6%)

Results absolutely identical to (3). Nothing worthwhile.

I didn't test configuration (6). Windows is not supposed to run without a pagefile, and even BV acknowledges that he encounters glitches, crashes, sound problems, programs refusing to load and so on. If it's not useable, what's the point?

Anyway, pagefile tuning for performance is useless. There are no gains to be had here. About the only thing it may be worth doing is to move the swap to another partition, but that's to lessen fragmentation, not for performance.

Passwords

From BV's site:

This is only valid for Windows XP Home: Do this NOW!! Everyone on XP Home, by default, has Administrator privileges and the User name is "Owner." If I know that, so does everyone else on the planet. Change the name and / or password your account. If anything, password it. NEVER have an account unprotected! EVER!

From STaSh, AT forum member:

Passwords are fine, but since most people will probably use a password like 'password' this tidbit actually LOWERS the security of the system. Why? Because BY DEFAULT, accounts with no passwords are not allowed to connect over the network. Only to the console. If you are not worried about physical access (like most people who use XP Home), an account with no password is very secure.

Clear enough?

Get rid of System Restore Service and Indexing Service.

System Restore Service can be invaluable, if you like to use the latest drivers, change hardware, try out new software and so on. Yes, it will use hard disk space. Hard disk space costs about 50 cents per gigabyte. What do you care? Even if it uses a full gigabyte (which I doubt), I'd pay 50 cents for peace of mind.

Indexing Service is not started automatically anymore. If you don't like it and don't need it, don't use it, and it will never bother you.

Automatic Updates

From BV himself:

I turn off Automatic Updates... I highly recommend you DO NOT disable this function.

Alright?

Themes

BV recommends disabling them. As Parts I and II show, there's no performance gain to be had from disabling them. If you like the pretty looks, just use them. You won't notice a difference on any system less than 5 years old, and anything older shouldn't run XP anyway.

That's all. Please share this review, since BV encourages people to do potentially harmful acts in the name of performance that can't be realized. Help to stop him. Post it on other forums, show it to your friends, and so on. Heck, mail it to BV. Maybe he'll give up and go back to ranting about hamburgers. As technically inclined people, you know it's us who usually end up fixing the borked installs of family members, friends and newbies who end up breaking their Windows installations trying to do this stuff, so let's take a little preventative action and save ourselves some work!

eeyrjmr · Sep 4, 2005

couple of things.

your RAM figures between different installtions do not add up.
You have already stated that is should be the same

And the statement abt disabling Themes makes no difference to performance: Try running Simulink with a big model and moving stuff around - there is severe GFX lag (on the desktop!!!) that goes away when themes are disabled

Fresh Daemon · Sep 4, 2005

your RAM figures between different installtions do not add up.

Where? If you point out an error I'll correct it.

Try running Simulink with a big model and moving stuff around - there is severe GFX lag (on the desktop!!!) that goes away when themes are disabled

On what system? Does it occur on different systems? Is it repeatable? Have you eliminated all other variables?

Edit: Ah, I see where you mean. What I actually said first was:

"However, bear in mind that on a system with 2GB of RAM the absolute saving will be practically identical."

This was my speculation before I ran Part II. After running that, I noted:

"The absolute and percentile RAM savings here are even less than on the low-end system. The PIII box uses more memory because it has more attached hardware, but even so, it seems the more RAM you have (and the more peripheral hardware), the less these tweaks are worth."

Hope that clears it up.

eeyrjmr · Sep 4, 2005

Fresh Daemon said:
On what system? Does it occur on different systems? Is it repeatable? Have you eliminated all other variables?

This occurs on my system and at work
Mine is a
P4:2.4GHz
1Gig RAM

Works are I think are
P4 1.8GHz
256Meg RAM

both get GFX lag when simulation is running and a window refocused,Resized, moved.
Turn themes off and no problem.

Need to move thing around to look between 4 different scopes to see if mid-simuation there is a problem with the control laws and stop the simulation thus saving time.

In those instances as Much RAM and CPU usage as possible (even if it is 1%) makes alot off difference when the average simulation time is like 1Hour

Tordek · Sep 4, 2005

eeyrjmr said:
This occurs on my system and at work
Mine is a
P4:2.4GHz
1Gig RAM

Works are I think are
P4 1.8GHz
256Meg RAM

both get GFX lag when simulation is running and a window refocused,Resized, moved.
Turn themes off and no problem.

Need to move thing around to look between 4 different scopes to see if mid-simuation there is a problem with the control laws and stop the simulation thus saving time.

In those instances as Much RAM and CPU usage as possible (even if it is 1%) makes alot off difference when the average simulation time is like 1Hour

thats a problem with Matlab not Windows... And you as a seemingly usual matlab user should know how buggy that program could get...

eeyrjmr · Sep 4, 2005

chinoquezada said:
thats a problem with Matlab not Windows... And you as a seemingly usual matlab user should know how buggy that program could get...

Run s fine in Linux

The fact is I can stop all services not needed in linux and get great sim time (using a light-weight WM)

The fact is, you might say it is a MAtlab problem, but the other way to look at it is it is a sim program and the more CPU time and more RAM it has (even a fraction) makes a difference

So isn't it Window's Mem management that CPU scheduler that is the problem?

Zardoz · Sep 4, 2005

I like what you have done here... You have put a price/benefit on the tweaks BV has done. As with most of BV tweaks you are disabling features in windows you may not want to do. but you show that with any high-end system that the cost per tweak is not really worth it.

I think BV has some good info, but as you show. it's not really worth it. you are much better spending money for some extra hardware to get way more performance with out disabling anything.

I for one thank you for your work...

rolo · Sep 4, 2005

Quick vernacular fix for your article: on Windows the correct term is indeed "page file," not "swap file."

The page file is a backing store for virtual pages (one page is 4 KB). Another backing store is physical memory. A full discussion of virtual memory is way past the scope of this post

Needless to say it's a complex topic and one that not many people fully understand.

Fresh Daemon · Sep 4, 2005

Quick vernacular fix for your article: on Windows the correct term is indeed "page file," not "swap file."

Thank you, I have edited and corrected. BV uses the terms interchangeably, and I knew one of them was wrong, I just forgot which after hours of calculating average 3DMark scores.

I think BV has some good info, but as you show. it's not really worth it.

The irony is that his "good info", e.g. on what the services do, is mostly lifted straight from Microsoft's website. It's his own content that's useless. Unless you like rants about burgers, of course.

I for one thank you for your work...

You're welcome.

RedShark · Sep 5, 2005

Great thread, hope more BV zealots see it.

djnes · Sep 5, 2005

Excellent write up. Thanks for taking the time to do so.

Met-AL · Sep 5, 2005

Excellent. I have been saying this all along...but mention it here and a flame fest erupts.

Looks like he is moving to the Linux world. Have fun you Linux gurus when he starts his crap with Linux.

Regardless, I have purchased a Mac Mini to experiment with OS X and have ventured deeper into the Linux realm.

SmokeRngs · Sep 5, 2005

You have some flaws in the way you did your comparisons. You didn't specify the exact RAM you used for your price calculations. I'm assuming it's DDR which flaws your calculations since I seriously doubt either system uses DDR. You would have to go with RAM that the system uses to do it correctly. Using the prices for something the system can't use is a flawed premise. I really don't care about those calculations but the way you have it done destroys any value it has. You'd be better off going back and fixing that.

Also, your pagefile tests are not very good considering your trying to monitor performance differences due to the pagefile. Most people already know that hard drive access doesn't affect games much except for loading of maps and levels. Your focus on gaming does not actually bring the pagefile into the equation much. There's a lot of other software out there that does make use of the pagefile and would benefit more than gaming from having the pagefile on a different physical drive during heavy drive access times. It's very similar to uncompressing files. Read the file from one drive and have it uncompress to another physical drive, you will find that is the most efficient and quickest way of doing it instead of trying to read and write to the same drive.

Another point of contention I have is disabling themes. I disable them on every system I have. I also disable pretty little things like shadows and window animation. They do not help me in any way whatsoever and do slow down what I can do. I don't want to wait the extra time for a window or menu to pop up due to animation. I want the thing there as soon as I click on it. And all the animation and pretty features do slow down general windows usage for the display, especially when doing other system intensive tasks in the background. Besides, I can't stand the cartoony look of XP and go back to the old Win2k method of viewing the OS.

Also, I take exception to your opinion that a PII system should not have WinXP on it because it really doesn't have the speed to run it. I have a dually PII 266 system with 256 meg of RAM that I use daily. I have no trouble whatsoever running general apps on it. This is also with two F@H clients chugging along using up a bit of RAM each. The system runs general office apps with no trouble whatsoever. No, it doesn't run them as fast as the system in my sig, but then again, I wouldn't expect a PII 266 to outrun a 2.5 Ghz AMD.

I think you should rethink and review the tests you have used. Some of them are mostly worthless for the conclusions you were trying to draw. It's misinformation at it's worst. You need the correct tests to draw a proper conclusion. Do not do a disservice to the people reading this. If you go back and use different testing methodologies and come up with the same results, that is just fine. But the way you have things done is not correct.

Lyquist · Sep 5, 2005

I appreciate your work. I for one just let windows do it's thing. It seems like whenever I apply a "tweak" I tend to lose some kind of functionality or notice erractic behavior in windows, but that is just my experience.

Lyquist

Tordek · Sep 5, 2005

SmokeRngs said:
Another point of contention I have is disabling themes. I disable them on every system I have. I also disable pretty little things like shadows and window animation. They do not help me in any way whatsoever and do slow down what I can do. I don't want to wait the extra time for a window or menu to pop up due to animation. I want the thing there as soon as I click on it. And all the animation and pretty features do slow down general windows usage for the display, especially when doing other system intensive tasks in the background. Besides, I can't stand the cartoony look of XP and go back to the old Win2k method of viewing the OS.

Most users like their interface with themes. I do.

Disabling them to gain performance in a test would be like cheating on 3dmark (nvidia n ati). Some users want their OS with the graphics included.

SmokeRngs said:
Also, I take exception to your opinion that a PII system should not have WinXP on it because it really doesn't have the speed to run it. I have a dually PII 266 system with 256 meg of RAM that I use daily. I have no trouble whatsoever running general apps on it. This is also with two F@H clients chugging along using up a bit of RAM each. The system runs general office apps with no trouble whatsoever. No, it doesn't run them as fast as the system in my sig, but then again, I wouldn't expect a PII 266 to outrun a 2.5 Ghz AMD.

Wrong.
I have installed and tested pII 266 systems with 512 MB of RAM and the OS run flawlessly. Now if you set out to do some processor intensive tasks, there you would hit a brick wall. But normal operation holds no trouble to these.

SmokeRngs said:
I think you should rethink and review the tests you have used. Some of them are mostly worthless for the conclusions you were trying to draw. It's misinformation at it's worst. You need the correct tests to draw a proper conclusion. Do not do a disservice to the people reading this. If you go back and use different testing methodologies and come up with the same results, that is just fine. But the way you have things done is not correct.

I agree with your pagefile test comment. Other than that the tests were good enough for me...

ThreeDee · Sep 5, 2005

Thanks for your time in this and sharing your results...

Fresh Daemon · Sep 6, 2005

Thanks to all who made supportive comments!

You have some flaws in the way you did your comparisons. You didn't specify the exact RAM you used for your price calculations. I'm assuming it's DDR which flaws your calculations since I seriously doubt either system uses DDR.

Do you have any real reason to suspect that memory useage patterns after these tweaks would be substantially different on a DDR system as opposed to an SDR system? Because if not, the price calculations are accurate. I don't believe that memory types affect the amount of memory used, only the speed at which it can be accessed. Perhaps you have evidence to the contrary.

Also, your pagefile tests are not very good considering your trying to monitor performance differences due to the pagefile. Most people already know that hard drive access doesn't affect games much except for loading of maps and levels.

No, if you read the tests you will find that running the game while making extensive use of the swapfile caused about a 20% loss in framerate. If what you said was true, I should have seen slower loading times as Windows swapped everything but the game to disk and then identical framerates, but that was not the case at all. The swapfile size in all cases was more than adequate to hold all the running programs several times over, should it have been possible.

You may also wish to take note of the memory and hard disk benchmarks from PCMark. They also show no noticeable difference from swapfile tweaking.

I may run more intensive tests of swapfile optimization in the future. Perhaps Photoshop benchmarks. I will update as and when.

Also, I take exception to your opinion that a PII system should not have WinXP on it because it really doesn't have the speed to run it. I have a dually PII 266 system with 256 meg of RAM that I use daily.

I did not say that a PII system could not run WinXP. I said that a PII-300 with 160MB of RAM should not run WinXP. This was borne out by my experiments: you can either run the thing and only have 70MB of RAM left over for apps (not a lot), or you can disable so much stuff that it becomes less functional than 2K, NT or 98.

I also disable pretty little things like shadows and window animation. They do not help me in any way whatsoever and do slow down what I can do.

Well, I'm interested in seeing some sort of evidence that that is the case. But as what you've said so far in your post has been wrong, you will excuse me if I don't take your word for it.

Looks like he is moving to the Linux world. Have fun you Linux gurus when he starts his crap with Linux.

I'm sure the floodgates will soon open on the Linux help forums. Is he some sort of virus? He's porting his OS-breaking talents to other platforms?

Sunni_Phono · Sep 6, 2005

Excellent post.

Carnival Forces · Sep 6, 2005

OS tweaks do have a major impact, specifically w/ Windows XP. Search for a thread in which i posted something along these lines: i tested Windows (win xp pro ) w/ a normal restart vs. "Safe Mode with Networking"

there was something like 4 minutes per frame difference in Folding @ Home -- that's massive, that's hours shaved off work unit completion time.

you're wrong.

Fark_Maniac · Sep 6, 2005

/puts the flame suit on...

It is hard to write a review, on this forum, on what is the optimum settings for an OS/Drivers/your serta mattress. This forum has everything from "How do you turn a PC on?? lolz" to "Would lapping the heatsink to a LiquidMetal setup bring major results with AS5?" We have all sorts of peeps here and from what I read, you thread on each side of the spectrum as well as some more opinionated subjects.

For the noob...this would probably be a decent jumping board to get a general picture.

For the more advanced PC user, they already have their opinions and will not change their minds without seeing it in print and/or screen shots of test projects that have been mulled over by several members.

I believe that a more advanced user will perform some disables of services more out of user preferences, as stated by SmokeRngs. He disables themes for two reasons. He does not like the eye-candy...that's a personal preference. He also does not believe that the shadows will benefit him. He may be coming from the mindset that with a base of NT-like boxed menus and windows w/o shadows...the effect of adding shadows to said menus and windows will add extra clock cycles for either the CPU or GPU. I would tend to agree with this mindset.

I had to chuckle a bit at chinoquezada for this "Wrong" comment. SmokeRngs is saying that a PII system can run XP just fine...just like you do.

On that subject, making a blanket statement that XP should not be used on a PII system is making too much of a generalization. A PII system with XP would make a great file server/gateway/http/ftp server. Adding 'heavy use workstation' would be correct.

Please don't take this as myself being an in your face arse....but first draft reviews that set out to prove a point, generally have some short comings. Starting a thread as wanting to have a collaborative project would produce much better results.

Tordek · Sep 6, 2005

Carnival Forces said:
OS tweaks do have a major impact, specifically w/ Windows XP. Search for a thread in which i posted something along these lines: i tested Windows (win xp pro ) w/ a normal restart vs. "Safe Mode with Networking"

there was something like 4 minutes per frame difference in Folding @ Home -- that's massive, that's hours shaved off work unit completion time.

you're wrong.

Oh you mean safe mode with no sound, no vid card drivers loaded, 256 bit color, and pretty much nothing else to do in windows but look at Folding @ Home process work units and run spyware/virus scans?

Windows in normal startup has to load drivers and such so the commit charge rises. This is inevitable.

What we are saying is that the difference between a normal windows startup and a "tweaked" windows startup isnt noticeable performance wise... And not worth the loss of funcionality and/or visual appeal (for those of us who like it...)

Fark_Maniac said:
For the more advanced PC user, they already have their opinions and will not change their minds without seeing it in print and/or screen shots of test projects that have been mulled over by several members.

QFT

Fark_Maniac said:
I had to chuckle a bit at chinoquezada for this "Wrong" comment. SmokeRngs is saying that a PII system can run XP just fine...just like you do.

Damn you are right...

Sorry for that... We all make mistakes...

ameoba · Sep 6, 2005

The reason the tweaks didn't work is 'cuz you didn't say "RABBIT RABBIT" when you got out of bed (LEFT FOOT FIRST!!!) on the 2nd tuesday of the month. If you'd followed this simple extra step, your results would've shown radically different results.

MrkXCeL · Sep 6, 2005

if it don't get better - beat it till it does

now tell me again - why did I spend all that time scratching my head how many partitions I would need to have my SCRATCH/OS/PROGRAMs all seperated from eachother -F@$& triple boot gowd deng waste of freakin ... has the a64x2 price dropped yet? anyone ... please ... anyone ? save me ... need new PC now

rolo · Sep 6, 2005

Carnival Forces said:
OS tweaks do have a major impact, specifically w/ Windows XP. Search for a thread in which i posted something along these lines: i tested Windows (win xp pro ) w/ a normal restart vs. "Safe Mode with Networking"

there was something like 4 minutes per frame difference in Folding @ Home -- that's massive, that's hours shaved off work unit completion time.

you're wrong.

So you're using one example that involves Safe Mode to debunk his entire article, something which doesn't relate to what he wrote anyway?

Granted, I don't put full faith in benchmarks having a complete translation over to real use performance. The main thing to take away from the rough draft article is that these so-called "tweaks" are not worth the effort. Most of the gain is placebo.

SmokeRngs · Sep 6, 2005

rolo said:
So you're using one example that involves Safe Mode to debunk his entire article, something which doesn't relate to what he wrote anyway?

Granted, I don't put full faith in benchmarks having a complete translation over to real use performance. The main thing to take away from the rough draft article is that these so-called "tweaks" are not worth the effort. Most of the gain is placebo.

This is one of my main problems with his testing. He is not using the proper tools to do the job. Using 3DMark or Quaketo test in a situation where pagefile usage (disk access) is fhe focus will not produce any meaningful results.

In my previous post I had completely avoided the synthetic vs realworld benchmark situation. I have my beliefs on that which I do not care to fully share here since this isn't the soapbox. I will say that in the case of PCMark, I have seen some really weird scores come from it out of nowhere and therefore, do not trust it much. That is all I am going to say on that subject since I would prefer to continue along the line of analyzing the methods used for testing.

Also, as far as the Safe Mode comment goes, just think of what is running in the background to cause such a difference in times for something vs running it "normal". Yes, drivers and such are going to use RAM but Safe Mode also uses drivers, just ones not specifically optimized for your hardware. Therefore, it should not actually run as well as Windows normally would. Safe Mode, is just that a "safe" loading of generic Windows drivers in order to troubleshoot problems with a system. It's main function is stablity, not speed. I'm not telling anyone they need to run their system in Safe Mode, but if there are performance gains going from a "normal" setup of Windows to Safe Mode, there must be something there that is seriously degrading performance.

Fresh Daemon · Sep 7, 2005

OS tweaks do have a major impact, specifically w/ Windows XP. Search for a thread in which i posted something along these lines: i tested Windows (win xp pro ) w/ a normal restart vs. "Safe Mode with Networking"

That's not tweaking, that's intentionally castrating your system. Even BV claims that a system should be useable for common tasks. An XP box running in safe mode doesn't understand a network or the internet, can't use more than 256 colours, has no sound, can't play games, etc. So if that's what you think competes with a real, functional desktop, then go ahead. I would like to see screenshots of your safe-mode XP box surfing the web, ripping MP3s, downloading torrents, applying Photoshop filters and playing Doom 3, please!

Starting a thread as wanting to have a collaborative project would produce much better results.

I would love it if others would run similar tests. This is a first draft and I intend to refine it based on the feedback I receive. Suggesting more benchmarks I could run and different configurations I could try is constructive criticism and is welcome, claiming that running in Safe Mode debunks my findings or claiming it "just feels faster" without a shred of objective evidence is not.

In my previous post I had completely avoided the synthetic vs realworld benchmark situation.

This is why I also ran game benchmarks. Admittedly those are still benchmarks, however, they accurately represent performance you would actually get when running games. The reason we have benchmarks is to give objective results which people can then apply their subjective opinions to. For instance, we can say that X system runs Far Cry at 40fps. Some people will say, "40fps is pretty good. I'd be happy with that." Others may say, "40fps seems choppy to me. I should get something faster." But merely saying, "Far Cry seems pretty fast to me on X system" is meaningless to everyone else.

Tordek · Sep 7, 2005

Fresh Daemon said:
That's not tweaking, that's intentionally castrating your system. Even BV claims that a system should be useable for common tasks. An XP box running in safe mode doesn't understand a network or the internet, can't use more than 256 colours, has no sound, can't play games, etc. So if that's what you think competes with a real, functional desktop, then go ahead. I would like to see screenshots of your safe-mode XP box surfing the web, ripping MP3s, downloading torrents, applying Photoshop filters and playing Doom 3, please!

He did say he used "with networking"

Aside from that, completly agree. No use running in safe mode.

Fark_Maniac · Sep 7, 2005

Fresh Daemon said:
I would love it if others would run similar tests. This is a first draft and I intend to refine it based on the feedback I receive. Suggesting more benchmarks I could run and different configurations I could try is constructive criticism and is welcome, claiming that running in Safe Mode debunks my findings or claiming it "just feels faster" without a shred of objective evidence is not.

there have been tests run, collaboratively, in the past attempting to measure the page file aspect. it had a bajillion posts and views. I believe the benchmark I used was Winstone or something like that. It basically compared a 1GB PF vs 512MB PF vs 0MB PF and the effects on file accesses and performance. Supprisingly, the gains were negligible running w/o a pagefile. To discuss THAT aspect further, discuss it in said thread, not here.
Edit: http://hardforum.com/search.php?searchid=1698643
search for "pagefile" that I've participated in

I believe that the reason "safe mode" was brought into the discussion was to intruduce the affect of having services disabled and what effects would be brought about. Of course, no productive desktop or game machine would run in safe mode. But in this case, "safe mode w/ networking" would be a valid run mode to show disabled services effects.

holyroller1 · Sep 7, 2005

Fark_Maniac said:
there have been tests run, collaboratively, in the past attempting to measure the page file aspect. it had a bajillion posts and views. I believe the benchmark I used was Winstone or something like that. It basically compared a 1GB PF vs 512MB PF vs 0MB PF and the effects on file accesses and performance. Supprisingly, the gains were negligible running w/o a pagefile. To discuss THAT aspect further, discuss it in said thread, not here.
Edit: http://hardforum.com/search.php?searchid=1698643
search for "pagefile" that I've participated in

I believe that the reason "safe mode" was brought into the discussion was to intruduce the affect of having services disabled and what effects would be brought about. Of course, no productive desktop or game machine would run in safe mode. But in this case, "safe mode w/ networking" would be a valid run mode to show disabled services effects.

The problem is that Safe Mode doesn't JUST disable services. Apples, Oranges

Fark_Maniac · Sep 7, 2005

holyroller1 said:
The problem is that Safe Mode doesn't JUST disable services. Apples, Oranges

okay, so educate me. don't leave me hanging.

the only thing I can see that would degrade performance in safe mode would be that vid card drivers are not loaded, which would make the proc do all the display calculations.

Phoenix86 · Sep 7, 2005

*tagged for reading*

This deserves some attention.

rolo · Sep 7, 2005

Fark_Maniac said:
okay, so educate me. don't leave me hanging.

the only thing I can see that would degrade performance in safe mode would be that vid card drivers are not loaded, which would make the proc do all the display calculations.

Safe Mode only loads what's necessary to barely get the system to the desktop, eschewing pretty much any non-generic drivers which have the potential to prevent that. No video acceleration. I believe there's no sound, either (other than the *beep*!). And your motherboard drivers contain all sorts of code to enable faster, hardware controlled disk access (as opposed to software controlled, aka slow) which is not loaded.

It's not that performance isn't being degraded. It's just not being enabled at all.

ameoba · Sep 8, 2005

if you were doing any disk-heavy activity in safe mode, you'd notice a major decrease in performace...

Tordek · Sep 8, 2005

ameoba said:
if you were doing any disk-heavy activity in safe mode, you'd notice a major decrease in performace...

i've noticed, got any idea why?

holyroller1 · Sep 8, 2005

Fark_Maniac said:
okay, so educate me. don't leave me hanging.

the only thing I can see that would degrade performance in safe mode would be that vid card drivers are not loaded, which would make the proc do all the display calculations.

Do your own research.

Don't act like an @ss next time and you may get a more obliging response.

Fark_Maniac · Sep 8, 2005

holyroller1 said:
Do your own research.

Don't act like an @ss next time and you may get a more obliging response.

If I were looking to act that way I would say this; "I'm sorry sweetie...next time I'll remember to put in a little smile."
yes, sweetie, I could do my own research, then report back here. However, if he knows, then it would educate us all (a good thing), and back up the contradiction.

any rate, I think we can all agree that safe mode where no 3rd party drivers are loaded and services are neutered. I'm too tired to discuss this...been pulling cables in a data center all night...

eeyrjmr · Sep 8, 2005

If someone wants to turn off some services that they dont want and then say THEIR computer "feels" more responsive

How is anyone to have a go at them

That is why I dont like all those threads abt "quackviper" he might of pulled together all the infomation and added his bits, so what, the fact is the infomation was then on one page as opposed to 20 from MS site

Dont have a go at someone cause they are doing what they want with their computer
YOu have the ability to change stuff on your computer (it is yours afterall)

What would you prefer MS NOT giving you the option to change serices?

Phoenix86 · Sep 8, 2005

eeyrjmr, the thread is titled "performance-oriented windows tweaking debunked" not "preference-oriented windows tweaking debunked".

If you folks want to argue "if someone *should* vs. if someone *could*" we'll be here 'till the cows come home. I really think in the interest of keeping this thread on the topic of performance tweaks we should drop the whole "I like it this way" end of the discussion.

Fair enough?

As far as testing comparisons in safe mode, don't. rolo's description is pretty accurate as to why it's a bad idea.

As SmokeRngs noted, testing under Part III is flawed.

The processes you use to test are not testing page file performance, they are testing application performace. These tests are good for general performace indicators, but because of the way paging work (once it's paged to disk, and everything you need is loaded in RAM) the tests won't really notice *how* it got paged. See, the tests are running *after* paging has happened. You'd need to monitor disk access while paging to get an idea on how effective the paging process is.

Then there's the fork in whole discussion, systems with lots of RAM. These days people simply don't need to tweak the PF because they are generally running enough RAM to supply the entire system's memory requirements w/o paging. Well, if you aren't paging because you have enough RAM, what good does tweaking the page file do?

Honestly I would leave that part out of this article or make them two different articles if you really want to test page files. It's a can of worms.

TGA · Sep 8, 2005

Phoenix86 said:
As SmokeRngs noted, testing under Part III is flawed.

The processes you use to test are not testing page file performance, they are testing application performace. These tests are good for general performace indicators, but because of the way paging work (once it's paged to disk, and everything you need is loaded in RAM) the tests won't really notice *how* it got paged. See, the tests are running *after* paging has happened. You'd need to monitor disk access while paging to get an idea on how effective the paging process is.

Then there's the fork in whole discussion, systems with lots of RAM. These days people simply don't need to tweak the PF because they are generally running enough RAM to supply the entire system's memory requirements w/o paging. Well, if you aren't paging because you have enough RAM, what good does tweaking the page file do?

Honestly I would leave that part out of this article or make them two different articles if you really want to test page files. It's a can of worms.

I would have to agree with you on that statement. However I cant agree with what you said first. Its not only his page file testing that is flawed. His testing of performance with themes disabled is also flawed and for much the same reasons as the page file tests. His tests are run once the windows have been opened and does nothing to account for the time it takes to open those windows.

I dont know about the rest of you but I open and close windows all day long on my PC. Having the fancy GUI crap turned on will and does slow down other tasks when opening new windows and increases the time it takes to open any new window. Especially on dated hardware like he used in those tests. It would not be easy to set up a test that could be repeated over and over consistently to show what impact the themes and such have on performance but I am sure it could be done.

Anyway I have to agree that most service tweaking is a bad idea and a waste of time. However I see no befit to using flawed tests to try and debunk BVs flawed logic. I just get the impression you went in to this trying to prove that these tweaks do nothing.

Phoenix86 · Sep 8, 2005

You can disable the fancy graphic w/o disabling themes. I think that satisfies both camps, no?

It's not the service that's generating the overhead it's the use of the service.

In general there are ways to disable the unwanted side effect w/o disabling the service. I'm kind moot on themes, yeah it's basic function is to add pretty stuff, and it's easy to diable all that with one whack...

Anyways, that's back to the whole preferenc/opinion thing. You are easily change my ideas on facts, you'll have a harder time changing my opinions. I know that sounds simple, but a lot of people miss it.

So about the tests on the services. I don't see a problem with the testing method at all. It doesn't test every aspect, but it gives you an idea of how single applications test. This is keenly usefull to gamers in peticular. He could add the test that open office apps, name escapes me, and test performance there. IIRC it runs a script that opens several apps.

Performance-oriented Windows tweaking debunked

Weaksauce

[H]ardness Supreme

Weaksauce

[H]ardness Supreme

[H]ard|Gawd

[H]ardness Supreme

2[H]4U

Gawd

Weaksauce

[H]ard|Gawd

Fully [H]

Supreme [H]ardness

[H]ard|DCer of the Month - April 2008

2[H]4U

[H]ard|Gawd

[H]F Junkie

Weaksauce

Limp Gawd

Supreme [H]ardness

2[H]4U

[H]ard|Gawd

Supreme [H]ardness

[H]ard|Gawd

Gawd

[H]ard|DCer of the Month - April 2008

Weaksauce

[H]ard|Gawd

2[H]4U

n00b

2[H]4U

Supreme [H]ardness

Gawd

Supreme [H]ardness

[H]ard|Gawd

n00b

2[H]4U

[H]ardness Supreme

Supreme [H]ardness

Gawd

Supreme [H]ardness