Windows XP Maintenance and Optimization Guide


Recommended Posts

:rolleyes:

It seems to me that you response is that of a great intellectual who is willing to share his divinely given truths with me, the mortal fool, to educate and guide me to the light.

EDIT:

On second thought, lets see... that was in the power users suggestions (so not recommended for average users...) of the service tweak sections, so if you do not agree, then by all means you are not required to turn it off.

Since I started tweaking, I always turned off the Firewall, as either I had a firewall enabled on my router or seen no use for protection in the first place, and this never gave me any issues.

So, I might clarify that if you have a router firewall than disabling it might be a safer.

:)

Also, I am going to add some help for those who accidentally opened a virus/malware filled executable in the future revisions.

Edited by Udedenkz
This guide is for Windows XP and XP64 only, if you have Vista or Windows 7 then this guide is not for you. Vista is simply proven to be slower than XP and as such I recommend installing XP instead or dual booting XP and Vista, for better performance.

Are we still in early 2007? Vista is not slower than XP anymore. Your recommendation to install an antiquated OS like XP is baseless as there is no discernable performance gain by sticking to XP at this point in time. This may have been true over 2 years ago, but it isn't now.

Vista is not slower than XP anymore. Your recommendation to install an antiquated OS like XP is baseless as there is no discernable performance gain by sticking to XP at this point in time.

/sigh

Because every computer is made for video editing and video games...

Because every computer meets the minimum requirements of Vista...

Because Vista is still faster than XP on lower spec machines, such as netbooks, that barely meet the minimum requirements...

Because everyone has the same computer and everyone chooses to believe the same benchmarks...

Because a difference of opinion and free thought don't exist in a world with variables...

--

Nice guide, Ill be trying some of these tweaks on my EeePC.

--

Note: Annoyance ? Do Not Highlight New Applications, Self Explanatory

For those not comfortable with the registry, you can also right click on the taskbar and select "Properties" --> Taskbar and Start Menu Properties --> Start Menu Tab --> Customize --> Advanced Tab --> Uncheck "Highlight newly installed programs."

FACT ? Contrary to popular belief, Macs get viruses too.

I would like to see proof of this, there has never been an actual virus for the Mac. As a Windows user, this might just be wishful thinking on your part, Macs have vulnerabilities but Virus are not one of them. I'm a Window user as well, but believe me if a Virus ever hits the Mac/OSX everybody in the world will know in a matter of minutes of it being posted anywhere in the world as that will really be a first. No, proof of concept or script exploits are not Virus...

Security Center ? That thing that might nag you a lot.

Yes, it might alert you that there are problems with your OS, Firewall or Anti-virus software. Why would we want to know that?

Automatic Updates ? No point in running this all the damn time, only run when you want to install updates. Once a month would be a good idea.

This is just silly, why would you want to wait a month to patch your OS? What is the overhead for having this service running in the background?

Power Users, Performance, (I forgot what this is, seems to remove some less used functionality, netmeeting, and some other things)

Are you serious? Wow, you are telling people to use tweaks that you don't even remember what they do? Never a good idea.

You could also grab TweakUI from Microsoft and avoid using some of the reg hacks for some of the stuff you mention. It's a little safer and can be UN-done if the effect is not noticeable or adverse. Visit the Power Toys site that Microsoft has up and you might find other toys to play with as well.

How much is doing all of this going to speed up my computer? How much functionality is going to be lost? People have been debunking some of these "tweaks" for years for example XP Myths and Lifehacker.

I'm not trying to be mean or an ass, but some of the stuff you mention is putting the user at risk for virus and OS errors and just doesn't seem worth the speed you might gain. You also might want to remove the adblock stuff since it's discussion isn't allowed here, because Neowin uses ads to keep the site runn;)g. ;)

Also Foobar kicks Winamp's ass all over the place.

Good Luck.

Today's Changes,

- Due to the word limit, all the registry tweaks have been moved into a zip attachment. Just explore or extract the zip file, double click on any tweak you desire, and add it to registry.

- More tweaks have now been sorted out.

- Music player, XMPlay has been added as a suggestion.

- Internet Explorer 8 added with link.

- Security section and Anti-Virus section has a minor update

- Link Fixed

- WMP11 suggestion has been added

- Some Spelling Mistakes Fixed

- Speed Up Interface Without Disabling Themes updated.

To Do,

- More Registry Tweaks

- Sort Better Existing Tweaks

- Organize better Services Section

- Add XPAntiSpy and TweakUI

Heh, are you going to take About.com's word that there are Mac Virus? Really? Did you read what they had posted?

Some try to answer this question literally, based on the strict definition of 'virus' - i.e. malicious software that infects other files. But the term 'virus' is used much more loosely these days and in that context refers to malicious software in general (or what the industry terms 'malware').
Inqtana is a proof-of-concept worm, coded with a stop date of February 24th and confining its spread to very specific bluetooth addresses. In other words, Inqtana in its current form poses no real threat to Mac users.

So, No Actual Virus Here.

Are you going to keep adding to this guide without addressing the concerns raised by myself and others? If you are just going to ignore the questions raised by myself and others, then let me know and I will ignore this thread.

The best way to optimize Windows XP is to leave it untouched and have SP3 Integrated.

Create a user account that is "limited" and use that account, this will filter unwanted damage from various activities while browsing the internet. It's safe and super effective. Because everytimes you install an app you have to right click 'Run as' and put in your admin credentials to allow the install process to continue.

I don't understand why you people have to fight over OS X security on a Windows XP optimization thread. This is just stupid and off topic.

That is what I was just about to say. :)

If you want to contribute, give suggestions like EZ8 did - suggestions that have something to do with the guide.

Update,

- I just discovered a neat little trick that might help with un-deletable files (Have Not Tested Yet): You can login to an account level higher than Administrator which gives you access to things like System Volume Information folder. I think this is a pretty neat power user feature, or you know to show off. :) I mean, not many might realize that you can login to SYSTEM.

- I also gather a few more registry tweaks that I will add later on, maybe even like on Thursday as I do not feel like doing it now - besides I will be busy this weekend.

Good Night Y'All.

I don't understand why you people have to fight over OS X security on a Windows XP optimization thread. This is just stupid and off topic.

Since my reply had to do with what was in the guide, how is it stupid or off topic?

That is what I was just about to say. :)

If you want to contribute, give suggestions like EZ8 did - suggestions that have something to do with the guide.

Uh, didn't the original guide claim as "Fact" that Macs got Virus? Yes, then you added a link to About.com that got loose with the definition of Virus and decided that malware was the same thing. It also linked to several "proof of concept" virus that are of no threat to OSX.

I don't know why you took a shot at the Mac platform in an XP optimization guide either, but you did.

As for the guide, I posted my concerns about 1) Turning off Automatic Updates 2) Disabling the Security Center 3) Listing reg hacks that he could "forgot what they did" 4) How much of a gain in performance you would get for these tweaks.

Really in all honesty, has anyone bench marked any of these "tweaks"? What were the results? Just for ****s and giggles I will format my empty partition with a clean fully updated XP Pro and bench mark it with 3dMark and PCMark05 and then do every tweak here and post the results as soon as I am done. My system is an HP a6010n - I've added an BFG8800GT 512, upped the ram to 2 gigs and added an Antec NeoPower 550 HE power supply. Hopefully I can get this done by Monday or so. See you.

I can't seem to edit my post above - maybe a time limit on editing or something, anyway PCMarK05 and 3DMark05 are constantly crashing and I can't figure out why. They both start up and start benchmarking, but then they just stop and say testing aborted?

I will have to look for another benchmarking tool, does anyone have a suggestion. I also decided not to reformat my current XP Pro partition, but have used BlackVipers XP Pro Default service setting reg file to reset all the service to their default just to save myself some time.

As for the guide, I posted my concerns about 1) Turning off Automatic Updates 2) Disabling the Security Center 3) Listing reg hacks that he could "forgot what they did" 4) How much of a gain in performance you would get for these tweaks.

Really in all honesty, has anyone bench marked any of these "tweaks"? What were the results? Just for ****s and giggles I will format my empty partition with a clean fully updated XP Pro and bench mark it with 3dMark and PCMark05 and then do every tweak here and post the results as soon as I am done. My system is an HP a6010n - I've added an BFG8800GT 512, upped the ram to 2 gigs and added an Antec NeoPower 550 HE power supply. Hopefully I can get this done by Monday or so. See you.

#1 - I think I explained this properly in the guide and enforced that it should be disabled at the Power User's discretion - or no?

#2 - I am running without it and all is good, once again at the Power User's discretion - some people might be virus prone malware magnets so they might need security - last time I clicked a malicious executable (which gave me services, startup entries, drivers, and dll files to bork my system) - I just manually removed it.

#3 - That has been fixed before your post - and actually will be updated again after I complete this post :)

#4 - Which Tweaks? Services? Registry? Drivers? Latest Service Pack? Also it all depends on your PC, so a Pentium 3 computer would receive higher performance boost while a Core 2 might not. Testing the minimum system requirements setup might be a good idea.

I can't seem to edit my post above - maybe a time limit on editing or something, anyway PCMarK05 and 3DMark05 are constantly crashing and I can't figure out why. They both start up and start benchmarking, but then they just stop and say testing aborted?

I will have to look for another benchmarking tool, does anyone have a suggestion. I also decided not to reformat my current XP Pro partition, but have used BlackVipers XP Pro Default service setting reg file to reset all the service to their default just to save myself some time.

I recommend reinstalling XP if you XP is borked (as it seems that it is) then installing latest XP updates and drivers then benchmarking it. Then applying service tweaks, GUI speed tweaks, PF tweaks, Registry Tweaks, whatever and then benchmarking it again and finally comparing XP vs. Tweaked XP.

You also need a programme that can test GUI speed and explorer.exe browsing / response time speed as many registry tweaks concentrate on those areas.

Good luck!

Well, here are the results. Bear in mind I read each test one right after another and only ran them three times. Your mileage may vary.

PCMarks05

BlackVipers default services reg file. No tweaks............................1) 5606 2) 5665 3) 5648 Average - 5639

After all tweaks in this guide and every reg file applied.................1) 5657 2) 5632 3) 5645 Average - 5644

Vista PCMark05 Fresh HP restore disk with all updates.................1) 5745 2) 5764 3) 5765 Average - 5758

PerformanceTest 07

before tweaks................1) 697.2 2) 697.9 3) 698.4 Average - 697.8

after tweaks...................1) 698.8 2) 699.4 3) 700.3 Average - 699.5

Vista...............................1) 686.8 2) 700.0 3) 704.2 Average - 697.0

Boot time

before tweaks...................1) 1:02.3 2) 59.2 3) 1:00.5 Average - 1:00.6

after tweaks...................1) 1:05.2 2) 1:05.1 3) 1:05.2 Average - 1:05.1

Vista ...............................................1) 1:01.7 2) 1:01.3 3) 100.7 Average - 1:01.3

So after tweaks my XP boot got slower by an average of 5 seconds, but PCMark05 show a gain of +5 or increase of 0.08% and Performance Test show an increase of 1.7 or an increase of 0.24%. I also made the PC less secure by turning of Automatic Updates and not having a Installed Anti-Virus.

It's also worth noting that Vista booted faster than the tweaked XP, but slower than the non-tweaked XP. PCMark05 also showed Vista faster than the tweaked XP by 1.98%, however Performance Test showed Vista to be slower, probably due to the DX10 rendering test not included in XP scores.

I guess these tweaks could actually make XP faster on old hardware, but the recommended hardware is a 300mHz cpu and 128 megs of ram and most netbooks are faster than that. My results are on more modern hardware and the tweaks just don't seem worth while, but hey it's your computer. :)

All in all, if you want to tweak your XP don't do it with reg hacks go get TweakUI from Microsoft and do them safer and with the ability to reverse them if you decide that they aren't worth it. These "guides" are all over they net and they just rehash the same old tweaks even when they have been proven to make much of a difference. Maybe the should include benchmarks to back up their claims of increased performance, but that's probably not going to happen.

P.S. I tried to get my wife to let me use her P4 1.6gHz 2gig or ram VPR Matrix to test these tweaks, but when she saw the resulting loss of functionality she said and I quote "Hell No!" Gotta love her, I didn't think my years of preaching safety on the net got to her. The big thing was the loss of commonly used apps and recent documents in the start menu, oh well can't win them all. ;)

*snip*

Thanks for contributing! :D

The difference between our experiences in tweaking is rather grand...

Few things I noticed,

- I noticed a one minute boot time... I find that rather surprising for your high-end rig (that seriously is slow). Did you config your BIOS for a fast boot?

- For a fair comparison, did you use the same driver versions and did you disable prefetch/superprefetch features in Vista before benchmarking? I am assuming that you did use a fresh copy of XP. I am assuming that all tests also were conducted on a clean boot without any 3rd party services either, right? Same drivers enabled / disabled? Did you also skip the username / password login on both systems? Which glitter elements of the GUI did you disable? etc...

- Which OS components have you removed with nlite? Which startup components have you disabled with jv Power Tools? Did you disable WFP and patch TCP/IP?

- Do I still need to make it more clear for the reader on how to reverse the effects of the registry tweaks? This is important actually, if the guide is not as clear as it should be, I would need to make it so.

- Did you use the tweaks provided when they were still in the guide or when they already were in a .zip file? I significantly revised that part.

- Also the main score in PerformanceTest does not tell anything, I would appreciate if you could post detailed results.

- Also you did not specify RAM usage before and after, which is very important.

- How come no 3DMark 2006+?

Thanks for contributing! :D

The difference between our experiences in tweaking is rather grand...

Few things I noticed,

- I noticed a one minute boot time... I find that rather surprising for your high-end rig (that seriously is slow). Did you config your BIOS for a fast boot?

Yes, fast boot is enabled, boot time includes password input. Calling an HP a6010n high-end is a bit of a stretch. ;)
For a fair comparison, did you use the same driver versions and did you disable prefetch/superprefetch features in Vista before benchmarking?
Yes, well as much as possible, two different versions of the OS can't have every driver version the same. No, Vista was a clean HP Vista SP1 that came with the PC. I updated it and stopped all the OEM services and removed them with Revo Uninstaller. I wanted Vista to be just like i got it on my PC without all the OEM stuff.
I am assuming that you did use a fresh copy of XP. I am assuming that all tests also were conducted on a clean boot without any 3rd party services either, right?
I tweaked my copy of XP Pro that I had been using for about 2 weeks. Comodo Internet Security was running on all three tests and on all OS versions.
Same drivers enabled / disabled? Did you also skip the username / password login on both systems? Which glitter elements of the GUI did you disable? etc...
Yes, no changes were made to XP except the tweaks. No, I didn't skip login / password. Themes disabled and "best performance" checked on XP, Aero was running on Vista.
Which OS components have you removed with nlite? Which startup components have you disabled with jv Power Tools?
No to Nlite. I disabled all 3rd party start up entries except CIS, but used Autoruns to check and stop for services. Then rebooted and checked again to ensure the service was stopped.
Did you disable WFP and patch TCP/IP?
Yes, and No.
Do I still need to make it more clear for the reader on how to reverse the effects of the registry tweaks? This is important actually, if the guide is not as clear as it should be, I would need to make it so. Did you use the tweaks provided when they were still in the guide or when they already were in a .zip file? I significantly revised that part.
Yes, providing a removal/reversal would be a good idea and I used the Zip file.
Also the main score in PerformanceTest does not tell anything, I would appreciate if you could post detailed results. Also you did not specify RAM usage before and after, which is very important. How come no 3DMark 2006+?
Sorry, XP and the results of the tests were lost when Vista was re-installed. Ram usage was dropped by about 3-5%, I can't remember the exact numbers or find my paper. 3DMark is for gaming and if you are running a 300mHz cpu with 128 megs of ram, minesweeper and solitaire are the only games you can play. ;)
This topic is now closed to further replies.
  • Recently Browsing   0 members

    • No registered users viewing this page.