Windows XP SP3 Shows Noticeable Speed Increase:

[RJM]

http://technologyexpert.blogspot.com/2007/...d-increase.html

[ripken204]

hmm, intersting. too bad im using vista tho. but this will help a great deal of people. i want to see some real world results tho.

[Glenn9999]

IMO, this is a better link - it's what the original link is referring to:

http://exo-blog.blogspot.com/2007/11/windo...ance-gains.html

or http://exo-blog.blogspot.com/2007/11/updat...ram-office.html

The testing tool used on that page is DMS Clarity Studio, downloadable at http://www.xpnet.com if you wish to test it for yourself.

Edited November 28, 2007 by Glenn9999

[ripken204]

i'm talking about real benchmarks, not just pointless numbers. like who cares about the performance in MS Word? lol.

[PC_LOAD_LETTER]

They have not included enough data about the tests to be making these kinds of claims.

is the SP2 machine running SP2 and no other patches (SP2 PURE) or is it running SP2 + all publicly available updates?

If its SP2 pure then I would expect these type of results seeing as SP2 is over 3 years old and none of that hardware is. If its SP2 and all available updates then you may consider me impressed.

Don't get me wrong, I've been fighting the urge to install SP3 beta on every machine Ive built this month just to avoid the Microsoft Update Download-O-rama I have to do even with my Slipstreamed SP2+ CD. I can hardly wait for SP3 to go public

[ripken204]

They have not included enough data about the tests to be making these kinds of claims.

is the SP2 machine running SP2 and no other patches (SP2 PURE) or is it running SP2 + all publicly available updates?

If its SP2 pure then I would expect these type of results seeing as SP2 is over 3 years old and none of that hardware is. If its SP2 and all available updates then you may consider me impressed.

Don't get me wrong, I've been fighting the urge to install SP3 beta on every machine Ive built this month just to avoid the Microsoft Update Download-O-rama I have to do even with my Slipstreamed SP2+ CD. I can hardly wait for SP3 to go public

i didnt even think about it that way. i bet you they are using a pure SP2

[cluberti]

It's really bad to have one benchmark test and then say "x is so many percent faster than y" - yeah, OK, maybe. Does anyone remember the same types of claims and benchmarks about Win98 vs Win2K, or better yet, Win2K vs WinXP? Back in 2001, XP was the worst thing there was, and now that the hardware's caught up, it's the best thing since sliced bread.

Whatever - it's about time a Windows OS with good security and forced secure coding practices (as much as is possible) is finally available. And to everyone who says "ditch Vista and buy XP", you still just bought another Windows license .

[Zxian]

Whatever - it's about time a Windows OS with good security and forced secure coding practices (as much as is possible) is finally available. And to everyone who says "ditch Vista and buy XP", you still just bought another Windows license .

Everytime people say that Windows is less secure than "the rest", I point them to the article that showed less than 30 critical updates in the first year of Vista's release. No other major operating system has matched that yet.

I'll be moving to Vista when SP1 comes out - I'd just rather wait until I can slipstream the installation source.

[bj-kaiser]

... Everytime people say that Windows is less secure than "the rest", I point them to the article that showed less than 30 critical updates in the first year of Vista's release. No other major operating system has matched that yet. ...

http://www.openbsd.org

Only two remote holes in the default install, in more than 10 years!

BTW: And then there is the question how many holes have been patched with those 30 updates?

[Zxian]

http://blogs.zdnet.com/Bott/?p=320

That's the original article I was referring to. OpenBSD is a good operating system (I've tested it as a webserver base), but it's nowhere near as big of a target as Vista is. Remember how people used to say that Firefox was "more secure" than IE? There's another article showing that over the past year, IE7 has had fewer security holes than Firefox. If OpenBSD were to show enough market share to make a difference, I'm sure that the number 2 would quickly grow. I guess my definition of "major operating system" was a little unclear - something that you might actually hear people talk about in passing reference to computers. OpenBSD is still one of those "specialty" OSes.

[cluberti]

http://www.openbsd.org

Only two remote holes in the default install, in more than 10 years!

BTW: And then there is the question how many holes have been patched with those 30 updates?

I love OpenBSD, but I'd never consider it as my primary desktop OS . As to how many holes, probably around 30. Windows is a beast, number of lines of code-wise, and hotfixes need to be tested extensively before public release - so code changes need to be as small as possible (meaning generally a 1:1 fix rate to hotfix). How many still exist to be found is probably quite large, I'm sure, but all software written by humans will have bugs. Some avoidable, maybe, and some that you just don't see coming at all.

[PC_LOAD_LETTER]

Ok getting back on topic:

Meet SP2, SP2point9, and SP3

All machines are 3.2 Ghz P4s (dont have any dual cores in the office ATM) with 1GB Ram with Intel onboard video

the OSes were installed as follows:

XP RTM installed

Drivers and Office 2003 & 2007 installed

Microsoft update run on the SP2 Patched Machine and the SP3 machine with all updates (including optionals) -SP2 PURE machine received MS installer 3.1, WGA, and all office updates.

SP3 Machine updated to SP3 via Microsoft Update

Auto Updates & Security Center services disabled (no AV installed and didnt want the nag to interfere with benchmarks)

I didnt bother with 3dmark since the video card in these machines will barely even play WOW but i did run Aquamark:

I was initially gonna run 3 test sets but ended up running 5

the results were as follows: (placed in code tags since i cant do tables here)

	 SP2 PURE		|	 SP2 Patched	 |	 SP3 v3244
--------------------------------------------------------------------
	  10,011		 |	  10,046		 |	  10,021
	  10,041		 |	  10,029		 |	   9,996
	  10,026		 |	  10,002		 |	   9,919
--------------------------------------------------------------------
rebooted all machines
--------------------------------------------------------------------
	  10,079		 |	  10,031		 |	  10,021
	  10,027		 |	  10,005		 |	  10,031
averages------------------------------------------------------------
	  10036.8		|	  10022.6		|	  9997.6 (10024.3 with below 10000 scores ommited)

I rebooted all the machines between tests 3 & 4 because SP3 wasn't doing well at all

after Aquamark tests, the machines were all rebooted and sit idle for 15 minutes before starting office bench

First SP3s results:

looks to average around 57

SP2 Patched results:

averages slightly lower at around 56

SP2 PURE results

did that every single time i ran it. i may rebuild this machine using SP2 slipstream install and see if officebench runs at all on SP2 PURE

I may have to round up some dualcores for a test too but until then what do you guys think of these "results"?

Edited December 1, 2007 by geek