InfoWorld Tests Show XP Slower Than W2K I was quite surprised to see the results of this test by InfoWorld. It's hard to believe. I'm sure that MS is extremely unhappy with these results, and they commented they were not able to replicate them either. I'm sure many more words will be written about this, but InfoWorld are a bunch of smart cookies and they have done this for over 20 years, I have been reading them for that long. A lot of my industry knowledge comes from this mag. It's not like the first bunch of rookies tells us that WinXP is actually slower than W2K. If it's true, it's a black eye for some one for sure. As per "fair use" I'm copying two paragraphs of the InfoWorld article, and then I'll send you to the actual full article. This is pretty amazing. When I received the email with this news I said to myself "WHOA NELLIE !!" Here goes: "HOPELESS OPTIMISM must be a fundamental part of human nature, because we want to believe that new operating systems truly represent an improvement on their predecessors. It's easy to point to certain features in a new OS as examples of progress, but end-users often find that a new OS performs like molasses compared to the version they were using. As a result, CTOs wanting to capitalize on the benefits of a new OS may find that new hardware investments are necessary -- and expensive -- requirements. "Unfortunately, Microsoft's Windows XP appears to be maintaining that tradition, as indicated by results of independent testing performed by CSA Research and confirmed by our work in the InfoWorld Test Center. Our tests of the multitasking capabilities of Windows XP and Windows 2000 demonstrated that under the same heavy load on identical hardware, Windows 2000 significantly outperformed Windows XP. In the most extreme scenario, our Windows XP system took nearly twice as long to complete a workload as did the Windows 2000 client. Our testing also suggests that companies determined to deploy Windows XP should consider ordering desktop systems with dual CPUs to get the most out of the new OS." Here is the article. This is a 'must read' my friends: And I'll keep you up to date regarding the inevitable sequels of this saga. http://www.w2knews.com/rd/rd.cfm?id=110101-WaitingForXP

"In fact, until 2GHz desktop PCs become commonplace, we have a hard time recommending widespread adoption of Windows XP at all"

The line that got me.

Having just imported Win 2000 Pro from the UK I am sitting here with a self-satisfied feeling. Microsoft ordered the withdrawal of all pre-XP Operating systems from the shops here in Ireland last week to "encourage sales" of XP.

If the tests are sound it seems that XP might have to be in a one-horse-race to compete and MS probably know that.

"Microsoft claims it's been unable to duplicate our results, but hasn't supplied us with a better explanation or identified a major flaw in our testing."

Not surprising, if they got similar results they are hardly going to confirm XP's shortcomings during the current publicity blitz. MS must hate this.

Great post jamball77, hopefully it will save people from jumping too quickly.

STORM

There has been a lot of talk about this test and the speed issues it purports to show about Windows XP. There is a serious problem though - this benchmark ("OfficeBench") isn't based on any real world performance criteria. it is simply a set of scripts that try and drive the office applications through a number of tests designed (in theory) to stress test the system they are on.

The problem is that the results are very dependant on the interaction of several complex pieces of software being used in a way that they were not, in fact, intended to be used. To try and extrapolate that to a overall picture of the comparative merits of Operating Systems is - well, silly.

For reference, here are some links:

• An older MS posting from 08/18/2001 on OfficeBench results with XP.
• An interesting AnandTech article on how their users like WindowsXP so far.
• The InfoWorld article.
• Discussions on the InforWorld  article. I find it interesting how strong the presence of a rep from the testing software company is. The tests are a commercial product BTW and obviously stand to benefit if companies feel compelled to purchase the test product to verify the results.

All I can tell you is what we have found here - WinXP responds faster than Win2000 - especially under load. it just feels faster. It might be slower for a script flashing through office applications at some inhuman speed ... but for me when I am using it... it's faster.

Oh.. and Poser works much faster :)

I don't understand the surprise. They only tripled the size of the code and added lots more fancy doddads. Sort of like trying to upgrade the performance of a car by giving it a new paint job and putting 2000 pounds of lead in the trunk. "Trust us, it will work better." Kevin

Question: Was it Windows XP Home Edition, or Pro? if its pro, there have been bench marks that have it with nearly 30% higher performance rating than Windows 2000. I never believe the results of just ONE test. For me, you have to provide several different tests along with hard core evidence. I want to see it for myself. I'm on Win2k and I wouldn't go to any other OS. Its stable, easy to work with, no stupid ALT(Command on a mac) things to learn like Linux and Mac, its decent looking, doesn't look like a hippy puked on it, and it runs all of my apps at the same time without a problem. All this on an AMD 1.0ghz T-Bird that is OC'd to 1.25 . Go fig.

If the new XP can make a old P II 300 run like rocket then Intel and the rest of PC maker will be out of job. Micro$oft just doing a favour to its business partners.

I'm waiting for some of the independents like Tom's Hardware to give us the green flag. XP essentially has the same kernel as 2000, that was the whole point with the stability issue. In a way it would be surprising if it were faster. AS Kevin said the added bells and whistles do little to make it aerodynamic. Also the self healing and auto update features that sunk ME are an integral part of XP. the real benchmark is how much ram or CPU can I buy for the $100 or $200 it takes to upgrade.

by the way most all the first comments were stuff I just cut and pasted, can't take the credit, just thought it was on target.

Anand tech ran the same tests a little bit ago and found the same thing! Win2000 was faster on thier tests also. Right now the latest article they have posted is a satisfaction only article. It's worth looking into thier archive to find the TEST between XP and Win2000. It's an eye opener.

yeah only on Office software. Its over all performance which is what everyone is concerned about is a lot faster than Win 2k

Anandtech also ran the same "Officebench" test software - a packadge that has a spotty record for measuring anything real as far as user interactive tasks go. I know that XP lets me use max and Poser and Photoshop a lot faster than Win2K did - but I think that is subjective as they tweaked the VM system to be more responsiv. But >I< work faster , and that's all I care about :)