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 :)