Is XP performance worse than Win2k, or just the same?
Latest test shows marginal gains at best
A pretty comprehensive set of benchmarks over at The Tech Report comes up with the possibly less than earth-shattering conclusion that WinXP is pretty much neck and neck with Win2k, from a performance point of view. XP seems to be slightly better than Win2k when it comes to office productivity, but is pretty well tied with it on graphics and gaming, in some cases even marginally losing.
Tech Report also threw in a WinME configuration for the tests, and this - also unsurprisingly - revealed itself as a comparative dog. But although you could think of the results as being predictable, given that WinXP is a development of Win2k, they're useful because they don't confirm an earlier benchmarking session, conducted for Infoworld by CSA Research, that showed XP seriously underperforming Win2k machines of the same configuration. That test concluded that XP got slower the more the load increased, and suggested it mightn't be right for widespread deployment until 2GHz desktops became commonplace.
It's difficult to see how the two tests came to such markedly different conclusions, but there are arguments in favour of both. You wouldn't expect base performance to be that different, but on the other hand XP's bells and whistles will have a tendency to slow things up. And on the third hand, any optimisations Microsoft has done between the two operating systems would tend to cancel out the performance impact of the candy factor, possibly resulting in a draw.
The Infoworld tests don't say how much RAM was used in the machines and this, as Tech Report suggests, may explain the difference, because XP likes memory a lot. It might also be worth noting that the Infoworld tests used Pentium III and Pentium IV, while Tech Report used AMD Athlon. So it might just be the case that the input AMD had in processor optimisation had an effect. ®
Sponsored: Benefits from the lessons learned in HPC