Don't mean to nitpick too much, but did you mean to say that this test was CPU limited? Flat FPS lines at varying resolutions would seem to indicate to me that the CPU was the bottleneck and not the GPU.
Slackware might work, but I don't see how it's any more representative of Linux (let alone desktop Linux) than Ubuntu is.
Slackware's kernel, drivers and packages are mainline vanilla ones without any Slackware-specific patches. Plus, the kernel configuration options are tuned for performance. Ubuntu is heavily patched. I'm not saying that it's a bad thing or that Ubuntu is a bad OS. I'm just saying that it's not truely Linux. And run the command "zcat /proc/config.gz" on both Ubuntu and Slackware. You'll find that there's lot more bloat in Ubuntu's kernel than Slackware's.
The article is only a Ubuntu vs Windows comparison, not Linux vs Windows. Ubuntu does not represent Linux. If you want to compare performance of Windows and Linux, you should choose Linux From Scratch (LFS) or Slackware.
Linux is what Linux users use, and like it's mentioned in the article most Linux installations are Ubuntu.
If you are so eager to criticise, find time to carry out these tests on your own on either Gentoo or LFS.
But then we will argue vehemently about the GCC flags which have to be used in order to compile those two distros.
Ubuntu is an established Linux distro. If you don't understand it, or don't like this fact, please, do not comment here and please, refrain from reading Phoronix articles.
Despite Linux graphical drivers have improved a lot in last years, Windows graphical performance is still the stronget aspect of Windows platform. Maybe that's because this aspect has been worked for years in Windows and Linux graphical system and drivers have to walk this way as well. Hopefully in two years we can talk about the opposite result! :-)
I thought the conclusion was that Windows and Ubuntu were pretty much the same, with some wins for Windows and some wins for Ubuntu. Michael actually says so specifically. On the last page, he quotes a myth that has been busted. Maybe he should have explicitly said "This is clearly false; the performance is pretty much the same."?
I'm actually encouraged by these tests. Ubuntu actually was ahead of Win7 in a few of the cases and generally was not that far behind, Intel graphics notwithstanding, but if you have an Intel graphics chip you're not going to be doing any serious gaming anyway. In many of the tests Ubuntu was ahead at low res settings where it was more CPU driven then fell behind at high res settings. This tells me that regardless of how far nVidia and ATI have come with their Linux drivers, they are still inferior to their Windows counterparts, which is something I think most of us knew already.
Windows is still a lot better for desktop, where latency and GUI response matters. There's is no good DE today, none of them are optimized or are memory efficient. They're are slow, featureless, and they suck.