Benchmarks: Mandriva, PCLinuxOS, Ubuntu, openSUSE
Phoronix: Benchmarks: Mandriva, PCLinuxOS, Ubuntu, openSUSE
Last week we delivered benchmarks of Fedora 13 Alpha and Ubuntu 10.04 (along with testing the Fedora 11 and 12 too), but today we have a new set of comparative benchmarks that are covering the latest development versions of Ubuntu 10.04, Mandriva 2010.1, PCLinuxOS 2010, and openSUSE 11.3. Here they are.
Woow,...I was thinking of trying PCLinuxOS for a long time,I newer imagined that it was that good.
Seriously,BFS killed all distros in these tests.Too bad that it doesn't use all newest packages .
One thing missing from the article is which of the tested distros were running with KMS enabled and which ones weren't. I suspect this is were the performance difference in the OpenArena and Tremulous benchmarks comes from.
64-bit vs 32-bit
As noted in the article, PClinuxOS is not distributed for 64 bits, so the tests used 32 bits distributions.
We've seen here in Phoronix that 64-bits gives better performance than 32-bits all things equal, so if you have a 64 bit machine, and you try the best release from each distro (64 bit for all except PCLinuxOS), the differences would be smaller (and who knows what distro would take the gold, it would be nice to see those tests)
I thought of the same. KMS and in case of UMS screen size matters as well.
Originally Posted by monraaf
If screen size was set bigger than current display resolution, it could cause worse performance (counts to UMS only).
It's very like KMS vs. UMS. Quake3(live) on a X1500 radeon is unplayable with KMS. With UMS i get 125FPS pretty much all the time....
Originally Posted by Zajec
If you boot with radeon.modeset=0 you can disable it, X/mesa should still work.
Would be nice if phoronix actually did a comparison between KMS/UMS, or just did a simple check if mandriva/pclinux has disabled KMS....).
Disappointing that phoronix fails to give any clarification about this, and update the article accordingly. Now some people might attribute the OpenGL performance gains to BFS or to some sort of 'magical' Mandriva patches.
Originally Posted by frej
That's intresting. Why screen size affects only UMS performance and not KMS?
Originally Posted by Zajec
In case of KMS we use memory management and we can allocate memory for screen dynamically. It's not possible in UMS where we do not use memory management.
Originally Posted by mar04
It may happen you allocate memory for bigger size than you actually need (it's for UMS) and it impacts performance making it worse.
Good joke. It doesn't perform better in every test. It usually kills mouse and keyboard input.
Originally Posted by kUrb1a