Benchmarking Recent Mesa 3D Releases

[phoronix]

Phoronix: Benchmarking Recent Mesa 3D Releases

With Mesa 7.8 arriving this month, we took the time to benchmark a few recent releases of the Mesa 3D stack with the Radeon DRI driver to see how the OpenGL performance has changed -- if at all -- over the past few months. In this article are our R500 Mesa benchmarks from the Mesa 7.6, 7.7, 7.8-rc1, and 7.9-devel releases.

http://www.phoronix.com/vr.php?view=14703

[Flyser]

Whats the point of a benchmark with 3 very simmilar engines?
What about intel mesa performance?

[werdanith]

Since it's not mentioned anywhere in the article, is this UMS or KMS performance?
(My bet is on UMS)

[HokTar]

Probably it would have been more interesting if you tested an r600+ card.
On the other it's nice to see how stable are recent mesa versions.

[FireBurn]

I bet the reason there wasn't much variance is he used the same kernel DRM and libdrm for each test

I bet you would have seen bigger differences if the mesa version had stayed the same and the drm parts had changed

[Michael]

Since it's not mentioned anywhere in the article, is this UMS or KMS performance?
(My bet is on UMS)

KMS, not UMS.

[Michael]

Whats the point of a benchmark with 3 very simmilar engines?

While all based on ioquake3, the different games have proved in the past to perform quite differently. Sure wish more titles could be used, but not too many work well with Mesa and can be benchmarked quite well.

[mendieta]

R600/R700 tests, pretty please !!!!!!!

I think those would show a lot more interesting results!

Cheers, and thanks for the benchmarks!

[V!NCENT]

@Michael:

Could you please re-run the tests with the noveau card and substract that performance from the Radeon benchmarks?

Gallium3D is ofcourse split up into KMS, Mesa and the GPU specific driver.

Subtracting would show the current performance of KMS+Mesa a little and see if there are improvements Gallium-wide.

It would be nice to see if specific drivers would go up/down and Gallium go up/down.

Mixing both doesn't tell us anything because it could be GPU driver specific....