Announcement

Collapse
No announcement yet.

ATI R500 Gallium3D Performance In June 2010

Collapse
X
  • Filter
  • Time
  • Show
Clear All
new posts

  • ATI R500 Gallium3D Performance In June 2010

    Phoronix: ATI R500 Gallium3D Performance In June 2010

    The past several months have been very exciting in the world of Gallium3D, the new graphics driver architecture for Linux and other operating systems that has been in development for years. This year we have witnessed the emergence of LLVMpipe to accelerate OpenGL commands on the CPU, Nouveau's Gallium3D driver starting to work well, and many other advancements. Over the past few months we have also been pleased with how the "R300g" driver has taken shape with this Gallium3D driver for ATI Radeon R300/400/500 series hardware (up through the Radeon X1000 series) stabilizing, performing well, and advancing beyond the classic Mesa 3D R300 driver. Today we have a fresh set of benchmarks looking at this ATI Gallium3D driver that soon will become the default.

    http://www.phoronix.com/vr.php?view=15000

  • #2
    The suspiciously flat line performance of the Gallium3D drive in most tests makes me think that this driver is still very CPU-bond (software fallbacks?). Can anyone confirm or dismiss this?

    Comment


    • #3
      Could you do the same test with low end R500 or mid end R300/R400 cards?

      I'm always going to 800x600 on my Radeon 9600 to get better FPS in Nexuiz on classic Mesa. I wonder how good/bad FPS I can get with Gallium3D now.

      Comment


      • #4
        As you can imagine probably it would be interesting which commit caused the regression. But maybe it's not so easy to find out with git-biselect because there is so much working at the code...?

        Comment


        • #5
          Those benchmarks are good as they spot obvious potential issues.

          Comment


          • #6
            It would be even more interesting if catalyst 9.3 numbers were show too.

            Comment


            • #7
              Originally posted by Jimbo View Post
              It would be even more interesting if catalyst 9.3 numbers were show too.
              Well, from an open source fan boy, catalyst can go to hell...

              Comment


              • #8
                Originally posted by sylware View Post
                Well, from an open source fan boy, catalyst can go to hell...
                Well, from a fellow open source fan boy, I'd quite like to see the comparision with that last usable fglrx too.

                C'mon, 3D accel is generally the only reason so many people use fglrx in the first place, so a comparision to show how far the open drivers are coming along in a series of 3D benchmarks is relevant...

                Comment


                • #9
                  Catalyst will go to hell much sooner if the OSS stack can match (or approach) its 3D performance.

                  So bring on the comparison, so we can see how far we still have to go!

                  Comment


                  • #10
                    Is this with VSync on or off?

                    Comment

                    Working...
                    X