Announcement

Collapse
No announcement yet.

Open-Source ATI R600/700 Mesa 3D Performance

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

  • Open-Source ATI R600/700 Mesa 3D Performance

    Phoronix: Open-Source ATI R600/700 Mesa 3D Performance

    As we alluded to last week, we have been in the process of benchmarking many Radeon HD 2000/3000/4000 series graphics cards using the open-source ATI Linux graphics stack with the Mesa R600/700 DRI driver. We have now carried out our first batch of R600/700 3D tests using this constantly evolving open-source driver to provide OpenGL acceleration and here are the results.

    http://www.phoronix.com/vr.php?view=14575

  • #2
    That's interesting - I would have expected the HD3650 and HD2600 to be relatively close in performance, but the 3650 seemed to be quite a bit faster across the board. Was there anything "special" about the 2600 - I'm sure you would have mentioned if it was AGP

    Comment


    • #3
      You can obviously see that tremoulus is bottlenecked somewhere else than the graphics performance. The framerates stay the same on most resolutions for most of the cards.
      The results arent much dependent on the performance of the graphics card alone (all 48xx lines are virtually identical).
      This is so obvious, Mike should have commented on this, instead of just saying "Tremoulus is more demanding". Something else is afoot there.

      Comment


      • #4
        Something is definitely fishy with the HD2600 results. The Xv and mesa code is more or less identical for all the r6xx chips and there are no aspects of that code that require more CPU on one asic compared to another. I can see the gpu performance being somewhat variable, but the cpu usage shouldn't vary much.

        Comment


        • #5
          Interesting would be to compare those Result with fglrx results

          Comment


          • #6
            Bad and good surprises

            I thought that my HD3650 would have been nearer to the HD3850... it's almost half the performance!
            It's quite a low level rather than a mid-range! (at its times I mean... nowadays it ain't even worth considering). ok, I bought it as a low budget transition card, but still...

            I've been following Phoronix for quite a long time, but I didn't get that open source drivers are 3D capable in R600/700
            They're even capable of running games. I guess that 3D performance is lower than fglrx... but what about quality comparison? and features?
            Netrunner Linux - Rolling Release ; Nexus 5 ROM Chroma 5.1 ; NAS 6TB on FreeNAS

            Comment


            • #7
              Am I the only one who thinks the graphing of the results is absolutely abhorrent? To get any valuable information out of the mess that is, for example, the GL and XV video performance graphs, they should have been at least two times their size. It'd also be nice to have access to the numeric data in tabular form as an option. More often than not, I'd like to compare numbers instead of trying hard to figure out which shade of $colour represents which card…

              Comment


              • #8
                Wait, you don't like some of the results so you won't show them? What the heck!

                It would be nice to know if things have failed miserably or are not sufficiently different to even see a difference, too, you know: you even already have all the data.

                Comment


                • #9
                  I think the performance of the OpenGL stuff is miserabel, because I have also Mesa 7.7 installed and Nexuiz is only playable with low details and resolution. With fglrx I was able to play 1680X1050 with the highest detail level. I hope Mesa 7.8 will bring good OpenGL performance to R600 cards.

                  Comment


                  • #10
                    Originally posted by bridgman View Post
                    That's interesting - I would have expected the HD3650 and HD2600 to be relatively close in performance, but the 3650 seemed to be quite a bit faster across the board. Was there anything "special" about the 2600 - I'm sure you would have mentioned if it was AGP
                    Of course not, I haven't touched AGP in years. It's just a standard HD 2600PRO.
                    Michael Larabel
                    http://www.michaellarabel.com/

                    Comment


                    • #11
                      Video decoding: GL2
                      Video decoding: X-Video
                      Neither of them does decode the video, they just do colorspace conversation and scaling.

                      Comment


                      • #12
                        Nice test by the way...not to shaby.

                        What's wrong with HD2600PRO in the last test...looks like budget cards are actually made junk that can be slow.

                        Not yet for series 5 support well it's OK...i'm not in gaming nor using Blender so, Nexuiz can even run on my poor G31 almost ok, it's fine with me.

                        Keep up the good work...donations are coming someday pretty soon.
                        Linux is a lifesaver...

                        Comment


                        • #13
                          Originally posted by bolzerrr View Post
                          Interesting would be to compare those Result with fglrx results
                          +1

                          Exactly what I was thinking. It would be nice to know when we are nearing (maybe even exceeding at some point?) the performance of the official drivers.

                          Comment


                          • #14
                            How about Open Source vs Catalyst?

                            Michael,

                            Dou you think you can run some tests for just one given card, comparing performance with the Open Source stack against the Catalyst driver?

                            Here is my own test (I wonder if the factor 5 in 3D performance is representative):
                            http://global.phoronix-test-suite.co...483-2293-23385

                            My test is current Kubuntu Lucid open source, against Catalyst in Karmic.

                            Comment


                            • #15
                              Originally posted by mendieta View Post
                              Michael,

                              Dou you think you can run some tests for just one given card, comparing performance with the Open Source stack against the Catalyst driver?

                              Here is my own test (I wonder if the factor 5 in 3D performance is representative):
                              http://global.phoronix-test-suite.co...483-2293-23385

                              My test is current Kubuntu Lucid open source, against Catalyst in Karmic.
                              When I get back to the office next week I plan to run Catalyst vs. Open-Source on an R600, RV670, and RV770.
                              Michael Larabel
                              http://www.michaellarabel.com/

                              Comment

                              Working...
                              X