Announcement

Collapse
No announcement yet.

AMD Catalyst vs. Linux 3.7 + Mesa 9.1-devel Gallium3D Performance

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

  • #11
    Please test unigine heaven
    ## VGA ##
    AMD: X1950XTX, HD3870, HD5870
    Intel: GMA45, HD3000 (Core i5 2500K)

    Comment


    • #12
      Originally posted by przemoli View Post
      So you have some defined test suite for performance? Can you recommend some PTS test ? Are such perf benchmarks beneficial for you? If not those what else can be tested?
      I assess the performance based on Phoronix articles and some games I like to use for testing, I also sometimes run some well-known AAA games under Wine.

      Comment


      • #13
        Progress is progress. My heartiest appreciation goes out to the devs.

        Comment


        • #14
          Run the same tests on the same kernel. I'm betting the proprietary driver stomps heavier. You could run it on Linux 3.6.6 against the 3.7 trunk and see how less impressive the open source driver is but you know that already.

          Comment


          • #15
            Marc Driftmeyer the kernel actually has nothing (much) to do with the AMD proprietary drivers.

            The later kernel is used in testing the OS driver because of the later kernel drivers.

            Looking at the figures most of those games are more than playable on the OS driver.

            Comment


            • #16
              Originally posted by danwood76 View Post
              Marc Driftmeyer the kernel actually has nothing (much) to do with the AMD proprietary drivers.

              The later kernel is used in testing the OS driver because of the later kernel drivers.

              Looking at the figures most of those games are more than playable on the OS driver.
              But he's betting that it stomps. And you know that already.

              How can argue with THAT? Huh?

              Comment


              • #17
                I try to compare Intel vs. AMD here:
                Nexuviz 2.5.2
                AMD 6 times faster than Intel

                OpenArena 0.8.8
                Intel about 5 fps faster than AMD

                Warsow 0.61 and 1.0
                About the same speed

                Xonotic 0.6
                depends
                Intel is 30 times faster than the newer AMD driver and mesa
                About the same speed with the older AMD driver and mesa

                Phoronix, Linux Hardware Reviews, Linux hardware benchmarks, Linux server benchmarks, Linux benchmarking, Desktop Linux, Linux performance, Open Source graphics, Linux How To, Ubuntu benchmarks, Ubuntu hardware, Phoronix Test Suite

                Phoronix, Linux Hardware Reviews, Linux hardware benchmarks, Linux server benchmarks, Linux benchmarking, Desktop Linux, Linux performance, Open Source graphics, Linux How To, Ubuntu benchmarks, Ubuntu hardware, Phoronix Test Suite

                Comment


                • #18
                  I applaud the efforts that have been made with the radeon driver. It came from nothing - to the offer good stability and a good subset of the OpenGL requirements for modern 3D gaming. But last time I tested it on my laptop with a **real** game (S.T.A.L.K.E.R. : SOC via Wine) I got something like 6 FPS (Catalyst gives more like 30 FPS). The lack of full shadows isn't too noticeable - but the lamentable framerate is...

                  Since I have a (very recent!!) legacy 4650M I guess most distros are gradually going to stop supporting it (my ARCH and Gentoo installs might limp on longer than the likes of Ubuntu - which xorg-server updates are the killer).

                  I only have an AMD GPU in my laptop because Nvidia had nothing decent when I bought it. While the Catalyst driver was very slowly getting better over the years - I'm now stuck on some stupid legacy driver (limiting xorg-server and kernel updates).

                  Trouble is I can't help looking at my truly ancient desktop Geforce 8800GTX with the bleeding edge 310.xx Nvidia beta driver - working like a champ and playing Black Mesa Source (via Wine) rather well!

                  Comment


                  • #19
                    Hi,

                    I guess the explanation for such a constant gap between the open source driver and fglrx is one single fundamental design difference between the two drivers, that's causing such an inefficiency on kernel/mesa/xf86-video-ati.

                    If TTM is the culprit, why isn't there any patches already (4 years!) for addressing the buffers migration issue? AFAIK, that's what's Chris W. has done w/ SNA for the Xorg Intel driver.

                    -Ilyes

                    Comment


                    • #20
                      I expect the performance gap is caused by a dozen or more design differences, not one. Just a few off the top of my head :

                      - tiling (mostly done AFAIK)
                      - hyper-z (started)
                      - shader compiler (WIP)
                      - threading (command submission is in separate thread on r300g, not sure about r600g)
                      - memory manager heuristics (this is what would probably help with buffer migration)
                      - adaptive load balancing
                      - adaptive memory reconfiguration

                      BTW I don't think Marek is saying "we've known this for 4 years", I think he's saying "I just looked recently and the problem with these specific applications seems to be buffer migration".

                      If the developers believed that one single issue was responsible for most of the performance differences then I think it's pretty safe to assume they would be all over it, but I don't think that is the case.

                      Some apps (typically the very slowest) may have one single issue that contributes most of the slowdown RELATIVE TO OTHER APPLICATIONS but that's not the same as saying one single issue contributes to most of the performance gap between the open source stack and the Catalyst stack.
                      Test signature

                      Comment

                      Working...
                      X