Announcement

Collapse
No announcement yet.

AMD vs NVIDIA drivers, that big difference?

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

  • AMD vs NVIDIA drivers, that big difference?

    I've been using AMD/ATI HD4870 on Kubuntu x64 for a while now, and then I decided to try out (borrowed) NVIDIA GTX 260 - which is a slightly worse card (on Windows). I also tried NVIDIA 8600 GT which is a quite crappy card.

    The results were quite sad, and I'm considering that I failed (Multiple times? Seems odd.) with the AMD/ATI drivers. Please note that glxgears isn't much of a benchmark, but it does show me something.

    fgl_glxgears
    5500 FPS NVIDIA GTX 260
    2100 FPS NVIDIA 8600 GT
    2200 FPS ATI HD4870

    glxgears
    110k frames / 5 seconds - GTX 260
    40k frames / 5 seconds - HD4870 and 8600 GT


    Here sure is a big difference.

    I also tried Wine World of Warcraft, which resulted in about 2-3 times more FPS on GTX 260. And the FPS on 8600 GT was as good or better than HD4870.
    I did try a small particle OpenGL program I wrote, which resulted in about two times more particles at higher FPS - on both NVIDIA cards (CPU bound) compared to HD4870.

    Do anyone else have any experience using ATI and NVIDIA? It feels like the performance on ATI should be better than this - or is this a common problem?
    I'm right now considering buying a NVIDIA Fermi card (as soon as Coolbits is working, and ETA yet?).

    NVIDIA Drivers:
    256.53
    ATI Drivers:
    10.8, 10.7, 10.5.

  • #2
    ignore ANY number glxgears ever says. It essentially benchmarks page-flips, something that shouldn't be done more than 60 times per second anyway.

    Wine is unfortunately developed on nvidia cards and tweaked accordingly. If you're using a lot of wine for 3d apps, you may be happier with nvidia.

    If you want real numbers, use a native real-world application. Not glxgears, not wine, not simple hand-written openGL programs. Try nexuiz or doom3 or something.

    In my experience, both ATI and nvidia are close to the 3d performance of their windows counterpart on desktop cards. On my notebook, the nvidia drivers just suck and fall way behind their windows counterpart for some reason.

    Comment


    • #3
      I'll run Unigine Heaven on all cards.

      Comment


      • #4
        Well you were sort of right, the performance in Unigine Heaven did give some more sense making results.

        NVIDIA GTX 260
        FPS:
        36.1
        Scores:
        910
        Min FPS:
        19.6
        Max FPS:
        72.8

        ATI HD 4870
        FPS:
        38.0
        Scores:
        957
        Min FPS:
        12.9
        Max FPS:
        78.5

        Unigine Heaven settings:
        Code:
        export LD_LIBRARY_PATH=./bin:$LD_LIBRARY_PATH
        ./bin/Heaven_x64    -video_app opengl \
                            -sound_app openal \
                            -extern_define RELEASE \
                            -system_script heaven/unigine.cpp \
                            -engine_config ../data/heaven_2.1.cfg \
                            -console_command "gl_render_use_arb_tessellation_shader 0 && render_restart" \
                            -data_path ../ \
                            -video_fullscreen 1 \
                            -video_mode -1 \
                            -video_width 1680 \
                            -video_height 1050
        I'm about to do some more testing on Wine now.

        Comment


        • #5
          (Apparently I can't edit my last post(s)). Well it's hard to tell, it seems like it is a bit worse FPS in World of Warcraft (still the same in glxgears, but hey, Unigine proved everyone else's theory - glxgears sucks).
          The question now is if I'd gain a bit more FPS if I bought a new ATI card later on. I'll see about that, but until then I'll not try to buy a new NVIDIA card (unless I find 460 cheap).

          Comment


          • #6
            glxgears doesn't "suck". What sucks is that people try using it for a purpose it is not made for.

            glxgears is a simple demonstration/test that proves that a card is "working". It is NOT A BENCHMARK and should not be used as one.

            Comment


            • #7
              Originally posted by Glaucous View Post
              (Apparently I can't edit my last post(s)). Well it's hard to tell, it seems like it is a bit worse FPS in World of Warcraft (still the same in glxgears, but hey, Unigine proved everyone else's theory - glxgears sucks).
              The question now is if I'd gain a bit more FPS if I bought a new ATI card later on. I'll see about that, but until then I'll not try to buy a new NVIDIA card (unless I find 460 cheap).
              Well, if you are interested in the new Unigine games coming out, you might want to go for a Radeon HD 5xxx series card. Many people don't know this, but ATI has had hardware tesselation since the HD 2xxx series, and it has been speced in OpenGL since the early 2000s. Anyway though, the 5xxx series has MUCH faster tesselation. I have a Radeon HD 5750 (way overclocked though) and I have been very happy with the performance.

              Comment


              • #8
                Also either the distro you guys run glxgears on is broken or both nvidia and fglrx are broken since glxgears in the sane mode is synced to refresh rate.

                Comment


                • #9
                  Originally posted by nanonyme View Post
                  Also either the distro you guys run glxgears on is broken or both nvidia and fglrx are broken since glxgears in the sane mode is synced to refresh rate.
                  I've never seen glxgears synced to anything. If yours is, then yours is the oddity.

                  Comment


                  • #10
                    Originally posted by droidhacker View Post
                    I've never seen glxgears synced to anything. If yours is, then yours is the oddity.
                    afaik the only distro I saw or experienced this on was Ubuntu

                    Comment


                    • #11
                      Originally posted by Glaucous View Post
                      Well you were sort of right, the performance in Unigine Heaven did give some more sense making results.

                      NVIDIA GTX 260
                      FPS:
                      36.1
                      Scores:
                      910
                      Min FPS:
                      19.6
                      Max FPS:
                      72.8

                      ATI HD 4870
                      FPS:
                      38.0
                      Scores:
                      957
                      Min FPS:
                      12.9
                      Max FPS:
                      78.5

                      Unigine Heaven settings:
                      Code:
                      export LD_LIBRARY_PATH=./bin:$LD_LIBRARY_PATH
                      ./bin/Heaven_x64    -video_app opengl \
                                          -sound_app openal \
                                          -extern_define RELEASE \
                                          -system_script heaven/unigine.cpp \
                                          -engine_config ../data/heaven_2.1.cfg \
                                          -console_command "gl_render_use_arb_tessellation_shader 0 && render_restart" \
                                          -data_path ../ \
                                          -video_fullscreen 1 \
                                          -video_mode -1 \
                                          -video_width 1680 \
                                          -video_height 1050
                      I'm about to do some more testing on Wine now.

                      Actually, I ran a few tests to see about performance on linux with a radeon 5770... here's the script i used to launch the program:

                      Code:
                      #!/bin/sh
                      
                      export LD_LIBRARY_PATH=./bin:$LD_LIBRARY_PATH
                      ./bin/Heaven_x64	-video_app opengl \
                      					-sound_app openal \
                      					-extern_define RELEASE \
                      					-system_script heaven/unigine.cpp \
                      					-engine_config ../data/heaven_2.1.cfg \
                      					-console_command "gl_render_use_arb_tessellation_shader 1 && render_hdr 8 && render_srgb 1 && render_restart" \
                      					-data_path ../ \
                      					-video_fullscreen 1 \
                      					-video_mode -1 \
                      					-video_width 1920 \
                      					-video_height 1200\
                      					-video_multisample 0
                      Note: Change video_multisample param is to control Anti-Aliasing... where 0= off, 1=2xAA, 2=4xAA, 3=8xAA.

                      Test Run with Tessellation off and 0xAA:


                      Test Run with Normal Tessellation and 0xAA to 8xAA:



                      Comment


                      • #12
                        Originally posted by kernelOfTruth View Post
                        afaik the only distro I saw or experienced this on was Ubuntu
                        Same on Fedora.

                        Comment


                        • #13
                          Code:
                          [nanonyme@confusion ~]$ glxgears
                          Running synchronized to the vertical refresh.  The framerate should be
                          approximately the same as the monitor refresh rate.
                          302 frames in 5.0 seconds = 60.202 FPS
                          300 frames in 5.0 seconds = 59.992 FPS
                          301 frames in 5.0 seconds = 60.008 FPS
                          What it's supposed to look like if glxgears if "display is synced to vblank" like glxgears.c says.

                          Comment


                          • #14
                            ps. Display is 60Hz

                            Comment


                            • #15
                              @Glaucous

                              remember for the future any benchmark result higher than 60fps or stereoscopic 3D 120fps are invalid benchmarks,

                              you get zero positiv effect if you have 10000000fps in any kind of programm.

                              so you can't benchmark a modern gpu with quake3 or glxgears

                              and in my point of view you can't benchmark a OpenGL4.1 card with a openGL2.1 programm or an openGL1.5 programm.

                              and you can not test your hardware with an benchmark only programme like the unigine tests in the phoronix test

                              only real apps/games can be a basic for an REAL test result

                              in CPU speech its like an SSE1 programm test on an SSE4 cpu you never geht any valid result of the real speed of your hardware...

                              same vor Directx9 on DX10+ hardware.

                              Comment

                              Working...
                              X