Announcement

Collapse
No announcement yet.

A Big Comparison Of The AMD Catalyst, Mesa & Gallium3D Drive

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

  • A Big Comparison Of The AMD Catalyst, Mesa & Gallium3D Drive

    Phoronix: A Big Comparison Of The AMD Catalyst, Mesa & Gallium3D Drive

    As was alluded to in our New Year greeting, we have been working on a massive graphics card / driver comparison under Linux. Beginning with ATI/AMD hardware, we have tested a series of graphics cards spanning the Radeon X1000, HD 2000, HD 3000, HD 4000, and HD 5000 generations using the very latest drivers. These drivers include the official Catalyst 10.12 Linux release as well as the very latest development code for the open-source Mesa and Gallium3D drivers. The results for seven ATI GPUs spanning four generations with three drivers are quite interesting.

    http://www.phoronix.com/vr.php?view=15604

  • #2
    Interesting to note that the older r500 card outperformed the newer competition in most of the tests. Marek has been doing a ton of work on r300g, and it looks like it's paying off. Thank you, Marek, and everyone else working on the driver. r600g is coming along much more slowly...

    It would have been nice to run Unigine and our other more demanding graphics tests in this large comparison, but they are simply incompatible with the open-source drivers at this point and for the near future.
    Note that i have seen (working) screenshots of Unigine Sanctuary on a r500 card. I think all it needs it the floating texture support which is present in one of Marek's private branches. There's been some talk lately of merging that into master behind a enable-patented-extensions configure switch. It would be really nice to have a test that is hitting the shaders hard rather than just calling old GL 1 functions. Something like Doom3, or another proprietary game could perhaps work until the Unigine demos are going?

    Comment


    • #3
      There is a mistake in page 6, where it says:
      the Tremulous game did slightly better with the classic Mesa driver than Catalyst
      where you probably meant to say
      the Tremulous game did slightly better with the classic Mesa driver than Gallium3D

      Now, about the benchmarks, I started reading Phoronix to track the features of the drivers of my card, a Radeon X1950 whose performance had always been abysmal on Linux, even when Catalyst still worked. Over the years the behavior shifted from abysmal to somewhat decent, and in the meanwhile I gained a new appreciation for open source drivers.

      At some point I contemplated about getting a new card, but then thought why would I want to do that? My graphics card is probably the best performing open-source-powered piece of graphics hardware in the world.
      This article only serves to demonstrate that the above sentiment is still accurate.

      Comment


      • #4
        Aside from VDrift (probably a bug in Catalyst drivers), open source drivers have abysmal performance. Just enough to run compiz.

        Comment


        • #5
          It would be nice if you establish frontpage button "Catalyst/G3D weekly comparison". Take one extreme case of Catalyst/G3D difference (like Lightsmark, there is 430fps / 4fps in one case) and then do wekly update of this graph. Same for NVIDIA.

          Comment


          • #6
            How come the open source drivers perform so poorly in comparison to Catalyst? I mean, obviously they are in development still and so lower performance is expected. But is there a concrete reason/missing feature which causes them to be so far behind?

            Comment


            • #7
              Originally posted by birdie View Post
              Aside from VDrift (probably a bug in Catalyst drivers), open source drivers have abysmal performance (probably a bug in the gallium drivers). Just enough to run compiz.
              fixed that for you o_O
              actually im pretty impressed with some of the benchmarks.
              1/4 of the catalyst performance is pretty good i think.
              and afaik they havent really started doing any optimizations on the cards yet.
              would have loved to see how the old catalyst drivers for r300 performed.
              but thanks for the benchmarks michael!

              Comment


              • #8
                Originally posted by Kazade View Post
                How come the open source drivers perform so poorly in comparison to Catalyst? I mean, obviously they are in development still and so lower performance is expected. But is there a concrete reason/missing feature which causes them to be so far behind?
                Check out the radeon feature matrix - all drivers, all incomplete, all missing the vast majority of 3D related features. Missing as in "To do" - nobody is even working on it. Now that is what I call AMD support.

                Comment


                • #9
                  I bought an ATI 4870 in the hope of having nice open-source stack in few months, when ATI opened-up 3 years ago.
                  I want to play recent games like Quake Wars, Heroes of Newerth or any Unigine based game when it comes out, so it excludes the possibility of getting open-source right now.
                  fglrx is working fine on my gentoo, but I'm still being angry the video tearing, which is just enormous. And that video tearing is just left as garbage by ATI, I don't wonder why, as I don't think this should be so complicated to correct if the only wanted to.
                  Nvidia is able to deal with tearing, why not ATI ?

                  Anyway, for all that reasons, I think i'll go back to nvidia on my next video card, as their blobs that I have to use anyway on ATI are better as they don't tear video. I have been using nvidia blobs for years before my 4870 and I didn't noticed any particular problem, so I don't mind reverting to nvidia.

                  Comment


                  • #10
                    The 11.1 drivers (coming out this month) are supposed to fix the tearing issue.

                    Comment


                    • #11
                      Originally posted by Kazade View Post
                      How come the open source drivers perform so poorly in comparison to Catalyst? I mean, obviously they are in development still and so lower performance is expected. But is there a concrete reason/missing feature which causes them to be so far behind?
                      developers developers developers

                      Comment


                      • #12
                        Well, talking about the results, I'm satisfied that even classic Mesa / Gallium3D OS ATI drivers can run most standard OpenGL games on Linux , (although their frame rates are almost an order of magnitude slower than Catalyst). With some time, I hope (maybe Gallium3D) Mesa drivers can run at best, about 50% the performance of 3D proprietary Catalyst. But, until then, I'll stick with Catalyst (I need some fast 3D perfomance ATM).

                        And... there's also another point that wasn't tested here (and I'd liked to see there): Currently, if you use Wine, you're almost out of luck if you use any of the OSS ATI driver stacks. Some games need SM3.0 and that's only avaiable with OpenGL3 or higher... (currently, only Catalyst supports it. )

                        My 2c, cheers

                        Comment


                        • #13
                          Originally posted by BlackStar View Post
                          The 11.1 drivers (coming out this month) are supposed to fix the tearing issue.
                          I hope they also fix the UVD1/UVD1+ H264/VC-1 problems in r600 generation cards (HD2xxx/HD3xxx) on both Windows / Linux OS's.

                          Cheers

                          Comment


                          • #14
                            the open source 3D drivers just don't cut it. the gap is huuuuuuuuuge.
                            one may wonder what is the difference between the drivers. That's not just "poorly" optimized drivers, some things must be plain missing (and probably kept secret by ATI)

                            Comment


                            • #15
                              This article is just sad The drivers for any recent card suck and it's 2011 by now.

                              Maybe by 2018 we get an r600 that's somewhat decent?

                              Comment

                              Working...
                              X