Announcement

Collapse
No announcement yet.

AMD's Catalyst Evolution For The Radeon HD 7000 Series

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

  • AMD's Catalyst Evolution For The Radeon HD 7000 Series

    Phoronix: AMD's Catalyst Evolution For The Radeon HD 7000 Series

    It used to be -- at least when using the Windows Catalyst drivers -- that within the first few months of AMD releasing new Radeon graphics hardware that Catalyst driver optimizations would deliver measurable improvements in this short span. For the Radeon HD 7000 series, which is built upon an entirely new GCN architecture, is this still the case? Here are benchmarks of all the AMD Catalyst Linux drivers that have been released this year and then benchmarked on an AMD Radeon HD 7950 graphics card.

    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

  • #2
    This is why I hate those "the drivers are not mature yet" guys. You can actually judge a card based on its initial showing: that's pretty much the performance you'll get. If there are serious bugs (like in this review), they stand out by themselves.

    Comment


    • #3
      Where can we find fglrx 9.0.0 ?

      Comment


      • #4
        I might be completely wrong here; but wasn't it said GCN is much easier to program for in
        terms of shader compiler optimizations than previous architectures?
        If so, that may explain why performance figures do not differ significantly between driver revisions.

        Comment


        • #5
          Originally posted by entropy View Post
          I might be completely wrong here; but wasn't it said GCN is much easier to program for in
          terms of shader compiler optimizations than previous architectures?
          If so, that may explain why performance figures do not differ significantly between driver revisions.
          Maybe because the tested games are faaaar from demanding for this graphics card.



          I would highly suggest to rethink this kind of tests. >400 fps what in hell tells us such a result? Nothing! I know there not many demanding current games out there. So at least quality enhancing features should be activated by default for such tests. I mean super sampling anti-aliasing and such stuff. The cards must be used to their capacity! In general it should be investigated if there are more demanding OpenGL benchmarks out there available for Linux. WINE is not an option because of the fast development of this project which also my encompass performance changes due to optimizations regarding WINE and not the display driver. Overall its not easy but testing an 8 years old game like Doom 3 doesn't tell us anything. At least there are very demanding graphics mods out there for Doom 3 which may be compatible with the Linux version. Then this would be an option.

          Here is a link to an interesting Doom 3 Mod which enhances the graphics: http://www.moddb.com/mods/cverdzislav

          Comment


          • #6
            Originally posted by bug77 View Post
            This is why I hate those "the drivers are not mature yet" guys. You can actually judge a card based on its initial showing: that's pretty much the performance you'll get. If there are serious bugs (like in this review), they stand out by themselves.
            It's probably much more true on Windows for this card. They have all those per-application optimizations built into their drivers, catalyst A.I. swapping out shaders with more optimized versions, etc. there, which aren't necessary or present for the simple OSS games Michael tests on Linux. Although they might have some of that for Unigine.

            Comment


            • #7
              Maybe Unigine Heaven would be a good test ?
              Test signature

              Comment


              • #8
                Has the OpenCL performance changed much on these cards?

                Comment


                • #9
                  Oh yeah Unigine Heaven is available for Linux and seems to be a good choice. @Michael Include this in ur next benchmarks, plz.

                  Comment


                  • #10
                    Phoronix must never test the official closed drivers again, you seem ridiculous doing that. Radeon7000 has 4Tflops@32bit wile Kepler has 3.5Tflops@64bit or 7Tflops@32bit(1cuda-core has 2 ALUs). Kepler is 70+% faster than Radeon7000 and twice than Fermi. There is a reason why in benchmarks is only 20% faster than Radeon7000 and 40% than Fermi: When you have 2*GPUs you only have +50% performance, thats not because they don't scale well (thats idiotic and impossible for stream processing), but because the driver turns in quality and precision mode. The loser company uses this trick against Unigine for example, to gain more frames wile losing quality. The benchmark is unable to measure the quality difference, because the benchmark speaks to driver and not directly to hardware. Nvidia has done in the past exactly what AMD does today, when the gtx7800(24pixel processors) it was 20% faster than x1900(48pixel processors). Then all benchmarks changed and gtx7950(2*gtx7800) was 20% faster, wile Nvidia threatens Microsoft that they develop their own API because D3D helps Radeon.

                    Comment

                    Working...
                    X