Announcement

Collapse
No announcement yet.

Sapphire Radeon HD 6870 On Linux

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

  • Sapphire Radeon HD 6870 On Linux

    Phoronix: Sapphire Radeon HD 6870 On Linux

    In this review today at Phoronix we are testing out the Sapphire Radeon HD 6870 Vapor-X 1GB graphics card to see how this popular AMD Radeon graphics processor is performing under Linux.

    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
    Michael, is it possible to run same tests on this cards using windows catalyst driver, linux catalyst driver and linux opensource driver? I know that unigine tests, for example perform same both on windows and linux when on nvidia hardware and their blob. But how is it about amd?

    And thanks for the review!

    Comment


    • #3
      Lastly, when switching over to Unigine Heaven, which is the most demanding Linux tech demo available from Unigine Corp, the RV770-generation of ATI graphics processors were actually faster than the Radeon HD 5700 and Radeon HD 6800 series.
      This is no surprise *at all*, because the RV770 chips don't have standard "DX11" tessellation support, thus won't actually render the same thing.

      Comment


      • #4
        Originally posted by SavageX View Post
        This is no surprise *at all*, because the RV770 chips don't have standard "DX11" tessellation support, thus won't actually render the same thing.
        The unigine-heaven test profile isn't using tessellation enabled.
        Michael Larabel
        https://www.michaellarabel.com/

        Comment


        • #5
          Originally posted by Michael View Post
          The unigine-heaven test profile isn't using tessellation enabled.
          Which settings does it use? For comparison, I'm getting about 24fps on a 460GTX paired with a Core2 E8400@stock. With normal tesselation, max AA and AF.

          Comment


          • #6
            Originally posted by Michael View Post
            The unigine-heaven test profile isn't using tessellation enabled.
            Oh.
            qwertzuiop

            Comment


            • #7
              Originally posted by Michael View Post
              The unigine-heaven test profile isn't using tessellation enabled.
              I think this should get mentioned prominently whenever you use it, then, because most people view the heaven benchmark as primarily about tesselation. That's the first thing that comes to mind, anyway.

              Comment


              • #8
                That's all sweet, but without VDPAU not buying (not to mention opengl bugs).
                Also, it's rather difficult to see which line what color is (temperature, fan speed charts), so either make them thicker or whatnot.

                Comment


                • #9
                  If you wish to see how your system's graphics performance compares to the Radeon HD 6870, you can run phoronix-test-suite benchmark 1102134-IV-SENSORS4225 to replicate the same tests and settings as used in this article so you can see how your system compares side-by-side to these results, when using Phoronix Test Suite 3.0.
                  This is nice and all Michael however since the amount of people that have a 30" monitor is extremely small perhaps future benchmarks should have more then one resolution tested. Runs at 1920x1080 should probably be used as the "base resolution". I'm not saying that 2560x1600 runs should not be included but for readers a resolution run of 1920x1080 would allow more people to do a head to head comparison.

                  Comment


                  • #10
                    Originally posted by deanjo View Post
                    This is nice and all Michael however since the amount of people that have a 30" monitor is extremely small perhaps future benchmarks should have more then one resolution tested. Runs at 1920x1080 should probably be used as the "base resolution". I'm not saying that 2560x1600 runs should not be included but for readers a resolution run of 1920x1080 would allow more people to do a head to head comparison.
                    If you run that comparison it already has 1920 x 1080 results inside.
                    Michael Larabel
                    https://www.michaellarabel.com/

                    Comment

                    Working...
                    X