Announcement

Collapse
No announcement yet.

AMD Radeon HD 8670D: Gallium3D vs. Catalyst

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

  • AMD Radeon HD 8670D: Gallium3D vs. Catalyst

    Phoronix: AMD Radeon HD 8670D: Gallium3D vs. Catalyst

    This morning there were the RadeonSI Gallium3D vs. AMD Catalyst Linux benchmarks for the high-end Radeon HD 7850/7950 "Southern Islands" graphics cards. While the new Southern Islands GPUs understandingly have a long way to catch up on their new open-source Linux Gallium3D driver compared to Catalyst, how is the AMD Radeon HD 8670D "Richland" APU performance between the open and closed-source drivers? Here are some benchmarks.

    http://www.phoronix.com/vr.php?view=18874

  • #2
    Doesn't make sense to do benchmarking of these cards while their most apparent and prominent performance bottleneck is being alleviated in a version that is +1 of the current one.

    Will be nice to compare differences though...

    Comment


    • #3
      well having in mind with kernel 3.10 APU can't reclock easily, it should improve by a lot with 3.11, i expect michael to try it once RC1 is out in the open

      Comment


      • #4
        Still no benchmarks with Source games like Left 4 Dead 2?

        Comment


        • #5
          I will not even open page: without DPM it's completely useless. Also please benchmark an R600g (pre-radeonsi) APU too.

          Edit: Left 4 Dead 2 please.
          ## VGA ##
          AMD: X1950XTX, HD3870, HD5870
          Intel: GMA45, HD3000 (Core i5 2500K)

          Comment


          • #6
            A10 6800k with the 8670 HD is an r600 APU (it's VLIW4)
            Last edited by ObiWan; 07-05-2013, 05:30 PM.

            Comment


            • #7
              Originally posted by ObiWan View Post
              A10 6800k with the 8670 HD is an r600 APU (it's VLIW4)
              Indeed. And like all AMD APUs, it defaults to a low power state upon boot.

              So either you use a kernel with dynamic power management, or you manually set high profile, otherwise it's not a meaningful test. I'm assuming that the power state was left in a "default" state here.

              It's a lot like comparing a car with automatic transmission to a car with manual, but refusing to shift out of the first gear, because this is "the default" that new users will see. :P

              Comment


              • #8
                Originally posted by Rexilion View Post
                Doesn't make sense to do benchmarking of these cards while their most apparent and prominent performance bottleneck is being alleviated in a version that is +1 of the current one.

                Will be nice to compare differences though...
                Especially considering he's also using ram that is HALF of the rated stock speed for these chips. What little meaningless data he does get using the old kernel is further thrown off by having the GPU be constantly starved for data from memory.

                Phoronix should be renamed UselessReviews.com

                Comment


                • #9
                  1066 Mhz is DDR3 2133,

                  DDR3 1066 would be 533 Mhz

                  Comment


                  • #10
                    Originally posted by darkbasic View Post
                    I will not even open page: without DPM it's completely useless. Also please benchmark an R600g (pre-radeonsi) APU too.

                    Edit: Left 4 Dead 2 please.
                    Originally posted by ChemicalBrother View Post
                    Still no benchmarks with Source games like Left 4 Dead 2?
                    It's not like Valve doesn't make a dedicated benchmarking tool... I've got it installed here, and it most definitely is a native Linux program.

                    Comment


                    • #11
                      Originally posted by Kivada View Post
                      It's not like Valve doesn't make a dedicated benchmarking tool... I've got it installed here, and it most definitely is a native Linux program.
                      Last check, the process still doesn't exit cleanly after test completion.
                      Michael Larabel
                      http://www.michaellarabel.com/

                      Comment


                      • #12
                        Originally posted by Michael View Post
                        Last check, the process still doesn't exit cleanly after test completion.
                        Yeah, that's a pretty bad bug in PTS if it can't handle that.

                        Comment


                        • #13
                          Originally posted by smitty3268 View Post
                          Yeah, that's a pretty bad bug in PTS if it can't handle that.
                          It's not a bug in PTS but a bug in steam that when quitting that the process remains alive, can be reproduced from the command-line.
                          Michael Larabel
                          http://www.michaellarabel.com/

                          Comment


                          • #14
                            Originally posted by Michael View Post
                            It's not a bug in PTS but a bug in steam that when quitting that the process remains alive, can be reproduced from the command-line.
                            Requiring proprietary apps to behave in any special way just to run in PTS sounds like a bug in PTS to me, and others have been able to get benchmark results out of Steam apps.

                            Call it what you will, I'm going to continue saying it's a PTS bug.

                            Comment


                            • #15
                              Originally posted by smitty3268 View Post
                              Requiring proprietary apps to behave in any way just to run in PTS sounds like a bug in PTS to me.

                              Call it what you will, I'm going to continue saying it's a PTS bug.
                              Any other game with automated test abilities works just fine, but after you tell a Steam game to quit from its console, even when it was launched directly from the command-line, the screen goes away and returns to the desktop, but the process actually remains alive. That's an issue with the upstream (steam).
                              Michael Larabel
                              http://www.michaellarabel.com/

                              Comment

                              Working...
                              X