Announcement

Collapse
No announcement yet.

Mesa 11.0 vs. 11.2-devel Tests For Intel Skylake

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

  • Mesa 11.0 vs. 11.2-devel Tests For Intel Skylake

    Phoronix: Mesa 11.0 vs. 11.2-devel Tests For Intel Skylake

    Given the fairly slow news day due to the holidays this week in the United States, here are some extra benchmarks to share this evening: some Mesa 11.0.2 vs. 11.2-devel Git benchmarks with an Intel Skylake system,..

    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
    That's the second article where this has happened. What's the point of the PTS if it can't get the basics right

    Comment


    • #3
      I guess once again 4K results on Xonotic, but FullHD results advertised

      Comment


      • #4
        Originally posted by FireBurn View Post
        That's the second article where this has happened. What's the point of the PTS if it can't get the basics right
        To what are you talking about?
        Michael Larabel
        https://www.michaellarabel.com/

        Comment


        • #5
          Originally posted by Michael View Post

          To what are you talking about?
          This:
          Note the reported P-State governor difference between reboot...
          You tried handwaving it away in the article, but it's basically rendering the whole point of PTS moot. These comparisons are useless while this issue continues.

          Comment


          • #6
            Originally posted by FireBurn View Post
            What's the point of the PTS if it can't get the basics right
            The culprit isn't PTS...
            ## VGA ##
            AMD: X1950XTX, HD3870, HD5870
            Intel: GMA45, HD3000 (Core i5 2500K)

            Comment


            • #7
              Originally posted by smitty3268 View Post

              This:

              You tried handwaving it away in the article, but it's basically rendering the whole point of PTS moot. These comparisons are useless while this issue continues.
              It's not PTS, all PTS does is get the contents of what's shown in the P-State/CPUFreq sysfs. It's straight getting the file contents without any operations or anything on it. It's the P-State driver or whatever it depends upon for determining default governor type that's acting up. If it weren't for PTS, it wouldn't even be known the difference unless manually cat'ing the file each time.
              Michael Larabel
              https://www.michaellarabel.com/

              Comment


              • #8
                I bought a skylake also and I'm super sad because I can't change the CPU governor.

                Did some success to get ``sudo cpupower frequency-set -g performance'' to do something? My cores are at 800MHz instead of 4GHz... It makes it impossible to reliably make music using realtime synthesizers...

                Comment


                • #9
                  Originally posted by Michael View Post

                  It's not PTS, all PTS does is get the contents of what's shown in the P-State/CPUFreq sysfs. It's straight getting the file contents without any operations or anything on it. It's the P-State driver or whatever it depends upon for determining default governor type that's acting up. If it weren't for PTS, it wouldn't even be known the difference unless manually cat'ing the file each time.
                  It doesn't matter whose fault it is. The bottom line is that your results are not repeatable or a fair comparison.

                  Comment

                  Working...
                  X