Announcement

Collapse
No announcement yet.

AMD Ryzen 9 3900X Benchmarks On 11 Linux Distributions

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

  • AMD Ryzen 9 3900X Benchmarks On 11 Linux Distributions

    Phoronix: AMD Ryzen 9 3900X Benchmarks On 11 Linux Distributions

    Now that BIOS updates over the past month have resolved the early boot issue with Ryzen 3000 processors and thus the new AMD CPUs playing nicely with modern Linux distributions, here is the long-awaited benchmark comparison of the Ryzen 9 3900X + X570 system benchmarked across an array of different Linux distributions... In fact, 11 Linux OS releases in total were tested on this high-end 12-core / 24-thread desktop processor.

    http://www.phoronix.com/vr.php?view=28224

  • #2
    Some of those performance differences are so drastic it's as if you were to buy a CPU 2 performance tiers down.

    Comment


    • #3
      Once again, good old boring Debian Buster more than holds its own. Very interesting results.

      Comment


      • #4
        Typo:

        Originally posted by phoronix View Post
        Endeaovour OS and Manjaro were at

        Comment


        • #5
          It'd be nice to know what is so bad in Fedora that makes it the worst.

          Comment


          • #6
            dav1d results
            4K
            Manjaro 154.768
            Endeavour 155.5
            Debian Testing 161.64
            Ubuntu 163.327
            Debian 163.475
            Ubuntu dated 163.62
            Fedora 165.5
            Solus 167.115
            Clear 176.27
            1080p
            Manjaro 348.7899
            Endeavour 350.82765
            Ubuntu 441
            Debian Testing 441.54
            Ubuntu dated 442.62899
            Debian 450.93867
            Solus 479.76
            Fedora 480.4 (exactly)
            Clear 545.08

            Comment


            • #7
              What kind of magic does clear linux use and why not any other distro uses it???

              Comment


              • #8
                The only issue I can see is that on Selenium tests, Debian & Debian Testing are running Firefox 60, while the rest are running Firefox 68.

                I get that these are bone stock installs with the bundled stuff, but v60 represents 4% of the installed base, v68 represents 72%.

                Not a flaw in the test methods, just a heads up that the results for Debian may not represent what really happens today.

                Comment


                • #9
                  Originally posted by 89c51 View Post
                  What kind of magic does clear linux use and why not any other distro uses it???
                  Probably they use CPU extensions not found on older processors that other distributions still wants to support.
                  Among other things.

                  Comment


                  • #10
                    I'm surprised how the rolling release Manjaro/Arch is behind in these. It seems there's something wrong with how they build their packages. Doesn't make sense for newer to be slower in pretty much all tests.

                    Comment

                    Working...
                    X