Announcement

Collapse
No announcement yet.

Fedora vs. Ubuntu Linux Benchmarks On Intel Haswell

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

  • Fedora vs. Ubuntu Linux Benchmarks On Intel Haswell

    Phoronix: Fedora vs. Ubuntu Linux Benchmarks On Intel Haswell

    Our latest Intel Core i7 4770K "Haswell" Linux benchmarks come in the form of comparing the performance of Ubuntu 13.04, an Ubuntu 13.10 development snapshot, and Fedora 19.

    http://www.phoronix.com/vr.php?view=18844

  • #2
    Whatsup with john the ripper being so much slower on fedora?

    Comment


    • #3
      It's also slower in Ubuntu 13.10, so it may be due to a regression somewhere in the kernel bits I think.

      It's me, or is Michael beginning to use more Fedora with the KDE Desktop than the Gnome one ?

      Comment


      • #4
        Originally posted by Morpheus View Post
        It's me, or is Michael beginning to use more Fedora with the KDE Desktop than the Gnome one ?
        I think he's continuing using OS X

        Comment


        • #5
          Originally posted by Morpheus View Post
          It's also slower in Ubuntu 13.10, so it may be due to a regression somewhere in the kernel bits I think.
          Maybe, it could be the CPU incorrectly going into one of the new s01x states. Or the compile flags changed and CPU acceleration of DES has been disabled.

          Comment


          • #6
            I dont think that tests are so usefull. ok also nice to have some gaming-benchmarks.

            but I would found it more interesting how fast the desktop is in comparsion, I find the difference between gnome 3.4 or 3.6 on a ubuntu basis much slower than the new fedora 19 gnome 3.8 experince and also there are factors like in fedora tmpfs is activated or setupd and stuff like that.

            so how much faster is a fedora current gnome vs latest gnome install on a ubuntu, gnome-shell and also in comparsion to unity. how good is ubuntu for a desktop if speed matters. is it extremly slower or only a bit?


            so is ubuntu a valid distro fro a modern distro or is only good for servers if you dont like debian or something like that?

            Comment


            • #7
              Originally posted by blackiwid View Post
              I dont think that tests are so usefull. ok also nice to have some gaming-benchmarks.

              but I would found it more interesting how fast the desktop is in comparsion, I find the difference between gnome 3.4 or 3.6 on a ubuntu basis much slower than the new fedora 19 gnome 3.8 experince and also there are factors like in fedora tmpfs is activated or setupd and stuff like that.

              so how much faster is a fedora current gnome vs latest gnome install on a ubuntu, gnome-shell and also in comparsion to unity. how good is ubuntu for a desktop if speed matters. is it extremly slower or only a bit?


              so is ubuntu a valid distro fro a modern distro or is only good for servers if you dont like debian or something like that?
              One of the big problems with measuring desktop user interface speed is that one of the most important metrics to the user is latency, but this is rarely measured in benchmarks. Google did quite an interesting thing in Android by generating an "Application Not Responding" error when the ui latency exceeds a certain time. This forces developers to think in terms of keeping synchronous processing off the ui thread so the ui remains responsive. I wish there was a way to enforce the same thing for Linux desktop apps; freezing and hung interfaces should be a relic of the past.

              Comment


              • #8
                Originally posted by Morpheus View Post
                It's also slower in Ubuntu 13.10, so it may be due to a regression somewhere in the kernel bits I think.

                It's me, or is Michael beginning to use more Fedora with the KDE Desktop than the Gnome one ?
                I presume the use of KDE is just to provide a check in case GNOME's use of 3D acceleration to render the desktop affects performance in 3D applications.

                Comment


                • #9
                  KDE uses composition too for desktop effects, so I'm not sure it is the reason, or maybe KDE is less harming performance than GNOME now. My remark was also on the fact that we see more benchmarks running on another distro (Fedora to name it), and it's good, as Ubuntu is not the only Linux spin in the wild.

                  Comment

                  Working...
                  X