Announcement

Collapse
No announcement yet.

6-Way Linux Desktop 2D/3D Performance Comparison On Ubuntu 14.04 LTS

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

  • #16
    Originally posted by Vim_User View Post
    So once again Micheal has not benchmarked actual performance, but performance under defaults of Ubuntu developers. Once again a useless benchmark, since the DEs/WMs were not run with same settings. A sad performance for a person running a benchmark site.
    I personally find the benchmark very interesting. Ubuntu is one of the biggest Linux distros for the average desktop user - whether you like it or not. Furthermore the optimum settings should be default and not require further subtle tweaking. If the DE/WM developers want the best exposure and end user experience they should pay attention to and guide down-stream integration.

    If you believe one DE is misrepresented you are welcome to elaborate and suggest concrete and specific improvements. Else I am forced to conclude it is just a sad comment for a person posting nonconstructive messages.

    Comment


    • #17
      Okay this made me curious so I ran a few benchmarks using Unigine-Heaven 4.0 myself. I tried multiple GL rendering backends, suspending effects for fullscreen apps and raster/native modes.

      XFCE 4.10 vs KDE 4.12.3 setup:
      Kernel: 3.13.6 x86_64 (Arch Linux)
      Mesa: 10.1
      CPU: Intel Q9550
      GPU: RadeonHD 4890
      RAM: 4GB DDR2

      Ungine-Heaven setup:
      Quality preset: Low
      AA: Off
      VSync: On
      Fullscreen: Yes
      Resolution: 1920x1080

      Code:
      DE   - Min - Avg  - Max  - GL  - Sus - Raster
      XFCE - 7.9 - 26.0 - 47.6 --------------------
      KDE  - 7.8 - 26.0 - 47.3 - 3.1 -  *  -  *  -
      KDE  - 7.8 - 25.9 - 45.3 - 3.1 -  *  -     -
      KDE  - 7.7 - 25.8 - 46.3 - 3.1 -     -  *  -
      KDE  - 7.7 - 25.8 - 46.5 - 3.1 -     -     -
      KDE  - 7.7 - 26.0 - 47.1 - 2.1 -  *  -  *  -
      KDE  - 7.7 - 25.8 - 46.4 - 2.1 -     -  *  -
      KDE  - 7.8 - 26.0 - 46.0 - 1.2 -  *  -  *  -
      KDE  - 7.7 - 25.8 - 48.5 - 1.2 -     -  *  -
      I'm using OGL 3.1 rendering backend for daily use and in my humble opinion that very small FPS drop is totally worth it considering the extra eyecandy and unmatchable flexibility KDE provides. And if you are a die hard gamer, nothing stops you from having a lightweight OpenBox or XFCE setup beside KDE

      Comment


      • #18
        Originally posted by Veto View Post
        I personally find the benchmark very interesting. Ubuntu is one of the biggest Linux distros for the average desktop user - whether you like it or not. Furthermore the optimum settings should be default and not require further subtle tweaking. If the DE/WM developers want the best exposure and end user experience they should pay attention to and guide down-stream integration.
        Wait, what? Now it is the fault of KDE developers when Ubuntu maintainers choose defaults that are not sane, because they are guiding them not enough? Are you kidding or should this be a serious comment?
        If you believe one DE is misrepresented you are welcome to elaborate and suggest concrete and specific improvements. Else I am forced to conclude it is just a sad comment for a person posting nonconstructive messages.
        Seriously? Did you even read the comments in this thread, were clearly was pointed out that Michael once again has not enabled un-redirecting for fullscreen rendering? Do you see the short test from siavashserver that clearly shows that there is no performance difference between KDE and XFCE if you enable that option? And how about the countless posts for every DE benchmark Michael delivers, where exactly this is pointed out?
        Talk about nonconstructive again.

        Comment


        • #19
          I could not reproduce the bad KDE performance on ArchLinux using all available Unigine benchmarks (Tested xfce, fluxbox and latest KDE).

          Comment


          • #20
            Originally posted by Kemosabe View Post
            I could not reproduce the bad KDE performance on ArchLinux using all available Unigine benchmarks (Tested xfce, fluxbox and latest KDE).

            siavashserver's results also indicate pretty much no difference (0.8% is probably not a statistically significant difference).

            Comment


            • #21
              Originally posted by Vim_User View Post
              Wait, what? Now it is the fault of KDE developers when Ubuntu maintainers choose defaults that are not sane, because they are guiding them not enough? Are you kidding or should this be a serious comment?
              Seriously? Did you even read the comments in this thread, were clearly was pointed out that Michael once again has not enabled un-redirecting for fullscreen rendering? Do you see the short test from siavashserver that clearly shows that there is no performance difference between KDE and XFCE if you enable that option? And how about the countless posts for every DE benchmark Michael delivers, where exactly this is pointed out?
              Talk about nonconstructive again.
              The headline of the article specifically mentions the benchmark is performed under Ubuntu 14.04 LTS. The point is not to make a particular DE look pretty, but to show what a person installing Ubuntu with the chosen DE can expect. So please tell us, why Michael should fiddle with some obscure settings, which would be unknown to most users before making his benchmarks?

              I see many other constructive posts here, presenting data and narrowing down issues based on Michaels benchmark - moving the world forward. If you believe there is a bug in Ubuntu, you could probably have posted it in Launchpad in less time than spent making derogatory posts here about Michael, Ubuntu and whoever.

              I do not expect any constructive or respectful answer from you - so have a nice life...
              Last edited by Veto; 03-12-2014, 06:13 PM.

              Comment


              • #22
                KDE

                So this basically tells me that KDE sucks on Ubuntu.

                Comment


                • #23
                  Originally posted by razaccour View Post
                  So this basically tells me that KDE sucks on Ubuntu.
                  Not really. It basically tells you that Ubuntu's KDE maintainers suck. As you can see, if you use sane settings it is fast as the other DEs.

                  Comment


                  • #24
                    It is what it Is

                    Originally posted by Vim_User View Post
                    Not really. It basically tells you that Ubuntu's KDE maintainers suck. As you can see, if you use sane settings it is fast as the other DEs.
                    I'm not playing the blame game saying it's KDE devs fault. To the user it doesn't matter who's to blame, it is what it is. I agree with you though.

                    Comment

                    Working...
                    X