Announcement

Collapse
No announcement yet.

The XanMod Kernel Is Working Well To Boost Ubuntu Desktop / Workstation Performance

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

  • The XanMod Kernel Is Working Well To Boost Ubuntu Desktop / Workstation Performance

    Phoronix: The XanMod Kernel Is Working Well To Boost Ubuntu Desktop / Workstation Performance

    It's been four years since last testing out the XanMod kernel as a spin of the Linux kernel with various patches and extra tuning designed to offer better desktop/workstation performance, similar to the Liquorix kernel. But given the recent Liquorix kernel testing and discussions over kernel schedulers and more, here are some fresh benchmarks of the latest XanMod kernel. Long story short, I am quite impressed by these latest XanMod results.

    http://www.phoronix.com/vr.php?view=28770

  • #2
    These kernel testings are great however the gaming part could really use a 95 percentile value instead of only min/max/avg
    Especially with VR use rising up, making sure that a majority of frames do not take more time than the refresh threshold is more important than anything else.
    Last edited by sheepdestroyer; 01-10-2020, 01:20 AM.

    Comment


    • #3
      Since it wasn't mentioned in the article: the Xanmod binary does not have BMQ enabled, so this is with default scheduler. BMQ is only available in the experimental binaries or you can enable it when you build it yourself.

      Comment


      • #4
        Originally posted by sheepdestroyer View Post
        These kernel testings are great however the gaming part could really use a 95 percentile value instead of only min/max/avg
        Especially with VR use rising up, making sure that a majority of frames do not take more time than the refresh threshold is more important than anything else.
        I tend to agree. Minimum and maximum values can be very sensitve while averages can misrepresent intermittent data. If it is at all possible, I'd prefer to see overlayed PDFs of the frame rate or frame time.

        Comment


        • #5
          Originally posted by TuxTuxTux View Post
          Since it wasn't mentioned in the article: the Xanmod binary does not have BMQ enabled, so this is with default scheduler. BMQ is only available in the experimental binaries or you can enable it when you build it yourself.
          It seems CFS with low-latency kernel and tuned CFS low-latency kernel gives the best results. 'Thanks' to Ubuntu for throwing server kernel at their users.

          Comment


          • #6
            It seems that it ships with ZFS modules, which is a deal breaker for me.

            Comment


            • #7
              Originally posted by DoMiNeLa10 View Post
              It seems that it ships with ZFS modules, which is a deal breaker for me.
              Why? If you don't use ZFS, the modules aren't loaded/used are they?

              Comment


              • #8
                Ugh, throughput tests being used to benchmark low-latency kernel again... Thanks for testing it for performance (throughput) regressions I guess.

                We really need ways to measure LATENCY to test things like this. Like test long does launching gedit take when your system is at 100% CPU and disk usage? What is the worst 1% FPS for a given game? How long does a click on the start menu take until the menu opens? How long does it take to switch from LibreOffice to Chrome to Firefox? How long does it take to open a new tab and open a simple page from disk in Chrome/Firefox? And so on and so forth.

                I know it's difficult to test for latency, but there's very little point testing throughput on low-latency kernels...

                Comment


                • #9
                  Originally posted by DoMiNeLa10 View Post
                  It seems that it ships with ZFS modules, which is a deal breaker for me.
                  It does not. Same as with BMQ, this is only enabled in the experimental builds.

                  Comment


                  • #10
                    What I see in these benchmarks is that kernel release freshness contributes to performance much more than out-of-tree patches.

                    Comment

                    Working...
                    X