Announcement

Collapse
No announcement yet.

Ubuntu 12.04.4 vs. 13.10 vs. 14.04 LTS Desktop Benchmarks

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

  • Ubuntu 12.04.4 vs. 13.10 vs. 14.04 LTS Desktop Benchmarks

    Phoronix: Ubuntu 12.04.4 vs. 13.10 vs. 14.04 LTS Desktop Benchmarks

    With Ubuntu 14.04 LTS being released today, here's some fresh benchmarks comparing the Ubuntu 14.04 64-bit performance against Ubuntu 13.10 and Ubuntu 12.04.4 LTS. In this article are desktop and gaming benchmarks comparing these versions of Ubuntu Linux.

    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
    I don't play much games so this seems promising to me. Anyhow I still hope graphics performance will be better in the future with the Ubunut 14.04 LTS version.

    So in my case, I'll probably upgrade.

    Comment


    • #3
      Do a gaming test benchmarking Valve's games like Half Life 2 and Team Fortress 2.

      Also benchmark some other games like Metro Last Light, etc. Some games that people actually play.

      Comment


      • #4
        Originally posted by SangeetKhatri View Post
        Do a gaming test benchmarking Valve's games like Half Life 2 and Team Fortress 2.

        Also benchmark some other games like Metro Last Light, etc. Some games that people actually play.
        I can't wait to see the answer you get. You're new here.

        Comment


        • #5
          i have been saying this for a few months. Steam games performance has almost halved on mesa 10.2 vs 9.2.5

          Comment


          • #6
            Originally posted by phill1978 View Post
            i have been saying this for a few months. Steam games performance has almost halved on mesa 10.2 vs 9.2.5
            Possibly because of MSAA being disabled before vs enabled now. Thats guaranteed to drop performance quite a bit.

            If on Intel set

            Code:
            INTEL_FORCE_MSAA=0
            in your ~/.bashrc file.

            If you're on non-Intel set

            Code:
            GALLIUM_MSAA=0
            in your ~/.bashrc file

            Logout / reboot. Then try and see if performance is better.
            All opinions are my own not those of my employer if you know who they are.

            Comment


            • #7
              Originally posted by SangeetKhatri View Post
              Do a gaming test benchmarking Valve's games like Half Life 2 and Team Fortress 2.

              Also benchmark some other games like Metro Last Light, etc. Some games that people actually play.
              Not until the Metro Last Light game developers, etc, add support for automated benchmarking to their Linux port...

              See http://www.phoronix.com/scan.php?pag...tem&px=MTUyNTU
              Michael Larabel
              https://www.michaellarabel.com/

              Comment


              • #8
                Possibilities that I see for 14.04 regressing:
                1) Increased graphical fidelity as previously mentioned (e.g. MSAA)
                2) HyperZ being disabled by default now... I'm not sure if that made it into 14.04 or if that's just in mesa-master currently.
                3) If radeonsi is being built by default now, that likely means that r600g was built against LLVM. Maybe there's missing optimizations in LLVM vs vadim's SB backend.
                4) It's Thursday and not Friday, and therefore the universe dislikes things whose versions have more than one '4' in it.

                Comment


                • #9
                  Originally posted by Veerappan View Post
                  Possibilities that I see for 14.04 regressing:
                  1) Increased graphical fidelity as previously mentioned (e.g. MSAA)
                  2) HyperZ being disabled by default now... I'm not sure if that made it into 14.04 or if that's just in mesa-master currently.
                  3) If radeonsi is being built by default now, that likely means that r600g was built against LLVM. Maybe there's missing optimizations in LLVM vs vadim's SB backend.
                  4) It's Thursday and not Friday, and therefore the universe dislikes things whose versions have more than one '4' in it.
                  i can officially tell you its not number 1. MSAA works like a dream on 9.2.5 where as previously it did not at all. I can run x4 msaa on counter strike with no mouse lag and although the frame rate is lower it is not gimped. In fact AA is gimped again on 10.1. and above as is the frame consistency / sync. The kernal improvements on 3.12 and onwards made a big improvement to AMD but i cant say the same for x.server 1.7.3.2 and mesa 10.2 im on older versions for now until it is sorted out.

                  I also noticed a complete stop in Oibaf/edgers updates on Mint/Ubuntu. I was getting them daily for months and for over a month now nothing.. not one update.

                  at the moment it feels like AMD is too far in bed in the corporate world of windows and consoles to care. Typically AMD cant see the woods for the trees.
                  Last edited by ForkedPython; 17 April 2014, 07:29 PM.

                  Comment


                  • #10
                    hm, am I the only one who noticed regression in almost every test? And why are you calling it desktop tests whereas these test are actually graphics driver tests? What an intel or nvidia user can learn from it?

                    Comment

                    Working...
                    X