Announcement

Collapse
No announcement yet.

Manjaro vs. Ubuntu vs. Fedora vs. OpenSUSE Benchmarks

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

  • Manjaro vs. Ubuntu vs. Fedora vs. OpenSUSE Benchmarks

    Phoronix: Manjaro vs. Ubuntu vs. Fedora vs. OpenSUSE Benchmarks

    The latest Linux distribution benchmarks to share at Phoronix are a comparison of Manjaro Linux 0.8.8, Ubuntu 13.10, Ubuntu 14.04 LTS in its current development state, openSUSE 13.1, and Fedora 20. All tests were done from an Intel Core i5 4670 Haswell system to look at the current state of various Linux distributions when it comes to various areas of open-source performance.

    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
    Didn't know that Manjaro is that out-dated for something that claims to be based on Arch.

    Comment


    • #3
      There's a kernel 3.13 package for Manjaro, should make a big difference, specially in GPU benchmarks.

      Manjaro uses older packages than Arch, as it is supposed to be stability-focused. There are different repositories you can choose from, depending on how up-to date you want to be. Here's an explanation:

      Comment


      • #4
        Those benchmarks could use a "More is better" \ "Less is better" captions.

        Comment


        • #5
          Originally posted by glock24 View Post
          There's a kernel 3.13 package for Manjaro, should make a big difference, specially in GPU benchmarks.
          Well, not to mention that it's still using Mesa 9.2.3 (which is not even the latest bug fix release in the 9.2 series) instead of Mesa 10 like in Arch...

          Comment


          • #6
            Originally posted by damg View Post
            Well, not to mention that it's still using Mesa 9.2.3 (which is not even the latest bug fix release in the 9.2 series) instead of Mesa 10 like in Arch...
            Manjaro stable uses mesa 10 since early january. Mesa 9.2 isnt even in the stable repositories anymore ...

            Manjaro stable is usually 2 weeks later than arch unless there is breakage. If you dont care for that you can use Manjaro unstable which is about 24 hours later than arch irregardless of breakage.

            Also manjaro install disks default to LTS kernel, and unlike arch they wont update major numbers, only minor numbers automatically. If you want a more recent kernel you need to tell the system, its fully supported, just not automatic due to the possibility of regressions.

            oh, and hi forum, first post .
            Last edited by SebastianB; 05 February 2014, 08:38 AM.

            Comment


            • #7
              Thanks a lot Michael

              Thanks a lot for this benchmarks Idolove, not all is Ubuntu or derivates, and lately some SteamOS
              Manjaro is a great Arch fork and perhaps their Ubuntu alike.

              I would add Antergos and SteamOS and or debian to the mix too but thanks a lot
              I switched to Manjaro even I still keep my Xubuntu as dual boot and I feel it is faster in the long term thanks to AUR
              AUR is much better than PPAs even for upgrading

              As an MS WOS suffers after a clean install with antivirus and with every program you add to the register
              My Xubuntu suffers with PPAs, and I do like to have the AUR / PPAs software

              Last but not least a fair benchmark would be with all the distros with the SAME kernel version, 3.12, 3.13 or both

              Comment


              • #8
                Arch > all for benchmarking latest linux versions. Nothing is as recent as it, not debian sid, and not fedora.

                Manjaro has imho the best kernel selection though, from 3.4 to 3.13 all installed with a single command and kept up to date. I actually use it for an old laptop for which 3.4 is the last kernel giving working wifi. Quite strange using a bleeding edge distro on deprecated hardware ... but none of the other major distros work for it anymore.

                Comment


                • #9
                  Wouldn't Antergos be a better choice than Manjaro, since it's using basically vanilla Arch repos? The goal is essentially to benchmark Arch without the setup time/interaction required, so it seems like a better choice.

                  Comment


                  • #10
                    Originally posted by benmoran View Post
                    Wouldn't Antergos be a better choice than Manjaro, since it's using basically vanilla Arch repos? The goal is essentially to benchmark Arch without the setup time/interaction required, so it seems like a better choice.
                    Only at first glance. Most benchmarks on phoronix are pretty kernel dependent, i.e. you may always want the latest userland, latest kernel is another beast though. Frequently there are issues with propriety drivers and latest kernels, or bad regressions. So being able to say:

                    I want Kernel x.y(y+1,y+2,...) with latest mesa/nvidia/fglrx is neat. Manjaro also ensures that every kernelversion has its corrosponding propriety drivers in the repository in binary form so, meaning no buildfailures.

                    For example, i have kernel 3.10 + nvidia installed. Now i tell the system:
                    sudu mhwd-kernel -i linux312

                    That will automatically install latest version of 3.12 series with nvidia propriety drivers and make a grub boot entry for it. Same for 3.11, 3.13 or 3.8, whatever you want. After installation normal system update (pacman -Syu) will keep all installed kernels and propriety drivers at latest version. Like i said, its neat.

                    The actual userland ... manjaro unstable basicly is vanilla arch. They dont change the packages, merely hold them for a time depending on repository. They do add a couple custom ones like the ones enabling the kerbel stuff above, but i dont think those would affect a benchmark.

                    Comment

                    Working...
                    X