Announcement

Collapse
No announcement yet.

A 10-Way Linux Distribution Battle To Kick Off 2016

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

  • A 10-Way Linux Distribution Battle To Kick Off 2016

    Phoronix: A 10-Way Linux Distribution To Kick Off 2016

    As our first multi-way Linux distribution comparison of 2016, I took ten different modern Linux distribution releases and benchmarked them on the same Intel Haswell system. Being benchmarked were various releases of Ubuntu, OpenSUSE, Debian, Clear Linux, Fedora, Antergos, and CentOS.

    http://www.phoronix.com/vr.php?view=22686

  • #2
    So Debian wins and Debian loses, and some packages should look at their default compile flags, and consider adding runtime CPU checks.

    Comment


    • #3
      How could you ever forget "quirky linux" ?
      For example this one, full install on hard disk , is very fast :
      http://barryk.org/news/?viewDetailed=00299
      That's what I have been using every day for months.

      Comment


      • #4
        i think debian stretch/testing using gcc 5.3.1 not 4.3.2 on the OP it says 4.3.1

        Comment


        • #5
          does anyone still use Jessie?

          Comment


          • #6
            Originally posted by edmon View Post
            does anyone still use Jessie?
            Everybody uses jessie, since it's the current stable version and only a masochist would use testing for a server.
            ## VGA ##
            AMD: X1950XTX, HD3870, HD5870
            Intel: GMA45, HD3000 (Core i5 2500K)

            Comment


            • #7
              Originally posted by sarfarazahmad View Post
              i think debian stretch/testing using gcc 5.3.1 not 4.3.2 on the OP it says 4.3.1
              Whoops, just a simple typo on that page. Now fixed, thanks.
              Michael Larabel
              http://www.michaellarabel.com/

              Comment


              • #8
                Originally posted by darkbasic View Post
                Everybody uses jessie, since it's the current stable version and only a masochist would use testing for a server.
                No, for desktops, testing has plenty of uses, especially if you don't want to be stuck with ESR browsers (which update 7 times less often than regular ones).

                If you don't need high availability, using testing on a server is perfectly acceptable.

                Comment


                • #9
                  The "new" Xeon Haswell is a tiny bit too new to be benchmarked with Debian 8.2. To gain full speed (without gfx) you can simply use a kernel from Debian backports if you want official support. Our Kanotix Special images use a mesa backport together with Linux 4.2 from Ubuntu 15.10 right now, but as soon as i update fglrx with kernel 4.4 support this might change. Interestingly 4.4 needed 0,2-0,3 s longer with systemd-analyze (rest stock Kanotix, around 3.881s with last test)- booted via UEFI directly without grub/initrd from an ext4 filesystem and Haswell i5-4670.

                  https://packages.debian.org/source/j...ackports/linux

                  It is not needed to replace the full os, just to get support for a new cpu. You can easyly get full Skylake support for Debian 8 too.

                  Comment


                  • #10
                    Originally posted by Calinou View Post
                    If you don't need high availability and fast response time to security problems, using testing on a server is perfectly acceptable.
                    Fixed that for you. Even the Debian Testing Security team admits that security support for Testing is not as good as for Stable.

                    Comment

                    Working...
                    X