Announcement

Collapse
No announcement yet.

Intel Linux Driver Performance Still Slower Than Windows 7

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

  • #31
    ???

    Originally posted by CrvenaZvezda View Post
    Yes, please stop using Ubuntu. It's getting more and more meaningless to use Ubuntu to test Linux performance. Use a proper Linux distribution like Fedora or openSUSE.
    why?

    fedora sucks with proprietary drivers... openSuse exists yet?

    itś a non sense using a distro with few people using

    Comment


    • #32
      Originally posted by valeriodean View Post
      In the first page I see:

      Plus the fact that (in a previous article) canonical will use XMir for the 13.10 for the open driver with a fallback in case you install the binary blob.
      If canonical has decided to push XMir for the 13.10 release then it is pretty obvious that they need to test it and so turn it on in the dev snapshot, isn't it?
      The last time I checked, intel provide open source driver only, so how can you sure that ubuntu 13.10 dev on intel driver do not use XMir?
      There is an easy way to be sure: drop ubuntu for the graphic drivers testing purpose.
      "Display Server: X Server 1.14.2". It's on the first page.

      Comment


      • #33
        Originally posted by Andrecorreia View Post
        why?

        fedora sucks with proprietary drivers... openSuse exists yet?

        itś a non sense using a distro with few people using
        Care to back it with some facts or youre just trolling

        Comment


        • #34
          Ubuntu have official daily's of Linux kernel. Its popular. AMD/Nvidia proprietary drivers target Ubuntu specifically. There are already ppa's for X.org/Mesa test builds. Steambox will use it. All the OEM Linux stuff circle around Ubuntu. (And most probably PTS corporate users require good Ubu support)

          Ubuntu is quite good choice for testing desktop Linux performance. (And Linux stand there for kernel, not userland :P)

          And Ubuntu is the distro that is most likely to be first Linux desktop distro anyone new will use.

          Comment


          • #35
            Originally posted by przemoli View Post
            Ubuntu have official daily's of Linux kernel. Its popular. AMD/Nvidia proprietary drivers target Ubuntu specifically. There are already ppa's for X.org/Mesa test builds. Steambox will use it. All the OEM Linux stuff circle around Ubuntu. (And most probably PTS corporate users require good Ubu support)

            Ubuntu is quite good choice for testing desktop Linux performance. (And Linux stand there for kernel, not userland :P)

            And Ubuntu is the distro that is most likely to be first Linux desktop distro anyone new will use.
            Until now.

            Comment


            • #36
              Originally posted by przemoli View Post
              Ubuntu have official daily's of Linux kernel. Its popular. AMD/Nvidia proprietary drivers target Ubuntu specifically. There are already ppa's for X.org/Mesa test builds. Steambox will use it. All the OEM Linux stuff circle around Ubuntu. (And most probably PTS corporate users require good Ubu support)/
              Popular yet hardly profitable in enterprise environment. Canonical has habit to taking credit from other works through PR.
              OEM Linux stuff mainly target profitable entity meaning Canonical played very little role.
              Last edited by finalzone; 07-20-2013, 03:03 PM.

              Comment


              • #37
                Originally posted by artivision View Post
                Phoronix must start using the most advanced real Linux destro -=Fedora=-. As they have a GPU-drivers PPA (debian), they must have another for Fedora (rpm).
                It certainly makes sense. I really don't care which LINUX that Larabel chooses to use, but at this point it's fairly obvious that Ubuntu wants to go off and do it's own thing. It doesn't want to be a Linux. Would Larabel constantly benchmark and use Android as his base and then still claim to be a Linux site? Technically, he could get away with it. He could also get away with using MacOS as his base and calling his website a BSD website. But people do not run around calling MacOS "BSD", they don't run around calling Android "Linux", and Ubuntu is putting itself in this same category. It's not Linux. It's Ubuntu. It's not BSD, it's MacOS. It's not Linux, it's Android. It's not Linux, it's Ubuntu.

                I use Fedora specifically because of Fedora's graphics stack, BTW. So I would have a preference in that direction, but any Linux that actually wants to be a Linux is just fine.

                I'll give you an example: http://www.phoronix.com/scan.php?pag...tem&px=MTQxNTc

                Canonical submitted 15 patches to Mesa for Mir.

                WHEN HAS CANONICAL EVER PUT FORTH PATCHES FOR MESA OR XORG OR ANY OTHER PROJECT?!?! WTF!

                Now sure, I'm confident that one patch happened 4 years ago blah blah blah, so what? Canonical is notoriously uninterested in being an upstream partner like Redhat, Arch, Mint, Debian, Gentoo, and (I do honestly mean to name every one of them but I know I'll make the post unreadable). But you get the point.

                Now that it suits Canonical they submit patches? WTF. They're leeches. They're parasites.

                They so rarely contribute back to the projects that they take advantage of, that it was actually a headliner:
                http://www.phoronix.com/scan.php?pag...tem&px=MTEwNTc

                And here's how it opens:

                While Canonical is known -- and commonly criticized -- for not investing in making heavy, low-level upstream Linux contributions,

                Comment


                • #38
                  Originally posted by Calinou View Post
                  Meanwhile: *buntus are still the best compromise around.
                  I'm not talking about compromising. *buntu hasn't been much of a compromiser over the last few years now have they? All they (normally) ever do is take, take, take, take, take, take.

                  Canonical so rarely ever contributes back to upstream, that it's newsworthy items here at Phoronix whenever it does happen.

                  It's "man bites dog" for them to contribute. The normal "dog bites man" story is this: Yet another day, and *buntu still doesn't contribute.

                  Take, take, take, take, take, take, take, take, take, take, take, take, take, take, take, take, take, take.

                  Comment


                  • #39
                    Originally posted by halfmanhalfamazing View Post
                    ...
                    I'm sure the direct Ubuntu derivatives would contribute, except the fact that most of them (especially Xubuntu and I think Lubuntu) both have a very, very small development team.

                    Whereas Ubuntu has Canonical and like 57859349534954 developers.

                    Comment


                    • #40
                      Originally posted by Andrecorreia View Post
                      maybe because is the future?!

                      until brings all features like windows mesa drivers will be a "little" worst than windows. the evolution says Mesa intel needs one more year to catch up windows on perfomance
                      English requires subject pronouns.

                      Comment


                      • #41
                        Gorgeous!

                        Originally posted by siride View Post
                        English requires subject pronouns.
                        You made my day!

                        Comment


                        • #42
                          Originally posted by sireangelus View Post
                          To Michael: i think you're missing the purpose of latency benchmarking. It's not only to have a average reading, but what matters most is how many, how frequent and the severity of the frame spikes, because that's the major cause of "perceived lagging"(beyond networking), usually a slightly slower frame average but with a much more stable frame latency is preferable over an inflated fps average with lots of spikes(like we see on this review, windows have It worse.)
                          This is also why I prefer to play LoL on Linux even if the avg on windows is way higher.. I have frame latency problems even with VSync on and an apparent rock-solid 60fps
                          lol.. guys... please.. don't go off topic.. let's talk about this instead..

                          Comment


                          • #43
                            Originally posted by artivision View Post
                            Phoronix must start using the most advanced real Linux destro -=Fedora=-. As they have a GPU-drivers PPA (debian), they must have another for Fedora (rpm).
                            No, you benchmark the most used distro, since thats the one people care the most about.

                            Don't like it? Come up with a more popular distribution.

                            Comment


                            • #44
                              We are not in the corporate sector

                              Originally posted by gamerk2 View Post
                              No, you benchmark the most used distro, since thats the one people care the most about.

                              Don't like it? Come up with a more popular distribution.
                              This argument only applies to the corporate sector.
                              As this article is only meant to measure the GNU/Linux Intel driver performance (which is a Kernel-matter), the desktop environment shouldn't be the bottleneck here.
                              Ubuntu's Unity is a _huge_ one and we should avoid it.

                              Comment


                              • #45
                                Originally posted by frign View Post
                                This argument only applies to the corporate sector.
                                As this article is only meant to measure the GNU/Linux Intel driver performance (which is a Kernel-matter), the desktop environment shouldn't be the bottleneck here.
                                Ubuntu's Unity is a _huge_ one and we should avoid it.
                                Yeah, Unity is such a _huge_ bottleneck: http://www.phoronix.com/scan.php?pag...tem&px=MTM5MzY

                                Comment

                                Working...
                                X