Announcement

Collapse
No announcement yet.

Reasons For Losing Motivation In Wayland

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

  • #71
    Originally posted by brosis View Post
    No it doesn't, it just confirms that you overlook things - the 5750-sb is opensource driver with Vadim optimizations and it outperforms catalyst in several tasks and stays on 80% in others.
    You have a very selective perception. A single card that uses special optimizations outperforms the catalyst in a few tests.

    Originally posted by brosis View Post
    Where do your 35% go?
    Look at the rest of the test, they benchmarked quite a few cards there.

    Originally posted by brosis View Post
    You kinda have zero clue
    Says the one who actually claims:

    Originally posted by brosis View Post
    opensource driver provides everything that catalyst provides
    Laughable, you are in total denial of reality.

    Originally posted by brosis View Post
    And if you start calling me "opensource zealot" I will personally contact Michael to ban you asshat, because just due to me prefering opensource anything does not give you a crap to open your shitty mouth.
    Please do that. If you can't stand criticism on your precious open source drivers then i guess the only way to keep your views intact is by trying to get people banned. This is exactly the behaviour one expects from zealots.

    Comment


    • #72
      Originally posted by smitty3268 View Post
      You don't know that. Kubuntu, for example, is clearly considering not shipping Mir.

      It remains to be seen exactly how complicated it will be to strip out. If it's easy, i bet some won't. If it's baked into the very core of Ubuntu and almost impossible to remove, then you're probably right and it will remain in all derivatives.
      Originally posted by dee. View Post
      Why would an xfce distro want an extra display server that they couldn't even make use of, just running under x for no particular reason? Doesn't make any sense for them.

      I think if Ubuntu makes it too hard to strip Mir, a lot of derivatives are going to be rebasing or quitting...
      To answer both:

      AFAIK the Kubuntu and Xubuntu guys have no plans to adapt KWin/Xfwm to support Mir natively. At some point, however, running Xorg stand-alone will no longer be supported (I expect 2016's LTS release to do this at the latest). Wayland may be inherited by *Ubuntu from Debian but also not in an officially supported way.

      One of the arguments for easy access for newcomers is the semi-automatic driver installer. Why should Canonical offer any but Mir-comparible drivers?
      Should Wayland and Mir require different drivers and neither Wayland not Xorg drivers are to be found in the repos, what would be the approach of least work? XMir obviously.

      Comment


      • #73
        Originally posted by duby229 View Post
        For my needs the OSS driver is obviously the better choice.
        I'm happy for you.

        Originally posted by duby229 View Post
        I can't think of a single scenario where guaranteed daily or even hourly kernel panics are "superior in every way".
        Daily or hourly kernel panics?

        Originally posted by duby229 View Post
        Come on. Lets be realistic here.
        Look who's talking.

        Originally posted by duby229 View Post
        These are the facts whether you like them or not.
        I don't think so. You lost your credit when you claimed you had hourly kernel panics. This clearly is not the standard experience a user will have with the catalyst driver. There might have been a driver version which did not work well with your card and led to hourly kernel panics but this is not the standard experience, not even close to it.

        Comment


        • #74
          Originally posted by Temar View Post
          I'm happy for you.



          Daily or hourly kernel panics?



          Look who's talking.



          I don't think so. You lost your credit when you claimed you had hourly kernel panics. This clearly is not the standard experience a user will have with the catalyst driver. There might have been a driver version which did not work well with your card and led to hourly kernel panics but this is not the standard experience, not even close to it.
          Read these very forums. The facts are what the facts are whether you like them or not.

          EDIT: First you try to claim that the proprietary drivers are some kind of panacea, then you try to claim that the OSS drivers are some kind of bane. Seems to me like you are the one who lost his credit. You've done this with no evidence, no links, no benchmarks, no quotes. Nothing. You have nothing.
          Last edited by duby229; 13 June 2013, 01:44 PM.

          Comment


          • #75
            Originally posted by brosis View Post
            This is same old BS as with Logitech crap. If Logitech refuses to support Linux, its Logitech sucking, not Linux.
            If Nvidia or Catalyst don't support Wayland, its them sucking, not Wayland. Also, Wayland is new technology, so chicken-egg. This problem never exists with opensource!
            Agree.


            BTW this is just tabloid like "news". Please Mr Michael keep it professional, don't post bullshit, thanks.

            Comment


            • #76
              Originally posted by Temar View Post
              You have a very selective perception. A single card that uses special optimizations outperforms the catalyst in a few tests.
              I will run full test suit in several days; its not special optimizations - they are in GIT now; and the idea is NOT to outperform catalyst, but to IMPROVE the existing situation - the driver runs at 80%-110% of catalyst, which is SUFFICIENT.

              Originally posted by Temar View Post
              Look at the rest of the test, they benchmarked quite a few cards there.
              The rest does not interest much, because this was intersection between Vadim's testing and that of Michael - from intersection, one already sees +30%..+100% improvement on same card (5750) running similar tests, between opensource (Michael), catalyst(Michael) and opensource (Vadim).

              Originally posted by Temar View Post
              Says the one who actually claims:

              Laughable, you are in total denial of reality.
              YOUR reality does NOT interest me! Opensource driver provides everything I need already.

              Originally posted by Temar View Post
              Please do that. If you can't stand criticism on your precious open source drivers then i guess the only way to keep your views intact is by trying to get people banned. This is exactly the behaviour one expects from zealots.
              I am denying any criticism, I am denying your blatant claims - you are NOT using opensource driver on radeon; you have NO IDEA how far it progressed; I never denied the weak points of opensource driver; and I am not imposing you which driver you should run.
              For disrespecting very spirit of opensource - the core spirit of Linux - and harrasing its users - people should be punished, yes.

              Comment


              • #77
                Originally posted by duby229 View Post
                I deal with the catalyst driver regularly. I maintain a network for medical providers. Mostly linux thin clients. I have evaluated catalyst in dozens of scenarios. For my needs the OSS driver is obviously the better choice.

                I can't think of a single scenario where guaranteed daily or even hourly kernel panics are "superior in every way". Come on. Lets be realistic here.

                EDIT: The OSS drivers have superior desktop performance, hibernation works, I've never seen a kernel panic yet, Multimonitor configurations work great, 3d rendering is accurate and performance has exceeded early expectations.

                These are the facts whether you like them or not.
                Two AMD systems here, both with Catalyst. No kernel panics, no rendering glitches, hibernation works without any problems. Oh, and performance is not on par with the Windows Catalyst, but still better than the FOSS driver. And it has proper power management.

                A few years ago I worked for a OEM that delivered all PC systems with Ubuntu (8.04, later 10.04) by default, all with proprietary drivers. There were no problems at all with the AMD drivers (and FWIW, also not with the Nvidia drivers).

                These are the facts, whether you like them or not.

                Comment


                • #78
                  I never claimed that they won't work for some people. But that is few and far between. Just because you are able to be ignorant of it's buggyness doesn't mean that everyone else can. And -that- is the fact.

                  Comment


                  • #79
                    Originally posted by duby229 View Post
                    I never claimed that they won't work for some people. But that is few and far between. Just because you are able to be ignorant of it's buggyness doesn't mean that everyone else can. And -that- is the fact.
                    If you mean with "some people" literally thousands of delivered systems than you are right, those "few systems" obviously don't count as facts. Welcome to lala-land, where one persons experience count more than 1000s of successful deployments.

                    Comment


                    • #80
                      Originally posted by Vim_User View Post
                      If you mean with "some people" literally thousands of delivered systems than you are right, those "few systems" obviously don't count as facts. Welcome to lala-land, where one persons experience count more than 1000s of successful deployments.
                      What you fail to realize is that a large proportion of those buyers -do- have problems and complain loudly. Many of them come right here to this very forum. Really when was the last time you looked at AMD bug tracker? It really is worth a good look.

                      Comment

                      Working...
                      X