Announcement

Collapse
No announcement yet.

Windows 7 & Windows 8 vs. Ubuntu 13.04 & Fedora 18

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

  • #41
    The goal was to compare the performance of the Intel driver on Linux und Windows.

    Originally posted by startzz View Post
    it doesnt really show real performance, because linux doesnt have any aaa or aa games
    "AAA" games with high end graphic effects would perform poor on intel graphics. It's useless to compare results of 3-5 fps.
    Plus the games to test should run natively on Linux and use OpenGL render path on both systems...

    Originally posted by startzz View Post
    its intel graphics we are talking about, its made for longer battery life, not for performance, so...
    Right, but comparing the performance can give us clues about the state of the driver development and potential to reach in the future...

    Originally posted by startzz View Post
    [..] linux drivers still is very bad.
    As the results show, it's a little worse than on Windows, but not "very bad".

    Originally posted by startzz View Post
    windows 8 is not made to be fastest, especially with intel graphics...
    The first test compared Ubuntu 13.04 (Unity) and Windows 8.
    After some arguments, the test was extendet with Ubuntu 13.04 XFCE and KDE.
    After further arguments, the test is now extendet with Windows 7 and Fedora.
    It's impossible to cover every distribution, especially if you start tweaking. The most common default options seem to be covered...

    Originally posted by startzz View Post
    And why the hell author of these stupid comparisons is so dumb ? Why he cant compare a real graphic cards, that were made for gaming, like geforce 690 gtx or radeon 7900 or something like that ????? [..]
    It's hard to compare Intel driver performance on NVIDIA or AMD hardware.
    Maybe it's not the author...
    ... but the reader?

    @ Michael: Thanks for the comparisons. And don't let the unjustified criticism discourage you.

    Comment


    • #42
      [QUOTE=liam;324234]
      Originally posted by jayrulez View Post

      Here's the only link I could find that mentioned the l4 kernel and galaxy s3v(well, actually its a comment to an article...): http://www.osnews.com/thread?544120
      The press release linked to is completely useless, and that company's site has mostly ad-speak level descriptions of the tech. There are some links to some pdfs at their site but all returned the http 404 code.
      Don't forget about windriver. They are a pretty big player as well.
      QNX runs on several architectures, but that, of course, doesn't mean that it scales efficiently. I've looked for some numbers since it claims to scale well, but couldn't find any. I'd especially be curious how well it handles massive smp since that seems to have been a fairly late feature added to what has been a fairly old kernel. I would expect it to be damn good at horizontal scaling, though.
      BTW, no problem with the assumption. This is phoronix after all
      Thanks man, I somehow forgot that I read that comment.

      I have some numbers for QNX compared to Linux and another Kernel on ARM and x86. I will post for you when I can locate them.

      Current QNX is not that old though. Stuff was rewritten for a more modern era, I think that was around QNX 6 Neutrino (not quite sure I recall correctly). It handles SMP well and is really good for distributed computing(see qnet).

      Comment


      • #43
        Originally posted by kraftman View Post
        It is, because you can tune it easily. Trade offs are everywhere, but in Linux you have control over them.
        How do you easily tune Linux for transparent distributed computing or fault isolation?

        It is not easy, and it cannot be done without massive breaking changes.

        Maybe because he wants to have full performance and use Freedom operating system? Maybe because he doesn't want his data to be sent to m$?
        Can the man speak for himself?

        There's nothing better, so they're right.
        I see you are one of those fan girls... sorry, I meant fan boys.

        And that's meaningless, because qnx is crap compared to Linux. Qnx supports arm, power, x86, but so what? It's damn slow, featureless crap.
        Based on your previous comment, I think it would be a waste of effort to try to correct you.

        Nope. Most of the kernels aren't so flexible and when comes to proprietary ones you can nearly tune nothing.
        I know quite a few kernels that are quite flexible and allows such changes to be made without being invasive. Whether it's proprietary or open source doesn't matter. The developer of the kernel can make it do what he wants it to do. I don't suppose you yourself are making massive changes to the linux kernel to get it to do what you want to do.

        Who cares? It's open source and you can take what you want and do what you want from it. I see your logic fails.
        What was the point of that comment? We all know Linux kernel is open source and with enough effort, you could get it to do what you want.

        However, say you want a kernel that enables transparent distributed computing(even across architectures): It would be much easier to use another kernel that already enables this rather than try to tame Linux for this. If you managed to change Linux to do that, it wouldn't even be Linux anymore.
        Last edited by jayrulez; 06 April 2013, 11:48 AM.

        Comment


        • #44
          Originally posted by Michael View Post
          If you are on a modern web-browser (anything supporting SVG, a.k.a most anything in past few years aside from IE), you get the SVG graphs that all have links to the OpenBenchmarking.org results... But if you're using AdBlock or similar, it seems to think that result graphs are ads, so it might be blocking you from seeing it.

          Yes, on OpenBenchmarking.org are all of the key system log files.
          Hum. I can see SVGs, I'm not using AdBlock, and yet I can't find the links. I'm probably just looking at the wrong place, though. Where exactly is that link placed at?

          Comment


          • #45
            Originally posted by Michael View Post
            What's your HTTP user agent string there?
            Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKit/537.31 (KHTML, like Gecko) Chrome/26.0.1410.43 Safari/537.31

            Edit- and headers that work:
            Mozilla/5.0 (Windows NT 6.1; WOW64; rv:21.0) Gecko/20100101 Firefox/21.0
            Mozilla/5.0 (compatible; MSIE 10.0; Windows NT 6.1; WOW64; Trident/6.0)
            Last edited by smitty3268; 06 April 2013, 04:15 PM.

            Comment


            • #46
              Originally posted by GreatEmerald View Post
              Hum. I can see SVGs, I'm not using AdBlock, and yet I can't find the links. I'm probably just looking at the wrong place, though. Where exactly is that link placed at?
              There's a little box icon (i think?) in the bottom left corner of all the charts.

              Comment


              • #47
                Originally posted by smitty3268 View Post
                There's a little box icon (i think?) in the bottom left corner of all the charts.
                Oh! OK, I see it. Although it's super tiny and I never would have thought it's clickable.

                Comment


                • #48
                  Thank you!

                  Originally posted by curaga View Post
                  Re float and S3TC questions, have some direct links:
                  OpenBenchmarking.org, Phoronix Test Suite, Linux benchmarking, automated benchmarking, benchmarking results, benchmarking repository, open source benchmarking, benchmarking test profiles

                  OpenBenchmarking.org, Phoronix Test Suite, Linux benchmarking, automated benchmarking, benchmarking results, benchmarking repository, open source benchmarking, benchmarking test profiles


                  Both have s2tc it seems, only ubuntu has float.
                  Thank you curaga! That was just the answer I was looking for. Now I wonder if the same is true for radeon tests, since they require a variable to be exported in order to activate S3TC feature. =-)
                  I'll be looking for it next time.

                  Comment


                  • #49
                    Originally posted by Ren H?ek View Post
                    The goal was to compare the performance of the Intel driver on Linux und Windows.

                    It's hard to compare Intel driver performance on NVIDIA or AMD hardware.
                    Maybe it's not the author...
                    ... but the reader?

                    @ Michael: Thanks for the comparisons. And don't let the unjustified criticism discourage you.
                    Yes, but it still makes no sense, because no real gamers cares about intel drivers or any sh1t like that, especially on linux, if you wanna play games, then at least buy a pc... for a price of mobile phone you can buy a very good pc with good graphics card, it will be many times better for gaming... Whats the point of watching gay porn, if you are straight ? doing these comparisons are the same freakin thing, just in technological way... You better test how hard have you to hit wall with your head to die
                    Last edited by startzz; 07 April 2013, 06:35 AM.

                    Comment


                    • #50
                      What the hell?

                      Originally posted by startzz View Post
                      Yes, but it still makes no sense, because no real gamers cares about intel drivers or any sh1t like that, especially on linux, if you wanna play games, then at least buy a pc... for a price of mobile phone you can buy a very good pc with good graphics card, it will be many times better for gaming... Whats the point of watching gay porn, if you are straight ? doing these comparisons are the same freakin thing, just in technological way... You better test how hard have you to hit wall with your head to die
                      You are loathsome.

                      Comment

                      Working...
                      X