Announcement

Collapse
No announcement yet.

Windows 10 vs. Ubuntu 16.04 Gaming With NVIDIA's GTX 1070 & GTX 1080

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

  • #61
    Ultra setting is without TressFX, therefore there should be no diff. But there is a Highest setting which is like Ultra but with TressFX. So if somebody compared Highest it might be possible that Windows is slower.

    Comment


    • #62
      This thread is old, but I would like to say something about using Vulkan. You already said that it is able to better exploit multicores.
      Linux kernel has better schedulers for that. So performance increase can be higher (but don't dream on tens of percents, at most 1-5% extra maybe). Game applications have a workload similar to HPC applications. Physics computations, linear algebra, graph exploration, etc. But of course, due to the real time nature of games, their problem instances are smaller in size.

      And I'll continue saying that once GNU Linux distros like Ubuntu ditch X protocol and server, the benchmarks with equal performance should be better on the Ubuntu side. You are taking out of the way an unnecessary step. The same idea as changing from OpenGL to Vulkan, but probably with less impact.

      The Feral ports are to blame in the bad performing titles. Their wrapper hurts performance, but it allows to have a catalog of games.

      I suppose Valve sees all of this as a long term project. They get money from the store. SteamOS is literally free. Don't know about steam machines partners though.
      And well, they already support OS X. Most of the GNU Linux ports seem to share code with OS X ports (graphics settings are a hint on that).

      Comment


      • #63
        Originally posted by debianxfce View Post


        Reality hits you hard, for example using wayland is slower than X.
        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


        Vulcan games can be faster than OpenGL, because of lower level control of the hardware. It is not the same idea than removing one server in X to wayland.
        *facepalm*

        the programs run in that benchmark were running through xwayland...

        Comment


        • #64
          Lolz, as other user said, the benchmark was running through Xwayland. X11 is a network protocol. It has naturally more overhead than Wayland. Today there are better approaches to remote interfaces like VNC or RDP (I am not sure about the latter). Also, security in X11 is a major issue.

          Comment


          • #65
            Maybe your 4G connection is behind NAT. Just look at the IP you got from your ISP and compare it to your external IP. Some providers use NAT, filter DNS or use a transparent proxy to compress HTML (strips blanks). It ist impossible to parse this HTML output in scripts if you expect something else. So i would say your 4G comment ist pure nonsense (i used 4G at home a couple of years).

            Comment


            • #66
              Nope, it is not, you can also have got a normal external ip, then it can be found like a dsl connection. It really depends on YOUR specific config. I really prefer a pure ip, to be able to provide services to the Internet - you can not connect VPN to your home box without.

              Comment


              • #67
                Well, ssh or whatever, just combine it with dyndns.

                Comment

                Working...
                X