Announcement

Collapse
No announcement yet.

Radeon vs. NVIDIA With Windows 10 & Ubuntu Linux

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

  • Eliasvan
    replied
    Originally posted by Michael View Post

    Eliasvan are you still involved in SuperTuxKart? The past few STK releases I seem to be unable to get around the 120 max_fps. The CLI option no longer seems to work nor any other max FPS override. Any idea? Thanks.
    Dear Michael,

    I'm sorry for the very late response. No, I currently don't have time anymore to update those tests unfortunately.

    Best regards,
    Elias

    Leave a comment:


  • Michael
    replied
    Originally posted by Eliasvan View Post
    Eliasvan are you still involved in SuperTuxKart? The past few STK releases I seem to be unable to get around the 120 max_fps. The CLI option no longer seems to work nor any other max FPS override. Any idea? Thanks.

    Leave a comment:


  • Eliasvan
    replied
    Michael
    Update: supertuxkart-1.5.2

    Leave a comment:


  • Eliasvan
    replied
    Originally posted by Eliasvan View Post

    Michael
    Fixed, please find the updated profile together with the change-log at http://openbenchmarking.org/test/eli...rtuxkart-1.5.0
    matthewtrescott agreed with the changes, for reference see our discussion at http://irc.wxm.be/logs/%23supertuxkart.2017.12.log
    Could you please push it to https://github.com/phoronix-test-sui...ee/master/pts/ since I'm not authorized to do so?

    Tested with Windows 10 and Ubuntu 16.04.

    Note that there's no zip archive available for the Windows build of Supertuxkart, only an NSIS installer, which means simply extracting is not possible and therefore I had to use the installer's command-line options set the install destination to the test directory.
    Hence during the installation UAC might ask to confirm (but I imagine that's also the case for some other Windows tests).
    You also might want to pull https://github.com/phoronix-test-sui...suite/pull/229 in order to get rid of some hacky code I had to use in the "install_windows.sh" shell script, see the pull request's / commit's description for more info.

    Feel free to ask any questions.
    Michael
    Update: supertuxkart-1.5.1
    Any feedback please?

    Please also take a look at following issues on GitHub of phoronix-test-suite: #231 and #232:
    I provided a fix for them, please merge them if you agree.

    Leave a comment:


  • Eliasvan
    replied
    Originally posted by Michael View Post

    I don't think the supertuxkart test profile currently has any Windows support.
    Michael
    Fixed, please find the updated profile together with the change-log at http://openbenchmarking.org/test/eli...rtuxkart-1.5.0
    matthewtrescott agreed with the changes, for reference see our discussion at http://irc.wxm.be/logs/%23supertuxkart.2017.12.log
    Could you please push it to https://github.com/phoronix-test-sui...ee/master/pts/ since I'm not authorized to do so?

    Tested with Windows 10 and Ubuntu 16.04.

    Note that there's no zip archive available for the Windows build of Supertuxkart, only an NSIS installer, which means simply extracting is not possible and therefore I had to use the installer's command-line options set the install destination to the test directory.
    Hence during the installation UAC might ask to confirm (but I imagine that's also the case for some other Windows tests).
    You also might want to pull https://github.com/phoronix-test-sui...suite/pull/229 in order to get rid of some hacky code I had to use in the "install_windows.sh" shell script, see the pull request's / commit's description for more info.

    Feel free to ask any questions.

    Leave a comment:


  • varikonniemi
    replied
    I would be ashamed to release a program that is not equally optimized on all platforms it supports.

    Leave a comment:


  • gregzeng
    replied
    So much easier to skim-read your reviews now. Thank you. Still however waiting for the "Executive Summary" or "Abstract", where the summary is: In x number of benchmarks, the xx DRIVER significantly outperformed the yy DRIVER. In the z1 numbers of z2 tests, Linux (brand-version) significantly outperformed Windows (brand-version). Hardware was cc CPU, dd GPU, ee SSD.

    Unstated in the report, but assumed, we have
    (1) Linux on EXT4, and Windows on Microsoft NTFS, appropriate to the original operating systems.
    (2) standard malware-protection & default settings as standard, were unchanged on all operating systems; Windows Defender (default).
    (3) The video-GPU drivers (version numbers not given) were updated to the latest STABLE versions?
    (4) The Linux kernel was also the latest STABLE (not RC) version as well? Windows (version?) was also the stable version, not the Insiders Preview?

    Eventually we hope that technical journalism is the computer industry will reach the high publishing standards of other engineering & management sciences.
    Last edited by gregzeng; 04 December 2017, 06:00 PM.

    Leave a comment:


  • nomadewolf
    replied
    Originally posted by Brisse View Post
    Michael
    Have you ever tried benchmarking games on Clear Linux? I wonder if those optimizations have any impact on gaming.
    Good point.
    You got me curious also.

    Leave a comment:


  • mitcoes
    replied
    I would like to read benchmarks for MS WOS 10 vs Ubuntu (WINE) vs Ubuntu (native) to let us know more.

    Leave a comment:


  • ruthan
    replied
    Get rid of OpenArena from test suite, for all other games are Windows same or better, in OpenArena is Linux way better => OpenArena results are not valid, its Windows code is poor.

    Leave a comment:

Working...
X