Results 1 to 9 of 9

Thread: Nearing Autonomous Linux Benchmark Selections

  1. #1
    Join Date
    Jan 2007
    Posts
    15,093

    Default Nearing Autonomous Linux Benchmark Selections

    Phoronix: Nearing Autonomous Linux Benchmark Selections

    When it comes to advancing Linux and open-source benchmarking, a number of breakthroughs are on the horizon with the Phoronix Test Suite and OpenBenchmarking.org...

    http://www.phoronix.com/vr.php?view=MTE0MTE

  2. #2
    Join Date
    Sep 2008
    Location
    Vilnius, Lithuania
    Posts
    2,599

    Default

    That's good to hear. One of the things that is really needed for OpenBenchmarking is a way to have advanced searching and filtering, because as of right now finding anything is nigh-impossible. For instance: http://openbenchmarking.org/results/...0II%20X4%20920 - the top hits have no relation to what was searched at all, and the tests that were performed are not shown (let alone filtering by them).

    By the way, the hardware index has categories, but how about displaying the performance data in a numeral way, so that different components could be compared directly? Adding such info to the product page would also be nice.

    Oh, and has anyone sent an update on the colouring scheme scripts yet?

  3. #3

    Default

    Quote Originally Posted by GreatEmerald View Post
    One of the things that is really needed for OpenBenchmarking is a way to have advanced searching and filtering, because as of right now finding anything is nigh-impossible. For instance: http://openbenchmarking.org/results/search/AMD Phenom II X4 920 - the top hits have no relation to what was searched at all, and the tests that were performed are not shown (let alone filtering by them).
    Yeah, a much-improved search implementation is due for a PTS commercial client within the next few weeks.

    Quote Originally Posted by GreatEmerald View Post
    Oh, and has anyone sent an update on the colouring scheme scripts yet?
    Nope, any color-selection improvements for pts_Graph are welcome though!

  4. #4
    Join Date
    May 2011
    Posts
    353

    Default

    Your CPU information looks good, but for the Intel Core 2 Duo T5450, which is in my laptop, it says that SSE2 is supported while this CPU is capable of (S)SSE3. Anyway, SSE is not benchmarked that often so maybe it's not such a problem.

  5. #5
    Join Date
    Sep 2010
    Posts
    472

    Default

    This all is really nice.

    I have a graphic card that is physically damaged. I get artefacts when using some applications.
    Is or will there be any way to either tell this to the test suite or the test suite figuring that out by itself?

    With the tracking of what OpenGL extensions are being used, the same logic will also become applicable to graphics cards and graphics drivers.
    Does this mean the test suite is going to be used for tracking bugs/features/regressions in OpenGL drivers?

  6. #6

    Default

    Quote Originally Posted by plonoma View Post
    This all is really nice.

    I have a graphic card that is physically damaged. I get artefacts when using some applications.
    Is or will there be any way to either tell this to the test suite or the test suite figuring that out by itself?
    There are image quality tests -- http://www.phoronix.com/vr.php?view=14380 -- assuming that its not like on your display output only where the card is damaged.

    Quote Originally Posted by plonoma View Post
    Does this mean the test suite is going to be used for tracking bugs/features/regressions in OpenGL drivers?
    It can already be used for that. Though in terms of a Phoromatic Tracker, I've been meaning to setup a per-commit one for Intel. Though I don't think so for NVIDIA/AMD, since Intel hardware is low-power and Intel also sends me out sufficient quantities of Intel hardware to use compared to the other vendors.

  7. #7
    Join Date
    Jan 2009
    Posts
    1,705

    Default

    Michael is this the line in the pts_Graph that determines the colours of the lines??

    Code:
    $config['color']['paint'] = array('#065695', '#e12128', '#009345', '#1b75bb', '#a3365c', '#2794A9', '#ff5800', '#221914', '#AC008A', '#E00022', '#3A9137', '#00F6FF', '#8A00AC', '#949200', '#797766', '#5598b1', '#555555', '#757575', '#999999', '#CCCDDD');

  8. #8

    Default

    Quote Originally Posted by 89c51 View Post
    Michael is this the line in the pts_Graph that determines the colours of the lines??

    Code:
    $config['color']['paint'] = array('#065695', '#e12128', '#009345', '#1b75bb', '#a3365c', '#2794A9', '#ff5800', '#221914', '#AC008A', '#E00022', '#3A9137', '#00F6FF', '#8A00AC', '#949200', '#797766', '#5598b1', '#555555', '#757575', '#999999', '#CCCDDD');
    No, within pts_Graph.php it's the public static function color_cache($ns, $id, &$colors) function.

  9. #9
    Join Date
    Sep 2010
    Posts
    472

    Default

    Quote Originally Posted by Michael View Post
    There are image quality tests -- http://www.phoronix.com/vr.php?view=14380 -- assuming that its not like on your display output only where the card is damaged.
    It's indeed not on my display output but in the GPU's 3d processing functionality.

    I notice some things not aligned as they should in an earlier computer and from my other family members.
    And very obvious some levels in UT2004 seem to have triangular shaped holes in the ground the texture flickers between following the ground and just connecting the dots in a straight line.

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •