Results 1 to 7 of 7

Thread: Weird benchmark msg

  1. #1
    Join Date
    Jul 2003
    Location
    Sydney, Australia
    Posts
    5,644

    Weird benchmark msg

    I'm getting this message and I haven't seen it on my other computers. Any ideas how to fix it or if it does any harm?

    BOINC v6.4.5
    Win XP Pro 64bit SP2
    ASUS P5K-E mobo
    4GB Corsair Dominator DDR2 1066 RAM
    Intel Q6600 @ 2.7GHz
    Gigabyte Nvidia 9800GT 512MB
    ForceWare 177.79 (according to Riva Tuner 2.21)
    GPU temp 60C fan @ 100%

    16/12/2008 12:32:00 PM||Running CPU benchmarks
    16/12/2008 12:32:00 PM||Suspending computation - running CPU benchmarks
    16/12/2008 12:32:32 PM||Integer benchmark ran only 0.875000 sec; ignoring


  2. #2
    Join Date
    Mar 2008
    Location
    N.E. Wisconsin
    Posts
    202
    Have you run any stress tests such as Prime95 with your current OC settings?

  3. #3
    AMDave's Avatar
    AMDave is offline Seeker of the exit clause Moderator
    Site Admin
    Join Date
    Jun 2004
    Location
    Deep in a while loop
    Posts
    9,612
    Quote Originally Posted by vaughan View Post
    I'm getting this message and I haven't seen it on my other computers. Any ideas how to fix it or if it does any harm?

    BOINC v6.4.5
    ...
    GPU temp 60C fan @ 100%
    ...
    16/12/2008 12:32:32 PM||Integer benchmark ran only 0.875000 sec; ignoring
    focussing on the important bits above, you have the latest BOINC version, the GPU is at ideal temp when the fan is at max.

    The problem is the benchmark.
    This error occurs when BOINC has instructed all wu's to suspend so that it has a clear run at the CPUs, but there is still something running with 'high priority' when the benchmark started.

    I suggest you manually suspend all wu's for a moment.
    Check nothing is chewing up the CPU/GPU
    run a manual benchmark and see if you get the same result

    My guess is that one of the work units was really tardy about suspending and had not stopped by the time the benchmark started.

    Give it a go and see what happens.

    (Don't forget to resume your work units)
    . . . . . ___
    . . . . . . .\___/\______
    . . . . . . . \__AMD___\\__
    ---------------------------------------------

  4. #4
    AMDave's Avatar
    AMDave is offline Seeker of the exit clause Moderator
    Site Admin
    Join Date
    Jun 2004
    Location
    Deep in a while loop
    Posts
    9,612
    1st picked up in BOINC 6.2.x where they inserted the "ignore benchmark if run in less than 3 seconds".
    http://boinc.berkeley.edu/dev/forum_...rap=true#17311

    I think this is when they started pushing in the GPGPU code, which would correlate to the GPU feeder process being in a tight loop on the CPU.

    In almost all cases the next benchmark runs just fine.

    As above, my guess is a tardy wu suspension (eg GPU feeder) OR a non-BOINC process is also slapping the CPU/GPU around at the time the benchmark started.
    Last edited by AMDave; 12-16-2008 at 01:27 AM.
    . . . . . ___
    . . . . . . .\___/\______
    . . . . . . . \__AMD___\\__
    ---------------------------------------------

  5. #5
    Join Date
    Mar 2007
    Posts
    638
    I guess i would try throwing out the Intel CPU and replace it with an AMD :p other then that I wouldn't know what else to try.



  6. #6
    AMDave's Avatar
    AMDave is offline Seeker of the exit clause Moderator
    Site Admin
    Join Date
    Jun 2004
    Location
    Deep in a while loop
    Posts
    9,612
    ha ha ha

    15-Love to Jason.
    . . . . . ___
    . . . . . . .\___/\______
    . . . . . . . \__AMD___\\__
    ---------------------------------------------

  7. #7
    Join Date
    Jul 2003
    Location
    Sydney, Australia
    Posts
    5,644
    Thanks Dave.
    LOL Jason - maybe if the Phenom 2 lives up to its hype

    I think it was something else running. New benchmark is:

    16/12/2008 6:42:57 PM||Benchmark results:
    16/12/2008 6:42:57 PM|| Number of CPUs: 3
    16/12/2008 6:42:57 PM|| 2546 floating point MIPS (Whetstone) per CPU
    16/12/2008 6:42:57 PM|| 7790 integer MIPS (Dhrystone) per CPU

    Note: 3 CPUs as I have Folding GPU client on this box. I paused it to run the BOINC benchmark.


Posting Permissions

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