1. Welcome to TechPowerUp Forums, Guest! Please check out our forum guidelines for info related to our community.

nans detected on GPU, UNSTABLE_MACHINE

Discussion in 'General Software' started by hat, Jan 16, 2009.

  1. hat

    hat Enthusiast

    Joined:
    Nov 20, 2006
    Messages:
    17,166 (5.57/day)
    Thanks Received:
    2,184
    Location:
    Ohio
    What's up with the F@H GPU client? I keep getting errors that say "nans detected on GPU, UNSTABLE_MACHINE error". I googled it and there are lots of people on the F@H forums having this problem. People who don't overclock anything. What's up with F@H?
     
    Crunching for Team TPU More than 25k PPD
  2. hat

    hat Enthusiast

    Joined:
    Nov 20, 2006
    Messages:
    17,166 (5.57/day)
    Thanks Received:
    2,184
    Location:
    Ohio
    bump.
    nobody can confirm this?
     
    Crunching for Team TPU More than 25k PPD
  3. phanbuey

    phanbuey

    Joined:
    Nov 13, 2007
    Messages:
    5,225 (1.92/day)
    Thanks Received:
    990
    Location:
    Miami
    Its a bit after you posted this, but I have the same problem on my second GPU - GPU is stable. I get NANs only on 57XX WU's
     
  4. W1zzard

    W1zzard Administrator Staff Member

    Joined:
    May 14, 2004
    Messages:
    15,316 (3.83/day)
    Thanks Received:
    12,483
    ati or nvidia? which driver version? which gpu client version?
     
    phanbuey says thanks.
  5. Solaris17

    Solaris17 Creator Solaris Utility DVD

    Joined:
    Aug 16, 2005
    Messages:
    17,565 (4.96/day)
    Thanks Received:
    3,845
    Location:
    Florida
    wtf is a nans?
     
  6. phanbuey

    phanbuey

    Joined:
    Nov 13, 2007
    Messages:
    5,225 (1.92/day)
    Thanks Received:
    990
    Location:
    Miami
    NANs - not a number... its nvidia GTX 260 - only the second gpu (GPU 1) - GPU console version 6.23 - drivers 182.05. Consoles are set with -gpu 1/0 flags, CPU affinity locks are disabled. Two different consoles in two different folders.

    It only happens after a 353 point WU from 57XX, and only after the WU finishes. I have to delete queue.dat and work - then it downloads a different WU and runs sometimes for 24 hours until it gets another 353 point 57XX. Meanwhile GPU 0 chugs along just fine... ive swapped the cards - no matter which card is GPU 1, it crashes - regardless of clocks (it crashed the other day at 500 core 1200 shaders and 900mem) - same card is game stable at 756 core and 1566 shaders.
     
    Last edited: Feb 11, 2009
  7. stanhemi

    stanhemi New Member

    Joined:
    Sep 12, 2008
    Messages:
    802 (0.33/day)
    Thanks Received:
    351
    Location:
    Montreal,-Canada.
    Failing projects

    480 points WUs (project ranges : 5013-5016, 5504-5507, 5801)
    384 points WUs (project range : 5757-5764)
    511 points WUs (project range : 5749-5756)
    353 points Wus (project range : 5765-5772)


    GTX2 xxx series
    9xxx series
    8xxx series
    Failing OS

    Windows XP 32 bits
    Windows XP 64 bits
    Windows Vista 32 bits
    Windows Vista 64 bits
     
    phanbuey says thanks.
  8. phanbuey

    phanbuey

    Joined:
    Nov 13, 2007
    Messages:
    5,225 (1.92/day)
    Thanks Received:
    990
    Location:
    Miami
    yep 5765 and up 353 pointers... what can I do about this? Its annoying to have to delete the work files and the queue everytime. Can I disable these WU's from getting assigned to me from the client end?
     
  9. stanhemi

    stanhemi New Member

    Joined:
    Sep 12, 2008
    Messages:
    802 (0.33/day)
    Thanks Received:
    351
    Location:
    Montreal,-Canada.
    phanbuey says thanks.
  10. phanbuey

    phanbuey

    Joined:
    Nov 13, 2007
    Messages:
    5,225 (1.92/day)
    Thanks Received:
    990
    Location:
    Miami
    that's what I was afraid of. Thanks for the confirmation, stanhemi :toast:
     

Currently Active Users Viewing This Thread: 1 (0 members and 1 guest)

Share This Page