Trouble with GPU client after it completes a work unit

[AMD_Gamer]

trouble with GPU client after it completes work unit

this is what my F@Hmon output shows, says something about UNSTABLE MACHINE and time expiring?

06:51:36] Project: 5214 (Run 4, Clone 14, Gen 13)
[06:51:36]
[06:51:36] Assembly optimizations on if available.
[06:51:36] Entering M.D.
[06:51:42] mdrun_gpu returned -1
[06:51:42] Going to send back what have done.
[06:51:43] logfile size: 5528 info=5528 bed=25 hdr=1
[06:51:43] - Writing 6066 bytes of core data to disk...
[06:51:43] Done: 5554 -> 2052 (compressed to 36.9 percent)
[06:51:43] ... Done.
[06:51:43]
[06:51:43] Folding@home Core Shutdown: UNSTABLE_MACHINE
[06:51:47] CoreStatus = 7A (122)
[06:51:47] Sending work to server


[06:51:47] + Attempting to send results
[06:51:47] + Results successfully sent
[06:51:47] Thank you for your contribution to Folding@Home.
[06:51:51] EUE limit exceeded. Pausing 24 hours.

if i restart the computer it will work through a whole WU like it does not have a problem but then it will do this after starting the second one?

and the log i posted above keeps repeating itself

 

[AMD_Gamer]

now if i restart the computer , quit the client then launch it again, it says that my gpu is not supported or i need the supported drivers?

 

[capreppy]

What card and clocks are you using? What are temps?

I've been playing with my 8800GTX. Even if I think it is stable (using ATI Tool and checking for artifacts) and temps are reasonable, it would still occassionally crash using the GPU2 client. I've backed down my Core and Memory OCs and bumped up my Shader OCs. Temps are very reasonable (100% fan) and I no longer get EUEs. It unfortunately takes some time trying to find clocks that the GPU2 client thinks is stable.

 

[AMD_Gamer]

8800GT default clocks

 

[Sunin]

What drivers?
you should use 177.35 off of nvidia's cuda site!

 

[AMD_Gamer]

What drivers?
you should use 177.35 off of nvidia's cuda site!

those are what i am using

 

[relic]

I had a similar problem.
It was solved by uninstalling all video drivers (back to standard VGA) rebooting and reistalling the 177.35s.

Apparently the CUDA-enabled drivers don't always overinstall correctly and require you to uninstall previous drivers first.

 

[AMD_Gamer]

I had a similar problem.
It was solved by uninstalling all video drivers (back to standard VGA) rebooting and reistalling the 177.35s.

Apparently the CUDA-enabled drivers don't always overinstall correctly and require you to uninstall previous drivers first.

these were the first drivers i installed, but i will try reinstalling anyways

 

[Sunin]

When you say stock... was this an OC2 card or an Super Clocked?

If not then I would look at the CPU OC and see if that maybe forcing the issue.

Worst case uninstall and reinstall.

 

[AMD_Gamer]

i did some research on the folding@home forums, they say this problem is caused by using windows remote desktop, and last night when i encountered this i had remote desktop open, so i will try without it open and see if it works,

 

[relic]

i did some research on the folding@home forums, they say this problem is caused by using windows remote desktop, and last night when i encountered this i had remote desktop open, so i will try without it open and see if it works,

Didn't realize you were using RDP.
Yes the GPU2 client is incompatable with RDP. You'll need to use VNC or similar for remote access.

 

[AMD_Gamer]

Didn't realize you were using RDP.
Yes the GPU2 client is incompatable with RDP. You'll need to use VNC or similar for remote access.

yeah i got UltraVNC up and running i think this will fix the problem, i will report back later

 

[AMD_Gamer]

ok i fixed the problem! it was me using RDP so i now use UltraVNC instead now my box is folding away without any trouble!