PDA

View Full Version : New issue has risen!!


GIXXERGUY6
03-08-10, 03:45 PM
new card BFG9800GTX+ OC has worked great stock is 760/1886/1125 I set it at 740/1998/1100 and it's completed about 50 units since I got it.

Now I have this issue[21:40:05]
[21:40:05] *------------------------------*
[21:40:05] Folding@Home GPU Core
[21:40:05] Version 1.31 (Tue Sep 15 10:57:42 PDT 2009)
[21:40:05]
[21:40:05] Compiler : Microsoft (R) 32-bit C/C++ Optimizin
.00.50727.762 for 80x86
[21:40:05] Build host: amoeba
[21:40:05] Board Type: Nvidia
[21:40:05] Core :
[21:40:05] Preparing to commence simulation
[21:40:05] - Looking at optimizations...
[21:40:05] DeleteFrameFiles: successfully deleted file=work
[21:40:05] - Created dyn
[21:40:05] - Files status OK
[21:40:05] - Expanded 70507 -> 383588 (decompressed 544.0 p
[21:40:05] Called DecompressByteArray: compressed_data_size
88, decompressed_data_size=383588 diff=0
[21:40:05] - Digital signature verified
[21:40:05]
[21:40:05] Project: 6604 (Run 4, Clone 15, Gen 0)
[21:40:05]
[21:40:05] Assembly optimizations on if available.
[21:40:05] Entering M.D.
[21:40:11] Tpr hash work/wudata_01.tpr: 1317566169 1025379
1085236572
[21:40:11]
[21:40:11] Calling fah_main args: 14 usage=100
[21:40:11]
[21:40:12] mdrun_gpu returned
[21:40:12] Going to send back what have done -- stepsTotalG
[21:40:12] Work fraction=0.0000 steps=0.
[21:40:16] logfile size=4941 infoLength=4941 edr=0 trr=25
[21:40:16] + Opened results file
[21:40:16] - Writing 5479 bytes of core data to disk...
[21:40:16] Done: 4967 -> 1843 (compressed to 37.1 percent)
[21:40:16] ... Done.
[21:40:16] DeleteFrameFiles: successfully deleted file=work
[21:40:16]
[21:40:16] Folding@home Core Shutdown: UNSTABLE_MACHINE
[21:40:19] CoreStatus = 7A (122)
[21:40:19] Sending work to server
[21:40:19] Project: 6604 (Run 4, Clone 15, Gen 0)
[21:40:19] - Read packet limit of 540015616... Set to 52428


[21:40:19] + Attempting to send results [March 8 21:40:19 U
[21:40:19] - Reading file work/wuresults_01.dat from core
[21:40:19] (Read 2355 bytes from disk)
[21:40:19] Connecting to http://171.64.65.61:8080/
[21:40:20] Posted data.
[21:40:20] Initial: 0000; - Uploaded at ~3 kB/s
[21:40:20] - Averaged speed for that direction ~3 kB/s
[21:40:20] + Results successfully sent
[21:40:20] Thank you for your contribution to Folding@Home.

and I don't know if it is the WU or not. I haven't restarted or changed drivers yet because the XFX9800GTX+ is still working away and having no real issues.

Someone help I am down a GPU :(

ChasR
03-08-10, 03:54 PM
p6600 to p6604 are new, so it could be the WU. One instant U_M doesn't cost you but a few seconds. It's when you go into 24 hour shutdown that you've got a problem.

GIXXERGUY6
03-08-10, 04:35 PM
p6600 to p6604 are new, so it could be the WU. One instant U_M doesn't cost you but a few seconds. It's when you go into 24 hour shutdown that you've got a problem.

oh no I seen 3 24hr shut downs even after restarting the client. I went back to 195.81 and we'll see what happens.

clocks BFG 760/1998/1125
XFX 740/1944/1100

joedymueller
03-08-10, 04:39 PM
are you running in SLI? i turned my sli off on my 295. and i havnt had a problem with GPU since.

KGH
03-08-10, 04:47 PM
oh no I seen 3 24hr shut downs even after restarting the client. I went back to 195.81 and we'll see what happens.

clocks BFG 760/1998/1125
XFX 740/1944/1100

just curious, why you have the Core clock on so high?
I dont think it's a factor at all.
what is the temp when you fold those new WUs?

GIXXERGUY6
03-08-10, 04:51 PM
just curious, why you have the Core clock on so high?
I dont think it's a factor at all.
what is the temp when you fold those new WUs?

my temps stay around 65-70 for GPU0 which is XFX and 52-60 for BFG

the core clocks are stock clocks. BFG is OEM OC'd.

GIXXERGUY6
03-08-10, 04:52 PM
the BFG is working on a 6602 and is 19% done so far
the XFX is working on a 6604 and is 48% done so far

Mark620
03-08-10, 05:00 PM
The only thing on the gpu I even try to OC is the shaders, everything else is stock.

the_cultie
03-09-10, 01:04 PM
I got some of those 66xx units and its given me 5 errors while I was at work so not my gpu is down for 24hrs. My day just keeps getting worse. Is there any way I can get my gpu folding quicker? I havnt changed any drivers or settings on my card. I've been folding for 1 year now with the same oc :mad:

Mark620
03-09-10, 05:24 PM
cultie,

Clean the HS&F if that does not fix it then:
Card is getting worn out.
Performance degrades over time.
back it off one step...

the_cultie
03-10-10, 10:09 AM
I got it folding again. I restarted the PC (was artifacting in game), deleted the work folder, log and queue.dat file and away it went. Has folded a few of them 66xx units no problem. As for temps they never go above 35°C, water cooled and I actually cleaned my radiator on Sunday. Will see how it goes before I reduce the OC. At least I'm folding again :)