8600GT + F@H

[mrSHEiK124]

This is my log file at the moment. This card is not overclocked, has adequate power (it also fails when plugged into the same computer that's happily folding away with a GTX260), and I've tried 4 or 5 different drivers. It's even more annoying when it gets to 70-80% and just EUEs out of nowhere. Any ideas?

--- Opening Log file [February 16 19:47:52 UTC]


# Windows GPU Console Edition #################################################
###############################################################################

Folding@Home Client Version 6.23

http://folding.stanford.edu

###############################################################################
###############################################################################

Launch directory: C:\Users\Abid\AppData\Roaming\Folding@home-gpu


[19:47:52] - Ask before connecting: No
[19:47:52] - User name: sheik124 (Team 198)
[19:47:52] - User ID: 414E30C710007E50
[19:47:52] - Machine ID: 2
[19:47:52]
[19:47:52] Loaded queue successfully.
[19:47:52] Initialization complete
[19:47:52] - Preparing to get new work unit...
[19:47:52] + Attempting to get work packet
[19:47:52] - Connecting to assignment server
[19:47:53] - Successful: assigned to (171.67.108.11).
[19:47:53] + News From Folding@Home: GPU folding beta
[19:47:53] Loaded queue successfully.
[19:47:57] + Closed connections
[19:47:57]
[19:47:57] + Processing work unit
[19:47:57] Core required: FahCore_11.exe
[19:47:57] Core found.
[19:47:57] Working on queue slot 04 [February 16 19:47:57 UTC]
[19:47:57] + Working ...
[19:47:57]
[19:47:57] *------------------------------*
[19:47:57] Folding@Home GPU Core - Beta
[19:47:57] Version 1.19 (Mon Nov 3 09:34:13 PST 2008)
[19:47:57]
[19:47:57] Compiler : Microsoft (R) 32-bit C/C++ Optimizing Compiler Version 14.00.50727.762 for 80x86
[19:47:57] Build host: amoeba
[19:47:57] Board Type: Nvidia
[19:47:57] Core :
[19:47:57] Preparing to commence simulation
[19:47:57] - Looking at optimizations...
[19:47:57] - Created dyn
[19:47:57] - Files status OK
[19:47:57] - Expanded 46723 -> 252912 (decompressed 541.3 percent)
[19:47:57] Called DecompressByteArray: compressed_data_size=46723 data_size=252912, decompressed_data_size=252912 diff=0
[19:47:57] - Digital signature verified
[19:47:57]
[19:47:57] Project: 5765 (Run 9, Clone 202, Gen 113)
[19:47:57]
[19:47:57] Assembly optimizations on if available.
[19:47:57] Entering M.D.
[19:48:03] Working on Protein
[19:48:05] Client config found, loading data.
[19:48:05] Starting GUI Server
[19:51:42] Completed 1%
[19:54:44] Run: exception thrown during GuardedRun
[19:54:44] Run: exception thrown in GuardedRun -- Gromacs cannot continue further.
[19:54:44] Going to send back what have done -- stepsTotalG=15000000
[19:54:44] Work fraction=0.0191 steps=15000000.
[19:54:48] logfile size=10564 infoLength=10564 edr=0 trr=23
[19:54:48] - Writing 11100 bytes of core data to disk...
[19:54:48] Done: 10588 -> 3887 (compressed to 36.7 percent)
[19:54:48] ... Done.
[19:54:48]
[19:54:48] Folding@home Core Shutdown: UNSTABLE_MACHINE
[19:54:52] CoreStatus = 7A (122)
[19:54:52] Sending work to server
[19:54:52] Project: 5765 (Run 9, Clone 202, Gen 113)


[19:54:52] + Attempting to send results [February 16 19:54:52 UTC]
[19:54:53] + Results successfully sent
[19:54:53] Thank you for your contribution to Folding@Home.
[19:54:57] - Preparing to get new work unit...
[19:54:57] + Attempting to get work packet
[19:54:57] - Connecting to assignment server
[19:54:58] - Successful: assigned to (171.67.108.11).
[19:54:58] + News From Folding@Home: GPU folding beta
[19:54:58] Loaded queue successfully.
[19:54:59] + Closed connections
[19:55:04]
[19:55:04] + Processing work unit
[19:55:04] Core required: FahCore_11.exe
[19:55:04] Core found.
[19:55:04] Working on queue slot 05 [February 16 19:55:04 UTC]
[19:55:04] + Working ...
[19:55:04]
[19:55:04] *------------------------------*
[19:55:04] Folding@Home GPU Core - Beta
[19:55:04] Version 1.19 (Mon Nov 3 09:34:13 PST 2008)
[19:55:04]
[19:55:04] Compiler : Microsoft (R) 32-bit C/C++ Optimizing Compiler Version 14.00.50727.762 for 80x86
[19:55:04] Build host: amoeba
[19:55:04] Board Type: Nvidia
[19:55:04] Core :
[19:55:04] Preparing to commence simulation
[19:55:04] - Looking at optimizations...
[19:55:04] - Created dyn
[19:55:04] - Files status OK
[19:55:04] - Expanded 45339 -> 251112 (decompressed 553.8 percent)
[19:55:04] Called DecompressByteArray: compressed_data_size=45339 data_size=251112, decompressed_data_size=251112 diff=0
[19:55:04] - Digital signature verified
[19:55:04]
[19:55:04] Project: 5769 (Run 3, Clone 168, Gen 177)
[19:55:04]
[19:55:04] Assembly optimizations on if available.
[19:55:04] Entering M.D.
[19:55:11] Working on Protein
[19:55:11] Client config found, loading data.
[19:55:12] Starting GUI Server

 

[mrSHEiK124]

Ugh, this is ridiculous. I'm just going to run GPUGrid on this 8600GT and switch to CPU F@H.

[19:58:32] Completed 1%
[20:42:22] Completed 14%
[20:47:23] Run: exception thrown during GuardedRun
[20:47:23] Run: exception thrown in GuardedRun -- Gromacs cannot continue further.
[20:47:23] Going to send back what have done -- stepsTotalG=15000000
[20:47:23] Work fraction=0.1500 steps=15000000.
[20:47:27] logfile size=13079 infoLength=13079 edr=0 trr=23
[20:47:28] - Writing 13615 bytes of core data to disk...
[20:47:28] Done: 13103 -> 4304 (compressed to 32.8 percent)
[20:47:28] ... Done.
[20:47:28]
[20:47:28] Folding@home Core Shutdown: UNSTABLE_MACHINE
[20:47:31] CoreStatus = 7A (122)
[20:47:31] Sending work to server
[20:47:31] Project: 5769 (Run 3, Clone 168, Gen 177)


[20:47:31] + Attempting to send results [February 16 20:47:31 UTC]
[20:47:31] + Results successfully sent
[20:47:31] Thank you for your contribution to Folding@Home.
[20:47:35] - Preparing to get new work unit...
[20:47:35] + Attempting to get work packet
[20:47:35] - Connecting to assignment server
[20:47:36] - Successful: assigned to (171.67.108.11).
[20:47:36] + News From Folding@Home: GPU folding beta
[20:47:36] Loaded queue successfully.
[20:47:37] + Closed connections
[20:47:42]
[20:47:42] + Processing work unit
[20:47:42] Core required: FahCore_11.exe
[20:47:42] Core found.
[20:47:42] Working on queue slot 06 [February 16 20:47:42 UTC]
[20:47:42] + Working ...
[20:47:42]
[20:47:42] *------------------------------*
[20:47:42] Folding@Home GPU Core - Beta
[20:47:42] Version 1.19 (Mon Nov 3 09:34:13 PST 2008)
[20:47:42]
[20:47:42] Compiler : Microsoft (R) 32-bit C/C++ Optimizing Compiler Version 14.00.50727.762 for 80x86
[20:47:42] Build host: amoeba
[20:47:42] Board Type: Nvidia
[20:47:42] Core :
[20:47:42] Preparing to commence simulation
[20:47:42] - Looking at optimizations...
[20:47:42] - Created dyn
[20:47:42] - Files status OK
[20:47:42] - Expanded 45415 -> 251112 (decompressed 552.9 percent)
[20:47:42] Called DecompressByteArray: compressed_data_size=45415 data_size=251112, decompressed_data_size=251112 diff=0
[20:47:42] - Digital signature verified
[20:47:42]
[20:47:42] Project: 5771 (Run 6, Clone 187, Gen 160)
[20:47:42]
[20:47:42] Assembly optimizations on if available.
[20:47:42] Entering M.D.
[20:47:49] Working on Protein
[20:47:50] Client config found, loading data.
[20:47:50] Starting GUI Server
[20:51:12] Completed 1%
[21:14:26] Completed 8%
[21:14:26] mdrun_gpu returned
[21:14:26] NANs detected on GPU
[21:14:26]
[21:14:26] Folding@home Core Shutdown: UNSTABLE_MACHINE
[21:14:28] CoreStatus = 7A (122)
[21:14:28] Sending work to server
[21:14:28] Project: 5771 (Run 6, Clone 187, Gen 160)
[21:14:28] - Error: Could not get length of results file work/wuresults_06.dat
[21:14:28] - Error: Could not read unit 06 file. Removing from queue.
[21:14:28] - Preparing to get new work unit...
[21:14:28] + Attempting to get work packet
[21:14:28] - Connecting to assignment server
[21:14:29] - Successful: assigned to (171.64.65.106).
[21:14:29] + News From Folding@Home: GPU folding beta
[21:14:29] Loaded queue successfully.
[21:14:30] + Closed connections
[21:14:35]
[21:14:35] + Processing work unit
[21:14:35] Core required: FahCore_11.exe
[21:14:35] Core found.
[21:14:35] Working on queue slot 07 [February 16 21:14:35 UTC]
[21:14:35] + Working ...
[21:14:35]
[21:14:35] *------------------------------*
[21:14:35] Folding@Home GPU Core - Beta
[21:14:35] Version 1.19 (Mon Nov 3 09:34:13 PST 2008)
[21:14:35]
[21:14:35] Compiler : Microsoft (R) 32-bit C/C++ Optimizing Compiler Version 14.00.50727.762 for 80x86
[21:14:35] Build host: amoeba
[21:14:35] Board Type: Nvidia
[21:14:35] Core :
[21:14:35] Preparing to commence simulation
[21:14:35] - Looking at optimizations...
[21:14:35] - Created dyn
[21:14:35] - Files status OK
[21:14:35] - Expanded 68462 -> 357580 (decompressed 522.3 percent)
[21:14:35] Called DecompressByteArray: compressed_data_size=68462 data_size=357580, decompressed_data_size=357580 diff=0
[21:14:35] - Digital signature verified
[21:14:35]
[21:14:35] Project: 5761 (Run 12, Clone 42, Gen 94)
[21:14:35]
[21:14:35] Assembly optimizations on if available.
[21:14:35] Entering M.D.
[21:14:42] Working on Protein
[21:14:43] Client config found, loading data.
[21:14:44] Starting GUI Server
[21:19:24] Completed 1%
[21:24:05] Completed 2%

 

[Insidious]

I've read thread after thread about this card struggling with F@H. They have something borked up in a way that causes this and I don't think it was ever resolved (I doubt they tried much)

I think you are wise not to bother trying to fix F@H on an 8600GT. The problem is not at your end. (afaik_

-Sid

 

[Drsignguy]

Follow Sids insites. All good as he know much more than me but I do know that you have to run simular cards with each other if you run 2 Gpu's.

Also, I know you said your card isn't overclocked, but I have tried and succeded at dropping the core/shader and memory clocks down and then bringing back up the shader clock to stock and/or higher. Give that a shot and see if that helps.

I did this with the MSI GTX 260/192 that is an OC card and It doesn't like F@H at it's default core speed. I dropped from 620 to 576 and the shaders are up from 12?? to 1512 and memory is now at 999. It has been working just great since I did this. :thumbsup:

May luck be with you

 

[geokilla]

Downclock your card by 10%, run stress testing like OCCT 3.0 or FurMark or w/e and see what the temps are and whether or not if you get any artifacts. Also try using Driver Cleaner. It should help. Also, maybe you just need to reconfigure your folding client. Also check what the error codes are.

 

[Denithor]

You're running the console client, right? I had real stability problems with the graphical tray client on even a GTX 260.

And, BTW, that 8600GT isn't going to crank a lot of ppd for you even if you get it working. These days the heavier WUs will bog down an 8800GS/9600GSO pretty badly (I've seen my heavily overclocked model only putting out 2200 ppd on some WUs).

 

[TAandy]

Originally posted by: Insidious
I've read thread after thread about this card struggling with F@H. They have something borked up in a way that causes this and I don't think it was ever resolved (I doubt they tried much)

I think you are wise not to bother trying to fix F@H on an 8600GT. The problem is not at your end. (afaik_

-Sid

not trying to be a smart arse, but i had no problems with f@h on my 8600gt
but that was on the tray version, not the console

 

[Insidious]

Originally posted by: TAandy

not trying to be a smart arse, but i had no problems with f@h on my 8600gt
but that was on the tray version, not the console

It was the same with the threads at the F@H forums. Some people said they were fine and many said they were not.
To give you an idea how severe the issue was, Stanford posted, both development team members and moderators that there was a problem. I don't know if you ever go to those threads, but they pretty much have to be gored in the azz by a unicorn to admit there are any issues with their project.

The answer there was the same answer I'll give here (basically already did). If it is working for you far out! If not, you are amoung those who are probably not going to see much luck.

-Sid

PS: I don't see differing perspectives as being a smart azz or anything bad at all. We need all the inputs we can get!

So now let's input a :beer:

 

[TAandy]

Originally posted by: Insidious

Originally posted by: TAandy

not trying to be a smart arse, but i had no problems with f@h on my 8600gt
but that was on the tray version, not the console

It was the same with the threads at the F@H forums. Some people said they were fine and many said they were not.
To give you an idea how severe the issue was, Stanford posted, both development team members and moderators that there was a problem. I don't know if you ever go to those threads, but they pretty much have to be gored in the azz by a unicorn to admit there are any issues with their project.

The answer there was the same answer I'll give here (basically already did). If it is working for you far out! If not, you are amoung those who are probably not going to see much luck.

-Sid

PS: I don't see differing perspectives as being a smart azz or anything bad at all. We need all the inputs we can get!

So now let's input a :beer:

i'll drink to that :beer:

 

[mrSHEiK124]

Originally posted by: Denithor
You're running the console client, right? I had real stability problems with the graphical tray client on even a GTX 260.

And, BTW, that 8600GT isn't going to crank a lot of ppd for you even if you get it working. These days the heavier WUs will bog down an 8800GS/9600GSO pretty badly (I've seen my heavily overclocked model only putting out 2200 ppd on some WUs).

I will try the console client. I'm using the tray client on my GTX265 (I refuse to call them GTX260 Core216 any longer, what the hell was nVIDIA thinking? if you can have a GTX285 why not a GTX265? GTX270?) and it works fine. Worked great on my 9600GSO when I had it too. Weird.

As of know it's just erroring in guarded run or something right around 50% on every WU it does.

 

[mrSHEiK124]

OMG, it finished a WU!

[16:07:24] Project: 5769 (Run 7, Clone 311, Gen 231)

 

[Insidious]

:laugh:

 

[Drsignguy]

Well congratulations! :thumbsup: and btw, it's about time