• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

Need GPU setup help

I

Intelman34

Limp Gawd
Joined
Mar 5, 2003
Messages
467
I got my GTX 295 in the mail today and am having some troubles getting it folding.

I don't have dummy plugs yet so I have just been testing, but im a bit concerned that even 1 instance of the client throws back "unstable machine". This is just simply downloading and installing the GPU client. Nothing special and no other instances running.

I also have a GTX 260 in the machine, but again, nothing is setup for that.

This is Win 7 64bit.

Does anyone know of good, accurate (recent), multi gpu setup guides?

Thanks!
 
Last edited:
What video card is your monitor plugged into?
Which 295? Sandwich or Co-op?
 
Monitor is plugged into the main DVI on the GTX 295. It is a dual PCB.

It downloads a work packet, starts working, then the monitor flashes once very quickly (a blink), and folding stops with unstable machine.

Light on back of card is green.
 
First Take the GTX 260 out for now

If not then:
Uninstall drivers
reboot
driver sweep
reboot
install new drivers
reboot
try again
 
same result... either way.

I downloaded the OCCT gpu tester and it runs, but has a few thousand errors after only a few min. Could something be up with the card?
 
Its possible.

I will say, not many people here including myself have had good luck folding on a 295 using Vista/7
 
I kept getting errors on 2 of my 4 cores on my dual 295 setup under 7
had to switch back the xp64, and now I'm going fine. You could try hdmi dummies but results are hit and miss atbest, I had one hdmi dummy work, one didn't

good luck, VERY small OC puts them at 16k per card
Posted via [H] Mobile Device
 
I am going to stick with 7 for sure. Shouldn't it at least get going on one of the GPUs? I am going to try the dummy plugs soon, but I figured I could at least get 1 instance going.
 
I had one each fine, and my best record was 3 of 4 under 7 using a hdmi dummy

you could also try that registry hack to force monitor detection, did t work for me but didfor others
I would suggest the dummy option, or get another monitor.
Posted via [H] Mobile Device
 
Also, if I try to disable SLI on the 295, the screen just goes blank and eventually reverts to how it was.
 
Folding seems to be causing the driver crash. After the screen blinks and folding shows unstable machine, the card will not clock above 400mhz if I fire up 3dmark.

The card runs 3dmark just fine with no errors or visible artifacts.

Now I am just confused.
 
It runs FurMark 1.8.0 xtreme burn test with no issue as well. Full 3d mode kicks in, clocks go to where they should, and both cores are at 100% usage.

I can also disable SLI after another fresh install of drivers.

Still no folding.
 
Heres a snapshot of the log:

This is with a single GPU install just trying to get 1 instance running. Once the first "unstable machine" hits, the screen blinks, and the card downclocks to 400mhz. Before that it runs at full (576mhz) 3d speeds. Once it blinks and goes to 400mhz, it will not increase speed anymore until I restart.

Any ideas yet? I'd like to get this and the GTX 260 back rolling!

[15:09:04] Decompressed FahCore_11.exe (1908736 bytes) successfully
[15:09:09] + Core successfully engaged
[15:09:14]
[15:09:14] + Processing work unit
[15:09:14] Core required: FahCore_11.exe
[15:09:14] Core found.
[15:09:14] Working on queue slot 01 [March 7 15:09:14 UTC]
[15:09:14] + Working ...
[15:09:14]
[15:09:14] *------------------------------*
[15:09:14] Folding@Home GPU Core
[15:09:14] Version 1.31 (Tue Sep 15 10:57:42 PDT 2009)
[15:09:14]
[15:09:14] Compiler : Microsoft (R) 32-bit C/C++ Optimizing Compiler Version 14.00.50727.762 for 80x86
[15:09:14] Build host: amoeba
[15:09:14] Board Type: Nvidia
[15:09:14] Core :
[15:09:14] Preparing to commence simulation
[15:09:14] - Looking at optimizations...
[15:09:14] DeleteFrameFiles: successfully deleted file=work/wudata_01.ckp
[15:09:14] - Created dyn
[15:09:14] - Files status OK
[15:09:14] - Expanded 64638 -> 341507 (decompressed 528.3 percent)
[15:09:14] Called DecompressByteArray: compressed_data_size=64638 data_size=341507, decompressed_data_size=341507 diff=0
[15:09:14] - Digital signature verified
[15:09:14]
[15:09:14] Project: 5784 (Run 0, Clone 35, Gen 54)
[15:09:14]
[15:09:14] Assembly optimizations on if available.
[15:09:14] Entering M.D.
[15:09:20] Tpr hash work/wudata_01.tpr: 2188830699 1211495396 3434354702 1536426367 2365523595
[15:09:20]
[15:09:20] Calling fah_main args: 14 usage=100
[15:09:20]
[15:09:20] Working on Gromacs Runs One Microsecond At Cannonball Speeds
[15:09:21] mdrun_gpu returned
[15:09:21] Self-test failure
[15:09:21]
[15:09:21] Folding@home Core Shutdown: UNSTABLE_MACHINE
[15:09:24] CoreStatus = 7A (122)
[15:09:24] Sending work to server
[15:09:24] Project: 5784 (Run 0, Clone 35, Gen 54)
[15:09:24] - Read packet limit of 540015616... Set to 524286976.
[15:09:24] - Error: Could not get length of results file work/wuresults_01.dat
[15:09:24] - Error: Could not read unit 01 file. Removing from queue.
[15:09:24] - Preparing to get new work unit...
[15:09:24] + Attempting to get work packet
[15:09:24] - Connecting to assignment server
[15:09:25] - Successful: assigned to (171.67.108.21).
[15:09:25] + News From Folding@Home: Welcome to Folding@Home
[15:09:25] Loaded queue successfully.
[15:09:25] + Closed connections
[15:09:30]
[15:09:30] + Processing work unit
[15:09:30] Core required: FahCore_11.exe
[15:09:30] Core found.
[15:09:30] Working on queue slot 02 [March 7 15:09:30 UTC]
[15:09:30] + Working ...
[15:09:31]
[15:09:31] *------------------------------*
[15:09:31] Folding@Home GPU Core
[15:09:31] Version 1.31 (Tue Sep 15 10:57:42 PDT 2009)
[15:09:31]
[15:09:31] Compiler : Microsoft (R) 32-bit C/C++ Optimizing Compiler Version 14.00.50727.762 for 80x86
[15:09:31] Build host: amoeba
[15:09:31] Board Type: Nvidia
[15:09:31] Core :
[15:09:31] Preparing to commence simulation
[15:09:31] - Looking at optimizations...
[15:09:31] DeleteFrameFiles: successfully deleted file=work/wudata_02.ckp
[15:09:31] - Created dyn
[15:09:31] - Files status OK
[15:09:31] - Expanded 64638 -> 341507 (decompressed 528.3 percent)
[15:09:31] Called DecompressByteArray: compressed_data_size=64638 data_size=341507, decompressed_data_size=341507 diff=0
[15:09:31] - Digital signature verified
[15:09:31]
[15:09:31] Project: 5784 (Run 0, Clone 35, Gen 54)
[15:09:31]
[15:09:31] Assembly optimizations on if available.
[15:09:31] Entering M.D.
[15:09:37] Tpr hash work/wudata_02.tpr: 2188830699 1211495396 3434354702 1536426367 2365523595
[15:09:37]
[15:09:37] Calling fah_main args: 14 usage=100
[15:09:37]
[15:09:37] Working on Gromacs Runs One Microsecond At Cannonball Speeds
[15:09:38] mdrun_gpu returned
[15:09:38] Going to send back what have done -- stepsTotalG=0
[15:09:38] Work fraction=0.0000 steps=0.
[15:09:42] logfile size=0 infoLength=0 edr=0 trr=25
[15:09:42] + Opened results file
[15:09:42] - Writing 637 bytes of core data to disk...
[15:09:42] Done: 125 -> 124 (compressed to 99.2 percent)
[15:09:42] ... Done.
[15:09:42] DeleteFrameFiles: successfully deleted file=work/wudata_02.ckp
[15:09:42]
[15:09:42] Folding@home Core Shutdown: UNSTABLE_MACHINE
[15:09:45] CoreStatus = 7A (122)
[15:09:45] Sending work to server
[15:09:45] Project: 5784 (Run 0, Clone 35, Gen 54)
[15:09:45] - Read packet limit of 540015616... Set to 524286976.


[15:09:45] + Attempting to send results [March 7 15:09:45 UTC]
[15:09:45] + Results successfully sent
[15:09:45] Thank you for your contribution to Folding@Home.
[15:09:49] - Preparing to get new work unit...
[15:09:49] + Attempting to get work packet
[15:09:49] - Connecting to assignment server
[15:09:49] - Successful: assigned to (171.67.108.21).
[15:09:49] + News From Folding@Home: Welcome to Folding@Home
[15:09:50] Loaded queue successfully.
[15:09:50] + Closed connections
[15:09:55]
[15:09:55] + Processing work unit
[15:09:55] Core required: FahCore_11.exe
[15:09:55] Core found.
[15:09:55] Working on queue slot 03 [March 7 15:09:55 UTC]
[15:09:55] + Working ...
[15:09:56]
[15:09:56] *------------------------------*
[15:09:56] Folding@Home GPU Core
[15:09:56] Version 1.31 (Tue Sep 15 10:57:42 PDT 2009)
[15:09:56]
[15:09:56] Compiler : Microsoft (R) 32-bit C/C++ Optimizing Compiler Version 14.00.50727.762 for 80x86
[15:09:56] Build host: amoeba
[15:09:56] Board Type: Nvidia
[15:09:56] Core :
[15:09:56] Preparing to commence simulation
[15:09:56] - Looking at optimizations...
[15:09:56] DeleteFrameFiles: successfully deleted file=work/wudata_03.ckp
[15:09:56] - Created dyn
[15:09:56] - Files status OK
[15:09:56] - Expanded 64984 -> 343707 (decompressed 528.9 percent)
[15:09:56] Called DecompressByteArray: compressed_data_size=64984 data_size=343707, decompressed_data_size=343707 diff=0
[15:09:56] - Digital signature verified
[15:09:56]
[15:09:56] Project: 5783 (Run 3, Clone 86, Gen 79)
[15:09:56]
[15:09:56] Assembly optimizations on if available.
[15:09:56] Entering M.D.
[15:10:02] Tpr hash work/wudata_03.tpr: 3573632084 1845282362 545884572 1506128375 2086401571
[15:10:02]
[15:10:02] Calling fah_main args: 14 usage=100
[15:10:02]
[15:10:02] Working on GROwing Monsters And Cloning Shrimps
[15:10:03] mdrun_gpu returned
[15:10:03] Self-test failure
[15:10:03]
[15:10:03] Folding@home Core Shutdown: UNSTABLE_MACHINE
[15:10:06] CoreStatus = 7A (122)
[15:10:06] Sending work to server
[15:10:06] Project: 5783 (Run 3, Clone 86, Gen 79)
[15:10:06] - Read packet limit of 540015616... Set to 524286976.
[15:10:06] - Error: Could not get length of results file work/wuresults_03.dat
[15:10:06] - Error: Could not read unit 03 file. Removing from queue.
[15:10:06] - Preparing to get new work unit...
[15:10:06] + Attempting to get work packet
[15:10:06] - Connecting to assignment server
[15:10:06] - Successful: assigned to (171.67.108.21).
[15:10:06] + News From Folding@Home: Welcome to Folding@Home
[15:10:06] Loaded queue successfully.
[15:10:07] + Closed connections
Click to expand...
 
Go to the shortcut properties, select compatibility and check the box that says run as an administrator.
 
Just an update... card was bad. Got new card today, put on dummy plug, clocked it to 650mhz, followed Stanfords multi gpu setup and its been chugging away since.
 
You must log in or register to reply here.
Back
Top