Folding Forum

OK today I changed out GPU in my desktop I was running a GTS250



and I installed a

GT 520

I uninstalled all drivers / software prior do doing this Then I installed 301.42, which was the same driver that I had installed for the 250. (fresh DL BTW) Now my GPU client is not running. I have tried uninstalling F@H and re-installing it. It is still hung up and I am not sure where to go from here. I also can not post links to my cards yet. Let me know if you need them.

I see that I can not attach a file (understandable) So I will paste the last log entry here:


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


[17:46:05] - Ask before connecting: No
[17:46:05] - User name: DaveQual (Team 178579)
[17:46:05] - User ID: 6313976A264ECCA2
[17:46:05] - Machine ID: 3
[17:46:05]
[17:46:05] Gpu type=3 species=21.
[17:46:05] Loaded queue successfully.
[17:46:05] Initialization complete
[17:46:05]
[17:46:05] + Processing work unit
[17:46:05] Core required: FahCore_11.exe
[17:46:05] Core found.
[17:46:05] Working on queue slot 02 [July 23 17:46:05 UTC]
[17:46:05] + Working ...
[17:46:06]
[17:46:06] *------------------------------*
[17:46:06] Folding@Home GPU Core
[17:46:06] Version 1.31 (Tue Sep 15 10:57:42 PDT 2009)
[17:46:06]
[17:46:06] Compiler : Microsoft (R) 32-bit C/C++ Optimizing Compiler Version 14.00.50727.762 for 80x86
[17:46:06] Build host: amoeba
[17:46:06] Board Type: Nvidia
[17:46:06] Core :
[17:46:06] Preparing to commence simulation
[17:46:06] - Ensuring status. Please wait.
[17:46:06] - Looking at optimizations...
[17:46:06] - Working with standard loops on this execution.
[17:46:06] - Previous termination of core was improper.
[17:46:06] - Going to use standard loops.
[17:46:06] - Files status OK
[17:46:06] - Expanded 46751 -> 252912 (decompressed 540.9 percent)
[17:46:06] Called DecompressByteArray: compressed_data_size=46751 data_size=252912, decompressed_data_size=252912 diff=0
[17:46:06] - Digital signature verified
[17:46:06]
[17:46:06] Project: 5766 (Run 1, Clone 372, Gen 2265)
[17:46:06]
[17:46:06] Entering M.D.
[17:46:06] Will resume from checkpoint file
[17:46:06] Tpr hash work/wudata_02.tpr: 998946706 4084355376 4143226230 3090724814 4160619262
[17:46:06]
[17:46:06] Calling fah_main args: 14 usage=100
[17:46:06]
[17:47:07] Gpu type=3 species=21.

Hello DaveQual, welcome to the forum.

If you installed the new card in the middle of a work unit, the old work unit may not work with the new card.

Stop the client. Deleted the Work folder, the queue.dat file, and the unitinfo.txt file. Start it up again... fixed?

You also need to be running v6.41 client version, or higher.

7im I have deleted every trace of F@h that I can find EXCEPT for my CPU. and it is still acting up.

The drivers look fine for fermi, which gpu version are you using? I just installed GTX560 on that driver version in 6.41 and then in v7 and both worked well.

I installed 301.42 on suggestion from Nvida's site. Which I believe was the same as I had on the 250 GTS.

DaveQual wrote:I installed 301.42 on suggestion from Nvida's site. Which I believe was the same as I had on the 250 GTS.

Sorry I meant GPU client version the driver version should be good.

I just dug through over 1000 files and found a trace of F@H. I am about 95% I have it working again.... only time will tell.

I just installed GPU systray 6.41 for the third time today. I think I was running 6.23 prior to the card swap.

Hello sophia12, welcome to the forum.

Deletion is always a last resort. Please confirm why you would need to do this procedure...