• 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.

Problems with GPU client

M

Mr. Pedantic

[H]ard|Gawd
Joined
Sep 19, 2009
Messages
1,707
I've got a problem with my GPU client, when I turn the computer on. The GPU client defaults to 2D clocks, and I can only get it onto 3D clocks by restarting the service.

Before:

Code: 
--- Opening Log file [January 10 16:32:42 UTC] 


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

                       Folding@Home Client Version 6.23

                          http://folding.stanford.edu

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

Launch directory: C:\PROGRA~1\FAH\GPU
Service: C:\PROGRA~1\FAH\GPU\GPU.exe
Arguments: -svcstart -d C:\PROGRA~1\FAH\GPU -verbosity 9 -local -forceasm 

Launched as a service.
Entered C:\PROGRA~1\FAH\GPU to do work.

Warning:
 By using the -forceasm flag, you are overriding
 safeguards in the program. If you did not intend to
 do this, please restart the program without -forceasm.
 If work units are not completing fully (and particularly
 if your machine is overclocked), then please discontinue
 use of the flag.

[16:32:42] - Ask before connecting: No
[16:32:42] - User name: Mr._Pedantic (Team 33)
[16:32:42] - User ID: 508DC76E16C08206
[16:32:42] - Machine ID: 2
[16:32:42] 
[16:32:42] Loaded queue successfully.
[16:32:43] 
[16:32:43] + Processing work unit
[16:32:43] - Autosending finished units... [January 10 16:32:43 UTC]
[16:32:43] Core required: FahCore_11.exe[16:32:43] Trying to send all finished work units

[16:32:43] + No unsent completed units remaining.
[16:32:43] - Autosend completed
[16:32:43] Core found.
[16:32:44] Working on queue slot 07 [January 10 16:32:44 UTC]
[16:32:44] + Working ...
[16:32:44] - Calling '.\FahCore_11.exe -dir work/ -suffix 07 -priority 96 -checkpoint 3 -service -forceasm -verbose -lifeline 2644 -version 623'

[16:32:44] 
[16:32:44] *------------------------------*
[16:32:44] Folding@Home GPU Core - Beta
[16:32:44] Version 1.24 (Mon Feb 9 11:00:12 PST 2009)
[16:32:44] 
[16:32:44] Compiler  : Microsoft (R) 32-bit C/C++ Optimizing Compiler Version 14.00.50727.762 for 80x86 
[16:32:44] Build host: amoeba
[16:32:44] Board Type: AMD
[16:32:44] Core      : 
[16:32:44] Preparing to commence simulation
[16:32:44] - Assembly optimizations manually forced on.
[16:32:44] - Not checking prior termination.
[16:32:44] - Expanded 98567 -> 492188 (decompressed 499.3 percent)
[16:32:44] Called DecompressByteArray: compressed_data_size=98567 data_size=492188, decompressed_data_size=492188 diff=0
[16:32:44] - Digital signature verified
[16:32:44] 
[16:32:44] Project: 5732 (Run 1, Clone 23, Gen 584)
[16:32:44] 
[16:32:45] Assembly optimizations on if available.
[16:32:45] Entering M.D.
[16:32:51] Will resume from checkpoint file
[16:32:51] Tpr hash work/wudata_07.tpr:  3148706236 48884035 407693985 3137473516 3103523013
[16:32:54] Working on Protein
[16:32:54] Client config found, loading data.
[16:32:54] Starting GUI Server
[16:32:56] Resuming from checkpoint
[16:32:56] fcCheckPointResume: retreived and current tpr file hash:
[16:32:56]    0   3148706236   3148706236
[16:32:56]    1     48884035     48884035
[16:32:56]    2    407693985    407693985
[16:32:56]    3   3137473516   3137473516
[16:32:56]    4   3103523013   3103523013
[16:32:56] Verified work/wudata_07.log
[16:32:56] Verified work/wudata_07.edr
[16:32:56] Verified work/wudata_07.xtc
[16:32:56] Completed 26%
[16:43:07] Completed 27%
[16:45:16] Service stop request received.
[16:45:16] ***** Got a SIGTERM signal (2)
[16:45:16] Killing all core threads

Folding@Home Client Shutdown.

Note that the frame time is just over 10 minutes - pretty slow. That is about 600-odd PPD.

After restarting the service:

Code: 
--- Opening Log file [January 10 16:45:16 UTC] 


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

                       Folding@Home Client Version 6.23

                          http://folding.stanford.edu

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

Launch directory: C:\PROGRA~1\FAH\GPU
Service: C:\PROGRA~1\FAH\GPU\GPU.exe
Arguments: -svcstart -d C:\PROGRA~1\FAH\GPU -verbosity 9 -local -forceasm 

Launched as a service.
Entered C:\PROGRA~1\FAH\GPU to do work.

Warning:
 By using the -forceasm flag, you are overriding
 safeguards in the program. If you did not intend to
 do this, please restart the program without -forceasm.
 If work units are not completing fully (and particularly
 if your machine is overclocked), then please discontinue
 use of the flag.

[16:45:16] - Ask before connecting: No
[16:45:16] - User name: Mr._Pedantic (Team 33)
[16:45:16] - User ID: 508DC76E16C08206
[16:45:16] - Machine ID: 2
[16:45:16] 
[16:45:16] Loaded queue successfully.
[16:45:16] 
[16:45:16] + Processing work unit
[16:45:16] - Autosending finished units... [16:45:16]
[16:45:16] Core required: FahCore_11.exe
[16:45:16] Trying to send all finished work units
[16:45:16] + No unsent completed units remaining.
[16:45:16] - Autosend completed
[16:45:16] Core found.
[16:45:16] Working on queue slot 07 [January 10 16:45:16 UTC]
[16:45:16] + Working ...
[16:45:16] - Calling '.\FahCore_11.exe -dir work/ -suffix 07 -priority 96 -checkpoint 3 -service -forceasm -verbose -lifeline 3012 -version 623'

[16:45:17] 
[16:45:17] *------------------------------*
[16:45:17] Folding@Home GPU Core - Beta
[16:45:17] Version 1.24 (Mon Feb 9 11:00:12 PST 2009)
[16:45:17] 
[16:45:17] Compiler  : Microsoft (R) 32-bit C/C++ Optimizing Compiler Version 14.00.50727.762 for 80x86 
[16:45:17] Build host: amoeba
[16:45:17] Board Type: AMD
[16:45:17] Core      : 
[16:45:17] Preparing to commence simulation
[16:45:17] - Assembly optimizations manually forced on.
[16:45:17] - Not checking prior termination.
[16:45:17] - Expanded 98567 -> 492188 (decompressed 499.3 percent)
[16:45:17] Called DecompressByteArray: compressed_data_size=98567 data_size=492188, decompressed_data_size=492188 diff=0
[16:45:17] - Digital signature verified
[16:45:17] 
[16:45:17] Project: 5732 (Run 1, Clone 23, Gen 584)
[16:45:17] 
[16:45:17] Assembly optimizations on if available.
[16:45:17] Entering M.D.
[16:45:23] Will resume from checkpoint file
[16:45:23] Tpr hash work/wudata_07.tpr:  3148706236 48884035 407693985 3137473516 3103523013
[16:45:23] Working on Protein
[16:45:23] Client config found, loading data.
[16:45:23] Starting GUI Server
[16:45:25] Resuming from checkpoint
[16:45:25] fcCheckPointResume: retreived and current tpr file hash:
[16:45:25]    0   3148706236   3148706236
[16:45:25]    1     48884035     48884035
[16:45:25]    2    407693985    407693985
[16:45:25]    3   3137473516   3137473516
[16:45:25]    4   3103523013   3103523013
[16:45:25] Verified work/wudata_07.log
[16:45:25] Verified work/wudata_07.edr
[16:45:25] Verified work/wudata_07.xtc
[16:45:25] Completed 27%
[16:48:25] Completed 28%
[16:51:25] Completed 29%
[16:54:26] Completed 30%
[16:58:33] Completed 31%

Frame times are now about 3 minutes - about 2500 PPD.

I can manually restart the service for now, seeing as it's university summer break, but when I get back to uni I would like to be able to schedule a wake-up from the BIOS (which I've already done) during the day and have it run with the GPU client using 3D clocks instead of 2D. I've tried deleting and reinstalling the GPU client, to no avail. I could create a batch file and schedule a task to restart the service, but that is far from ideal and I'd like to hear if you guys had any solutions for this.

Thanks.
 
I think that stanford advises not to run the GPU client as a service. I think service support is experimental at best.

I've had more luck starting it as a scheduled task and then using a small utility like tray it to minimize and hide it.

 
This is XP, not Vista. As far as I'm aware, the service works perfectly well in XP, and it has up until about a week ago.
 
You must log in or register to reply here.
Back
Top