• 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.

Early Unit End -GPU beta

M

M1ster_R0gers

[H]ard [H]eaded
Joined
Jun 8, 2000
Messages
1,525
I installed the appropriate catalyist drivers from the faq and downloaded directx to try this out on my main rig but unfortanutely I am getting an EARLY UNIT END error downloading the work file.

Any suggestions from those of you who have more experience?
 
M1ster_R0gers said:
I installed the appropriate catalyist drivers from the faq and downloaded directx to try this out on my main rig but unfortanutely I am getting an EARLY UNIT END error downloading the work file.

Any suggestions from those of you who have more experience?
Click to expand...

Post the relevant parts of FAHLOG.TXT. (the stuff for just this work unit)
 
--- Opening Log file [October 2 23:18:22]


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

Folding@Home Client Version 5.90beta

http://folding.stanford.edu

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

Launch directory: C:\Documents and Settings\noah\Desktop\FOLDING
Executable: C:\Documents and Settings\noah\Desktop\FOLDING\FAH5.05beta1-console.exe


[23:18:22] - Ask before connecting: Yes
[23:18:22] - Use IE connection settings: Yes
[23:18:22] - User name: m1ster_r0gers (Team 33)
[23:18:22] - User ID: 5D773BB83C1511C4
[23:18:22] - Machine ID: 1
[23:18:22]
[23:18:22] Loaded queue successfully.
[23:18:22]
[23:18:22] + Processing work unit
[23:18:22] Core required: FahCore_10.exe
[23:18:22] Core found.
[23:18:22] Working on Unit 01 [October 2 23:18:22]
[23:18:22] + Working ...
[23:18:22]
[23:18:22] *------------------------------*
[23:18:22] Folding@Home GPU Core - Beta
[23:18:22] Version 0.05 (Sat Sep 30 16:08:52 PDT 2006)
[23:18:22]
[23:18:22] Compiler : Microsoft (R) 32-bit C/C++ Optimizing Compiler Version 13.10.3077 for 80x86
[23:18:22] Build host: CYGWIN_NT-5.1 vishal-gpu 1.5.19(0.150/4/2) 2006-01-20 13:28 i686 Cygwin
[23:18:22] Preparing to commence simulation
[23:18:22] - Looking at optimizations...
[23:18:22] - Files status OK
[23:18:22] - Expanded 82993 -> 443705 (decompressed 534.6 percent)
[23:18:22]
[23:18:22] Project: 2711 (Run 0, Clone 302, Gen 0)
[23:18:22]
[23:18:22] Assembly optimizations on if available.
[23:18:22] Entering M.D.
[23:18:34] Completed 0
[23:18:34] Starting GUI Server
[23:18:35] mdrun_gpu returned 114
[23:18:35]
[23:18:35] Folding@home Core Shutdown: EARLY_UNIT_END
[23:18:38] CoreStatus = 72 (114)
[23:18:38] Sending work to server
[23:18:38] - Error: Could not get length of results file work/wuresults_01.dat
[23:18:38] - Error: Could not read unit 01 file. Removing from queue.
[23:18:38] - Preparing to get new work unit...
[23:18:38] - Presenting message box asking to network.
[23:18:41] + Attempting to get work packet
[23:18:41] - Connecting to assignment server
[23:18:41] - Successful: assigned to (171.65.103.159).
[23:18:41] + News From Folding@Home: Welcome to our GPU beta.
[23:18:41] Loaded queue successfully.
[23:18:43] + Connections closed: You may now disconnect
[23:18:48]
[23:18:48] + Processing work unit
[23:18:48] Core required: FahCore_10.exe
[23:18:48] Core found.
[23:18:48] Working on Unit 02 [October 2 23:18:48]
[23:18:48] + Working ...
[23:18:48]
[23:18:48] *------------------------------*
[23:18:48] Folding@Home GPU Core - Beta
[23:18:48] Version 0.05 (Sat Sep 30 16:08:52 PDT 2006)
[23:18:48]
[23:18:48] Compiler : Microsoft (R) 32-bit C/C++ Optimizing Compiler Version 13.10.3077 for 80x86
[23:18:48] Build host: CYGWIN_NT-5.1 vishal-gpu 1.5.19(0.150/4/2) 2006-01-20 13:28 i686 Cygwin
[23:18:48] Preparing to commence simulation
[23:18:48] - Looking at optimizations...
[23:18:48] - Created dyn
[23:18:48] - Files status OK
[23:18:48] - Expanded 82993 -> 443705 (decompressed 534.6 percent)
[23:18:48]
[23:18:48] Project: 2711 (Run 0, Clone 302, Gen 0)
[23:18:48]
[23:18:48] Assembly optimizations on if available.
[23:18:48] Entering M.D.
[23:19:00] Completed 0
[23:19:00] Starting GUI Server
[23:19:00] mdrun_gpu returned 114
[23:19:00]
[23:19:00] Folding@home Core Shutdown: EARLY_UNIT_END


It keeps cycling through this continually. I have redone everthing once over. It seems to be getting cut off early during the download
 
Delete the whole folder and reload (to reload core 10). This might help.
 
Just tried that. no-go

I have been trying to find out about this

[23:19:00] mdrun_gpu returned 114

Can anyone verify that is a correct value?
 
Yeah I'm getting that after the 5th completed one now too.

EDIT: don't think 6.5 will work on the x1950 :/

EDIT2: oh wait after looking at the log file, its going to the next unit so nevermind. i think these are just test units maybe for the beta?

[22:19:57] Working on Unit 01 [October 2 22:19:57]
[22:19:57] + Working ...
[22:19:57]
[22:19:57] *------------------------------*
[22:19:57] Folding@Home GPU Core - Beta
[22:19:57] Version 0.05 (Sat Sep 30 16:08:52 PDT 2006)
[22:19:57]
[22:19:57] Compiler : Microsoft (R) 32-bit C/C++ Optimizing Compiler Version 13.10.3077 for 80x86
[22:19:57] Build host: CYGWIN_NT-5.1 vishal-gpu 1.5.19(0.150/4/2) 2006-01-20 13:28 i686 Cygwin
[22:19:57] Preparing to commence simulation
[22:19:57] - Looking at optimizations...
[22:19:57] - Created dyn
[22:19:57] - Files status OK
[22:19:57] - Expanded 83011 -> 443705 (decompressed 534.5 percent)
[22:19:57]
[22:19:57] Project: 2711 (Run 0, Clone 56, Gen 0)
[22:19:57]
[22:19:57] Assembly optimizations on if available.
[22:19:57] Entering M.D.
[22:20:06] Completed 0
[22:20:06] Starting GUI Server
[22:26:28] Completed 1
[22:32:31] Completed 2
[22:38:35] Completed 3
[22:44:38] Completed 4
[23:04:52] Completed 5
[23:09:31] mdrun_gpu returned 114
[23:09:31]
[23:09:31] Folding@home Core Shutdown: EARLY_UNIT_END
[23:09:33] CoreStatus = 72 (114)
[23:09:33] Sending work to server
[23:09:33] - Error: Could not get length of results file work/wuresults_01.dat
[23:09:33] - Error: Could not read unit 01 file. Removing from queue.
[23:09:33] - Preparing to get new work unit...
[23:09:33] + Attempting to get work packet
[23:09:33] - Connecting to assignment server
[23:09:36] - Successful: assigned to (171.65.103.159).
[23:09:36] + News From Folding@Home: Welcome to our GPU beta.
[23:09:36] Loaded queue successfully.
[23:09:39] + Closed connections
[23:09:44]
[23:09:44] + Processing work unit
[23:09:44] Core required: FahCore_10.exe
[23:09:44] Core found.
[23:09:44] Working on Unit 02 [October 2 23:09:44]
[23:09:44] + Working ...
[23:09:44]
[23:09:44] *------------------------------*
[23:09:44] Folding@Home GPU Core - Beta
[23:09:44] Version 0.05 (Sat Sep 30 16:08:52 PDT 2006)
[23:09:44]
[23:09:44] Compiler : Microsoft (R) 32-bit C/C++ Optimizing Compiler Version 13.10.3077 for 80x86
[23:09:44] Build host: CYGWIN_NT-5.1 vishal-gpu 1.5.19(0.150/4/2) 2006-01-20 13:28 i686 Cygwin
[23:09:44] Preparing to commence simulation
[23:09:44] - Looking at optimizations...
[23:09:44] - Created dyn
[23:09:44] - Files status OK
[23:09:44] - Expanded 83011 -> 443705 (decompressed 534.5 percent)
[23:09:44]
[23:09:44] Project: 2711 (Run 0, Clone 56, Gen 0)
[23:09:44]
[23:09:44] Assembly optimizations on if available.
[23:09:44] Entering M.D.
[23:09:54] Completed 0
[23:09:54] Starting GUI Server
[23:15:52] Completed 1
[23:21:51] Completed 2
[23:27:50] Completed 3
[23:33:49] Completed 4
[23:39:48] Completed 5
[23:42:44] mdrun_gpu returned 114
[23:42:44]
[23:42:44] Folding@home Core Shutdown: EARLY_UNIT_END
[23:42:48] CoreStatus = 72 (114)
[23:42:48] Sending work to server
[23:42:48] - Error: Could not get length of results file work/wuresults_02.dat
[23:42:48] - Error: Could not read unit 02 file. Removing from queue.
[23:42:48] - Preparing to get new work unit...
[23:42:48] + Attempting to get work packet
[23:42:48] - Connecting to assignment server
[23:42:49] - Successful: assigned to (171.65.103.159).
[23:42:49] + News From Folding@Home: Welcome to our GPU beta.
[23:42:49] Loaded queue successfully.
[23:42:53] + Closed connections
[23:42:58]
[23:42:58] + Processing work unit
[23:42:58] Core required: FahCore_10.exe
[23:42:58] Core found.
[23:42:58] Working on Unit 03 [October 2 23:42:58]
[23:42:58] + Working ...
[23:42:58]
[23:42:58] *------------------------------*
 
Thanks for everyones help, 6.5 did it. l would take all mentions of 6.1 away in the faq.
 
Just thought i would mention that i got the EUE error running it in 2D mode. After getting the Overclocking setup going @ 1.4V took care of it. It is a shame that Stanford didn't set the GPU client up to automatically switch over to 3D mode.

For some reason i also had to download D3dx9_30.dll.. Even though i had the latest DX9 installed, FAH couldn't find that file.
 
The only way to get 3D clocks would be to run in full-screen mode, and then you couldn't use your machine. The actual issue seems to be a combination of certain WU's with 6.10, not the voltage, but we need more testing. We've asked people to switch to 6.5 for now, at normal and 3D clocks/voltage to see if it's mainly a driver issue.
 
mhouston said:
The only way to get 3D clocks would be to run in full-screen mode, and then you couldn't use your machine. The actual issue seems to be a combination of certain WU's with 6.10, not the voltage, but we need more testing. We've asked people to switch to 6.5 for now, at normal and 3D clocks/voltage to see if it's mainly a driver issue.
Click to expand...
You could also force the 3dclock to be run as 2dclock...

Get AtiTool and disable the ATI Poller, run the tool and adjust the speeds to the 3d clocks of 650/775 for 1900XTX

Here's a how to do - http://www.madshrimps.be/?action=gethowto&number=2&howtopage=161&howtoID=61
 
We use ATITool all the time to set the clocks/voltage/fans manually. What I mean is that there isn't a way for us at FAH to do this easily without going to full-screen or getting ATI to app-detect us to raise the clocks.
 
mhouston said:
We use ATITool all the time to set the clocks/voltage/fans manually. What I mean is that there isn't a way for us at FAH to do this easily without going to full-screen or getting ATI to app-detect us to raise the clocks.
Click to expand...
Oh my bad, now I understand...
 
mhouston said:
We use ATITool all the time to set the clocks/voltage/fans manually. What I mean is that there isn't a way for us at FAH to do this easily without going to full-screen or getting ATI to app-detect us to raise the clocks.
Click to expand...

I suggest since this has made a splash in a press release on the ATI website that they fix this for you. Along those lines are they going to start compatibility testing and incorporating f@h into their driver code?
 
M1ster_R0gers said:
I suggest since this has made a splash in a press release on the ATI website that they fix this for you. Along those lines are they going to start compatibility testing and incorporating f@h into their driver code?
Click to expand...
Seconded..
 
We are in their regressions, but we never ran long enough WUs to trigger the current bug. ATI are in the loop and engineers are staying way late to dig into this. They are engaged, but it's obvious we need to provide them a better regression test...
 
I noticed in the FAQ that they recommend against running the GPU client as a service. What other options do i have besides starting it manually each time?
I would prefer not to use the graphical client because i hear it is slower.
 
Batch script and then put it in your startup folder. I haven't tested this, but it should work. The issue of running as a service is that you own the console and the service runs as a different user, usually SYSTEM, so it doesn't get a GPU context.
 
mhouston said:
We are in their regressions, but we never ran long enough WUs to trigger the current bug. ATI are in the loop and engineers are staying way late to dig into this. They are engaged, but it's obvious we need to provide them a better regression test...
Click to expand...

I am glad to hear all of this. Keep up the good work with all companies involved.
 
mhouston said:
Batch script and then put it in your startup folder. I haven't tested this, but it should work. The issue of running as a service is that you own the console and the service runs as a different user, usually SYSTEM, so it doesn't get a GPU context.
Click to expand...
even nicer would be something in WSH that could be run under the 'task scheduler'

if(gpu client is running)
quit;
else
start_gpu_client();
quit;

I have no idea how to code that though. You could make it check every hour. Even more sophisticated would be a verion that checked for any 3d apps first...
 
You must log in or register to reply here.
Back
Top