News:

Follow BoincTasks on Twitter Facebook        Visit our website here.
BoincTasks cloud login is working again

Main Menu
Menu

Show posts

This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.

Show posts Menu

Messages - book_wyrm

#1
I've just tested v4.00 on 2 Milkyway GPU tasks and it's worked flawlessly with 'Connect with BOINC Client' checked.
I think you've solved my problem, thank you :D
#2
I'll look forward to the future versions. I'll make sure to test them to see whether the milkyway problem still affects me or not.
Also, do you think that this is specifically a TThrottle problem? What are the chances of this being a milkyway problem? I have now gone through 3 einstein@home tasks and temperature control works perfectly fine with those as well.
#3
I've tried the new 3.60 version with 2 milkyway tasks and the same problems still persist. I think it could just simply be a combination of my hardware, software and milkyway@home if it doesn't seem to be happening to anyone else. I can't see anyone else on the MW forums having this trouble. I'm currently trying Einstein@home and so far it seems that TThrottle is working properly.
#4
Tried a few Milkyway WU. It seems running in Admin mode has no effect. I've tried a few units with 'connect with boinc client' checked and a few units with it unchecked.
#5
The stable version 6.10.60
#6
It dropped to 60 when I snooze the GPU. I've set TThrottle to have a max temp of 75.

Edit: After unsnoozing the GPU the MW task that was running restarted and gpu started to overheat again. I had to restart TThrottle as usual.
#7
I've PMed the log to you. It's after a TThrottle restart after a high GPU temp.
I'm using the stock cuda application from Milkyway.
#8
Shall I stop posting overheating task information until you get back? Do you have enough information now? Let me know if you need anything else.
#9
CPU:0, GPU:1, PID:4072 (5)   Slot:1   http://milkyway.cs.rpi.edu/milkyway/   de_separation_10_3s_fix20_1_114888_1304776310
Gpu: milkyway_0.52_windows_intelx86__cuda_opencl.exe, PID: 4072, Threads: 5

'Connect with BOINC client' was checked when task switched
#10
The task in the last post had a compute error but it may be of interest to note that after running 721.57 seconds in real time the task only required 6.04 seconds in CPU time.
#11
WU in last post took 3,519.92 seconds in real time and 24.69 in CPU time.

Gpu: milkyway_0.52_windows_intelx86__cuda_opencl.exe, PID: 4068, Threads: 5
CPU:0, GPU:1, PID:4068 (5)   Slot:1   http://milkyway.cs.rpi.edu/milkyway/   de_separation_10_3s_fix20_1_114904_1304776310
'Connect to BOINC task' was unchecked when the tasks switched.
#12
WU in last post took 3,181.23 seconds in real time and 16.75 seconds CPU time

Current WU:
CPU:0, GPU:1, PID:3776 (5)   Slot:1   http://milkyway.cs.rpi.edu/milkyway/   de_separation_10_3s_fix20_1_114911_1304776310
Gpu: milkyway_0.52_windows_intelx86__cuda_opencl.exe, PID: 3776, Threads: 5
'Connect with BOINC client' was checked when task switched. Still needs TThrottle to restart to throttle temps.
#13
The task for the post above took 3,414.82 seconds real time and 25.62 seconds CPU time

My current task at the moment is overheating again.....
Gpu: milkyway_0.52_windows_intelx86__cuda_opencl.exe, PID: 5020, Threads: 5
CPU:0, GPU:1, PID:5020 (5)   Slot:1   http://milkyway.cs.rpi.edu/milkyway/   de_separation_10_3s_fix20_1_114912_1304776310
'Connect with BOINC client' was off this time when the task switched
#14
The task in my last post took 3,131.24 seconds real time and 15.49 CPU time

CPU:0, GPU:1, PID:3332 (5)   Slot:1   http://milkyway.cs.rpi.edu/milkyway/   de_separation_10_3s_fix20_1_114924_1304776310
Gpu: milkyway_0.52_windows_intelx86__cuda_opencl.exe, PID: 3332, Threads: 5
This task initially had "Connect with BOINC client" ticked
#15
Milkyway is my backup project so I've not had an WU from it for a while

The problem still persists, my GPU still overheats until I restart TThrottle. Turning off 'Connect with BOINC client' has no effect anymore so I need to restart TThrottle no matter what settings I have on it.

Here's another task that's overheating:

Gpu: milkyway_0.52_windows_intelx86__cuda_opencl.exe, PID: 4292, Threads: 5
CPU:0, GPU:1, PID:4292 (5)   Slot:1   http://milkyway.cs.rpi.edu/milkyway/   de_separation_10_3s_fix20_1_114925_1304776310

I can now view my tasks on their website and it turns out that one of my tasks took about 18.88 seconds CPU time while it took 3,840.49 in real time with TThrottle running with GPU usage at about 33%
Is this relevant?
Previously, Milkyway was deleting task entries from my account information on their website as they were completed so I could only see what tasks I had running but not tasks that had been completed, until now.