nvidia Tmax and Twait fault?

It seems that a lot of GPU problems revolve around specific versions of drivers. Though NVidia has their own support structure, you can often learn from information reported by others who fold.

Moderators: Site Moderators, FAHC Science Team

Post Reply
affende
Posts: 4
Joined: Tue Feb 07, 2017 6:37 pm

nvidia Tmax and Twait fault?

Post by affende »

Nvidia Quadro 600
drivers 311.66

the GPU is folding rather slowly. 32m 44s estimated TPF

at the bottom of the log tmax and twait are mentioned. Are these settings adjustable?

Code: Select all

*********************** Log Started 2017-02-07T18:17:27Z ***********************
18:18:14:WU01:FS01:Connecting to 171.67.108.45:8080
18:18:15:WU01:FS01:Connecting to 171.67.108.45:80
18:18:15:WU01:FS01:Assigned to work server 171.67.108.157
18:18:15:WU01:FS01:Requesting new work unit for slot 01: READY gpu:0:GF108 [Quadro 600] from 171.67.108.157
18:18:15:WU01:FS01:Connecting to 171.67.108.157:8080
18:18:15:WU01:FS01:Downloading 5.16MiB
18:18:18:WU01:FS01:Download complete
18:18:19:WU01:FS01:Received Unit: id:01 state:DOWNLOAD error:NO_ERROR project:9415 run:139 clone:0 gen:15 core:0x21 unit:0x00000015ab436c9d585e06caa0ef97f3
18:18:19:WU01:FS01:Downloading core from http://fahwebx.stanford.edu/cores/Win32/AMD64/NVIDIA/Fermi/Core_21.fah
18:18:19:WU01:FS01:Connecting to fahwebx.stanford.edu:80
18:18:19:WU01:FS01:FahCore 21: Downloading 3.47MiB
18:18:20:WU01:FS01:FahCore 21: Download complete
18:18:21:WU01:FS01:Valid core signature
18:18:21:WU01:FS01:Unpacked 11.80MiB to cores/fahwebx.stanford.edu/cores/Win32/AMD64/NVIDIA/Fermi/Core_21.fah/FahCore_21.exe
18:18:21:WU01:FS01:Starting
18:18:21:WU01:FS01:Running FahCore: "C:\Program Files (x86)\FAHClient/FAHCoreWrapper.exe" C:/Users/user/AppData/Roaming/FAHClient/cores/fahwebx.stanford.edu/cores/Win32/AMD64/NVIDIA/Fermi/Core_21.fah/FahCore_21.exe -dir 01 -suffix 01 -version 704 -lifeline 5572 -checkpoint 15 -gpu 0 -gpu-vendor nvidia
18:18:21:WU01:FS01:Started FahCore on PID 6924
18:18:21:WU01:FS01:Core PID:6280
18:18:21:WU01:FS01:FahCore 0x21 started
18:18:22:WU01:FS01:0x21:*********************** Log Started 2017-02-07T18:18:22Z ***********************
18:18:22:WU01:FS01:0x21:Project: 9415 (Run 139, Clone 0, Gen 15)
18:18:22:WU01:FS01:0x21:Unit: 0x00000015ab436c9d585e06caa0ef97f3
18:18:22:WU01:FS01:0x21:CPU: 0x00000000000000000000000000000000
18:18:22:WU01:FS01:0x21:Machine: 1
18:18:22:WU01:FS01:0x21:Reading tar file core.xml
18:18:22:WU01:FS01:0x21:Reading tar file integrator.xml
18:18:22:WU01:FS01:0x21:Reading tar file state.xml
18:18:22:WU01:FS01:0x21:Reading tar file system.xml
18:18:22:WU01:FS01:0x21:Digital signatures verified
18:18:22:WU01:FS01:0x21:Folding@home GPU Core21 Folding@home Core
18:18:22:WU01:FS01:0x21:Version 0.0.18
18:18:27:WU01:FS01:0x21:Completed 0 out of 6250000 steps (0%)
18:18:27:WU01:FS01:0x21:Temperature control disabled. Requirements: single Nvidia GPU, tmax must be < 110 and twait >= 900
18:51:11:WU01:FS01:0x21:Completed 62500 out of 6250000 steps (1%)
I am just now back to folding after taking some time off .... so im starting fresh.

Thanks.
bruce
Posts: 20910
Joined: Thu Nov 29, 2007 10:13 pm
Location: So. Cal.

Re: nvidia Tmax and Twait fault?

Post by bruce »

FAHCore_21 does attempt to manage temperature in limited situations, and yes, the values can be set.

Personally, I don't recommend it. Turning off a hot GPU and waiting twait seconds is not the best way to handle a overheated GPU. Most GPU utilities allow you to manage settings for fan speed, clock-rate, power, etc. Adjusting those settings to maximize a uniform throughput that's limited by your system's cooling limitations would be much easier on your GPU than cycling it off and on.
affende
Posts: 4
Joined: Tue Feb 07, 2017 6:37 pm

Re: nvidia Tmax and Twait fault?

Post by affende »

bruce wrote:FAHCore_21 does attempt to manage temperature in limited situations, and yes, the values can be set.

Personally, I don't recommend it. Turning off a hot GPU and waiting twait seconds is not the best way to handle a overheated GPU. Most GPU utilities allow you to manage settings for fan speed, clock-rate, power, etc. Adjusting those settings to maximize a uniform throughput that's limited by your system's cooling limitations would be much easier on your GPU than cycling it off and on.

understood. thanks for the quick solution!
Joe_H
Site Admin
Posts: 7856
Joined: Tue Apr 21, 2009 4:41 pm
Hardware configuration: Mac Pro 2.8 quad 12 GB smp4
MacBook Pro 2.9 i7 8 GB smp2
Location: W. MA

Re: nvidia Tmax and Twait fault?

Post by Joe_H »

Your Quadro 600 is similar to the Geforce GT 430 or 440 with just 96 shaders and a reduced core clock. That is at about the bottom end of GPU's able to process current WU's within deadlines, so the 32 min TPF might just be about right depending on the project.
Image

iMac 2.8 i7 12 GB smp8, Mac Pro 2.8 quad 12 GB smp6
MacBook Pro 2.9 i7 8 GB smp3
Post Reply