Advanced search

Message boards : Graphics cards (GPUs) : Can't get any WU's

Author Message
Ichijin
Send message
Joined: 23 Jan 09
Posts: 2
Credit: 0
RAC: 0
Level

Scientific publications
wat
Message 6005 - Posted: 25 Jan 2009 | 19:12:59 UTC

I have a NVS 140M and here is the error

1/25/2009 2:07:04 PM||Starting BOINC client version 6.4.5 for windows_intelx86
1/25/2009 2:07:04 PM||log flags: task, file_xfer, sched_ops
1/25/2009 2:07:04 PM||Libraries: libcurl/7.19.0 OpenSSL/0.9.8i zlib/1.2.3
1/25/2009 2:07:04 PM||Data directory: C:\Documents and Settings\All Users\Application Data\BOINC
1/25/2009 2:07:04 PM||Running under account Ichijin
1/25/2009 2:07:04 PM||Processor: 2 GenuineIntel Intel(R) Core(TM)2 Duo CPU T7300 @ 2.00GHz [x86 Family 6 Model 15 Stepping 10]
1/25/2009 2:07:04 PM||Processor features: fpu tsc sse sse2 mmx
1/25/2009 2:07:04 PM||OS: Microsoft Windows XP: Professional x86 Editon, Service Pack 3, (05.01.2600.00)
1/25/2009 2:07:04 PM||Memory: 1.97 GB physical, 1.82 GB virtual
1/25/2009 2:07:04 PM||Disk: 15.42 GB total, 2.32 GB free
1/25/2009 2:07:04 PM||Local time is UTC -5 hours
1/25/2009 2:07:04 PM||Not using a proxy
1/25/2009 2:07:04 PM||CUDA devices found
1/25/2009 2:07:04 PM||Coprocessor: Quadro NVS 140M (1)
1/25/2009 2:07:04 PM|GPUGRID|URL: http://www.gpugrid.net/; Computer ID: 23957; location: (none); project prefs: default
1/25/2009 2:07:04 PM||No general preferences found - using BOINC defaults
1/25/2009 2:07:04 PM||Reading preferences override file
1/25/2009 2:07:04 PM||Preferences limit memory usage when active to 1007.11MB
1/25/2009 2:07:04 PM||Preferences limit memory usage when idle to 1812.80MB
1/25/2009 2:07:04 PM||Preferences limit disk usage to 1.00GB
1/25/2009 2:07:46 PM||No general preferences found - using BOINC defaults
1/25/2009 2:07:46 PM||Reading preferences override file
1/25/2009 2:07:46 PM||Preferences limit memory usage when active to 1007.11MB
1/25/2009 2:07:46 PM||Preferences limit memory usage when idle to 1812.80MB
1/25/2009 2:07:46 PM||Preferences limit disk usage to 1.00GB
1/25/2009 2:08:14 PM|GPUGRID|Sending scheduler request: Requested by user. Requesting 60480 seconds of work, reporting 0 completed tasks
1/25/2009 2:08:19 PM|GPUGRID|Scheduler request completed: got 0 new tasks
1/25/2009 2:08:19 PM|GPUGRID|Message from server: No work sent
1/25/2009 2:08:19 PM|GPUGRID|Message from server: Full-atom molecular dynamics for Cell processor is not available for your type of computer.
1/25/2009 2:08:19 PM|GPUGRID|Message from server: Full-atom molecular dynamics on Cell processor is not available for your type of computer.
1/25/2009 2:08:19 PM|GPUGRID|Message from server: Full-atom molecular dynamics is not available for your type of computer.
1/25/2009 2:08:27 PM|GPUGRID|Resetting project
1/25/2009 2:08:34 PM|GPUGRID|Sending scheduler request: Requested by user. Requesting 60480 seconds of work, reporting 0 completed tasks
1/25/2009 2:08:39 PM|GPUGRID|Scheduler request completed: got 0 new tasks
1/25/2009 2:08:39 PM|GPUGRID|Message from server: Not sending work - last request too recent: 20 sec
1/25/2009 2:08:41 PM|GPUGRID|Started download of logops3grid.png
1/25/2009 2:08:41 PM|GPUGRID|Started download of project_1.png
1/25/2009 2:08:42 PM|GPUGRID|Finished download of logops3grid.png
1/25/2009 2:08:42 PM|GPUGRID|Started download of project_2.png
1/25/2009 2:08:44 PM|GPUGRID|Finished download of project_1.png
1/25/2009 2:08:44 PM|GPUGRID|Started download of project_3.png
1/25/2009 2:08:45 PM|GPUGRID|Finished download of project_2.png
1/25/2009 2:08:46 PM|GPUGRID|Finished download of project_3.png
1/25/2009 2:08:50 PM|GPUGRID|Sending scheduler request: Requested by user. Requesting 60480 seconds of work, reporting 0 completed tasks
1/25/2009 2:08:55 PM|GPUGRID|Scheduler request completed: got 0 new tasks
1/25/2009 2:08:55 PM|GPUGRID|Message from server: Not sending work - last request too recent: 16 sec
1/25/2009 2:09:55 PM|GPUGRID|Sending scheduler request: To fetch work. Requesting 60480 seconds of work, reporting 0 completed tasks
1/25/2009 2:10:00 PM|GPUGRID|Scheduler request completed: got 0 new tasks
1/25/2009 2:10:00 PM|GPUGRID|Message from server: No work sent
1/25/2009 2:10:00 PM|GPUGRID|Message from server: Full-atom molecular dynamics on Cell processor is not available for your type of computer.
1/25/2009 2:10:00 PM|GPUGRID|Message from server: Full-atom molecular dynamics is not available for your type of computer.
1/25/2009 2:10:00 PM|GPUGRID|Message from server: Full-atom molecular dynamics for Cell processor is not available for your type of computer.

Thanks

Scott Brown
Send message
Joined: 21 Oct 08
Posts: 144
Credit: 2,973,555
RAC: 0
Level
Ala
Scientific publications
watwatwatwatwatwat
Message 6006 - Posted: 25 Jan 2009 | 19:49:22 UTC - in response to Message 6005.

While this does not answer your question, the NVS 140M is a 16 shader card (based on the GeForce 8400M GS with G86M processor), and is therefore not fast enough to complete any of the work at GPUGRID within the 4-day deadline.

As for why the card is experiencing download issues, perhaps you need to update the driver? What driver version are you using?

Ichijin
Send message
Joined: 23 Jan 09
Posts: 2
Credit: 0
RAC: 0
Level

Scientific publications
wat
Message 6007 - Posted: 25 Jan 2009 | 19:53:28 UTC

I'm using the most recent lenovo driver, so i can't really tell what version it is compared to the nvidia ones. I know it supports Cuda because BOINC sees it and I ran Folding@home-gpu with it

Profile Bender10
Avatar
Send message
Joined: 3 Dec 07
Posts: 167
Credit: 8,368,897
RAC: 0
Level
Ser
Scientific publications
watwatwatwatwatwatwat
Message 6012 - Posted: 25 Jan 2009 | 23:02:16 UTC - in response to Message 6007.
Last modified: 25 Jan 2009 | 23:05:37 UTC

Yes, It does support CUDA (2.0). But don't confuse running Folding@Home, with your GPU's ability to run on GPUgrid.

You also need to be running an Nvidia Driver of 177.67 or higher (CUDA enabled).

and is therefore not fast enough to complete any of the work at GPUGRID within the 4-day deadline.


This actually sums it up. The Project Wu time limit requirements pretty much exclude your card, because of the processing power required to meet the project Wu 4 day time limit. GPU's with a minimum of 64 Stream processors can finish a Wu inside the 4 day limit (maybe a little faster than that...;)

See this thread:

http://www.gpugrid.net/forum_thread.php?id=316

Your card (as you know) will actually work fine on Folding (and the other GPU projects out there), as they provide a less complex computation model and smaller Wu's (less run time).

I don't like being told 'You can't crunch here..(with your current hardware)...", but that is it in a nutshell. I think you could get your 140M to eventually get some Wu's, But you would not be able to meet the time limit...:(

But there are other GPU projects out there (3 that I know about) with lower hardware requirements. All of which you can run with your current GPU.
____________


Consciousness: That annoying time between naps......

Experience is a wonderful thing: it enables you to recognize a mistake every time you repeat it.

Profile Paul D. Buck
Send message
Joined: 9 Jun 08
Posts: 1050
Credit: 37,321,185
RAC: 0
Level
Val
Scientific publications
watwatwatwatwatwatwatwatwatwat
Message 6015 - Posted: 26 Jan 2009 | 7:09:54 UTC - in response to Message 6012.

But there are other GPU projects out there (3 that I know about) with lower hardware requirements. All of which you can run with your current GPU.


Folding@Home, SETI@Home ...

What are the others?
____________

Profile Edboard
Avatar
Send message
Joined: 24 Sep 08
Posts: 72
Credit: 12,410,275
RAC: 0
Level
Pro
Scientific publications
watwatwatwatwatwatwatwatwatwat
Message 6308 - Posted: 2 Feb 2009 | 14:48:28 UTC - in response to Message 6015.

···Seti@home beta

Profile DoctorNow
Avatar
Send message
Joined: 18 Aug 07
Posts: 83
Credit: 122,995,082
RAC: 0
Level
Cys
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 6310 - Posted: 2 Feb 2009 | 15:32:55 UTC - in response to Message 6015.

What are the others?

It's also possible now to run a new custom-made MilkyWay@Home-app on your GPU, but currently ONLY possible with an ATI-card and a 64-Bit Windows-system.
More details you can read in this thread.
____________
Member of BOINC@Heidelberg and ATA!

Profile Paul D. Buck
Send message
Joined: 9 Jun 08
Posts: 1050
Credit: 37,321,185
RAC: 0
Level
Val
Scientific publications
watwatwatwatwatwatwatwatwatwat
Message 6312 - Posted: 2 Feb 2009 | 15:39:42 UTC - in response to Message 6310.

What are the others?

It's also possible now to run a new custom-made MilkyWay@Home-app on your GPU, but currently ONLY possible with an ATI-card and a 64-Bit Windows-system.
More details you can read in this thread.


Yeah, I am aware of that coming application too ... but it is still pre-alpha and the only ATI cards I have are not of a class that could run it anyway ...

If they get it out the door soon I might just get a couple of the lower end ATI cards that can handle it just for the mean time till they get the Nvidia version done ... and I can hope that the rumors of a new Einstein application are not just smoke also ...

Profile Bender10
Avatar
Send message
Joined: 3 Dec 07
Posts: 167
Credit: 8,368,897
RAC: 0
Level
Ser
Scientific publications
watwatwatwatwatwatwat
Message 6313 - Posted: 2 Feb 2009 | 15:56:08 UTC - in response to Message 6015.
Last modified: 2 Feb 2009 | 16:01:25 UTC

But there are other GPU projects out there (3 that I know about) with lower hardware requirements. All of which you can run with your current GPU.


Folding@Home, SETI@Home ...

What are the others?


The playing field is changing quickly. The 'working' (GPU Cuda) projects I've found (and seti beta is kinda so-so..) are, Folding, Seti-beta, and RC5-Beta. There is also developement on Milkyway (non-cuda) and Einstein (not sure what the status here is...).

The 'working' CUDA projects (ex GPUgrid) will most likely complete work with even the slowest GPU (16-32 SP's), as they are using the less complex CC1.0 verses CC1.1 of GPUgrid.
____________


Consciousness: That annoying time between naps......

Experience is a wonderful thing: it enables you to recognize a mistake every time you repeat it.

Profile Paul D. Buck
Send message
Joined: 9 Jun 08
Posts: 1050
Credit: 37,321,185
RAC: 0
Level
Val
Scientific publications
watwatwatwatwatwatwatwatwatwat
Message 6315 - Posted: 2 Feb 2009 | 17:50:17 UTC - in response to Message 6313.

The playing field is changing quickly. The 'working' (GPU Cuda) projects I've found (and seti beta is kinda so-so..) are, Folding, Seti-beta, and RC5-Beta. There is also developement on Milkyway (non-cuda) and Einstein (not sure what the status here is...).

The 'working' CUDA projects (ex GPUgrid) will most likely complete work with even the slowest GPU (16-32 SP's), as they are using the less complex CC1.0 verses CC1.1 of GPUgrid.


In that I am BOINC heavy and in the past have had less than good luck getting several projects to run at the same time on a machine ... well ... I will just wait till they have BOINC versions.

I know a lot of people are jumping up and down at their favorite project demanding or prodding, what ever you want to call it, for a CUDA version of the applications. Heck, we don't even have BOINC working nicely with the CUDA and non-CUDA workloads.

I know I have dived off the deep end on GPU cards in the last couple weeks but that is just my insanity coming to the fore ... In the mean time I can stand to hang out here and do work for awhile ... when some other projects come along that are more my interest I will be dropping my share here, but, heck, I do almost every project out there ...

Profile Bender10
Avatar
Send message
Joined: 3 Dec 07
Posts: 167
Credit: 8,368,897
RAC: 0
Level
Ser
Scientific publications
watwatwatwatwatwatwat
Message 6326 - Posted: 2 Feb 2009 | 21:26:10 UTC - in response to Message 6315.
Last modified: 2 Feb 2009 | 21:26:38 UTC

As with some of us here, I have been involved in Computer / Electronics for a few decades as part of my work...Hmmm

Heck, we don't even have BOINC working nicely with the CUDA and non-CUDA workloads.


And I've had some fun trying to get different projects to run on my computers. With this new Cuda stuff...I have been able to get a couple of the new Cuda projects (Folding and RC5) to run alongside normal Boinc projects, without too much swearing or smashing of crockery...lol

Folding was easy to get running with Boinc. But I only had 1 GPU in that box. So I could not try it paired with GPUgrid. Maybe later. RC5 took a little coaxing, But I got it to run alongside GPUgrid on a GX2...it just needed some persuading ...hehe
____________


Consciousness: That annoying time between naps......

Experience is a wonderful thing: it enables you to recognize a mistake every time you repeat it.

Post to thread

Message boards : Graphics cards (GPUs) : Can't get any WU's

//