Hi,

Just trying to get this going at the moment, but keep getting this error:

16/09/2008 3:42:13 PM|PS3GRID|Message from server: Full-atom molecular dynamics for Cell processor is not available for your type of computer.

Initially I had installed BOINC as a service, and ran into the issue documented in this forum where CUDA will emulate a GPU (Coprocessor: Device Emulation (CPU) (1)), as BOINC didn't have privileges to access the actual hardware. The error currently in question was also present during this period.

Anyway, a quick non-service BOINC reinstall later and I am still getting it. A quick peek through the google shows it's apparently not a common issue, and nothing really pointed me in the right direction.

So, dear internet, do you think you can help me?

Here is the complete log output:

16/09/2008 3:41:46 PM||Starting BOINC client version 6.3.10 for windows_x86_64
16/09/2008 3:41:46 PM||This a development version of BOINC and may not function properly
16/09/2008 3:41:46 PM||log flags: task, file_xfer, sched_ops
16/09/2008 3:41:46 PM||Libraries: libcurl/7.18.0 OpenSSL/0.9.8g zlib/1.2.3
16/09/2008 3:41:46 PM||Data directory: C:\ProgramData\BOINC
16/09/2008 3:41:46 PM||Running under account davidz
16/09/2008 3:41:47 PM||Processor: 4 GenuineIntel Intel(R) Core(TM)2 Quad CPU Q9300 @ 2.50GHz [Intel64 Family 6 Model 23 Stepping 7]
16/09/2008 3:41:47 PM||Processor features: fpu tsc pae nx sse sse2 pni
16/09/2008 3:41:47 PM||OS: Microsoft Windows Vista: Ultimate x64 Editon, Service Pack 1, (06.00.6001.00)
16/09/2008 3:41:47 PM||Memory: 8.00 GB physical, 16.21 GB virtual
16/09/2008 3:41:47 PM||Disk: 372.61 GB total, 252.53 GB free
16/09/2008 3:41:47 PM||Local time is UTC +10 hours
16/09/2008 3:41:47 PM||Not using a proxy
16/09/2008 3:41:47 PM||CUDA devices found
16/09/2008 3:41:47 PM||Coprocessor: GeForce GTX 260 (2)
16/09/2008 3:41:47 PM|lhcathome|URL: http://lhcathome.cern.ch/lhcathome/; Computer ID: 9725804; location: (none); project prefs: default
16/09/2008 3:41:47 PM|SETI@home|URL: http://setiathome.berkeley.edu/; Computer ID: 4576102; location: home; project prefs: default
16/09/2008 3:41:47 PM|PrimeGrid|URL: http://www.primegrid.com/; Computer ID: 78334; location: (none); project prefs: default
16/09/2008 3:41:47 PM|PS3GRID|URL: http://www.ps3grid.net/; Computer ID: 9091; location: (none); project prefs: default
16/09/2008 3:41:47 PM||General prefs: from PS3GRID (last modified 16-Sep-2008 13:38:39)
16/09/2008 3:41:47 PM||Host location: none
16/09/2008 3:41:47 PM||General prefs: using your defaults
16/09/2008 3:41:47 PM||Reading preferences override file
16/09/2008 3:41:47 PM||Preferences limit memory usage when active to 4094.70MB
16/09/2008 3:41:47 PM||Preferences limit memory usage when idle to 7370.46MB
16/09/2008 3:41:47 PM||Preferences limit disk usage to 9.31GB
16/09/2008 3:41:47 PM|SETI@home|Restarting task 20au08ab.26857.12751.3.8.33_0 using setiathome_enhanced version 603
16/09/2008 3:41:47 PM|PrimeGrid|Restarting task psp_sr2sieve_9180776_0 using psp_sr2sieve version 104
16/09/2008 3:41:47 PM|SETI@home|Restarting task 24au08ag.9956.20522.15.8.194_1 using setiathome_enhanced version 603
16/09/2008 3:41:47 PM|PrimeGrid|Restarting task llrTPS_17141550_1 using llrTPS version 509
16/09/2008 3:42:02 PM|PS3GRID|Sending scheduler request: Requested by user. Requesting 84918 seconds of work, reporting 0 completed tasks
16/09/2008 3:42:13 PM|PS3GRID|Scheduler request completed: got 0 new tasks
16/09/2008 3:42:13 PM|PS3GRID|Message from server: No work sent
16/09/2008 3:42:13 PM|PS3GRID|Message from server: Full-atom molecular dynamics for Cell processor is not available for your type of computer.
16/09/2008 3:43:57 PM|PrimeGrid|Computation for task psp_sr2sieve_9180776_0 finished
16/09/2008 3:43:57 PM|PrimeGrid|Starting gcwsieve_5662123_0
16/09/2008 3:43:57 PM|PrimeGrid|Starting task gcwsieve_5662123_0 using gcwsieve version 105
16/09/2008 3:43:59 PM|PrimeGrid|Started upload of psp_sr2sieve_9180776_0_0
16/09/2008 3:44:03 PM|PrimeGrid|Finished upload of psp_sr2sieve_9180776_0_0
16/09/2008 3:45:29 PM|PrimeGrid|Computation for task llrTPS_17141550_1 finished
16/09/2008 3:45:29 PM|PrimeGrid|Starting psp_sr2sieve_9180713_0
16/09/2008 3:45:29 PM|PrimeGrid|Starting task psp_sr2sieve_9180713_0 using psp_sr2sieve version 104
16/09/2008 3:45:31 PM|PrimeGrid|Started upload of llrTPS_17141550_1_0
16/09/2008 3:45:36 PM|PrimeGrid|Finished upload of llrTPS_17141550_1_0

From you log, you appear to have two GTX260's?

Are they in SLi? GPU grid cannot run them if they are in SLi mode.

---

From you log, you appear to have two GTX260's?

Are they in SLi? GPU grid cannot run them if they are in SLi mode.

They are not in SLI mode. I'm running on three monitors and I don't have an SLI capable motherboard, so I would be surprised if I was mistaken.

Same problem here!

I don't change anything on my computer since almost 2 week.

Jim PROFIT

Di 16 Sep 2008 08:00:08 CEST|PS3GRID|Sending scheduler request: To report completed tasks. Requesting 245680 seconds of work, reporting 1 completed tasks
Di 16 Sep 2008 08:00:13 CEST|PS3GRID|Scheduler request completed: got 0 new tasks
Di 16 Sep 2008 08:00:13 CEST|PS3GRID|Message from server: No work sent
Di 16 Sep 2008 08:00:13 CEST|PS3GRID|Message from server: Full-atom molecular dynamics for Cell processor is not available for your type of computer.

You're not the only ones :-( I think this will be shown to everybody who is crunching on the GPU... :-/

Di 16 Sep 2008 08:00:08 CEST|PS3GRID|Sending scheduler request: To report completed tasks. Requesting 245680 seconds of work, reporting 1 completed tasks
Di 16 Sep 2008 08:00:13 CEST|PS3GRID|Scheduler request completed: got 0 new tasks
Di 16 Sep 2008 08:00:13 CEST|PS3GRID|Message from server: No work sent
Di 16 Sep 2008 08:00:13 CEST|PS3GRID|Message from server: Full-atom molecular dynamics for Cell processor is not available for your type of computer.

You're not the only ones :-( I think this will be shown to everybody who is crunching on the GPU... :-/

Work run out. New workunits availabl.e

gdf

Di 16 Sep 2008 08:00:08 CEST|PS3GRID|Sending scheduler request: To report completed tasks. Requesting 245680 seconds of work, reporting 1 completed tasks
Di 16 Sep 2008 08:00:13 CEST|PS3GRID|Scheduler request completed: got 0 new tasks
Di 16 Sep 2008 08:00:13 CEST|PS3GRID|Message from server: No work sent
Di 16 Sep 2008 08:00:13 CEST|PS3GRID|Message from server: Full-atom molecular dynamics for Cell processor is not available for your type of computer.

You're not the only ones :-( I think this will be shown to everybody who is crunching on the GPU... :-/

Work run out. New workunits availabl.e

gdf

What GDF, did not say is,
The message about "full atom" is for the other application which is for the PS3. The scheduler when it runs out of tasks for your type of computer, the GPU, still has work available for the PS3, which will not run on the GPU, so it sends this message, meaning there are some types of work, but none that will run on your computer. This happens also for linux vs windows. I get these types of messages frequently on other projects that have multiple apps, where in my account I have selected to only run one or two of those. That app runs out of work but the others still have work.

...The scheduler when it runs out of tasks for your type of computer, the GPU, still has work available for the PS3, which will not run on the GPU, so it sends this message, meaning there are some types of work, but none that will run on your computer. This happens also for linux vs windows.

That is good to know. I was getting these about 12 hours ago on my Win 32 Nvidia GPU machine. I kept trying several times and finally did download a work unit (although for the other reasons stated on these boards, it ran more slowly than the previous Windows GPU version).

One troublesome thing about the "No new work" message is that the scheduler backs off a full 24 hours. That seems excessive to me, when new work might be put up much sooner than that. Seems like a waste of time to be forced to wait that long, imho! ;)

One troublesome thing about the "No new work" message is that the scheduler backs off a full 24 hours.

Yes, especially with such a short-deadline-project like GPU-Grid 24h is too long. Is this a server setting or a global setting in the BOINC software?

MrS

---

Scanning for our furry friends since Jan 2002