Anything special we need to do to get these? I wanted to test few, but all I got was de_separation_14_2s_05_3 and ps_separation_14_2s_null_3_v4, which all error out on my HD3850 with the 0.82 app. Nothing with "test" in it.
____________
.

Had a few test WU on my NV 460SE and these run and valided without a problem.

I'm trying to address the errors people have been having

No problems with this antique that Im running with XP Pro SP 3 Boinc 6.10.58

We sure are getting a LOT of these test_1 jobs.
They all seem to run and validate fine , but that's on a 7970...
I hope these are Real usable results and not only for testing because we have run many hundreds of them. :-)

test_1 work units validate OK.
They validate even with the old good 0.91 sse3 intel optimized app. and with ati-CAL on the other end. Thats good. CPU-opti-app. vs GPU-opti-app. validation is tougher. :)

see?
http://milkyway.cs.rpi.edu/milkyway/workunit.php?wuid=179016095

I'm trying to address the errors people have been having

... with HD3850 (AGP) cards.

ps_separation_14_2s_05_3_test_1_etc

XP Pro SP3 (32 bit) all OK.
W7 Pro SP1 (64 bit) all fail.

To clarify, from yesterday, in a different thread.
XP & W7 drivers:-
88.81 / 11.8
Catalyst Control Center Version
XP 2011.0728.1723.29300 (works) BOINC 6.12.34
W7 2011.0728.1756.30366 (later - doesn't work) BOINC 6.12.34
W7 PC has been cold-booted, XP PC has not i.e., still running since yesterday.

So these new workunits are still all failing for you on 64 bit systems?

... with HD3850 (AGP) cards.

ps_separation_14_2s_05_3_test_1_etc

XP Pro SP3 (32 bit) all OK.
W7 Pro SP1 (64 bit) all fail.

Just tested some de_separation_14_2s_05_3_test_1_rand_* tasks on my ATI HD3850 (Win7 x64, ATI CAL application v0.82), all ended with exit code -1073740940 (0xc0000374).

All tasks for computer 293662
____________
.

I'm trying to address the errors people have been having

... with HD3850 (AGP) cards.

ps_separation_14_2s_05_3_test_1_etc

So these new workunits are still all failing for you on 64 bit systems?

Drove 16km with bycicle to the HD3850AGP XP32 MW-Machine to test again (not only this Project needed changes ;)). But still computing until 100% and then nothing happens (no End).

One of these i must abort again then:

http://milkyway.cs.rpi.edu/milkyway/result.php?resultid=227559812
____________
DSKAG Austria Research Team: http://www.research.dskag.at

Still trying.

Another GPU-task fail on the 64-bit, HD3850 W7 SP1 rig - spec as before ...

http://milkyway.cs.rpi.edu/milkyway/result.php?resultid=227629072

And another one:-

http://milkyway.cs.rpi.edu/milkyway/result.php?resultid=227629073

This Work-Unit freezes at 99.999 percent and hangs there!
(using 0 percent of CPU time, so the 2-logical-core CPU says its now 50 percent idle).

http://milkyway.cs.rpi.edu/milkyway/result.php?resultid=227925575

(It is running on app. 0.91 sse3 intel)

Shall I just abort it???
Or ... I will let it run for another 12 or 24 hours to see if something interesting happens... But it seems frozen. :)

Shall I just abort it???
Or ... I will let it run for another 12 or 24 hours to see if something interesting happens... But it seems frozen. :)

I had many errors.
http://milkyway.cs.rpi.edu/milkyway/results.php?userid=41296

Of course. I tried it many times with same result:
BOINC starts normally and this
WU is already frozen immediatelly, at: 91.819 %

http://milkyway.cs.rpi.edu/milkyway/result.php?resultid=227925575

I had to *KILL* / err. - abort the task in Boinc MAnager. I finally lost my God-like patience with it! :)
*Namaste*
Filip

Another frozen WU at 91.9 percent.
Aborted already on more than one platform! see? :)

http://milkyway.cs.rpi.edu/milkyway/workunit.php?wuid=180587743

Still trying.

Another GPU-task fail on the 64-bit, HD3850 W7 SP1 rig - spec as before ...

http://milkyway.cs.rpi.edu/milkyway/result.php?resultid=227629072

And another one:-

http://milkyway.cs.rpi.edu/milkyway/result.php?resultid=227629073

I'm getting alot of errors on my aeging PowerPC, only two tasks completed in the past several days:

http://milkyway.cs.rpi.edu/milkyway/results.php?userid=198978

Strangely enough, on any of my machines I don't have a problem running any of the project work units on my other computers (one Windows XP machine, one Windows 7 Ultimate machine, one Linux machine).

However, I do have one machine running FreeBSD 9-STABLE, and using the boinc client that came in the ports tree; and on this particular machine, when I get a batch of work units, a percentage of them fail instantly when they start, while others work fine and complete fine.

I'm not entirely sure if this is an issue with FreeBSD or not, I bring it up here because they're the same units from this new search that randomly fail.

EDIT: I just checked all my other machine's work units, and they're all working on a different search. But nonetheless, two of the six units it failed to run were pe_separation, and de_separation; so I can safely say I am getting faulty units with this new search, but I'm not to sure as to why others are failing.

Every one of these tasks I've gotten for the last few days has been ending with computational errors after a few seconds. Any ideas how to fix it? PC ID: 338106