Posts by halfempty

81) Message boards : Number crunching : New ATI OpenCL-based AstroPulse (rev516) released (Message 1088123)
Posted 18 Mar 2011 by halfempty
Post:
Thanks, that's what I thought. New app releases don't come around that often so it's not a big deal, but it must be a pain for you guys (developers/testers).
82) Message boards : Number crunching : New ATI OpenCL-based AstroPulse (rev516) released (Message 1088116)
Posted 18 Mar 2011 by halfempty
Post:
-hp - sets high priority class

Is this a higher priority than other BOINC processes only, or is it higher than normal system processes?

Also, is there any way to tell BOINC to re-read the app_info file, or do you have to restart the application?

TIA
83) Message boards : Number crunching : Nvidia Gts 450 expectations (Message 1081166)
Posted 25 Feb 2011 by halfempty
Post:
The app should be correct because some units get validadet.
Also i checked other hosts running the 450.
All using the same app.

Host ID http://setiathome.berkeley.edu/results.php?hostid=5499328

He reinstalled the optimized apps using unified installer 0.37.
Nvidia drivers updated as well.

I'm running the GTS450 also with the same app and driver, but with AMD and Win7_64. I get a number of inconclusive results, but they almost always validate. I remember seeing 3 invalid results over the past month I've been running the card, but your friend has a bunch.

The only thing that caught my eye is that his run times vary a lot more that mine. Were there any attempts at overclocking, or is there something else heavily competing with the card while it's crunching?

I should have suggested to buy a ATI instead.

I like ATI myself, but none of their newer cards that do double precision are in my price range, or fit in my case.
84) Message boards : Number crunching : My HD4870 GPU burned (Message 1079923)
Posted 21 Feb 2011 by halfempty
Post:
Coordinating a collection drive can be a thankless time consuming PITA. I would just like to say one thing:

THANK YOU MIKE!!!
85) Message boards : Number crunching : My HD4870 GPU burned (Message 1079375)
Posted 19 Feb 2011 by halfempty
Post:
Don't take me too seriously, just rooting for the home team.
86) Message boards : Number crunching : My HD4870 GPU burned (Message 1079050)
Posted 19 Feb 2011 by halfempty
Post:
Yes but I haven't given up hope on getting a 58x0 and want to make sure he has one for future development. ;-)
87) Message boards : Number crunching : My HD4870 GPU burned (Message 1078984)
Posted 19 Feb 2011 by halfempty
Post:
Don't have much $$$, but I'll be happy to contribute what little I can. If I may make a suggestion, instead of a single top of the line card, one each 5000 and 6000 series cards would help in testing apps for the different platforms.

Also agrree with a previous poster, check what cards will fit. I am currently crunching on a GTS450 since I had to retutn the HD5830 I ordered because it wouldn't fit in my case with those power connectors in the back.
88) Message boards : Number crunching : Closing the lid (Message 1077024)
Posted 14 Feb 2011 by halfempty
Post:
I have to keep the cases closed. My cats aren't as well behaved as msattler's. They love to play with and/or tear out the pretty colored wires.
89) Message boards : Number crunching : MultiBeam application for ATi GPUs released (Message 1071857)
Posted 29 Jan 2011 by halfempty
Post:
Thank you oh gods of MultiBeam! :-D

Now that my HD5750 can crunch more than AP, I have to figure out what to do with the new GTS450 that arrived today.
90) Message boards : Number crunching : cluster computer (Message 1067664)
Posted 17 Jan 2011 by halfempty
Post:
I'm not sure that a p-III or p-4 can calculate a seti work within the deadline.

Two months ago I retired (it retired itself) and OLD PII-450 I had running for a decade. Years ago I upgraded it to the "modern" BOINC software from Seti Classic and it always made its deadlines with no errors, even AP. Admittedly it took almost two weeks to process a single AP WU and had a whopping RAC of 40, but it made it! The temperature in the room went down a couple of degrees when the poor old beast died. :-)
91) Message boards : Number crunching : ATI MB rev170 Now 177 beta app testing (Message 1067482)
Posted 16 Jan 2011 by halfempty
Post:
Yeah, I'm with you. I can't afford to retire a card I haven't even received yet. :D
92) Message boards : Number crunching : ATI MB rev170 Now 177 beta app testing (Message 1067315)
Posted 16 Jan 2011 by halfempty
Post:
Any Idea when this will be public?

I was thinking of asking the same question. Wish I hadn't just ordered a 5830. Guess I can keep crunching AP until they run out and maybe the GPU gods will smile on us ATi users and it won't be too long. ;)
93) Message boards : Number crunching : Astropulse times with ATI app (Message 1062555)
Posted 1 Jan 2011 by halfempty
Post:

Sorry that I might have caused some confusion: I compiled my own version of OpenCL Astropulse

Thank you for that reply. At least this time I have an excuse for being clueless. :-)
94) Message boards : Number crunching : Astropulse times with ATI app (Message 1062400)
Posted 1 Jan 2011 by halfempty
Post:

Running 2 astropulse apps isn“t possible atm like Raistmer said.


Thanks for clearing that up, I was confused. Happens a lot!
95) Message boards : Number crunching : Astropulse times with ATI app (Message 1062313)
Posted 1 Jan 2011 by halfempty
Post:
Hope everyone had/is having a great New Year!


Claggy this result was crunched before i changed the values.
All look fine now.

If I may ask, what values did you change? I'm trying to run 2 AP WUs on an HD5750 using the r456 OpenCL app and am getting the same "Can't follow BOINC's proposal, will try to find free device.." error message. The second WU accrues time, but no progress until the first one finishes. The GPU averages about 34% with several periods of 0 activity during this time.


Previous 20


 
©2020 University of California
 
SETI@home and Astropulse are funded by grants from the National Science Foundation, NASA, and donations from SETI@home volunteers. AstroPulse is funded in part by the NSF through grant AST-0307956.