Yes, the discussion about validator problems, and the sending of replacement tasks for tasks which temporarily had the status 'unknown', but validated correctly at the second attempt.
I have 42 validated tasks, with 4000 credits identified as:
Binary Radio Pulsar Search (Perseus Arm Survey) v1.33 (BRP4cuda32nv270)
Is that BRP5 or BRP4? Is it a different application running or different data to the same application?
Just curious, I'm not loosing any sleep over this.
Oh and there were running on different machines with different GPU's but the times range between 18,336 and 31,544 seconds on Linux with nVidia cards running 2 or 3 WUs at a time. MacOS were in the 58,000 range with the stock ATI card.
The Arecibo were in the 1700-4000 range for nVidia and about 20,000 for Mac/ATI for 500 credits.
I have no complaints about credit allocation, at least not until we get to exchange them for BitCoins.
Binary Radio Pulsar Search (Perseus Arm Survey) v1.33 (BRP4cuda32nv270)
Is that BRP5 or BRP4? Is it a different application running or different data to the same application?
This is a BRP5 task ("BRP4cuda32nv270" is the name of the BOINC "plan class", which we are just re-using from the BRP4 run, hence the name). It is different data (from a different radio telescope, even), crunched by the same application, but using different parameters & templates (which tell the application "how" it should look for "which" signals, so to speak).
This WU was validated and yet
)
This WU was validated and yet it was sent out again days later: http://einsteinathome.org/workunit/165131284
Am I missing something?[/url]
RE: This WU was validated
)
Yes, the discussion about validator problems, and the sending of replacement tasks for tasks which temporarily had the status 'unknown', but validated correctly at the second attempt.
All in Technical News.
I'm a little confused about
)
I'm a little confused about what exactly BRP5 is.
I have 42 validated tasks, with 4000 credits identified as:
Binary Radio Pulsar Search (Perseus Arm Survey) v1.33 (BRP4cuda32nv270)
Is that BRP5 or BRP4? Is it a different application running or different data to the same application?
Just curious, I'm not loosing any sleep over this.
Oh and there were running on different machines with different GPU's but the times range between 18,336 and 31,544 seconds on Linux with nVidia cards running 2 or 3 WUs at a time. MacOS were in the 58,000 range with the stock ATI card.
The Arecibo were in the 1700-4000 range for nVidia and about 20,000 for Mac/ATI for 500 credits.
I have no complaints about credit allocation, at least not until we get to exchange them for BitCoins.
Joe
RE: Binary Radio Pulsar
)
This is a BRP5 task ("BRP4cuda32nv270" is the name of the BOINC "plan class", which we are just re-using from the BRP4 run, hence the name). It is different data (from a different radio telescope, even), crunched by the same application, but using different parameters & templates (which tell the application "how" it should look for "which" signals, so to speak).
Cheers
HB
Thanks HB!
)
Thanks HB!
I see the BRP5 1.34 size is
)
I see the BRP5 1.34 size is identical to the 1.34 BRP4 task size.