Strange behavior in WU granted credit
29 Mar 2005 20:47:05 UTC
Topic 188796
(moderation:
This work unit seems odd to me:
http://einsteinathome.org/workunit/619205
Anybody understand why the lowest value was credited?
The WU sent to 4 users, one of which seemed to have errors but the other three were given the lowest value not the mid value.
Any ideas?
Thanks
Strange behavior in WU granted credit
)
http://einsteinathome.org/workunit/565027
"Any ideas?":/
It can be an failure in the
)
It can be an failure in the "Matrix".
I don´t know why, but i can say, low credits are better than no credits. :D
> This work unit seems odd to
)
> This work unit seems odd to me:
>
> http://einsteinathome.org/workunit/619205
>
> Anybody understand why the lowest value was credited?
>
> The WU sent to 4 users, one of which seemed to have errors but the other three
> were given the lowest value not the mid value.
>
> Any ideas?
The first machine (running Linux) to return its result was overruled by the next 2 machines (running Windows). At that point the lower credit claimed for the 2 accepted results was granted. The last machine to return its result had no influence on the result and was granted the amount of credit that was already determined previously.
@Jerzy Ziembinski
>http://einsteinathome.org/workunit/565027
There is nothing unusual. The last result came in late, so no credit.
[url=http://einstein.phys.uwm
)
[url=http://einsteinathome.org/workunit/636326]WTF - Granted 0.0 credits!
Why invalid?
Freedom of Speech is a cherished right in every democracy and democratic institution.
>
)
> [url=http://einsteinathome.org/workunit/636326]WTF - Granted
> 0.0 credits!
>
> Why invalid?
Because it is not valid?
Seriously, if there was an error in the processing, it does not compare with the others by a wide margin, corruption in the data returned, etc.
We all get our invalid Results. In part that is because unlike many of the first generation DC projects, all BOINC projects are doing validation testing and redundent results, even CPDN has some models re-run ...