We need to reward only for successful tasks. For computation resources, it
needs to be based on the FLOP count. This can be estimated in several
ways, and no matter what is done gold plating benchmarks, it will always be
an estimate.
jm7
Raistmer
<[email protected]
> To
Sent by: "Martin" <[email protected]>,
<boinc_dev-bounce "BOINC Developers Mailing List"
[email protected] <[email protected]>
u> cc
Subject
09/28/2009 03:45 Re: [boinc_dev] [boinc_alpha] Card
PM Gflops in BOINC 6.10
>Do we "reward" for the result returned, or for the resource that a
>project has usefully utilised?
Reward should be for result returned of course, not for resource used.
1) Resource usage just can be faked by too eager for credits participant.
No "cheating" should be ensured.
2) What about case of "broken resource" ? I personally had such "resource"
few days - excessively OCed GPU.
It was fast resource, accordingly "pay for resource" conception it would
get much more than non-OCed GPU card.
But it returned completely broken results. For what reason should credits
encourage such configs ???
We do computations NOT FOR CREDITS themselves, but for scientific results!
If results are absent or wrong - no credits.
Will anyone pay for broken TV, for example, just because someone spend few
hours to assemble it ?
_______________________________________________
boinc_dev mailing list
[email protected]
http://lists.ssl.berkeley.edu/mailman/listinfo/boinc_dev
To unsubscribe, visit the above URL and
(near bottom of page) enter your email address.
_______________________________________________
boinc_dev mailing list
[email protected]
http://lists.ssl.berkeley.edu/mailman/listinfo/boinc_dev
To unsubscribe, visit the above URL and
(near bottom of page) enter your email address.
