Log in

View Full Version : BOINC: Fixed or Benchmarked Credits


Evil-Dragon
10-20-2006, 01:44 PM
Thought i would start a poll to see what people prefer. Please feel free to comment.
BlackAdder
10-20-2006, 03:59 PM
I think I prefer a time spent and benchmark credit system, however the old Seti system of one WU for one credit point has it's advantages too...don't know...thats a tough question , but it needs to be fair to all...
drezha
10-20-2006, 04:19 PM
Fixed credits please. Fair fixed credits I might add.

The folding way of doing things is best IMHO. (All units ran on a benchmark machine (with SSE off) and points award based on 110*No. of Days to complete unit)
Lagu
10-20-2006, 04:48 PM
If we should prefer a dynamic credits slower computers should be discriminated ant that should not be fair i think
Steve Lux
10-20-2006, 08:25 PM
I don't care how or if credits are awarded. It's the research that counts.
Brucifer
10-20-2006, 08:43 PM
Definitely fixed points.
AMD-USR_JL
10-21-2006, 03:34 AM
I want fixed credits too, but i'd want all the project admins to a agree on the same way of figuring credit per WU (maybe a FLOP counter?). I don't see this happening anytime soon, so i voted for dynamic credit.
gatekeeper53
10-21-2006, 04:58 AM
I go along with Steve.
Murray
10-22-2006, 05:07 AM
I must say that fixed credits such as Riesel Sieve's system is great if you have a fast computer, slower PCs and Laptops on the other hand benefit from dynamic credits...I am sure there will always be projects using both and in both cases crunchers can use the systems to their advantage.

.....
.....
if they are stats crazy...like most of us :)