Tell Us What You Think
Are you Happy with the points Ver. CPU Time
In the project your working
GPU's excluded
Printable View
Tell Us What You Think
Are you Happy with the points Ver. CPU Time
In the project your working
GPU's excluded
For a BOINC gypsy like me, that's impossible to answer. I rarely spend more than a week crunching a single project on my 'main' machines. For old flaky machines I tend to go for projects granting 'decent' credit or it's not worth the electricity keeping them running.
I'm happy with the credit from most projects.
There's a few I won't crunch due to stupidly low credit (Correlizer, SAT, Rosetta, EON, YAFU, Slinca, Mindmodelling, WEP-M2 to name the main offenders).
Others I won't crunch because of long WUs with no checkpointing. And some I won't crunch because they seem to be a complete waste of electricity. And some I won't crunch because they need a version of BOINC that sucks (pretty much anything after 5.10.45). And some I won't crunch because they only run on Linux. And some I won't crunch because the admins are d*cks :)
Funnily enough, 3 the 4 big-hitters for CPU work are down or out of work. I smell a conspiracy ;)
Anywoo, that leaves me with 41 projects on my list to choose from that offer enough credit to keep me happy. I remember when there were only 6 project in total to choose from ;)
End of waffle :)
Al.
Only 6 votes :Pokes:
Come on be a Man stand up and be counted:cheers:
For those of you that are unhappy with the points awarded you may want to try numberfields. The return on investment for CPUs is above average. It appears that it will only have a limited run so you better get them while you can.:hiya:
Oh, is the 'run time' problem fixed for older clients? I abandoned it due to dismal credit with 5.10.45 :(Quote:
Originally Posted by Bigred
Al.
Depends on the project in question. Some don't aware enough, so really should be giving more per hour to bring them closer to the average. I'm talking about Rosseta and WCG as 2 of the big offenders for low credit return.
My strat is usually to pair up one of the high payers with something that pays poorly. On the one hand I wouldn't want to do nothing but math projects all the time, on the other hand, the credit return for some is so ridiculously low, they really could stand to give more credits per task. But by running them together, things fall more in the middle, and the higher paying projects gives one the headroom to run some of the over nerfed and ridiculously low paying projects, without the RAC taking a big nose dive. It can then sorta balance out, in the middle somewhere :D