Re: More stats
"Mike" <[email protected]>
| Newsgroups | gmane.comp.distributed-net.rc5 |
|---|---|
| Message-ID | <[email protected]> |
I think either of these measures would be a bit misleading. A measurement of MIPS, not FLOPS, would vary significantly if measured per stat unit based on the CPU architecture and speed. What I mean is, an operation that takes 2 CPU cycles on an AMD may take only 1 on a Cell, and the same operation may take 3 on a SPARC without even comparing the speed of the CPU in ghz. I think the same goes for CPU year. The amount of work done by a P2 300 in a year vs a 3ghz Xeon or a PS3 Cell varies greatly. I suppose one could pic a baseline cpu to base CPU years on much like on the proxy rate page for rc5.. This would work for MIPS or CPU years - but these would both be a direct function of stats units per day. Don't get me wrong, I'm all for more geek stats, but I think we could be comparing apples and oranges.. I am interested to hear any other opinions on this one :) Mike _______________________________________________ rc5 mailing list rc5-Ra3b/QYEcJ3d140v2zMXi0fjHoOT/h/[email protected] http://lists.distributed.net/mailman/listinfo/rc5