Author: Robert Hyatt
Date: 10:46:00 01/30/03
Go up one level in this thread
On January 30, 2003 at 04:55:07, Vincent Diepeveen wrote: >On January 30, 2003 at 00:24:25, Jeremiah Penery wrote: > >>On January 29, 2003 at 23:20:11, Vincent Diepeveen wrote: >> >>>That already should give you the answer. Writing parallel programs is 1 thing. >>>Writing something that works well without inventing numbers yourself is another >>>thing. >> >>So, can you show us some non-invented numbers? > >Sure i can. You can lookout for an icga article there. dunno when. some months >probably, but i will post all outputs online too and it won't be 'lost outputs' >or invented numbers. > >More interesting from my viewpoint isn't the 8 cpu output though, i am aiming > >32 cpu's. And what a paper that will be. No doubt "proofing" that you can get a speedup of > 2.0 for two processors, all the time. That you have a totally deterministic search that doesn't vary at all from run to run. That your "locks" don't hurt performance at all... I can hardly wait... Please ask them to print that ICGA issue on very _soft_ paper, so I can get at least _some_ use out of it. :)
This page took 0 seconds to execute
Last modified: Thu, 15 Apr 21 08:11:13 -0700
Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.