Author: Dave Gomboc
Date: 23:44:58 09/17/01
Go up one level in this thread
On September 17, 2001 at 07:40:44, Dan Andersson wrote: >I have known of APHID for a long time. All I can say is. I'm not impressed. >There are way better work distribution algorithms. > >MvH Dan Andersson Mark didn't cheat by counting speedup by comparing a one node system with 32Mb of hash with a 1000 node system with 32Gb of hash. The numbers would be better if his results were distorted like the reported numbers for some other parallel searchers are. So compare running time of implementations before dismissing it as junk. Dave
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.