Author: Robert Hyatt
Date: 17:01:12 11/22/01
Go up one level in this thread
On November 22, 2001 at 14:06:37, Dan Andersson wrote: >Don't forget Heinz paper that proves diminishing returns for Fritz in a >self-play test. That is not what I call anecdotal. > >MvH Dan Andersson It is anectodal from the perspective I gave... namely that of playing A vs A (different depths) to extrapolate how A does at increasing depths against _anybody_. A vs A is already an uninteresting test since the programs are essentially identical except for search depth and nothing else. Drawing conclusions about how depth affects an engine by playing it against itself is not necessarily going to say _anything_ about how a program will do with deeper depths against _other_ opponents... Monty ran an experiment (not self-play) that suggested that deeper depths are _always_ better, at least thru 14-15 plies. I trust that at _least_ as much as A vs A testing.
This page took 0.01 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.