Author: Tord Romstad
Date: 09:12:37 10/30/03
Go up one level in this thread
Sorry for the empty posting -- I accidentaly hit the "Submit Follow Up" button. On October 30, 2003 at 09:29:44, Vincent Diepeveen wrote: > >91.4% is fine as a cutoff rate. It seems like most people think so. Nice to know. >Interesting other statistics are: fliprate > >Which is the chance that a node stored as <= alfa, that it gets >= beta now > >==> P( alfastored | bestscore >= beta ) I am not entirely sure I understand what you mean here. Do you mean to check the frequency of nodes where the position is found in the hash table, the upper bound stored in the hash table entry is <= alpha, but the search returns a value >= beta? It could be interesting to experiment with this as well, although it would look slightly different in my program because I have only one search bound. Tord
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.