Author: Jon Dart
Date: 08:22:36 08/15/04
Go up one level in this thread
On August 15, 2004 at 10:35:29, Stuart Cracraft wrote: >I am familiar with the lazy checkpoint 1. I assume the further >ones just somewhat lesser-contributors on down and that you have >retests of the condition at #1 for the new score, continually >exiting along the way if possible avoiding further evaluation, etc. > Yes, that it is it. The margin to exit can be different at various points in the eval. It also helps if you avoid cases where there are potentially large eval terms you have not calculated yet. E.g. I have a large bonus for pawn races where one side appears to win: if the pawn race code might be triggered I disable lazy eval in the sections of the code before the pawn race calculation. --Jon
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.