Author: Uri Blass
Date: 06:49:12 01/29/06
Go up one level in this thread
On January 29, 2006 at 07:17:18, mike schoonover wrote: >hi jouni, >ryans observation about all the tables is right. >i wonder though,with todays higher processor speeds >if programmers can now add all the stuff they left >out in the past because of speed penalties? >example;routines that were deemed impractical 3-4 years ago >may be ok today due to the speed increase. >it would be interesting to see some programmers >reconsider some stuff the left out in the past to see what happens. >regards >mike I never decided not to add stuff because of speed panelties. Movei does not know a lot of things but the reason is not speed but being too lazy to implement the stuff with no bugs. Calculating knowledge when the remaining depth is big enough is never expensive and if the knowledge is important then I am sure that it is possible to have productive rules where to use it. Note that I believe that chess programs need to evaluate internal nodes in order to do inteligent decisions in the search so evaluation of correct knowledge when the remaining depth is big enough can be always productive(unless we talk about knowledge that save one or 2 plies in the search and this type of knowledge is not real knowledge because it can be practically replaced by searching deeper). Uri
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.