Author: Bas Hamstra
Date: 13:56:59 01/21/03
Go up one level in this thread
>Yes, but of course before passing the attacked set back to the routine one >should mask it with some appropriate allowed to move/capture set. >May a piece capture a passers in eg. n <= 4 moves/captures, considering >opposite null moves, except pushing the passer? What's the "best" way to >develop a piece? Detecting trapped pieces, the flood stops after two >iterations etc... I have played a bit with pawnstructures, keep pawns so that important squares can be potentially dominated, while the opponent cannot. It is interesting but in my case it lead to kind of ehm, "exotic" pawnstructures. Maybe bad tuning of penalties, I don't know. But my experiences with this type of lookahead in practical gameplay are not very encouraging so far. A good example of where it does work, is Crafty's potential passer code (which BTW can be done loopless and paralel for all pawns, with some fill routines). Do you actually use these kind of ideas in your eval? Bas.
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.