Author: Michael Neish
Date: 19:30:42 11/28/00
Hi, When one program is said to be tactically stronger than other, is this usually the result of extensions? I've added the usual extensions to my search routine (check, recapture, single reply), but it seems that more are needed to make the program "strong" as opposed to "acceptable". I'm very impressed, for example, with how Hiarcs 7 often finds combinations at depth 4 which my program cannot see to depth 8 or 9. I tried adding some more extensions, such as enemy moves that attack squares near the King, with the idea that if moves are threatening, they ought to be searched more carefully. This either exploded the tree or was useful only in such restricted situations that it was not worth the extra load. Maybe some other factor is needed, such as threatening moves to the King when at least one or two other pieces are present, but of course, each extra condition slows down the program a notch. So, has anyone got any handy hints about adding more tactical acuity to one's program? I heard that Phalanx uses extensions very well for King attacks. Would it be worth studying its source code? Cheers, Mike.
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.