Author: Robert Hyatt
Date: 10:53:28 01/24/02
Go up one level in this thread
On January 23, 2002 at 02:24:29, Jouni Uski wrote: >It seems to have working trojan check... I even looked at source and noticed, >that code for that check was only may be 30 rows! I wonder if that code really >can detect in all positions, if white/black has winning attack after taking >trojan bishop/knight? > >Jouni It is simple code. It looks for the open h-file, with crafty castled and the opponent not castled, with the opponent having a rook + queen available for attacking on the file. It is simple, fairly "coarse" in what it looks for, but it _does_ prevent taking the piece. Sometimes when it might be perfectly safe to do so in fact. What it generally does is to find a way to take the piece _and_ force the trade of queens (not necessarily in that order) so that after opening the file, the queen is gone and the attack doesn't work.
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.