Author: Robert Hyatt
Date: 12:58:57 10/30/01
Go up one level in this thread
On October 30, 2001 at 15:53:24, Dieter Buerssner wrote: >On October 30, 2001 at 15:40:54, Robert Hyatt wrote: > >>if it does that it will almost certainly end in a draw. The KRB vs KRP is >>_not_ going to be lost. And in the meantime, it is going to save me a _lot_ >>of 0 scores, because my opponents think that KRB vs KRP is winning for the >>KRB. When it isn't. > >I don't understand, why you think, that it saves a lot draws. I would think, any >human opponent, who can reach such a position will very well know, what he is >doing, and not accidently trade his last pawn away. Computer opponents, that >have been a bit matured, will do the same (it is only a few minutes of work, to >do it this or another way). > >Regards, >Dieter It saves draws because I have seen _several_ chess engines win a piece, and then allow trades that take it to a piece up ending that is a dead draw. I don't want to draw something I might can win later on. Of course, I don't handle KRBP vs KRPP in my eval right now since both sides have pawns and I consider it winnable by either side. But if white gives up that pawn, then I want to know that now I can't win even though I have a material advantage. As far as how programs do it, I have seen way too many that simply "don't know" this. And it has saved me a few losses and turned them into draws. IE in a KRR vs KRBP ending crafty will promptly trade the rook for the opponent's last pawn if it gets too far advanced, knowing that is a draw. Rather than waiting until it is too late and the opponent builds a protective bridge and slips the pawn in for a new queen. If you handle it another way, that's fine. But it _must_ be handled or you end up drawing won endings, or losing drawn endings by letting your opponent hold onto a pawn too long.
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.