Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Keeping the last pawn

Author: Robert Hyatt

Date: 18:30:07 11/26/00

Go up one level in this thread


On November 26, 2000 at 18:19:44, Mike S. wrote:

>In general, trading is good for the side being ahead in material. This is basic
>chess knowledge, known to most programs. But in the endgame, it's often wrong to
>trade the last pawn, if the resulting material doesn't allow to win (a common
>rule known from beginner's books is, that even the advantage of a minor piece
>usually is not enough to win in a PAWNLESS endgame).


Actually the rule generally says "if ahead in material, trade pieces but
_not_ pawns...  if behind in material, trade pawns but not pieces..."

That solves your problem...




>
>This important motif can be tested with the following position from CSS 5/98,
>where Mr. R.Petrykowski reported from a game Fritz5/P133 vs. the Tasc R30. Does
>your program keep the last pawn?
>
>[D]8/4kn1p/R7/6p1/4K2P/8/8/8 w - - 0 49
>
>49.Kf5? gxh4, and the game ended in a draw.
>
>The only winning move seems to be 49.h5.
>
>I have tested some programs (with 4-piece TBs), for a few seconds on P3/700. The
>all would play h5, or found it fast respectively: Comet B02, Doctor? 3.0, Fritz
>5.32, Fritz 6 light, Hiarcs 6, Hiarcs 7.32, LiGo2000a, Junior 5 (0:15) -
>although some still show a big advantage after Kf5 gxh4 too.
>
>Regards,
>M.Scheidl



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.