Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: I want to make a new statement in chess

Author: Christopher R. Dorr

Date: 06:52:27 10/05/99

Go up one level in this thread


I think this flies in the face of vast experience. While changing these values
in favour of the knight may make CM6K play better in some circumstances, I doubt
that you could find a human master who would agree with your statements about
the relative values of bishop and knight.

Your statement that N's are better "up until the ending" is flat wrong. In
certain types of positions (locked pawn structures typical of closed systems,
French Defences, Czech Benoni's, etc.) the N is more useful than a bad bishop
(especially if it can get to a secured outpost square). But in the majority of
positions, a bishop will be of slightly more value that the knight, if because
of nothing other than sheer mobility. And this completely underestimates the
great power of the two bishops. Want to try an interesting experiment? Try
removing the 2 N's from one side, and the 2 B's from the other. I'd bet a lot
that the side *with* the 2 B's won a sizeable majority of the games.

One thing to remember about chess programs, and modifying their pawn/piece
values. Modern programs as so *vastly* stronger than average humans, that you
can modify them in many ways that actually decreases their true strength, and
still have them kill you almost all of the time, simply because of their
tactical brilliance. I'd venture that I could change the value of R=6 and Q=7
(obviously very incorrect numbers), and still have the average chess program
whump the average USCF 1900 9 out of 10 times.

Philidor's statement survived for hundreds of years because it encapsulated a
deep and profound truth about chess. And because it is right. I don't think your
statement about the relative value of knights vs. vishops does either. But I am
glad to see that the chess programs of today are creating discussions about the
philosophy of chess, in addition to giving us great playing partners, and lovely
games to look at.

Cheers,

Chris Dorr
USCF Master


On October 04, 1999 at 20:24:06, Marc van Hal wrote:

>In the time Philidor said that the pawns are the soul (skeleton) of chess
>I now say he was only partialy right.
>I want to improve this statement by saying pawns and knight are the soul of
>chess
>After I made my setup for a chessmaster personelety I went up giving the knight
>a higher valeu then the bischop.
>with this reason that it will use his knight more often and also makes strong
>squares for it
>The Bischop only is stronger in the very end of the endgame but in all other
>stages of the game the knight play a much bigger role in the game then the
>bischop that means about 80% of the game the knight plays a bigger role
>or you should play like a morphy stylist and brake open lines by doubeling pawns
>only then the bischop plays a bigger role in the 100% of the game
>but this is a hard style to folow and is very risky.
>Just think about it.



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.