Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Sacrifices of chess programs

Author: Mark Young

Date: 08:04:59 11/08/01

Go up one level in this thread


On November 08, 2001 at 10:20:37, Uri Blass wrote:

>I define a real sacrifice of chess program as a move that is losing material
>based on only material evaluation but is played by the program because the
>positional score is bigger than a pawn.

Most of the Real sacs I have seen are based on mating attacks. Do you define
this as a real sacrifice?



>
>How many real sacrifice are played by programs?
>
>It may be interesting to get information about the average number of real
>sacrifice per game by different programs.
>
>I define a sacrifice as something that is seen by only material evaluation as a
>mistake and the program need to choose it not because of the fact that the loss
>was too deep or because a bug in the search but because of the evaluation
>function.
>
>Here is an example of a real sacrifice that part of the programs play(after h4
>Nxh4 black is a pawn down) and example of a move that I believe that is simply a
>tactical blunder that is too deep for programs to see
>
>Abir Har aven - Uri Blass
>[D]rnbqkbnr/ppp1ppp1/8/3p3p/5P2/5NP1/PPPPP2P/RNBQKB1R b KQkq - 0 1
>
>Analysis by Junior 7:
>
>3...h4 4.Nxh4
>  ²  (0.32)   Depth: 3   00:00:00
>3...g6 4.Nc3
>  =  (0.06)   Depth: 3   00:00:00
>3...d4 4.Bg2
>  =  (-0.01)   Depth: 3   00:00:00
>3...c5 4.Nc3
>  =  (-0.06)   Depth: 3   00:00:00
>3...Nh6 4.Nc3
>  =  (-0.09)   Depth: 3   00:00:00
>3...Nf6 4.Nc3
>  =  (-0.20)   Depth: 3   00:00:00
>3...Bf5 4.Nc3
>  ³  (-0.34)   Depth: 3   00:00:00
>3...Bf5 4.Nh4 Be4 5.Bg2 Bxg2 6.Nxg2
>  ³  (-0.26)   Depth: 6   00:00:00
>3...Nf6 4.Nc3 Bf5
>  ³  (-0.30)   Depth: 6   00:00:00
>3...c5 4.Nc3 d4 5.Na4 Qd5
>  ³  (-0.39)   Depth: 6   00:00:00  1kN
>3...c6 4.c3 Qb6 5.Qa4
>  ³  (-0.46)   Depth: 6   00:00:00  2kN
>3...Qd6 4.Nc3 Qc5
>  ³  (-0.48)   Depth: 6   00:00:00  2kN
>3...Qd6 4.d4 Bf5 5.Nh4 Be4 6.Bg2 Bxg2
>  ³  (-0.41)   Depth: 9   00:00:00  14kN
>3...Qd6 4.c4 dxc4 5.Na3 c3 6.Bg2 Qb6 7.bxc3
>  ³  (-0.42)   Depth: 12   00:00:00  170kN
>3...h4 4.Bg2 h3 5.Bf1 c6 6.c3 Qb6 7.Qa4
>  ³  (-0.58)   Depth: 12   00:00:00  317kN
>3...h4 4.Rg1 hxg3 5.Rxg3 Qd6 6.d3 Bf5 7.Bd2
>  ³  (-0.54)   Depth: 15   00:00:04  2153kN
>
>(Blass, Tel-aviv 08.11.2001)
>
>Dov Rozenberg - Uri Blass
>[D]2r2rk1/p3q1pp/bp6/2bp1p2/4n3/1P2P1P1/PB1N1PBP/1QRR2K1 w - - 0 1
>
>I believe that 23.Nf1 is losing a pawn or a rook for a bishop and a pawn but
>programs positional scores are high here and simply hide the picture and I have
>not program with only material evaluation to check it.
>
>the analysis is 23.Nf1 Be2 24.Rxd5 Qf7 25.Re5 Rcd8 26.g4 fxg4 27.Rxe4 Qxf2+
>28.Kh1 Bf3 29.Bxf3 Qxf3+ 30.Kg1 g3 31.Qc2 gxh2+
>
>In the game my opponent chose 26.Rcxc5
>
>Is there a program with only material evaluation or at least mainly material
>evaluation in order to find tactical mistakes or sacrifices in a game and if
>there is more than one what is the best program to find material mistakes or
>sacrifices?
>
>Uri



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.