Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Is silicon silly?

Author: Uri Blass

Date: 10:45:44 12/29/04

Go up one level in this thread


On December 29, 2004 at 12:54:34, Drexel,Michael wrote:

>On December 29, 2004 at 11:39:10, Misael Banuelos wrote:
>
>>The next annotated game was published at the chess column of the Spanish
>>newspaper 'El Pais' on December 24, and it was written by chess journalist and
>>Fide Master Leontxo Garcia.
>>
>>Yesterday I decided to post this message in this forum and I have translated it
>>into English. I hope I didn't commit too many mistakes in this language.
>>
>>Regards.
>>
>>MB
>>
>>---
>>
>>
>>[D] 6k1/6b1/1B1p2qp/3Pp1p1/2r1p3/3n2P1/P2QB2P/R5K1 w - - 0 27
>>
>>Position after 26...Rxc4
>>
>>*Silicon is silly*
>>
>>White: Gelfand.
>>Black: Sutovski.
>>King's Indian Defence.
>>Pamplona, December 22, 2004.
>>
>>In the opening round Israeli Boris Gelfand manages very well a position that
>>computers don't understand yet.
>>
>>1.d4 Nf6 2.c4 g6 3.Nc3 Bg7 4.e4 d6 5.Be2 0-0 6.Nf3 e5 7.Be3 h6 8.0-0 Ng4 9.Bc1
>>Nd7 10.d5 Ngf6 11.Be3 Ng4 12.Bd2 f5 13.g3 fxe4 14.Nh4 g5 15.Ng6 Nxf2 16.Rxf2
>>Rxf2 17.Kxf2 Qf6+ 18.Kg1 Qxg6 19.Nb5 Nc5 20.Nxc7 Rb8 21.Be3 b6
>>
>>Black's advantange, a pawn, is very little important: the bishop on g7 is almost
>>useless and the pawns on a7, b6 and d6 are weak.
>>
>>22.Nb5 Nd3?!
>>
>>As the reader is going to understand very soon, it was better 22...a6
>>
>>23.Nxa7 Nxb2 24.Qd2 Nd3 25.Nxc8 Rxc8 26.Bxb6 Rxc4 (diagram)
>>
>>Gelfand has got a clear advantage, thanks to his passed pawn on a2. But chess
>>programs don't appreciate it, not even after a long calculation time. It is due
>>to the so called 'horizon effect': Silicon is uncapable of understanding that
>>the little pawn on a2 is a potential queen, something that any chess player of
>>certain skill sees it in tenths of second.
>>
>>27.a4! Qe8 28.a5! Ra4
>>
>>The computer shows that 28...Qb5 is advantageous for Black, without noticing
>>that after 29.Be3 Qxd5 30.a6 Rc8 31.a7 Qa8 32.Bg4 Rf8 33.Be6+ Kh8 34.Qa2, with
>>the threat of Bd5, 34...Nb4 35.Qb3, Black must resign.
>
>Complete nonsense...
>
>Is "the Computer" Fritz 5.32 on a P200??

No need for P200

Fritz5.32 and a lot of chess programs cannot see it even on better hardware.

>Fritz 8 and Shredder 8 quickly think white is slightly better after 28...Qb5.
>
>Michael

I know that Fritz8 can often sacrifice material for no reason(it lost a game
against older version of movei exactly for that purpose) so I dislike Fritz8.

The first program that I tried(guess which program) did not see quickly that
white is slightly better after 28...Qb5.

[D]6k1/6b1/1B1p3p/Pq1Pp1p1/2r1p3/3n2P1/3QB2P/R5K1 w - - 0 1

I tried a better program like fruit2 and it also does not see it in a short time
and even after some minutes can see only 0.00 score

Analysis by Fruit 2.0:


3.Be3 Ra4 4.Rxa4 Qxa4 5.Qc3 Bf8 6.Bg4 Be7 7.Qc8+ Kg7 8.Qb7 Qa1+ 9.Kg2 Qa2+
10.Kh1
  ³  (-0.48)   Depth: 12/36   00:00:08  7499kN
3.Be3 Ra4 4.Rxa4 Qxa4 5.Qc3 Bf8 6.Bg4 Be7 7.Qc8+ Kg7 8.Qb7 Qa1+ 9.Kg2 Qa2+
10.Kh1 Kf7 11.Be6+ Kf6
  ³  (-0.42)   Depth: 13/36   00:00:11  10971kN
3.Be3 Ra4 4.Rxa4 Qxa4 5.Qc3 Bf8 6.Bg4 Qb5 7.Be6+ Kh8 8.Bf5 Kg7 9.Qc7+ Kf6 10.Qc8
Qb1+ 11.Kg2 Qb2+ 12.Kh1
  ³  (-0.36)   Depth: 14/38   00:00:19  18674kN
3.Be3 Ra4 4.Rxa4 Qxa4 5.Qc3 Bf8 6.Bg4 Kh8 7.Qc8 Qa1+ 8.Kg2 Qb2+ 9.Kh1 Qb1+
10.Bg1 Qf1 11.Qf5 Qxf5 12.Bxf5
  ³  (-0.26)   Depth: 15/38   00:00:42  40167kN
3.Be3 Ra4 4.Rxa4 Qxa4 5.Qc3 Bf8 6.Bg4 Kg7 7.Qc7+ Kg6 8.Qb8 Kf7 9.Qb7+ Be7
10.Bh5+ Kf6 11.a6 Qa1+ 12.Kg2 Qa2+ 13.Kh1
  =  (-0.07)   Depth: 16/46   00:02:02  113462kN
3.Be3 Ra4 4.Rxa4 Qxa4 5.Qc3 Bf8 6.Bg4 Kg7 7.Qc7+ Kg6 8.Qb8 Kg7 9.Qc7+
  =  (0.00)   Depth: 17/48   00:03:58  222044kN

(, MyTown 29.12.2004)


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.