Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Polugaevsky-Petrosian 50th USSR Ch Final 1958

Author: Ricardo Gibert

Date: 16:55:08 11/09/02

Go up one level in this thread


On November 09, 2002 at 18:40:44, Uri Blass wrote:

>On November 09, 2002 at 18:13:35, Ricardo Gibert wrote:
>
>>Are there any chess engines that understand anything about strategic exchange
>>sacs?
>>
>>
>>The following is an instructive example of an exchange sac that yields a
>>strategicly won position:
>>
>>[D]r3r1k1/1pqn1pbp/p2p2p1/3P4/1RP5/4B3/P3BPPP/3Q1RK1 b - -
>>
>>From Polugaevsky-Petrosian 50th USSR Ch Final 1958
>>
>>19...Rxe3!! 20.fxe3 Nc5-+ 21.Qc2 Re8 22.Rf3 Bh6 23.Qc3 Qe7 24.Rb6? Na4 0-1
>>
>>Neither Nimzo8 nor Ruffian 1.0.1 have a clue in the above position.
>
>Movei can see only a small positional compensation for white(score 1.30 at depth
>11 and 1.20 at depth 12 after 19...Rxe3 20.fxe3 Nc5 when the material advantage
>is considered as slightly more than 1.30)
>
>Maybe with bigger positional scores it can find it.
>
>depth=11 +1.30 f1f3 c7a5 d1b1 a8e8 b4b6 a5d2 g1f2 f7f5 f2f1 g7e5 g2g3
>Nodes: 24362368 NPS: 116907
>Time: 00:03:28.39
>
>depth=11 +1.30 f1f3 c7a5 d1b1 a8e8 b4b6 a5d2 g1f2 f7f5 f2f1 g7e5 g2g3
>Nodes: 39418809 NPS: 118563
>Time: 00:05:32.47
>
>depth=12 +1.20 f1f3 c7a5 d1b1 a8e8 g1h1 g7h6 b4b6 h6e3 b6d6 a5d2 b1f1 f7f5
>Nodes: 41103072 NPS: 118613
>Time: 00:05:46.53
>
>Can you win Ruffian or Nimzo with black from the above position?
>
>Uri


I'm too weak!

Against Ruffian I tried several 2 minute games with a 6 sec increment and lost
all of them. Then I put it in analysis mode tried to see if I could beat it by
peeking at the analysis. This is a double edge procedure, because it does not
really understand the position.

I lost the first couple before deciding that winning White e-pawn is a strategic
error. It allows White to exchange off a pair of Rooks. After that discovery I
actually managed to win one by winning the exchange (Ne4-g5xRf3), but when I
tried to repeat this success and giving Ruffian a bit more time, aI won the
exchange again, but it then found a strong pawn sac idea c4-c5 to activate the
White squared Bishop and I lost. That's enough for me today.

However, I have the feeling that after about maybe a dozen more games I may
learn enough about the position to be able to hold my own, but only by peaking
at Ruffian's analysis. A tough and instructive position!



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.