Author: martin fierz
Date: 19:56:29 04/05/02
Go up one level in this thread
On April 05, 2002 at 08:23:55, Gian-Carlo Pascutto wrote: >[D]3r1rk1/p1pq2pp/3bp3/8/1nNpP1n1/1P3N2/P4PPP/RQBR2K1 b - - 0 16 > >PM played Rxc3! and ripped my thing of the board. > >(I'm not sure if it's the best move or the only way to win, >but I'm impressed nevertheless) > >-- >GCP any strong human will play this move without even thinking. with thinking, he may be afraid, but if you show this to any player who ever heard about the idea of an exchange sacrifice, he will suggest Rxf3. i'll tell you what the problem is: in all these chess books, people faithfully write: bishop: 3 pawns, knight: 3 pawns, rook: 5 pawns. this is just bogus. in the middle game a rook is not very powerful (endgames are different...). positions where one side can sac a rook and get a pawn for it are often ok for the side which is an exchange down (but it is doomed once all heavy pieces are off!). if the position remains fairly closed and you get a nice outpost for a minor piece, you are already ok. if, in addition, you can rip the opponent's pawn structure to pieces, you are just fine. if the king still happens to be behind that ripped pawn structure, well, it's a must! this position has all these ingredients (and more - all defenders are gone, you still have a knight to attack). your static evaluation for the position after Rxf3 gf3 Nxh2 should be negative for white. at least my static evaluation of this position is clearly negative. being human, i would worry that i miss some tactics, that i lose the knight on h2, but if not, black is clearly better here. there are many examples for nice exchange sacs in return for less than what you get here. in the dragon, for example, black will always play Rxc3 if white has to play bc3 in return. in the tarrasch french there are lots of Rxf3 positions, which look less promising than this one of yours. aloha martin
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.