Author: Will Singleton
Date: 18:39:57 12/06/05
Go up one level in this thread
On December 06, 2005 at 18:17:32, Vincent Lejeune wrote:
>TogaII is known to sac the exchange more often than others engine. Rybka seems
>to do the same. The sac exchange in middle game is really hard to estimate, you
>often get a more active piece (B or N) instead of a blockes rook. Other engines
>have simpler estimation and simply decide to not give material !
>
>Thanks,
>Vincent
>
That seems to be right. This is a case where Fruit/Toga/Rybka play a move
quickly due to eval, whereas others will never play it. As the analysis below
clearly shows, those programs are obviously clones of Amateur, but their
programmers have cleverly avoided the first 13-14 plies of search due to
fail-low pruning and other secret techniques:
Amateur 2.86:
14 -5 9611 39966361 Qe2 a4 Rg4 Ra5 e4 a3 Rg3
14 0 10863 44947113 Qd2 a4 Ra1 a3 Qc1 Qa7 Nb1 a2 Nc3 Qb6 Rb4 Na6 Ra4
Nc7 Rxa8 Rxa8 Rxa2 Rxa2 Nxa2 Nxb5
15 -4 17596 72251056 Qd2 a4 Ra1 Rc8 Ra3 f5 Qf2 Rc7 Rb4 Qb6 Qg3 f4 Qxg6 fxe3
Nxa4 Rxa4 Rxa4 Nxa4 Rxa4 Qxb5 Qxd6 Rxc2 Qxe5
15 21 19884 81607297 Rxc5 dxc5 Ne4 Qa7 d6 c4 Qd5 Qxe3 Kh2 Qa7 b6 Qxb6
Qxf7 Kh8 Qxg6 Qb7 Rf7
16 19 26508 108388508 Rxc5 dxc5 Ne4 c4 Qxc4 Rc8 Qb3 a4 Qd3 f5 Nf2 Rc5 c4
Qc8 d6 Rxc4 Qd5 Kh8 Qxe5
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.