Author: Uri Blass
Date: 07:30:22 01/27/06
Go up one level in this thread
On January 27, 2006 at 09:39:51, Derek Paquette wrote: >On January 26, 2006 at 17:50:15, Uri Blass wrote: > >>On January 26, 2006 at 17:41:37, Derek Paquette wrote: >> >>>On January 26, 2006 at 17:19:43, bryan wrote: >>> >>>>With Topalov showing that he will often exchange rooks for knights or bishops, >>>>would a chess engine ever choose to make such sacrifices? Are the piece values >>>>locked at the core of the evaluation function, or would it be common to lower >>>>the value of a rook bewlow that of a bishop or knight depending on the position? >>>> Specifically, would a computer ever play 18.Rxe4 as Topalov did against >>>>Aronian? >>> >>>I read this post and thought for a minute about which programs might be doing it >>>already, then I thought...even if no program does this, Topolov can't beat any >>>of the top 10 programs on the SSDF in a match anyway, so it doesn't matter. >> >> >>We do not know that he cannot beat them in a match. >>This is only a speculation. >> >>Note also that Fritz9 and rybka can find the sacrifice after enough time. >> >>>Maybe playing like Topolov is a weakness...? >> >>No >> >>I suspect that topalov learned some good ideas from analyzing positions with >>chess programs. >> >>Uri > >Topolov finished below the programs in Man vs Machine, Kasp can't beat them, >Kram can't beat them, Bareev can't beat them, and programs are just getting >stronger, if you were a betting man, would you bet on historical evidence such >as what i just said or on your gut instinct? Topalov finished below programs when he was weaker player. I also doubt if the games were very important for him. We need big prize money for winning the match and nothing for a loss in order to give more motivation for the top humans. 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.