Author: Uri Blass
Date: 15:01:11 07/15/02
Go up one level in this thread
On July 15, 2002 at 15:44:55, Gian-Carlo Pascutto wrote: >[D]r2qrbk1/5pp1/p1bp3p/1p2p3/1P2P1Q1/2P4P/P1B2PP1/R3RNK1 b - - 0 21 > >Sjeng played d5, the same move was played by Ponomariov earlier >this year in Linares against Adams. Alex Kure criticised it as being >premature (citing Adams). Sjeng got into a worse position and Brutus >slowly worked its way through a win. > >How long does your favorite program need to avoid it? > >[D]r2q1rk1/p2bbppp/1pn1p3/4P3/2pP1N2/2P5/P1B2PPP/1RBQ1RK1 w - - 0 15 > >Brutus played Qh5 and went on to win after the double piece >sacritice. I had a discussion with Tony Werten, who claimed >you had to search really really deep to see the compensation. >I think you just need good kingsafety :) I think that you just need good extensions. It is known facts that programs have no good knowledge to extend the right lines and Movei is no exception. Another question is what do you mean by a winning score. Suppose that I can see Qh5 with +0.97 score when another move has +0.56 score. Does it mean that Qh5 has a winning score? 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.