Author: Uri Blass
Date: 10:21:29 06/30/02
Go up one level in this thread
On June 30, 2002 at 12:26:30, Mike Byrne wrote: >On June 30, 2002 at 09:49:32, Art Basham wrote: > >>This is one of my favorite moves! >> >>[D]r2q1rk1/pb3ppp/2n5/1pbNP3/2p1Q3/P4N2/BP3PPP/R4RK1 w - - >> >>White plays Nf6+..! > >I'm not sure why it takes my modified crafty so long - almost 13 minutes This problem is a positional problem. Some programs can find it faster but it does not mean that they have to be better because they can also find wrong sacrifices. It is easy to see that Nf6+ gives white at least a draw so if a program believes that black is better it is going to have no problem to find it. The second way to find it is by having big positional score but in this case the program may also play wrong moves in other cases. It is easy for programmers to change their program to solve this problem but the target of programmers is not to solve this problem but to have a program that plays better. Even after the first 6 plies it is hard to see that white is winning. Movei cannot do it in a short time and has a score of 0.00 after the first 6 plies(1.Nf6+ gxf6 2.Bb1 Re8 3.Qh7+ Kf8). 4.Be4 is winning but movei cannot find 4.Be4 even at depth 10 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.