Author: Harald Faber
Date: 05:32:55 05/05/99
Go up one level in this thread
On May 05, 1999 at 07:51:57, Micheal Cummings wrote: >>>>r1b2rk1/pp2nppp/2n1p3/q1ppP1B1/3P2QP/P1P5/2P1NPP1/R3KB1R b KQ h3 0 1 >>>> >>>>This position is from a game between Rebel8 and Saitek Sparc(1996) >>>> >>>>Saitek Sparc like other programs falled in a trap and played 10...Nxd4 >>>> >>>>Can your program avoid 10...Nxd4. >>> >>>I don't have them here at work (it wouldn't help, CM doesn't run in NT) but I >>>guess the only programs "smelling" the danger can be Chessmaster and CStal. All >>>others certainly have no evaluation functions for this longtime danger. (I tried >>>Phalanx and WChess) >>>This is an excellent example for the importance of king safety. The black loss >>>is too far because black can't resist the temptation to give some checks which >>>push the loss beyond the horizon. >>>Any programmer confirming this? >> >>Chessmaster6000 cannot see the danger even some plies after Nd4 >> >>Uri > >After 30 mins CM6K still likes Nd4, do not know if it likes it with more time >though, All my programs only like Nd4. Shredder 2, and Rebel 10 all the same. So I think no program will find this, the loss is too far.
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.