Author: blass uri
Date: 00:39:54 05/11/98
Go up one level in this thread
On May 10, 1998 at 21:19:45, Fernando Villegas wrote: >Hi Bob: >I just downloaded crafty 14,9 and as much as never I have been capable >to get a decent library for it it happens I currently polay without it >againts your criature; is, besides, a way to do things differently. >Well, to my amazement Crafty followed thjis line: > >d2d4 d7d5 >c2c4 dxc >g1f3 Qd5?? >b1c3 Qa5 ?? >Bd2 Qb6?? >e2e4 c6 >Bxc4 Qxbp?? >O-0 e6 >a3 Qb6 (forced, of course) > >I left here the game. As you see, black has not developed nothing and >white is a model of development and many threats are coming. Just e5 >lets black desperately cramped, to begin with. > >Now my question: how a so powerful and reputed program can play so badly fritz5 evaluate after some minutes black has adventage in the final position. Junior4.6 evaluate the position as almost equality 0.12 pawns for white they suggest to play d5 after Qb6. I do not think Qxbp is a mistake because black has no good option fritz5 and Junior4.6 suggest this move. The mistakes were before this move at moves 3-5. after Qxbp the computer suggest Rb1 Qa3 with 2 good moves Nd5 to take the queen or Rb3 Qd6 Rxb7 Bxb7 Qb3 >just because he is not aided by the book? Don't you expect from a >modern, top program a kind of decent understanding even if he has no >book? Not even a beginner with a couple of lessons play Qxbp or just go >with the queen from one place to another. Or Crafty 14,9 just went badly >out of the production line? > >My feeling is -yes, feelings; I don't dare to have "ideas" any more in >this so technical group- that a good test of chess understanding of >modern programs is how well or bad they play the opening without the >ortopedic aid of a book. I thought that they had to have an inbuilt >device in the code to avoid such a awful play. My computer is a 200 MMX >intel. I hope you tell me that Crafty 14,9 was a disaster and has been >condemned to firing squad... :-) >fernando computers do not understand positions when one side has a material adventage and the other positinal adventage. my nunn match between Junior and fritz(1000 seconds for move) prove it I will post a game from this match when both program evaluate white positon as better (a pawn adventage and white has a pawn adventage on the board but could not castle) white lost in this game.
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.