Author: Marc van Hal
Date: 11:48:42 06/24/02
Go up one level in this thread
On June 24, 2002 at 13:45:59, pavel wrote: >On June 24, 2002 at 12:53:46, Marc van Hal wrote: > >> >>Well you know I am someone who prefers the threuth above nonsens >>I rather see a program who can reproduce at least 10 perfect games >>Then a 1000-0 score against Fritz . (the name fame and glory can all be false >>based on this.) >> >>Though mostlikely it will win more games if it can play perfect games! >>Fritz can come out of openingsbook in a position it doesn't like. >>It can be to positinal . >>Maybe the Nunn or Knaak openings would be better. >> >>The games which where shown on chessbits where all openings which where >>more prefereable for Rebel then for Fritz >>This already doesn't give me a good cleu of the strenght of RebelXP. >>Only of the weaknes of Fritz! >>So intresting for Frans You only can learn from mistakes and not from your won >>games. >> >>Getting info about the time used for epd diagrams >>Doesn't give me a cleu on how strong it is neither. >>It doesn't show how the game play is >>Which also was proven by the Rebel Century3 contest! >> >>Regards Marc > > >Define perfect game. >And tell me how will you know when it's a perfect game. > >pavs A perfect game is a game where every move is good. And keeps good after deeply analysing and I don't mean analyses of computer programs alone. So the games are of theoretical importance! The second defintion is givven in the first defintion. And believe me some programs have played such a game. Regards Marc
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.