Author: Don Dailey
Date: 17:11:38 06/24/98
Go up one level in this thread
On June 24, 1998 at 18:23:58, Eran wrote: >On June 24, 1998 at 13:11:29, Robert Hyatt wrote: > >>On June 24, 1998 at 11:46:06, Eran wrote: >> >>>On June 24, 1998 at 09:51:24, Robert Hyatt wrote: >>> >>>>On June 24, 1998 at 02:00:31, Eran wrote: >>>> >>>>>On June 24, 1998 at 01:12:06, Bruce Moreland wrote: >>>>> >>>>>> >>>>>>On June 24, 1998 at 00:49:24, Eran wrote: >>>>>> >>>>>>>Rebel9 found Re1xe6 in 1 second!! >>>>>> >>>>>>No ill will intended toward anyone, but doesn't Rxe6 *lose*? >>>>>> >>>>>>bruce >>>>> >>>>> >>>>>This is not about whether or not the move Rxe6 loses. The only thing is that I >>>>>tried to show to all of you that Rebel9 with different settings in the Option >>>>>menu clearly calculated faster with that specific chess position as mentioned >>>>>than anti-grandmaster chess software. Ed Schroder and anybody else may not know >>>>>that. So I just tried to show that to them. >>>>> >>>>>Now, as to the question of whehter or not the move Rxe6 loses, I cannot tell >>>>>clearly. It depends on who the opponent is and what his or her playing style is. >>>>> >>>>>Eran >>>> >>>> >>>>however, what's the point in finding settings in a program that suddenly >>>>makes it start playing *losing* moves? That is usually *easy* to do. THe >>>>hard part is to make it play *winning* moves... :) >>> >>>Okay, but if the move Bh6 is better than Rxe6, then I do not know why Rebel >>>won't play Bh6 after one or two minutes of calculation. >>> >>>I think Ed Schroder believes that the move Rxe6 is a better anti-grandmaster >>>move for instance in hope to make Black becoming tangled. However, it sounds to >>>me that the move Rxe6 is subject to controversy; it looks like "chicken-egg >>>cycle." It seems too difficult to know for sure whether or not Rxe6 is a good >>>anti-grandmaster move right now. >>> >>>Anyway, the anti-grandmaster features in Rebel10 are just an experiment. I hope >>>the anti-grandmaster chess software which Ed created would beat Vishy Anand in >>>July. I think that if the anti-grandmaster features are good, Ed would retain >>>them in Rebel10. Otherwise they would be thrown away. Correct me if I am wrong. >>> >>>Eran >> >> >>I don't think there is any controversy here at all. Rxe6 loses. It will >>lose against a computer. It will lose against a GM. It will lose against an >>IM. Who cares if it wins against masters. We can already do that without >>tossing material out the window... and self-attacking... > >So, why does Ed Schroder think anti-grandmaster moves are a little better such >as the move Rxe6? Ed is one of the best chess programmers in the world, and I am >still wondering why your saying is very different from what Ed thinks.... > >Eran Ed is not a Grandmaster, only a "top computer chess programmer." He could be wrong or the others could be wrong about that specific position. The anti-GM style of play and its value is a different issue altogether and there is plenty of room for other good programmers to agree or disagree about this one. The top people in any field usually disagree about many issues so don't be surprised. I have no doubt that this is possible, I am much less sure about how easy it will be to implement succesfully. If it's practical, someone will eventually do it. - Don
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.