Author: blass uri
Date: 02:18:44 05/02/99
Go up one level in this thread
On May 02, 1999 at 03:56:26, odell hall wrote: >Hi CCC > > >Hi > > I agree with IBM" s decision to withhold games From Kaspy Before the 1997 >Return Match. I have no doubt had Kaspy had atleast 20 games by deepblue he >would have easily defeated the machine. I do not agree. I think that seeing 20 games could help kasparov but this is not an easy task. If the machine is strong enough then Kasparov have no chance against it even after seeing 20 public games. I think it is grossly unfair to allow >the grandmaster access to Games before a match. My reason is that humans have >the advantage over the machine in that they have the ability to analyze the >style and weakness of the computer opponent, Yet computers do not have this >ability or chance, only if the programmers did not teach them to analyze the opponent games. thus they are at a considerable disadvantage. A programmer >can similate in a very crude fashion this innate human ability by imputing >thousands of games or positions into the computers memory, However my >understanding is that Most programmers are not sure if this technique is >beneficial or how much so. In any case it does not come close to the Abilities >that humans possess. In my opinion if the object of a match is to test the raw >ability or talent of the contenders, no such games should be given. Kaspy has >been known to defeat grandmasters at home before he ever goes to the board There are big numbers of openings. Humans often play the same opening again and again but against computers it is a different story and if he can only see games and cannot play against it at home then I cannot see him winning by opening preperation. , the >question is , Is this Raw chess ability? or Superior Analyses Training? Kaspy's >charge that he loss because he had no games from the computer is Valid, But the >Real debate should be Whatever such information should be made availabe when a >contest is about raw ability or talent. If Garry Kasparov is the Better Player >or Has more ability than deepblue he should be able to defeat his opponent >without knowing anything about it! If you want to be fair then the programmers of deeper blue should not analyze games of kasparov when they prepare deeper blue against kasparov. Ofcourse in Human Vs Human Play he would be >totally justified in making this cliam, but in a contest with the computer his >claims are not justified. I think that Yermonsinky will Probally easily defeat >Hiarcs7, since he has access to the program. In my opinion Matt Frank should >have gotten Yermo to agree to a match against a computer but not specify which >computer opponent he would play. This would have allowed a fair contest!! The case of Yermolinsky is different because Yermo can play against Hiarcs7 at home and not only see games of Hiarcs7. 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.