Author: stuart taylor
Date: 02:43:07 04/29/02
Go up one level in this thread
On April 28, 2002 at 20:26:45, pavel wrote: >On April 28, 2002 at 20:11:36, Mogens Larsen wrote: > >>On April 28, 2002 at 19:57:48, José Carlos wrote: >> >>> Yes, good news! >>> Some years ago, we were impressed and suprised every time a chess program >>>beated an IM or GM, and the programmer received a lot of congratulations and >>>compliments. >>> Now, one of the strongest players in the world wins a game against a program >>>and we find it a great success for humanity. It only can mean that we're now >>>used to see programs winning over and over against strong players. >> >>That's incorrect IMO. The primary effort thus far have been concentrated on >>discovering whether the engine was replaced, sabotaged or operated incorrectly >>when it lost. Not exactly the greatest achievement of humanity. >> >>Regards, >>Mogens > > >Behind every achievement, there is a "flaw". > >The fact is Smirin won (in what a way) against a program (whether GT or CT) in >the best possible setting. > >The argument whether GT is better than CT against human (or otherwise), whether >anti-human=on is better than anti-human=off against human (or otherwise), is not >imprtant, because you cannot be happy with any settings, specially if you lose a >game like that. > >And again you can never have a "perfect" setting, who knows maybe smirin didnt >have a "perfect" breakfast before this game? >Had he lost the game, would he blame the breakfast for it? > >pavs ;) I've even seen great famous games won by such as Alekhine, in which if he was playing the final part (coup de grace) against an average computer program of today, he would have LOST. And he won that against a fellow world champion class player (maybe it was Capablanca). There are many such cases. Similarly with mathematician Euwe, I remeber discovering such a flaw with a computer program, which was not in the annotations, but would have made the results of a famous game, the opposite! S.Taylor
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.