Author: Robert Hyatt
Date: 14:16:38 01/22/99
Go up one level in this thread
On January 22, 1999 at 01:07:47, KarinsDad wrote: >On January 21, 1999 at 23:37:28, Robert Hyatt wrote: > >[snip] > >>>If Crafty wins a computer/computer tournament against strong competition, would >>>you not feel good about it? If it's games had slight imperfections that other >>>computers found out under further analysis, would you be less happy with the >>>result of winning the tournament? >> >>Sorry, but you totally misread me. Ask on ICC. I'm much happier with a well- >>played loss than I am with a blunderful win. Because blunders won't always >>be lucky enough to win, while good play will eventually prevail... > >Ok, you've totally lost me here. If you have a well played loss, then you have >by default blundered, at least once. There is no other possibility. The >blunder(s) may be infinitesimally obscure, but you have made a mistake(s) none >the less. In my example, it was not that Crafty has made obvious blunders, but >that it took analysis from a series of other computers to find the obscure >mistakes (which it still makes in it's current incarnation) that it made. >Knowing that Crafty played extremely well, but not perfect was the question. > reread the above. "I'd be much happier with a well-played loss than with a blunderful win."... Well-played loss _may_ mean a blunder in one move. Or it could be choosing a bad opening but fighting almost perfectly but in a lost position. On the other hand, I do not get excited about winning when I can find lots of positional mistakes but then win on time or win due to a simple tactic the opponent overlooked. The two are far different. >I can understand the position (which is what I think you were saying) that you >would rather lose with a series of infinitesimally obscure mistakes than win >with a bunch of more obvious mistakes. That's a fair statement. correct... or lose with _no_ mistakes, because it is possible that I come out of book a pawn down to start with. And it may play wonderfully from that point forward and _almost_ win. > >But a different way of saying my question is: Would the fact that a bunch of >computers running around the world for several hours or days analyzing the games >that Crafty played in such a tournament and finding a series of infinitesimally >obscure mistakes which would have eventually resulted in one or more draws or >losses which would have prevented Crafty from winning the tournament detract >from your joy that Crafty won the tournament? Or would you just be proud of your >accomplishment and not worry that Crafty wasn't perfect? I would definitely worry. I have won my share of computer vs computer games, even at the world-championship-level, where my program was outplayed but won due to a trick it found. But you don't 'live' by tricks, you die by them. Because you can't plan on your opponent making mistakes (computers are getting better and faster). > >If the answer is yes, you would be happy and proud, doesn't the same apply to >this game? Especially since the player of the "bolt of lightning" move is the >one who had the victory. If the answer is no, then I must admit that I feel >sorry that you are missing one of the great joys of life. The thrill of victory >unspoiled by the petty comments (or analysis) of others. I enjoy watching it win... but that doesn't mean I don't/can't watch with a critical eye towards its mistakes. I can certainly enjoy brilliant move, and disdain dumb moves. But if crafty plays a brilliant move that later turns out to be wrong with further analysis, I certainly look at it to see if I can fix the problem... because I worry that the _next_ time it might not get away with 'smoke and mirrors.' So don't feel 'sorry' for me, I see the faults _and_ the good things, and both give me plenty of enjoyment, the former because I have to find solutions for them, the latter because I have to avoid breaking them when I fix other things... > >>> >>>The move is brilliant not because it is a guaranteed win, but rather because it >>>contributed to winning the game in a spectacular way, both on the board and >>>psychologically. It matters not whether a computer can detect that it is a >>>lemon. It only matters that Topolov could not. Isn't that the beauty of playing >>>chess? >> >>A matter of perspective, I suppose. But something about 'hoping the opponent >>won't play the best moves' leaves me a bit cold, I think... > >Correct me if I am wrong as I have not read all of the posts yet, but the move >Rxd4 has not been shown to be an obvious blunder. From what I've read, it hasn't >even been show to be a substantially weaker move. If it takes hours or days for >people running the best computer programs around to find weaknesses in the play, >doesn't that imply that at least good chess (if not great) was played? Of course >a computer program from 100 years in the future that plays nearly flawless chess >may find minor mistakes or even a lemon in a matter of 30 second on this game, >however, what would that matter? Fifty years ago, we would not be having this >conversation since there would be no computers and very few GMs who would have >been able to analyze the game to the degree that such minor mistakes could be >found. > >We don't live in a perfect world. But one aspect of the world that we do live in >is that at all levels of human versus human chess, there is always a human on >the other side of the board who along with his or her strengths, also has his or >her weaknesses (such as being psyched out by a very strong move). Garry used >several weapons in his arsenal, not just good chess playing. This is very smart >on his part and illustrates why he is the champion that he is. > >KarinsDad > >> >>> >>>If the move would have lost for Kasparov, most of the chess world would have >>>admired Topolov for coming up with the refutation over the board and the >>>sacrifice would have still been remembered. A footnote in chess history would >>>have been made in either case. That also contributed to making the win brilliant >>>(or at least spectacular). >>> >>>KarinsDad >>> >>>> It >>>>might have turned out even better for him. Or it might have lost if the machine >>>>played Rhe8 and that is good enough to hold on. >>>> >>>>Everyone knows my opinion of 'Kasparov, the man'. I still respect and admire >>>>'Kasparov, the chessplayer' however. And this was only about 'the chessplayer' >>>>and the move he played... was it good, bad, or just legal? >>>> >>>> >>>> >>>>> >>>>> >>>>>>Best regards, Jeroen
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.