Author: Omid David Tabibi
Date: 18:48:53 07/21/03
Go up one level in this thread
On July 21, 2003 at 21:27:44, Sherry Washington wrote:
>On July 21, 2003 at 20:01:30, John Merlino wrote:
>
>>On July 21, 2003 at 19:50:20, Russell Reagan wrote:
>>
>>>On July 21, 2003 at 19:05:01, John Merlino wrote:
>>>
>>>>This is true. In the 4-game match between CM9000 and GM Christiansen, Larry
>>>>played VERY exciting chess, with lots of tactics and wild positions. The result
>>>>was, of course, a 2.5-1.5 win for CM9000.
>>>
>>>It sounds like you're saying he barely lost the match despite his playing style
>>>in those games. Do you think he would have faired significantly better had he
>>>employed the "safer" style of play?
>>
>>Actually, if we had wanted, we could have worn him down in the last game and
>>possibly won 3.0 to 1.0. We were up a pawn in the endgame (2 rooks and 4 pawns
>>to 2 rooks and 3 pawns), so we theoretically could have won that. However, by
>>offering a draw we won the match, which was certainly all we could have asked
>>for.
>>
>>But, I would definitely say that, yes, if Larry had played a much more patient
>>style, he probably would have won. But the games would not have been anywhere
>>near as exciting to watch!
>>
>>jm
>
>
> What makes you think he would have won? Van wely certainly played patient as
>well as bareev all they could manage was draws. Christian has years of
>expierence playing computers on icc, he is very familiar with computer play yet
>could not win a match for money.
The problem with most Grandmasters is that they know very little of computer's
thinking process (and their experience doesn't matter). Once Kasparov was
playing against a program (I don't exactly remember which one), and in reply to
1.e4 he didn't play his usual 1...c5, but instead tried to "surprise" the
machine with 1...e5. When the machine replied instantly (obviously, out of
opening book), the puzzled Kasparov asked "how did you know that I was going to
play this?"
Grandmasters know that computers are strong in tactics, weak in endgames, etc,
but they don't know _why_ it is so. I believe that a Grandmaster with a good
understanding of computer's thinking process can still beat even the strongest
programs. The best example is Ilya Smirin's match against the top programs:
http://www.chessbase.com/events/events.asp?pid=122
Shredder Hiarcs Junior Tiger Total
Computers 0 ½ ½ ½ ½ ½ ½ 0 3.0
Ilya Smirin 1 ½ ½ ½ ½ ½ ½ 1 5.0
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.