Author: David Hanley
Date: 09:06:18 11/13/02
Go up one level in this thread
On November 13, 2002 at 04:32:11, Jorge Pichard wrote: >This was article was copied from, Mig is the editor-in-chief of >KasparovChess.com. > >"So we know that Deep Blue was much faster than anything around today, even >hundreds of times faster. But does that mean better? You can always count on >Amir Ban to cut through the hot air and he does so admirably here, "I've written >on this several times, and to summarize my position, it is that a machine that >is much better than all the others must be shown to play an objectively good >move that the others don't, or at least the others need much more time to play >it. That doesn't really follow. A chess game is not producing one good move, or even several. It required playing many good moves in a row, as well as no real blunders. You could show a great GM game to a bunch of 1200 players, and ask them to guess every move. At least one 1200 will guess right every time. That doesn't mean the GM is no better than the 1200 player. This is Amir's argument, though. > Falling back on theory and one >fluke result of six games is no substitute for empirical evidence. Deep blue played a lot of games. It blew over all the computers ( except fritz, one time ) in the ACM contests. It played a lot of GM games and matches. I don't think this put-down of DB is intended as an honest argument. dave
This page took 0.01 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.