Author: Robert Hyatt
Date: 05:18:44 03/09/99
Go up one level in this thread
On March 09, 1999 at 07:04:10, Nacho Bidnuz wrote: >On March 08, 1999 at 22:07:11, Peter Hegger wrote: > >>On March 08, 1999 at 20:14:22, odell hall wrote: >> >>> >>>On March 08, 1999 at 02:44:36, Jouni Uski wrote: >>> >>>>Because any human cannot beat Kasparov in match... >>> >>> >>>Why a "new deeblue"? The "old" deeblue was enough to do the Job. >> >>And a new deep blue running at 10x speed as the one that defeated kasparov >>(technically feasible) would crush him like a ripe grape. > >I find it increasingly difficult to understand what sort of mentality it takes >to believe that a machine that has a 5-7 score against Kasparov is actually the >better player. Okay, the 1997 version of Deep Blue was faster than the one he >trounced in 1996 (or so says IBM), but was it really better? yes it was better. The 'chess processor' was completely redesigned for the last match, to allow them to do the things in evaluation that they felt that they needed to compete (IE to find out which pieces are directly and indirectly attacking the king, and so forth). It was faster due to more processors, it was far 'smarter' due to a complete hardware redesign. >As far as I can >tell, the "improvement" might just as well have been due solely to the addition >of GM's Benjamin, DeFirmian and Fedorowicz to the IBM team. They just MIGHT >have been seeking a measure of revenge against the man who said he could give a >simul against the American Olympic team. no idea whether that is true or not, but I don't quite get the 'point'. IE revenge or not, what can you do to beat a "Kasparov"? Other than to make the program faster, smarter, more robust, etc... > I can understand people who don't like Kasparov for his arrogant and often >crude remarks about others, but I find it very hard to side with those who would >invent a fantasy Big Brother of a chessplaying monstrosity (e.g., Deep Blue >analysing ten trillion moves per second) to beat up this "school bully" of a >world champion, but does anyone seriously believe that a one-point margin of >victory in a brief match conclusively establishes the superiority of Deep Blue >over Kasparov? of course not. It proved that during _that_ week, DB was better. But that is all. It strongly suggests that DB is not significantly worse than Kasparov, and weakly suggests that DB may be better. But that is all... > One could dismiss such people as the sort of envious mediocrities who are >always hanging around to hate the good for being good. But what's there to >disparage about his chess? One could say the same about 'Deep Blue' in fact. Those that 'despise' this machine only despise it because it is 'big iron' made by IBM, rather than a micro program. Can you imagine the difference we would be seeing had that match been played (and won) by _any_ commercial chess program running on any kind of hardware at all? Things would be _vastly_ different, IMHO.
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.