Author: Dann Corbit
Date: 21:06:48 02/04/00
Go up one level in this thread
On February 04, 2000 at 21:04:35, Rich Shippy wrote: >Since deep blue used 2.c3 to beat Kasparov, is this the best move for white >against the Sicilian? What does computer analysis say? You probably want a database, not computer analysis results. That early in the openings, computer analysis stinks. It takes about 8 hours of computation to get even a decent choice for early opening book positions. For the openings, computers will occasionally find hidden tactical snares that are 6 fullmoves deep or so and nobody saw them yet. But these are very, very rare. Now, with a database, you can load the data from hundreds of thousands of GM games and see what they did. Then you can look at computer analysis and see if the computer is full of balogna or not. A bare computer, buzzing away for 6 minutes on a move a few plies away from the origin will produce tripe. Because openings are repeated so often, humans have thoroughly debugged them. You won't defeat a good human player by doing what a computer says to do in the opening unless it is just following along in a preanalyzed opening book composed from GM games (and possibly some computer analysis to bolster it). Eventually, we will be able to string together successive analysis results produced by computers and come up with something much better than we can right now. IOW -- you're barking up the wrong tree.
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.