Author: Dieter Buerssner
Date: 14:31:32 07/20/01
Go up one level in this thread
On July 20, 2001 at 16:47:29, Dann Corbit wrote: >[...] I think there will always be a gap between the >professional engines and the amateur engines for 3 reasons: >1. The professionals get to spend a lot more time developing >2. The data in the form of opening books will always be better because amateurs >are not capable of assembling a professional quality opening book It seems, that Chessbase is delivering Tiger and Shredder with a more or less "random" book, just like Crafty. Still, these programs obviously do well with their book. However, no doubt in "real" tournaments, they will use different books. We won't see those books. I guess, they are rather well tested against their main opponents in autoplayer games. Booklearning will not make any difference in a Swiss tournament, where each opponent is played once at most. So, it should be no big problem, to have a really tight book, which would be inferior for typical users for various reasons (including the booklearning in matches against the same opponent). >3. Most amateur innovations are openly discussed and most profesional >innovations are kept secret. 4. Professionals have much better testing facilities. If I remember correctly, Christophe Théron has mentioned, that he has 6 computers just for testing. I think, most amateur programmers have not those means. I believe, it is much faster to code a new idea, than to test, if the idea works. Or, if the idea is in some way parameterized (e.g. by evaluation parameters), to really even start to tune those parameters, without doing some gambls. Regards, Dieter
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.