Author: Mig Greengard
Date: 06:29:06 11/25/03
Go up one level in this thread
On November 25, 2003 at 08:36:03, Omid David Tabibi wrote: >Nunn positions are by no means equal. They are balanced, from the average point >of view. But even humans will prefer to play one side or another in Nunn >positions, not any given side. My point is not which positions in particular, just balloted to avoid the need for human interference with the book, or avoid a book at all. Of course there would be a preference but as long as they are all acceptably balanced and random that would be fine. 200 approved positions, for example. >>Otherwise it seems to me that you >>just try to cover up the weaknesses of your program by tweaking the book to >>avoid the positions it doesn't handle. > >Exactly, like humans do. As a human, when you don't like playing against French >for example, you choose 1.d4. To "cover up your weakness" in that opening. Each >program has a style of its own, and I don't see any problem in trying to avoid >uncomfortable positions. That means you are in favor of humans-as-chessplayers helping the machine, not only humans-as-programmers. I recognize that comps aren't yet ready to play without books, but the current state of heavy influence by a team, usually with GMs, on picking the openings - even for each game - is moving away from programming a computer to play chess. Having a significant percentage of decisive games in Graz, for example, decided in the opening books would be a shame and a major waste of time in my opinion. It's only marginally less silly in man-machine. When I test an engine I don't even bother with the opening book. It's just a distraction from how well it evaluates positions and finds tactics. >>And that should be contrary to the goal >>of making a good chessplaying machine. > >BTW, assume we ban the use of opening books. How are you going to enforce that >rule?! You don't let me have an opening book file, fine, I will create an >in-built book within the engine. There is no way you can stop that from >happening. This is very much a separate issue, and one I mention in my original message. I doubt a technical solution is impossible. For now I'm interested in the theoretical argument in favor of books, since they continue to be used enthusiastically. Are they just a necessary evil? Or does everyone think they can use it to their advantage? Is there no movement to abolish them or limit them?
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.