Author: David Blackman
Date: 06:37:48 08/12/00
Go up one level in this thread
On August 11, 2000 at 08:36:41, Dan Homan wrote: > >I should probably point out that by specifying "programmers" and people with >the "interest and determination", I've already included people with a good bit >of natural ability. I just don't think there is anything special about chess >programming that requires abilities above those necessary to create any other >kind of (reasonably complex) program. > > - Dan I think it's not too hard to write a program that plays legal moves. Any competant programmer should be able to do that with no more knowledge than the FIDE rule book on the desk beside them. Cooking up a minimax, or even alpha-beta search from vague descriptions that are available all over the net should also be easy enough. You could sneak a quick look at TSCP if the descriptions don't make enough sense. Beyond that, if you want to write a STRONG program, it gets much harder. The programming side is tough because it involves a lot of performance measurement and tuning and low level algorithm selection, and that side of programming is something that most programmers rarely run into any more. The chess heuristics side is even tougher if you don't know much about chess. It's hard to stop your program making mistakes when you don't even know it's making mistakes. That said, there are a few weak chess players, and even non chess players, who have written very strong programs. I guess you can read books, tune purely by game results, or maybe by neural net or something, and concentrate your main efforts on algorithms to go fast and minimise node count, both things a programmer without much chess understanding can easily measure.
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.