Author: David Blackman
Date: 22:00:29 02/13/00
Go up one level in this thread
On February 12, 2000 at 23:50:18, Vincent Vega wrote: >I don?t see any other use for 90% of amateur programs than allowing their >authors to learn programming. While learning to program is a good motivation, >it appears that many experienced programmers are also creating Yet Another Chess >Program. The reason it?s such a waste of time is that most of them just use the >obvious pattern - single threaded, alpha-beta, fast searcher. Why? Because >it?s easy. But there are so many real challenges out there: big knowledge >programs, programs that play human-like, use of genetic algorithms, use of >neural nets, parallel programs, distributed programs, programs using pattern >recognition, algorithms for quantum computers, etc. that are waiting for >advances. Instead what we get is another Crafty-wannabe. Sigh. I have mixed feelings about this. You can learn a lot about computer chess, and even about chess, by trying to build a strong program using conventional methods. The stuff you learn doing that will be very useful if you want to try something more radical later. Most people who suggest radical ideas for chess programs, but have not first written a conventional one, just have no idea what it's about. Their suggestions are either so lacking in detail as to be useless, or simply won't work. On the other hand, once you get a conventional program working, it gets a bit addictive, and it's hard to break free and try anything else.
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.