Author: Dann Corbit
Date: 02:34:17 04/28/00
Go up one level in this thread
On April 28, 2000 at 04:28:14, Aaron Tay wrote: [snip] >In fact, if you spent all this time on your program "BeanCounter", BeanCounter >might be giving Crafty a run for his/her/its? money now! No. I am doing things methodically in bits and pieces. Trying weird ideas, and I have not even strung them together yet. Crafty, TCB, and Phalanx dwell in the stratosphere of dedicated effort over a long period of time. The authors are very experienced chess programmers. I expect long strings of failures before I come to any success. Some of my ideas will (undoubtably) thud. And the design itself requires lots of processors (the more the merrier). So it won't be terribly interesting to people who have a single CPU (99% of the consumer market). I'm trying some things just because it is interesting to me, but I know in advance they are not likely to be successful. I'm just doing it for the programming fun of it (e.g. variable sized hash tables ala Popeye). Unusual data structures that hold and delete partial game trees, stuff like that. I'm trying really weird eval functions, and I also want to try fitting the data using a non-linear optimization function and a bunch of off the wall stuff like that. For me, the real goal is just to play around and have fun at it. However, if I should start to whip the competition, the goals might change. ;-)
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.