Author: Carey
Date: 16:05:33 10/05/05
Go up one level in this thread
I'm not a good chess player, and I'm not a good chess programmer, but I'd like to comment too... I think that these days, it'd be hard to write a chess program that couldn't play well. Today's hardware is just so darn fast that even a simple program should be able to do fairly well againt most people. (These days, we've got PDA's and cell phones that are more powerfull than most computers of just a few years ago.) And a deep search can mask a lot of evaluator problems. (Poor tuning, simplistic evaluator, etc.) An intersting experiment would be a new version of "TECH" program. (TECH 4, I guess.) For those of you who don't remember "TECH", it was by Jim Gillogly in the early 1970s. It's purpose was as a technology benchmark between different computers of the time. I think it'd be interesting to do something similar today. A deliberately simple evaluator with typical search features, and a large opening book and endgame tablebases. A nice standardized 'benchmark' program just to show what kind of play you can get with a relatively simple evaluator, and depending on the databases and the performance of the hardware.
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.