Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Bruce Moreland's Gerbil

Author: David Rasmussen

Date: 17:08:44 06/09/01

Go up one level in this thread


On June 09, 2001 at 18:07:49, Ricardo Gibert wrote:

>The following finger notes can be found on ICC:
>
>
>
>Information about gerbil(C) (Last disconnected Sat Jun 09 2001 03:34):
>
>          rating [need] win  loss  draw total   best
>Bullet      1427  [7]     1     0     0     1
>Blitz       2032         49     7     5    61   2032 (09-Jun-2001)
>
> 1: Gerbil
> 2:
> 3: A simple chess program by Bruce Moreland.
> 4:
> 5: First line of code written on 28-May-01.
>
>
>
>What's the scoop on this (very) new chess program? Why are you (Bruce) starting
>from scratch? What are your aims with this program? Is this intended to be a new
>version of TSCP or do you intend to be more ambitious? Inquiring minds want to
>know.

Bruce said in rec.games.chess.computer that Gerbil was a new educational engine
that he was working on. That is, an engine that is supposed to be as simple and
clear as possible for the purposes of understanding how a chess program works.
This came up, because we where talking about that I am thinking of doing the
same. It seems that most such programs are a compromise between simplicity and
performance. In my educational program, if there ever is one, the important
principles will be:

1. Simplicity and good design is _always_ more important than performance.
People could always look elsewhere for performance, whenever they understand the
basic design. The performance critical things are the things that should be
experimented with anyway, instead of just copied.

2. Good, simple design and performance are not enemies. They emphasize
eachother. A good design will eventually lead to fewer bugs and better
performance.



This page took 0.01 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.