Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: What constitutes a clone?

Author: Vasik Rajlich

Date: 03:00:24 02/17/05

Go up one level in this thread


On February 16, 2005 at 04:38:27, Andreas Guettinger wrote:

>I would recommend to start with a blank page. Starting from a given (freeware)
>program for me is cloning. I also don't take a short story, write ten chapters
>of my own and sell it as a book.
>
>If you look at code of other available engines to take over ideas is perfectly
>ok for me. However, never copy 'n' paste, never take over tables and arrays of
>evaluation data.
>
>When you write a chess engine, it's normal to start 3 - 5 times again from the
>beginning and rewrite your whole code due to change of important data
>structures, etc. This makes code quite unique.
>
>Do no optimize-cloning. Normally the routines of a major chess playing program
>(like i.e crafty) are faster than the ones written by yourself. Just live with
>it, leave it for later or try to find the bottleneck.
>
>For me, writing a decent playing chess program as a hobby project (maybe
>different if you do it for your thesis or as a professional) takes at least 3-5
>years. All engines that develop faster for me are highly suspicious.
>
>Just my two cents.
>(programming since 3 years and it still sucks) Andy

Fruit appears to have been developed in something like one year.

Ruffian according to the author took around two years.

I think most amateurs, myself included, waste some effort in the beginning doing
the wrong things. By the time they figure it out, a year or two is gone
(although it feels like much less).

Vas



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.