Author: Keith Hyams
Date: 02:58:49 08/23/05
I have been following the debate on cloning with interest. A substantial amount
of every modern program is a cut and paste job. Programmers,whether they realise
it or not, put pieces of code written by other people into their programs.
Preprocessor directives and .dll files are two examples of this. The only way in
which you may be able to avoid using pre written code is to choose to program in
machine code and it is debatable whether even this solves the problem .As the
languages we use become higher level, the amount of code pre written code that
we use increases. No sane person worries about this.
I am writing this because the idea of producing chess engines that can be used
as pre written code interests me. I like to speculate on the attributes that
would be required for one of these 'libraries' to be useful. For example Crafty
may not be useful for this purpose. Professor Hyatt has stated that Crafty is at
the stage where it will progress by evolution rather than revolution. Crafty is
complex and mature. Fruit, on the other hand may be more suitable SMK, the
author of Shredder,described the code of Fruit as being “very clean”. How about
TSCP?
I wonder how the approach that Professor Hyatt used to construct Crafty would
have been different had he started to write it today. If you were going to
construct such a basic engine you would look for simplicity, clarity and
development potential. What features would you include and what features would
you avoid if you were trying to achieve this?
Changing the subject, it seems to me that when I play matches in Chessbase, the
engine playing from bottom to top seems to do better than the engine playing
from top to bottom. Has anyone else noticed this?
Keith
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.