Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Any idea's on how long it takes to learn C++, then create a chessmonster

Author: Heiner Marxen

Date: 20:38:50 12/10/01

Go up one level in this thread


On December 10, 2001 at 18:12:05, Gareth McCaughan wrote:

>On December 10, 2001 at 16:28:39, Dann Corbit wrote:
>
>[quoting David Hanley:]
>>> lisp
>>
>> Be prepared to purchase a new parenthesis key for your keyboard once a month.
>> ;-)
>
>Ah, but think of all the wear and tear it saves on the
>semicolon key. Unless you put comments in your code, of
>course. :-)
>
>>
>>> SML
>>
>> What's SML?  Is it related to ML?
>
>SML = Standard ML. I would recommend OCaml in preference, though,
>since the OCaml compiler produces staggeringly fast code.

I do prefer to have a full implementation of the ML Standard.
Like with SML/NJ.

As a matter of fact, I have used SML/NJ to achieve the best known
results for the "busy beaver" function, once upon a time, in 1989,
followed up by several other records regarding this topic,
and since then, nobody has come up with other more efficient implementations.

OTOH, SML is not well suited for efficient implementations of chess,
as long as there are only comparatively simple algorithms to do
a top level chess program.

Once we detect a way/theory of chess, where the meta-level (reasoning)
is an important part of the strength of a program, we will have a good
reason to consider a high level language like SML to implement it.
As long as we think about chess in terms of (rotated) bit boards
(and not much more) we have no need for more than "C".  IMO.

Cheers,
Heiner
--
Heiner Marxen   heiner@insel.de         http://www.drb.insel.de/~heiner/



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.