🚀 go-pugleaf

> I am not sure what you refer to as "ancient Indian chess". If my
> memory is correct, they have quite a lot of variations. (No wonder,
> chess was invented there...).
>
> Now, as to modify an existing source code to change the moves of the
> pieces, that can be easily done IMHO. Beside some careful screening of
> the whole code for "embedded" rules (like captures en passant and
> castling), you should be able to make straightforward modifications to
> specific arrays that describe the moves. Gnuchess has (if I recall
> well, again), 2 such arrays: 1 for directions, and 1 for steps.
>
> Of course, the patched code will play incredibly bad, since the
> evaluation function that contains all the heuristics will be no longer
> valid. That also can be changed by code screening, and by modifying
> some arrays (in Gnuchess again).
>
> Overall, the skeleton of the program and the search algorithm should
> be ok, even though they have been specifically engineered and
> optimized for European chess (timing considerations, etc...).
>
> I think it might be as much time-consuming to reverse-engineer the
> code of a chess program and adapt it to your needs than to build it
> from scratch, with the benefit of the accumulated experience in
> European chess programs.
>
> The people who wrote Gnuchess will surely be of better advice.
>
> gilles

I have had a 3rd-year student spend 300+ hours re-engineering GNU Chess
for Japanese chess.

Now I have another student continuing with the project.

I expect (hope) to have a completely documented GNU chess code in 2 months.

All of the data elements of SHOGI have been completely described in about
50 pages of text last year.

I hope to have a description of the exact method of timing control, look-ahead
pruning technique, and transposition method within 4 months.

These will then be added to the current state of the program SHOGI.

I believe it is  m u c h  easier to work with code already written that
works well than trying to re-invent the algorithms.

However, GNU code is VERY hard to follow as the documentation is some of the
worst I have ever seen.

All of the work we do will be provided for the GNU foundation when it is
in a sensible form.

raymond

--
University of South Australia   | Act in haste and repent at leasure
P.O. Box 1                      | Code too soon and debug forever
Ingle Farm                      | Knobs, knobs everywhere,
South Australia                 |              just vary a knob to think!