Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Shortening Huffman coding + OT

Author: Guido

Date: 01:58:38 10/31/99

Go up one level in this thread


First of all  I thank all the persons who replied to my message with very
interesting arguments.

I am not expert in file compression algorithms and therefore I have nothing to
say about it, but only to try to understand.

But remaining in the original problem, I agree with KarinsDad that extracting
kings from the square, queens can be coded with only 5 bits (and 4 bits if in
the first or in the last row).

I agree also with Ratko V Tomic that in 12 bits we can  store positions of the
kings with every combinations of castling status.

Now, if we put these fixed 12 bits at the beginning of the bit string, we can
avoid every bit for the kings and also for the "castlable" rooks, so our board
will be made of a number of square variable between 58 and 62 squares (but we
already know where are kings and rooks to insert in order to define the first
and the last row).

In this case, full castling status reduces the dimensions of the string of  4*4
= 16 bits (rooks in original positions take only 4 bits), while in the case of
no castling rooks will be obviously present.

Regards
Guido








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.