Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Compact encoding of chess positions

Author: Reinhard Scharnagl

Date: 00:44:31 03/05/05

Go up one level in this thread


On March 04, 2005 at 17:33:44, Russell Reagan wrote:

>Final position:
>
>[D]8/r1Q1nQ2/3Qb1qb/q2QQQqk/1n1QKRrq/2NBQQ2/1R1NqB2/8 w - - 0 77

though being beside my restricted subset I will try to calculate its enconding
length (according to my scheme):

Part A: coordinates    =  12 Bits
        color/e.p.     =   2 Bits
Part B: filling matrix =  62 Bits
        color matrix   =  26 Bits
        piece codes    =  78 Bits
---------------------------------
Total                  = 180 Bits
---------------------------------

Facing for facts: that a lot of unrealistic promotions have been made,
that all officers are placed here at inner lines (2 to 7), that after
the 9 white Queens no white pieces are following and that no pawns at
all are used, the maximum of 164 Bits for the restricted realistic set
of positions is not surpassed too much.

Nevertheless such constructed positions are not of my interest.

The highly artificial nature of your example moreover is underlining,
that my restrictions are indeed realistic.

Reinhard.





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.