Author: Kolss
Date: 07:00:24 03/10/05
Go up one level in this thread
Hi, Here is a suggestion for a (perhaps most) parsimonious way of coding castle moves in Fischer Random Chess (FRC): If normal moves are coded as e.g. "e2e4", it would be nicest to have the same for the castle moves. As in FRC, however, this can end up being ambiguous for king moves (e.g. "f1g1" could be a normal king move or a castle move), this could be prevented by coding the castle move as [king_from_square][rook_with_which_king_castles_square]. Then the castle move would be coded as "f1h1", while the normal king move would be "f1g1". This also catches the case with king on b1, rook on a1: king move to c1 is "b1c1", long castle is "b1a1". "b1a1" cannot be a normal king move, as there is an own rook there, as should easily be recognized by the engine. This scheme of course also works for the "normal" chess position, and you don't need to add something like "f1g1k" or "f1g1c" for "castle". OK, just a suggestion, but I think one that should work OK, although it is not really intuitive for a chess player on the board... Any opinions?! By the way: How is a FRC castle executed on a virtual board on the screen (how do you input the castle move)? I never heard anyone talk about that problem: if f1g1 could be both a normal king move and a castle move, how is it recognized by the GUI? Best regards - Munjong.
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.