Author: Bo Persson
Date: 13:10:20 12/16/99
Go up one level in this thread
On December 16, 1999 at 10:59:48, Dan Ellwein wrote: >My original post came about as a result of wanting to know how many unique >movements there are in chess (or at least have a rough idea). > >Chess, being the complex game it is, I wanted to see just how many unique >movements this complexity was based on. > >What came to mind for consideration were the following: > >the movements of the 32 pieces >special moves like castling and en-passant >pawn moves being different from pawn captures >pawns being able to move 2 squares on the first move >and pawn promotion (although this may not apply) > >(there may be other possibilities to consider) Also, the Queen is very powerful, but alone and also very vulnerable the Rooks are not quite as strong, but there is two of them! the Knights can jump over other pieces, their attacks are hard to block. So is their movement the Pawns are weak, but they are many and work good in teams the two Bishops are equal, but also different (square colors). Potential for team work with many other pieces. Long range and lower value is good for support. the King is often threatened, but can move in any direction. It is also very strong in the end game, where it can block its opponent. An extremely successful mix! The complexity is not only in the different "movements", but also in forming groups of pieces where each "character" can use its own special abilities to support the others, working together towards a common goal. Hm, an interesting thought there... :-) >Since pawns do move differently than the way they capture, I decided to use as >one of my parameters for movement - Type. > >I came up with 3 types of movement which I called: Capture/NonCapture, >NonCapture Only, and Capture only. or, possibly, "Piece moves", "Pawn moves" and "Pawn captures"? (forgetting about castling for a moment, and separating the pawns from the major pieces). >Capture/NonCapture would include all the pieces except for the pawn. > >NonCapture Only would include the pawn moves (including the option of the pawn >moving 2 squares on its first move) slightly different from the castling (where the king and the rook make their combined first move), the pawn can make a double move *from its original location*. Ok, so it's only a small difference, but the pawns state can be deduced from its position (2nd row), while you have to keep track of king and rooks, as they could have moved previosuly and returned to there start position. > and castling (since a capture cannot occur on this move). No, but some other *very* special conditions apply, like some squares must be empty, and some squares must not be attacked by the opponent. Not to mention that two pieces that haven't even moved *once* before, suddenly move both at the same time. And they are passing thru (or over?) each other! >Capture Only would only include the pawn. En-passant would be covered in this >category. > >Also, since, pieces can move more than one square on a move (moving from one end >of the board to the other end a piece travels 7 squares), I decide to use as >another parameter of movement - Magnitude. > >Since chess is played on an 8x8 board, Magnitude would have a maximum limit of >7. > >So, Movement, then, (at least for my purposes) is a funtion of Type and >Magnitude. > >Type - having three categories: Capture/NonCapture, NonCapture Only, and Capture >Only. > >Magnitude - ranging from 1 (square) to 7 (squares). > >Looking at the 32 pieces we have the following: > >K's - 2 >Q's - 2 >R's - 4 >B's - 4 >N's - 4 >P's -16 > >So, from the 32 pieces, 6 pieces need to be considered - K,Q,R,B,N,P > >From these 6 pieces, 4 need to be looked at due to the Rook and Bishop movements >are incorporated into the Queen. > >So now we have - K, Q(which includes the Rook and Bishop), N, P Yes, some programs (like Crafty) keeps track of the combined piece types Rooks_and_Queens and Bishops_and_Queens to use in places like attack detection. >And this is how I came up with my table and came up with there being 11 unique >(discrete) movements in chess. > >If, instead of Magnitude having a range of 1 to 7, but we give it a range of 0 >to 7, then pawn promotion could be considered also. Replacing a pawn with >either a Q,R,B,N does not involve a movement (the Q,R,B,or N is placed on the >same square as the pawn was on - the actual promotion requires no movement). > >In this case there would be 12 unique movements in chess, and these 12 movements >would be what gives chess its complexity. > >PilgrimDan Still fascinated by the thoughts, but don't (yet?) know how to use it. Bo Persson bop@malmo.mail.telia.com
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.