Distributed Chess Theory Was: Re: [rc5] v3

Jason Bechtel jbechtel at eng.utoledo.edu
Tue Oct 28 17:35:05 EST 1997

> "Skip Huffman" <SHuffman at Atl.Carreker.Com> wrote:
> ] That does save one bit, but it significantly complicates the design of
> ] the client.  The client now has to keep a separate table for each type
> ] of piece so as to translate its "move code" into a board destination.
> ] The trade off may or may not be worth it.  Also it would make pawn
> ] conversions more complicated because a pawn that has reached the eighth
> ] rank may have become any of several pieces with different movement
> ] rules.

[snip snip]

This may be true, but we're getting away from the original premise.  The 
original idea was to pass all of the moves thus far in the game instead 
of the board position and all the sticky details (un-passant, castling, 
promotion, etc).  It sounded to me like a good idea.  It's compact 
(unless the game gets really long) and it allows the client to deduce
messy details (including draw-by-repetition) by itself!

> First, the client needs to know where the piece can move anyway,
> because it needs to be able to advance the board position by making all
> possible moves for all possible pieces, iteratively.

The client already knows where the piece can move.  It knows the rules
and all the moves so far, so it can construct the board and play from 

>    Another opportunity for optimization is to decrease the number of
> bits used to encode the piece.  It is known how many pieces are on the
> board and where they are, so the piece selection field of the move
> encoding only needs to select between the currently active pieces.

I don't really understand what you're trying to say here.  You want to
number the pieces, and have the list of still-active pieces be passed
along with everything else, so that another computer can renumber them 
with (perhaps) fewer bits?  You've already shot yourself in the foot by
having to pass on a numbered list of pieces.  Maybe I'm just missing 
your point.

My basic point is, trying to cut down the size by methods like this is 
not feasible.  They all depend on some prior knowledge.  If we just pass
along the move list, we have everything.

**Join the largest computer in the world:  http://www.distributed.net
1 in 18,446,744,070,000,000,000 keys works.  I think we can find it.**
To unsubscribe, send email to majordomo at llamas.net with 'unsubscribe rc5' in the body.

More information about the rc5 mailing list