Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Efficient hash algorithm?

Author: Dave Gomboc

Date: 22:35:45 09/27/98

Go up one level in this thread


On September 28, 1998 at 01:33:01, Dave Gomboc wrote:

>On September 27, 1998 at 23:45:58, Serge Desmarais wrote:
>
>>On September 22, 1998 at 10:11:05, William H Rogers wrote:
>>
>>>Hash tables are used mostly in two instances: 1. for opening books moves as is
>>>saves time calculation your initial responses to game openings; and 2. for end
>>>game calculations. That is, when you reach a certain material value i.e. "small"
>>>then you start hashing the chess board so you can store winning moves leading to
>>>checkmate, or stalemate if no other option is available.
>>>It must be stated though, that many programs keep opening books to a depth of 15
>>>to 20 moves deep. In theory a complete game could be stored this way, thus
>>>eliminating a chess engine completely, and that is not what I call CHESS!
>>>All of this pre-supposes that every player makes the best move possible in every
>>>game.
>>
>>
>>
>>A friend of mine made a joke about it, saying that with a giant opening book and
>> more men endgame tablebases, you could go from the opening libraries right into
>>the tablebases, thus not needing any chess engine anymore. And all move would be
>>almost played instantly, thus CHESS would have been solved as a game, like Tic
>>tac toe! :) Of course it is not serious...
>>
>>Serge Desmarais
>
>Chinook (world man-machine checkers champion) occasionally will do this, but
>it's not likely to happen in chess in the next little while. :-)
>
>Dave Gomboc

I should correct this slightly: even in these rather rare circumstances, there's
still a 17-ply or so search from where the program drops out of the opening book
to where it is hitting the endgame tables.

Dave Gomboc



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.