Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Pathway count to eight ply

Author: Steven J. Edwards

Date: 01:59:00 08/07/98

Go up one level in this thread


On August 07, 1998 at 01:09:56, Danniel Corbit wrote:

>All this masterful pounding of plies...  It is an interesting exercise in
>itself.  But I wonder how many of those legal permutations are actually useful.
>Did you find any interesting positions in the mix, or are you just throwing them
>away without bothering to look at them?

At eight microseconds per pathway, there's not much time for analysis.

I have found an earlier reference to the enumeration of eight ply long
trajectories from the initial array.  In the 1949 book _The Fireside Book of
Chess_ edited by Chernev and Reinfeld, there is on page 106 an estimate by Edwin
Anthony.  A figure of 318,979,584,000 is given, a value which is too high by a
factor of about 3.75.  Well, fifty years ago I guess that logarithms were high
tech.  Another figure is given for the twenty ply enumeration; it's
169,518,829,100,544,000,000,000,000,000.  Yes; that's fifteen significant
digits.  It may not be that far off; I'd guess that the real answer is somewhere
is the neighborhood of 10 to the 30th power.

I may do some more work with this using a variant of transposition tables to
store position subtree enumerations.  This could make an attempt at nine ply
enumeration reasonable.  Maybe even ten ply with enough memory.

-- Steven (sje@mv.mv.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.