Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Major Breakthrough on the 20 Byte Problem!!!

Author: David Eppstein

Date: 11:03:01 08/19/99

Go up one level in this thread


On August 18, 1999 at 20:25:06, KarinsDad wrote:
>Doesn't 2 * 2^162 + 16 * 2^161 > 2^165?
>>So even ignoring the cases you can't handle, you still have 166 bits required.
>>Why do you say the maximum required is 162?
>
>The cases are not part of the same position, but rather templates of given
>position types. For example, you could split positions into 17 cases: ones with
>16 pawns, one with 15 pawns, ..., ones with zero pawns.

Perhaps you misunderstood my question.  Any count of the number of bits required
to store a position has to include the number of bits used to tell which case
you are in.  Otherwise I could just divide into 2^180 cases and say that I store
the position in 0 bits.  It seems that when you are saying you require at most
162 bits (except for the cases you can't handle) that you are not counting this
part of the representation.



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.