Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: zobrist hashing

Author: vitor

Date: 21:36:46 05/29/99

Go up one level in this thread


On May 29, 1999 at 23:33:06, Robert Hyatt wrote:

>On May 29, 1999 at 22:29:15, Ricardo Gibert wrote:
>
>>Section 4.6 in Algoritms + Data Structures = Programs by Wirth is correct.
>>>
>>The 1983 edition of Algoritms by Sedgewick gives the typical incorrect
>>exposition of the topic, which you seem to echo rather closely.
>>>
>>The 1984 Edition of Fundamentals of Computer Algorithms by Horowitz & Sahni I
>>have is also incorrect.
>>>
>>I'll check Knuth when I pass by the bookstore today.  I'm pretty sure he gets it
>>right.  He is very meticulous & comprehensive of course.  Not very good at
>>explanations though.  Not a good writer.  He seems to set the standard,
>>nevertheless.  I'll probably wind up posting a quote from Wirth even though it
>>looks like it will be overly long. They don't let you use a copier in bookstores
>>unfortunately.
>>>
>>Just think about what you are saying.  "M mod N" would then be hashing a
>>function according to you.
>
>
>M mod N is most definitely a hashing function.  It is a poor one because only
>the low order bits play a part in the result.  But hashing it is.  As is the
>Zobrist approach... as is the approach of summing the characters in a name, and
>so forth...


so how does crafty/crayblitz convert the 64bit key to an index value? obviously
it doesnt have a 2^64 hasbtable.



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.