Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: zobrist hashing method (looking for on line resource)

Author: Gian-Carlo Pascutto

Date: 03:24:16 06/05/01

Go up one level in this thread


On June 05, 2001 at 06:11:09, Cheok Yan Cheng wrote:

>I try to read some code implematation from others. The most confused me is :
>1. Why they need to use a random number instead of fixed defined number?

You could use either. You just try to make sure that by XOR'ing a whole
lot together you don't end up with one of the originals.
That why we like to use random number generators with a large average
and minimum hamming distance, so we can be sure the above has a very
slim chance of happening.
You could construct a fixed sequence too, but that would be more
work than to just rely on the ramdom number generator.

>2. Is there any possible that the collision will occur?

Yes. How probable depends on the size of the keys.

--
GCP



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.