Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: About random numbers and hashing

Author: Ralf Elvsén

Date: 08:17:24 12/06/01

Go up one level in this thread


On December 05, 2001 at 15:56:32, Sven Reichard wrote:


>
>Summarizing I can say that I see no connection between the quality of hash codes
>and their Hamming distance. Using a good RNG like the one provided in GNU's
>stdlib will yield good hash codes ( you can actually prove that), and so I will
>take the codes as they are supplied by rand() or random() without messing with
>them and thereby most likely make them worse.
>
Your argument seems convincing, good thinking. All I can say is that
my hashtable worked clearly better when I switched from random numbers
"out of the box" from a random()-function, to a set I generated myself
using the Hamming-distance as a measure of quality. It doesn't
prove anything and was a long time ago. Maybe I should test again...

Ralf

>Sorry for the length of this message and my rambling on. Thanks for your
>patience and any feedback. Actually I bet a case of beer that nobody reads up to
>this point (just kidding).
>
>Later,
>Sven.



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.