Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Hashkey collisions (typical numbers)

Author: Dieter Buerssner

Date: 15:10:09 04/07/04

Go up one level in this thread


On April 07, 2004 at 11:59:58, Sune Fischer wrote:

>>>It is quite similar from the point on where the hash table is full.
>>
>>absolutely not!
>>the probability that *any* two 64-bit signatures are the same goes from 0...1 as
>>you increase table size and is about 0.5 at 2^32 entries (and goes to 1 very
>>rapidly from there).
>
>I suspect you may in the heat of the moment have thought
>that 2^32 is half of 2^64?
>
>It is not of course :)

I think, Martin is correct. This does not mean, that you will have many
collisions. It means, that each 2^33 probes, you will have one collision.

Regards,
Dieter




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.