Computer Chess Club Archives

Search

Messages

Subject: Re: hash collisions

Author: Russell Reagan

Date: 17:31:51 07/09/02

Go up one level in this thread

Very interesting. I have a few questions. What happens when you continue to
increase the "false match" rate? Surely it will start to collapse at some point.
If it doesn't, then something must be broken.

The other question that came to mind was about if the search was able to
withstand many false matches, would it increase performance at all to use a
32-bit hash key instead of a 64-bit hash key. I recall you speaking many times
of your (and others) tests that demonstrated that 32-bit hash keys generated far
more collisions, but now the question is, is that as bad of a thing as we
thought?

Russell

• Re: hash collisions Robert Hyatt 19:55:02 07/09/02
  • Re: hash collisions Bruce Moreland 20:29:18 07/09/02
    • Re: hash collisions Ricardo Gibert 22:02:38 07/09/02
      • Re: hash collisions Sune Fischer 01:30:28 07/10/02
        • Re: hash collisions Ricardo Gibert 06:30:30 07/10/02
          • Re: hash collisions Sune Fischer 08:04:34 07/10/02
            • Re: hash collisions Ricardo Gibert 17:42:45 07/10/02
              • Re: hash collisions Robert Hyatt 19:04:44 07/10/02
                • Re: hash collisions Ricardo Gibert 21:34:56 07/10/02
                  • Re: hash collisions Robert Hyatt 08:40:08 07/11/02
                  • Re: hash collisions [correction] Ricardo Gibert 21:52:28 07/10/02
  • Re: hash collisions Russell Reagan 20:19:12 07/09/02

This page took 0.01 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.