Computer Chess Club Archives


Search

Terms

Messages

Subject: HT, sanity check

Author: Georg v. Zimmermann

Date: 15:47:47 11/02/03


Hi,

is there some flaw in the following reasoning or does it sound correct:

Hash table

A) Conditions

1) perfect random numbers
2) 800 000 slots
3)   5 000 positions I want to store, all equally important
4) a simple scheme, without any linking between the entries (we dont try another
slot when 1 already full).

Is it correct that we get about 32 rejected positions ?

B)
all equal but linking between even and uneven numbered slots (which means we try
exactly 1 slot if another already full).

How many rejected positions now ? Whats the formular for X linked entries ?


Looking forward to a lot of FAST responses, preferably as fast as if I started
another SSDF threat. Cause it is 11:46 pm here and I have too much to work
tomorrow.

Georg




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.