Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Diminishing returns

Author: Robert Hyatt

Date: 20:38:53 04/29/04

Go up one level in this thread


On April 29, 2004 at 21:03:55, Dan Andersson wrote:

>>More correct would be: D+1 is always better than or equal to D.
> It's a safe assumption. But it is not strictly true.
> One example is when evaluation gives best move at depth D and at D+1 a tactic
>is available that at later depths doesn't pan out.
> Another is the fact that there exist so called pathological trees that have the
>property that the quality of the move is always less or equal at each deeper
>ply.
>
>MvH Dan Andersson

In demand paging, "Belady's Anomaly" shows that increasing the resident set size
increases the number of page faults in the LRU page replacement strategy.  Guess
what strategy _all_ current O/S paging systems use?  :)

Exceptions are just that...  exceptions.

Give me an extra ply _every_ time.  If it rarely bites me, I'll take the bite.
Most of the time it will bite my opponent instead...




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.