Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: To Recurse or not to Recurse...

Author: Bruce Moreland

Date: 09:28:29 10/31/00

Go up one level in this thread


There is one way to find out, and that is to try it and see.

A problem is that as your search routine gets more complex, it is harder to
manage everything that is going on.  Mine was non-recursive back in 1994 or so,
but I eventually tired of the added complexity.

I think the function call overhead is minimal.  What I wanted to do was snag an
extra register (BP).  It worked, kind of.

bruce

On October 31, 2000 at 08:30:04, Steve Maughan wrote:

>In my chess program I use recursion to carry out the basic alpha beta search.
>However while on one of the Delphi newsgroup I was told that recursion is a
>source of significant inefficiency and that _ALL_ recursive algorithms can be
>converted into an iterative form that is generally faster.  Upon searching the
>net I found this site:
>
>http://www.geocities.com/zabrodskyvlada/aat/a_recu.html
>
>This basically shows that the recursive version of a sort runs at half the speed
>of the alternative version.  Moreover, thinking back I remember that Ed Schroder
>mentioned that he doesn't use recursion.
>
>How many people do use recusion?
>
>Is there any literature on non recursive alpha / beta?
>
>Has anyone done any comparisons between recursive and none recursive searches?
>
>Regards,
>
>Steve



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.