Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: A final stab at big-O

Author: Ricardo Gibert

Date: 14:32:27 05/09/01

Go up one level in this thread


On May 09, 2001 at 16:15:44, Dann Corbit wrote:

>On May 09, 2001 at 16:10:09, Andrew Dados wrote:
>[snip]
>>Guaranteed fixed number of steps means O(1) afik.
>
>The number of steps to solve is a property of the algorithm.  All algorithms
>terminate after a fixed number of steps.  That't part of the definition of the
>term 'algorithm.'


By fixed, he means constant. All algorithms terminate in a finite number of
steps, but do not terminiate in constant number of steps.

The difference is that an O(1) algorithm is guaranteed to finish in a finite
number of steps fewer than some constant c.

If it is more than O(1), the number of steps continues to be finite, but no
constant c exists that delimits the number of steps executed.


>
>If any algorithm does not terminate in a fixed number of steps, then *by
>definition* is ISN'T and algorithm.
>
>Therefore, we KNOW that the algorithm will terminate (perhaps in arbitrary
>time).  If we want to estimate how long it will take, we need to know the big-O
>behavior of the algorithm.  To call chess O(1) is simply not useful.



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.