Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: What is a Knowledge program ?

Author: Charles L. Williams

Date: 12:00:26 05/06/99

Go up one level in this thread


On May 06, 1999 at 00:05:09, James Robertson wrote:

>On May 05, 1999 at 23:39:11, Charles L. Williams wrote:
>
>>On May 05, 1999 at 18:15:28, blass uri wrote:
>>
>>>
>>>On May 05, 1999 at 13:26:43, jose hernandez wrote:
>>>
>>>>
>>>>
>>>>For example Are a knowledge program like Rebel or Hiarcs better in blitz games
>>>>than a Fastest programs like Fritz or Junior ?
>>>>
>>>>or a Knowledge program is better in long controls than a fastest programs ?
>>>>
>>>>What do you think?
>>>
>>>suppose Ed finds an idea that makes Rebel
>>>5 times faster with the same evaluation function.
>>>
>>>Do you continue to call Rebel a knowledge program?
>>>
>>>If you comtinue to call Rebel a knowledge program then your definition is not
>>>based on the fact that Rebel is searching less nodes per second and I want to
>>>know your definition for knowledge programs.
>>>
>>>Uri
>>
>>
>>I'd think that a true knowledge-based program wouldn't have a search.
>>Otherwise, it's a search-based program and the debate becomes how much chess
>>knowledge is in the evaluation.
>>
>>
>>Chuck
>
>That doesn't quite work either, because the search itself is intrinsically
>knowledge. :)
>
>James


I see your point, so let me try another one.  Maybe programmers claim their
programs are "knowledge-based" when they can't get as many nps as the next guy.
:)

Chuck



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.