Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Chessmaster version test for Mr. John Merlino

Author: John Merlino

Date: 15:19:20 07/12/02

Go up one level in this thread


On July 12, 2002 at 17:55:58, Coxwell Strange wrote:

>Thanks for clearing that up for me.
>    Let's see I have this straight now...
>1. selective search does little but reduce time to depth, so we dont need that
>and can use brute search.
>2. hash does little more than reduce time to depth, so we can use the least
>possible.
>3. CPU does little but reduce time to depth, so.. we only need a 200mz,not a 2gz
>4. All of those "others" must be the guys all over the internet with their
>Chessmaster8000s, on PC200s, doing brute searches, and licking the crap out of
>tiger, fritz and all the others... :)... thx again for the thorough education of
>strength and computer chess.

Would you like me to whip up a program that can see to depth 20 at just about
any time control, even with slow hardware, yet still lose every game?

FIRST you need the knowledge, THEN you work on speed. What's the point of
getting to depth N if you are not able to properly evaluate a position at that
depth?

The best programs are not the best because they are the fastest. The King is, as
you have noticed, relatively slower than most of the other commercials. Why is
it as good as the other commercials then?

The one MAIN point you have failed to take into account is that when The King
reports a certain depth, it is reporting a MINIMUM search depth (unlike some
other programs that report a MAXIMUM search depth), and many of the lines that
it has searched have been searched WELL PAST that depth. Take a look at this
post:

http://www.talkchess.com/forums/1/message.html?240127

How do you think it is possible for The King to announce Mate in 11 (22 plies
away) at depth 9?

And, vice versa, there are probably several amateur programs available that are
faster searchers than any of the commercials. Yet, why are they not as strong as
the commercials?

You continue to fail to see the flaw in your argument....

jm



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.