Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Fritz5 disaster !

Author: Robert Hyatt

Date: 15:39:51 03/03/98

Go up one level in this thread


On March 03, 1998 at 13:14:30, Ed Schröder wrote:

>>Posted by Robert Hyatt on March 03, 1998 at 10:35:49:
>
>>>I.e., the 14 French opening games Rebel vs. Fritz which Howard mentioned
>>>seem to contain at least 10 doubles. Thence, their statistical relevance
>>>shrinks significantly -- namely to that of 4 other games without
>>>doubles!
>
>>>=Ernst=
>
>>Not necessarily.  IE what would you expect if You entered "X" in a
>>tournament and it played me (me, Bob Hyatt, not me, Crafty) and I
>>won.  And the next week we meet again, same colors, and I play the
>>same opening and win again?
>
>>I call that stupid programming.
>
>Yeah, and that same stupid program (Rebel8) has beaten Crafty even
>when Crafty had 100 x more time :)
>

and your point would be?  IE I bet you would *not* beat it twice with
the same opening...  so I don't know where the above came from.  nor do
I know what it means or what it has to do with counting doubles or not
counting doubles.  Remember, you cut the match off after one game.  So
there could not possibly be a double...



>
>>"doubles" are something that a good program avoids if it is
>>losing those games,
>
>Fixed in Rebel9.
>
>Still I believe SSDF should not count doubles.
>
>The goal of SSDF is to test the playing strength of a chess program
>and not smart book algorithms.
>
>- Ed Schroder -


I believe the purpose is different...  To compare the programs to each
other in the same methodology that a human would have to use to play the
program.  IE humans tend to play the same opening and it will be up to a
program to take evasive action if the human finds a way to get to it in
an opening variation.

I've seen my share of gripes about this very problem, when programs
(mine
included until about 1.5 years ago) would play the same losing opening
over
and over...

A program that won't fall for this tactic doesn't have to worry about
whether doubles will be counted or not, except for winning doubles that
it
will find if the opponent doesn't know how to vary...



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.