Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Chess strength of these programs?

Author: Amir Ban

Date: 14:56:22 03/12/98

Go up one level in this thread


On March 12, 1998 at 01:37:02, Komputer Korner wrote:

>
>Thoralf, 99.9% of everyone  respects your hard work and that of your
>testers but if you don't require all autotest drivers to be made public
>before you test, I am afaid that your list is doomed because one by one
>the programmers will withdraw their public autotest drivers since
>Chessbase refuses to release theirs. You may end up in a situation where
>they send their secret autotest drivers to your organization. This will
>be a sad day for others who will want to autotest as well. Overtime,
>even the secret arrangement won't work because some will not trust that
>it will remain secret and thus your list will lose potential candidates.
>An SSDF list with only 1/2 the participants will lose its credibility. I
>believe that only the following will be work. All autotest drivers and
>all software be commercially available. The only other solution could be
>to accept opening book upgrades from the programmers at any time. I do
>not profess to tell you what to do but I would hate to see all your good
>work come to naught over the issue of secret autotesters.

I saw the consenting opinions of Ed & Johan, so perhaps I'm the only one
who doesn't understand what your point is here, but I don't understand
it anyway. The Ossie Wiener letter made a similar point about the
autoplayer. I don't know if he was implying that there was funny
business going on with the autoplayer, or that supplying a private
autoplayer is cheating by itself (judging by the letter, he didn't know
himself what he was implying, and didn't care).

To your points: Providing an autoplayer to the public is an extra
feature, and maybe some users want to have it, but why should it be
required ? Providing one to the SSDF and to people like Eric makes
sense, because they really need it. Not providing an autoplayer to the
public if you have it is a defensive act against booking up (something I
think everyone except Ossie agree Fritz5 didn't do). Not providing an
autoplayer to the SSDF is not an option, unless you don't want to be
listed. Do you know of anyone who doesn't want to be listed ? If you are
not listed, you don't exist, or so it seems. My conclusion from this
whole business is not as Ossie writes that the SSDF will soon be dead,
or as you write that soon half of the programs won't want to be listed,
but that the SSDF has now become the overriding standard for us (to an
extent I think they never meant to be).

My problem with what you wrote, and with Ossie's dark questions about
the autoplayer, is that I have no idea how you can cheat with an
autoplayer, even if you wanted to. By the way, the only autoplayers used
so far by the SSDF are provided by two of the signers of the Ossie
Weiner letter. One of them, the DOS autoplayer, is available for a price
from Donninger. It's an executable, and what it actually does is hidden
in the source code, which is not available. The second, the Windows
autoplayer, is not available to the public, though the authors
Meyer-Kahlen and Donninger have given it for free to several developers
(myself included) in source code form. It is worth noting that Fritz5,
being a Win16 program, can't use either of these autoplayers, though I
don't know why they would have to even if they could. Did they signers
of the Ossie Weiner letter feel that ChessBase were cheating because
they used an autoplayer that they (the signers) did not develop ?

By the way, I'm not suspicious of the "standard" DOS autoplayer, even
though I don't have source and don't know exactly what it does (I'm sure
Bob Hyatt would like to know). One reason is that the author is above
suspicion, but an even better one is that I don't know what an
autoplayer can do to cheat.

Amir




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.