Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: standardisation abilities low

Author: Mike S.

Date: 17:12:45 09/27/00

Go up one level in this thread


On September 27, 2000 at 17:53:59, James T. Walker wrote:

>(...)
>It's a very common problem with chess programs.  Some count the first move in
>their mate announcements and some apparently feel they are making the move first
>and then announcing mate in "x".  I think either is OK but I wish there was a
>standard that evrybody followed.

Indeed, but I think this is one of the problems which is quite hopeless.
Standardisation abilities are low in computer chess. The PGN standard and
WinBoard are wonderful exceptions.

For example, I think some programs needed several generations to have a complete
PGN support at last. Starting from none at all, to rudimentary support with only
1 game per file for example, or without support of FEN starting positions, etc.
After these and other experiences, I don't expect too much from the computer
chess scene in this respect.

The mate announcing problem has been somewhat "blurred" by programs announcing
(long) mates which turn out to be possible with less moves later on anyway, and
by announcing the very first mate they find and moving immediatly. Users would
like to have the programs just showing the eval #xx and the pv, and to continue
the search for a shorter mate (as long as it has enough time on the clock). Some
programs (very few) do that; I don't know which one's at the moment.

Maybe this requires difficult modifications of the normal A/B search algorithms
or whatever in this situation (I'm no programmer); I seem to remember that this
has been discussed already somewhere. But I think the word "standard" wasn't
mentioned.

Regards,
M.Scheidl



This page took 0.01 seconds to execute

Last modified: Thu, 07 Jul 11 08:48:38 -0700

Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.