Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Usage of general.ctg book+CB learner by ChessMaster in SSDF testing

Author: Mike S.

Date: 17:09:17 10/31/03

Go up one level in this thread


On October 31, 2003 at 19:42:20, Geoff Lane wrote:

>(...)
>>CM9K books cannot be converted to other usable formats.
>
>http://www.chess-archive.com/ccc.php?art_id=313052

You didn't try that, or didn't take a look at the conversion result.

This only *seems* to work, but the moves and probabilities have been checked and
compared to the original version. The conversion result is totally broken and
useless!!

IIRC it also has been said that Fritz conversions of older CM books like
CM8000.OBK were faulty, but I'm not 100% sure about that.

As a replacement in other GUIs, I think a good large database of strong IM and
GM games could be used to generate a solid and simple opening tree. This is very
common with many freeware engines too. IMO the ply depth shouldn't be too big
for The King, which has good opening capabilities of it's own. Maybe even
CM9000's games database could be used for that, although it's in Tasc format.
Maybe it could be converted to be able to create a CTG tree from it, I don't
kono.

(Btw. to make it clear, *I* have never recommended to use general.ctg, but the
above idea.)

Maybe this would have been a better decision, to avoid any "product mixing
protests", although even some basic computerchess specific editing (not engine
specific, but too risky gambits and the like) would be missing in such a simple
tree. But I think King could handle that quite well.

Regards,
M.Scheidl



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.