Computer Chess Club Archives


Search

Terms

Messages

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

Author: Chessfun

Date: 10:11:44 10/31/03

Go up one level in this thread


On October 31, 2003 at 11:20:18, Mogens Larsen wrote:

>On October 31, 2003 at 10:58:01, Francesco Di Tolla wrote:
>
>>This is nonsense: if a program requires to be tested manually due to the lack of
>>any interoperability protocol, then I prefer to see CM in ChessBase gui with a
>>different book rather than no test at all.
>>But since it appears that the CM book can be converted I see no need to discuss
>>further.
>>Just convert it and switch learning off.
>
>It can't be that difficult to comprehend the idea behind SSDF, ie. testing
>commercial products as they appear to the customer. It isn't the strongest
>possible configuration by popular opinion or preference. Both are irrelevant. In
>some cases, however, testing can be difficult or impossible. Sometimes a
>compromise is possible as long as it doesn't distort performance of the product
>in question. How isn't crucial if the setup offers little or no distortion,
>which is the gist of the paragraph you commented on. No nonsense in sight.
>
>If The King can be tested with a converted CM9K book and Nalimov EGTB, I suppose
>the distortion would be minimal. Can CM9K books be converted or was it only true
>for CM8K and older?

CM9K books cannot be converted to other usable formats.

Sarah.



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.