Computer Chess Club Archives


Search

Terms

Messages

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

Author: Mike Byrne

Date: 14:26:24 10/31/03

Go up one level in this thread


On October 30, 2003 at 23:06:08, Chessfun wrote:

>On October 30, 2003 at 22:43:24, Mike Byrne wrote:
>
>>On October 30, 2003 at 22:27:54, Chessfun wrote:
>>
>>>On October 30, 2003 at 22:21:39, Christophe Theron wrote:
>>>
>>>>On October 30, 2003 at 17:58:29, Uri Blass wrote:
>>>>
>>>>>On October 30, 2003 at 17:10:25, Christophe Theron wrote:
>>>>>
>>>>>>On October 30, 2003 at 14:32:08, Gian-Carlo Pascutto wrote:
>>>>>>
>>>>>>>Hi,
>>>>>>>
>>>>>>>I don't understand why ChessMaster is allowed to use a Kure book (general.ctg)
>>>>>>>in the SSDF testing.
>>>>>>>
>>>>>>>Last time this happened (with Gromit) it was specifically said by Alex Kure
>>>>>>>that it was _not_ allowed to use the Fritz books to test anything other
>>>>>>>than Fritz. So why is ChessMaster using it now? Did Kure give permission
>>>>>>>(which I seriously doubt)?
>>>>>>>
>>>>>>>IMHO using the ChessBase learner for ChessMaster is also very questionable.
>>>>>>>
>>>>>>>--
>>>>>>>GCP
>>>>>>
>>>>>>
>>>>>>
>>>>>>I agree to both of your points.
>>>>>>
>>>>>>If Kure does not allow one of its books to be used, it is unfair to let CM9000
>>>>>>use it.
>>>>>>
>>>>>>CM9000 is not a ChessBase product, so it is unfair to allow the engine to use
>>>>>>the ChessBase book learning system.
>>>>>>
>>>>>>Thank you for bringing it to our attention. I'm extremely surprised that the
>>>>>>SSDF guys do something that is clearly against the spirit of their list.
>>>>>>
>>>>>>
>>>>>>
>>>>>>    Christophe
>>>>>
>>>>>CM9000 is not a chessbase product but I think that the ssdf should be allowed to
>>>>>test every combination of engine, book and learning.
>>>>>
>>>>>After all the interesting question is what is the best thing that people can buy
>>>>>and if people can buy chessmaster together with Fritz book and get something
>>>>>better than it is important to know it.
>>>>>
>>>>>SSDF have not enough time to test every combination and this is the reason that
>>>>>they should prefer the programmer choice in case that he suggests something(I
>>>>>guess that in this case Johan suggested nothing so they are free to test it in
>>>>>the way that they want to do it)
>>>>>
>>>>>I think that book makers should not be allowed to forbid people to test their
>>>>>book with another program.
>>>>>
>>>>>They earn enough from not allowing other people to use their book in world
>>>>>championship or from the fact that people know that the ssdf leader used their
>>>>>book.
>>>>>
>>>>>Chessbase also can earn money if people know that the programmer chose to test
>>>>>chessmaster under chessbase because it means that people may buy both chessbase
>>>>>interface and chessmaster.
>>>>>
>>>>>Uri
>>>>
>>>>
>>>>
>>>>Come on, Uri...
>>>>
>>>>Maybe you hope that some day the same favor will be allowed for Movei?
>>>>
>>>>It's ridiculous. The spirit of the SSDF has always been to test the programs
>>>>"out of the box", unless the programmer himself suggests a change in the default
>>>>setup.
>>>>
>>>>Here it's not an "out of the box" program that is tested, because it cannot be
>>>>tested automatically.
>>>>
>>>>OK, I can understand that.
>>>>
>>>>But WHY would this program receive TWO UNFAIR ADVANTAGES?
>>>
>>>
>>>You seem to also have missed that this isn't the default CM9K but a modified
>>>settings version SKR.
>>>
>>>Also there are others programs which have been tested previously by the SSDF
>>>using a general.ctg from CB. Those programs tested also were not part of CB's
>>>stable.
>>>
>>>Sarah.
>>>
>>
>>I respectfully disagree.  The rules were always "out of the box, best settings
>>by the author".
>
>Disagree with what?. I simply stated two facts.
>My opinion is that the default CM9K should be used, I just never stated it.
>
>Sarah.

Sorry Sarah - I misread the post.


>
>
>>
>>That should be CM9K STANDARD settings with the CM9K book.  If the book is not
>>converted, I would go with the the CM8K book which is convertible.
>>
>>This whole concept of testing with the settings not supplied by the original
>>author is a bad precedent and should be abandoned immediately.  In fact, if I
>>was a chess programmer I would dis-assocaiate my with SSDF program.  If some
>>other orgaziation undera a different name wanted to test under these conditions
>>- that would be fine - but the SSDF ratings themselves should not be
>>contaiminated with less than optimal settings as defined by the original AUTHOR.
>>
>>Period.
>>
>>
>>
>>>
>>>>Why would the SSDF give an advantage when the author and the manufacturer of the
>>>>program in question have provided no help to solve the issues of the book and
>>>>the GUI???
>>>>
>>>>If this is allowed now by the SSDF, I guess chess programmers do not have to
>>>>write an opening book. Just write an engine and send it to the SSDF, they will
>>>>add a GUI, book learning and a very strong opening book.
>>>>
>>>>In two years from now the Kure book will be standard and all the SSDF games will
>>>>end in a draw because all games will be Kure book against Kure book.
>>>>
>>>>Instead of giving CM9000 TWO unfair advantages it should:
>>>>* either not be tested at all
>>>>* or be tested without opening (and without book learning of course)
>>>>
>>>>What's the next step? "Dear SSDF, I have written a chess program that cannot
>>>>play the endgame, so please switch to Shredder when you reach the endgame stage.
>>>>Thank you."
>>>>
>>>>
>>>>
>>>>    Christophe



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.