Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Creating Opening books ==> don't use CAP data.

Author: Dann Corbit

Date: 01:43:01 07/11/00

Go up one level in this thread


On July 07, 2000 at 13:18:25, blass uri wrote:

>On July 06, 2000 at 23:44:40, Vincent Diepeveen wrote:
>
>>On July 06, 2000 at 21:27:00, Dann Corbit wrote:
>>
>>>On July 06, 2000 at 19:41:31, Vincent Diepeveen wrote:
>>>
>>>>Hello,
>>>>
>>>>this is a post i just did at crafty list after a long number of mails about
>>>>a project called CAP which people tend to believe they can use it for
>>>>creating their openingsbook. Some warnings are on their place, when you
>>>>want to let such a book compete with commercial books.
>>>>
>>>>I first showed that CAP goes completely wrong already quite soon in opening.
>>>>Like 1.d4,d5 2.c4,e5 3.dxe5 there best move is d4 all other moves you can
>>>>go home and rest but first resign the game.
>>>
>>>And (as I showed in the Crafty mailing list) CAP supports this choice.
>>
>>quote from your email:
>>
>>
>>------
>>>
>>> 1.d4,d5 2.c4,e5 3.dxe5  then what does your CAP play?
>>
>>dxc4
>>------
>>
>>You answered dxc4
>>
>>BEEP
>>wrong answer.
>>
>>Now you say suddenly it does play d4.
>>However you are faced with a big problem here.
>>
>>Program's choice.
>>
>>Some programs like d4, most will never play d4.
>>
>>Which one to trust?
>
>I believe that most top programs will play d4 if you give them enough time.
>
>Chessmaster6000(ss=10) found d4 in less than 10 minutes.
>Crafty17.11 is slower but it also found d4 in less than 4 hours(3:42:17) at
>depth 14.
>
>My hardware is pentiumIII450
>I used Crafty17.11 as an engine for Hiarcs7.32 with hash tables 128 Mbytes.

An interesting effect occurs when you analyze the alteratives separately.
Crafty finds that d4 is better than dxc4 in just a few minutes.

This is especially valuable if you have just two or three apparently good
alternatives.  Which one to choose?  Analyze each of the alternatives and see.
This does not always get the proper result.  It works only when the answer is
just over the horizon by one or two plies.

But if you continue that technique forward, on occasion you can find some very
interesting results.



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.