Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Different opening books?

Author: Harald Faber

Date: 15:17:19 06/09/98

Go up one level in this thread


On June 05, 1998 at 11:32:16, Komputer Korner wrote:

>>>>Will it be the future that programs will have (at least) 2 different
>>>>opening books, one for playing other programs and one for playing
>>>>humans? Both is important. Playing well against programs will give
>>>>reputation in SSDF/CCL etc. but may be not as successful playing humans.
>>>>So the need of another book, because I bet the "human-book" won't be OK
>>>>for playing programs... right?
>>>
>>>Just as important is a different style against humans.
>>
>>Is this really necessairy or would it be enough to adjust the book?
>
>Well if you could adjust the book all the way to the end of the game, in
>other words a perfect 200 move opening book you would have solved chess.
>Short of that, style is important along with king safety ....etc. Once
>out of book the Genius 5 style is hard to defeat by humans whereas the
>style of Fritz 5 opens itself up to positional sacs which defeat Fritz
>5. Sometimes you don't even have to sac against Fritz 5, it will beat
>itself trying to sac against you. However against other programs Fritz 5
>will often win by outsearching them.
> Computer and human chess are very different, therefore style counts a
>lot.

Yes, but how is the situation to handle that ie MCP7 has problems
against other programs but is very dangerous for humans? Wouldn't it be
enough to change the opening repertoire a little against other progs? So
that the resulting positions when leaving the book are not double-edged,
meaning that they may work against humans but not against programs.



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.