Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Engines

Author: Thomas Mayer

Date: 03:47:28 05/29/05

Go up one level in this thread


Hi Pedro,

On May 29, 2005 at 02:40:46, Pedro Castro wrote:

> 9252704*1*Start calc, move no: 0
> 9252975>1:time 30000
> 9252975>1:otim 30000
> 9252975>1:white
> 9252975>1:go
> 9252995<1:Book: bookwhite.opn
> 9252995<1:1 9998 0 53 f8e6
> 9253005<1:move f8e6
> 9253005*1*Mov. encontrado:Cf8-e6
> 9253115<1:1-0 {White mates}
>
> This position does not say anything, my engine makes that move and gives
> mate in one, but I do not use for example bitboard for the generator of
> moves.  When I edit a position I do not verify that he is legal, for example
> I do not verify if there is more of a king for each color or if there are
> pawns in row 8, etc.  As I imagine that it will do a pile of people and I
> imagine there that many engines will do just like me.

of course you are right, this single position does not say anything like every
other single position does not say much. The problem is: when there are so MANY
positions that show similarities then one get suspicious. That's the current
problem with Fafis. As long you do not have the source you can never be
absolutely sure. (Most of the cases, Deep <9> was the exception here, because it
simply behaves everywhere similar to Pepito)

Greets, Thomas



This page took 0.11 seconds to execute

Last modified: Thu, 07 Jul 11 08:48:38 -0700

Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.