Author: Brian Richardson
Date: 18:14:54 05/07/01
Go up one level in this thread
On May 07, 2001 at 16:30:26, Robert Hyatt wrote:
>On May 07, 2001 at 16:15:08, Brian Richardson wrote:
>
>>In general...
>>
>>I don't know about adding pawns to the mix (per the Crafty/Diep games), but in
>>evaluate.c, Crafty seems toI prefer the single queen to 2 rooks, or am I missing
>>something...
>>
>> else if (WhiteMajors == BlackMajors) {
>> if (WhiteQueens && !BlackQueens) score+=BAD_TRADE>>1;
>> else if (!WhiteQueens && BlackQueens) score-=BAD_TRADE>>1;
>> }
>
>That is where I correct the score to avoid tweaking with piece values and
>seeing nonsense moves that do bad things. This says that a queen is only
>a little worse than two rooks. Which means that if I have a queen and a pawn,
>vs two rooks (and ignoring everything else on the board) then the queen and pawn
>are a little better as the two rooks will never win.
>
>that is nothing compared to the bonus Vincent tosses in for Q vs 2R...
OK, but it looks like Crafty prefers the white queen to the 2 black rooks with
score += BAD_TRADE (and vice versa)...or do I have it backwards?
Thanks
Brian
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.