Author: Roberto Waldteufel
Date: 18:50:20 11/05/98
Go up one level in this thread
On November 05, 1998 at 18:36:15, Will Singleton wrote: > >On November 05, 1998 at 18:16:38, Dan Homan wrote: > >> >> if(p->side_to_move == 'white') return score; >> else return -score; >> > >Dan, > >It's funny, so much of this is semantics. I originally asked: > >"Now for the odd question: it seems I still have to negate the value from eval() >on odd plys to get the thing to work!" > >And now I find that you do the same thing. By "odd plys," I meant when >search-length is odd. Glad we got that straightened out! > >Will I think that it would be easier to recode your evaluation function so as not to give a wtm evaluation all the time, but to give the evaluation always from the point of view of the side to move. That way you just evaluate when you want, and you never have to worry about odd and even plies. You can simply add an extra paramater to the evaluation function: value=evaluate(position,side) Then all your plies (even and odd) can use exactly the same code, removing many IF/THEN/ELSE clauses, which should help the efficiency a bit. Best wishes, Roberto
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.