Re: [computer-go] Once and only once

[Don Dailey <drd@xxxxxxx>]

Are you using nega-max?  It's usually much harder to maintain code
like you are suggesting.   negamax makes it easy to handle the
search very uniformly, without any testing for whether it is a min or
a max node.

But I'm  not sure that's what you are talking about.

Sometimes I write a separate function for the ROOT node, because the
code can vary signficantly.

- Don


   Date: Tue, 05 Apr 2005 22:30:13 +0100
   From: Tom Cooper <t.cooper@xxxxxxxxxxxxxxxxx>
   X-Authenticated-Sender: t.cooper@xxxxxxxxxxxxxxxxx
   X-Spam-Processed: FirenetMailserver1, Tue, 05 Apr 2005 22:34:03 +0100
	   (not processed: message from trusted or authenticated source)
   X-MDRemoteIP: 80.229.247.11
   X-Return-Path: t.cooper@xxxxxxxxxxxxxxxxx
   X-MDaemon-Deliver-To: computer-go@xxxxxxxxxxxxxxxxx
   X-MDAV-Processed: FirenetMailserver1, Tue, 05 Apr 2005 22:34:08 +0100
   Reply-To: computer-go <computer-go@xxxxxxxxxxxxxxxxx>
   Sender: computer-go-bounces@xxxxxxxxxxxxxxxxx
   X-Spam-Score: -4.9
   X-Spam-Flag: NO
   X-Scanned-By: MIMEDefang 2.42

   Is there a consensus on list about whether tree search code should be 
   written 'once and only once' abstractly, or should it be duplicated and 
   written concretely, once for a black node and once for a white node?  Or 
   should one make an ad hoc decision, based on the nature of the search?


   _______________________________________________
   computer-go mailing list
   computer-go@xxxxxxxxxxxxxxxxx
   http://www.computer-go.org/mailman/listinfo/computer-go/

_______________________________________________
computer-go mailing list
computer-go@xxxxxxxxxxxxxxxxx
http://www.computer-go.org/mailman/listinfo/computer-go/