> Don has stated a couple of times that option (A) worked better for
> him. I chose option (B) without testing option (A) because I did not
> want to have to decide how many seconds to use to guess the opponent
> move before starting to think about my next move.
There is no need to spend any extra time if you choose option A. You
can just take the guess as the 2nd move from the principal variation
when you generate your original move.
Yes, I understand that deciding on zero seconds is an option. But
consider the following situation: Your engine makes a move very
quickly (perhaps it is a statically-recognized, large-group-saving
move or maybe your time management code demanded a fast move). Now
suppose your opponent thinks about it's move for a much longer time.
If your engine uses option A (with zero additional seconds), then you
are now pondering for a long time on a subtree that is probably
meaningless because your hastily guessed opponent response was not
based on sufficient exploration of that subtree. If you had chosen
option B, then you are using that large chunk of time to fairly ponder
on the entire tree. It just seems right. But on the other hand, it's
tough to argue against Don's empirical results. Like he said, it
doesn't hurt to try both in your engine -- it just takes time, like
everything else.
_______________________________________________
computer-go mailing list
computer-go@computer-go.org
http://www.computer-go.org/mailman/listinfo/computer-go/
