Re: [computer-go] Re: Scalability

[Chris Fant <chrisfant@xxxxxxxxx>]

Has anyone else tried averaging the evaluations for the quiesence
plies (perhaps weighted)?  I tried this without much success (it was
my only attempt at a q-search).  It seems like an obvious thing to do,
but I don't think I've ever heard the reasons why it is not a good
thing to do.
I used no branching.  One move was generated and played for something
like 2-3 plies.  And the evaluations of those plies were averaged
together.  I tried a few weighting schemes.
_______________________________________________
computer-go mailing list
computer-go@xxxxxxxxxxxxxxxxx
http://www.computer-go.org/mailman/listinfo/computer-go/