[Computer-go] Zen vs Tromp: now 1-1

Christoph Birk birk at obs.carnegiescience.edu
Fri Aug 27 12:06:31 PDT 2010


On Fri, 27 Aug 2010, Don Dailey wrote:
> Like I say some experimentation is to be expected, even in the real match.
> The issue that is of interest to me is whether the loss should be considered
> relevant or not.   Did  John play  stupid moves on purpose "just to see what
> the program would do?"

Why would he play a stupid move?

> Or did he play hard?

Playing "hard" and "expriment" are not mutually exclusive.
In a test game I would not necessarily see winning as the prime
objective, but rather gather experience how a programm reacts
to various styles.
Eg. opening style, life & death evaluation. In paricular
trying to create a big ko seems worth an experiment.

Christoph




More information about the Computer-go mailing list