Re: [gnugo-devel] Reward multi-purpose moves ?

[Arend Bayer]

On Fri, 8 Nov 2002, Portela Fernand wrote:

> I thought about it when I was looking at nngs3:320. I made a small test
> with the appended patch (which is meant as reference for discussion, not
> for inclusion in CVS) without hoping for much and I got following
> regression breakage:

[patch multiplies move value by 1.003^(number of move reasons) if this
number is at least 3]

> 7 PASSes
> 2 FAILs
> Any comments welcome.

Interesting. It could just mean that connect/disconnect move reasons are
undervalued. Since Gunnar changed these to refer to worms not dragons,
a single move can really have a big bunch of these.

Arend