[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: computer-go: minimax and go



Rong Zeng <rodneyzeng@xxxxxxxxxxxxxxxxx> wrote:

> And I think this static model is not a technique problem if those
> excellent game companies would like to produce an engine for it. It's the
> problem of inadequate programming power now. We'll see it earlier if they
> would do it.


I do not see what a video game engine has to do with go.
Those engines are optimized for quite different kind of games, where they
can take various shortcuts not available in go. They were also able to
design the game rules so that they would fit with their engines...

You are right, lots programming effort is needed to make better go
programs. Yet the Gnu Go project is doing quite well with a relatively small
team of volunteers. And many of the stronger programs available today are
the work of one man, or at most a small team. Just as in playing go, adding
brute force with no good plan will not help much, adding more programmers
into a project will only slow it down, unless done with great care. Not
necessarily with detailed planning, as the Open Source model of GnuGo shows,
but with a good coordination anyway.


-H

-- 
Heikki Levanto     LSD Levanto Software Development   heikki@xxxxxxxxxxxxxxxxx
               "In Murphy we Turst"