On Fri, Oct 30, 2009 at 10:50:05AM -0600, Brian Sheppard wrote:
> >> Parallelization *cannot* provide super-linear speed-up.
> >
> >I don't see that at all.
>
> This is standard computer science stuff, true of all parallel programs and
> not just Go players. No parallel program can be better than N times a serial
> version.
>
> The result follows from a simulation argument. Suppose that you had a
> parallel process that performed better than N times a serial program.
> Construct a new serial program that simulates the parallel process. There is
> a contradiction.
Of course, if I'm claiming that running M playouts in N threads is
better than running N*M playouts, I must agree that running N separate
groups of M playouts in sequence and letting them vote must yield the
same result - but thank you for this perspective, I haven't thought
about it like this. Clearly, there is a lot of room for experimentation
and measurements here...
--
Petr "Pasky" Baudis
A lot of people have my books on their bookshelves.
That's the problem, they need to read them. -- Don Knuth
_______________________________________________
computer-go mailing list
computer-go@computer-go.org
http://www.computer-go.org/mailman/listinfo/computer-go/
