David Doshay wrote:
On 17, Jul 2008, at 1:34 PM, terry mcintyre wrote:
--- On Thu, 7/17/08, David Doshay <[EMAIL PROTECTED]> wrote:
From: David Doshay <[EMAIL PROTECTED]>
My program runs on a cluster ... no way around that.
David, you're just not taking full advantage of Virtualization ...
simply emulate multiple VMs on a single computer; there may be a
performance penalty. ;)
It is actually one step easier than that. MPI lets one do it directly
without all of the multiple VMs. And hey, just like your smiley
indicates, if I emulate 64 machines I "only" have a factor of maybe 65
in speed hit!! Wow, my program could finish a one hour game in only 65
hours!
I do appreciate having some head-to-head competitions using similar
hardware, but I also want to see unlimited open class competition,
where people are able to run multiprocessors, and learn to scale
algorithms to big-n processors. We'll all have 1000 processors on our
desktops in a few years anyhow, might as well iron out the problems now.
getting further off-topic in a somewhat off-topic thread, I have my
doubts about common desktops moving much past 8 or so cores. That is
about the max number of things most users will try to run at one time.
Hmmm. I could use as many cores as you can give me - but the average
won't, just as you say.
The issue I think will come down to software technology - will
applications start popping up that are able to utilize more and more
cores? If so, then more cores will always equal more usable
capability, even for non-savvy users.
Maybe the truth is somewhere between? I already know of people with
core 2 duo computers that remain mostly idle even as they are sitting in
front of them.
- Don
Cheers,
David
_______________________________________________
computer-go mailing list
computer-go@computer-go.org
http://www.computer-go.org/mailman/listinfo/computer-go/
_______________________________________________
computer-go mailing list
computer-go@computer-go.org
http://www.computer-go.org/mailman/listinfo/computer-go/