Subject: "fastfib" in the gf package faster than "fib"
From: Zach
Date: Wed, 9 Jul 2008 09:30:56 -0600
> By the way, if you're doing *floating-point* calculations, probably
> the most cost-effective platform (FLOPS per dollar) right now is
> graphics accelerator cards (!), and there are people working on real
> physics using them. Not something I see Maxima using any time soon,
> though.
I have noticed some of these papers as well. In particular, one (Harvesting
Graphics Power for MD Simulations in Molecular Simulation) by Daan Frenkel's
group reporting that running molecular dynamics simulations on several `SLI'
linked video cards gains them a >100x speedup over the CPU (in some cases).
As MD simulation is what I do for money these days, I find this
interesting. I guess that is what 16 cores times 4 graphics cards all with
shared, fast memory, gets you. Made me want to get a fancy new video card
and the Cg language so I could tinker with it.
But, floating point calculations is not what Maxima is for, so of course
this is completely off topic. Just very cool to me.
Zach