Ben Chambers <ben### [at] pacificwebguycom> wrote:
> PaulSh wrote:
> > POV-Ray is finally making its transition to SMP systems, which is great news
> > for those of us who can afford them. However, in terms of raw CPU power the
> > latest NVIDIA graphics cards would seem to completely blow away anything in
> > the way of SMP solutions this side of a research lab. Their CUDA GPU
> > architecture is claimed to allow up to 128 independent processing units
> > each running at 1.35GHz to be thrown at computationally-intensive problems.
> > So, my first thought was not SETI or protein folding, but POV-Ray. Given
> > that V3.7 is going to be fully threaded, what would be the possibility of a
> > CUDA version? I guess that will depend on the time and abilities of someone
> > with a lot more time and a lot more ability than myself...
> >
> >
>
> No good, they're only single precision.  Plus, each shader unit would
> need access to the entire scene file, which would be a pain in the a**
> to code.
>
> ...Chambers

The counter-rotating quad-toroidal NUMA memory bus of the 7 or 8 SPEs in
IBM's Cell Processor
(Mercury Blades, Playstation 3) would support this, though. They're dual
precision vector coprocessors and can access the global memory that the dual
core PowerPC can through a shared memory interface (as well as their own
individual local RAM).

I'm up for tackling it once the 3.7 source becomes available. I'm stalling
until then because I don't want to reinvent the full-threading wheel that I
know they've nearly finished.

I'll hold my breath for the early CVS/SVN source access ;-)

Post a reply to this message