[rescue] Advice on Octanes
Francisco Javier Mesa-Martinez
lefa at ucsc.edu
Tue Sep 23 15:33:23 CDT 2003
On Tue, 23 Sep 2003, Joshua D. Boyd wrote:
> On Tue, Sep 23, 2003 at 01:17:19PM -0700, Francisco Javier Mesa-Martinez wrote:
> > > What radiosity solver were you using? I wonder why the comparitive
> > > difference between radiosity and raytracing.
> > >
> > > I can't easily think of any reason why it would behave like that.
> > Raycasting is very cache friendly, Radiosity since it involves large
> > amounts of recursion is not.
> I thought of that, but then, shouldn't the Octanes superior memory bus
> make it faster than a P3-800 which would at best have a PC133 memory
The memory system of the Octane is on par if not slower than a P3-800, the
P3 destroys the R10K in raw computation power, hence the Octane can have
all the memory bw it wants, but since it is running a computationally and
not data bound process it just can not compete with the P3 in those tasks.
More information about the rescue