In article <oef### [at] tritonschunteretctu-bsde>,
 Christoph Hormann <chr### [at] gmxde> wrote:

> > *Much* slower? I thought it was slightly slower..... At least it's still 
> > faster than using isosurfaces, right?
> 
> Well, i have not made detailed tests myself (especially none with a lot 
> of components) but it is significantly slower.

From my tests, common results range from 2 to 3 times slower, in some 
cases almost as fast as the original blob, and much faster than the 
isosurface primitive.


> It is much faster than the function based blob in IsoCSG because it does 
> not have to consider all components of the blobs at every point (only 
> those close enough to have an influence)

It actually does not yet do this type of bounding. It first tests the 
ray against the bounds of each component, gathering a list of components 
that affect the ray. Intersection solving is then done with this list, 
which may have only a couple out of several thousand total components. 
This "hit gathering" stage could itself be optimized by using 
heirarchial bounding, but that isn't implemented yet.

-- 
Christopher James Huff <cja### [at] earthlinknet>
http://home.earthlink.net/~cjameshuff/
POV-Ray TAG: chr### [at] tagpovrayorg
http://tag.povray.org/

Post a reply to this message