David El Tom <dav### [at] t-onlinede> wrote:
> Bruno Cabasson wrote:
> > I was proud of my blob-macro for elliptical torus in response to a recent
> > thread in p.b.newusers. It works fine for solid torii and renders fast, but
> > I noticed artifacts when I fill with glass or media. See the eye on the
> > right and the sparkles on the nose. Where could it come from?
> >
>
> could it be that you there are to much overlapping blob components at the
> start/end section ?
>
> to put it other words: for a closed path you have n components and n section
> (between the components) while for an open path you just have n-1 sections.
> So try to remove the last blob component  or adjust the step-width and see if
> there is still this artifact.
>
> ... dave

Thanks for the comment!

The linear density of the blob (expressed in number of spheres per torus
radius) is auto adjusted by the macro by over-sampling and intergrating the
distance travelled so far and droping a blob component at constant distance.
But, as you said, I also noticed the problem comes from the junction of the
two ends of the blob. I played with the loop control and only one more/less
sphere is sufficient to produce the artifact. I could not see whether I made
a mistake in my loop within the macro, and I could not find any satisfactory
workaround. But the artifact is very sensitive, so my conclusion is that my
macro is OK for solid objects with not too much reflection (the artifact is
not visible, and fast render). If needed, I use the sphere_sweep version
with appropriate bounding box.

     Bruno

Post a reply to this message