ingo <ing### [at] tagpovrayorg> wrote:

> Have you considered blob-sweeps? You could use the .z in the vector for
> the strength of the blob.


This looks like a solution. If the b_spline function could be converted to SDL
in such a manner that it would generate points and corresponding diameters at
specific intervals (the distance between points would have to remain constant
regardless of the curvature) then a blob element could be placed at each point
and the diameter could be converted to a blob strength.

This works perfectly when the two character strokes that are intersecting each
other differ greatly in diameter but when their diameters are similar then a
little bump results so some tweaking would be necessary.

Thanks for the suggestion Ingo! Even if the b_spline math turns out to be beyond
my capabilities, I think you have provided a solution. More testing would be
necessary to make sure that tight turns don't cause blob elements to connect
that shouldn't connect.

Here's the test code that generated the sample image:

#local BlobStrength = array [2] {.8, .4}
#local DeltaTheta = .1;

blob {
   threshold 20
   #for (I, 0, 1)
      #for (Theta, 0, 360 - DeltaTheta, DeltaTheta)
         sphere {<cos (radians (Theta)) * 6 + 4 * (I * 2 - 1), sin (radians
(Theta)) * 6, 0>, BlobStrength [I], 1}
      #end //#for
   #end //#for
} //blob

Post a reply to this message

Attachments:
Download 'blob.jpg' (83 KB)

Preview of image 'blob.jpg'