> Have you tried that out yet?

Nope. Like I tried to explain, the Surcoat macros only sample the surface of
an object evenly, with a small speed-up I came up with to make this process
more efficient. Any standard user can take the object and throw millions of
samples at it and hope to get a good coverage, I've done some discretization
steps and pre-testing to make things more efficient and ensure that there's
a fairly even coverage.

What you do with the data is something entirely different. So far, I've only
written a small macro-set which will take the samples and check their
vicinity to walls and how large they can be in order to model something that
looks like a layer of snow on the object. I'm currently planning a macro-set
which will place hairs on the samples, so that I finally have a toolset to
grow fur/hair onto objects.

For your application, a simple approach might be to cover the bottom of the
wing with samples. Then, select samples within a certain radius and check
the lowest sample, place a "drip" there. Repeat this, probably with varying
size and varying center for the radius-check (a quite obvious step, I hope
;-) and you should get something you could work with. You can of course take
a different approach or create some more overhead to make a more extensive
test for a drip, but the above technique might already be sufficient.
Note though that this can be quite slow and tedious to parse (checking
thousands of samples again and again if they are in a certain radius!? There
are several speed-ups to be implemented here, like saving the centers for
the radius-check within an array and successively updating the lowest point
for that center in a second array, effectively only parsing through all
samples once), so you should implement something to save the results to disk
right along with it.

Once I've got the Surcoat2Snow-Macros cleaned up, you'll have a source to
derive inspiration from for this kind of stuff. :-)

-- 
"Tim Nikias v2.0"
Homepage: <http://www.nolights.de>

Post a reply to this message