[Fluxus] pdata-size extending?

Kassen signal.automatique at gmail.com
Tue Aug 3 14:07:14 PDT 2010


Dear list,

Suppose I have a nice icosphere that's green and inside of this I have a
slightly smaller icosphere that's blue. Now I could deform the green one,
multiplying some of the position vectors by some number between 0 and 1,
according to the location. This would give me lakes and oceans in a nice and
simple way.

Sadly, once I do this, in practice, the lakes look like game pieces because
a resolution for icosphere that looks nicely round isn't high enough to
create organic shapes for these lakes; you can clearly make out the
triangles.

One solution might be to iterate over the triangles, figure out which ones
have at least one corner inside of the lake's radius and at least one
outside of it, split those up into -say- three new polygons (recursing if
needed) and only then deform the sphere. That should lead to only having a
higher polygon density where we need it, so it sounds like a good idea to
me. Now I have a problem; we can do all sort of things to the pdata array
and we extend it adding new types, but it seems that we can't extend it's
length while keeping the primitive in question as a object (at least that's
not covered in the manual). Is that right? If so would it be nice to be able
to append polygons to existing pdata arrays?

A alternative way might be to copy all data to a new object, except with a
slightly higher size, then concat that with the transformation matrix of the
original object, perhaps have some parent object to be able to maintain
animation functions that might affect the original primitive? I'm not sure
that that way I could make sure the new one would have the same name that
the old one had before I destroy it though. All in all that sounds like a
technique that would work but would have downsides and it sounds a bit
clumsy to me.

As pdata is a array we might be able to append to it like to a normal array?
Perhaps then forcing the engine the rebuild the object? Maybe there are
better trees to bark up and I picked the wrong one?

Yours,
Kas.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.pawfal.org/pipermail/fluxus-pawfal.org/attachments/20100803/cb1d5ff1/attachment-0002.htm>


More information about the Fluxus mailing list