Re: pfMorph and iv loader

New Message Reply Date view Thread view Subject view Author view

Marcus Barnes (marcus++at++multigen.com)
Thu, 2 May 1996 14:26:09 -0700

On Apr 29, 7:14am, ceti wrote:
> Subject: pfMorph and iv loader
> Well,
> I have two geometries built using Explore ( wavefront ), converted in iv
> file format. Geometries are same, just shapes are change . I use pfLoadFile
> to load them and I want to morph from one to the second using morphing
nodes.
> Here is the problem, when checking the two geosets, repartition of the
> vertex in the strips is not the same. GULP!

The current pfd[Geo]Builder utilities do not handle this. You need to be able
to process (at least) pairs of vertices as a single unit: the split (actual)
vertex and the final (morph) vertex.

> So I assume that the iv loader tries (and may be succed) to optimise the
> object but here I want it to be dummy.

Yes. Depending on the "suitability" of the source data, a non optimizing loader
can create pfMorph source attributes fairly easily. All the work associated
with sorting polygons by state wrt pfGeoSet's must now be leveraged towards
building source attributes for (few) pfMorph nodes. pfGeoSet(s) receive the
pfMorph destination attribute.

> Is there a way to avoid crossing optimising step or should I write a dummy
> loader ??

What's needed is a pfdMorphBuilder ;-)

... the work of mapping all the (formerly) geoset data into the interior
pfMorph node attributes efficiently (by state, tmeshed in pairs, etc.), with
the morph destinations binding back to the geosets (leaves), is non-trivial.

Regards.

--
    __  ___      ____  _ ______          Marcus Barnes, Member Tech. Staff
   /  |/  /_  __/ / /_( ) ____/__  ____  MultiGen Inc, 550 S. Winchester
  / /|_/ / / / / / __/ / / __/ _ \/ __ \ Blvd. STE 500, San Jose CA 95128
 / /  / / /_/ / / / / / /_/ /  __/ / / / PH:1-408-556-2654 FX:1-408-261-4102
/_/  /_/\__,_/_/\_\/_/\____/\___/_/ /_/  EMAIL: marcus++at++multigen.com

New Message Reply Date view Thread view Subject view Author view

This archive was generated by hypermail 2.0b2 on Mon Aug 10 1998 - 17:52:52 PDT

This message has been cleansed for anti-spam protection. Replace '++at++' in any mail addresses with the '@' symbol.