Subject:
|
Re: CW/CCW, vertex sequence, co-planar, convex, (115kB)
|
Newsgroups:
|
lugnet.cad.dev
|
Date:
|
Wed, 29 Sep 1999 20:50:44 GMT
|
Viewed:
|
522 times
|
| |
|
|
There are commands in 3DS and 3DSMAX to "unify" normals (normals are what we're
discussing... the vector that points perpendicular to a face, indicating which
side is "out", which side is shaded or rendered)... i would assume there is
documentation of good routines for determining the inside/outside of complex
objects, therefore allowing normals to be unified...
the biggest problem with the LDraw parts/primatives library is that they
weren't modeled in a robust modeling app... laying out all vertices by hand
made it possible for every type of human error or preference to play a role in
the library's creation... (i know my parts wouldn't pass a CW/CCW test) we've
all created bowties, concave polys, concident verts, etc... even elements
created by me in 3DSMAX and converted using MAX2DAT were not checked for
"normal-friendliness"...
looking back through all of the discussion from the LCAD mailing list made
something abundantly clear to me... we're starting to see problems because the
technology used to create LDraw was not consistent with true 3D
modeling/rendering techniques.. so now that we are seeing an emergence of new
programs that utilize "prefab" 3D techniques, we're seeing problems with the
basic construction of our library since LDraw didn't care about the orientation
of normals and other aspects of 3D design necessary for advanced rendering
to work correctly...
true 3D programs don't deal with infitintely thin objects well... anyone who
has ever built a doom level knows this ("hall of mirrors", for the
experienced)..
but many of our primatives are infiniteley thin (4-4cyli.dat) and were created
to be used as 2-sided objects with zero thickness... this way they could act as
either an outside *or* inside surface with normals facing in both directions...
this won't work with the direction we're headed in..
making the programs deal with the innacuracies and correct them "on the fly" is
wrong... totally re-rengineering the library is practically impossible.. but
there is a real issue getting started here.. and it's only going to get worse..
J
|
|
Message has 2 Replies:
 | | Re: CW/CCW, vertex sequence, co-planar, convex, (115kB)
|
| (...) It wouldn't be impossible to fix primitive usage so that primitives such as 4-4cyli.dat are always applied as infinitely thin 1-sided objects. Would that solve the problem? (...) I don't think re-engineering the library is impossible. (...) (25 years ago, 30-Sep-99, to lugnet.cad.dev)
| | | Re: CW/CCW, vertex sequence, co-planar, convex, (115kB)
|
| (...) I was curious, how do you add a stud to a part if you're using MAX2DAT ? Can it be done in 3DS or you have to edit the part later ? I thought that we've agreed that the new parts would have to pass L3P tests for coincident verts and bow-ties, (...) (25 years ago, 30-Sep-99, to lugnet.cad.dev)
|
Message is in Reply To:
| | Re: CW/CCW, vertex sequence, co-planar, convex, (115kB)
|
| (...) I miss that kind of discussion, those were very interesting subjects. Also, where's Jeff Findley ? (...) I've already thought about a ray intersection algorithm but I've never tried to implement it, I might try do it now. Some problems that I (...) (25 years ago, 29-Sep-99, to lugnet.cad.dev)
|
53 Messages in This Thread:
 
 
- Entire Thread on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
This Message and its Replies on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
|
|
|
|
