Subject:
|
Re: stl2dat conversion tool beta2
|
Newsgroups:
|
lugnet.cad.dev
|
Date:
|
Fri, 11 Feb 2000 17:29:13 GMT
|
Reply-To:
|
rui.martins@link.ptNOSPAM
|
Viewed:
|
1172 times
|
| |
|
|
> On Thu, 10 Feb 2000 11:36:31 GMT, Rui Martins <Rui.Martins@link.pt>
> wrote:
>
> > I haven't had time to check if the stl supports NURBS, or is it just a
> > normal poligon based format ?
> STL only support triangles. Nurbs (and other geometry) must be
> converted by the CAD software to triangles. That's
> why I choosed STL : it already contains triangles and that's very near
> LDraw format. (In a geometric point of view)
Ok, my cad program can do that, but I have to mess with it because the
options to tesselate a surface are somewhat complex, at least if you want
to be able to have some control on the number of polys generated.
Does STL allow only for triangles ? no Quads ? no N sided Polygons ?
> <SNIP>...
> > Buy the way, do you correctly generate these lines (expecially the
> > optional ones) ?
> Yes, I generate automatically the lines between two adjacent facet. If
> the angle is lower than a certain value (about
> 45°) the line is optional.
Is this the better way ?
Which way is the best way ?
Well, I havent tought about that iet.
The optional lines are required only to complement the silhuete of the
part, so if the angle is different from zero (0) then it's a candidate
for an optional line.
Since we use optional lines to define this silhuete for aproximated round
parts, this value should be defined by the less precision we accept
(8 sub divisions in a circle) which ields 45 has the angle of choice
(AKA default value), but his value should be configurable.
> However, the program only check strictly
> adjacent facets. I have to improve it so it check also
> partial adjacent facets.
I don't know exactly what you mean by partial adjacent faces, but if
what you mean is something like this:
1------------2
| |
3------------4-----5
| |
| |
6------------------7
legend:
- Numbers are points
where the top quad is defined by vertex sequence (1 3 4 2)
and the botton quad is define by vertex sequence (3 6 7 5).
where the partial adjacent face (edge) would be (3 4) & (3 5)
Just don't bother, because this is Plain BAD Modeling, and no conscious
modeller will do this, instead he should break the bottom poly into three
triangles using vertex 4, or define another point below vertex 4 and break
into two quads.
Tesselation algorithms will NEVER do this.
An inexperienced modeller will eventually do this, but it shouldn't
because:
This kind of modelling will make cracks appear bettween the two edges
when looking at it from some angles, due to inacurite (discrete)
rasterization on your monitor, especially if the edge is drawn in
different directions depending on the poly beeing drawn.
> Are you interested to be able to change this value ?
Of course !
> Are they other things that you wish to control ?
Nothing springs to mind right now.
Inte'
Rui Martins
|
|
Message has 1 Reply:
 | | Re: stl2dat conversion tool beta2
|
| (...) No ONLY Triangles. Look for STL format description (URL) >Buy the way, do you correctly generate these lines (expecially the (...) <SNIP> (...) The program check the angle between two adjacent facets if abs(angle ) < 1e-5 No Edge if abs(angle) (...) (25 years ago, 11-Feb-00, to lugnet.cad.dev)
|
Message is in Reply To:
| | Re: stl2dat conversion tool beta2
|
| (...) STL only support triangles. Nurbs (and other geometry) must be converted by the CAD software to triangles. That's why I choosed STL : it already contains triangles and that's very near LDraw format. (In a geometric point of view) (...) Yes, I (...) (25 years ago, 10-Feb-00, to lugnet.cad.dev)
|
14 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
|
|
|
|
