Subject:
|
Re: CW/CCW, vertex sequence, co-planar, convex, (115kB)
|
Newsgroups:
|
lugnet.cad.dev
|
Date:
|
Fri, 1 Oct 1999 20:36:47 GMT
|
Viewed:
|
2392 times
|
| |
|
|
[ I got a bit too much good French red wine, so maybe I
shouldn't try to answer now, but ... ]
Steve:
> On Fri, 1 Oct 1999 14:24:00 GMT, sparre@sys-323.risoe.dk (Jacob Sparre
> Andersen) wrote:
>
> > Lars:
> >
> > > Leonardo Zide wrote
> > It looks like (C)CW settings should propagate to included
> > files.
>
> No, they shouldn't. Just because the part XYZ is CW, a program can't
> assume that the primitive ABC is also CW.
Right.
> > What about:
> > 0 FACE DOUBLE-SIDED
> > 1 16 0 -4 0 6 0 0 0 4 0 0 0 6 4-4cyli.dat
>
> I like this. I don't like it for use on primitives, but DOUBLE-SIDED could
> solve the problem Leo pointed out, with being able to view the back side of
> patterns on transparent parts. Just make all the quads/triangles for
> patterns double-sided.
Shouldn't all elements of transparent parts be drawn?
> > > A part with no keywords is interpreted like it had begun
> > > with "0 FACE UNKNOWN".
> >
> > Shouldn't it rather inherit the setting from the file that
> > calls it? Or do we need a specific "0 FACE INHERITED" for
> > this purpose?
>
> I agree with Lars on this one. 0 FACE UNKNOWN should be the default for
> unmarked part files.
Yes. (I start to suspect that wine has a good influence on
my intellectual abilities)
> It seems like it would be as easy to write "0 FACE CW" as "0 FACE
> INHERITED". And generally, sub-files can't know anything about the state
> of the file calling them.
Yes.
> > Open primitives should definitly work with inherited (C)CW
> > settings, unless they are intended to be closed up when
> > used (like primitives for patterned parts).
>
> I don't get this. If a part is CCW and another is CW, how can a primitive
> be written to work with both by inheritance?
Hmm? I wonder what I was thinking?
> > > The only use for the UNKNOWN keyword I can see, is for a
> > > section of a part you want to make compliant at a later
> > > time.
> >
> > Yes.
>
> (It's also useful as a default value.)
Yes.
> There is a lot of "low hanging fruit", sections of the parts library which
> can be easily verified/corrected. Using the UNKNOWN keyword will let us
> put off fixing more difficult sections, in favor of fixing those parts
> which are easier (or more important) to fix.
Right.
> 0 FACE CW
> 0 FACE CCW
> 0 FACE DOUBLE-SIDED
> 0 FACE UNKNOWN
>
> Is there a practical difference between UNKNOWN and DOUBLE-SIDED? They'd
> have the same basic effect: the surfaces would always be drawn. But
> UNKNOWN could include artifacts like concave and bowtie quads.
No difference for the renderer, only for the parts authors.
DOUBLE-SIDED is _known_ to need both sides drawn, whereas
UNKNOWN just hasn't been sorted out yet.
Play well,
Jacob
------------------------------------------------
-- E-mail: sparre@cats.nbi.dk --
-- Web...: <URL:http://www.ldraw.org/FAQ/> --
------------------------------------------------
|
|
Message is in Reply To:
 | | Re: CW/CCW, vertex sequence, co-planar, convex, (115kB)
|
| (...) That would be nice, but it's unrealistic. (...) No, they shouldn't. Just because the part XYZ is CW, a program can't assume that the primitive ABC is also CW. (...) I like this. I don't like it for use on primitives, but DOUBLE-SIDED could (...) (27 years ago, 1-Oct-99, to lugnet.cad.dev)
|
53 Messages in This Thread:
 
 
- Entire Thread on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
|
|
|
|
