Subject:
|
Re: primer/tutorial on BFC?
|
Newsgroups:
|
lugnet.cad.dev
|
Date:
|
Sat, 3 Jan 2004 07:52:51 GMT
|
Viewed:
|
1890 times
|
| |
|
|
In lugnet.cad.dev, Lars C. Hassing wrote:
> In lugnet.cad.dev, Travis Cobbs wrote:
> > In lugnet.cad.dev, Steve Bliss wrote:
> > > That shouldn't work -- the rendering program should detect that you've mirrored the
> > > object, and adjust accordingly. Unfortunately, I see that it does 'work' in L3Lab.
> >
> > I do agree that it isn't supposed to work, because the BFC spec very carefully
> > states that any polygons drawn with an aggregate matrix that causes mirroring
> > must be flipped back.
>
> I think it works just fine.
You're right; I'm wrong. I'm still having a hard time visualizing it, but your
stud example below convinced me that negating the one value in the matrix will
indeed flip a 2D primitive like a disk, and this is the expected behavior. In
fact, if it weren't for L3Lab detecting the mirroring, and inverting the
winding, the disk WOULDN'T flip.
For others reading this, the stud does make a good example. If you mirror it
around Y, the whole thing flips over (including the disk on top). So instead of
pointing up, the disk now points down. If you remove everything but the disk
from the file, you will of course get the same behavior, and end up with a disk
that has been flipped. Consequently, a mirror around Y will flip a disk (or any
other 2D shape in the X-Z plane).
> > Or you could just use INVERTNEXT, which will work for both 2D and 3D objects,
> > and doesn't require any additional matrix math (Isn't that what it's there
> > for?):
> >
> > 0 BFC INVERTNEXT
> > 1 16 0 0 0 10 0 0 0 1 0 0 0 10 4-4disc.dat
>
> Yes, that would be nicer.
> As Tore correctly stated, L3Lab does support INVERTNEXT.
For clarity, I think INVERTNEXT would be preferable to mirroring the matrix. It
acts as expected in all circumstances, while a mirror matrix will only work for
a 2D shape that isn't perpendicular to the mirror plane.
--Travis Cobbs
|
|
Message is in Reply To:
 | | Re: primer/tutorial on BFC? [DAT]
|
| (...) I think it works just fine. L3Lab correctly detects that the matrix is mirroring, see View/Tree View... and click the BFC radio button, it should say INVERT in front of the line. When we discussed the BFC business the general assumption was (...) (21 years ago, 3-Jan-04, to lugnet.cad.dev)
|
19 Messages in This Thread:
 
 
 
- Entire Thread on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
|
|
|
|
