Subject:
|
Re: Some Words To BFC
|
Newsgroups:
|
lugnet.cad.dev
|
Date:
|
Thu, 6 Apr 2000 18:28:12 GMT
|
Viewed:
|
2853 times
|
| |
|
|
In lugnet.cad.dev, Rui Martins wrote:
> On Thu, 6 Apr 2000, Leonardo Zide wrote:
> > > I have researched (Odd/Even), (BSP) and (Winged Edge). There is also the brute
> > > force method of supersampling raycasting, but this one is a NO GO.
> >
> > I think I'm using what you call odd/even, I cast a ray from the center
> > of each face to both directions and check the number of faces hit. If it
> > can't determine the correct orientation (hit the same number of faces
> > from both sides) that I try again from other points in the face.
>
> When do you actually stop, when you find the correct orientation ?
> or after a few fixed iterations ?
> Eventually you could end up with an infinite loop if you are not carefull.
> That method does NOT really garantees any correct results due to the way our
> parts are defined.
>
> See ya
>
> Rui Martins
Would it help to get primitive files cleaned up, and then make use of
what is known of the faces in the primitive to decide the orientation of
faces in the part?
What I mean is, if you follow a ray from the center of a face, and the
first/closest face hit by the ray is pointed toward the face, then that
would indicate that the face being tested should face toward the other
face.
Of course that presupposes the file being cleaned up has all the
inversion issues squared away. And some primitives would be no help,
like studs.
Steve
|
|
Message is in Reply To:
 | | Re: Some Words To BFC
|
| (...) When do you actually stop, when you find the correct orientation ? or after a few fixed iterations ? Eventually you could end up with an infinite loop if you are not carefull. That method does NOT really garantees any correct results due to (...) (25 years ago, 6-Apr-00, to lugnet.cad.dev)
|
61 Messages in This Thread:
 
- Entire Thread on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
|
|
|
|
