| | | | |
| |
| Gary Williams wrote in message ...
> An orientation matrix doesn't contain enough information to invert an
> asymmetrical part; if you assume it does, this would result in a part with a
> mirrored shape.
>
> I propose that (1) INVERT be explicitly used instead of relying on negative
> determinants, and (2) no matrix should ever have a negative determinant.
>
> I don't know how many asymmetrical subparts there are, if any, so I don't
> know if this is a relevant point or not.
Well, what I meant to say is:
Matrices with negative determinants may be used for mirroring a part, BUT
the matrix should never be assumed to perform the actual inversion.
If a matrix has a negative determinant, the rendering program should reverse
the order that it processes that subfile's (and that subfile's subfiles')
vertices.
If an INVERT is encountered, then the rendering program should reverse the
vertex order again.
I apologize if this isn't making any sense, but to me it appears to be a
valid issue.
-Gary
| | | | | | | | | | | | | On Thu, 21 Oct 1999 01:07:30 GMT, "Gary Williams" <graywolf@pcpros.net> wrote:
[clipped good stuff]
> I apologize if this isn't making any sense, but to me it appears to be a
> valid issue.
You are making sense, it is a valid issue, I've been missing the boat on
inversion for awhile now, and I'll update the document.
Should there be an explicit statement about mirroring in the document, perhaps
in a DAT file authors section? (Am I repeating myself between messages?)
Steve
"Never trust anything that can think for itself, if you can't see where it keeps its brain."
| | | | | | |