John W. Barrus (barrus++at++merl.com)
Thu, 16 Jan 1997 08:47:01 -0800
When Performer creates tri-strips from imported polygons, it looks at the
face or vertex normals to determine which direction the first triangle in
the new tri-strip should be facing (instead of looking at the cross product
of the first two eges of the polygon).
We had a few models created in a modeler that is not very careful about
making sure the normals and edge cross-products correspond ((V1-V0)X(V2-V1)
* N should not be negative where N is the polygon or vertex normal, V0, V1
and V2 are vertices describing the triangle when using counter-clockwise faces.)
Since our normals were accidently reversed in one part of the model, some of
the faces seemed to be missing when perfly read in the model. What happened
was that faces with bad normals were getting reversed when they were
converted to tri-strips.
Of course, using the normal is faster than calculating the cross-product of
two vectors, but it would be nice if Performer provided an option to it's
tri-strip builder that checked the cross-product instead of the normal at
the expense of slightly slower load rate (or even a utility that checked it
and pointed out potential problems).
Any thoughts or suggestions?
John B.
Note new address and phone number:
John W. Barrus Research Scientist
Mitsubishi Electric Information Technology Center America
1060 Arques Avenue
Sunnyvale, CA 94086
phone: (408) 523-6806
fax: (408) 524-0613 e-mail: barrus++at++merl.com
=======================================================================
List Archives, FAQ, FTP: http://www.sgi.com/Technology/Performer/
Submissions: info-performer++at++sgi.com
Admin. requests: info-performer-request++at++sgi.com
This archive was generated by hypermail 2.0b2 on Mon Aug 10 1998 - 17:54:21 PDT