Michael T. Jones (mtj++at++babar.asd.sgi.com)
Tue, 15 Oct 1996 10:25:25 -0700
Well, as a matter of fact, the Graphicon 2000 did an ok job
on such things. It could render quads directly since it had four
edge evaluators that gave a thumbs-up/thumbs-down signal for all
the pixels inside the bounding box.
Rendering bow-ties (the phrase for such quads) is not really a
virtue in many cases:
1. you know that it's bad when you're loading it so why not
deal with it then rather than at the 11M/sec part of the
graphics pipeline.
2. color, lighting, clipping, and other such interpolation
related issues are almost sure to be inconsistently done
on any machine that renders more than 3-sided polygons
directly.
3. intersection results are going to give inconsistent normals
anywhere "in" the non-planar quad. This is the same as #2,
but it's generally host-based rather than in the pipeline.
My view on this is that points, lines, and triangles are not
only all that you need -- they're all that you want!
Michael
=======================================================================
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 - 17:53:46 PDT