news:3fe68802$1@news.povray.org...

> All three rendering systems you mentions (Mentalray, PRman, Mantra) use
> tessellation to map polygons onto the surface of a nurbs just before
> rendering. They actually render the poly's not the nurbs.

Perhaps I'm putting my foot in my mouth here, but isn't this done on-the-fly
and adaptative ? In a car modelled with Rhino for instance, the car body and
the gearshift are both complex objects requiring a lot of polys when seen in
close-ups. However, they may not need the same number of polys depending on
the view of the model. In an animation doing a fly over from the exterior of
the car into the interior, I imagine that the renderer would "decide" on the
optimal number of polys, depending on what objects are visible and close to
the camera, thus saving a lot of RAM.

> In fact it
> would seem faster to have already processesed the NURBS to creat the 500M
of
> poly's before render time so this is not done every time a frame is
> rendered. Is not speed the real issue.

I don't know. Possibly tesselating NURBS and rendering the polys is faster
than rendering them in a native form.

G.


-- 

**********************
http://www.oyonale.com
**********************
- Graphic experiments
- POV-Ray and Poser computer images
- Posters

Post a reply to this message