>http://zhengjj.freeyellow.com/html/publications.html
 >"Control point surfaces over non-four-sided surfaces"

This is really interesting, I was looking for such a paper for weeks, 
thanks Rune!

The cubic version (degree 3) of both, 3- and 5-sided patches should 
smoothly connect to regular 4-sided cubic bezier patches (as supported 
by POV) - I'll try to implement those into JPatch.

It would be interesting if it is possible to subdivide these patches 
into 3 or 5 4-sided patches respectively without loosing continuity, as 
described here

http://jpatch.sourceforge.net/developer/hash_complete/hash.html

under "5-Point Patches". This is beyond my math skills, maybe someone 
here can help?

If splitting the 3- and 5-sided patches into smaller 4-sided patches is 
possible the modeler could export bezier-patches only (no triangels) and 
thus leave the triangulation to the renderer (which would result in 
smaller files and better quality because the detail level can be set at 
render time).

Thanks,
-Sascha


Rune wrote:
> Using a different amount of samples won't change anything I'm afraid.
> You will probably just get a less detailed wrinkle instead of a highly
> detailed one...
> 
> To see a discussion on making a real 3 point bezier patch, see:
> http://news.povray.org/povray.advanced-users/26671/
> 
> To see it implemented (in pov-ray SDL):
> http://news.povray.org/povray.text.scene-files/26676/
> 
> To see an animation where it's implemented:
> http://news.povray.org/povray.binaries.animations/26690/
> 
> Rune
> --
> 3D images and anims, include files, tutorials and more:
> rune|vision:  http://runevision.com **updated Jul 25**
> POV-Ray Ring: http://webring.povray.co.uk
> 
>

Post a reply to this message