[cairo] Backend using subdivisional surfaces?

[Stuart Axon]

> ----- Original Message ----
> From: Carl Worth <cworth at cworth.org>
> To: Stuart Axon <stuaxo2 at yahoo.com>
> Cc: cairo at cairographics.org
> Sent: Tuesday, October 14, 2008 6:21:36 PM
> Subject: Re: [cairo] Backend using subdivisional surfaces?
> 
> On Mon, 2008-10-06 at 17:59 -0700, Stuart Axon wrote:
> > Hi,
> >  I filed bug #17733 asking for a blender uni-verse backend.  I've been looking at verse a bit more and it's base primitive is subdivisional surfaces. 
> > Forgetting about the verse part for a moment; how difficult would it be to make a backend that rendered to subdivisional surfaces?
> > Are there any artefacts that could occur in such a conversion?
> 
> Hi Stuart,                       
> 
> Could you describe in more detail what you mean here by a subdivisional
> surface?
> 
> Perhaps one way you could make that very concrete would be to propose a
> new cairo_subdivisional_surface_create function. What would the
> declaration of such a function be?
> 
> -Carl


Hi Carl,
  I'm going to have more of a look into this and get back to you when I have 
more details (I'm really learning about verse and cairo (and a lot of other 
technologies) all at the same time (slowly!)).

I'm not sure about the function definition as I have to get to know cairo a bit
more (but I will look into this).

I've posted the email I got on the verse mailing list which has some info to
start with.

There isn't much info here, but I'm forwarding the mail now so I don't forget
that it's there :)

- Hopefully I can learn more cairo and verse and get back with more useful
information in future.

S++

:Mail from verse mailing list follows:
> On Thu, Oct 16, 2008 at 4:07 PM, Stuart Axon <stuaxo2 at yahoo.com> wrote:
> > Is this the same thing used in verse? http://en.wikipedia.org/wiki/Subdivision_surface
>
> They're close, but not the full story. :|
>
> I'm not aware of a good reference-type description of Verse's
> subdivision surfaces, which is kind of silly. Perhaps Eskil has one?
>
> Catmull-Clark surfaces don't have *creasing* support, which is in
> Verse. With creases, you can express how smooth each corner and edge,
> which means you are not limited to all-rounded objects.
>
> I assume you've alread read the relevant part of the spec, here
> <http://verse.blender.org/fileadmin/verse/spec/n-geometry.html>, for
> (slightly) more on creasing.
>
> This is a very unsatisfying answer, I know. Sorry.
>
> Regards,
>
> /Emil
> _______________________________________________
> Verse-dev mailing list
> Verse-dev at blender.org
> http://lists.blender.org/mailman/listinfo/verse-dev

__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com