Re: [Gzz] PEG: vobn

[Tuomas Lukka]

On Thu, May 22, 2003 at 07:33:18PM +0300, Asko Soukka wrote:
> Thu, 22 May 2003, Tuomas Lukka kirjoitti:
> > For the fillet tests (and probably also other things)
> > we need N-coordinate system Vobs as a generalization
> > of the current 1, 2, 3 -cs vobs. N-coordsys Vobs will
> > be used to allow a vob which is a graph node that knows
> > all its neighbours at render time.
> 
> Will the current interfaces for 1, 2, 3 -cs vobs be replaced
> by this generalized one? Or are they left for effieciency?

They will remain for efficiency. N-coordsys vobs are probably
going to be rare but sometimes necessary.

For the multi-end fillet the only reason we need them is
that we need to sort the coordsyses for the other ends
of the connections by the angle they leave the currently
being-rendered node in, and the angle may (will) change
between frames.

> just curious

Excellent!

        Tuomas