Nurbs Curve support added to OCurves schema in dev clone.

[Lucas Miller]

None of the tools currently support it, but I've had it in place for about a month and it feels pretty stable.

Alex, what do you think of it?

http://code.google.com/r/millerlucas-dev/source/detail?r=af02caafb128524bf0336ead474938140f156bd7

[Alex Suter]

Code looks good.

 

From the tests it looks like the original style will still be supported. If a client was reading the new style with old code would it make any sense?

--
You received this message because you are subscribed to the Google Groups "alembic-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email to alembic-dev...@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[Lucas Miller]

Most of the readers that I am aware of should draw something that is a very rough approximation.

We also still need to pick whether the number of knots specified should be N+2M-1 or N+2M+1

http://download.autodesk.com/us/maya/2011help/API/class_m_fn_nurbs_curve.html

Lucas

[Alex Suter]

We use N+2M+1 internally, but it’s probably more useful for the world if we match Maya. Not sure what they use. Anyone know?

[Lucas Miller]

I believe Houdini likes N + 2M + 1 as well, but I know of another studio that prefers a N+2M-1.

This one might require a good old fashioned coin flip.

Lucas

[Lucas Miller]

More info from one of our curves guys:

Maya allows  N + 2M - 1 knots in the interface, but internally must still use N + 2M + 1, because that is how many a real NURBS curve/surface uses.  Maya just adds an extra knot at the beginning and end with value 0 and 1, respectively, for a normalized curve (or duplicates of the first/last knot), and the user can't change that.  In simplified terms, all Maya default NURBS curves interpolate the first and last control point, but if you want a general NURBS curve where this is not the case, you might want the first/last degree +1 knots to be different.  Therefore a general user should have total control over specifying all knots (N + 2M + 1), not just N + 2M - 1.

I think we should use the N + 2M + 1 form.

Lucas

[Alex Suter]

Definitely gets my vote.