[chiplotle-discuss] Catmull-Rom

francesco fantoni francesco at hv-a.com
Thu Mar 18 04:43:30 EDT 2010


Hi Victor,

I don't know if it is good to add the (x,y) parameter to these curve
classes, because I think it can not be used in an effective way for the
particular nature of the class. The user might think: "why am I supposed
to put this (x,y) if then the curve is defined by a list of absolute
points?". I think it may be confusing and sound useless to user.
On the other end I understand and valuate your concern about
consistency.
I don't really know what the right thing is, personally I would choose
the solution that is less confusing for a user, but you know the whole
project far better than I do, and of course if you think that breaking
the consistency is an issue, then I'll add the (x,y).

For what it concerns the rational Bezier, I think that in the case all
the points have the same weight, we fall back to a non-rational bezier
curve, because the equation is reduced to the normal Bezier equation

http://atrey.karlin.mff.cuni.cz/projekty/vrr/doc/man/progman/Rational-Bezier-curves.html

and actually if you try different sets of weights, when all weights are
set the same, whichever is their value, you get always the same curve.

The idea of having a cycle-through if the weight list is shorter than
control-points list is really a good idea, I will try to implement it,
and I will add the 'pen' parameter too.

thanks a lot for your kind appreciation and for the suggestions.


best, francesco


> 
> 
> 
> On Tue, Mar 16, 2010 at 12:49:50PM +0100, francesco fantoni wrote:
> > added Catmull compound class, for Catmull-Rom curve interpolation.
> > Similar to Path command, it draws a Catmull-Rom cspline, first and last
> > control points are not drawn as part of the curve but just define the
> > starting and ending tangent, the intermediate control points are on the
> > curve.
> > 
> > Catmull(control_points, xy=None, points_to_compute=None
> > 
> > best, francesco
> > 
> > _______________________________________________
> > chiplotle-discuss mailing list
> > chiplotle-discuss at music.columbia.edu
> > http://music.columbia.edu/mailman/listinfo/chiplotle-discuss
> > 
> _______________________________________________
> chiplotle-discuss mailing list
> chiplotle-discuss at music.columbia.edu
> http://music.columbia.edu/mailman/listinfo/chiplotle-discuss



More information about the chiplotle-discuss mailing list