David,
It's a pretty long definition, but basically what's happening is that
I have created a series of planes that intersect a series of
interpolated curves.  At each intersection, I will get somewhere
between 5 and 6 points, which I can connect with straight line
segments and join to create a series of 5 or 6 sided closed
polylines.  I then offset this curve 4" to get a inner curve (I'm
creating a ribbed structure and we don't want the ribs to be solid all
the way through).  Long story short, I have a set of outer curves and
a set of inner curves which I then re-orient to the XY plane for CNC
milling.  However, each rib section is actually the composite of 2
overlapping sections.  For the outside of the rib, the curve union
seems to work great.  However, it doesn't give me the desired output
for the inner curves.  So, the way I worked it out earlier was through
the process above.  By creating a planar srf out of the re-oriented
offset inner curves and then feeding those into a Solid Union.
However, if I do this in my large file, then it doesn't seem to work.
If I create a simplified scene, so I can test other options... I
usually bake in the inner curve layers and then reference them back
into GH.  When I do this it works fine.  So, this is why I was asking
if there was a difference.
BTW, man those new curve booleans look great.  They would have saved
me a lot of time and I can't wait to try them out.
Best,
Andy

On Feb 10, 7:56 am, David Rutten <[email protected]> wrote:
> Rchitekt,
>
> I just thought of a possibility. If the curves you generate in
> Grasshopper are not valid (or if they have otherwise funky
> properties), they might not work well as PlanarSurface input. However,
> the act of adding a curve to the Rhino document will automatically
> call a number of fixing-algorithms which might take away the problem.
> It's a long shot, but this may explain why grasshopper curves do not
> work whereas referenced curves do work. How are you creating your
> input curves?
>
> I have not been able to repeat the problem so far. If I build a
> network similar to the screenshot you posted, it works with both kinds
> of curves. But since we are not using the same version of Grasshopper,
> this is hardly conclusive evidence of anything...
>
> --
> David Rutten
> [email protected]
> Robert McNeel & Associates
>
> On Feb 10, 4:18 pm, David Rutten <[email protected]> wrote:
>
> > Hi Rchitekt,
>
> > first, it makes no sense to me. There should not be a difference
> > between a referenced curve and a curve generated by a Grasshopper
> > component.
> > I'll look into this to see how the most recent build behaves.
>
> > Second, you'll be glad to learn I've added Union, Difference and
> > Intersection components for closed curves to Grasshopper, so this will
> > be much easier in the 
> > future:http://groups.google.com/group/grasshopper3d/web/Curve%20Regions.png
>
> > --
> > David Rutten
> > [email protected]
> > Robert McNeel & Associates
>
> > On Feb 10, 3:47 am, Rchitekt <[email protected]> wrote:
>
> > > Sorry... I forgot to post the screenshot.  Here's it 
> > > is:http://grasshopper3d.googlegroups.com/web/Ref%20geometry%20vs%20volat...
>
> > > On Feb 9, 6:47 pm, Rchitekt <[email protected]> wrote:
>
> > > > Is there any difference in the way Grasshopper deals with Referenced
> > > > Geometry vs. Volatile Data (aside from the fact that one has
> > > > persistent data and the other is not).  The reason I ask if that I
> > > > have 2 closed polylines that overlap each other, and I'm trying to get
> > > > the inside polyline... almost like a Curve Difference instead of a
> > > > Curve Union.  So, I've created a simple definition that creates a
> > > > planar srf out of both polylines and then using the Solid Union it
> > > > creates an object with a set of different faces.  Using a little math
> > > > you can eliminate all of the smaller faces so that your left with the
> > > > one big face in the middle which is just like creating a Curve
> > > > Difference (I think).  Anyway, the code seems to work when I select
> > > > each curve from Rhino using referenced geometry.  However, if I feed
> > > > those same curves into the curve data using volatile data, the Solid
> > > > Union fails which throws off the rest of the definition.  So, my
> > > > question is, why would it work with referenced geometry but not with
> > > > volatile data?
> > > > I guess I can always bake the curves into my scene and then re-assign
> > > > them back into my definition but it seems like I'm having to go around
> > > > my butt to get to by elbow.
> > > > Thoughts?

Reply via email to