Facet with tangent planes?

I need to facet a random surface (ex hyperbolic parabolic) with tangent planes from random points on the surface. My problem is to trim the tangent planes.. Is there a better way to do it? (I'm a Rhino and grasshopper noob)
https://www.dropbox.com/s/wfb7faod3kuez3a/Facet_beta_2.gh

Replies to This Discussion

Permalink Reply by David Rutten on August 28, 2012 at 6:36am

I suppose this wasn't what you were after?

David Rutten

david@mcneel.com

Poprad, Slovakia

Attachments:

Facet_beta.gh, 6 KB

Permalink Reply by David Rutten on August 28, 2012 at 6:41am

Here's something else you're probably not looking for :)

David Rutten

david@mcneel.com

Poprad, Slovakia

Attachments:

Facet_beta.gh, 6 KB

Permalink Reply by Thomas Perkov on August 28, 2012 at 8:03am

Nope that is not what what in were looking after :-( But thanks for the fast answers... it's actually the same problem as:

http://www.grasshopper3d.com/forum/topics/trim-surfaces-1

but here i need grasshopper to chose the first plane(s)..

Permalink Reply by Arthur Lara on June 7, 2016 at 5:45am

Attachments:

planar_surf1.gh, 18 KB
planar_surf_voronoi2.gh, 15 KB

Permalink Reply by Arthur Lara on June 7, 2016 at 6:17am

Attachments:

planar_surf_voronoi3d.gh, 15 KB

Permalink Reply by Andrew Heumann on August 29, 2012 at 8:49am

It would be very helpful if you could explain what you're after a little better. What is the difference between the solution you're after and the ones David provided?

Also, it seems like the idea of "trimming the tangent plane with the hyperbolic surface" is a little bit tricky in and of itself; where should the plane be cut exactly? A tangent plane extended to infinity on a hyperbolic surface will intersect it precisely along a sort of "X" - not sufficient to "clip" it on all sides.

Even if you offset the plane in one direction or the other, you will only ever get two intersection curves, which will not intersect; this doesn't provide a closed bounding region with which to trim your facets.

If I've got it backwards, and you're trying to trim the hypar with the many planes, you still have effectively the same problem.

It would be especially helpful if you could include a drawing or diagram to help explain what you're after.

Permalink Reply by Thomas Perkov on August 29, 2012 at 9:49am

The Idea is to trim all tangent planes.

After the tangent planes are found I don't need the hyperbolic parabolic, I only use the hyperbolic parabolic to finde the planes.

If I move the points on the surface i would get a different pattern.

Why the hyperbolic parabolic? It's double-curved and harder to facet.. If I can facet the hyperbolic parabolic i can facet all surfaces with different patterns made by controling the points.

When I say "random" I mean any points og surfaces given by me ;-)

Permalink Reply by Andrew Heumann on August 29, 2012 at 9:51am

I still don't understand how you want to trim the tangent planes. is the idea to trim them against each other?

Permalink Reply by Thomas Perkov on August 29, 2012 at 9:53am

yup

Permalink Reply by Thomas Perkov on August 29, 2012 at 10:01am

Permalink Reply by Andrew Heumann on August 29, 2012 at 10:51am

Thanks; the graphic is helpful in clarifying what you're after.

However, I'm not sure it's so simple as that. David's second example illustrates something very important about the nature of the problem, which I'll try to illustrate in diagrams:

Take a single facet from david's solution (Which satisfies the requirement of producing a set of tangent planes based on random points):

And extend it and its neighboring facets to be larger, intersecting planes:

Then color code all the neighboring planes, and then take the intersection of each tangent plane with the highlighted facet.

If you look at just the plane and the intersection curves:

You can see that, for one, there's any number of combinations of these lines that could be used to trim that facet, and moreover that no combination seems to be a valid one, matching at edges with all neighbors the way your diagram shows.

All I mean to communicate by this is:

1. "trim against each other" is much, much more complicated than it sounds

2. a predetermined collection of surfaces tangent to a non-convex doubly-curved surface may not have a valid set of intersections that create a clean, planar polygonalization.

3. a self-described "grasshopper noob" might want to start with a simpler problem :) If you limit the problem to purely convex shapes the problem is much, much easier.

Permalink Reply by Thomas Perkov on August 29, 2012 at 1:22pm

First of all Thank for the awesome response ;-) Could you upload the .ghx file? :-D

I have started a lot of project in grasshopper, this is the toughest problem.