Rotate Loft or boundary surfaces from one attractive point

Hello,

When I move the attractive point in Rh , I see that the different lofts don't rotate properly with respect to the center of the attractive point and the angle.
I think my problem is when the different angles take a negative value I don't know how GH manage this.
How to solve this problem ?

Another question :
For experts, is the way I take for finding the normal vector, generated from the tangent vector of each point, are well done?

Thank a lot for your help

Attachments:

elevation facade.3dm, 49 KB
test elevation facade with one attractive point.gh, 24 KB

Replies to This Discussion

Permalink Reply by Joseph Oster on May 14, 2017 at 11:10am

Instead of fixing your rotation problem, I simplified your code by using the vector to the attractor point to construct the lofted surfaces. So each one is constructed with appropriate orientation.

Attachments:

testelevationfacadewithoneattractivepoint_2017May14a.gh, 15 KB

Permalink Reply by Samco on May 14, 2017 at 2:22pm

Joseph,

Many thanks for your help and quick response!
Waouh, big lesson for me !
Why do I do complicated things when you can do them simply?
Enormous simplification of the actual code
That's what I'm learning.
Great Respect for the time you spent for me

Permalink Reply by Joseph Oster on May 14, 2017 at 2:53pm

My pleasure. The rotation approach you started with is perfectly valid and can be solved, I just got lazy and took a short-cut. There are so many ways. Keep it simple when possible.

Permalink Reply by Samco on May 15, 2017 at 3:28am

Hello Joseph,

If you have a little bit time for me
I try to stay simple, and I just want more control point instead of arc but I don't understand why some boundary surfaces don't calculate and display.

What's wrong ?

Thanks

Attachments:

testelevationfacadewithoneattractivepoint_2017May15b.gh, 13 KB

Permalink Reply by Joseph Oster on May 15, 2017 at 8:07am

The problem is that the top points (derived by dividing the top curve) are not co-planar with the other three points. You can see this by eliminating the {0;3} input to 'Entwine'. The three remaining points form a planar curve which makes a surface.

Permalink Reply by Samco on May 15, 2017 at 1:14pm

Effectively ....
Fatal and trivial error..Sorry I tired.
Thank you again for your intervention

Permalink Reply by Samco on May 15, 2017 at 1:53pm

Not clean clean... but very simple with Srf4pt !

Attachments:

testelevationfacadewithoneattractivepoint_2017May15c.gh, 14 KB

Permalink Reply by Joseph Oster on May 15, 2017 at 6:15pm

Sure, that works. Sorta... you can see the twist from the top view.

You can get a planar surface, arc or trapezoid, by determining the top points differently. For example, instead of dividing that top line to get those points, create planes parallel to the vectors for each bottom point, then get the intersection of those planes with that top curve to get the top points. 'Simplify' or other means might be necessary get the data paths to match again to create curves and/or surfaces.

Permalink Reply by Joseph Oster on May 15, 2017 at 6:36pm

All planar surfaces because top points are defined that way! See yellow group:

Attachments: