Pseudo Attractor problem (tree structure)

Hello Guys,

The idea here is to have a series of curves affecting the z coords of some points along intersecting curves

This is my method:

find the intersection points of multiple curves. Then find the closest point to these intersection Pts. and move them cording to a graph mapper and then rebuild curves rebuild the curves.

But I encountered two main problems:

1- After affecting the curves and moving the points in order to reconstruct the curves again. I get a tree structure, that even flatting won't solve. So the reconstruction is not properly done

2-When a curve intersect more then one time the same curve (2 intersection pts) Only one of them is affecting the points.

Here are the rhino files as well as the definition

Any idea

Attachments:

attrraction_lines.3dm, 29 KB
attractor-03.gh, 9 KB

Replies to This Discussion

Permalink Reply by Michael Pryor on January 28, 2013 at 11:05am

"Then find the closest point to these intersection Pts" as far as i see u have 2 items. curves and intersection points. So what are the points you are testing against to be closest to the intersection points?

Permalink Reply by Claudio on January 28, 2013 at 11:23am

Hi Michael,

Hope this screenshot is clear. thank you for your help

What I meant by ''find the closest points to the intersection Pts'' is that the Straight lines are flat then they are dived in many points and the closest points to the attractor Pt (intersection PTs) are then moved in order to reconstruct the line again(now with an indentation in their shape).

The what if I want to have more then one intersecting curve (the wavy ones) how would I reconstruct the lines again.....Hmmmm?

Permalink Reply by Michael Pryor on January 28, 2013 at 1:03pm

Attachments:

EXAMPLE.3dm, 50 KB
attractor04.gh, 8 KB

Permalink Reply by Claudio on January 29, 2013 at 9:57am

Thank you Michael I'll try your definition..

cheers

Permalink Reply by Claudio on January 30, 2013 at 5:10pm

Hi Michael,

I tried your definition and modified a little bit. It does what is supposed to but I encountered an other problem. It seams that the attraction part of it creates a noise that translate in to a strange curvature along the curves (see image).

Is it possible to get rid of this noise and have a smooth if not regular curvature along the curves. I've uploaded the new definition as well as the rhino file

thank you for your help

Attachments:

attractor06.gh, 12 KB
EXAMPLE.3dm, 56 KB

Permalink Reply by Michael Pryor on January 30, 2013 at 11:03pm

well actually the definition is fine. What is happening here is because you are basically using intersections as attractors. When your curve runs parallel to your straight lines there is no attractor there until it passes through a line. If you look under where your arrow is there is a gap also between the attractor points. This is why. A way to smooth this out would be

1. remove the flip matrix, poly line, and intersect components. Connect the curves that connected into the b of intersect directly into g of pull.

2. Make the U and V more dense for surface division to get more lines to intersect with.

In both cases density should even things out plus adjusting graph mapper settings.

RSS