Grasshopper

algorithmic modeling for Rhino

surface_01.jpg

 

I need to create this surface. I started by a rectangular grid.

Now I need to branch that grid in items of the same order,

like (0,1), (1,1), (2,1) and move them in z-direction.

Is that the best way to manipulate the points? 

Any advice wold be much appreciated! 

Thank you

Views: 1064

Attachments:

Replies to This Discussion

Hmm ...

1. NEVER EVER use the offset curve on surface thingy: slower than a stationary Harley Davidson.

2. Map the rectangles into the surface in question. Obviously we are talking trees here since this is not a "linear" arrangement (List could be OK in such an occasion).

3. Use some attractors for the push/pull "gradual" effect. Check that the values are sampled in an apples-to-apples basis with regard the data structure from stage 2. If not > Armageddon + Chaos + Mama Mia.

4. Extrude the mapped rectangles from stage 2 according values from stage 3.

That said I have several "similar" examples ... but not using components. Do by accident speak C#?

Just recalled that I've made some time ago this demo for a fellow new user with regard map matters.

Since this uses Lunchbox (that doesn't make trees) is a bit confusing - but I could add a line of C# that arranges things properly in a  tree manner (also doable with components).

Load R file first.

Anyway ... notify if you want a demo related with any attractor List imaginable in any mode (but as I said ... using solely code)

The usual behavior (no attachment possible) - 1

Attachments:

The usual behavior (no attachment possible) - 2

Attachments:

I'm trying something like that... How can I control the Z vector differently in the rows and the columns... 

It worked out in the X-direction, How can it work in the Y-direction for all the points?

Attachments:

Good news: see attached (use the already posted Rhino file).

Bad news: I had a very limited time to do this thus I've used some C#(*) from numerous other cases. Given the opportunity the Lunchbox List is "mapped" into a proper Tree.

(*) that C# way is like cancer: more is better.

On the other hand definitely some good Samaritan (who likes the component way) will post some non-C# attractor(s) approach.

Attachments:

Obviously if you don't need Lunchbox (for other type of "cell" arrangements) you can use the good old RectGrid thingy (and get rid some list>tree puzzles).

Notify if you need some other type(s) of control on the extrusion vectors (as they are are surf normalsAt centroid which yields a "bit" odd results if the hosting surface is not "continuously" convex - or concave depending upon the flip boolean value).

Attachments:

RSS

About

Translate

Search

© 2024   Created by Scott Davidson.   Powered by

Badges  |  Report an Issue  |  Terms of Service