Grasshopper

algorithmic modeling for Rhino

Which components can model such a wavy texture on top of these Zaha Hadid vases?

Views: 1874

Replies to This Discussion

Hmm ... it's the second time that I've seen a "similar" puzzle.

In general - and roughly speaking - I would suggest to "distort" a nurbs via control points ... a task NOT for a novice by any means, nor something that it can been described by "listing" components (BTW: you can divide in U/V the distorted nurbs if "segments" are in demand).

I have code to do that but it's a quite complex script in C# (no GH components at all) for internal use:

It works (a) by rebuilding a nurbs by a user controllable amount of U/V points then (b) creates any imaginable SUBSET of control points ("shifting" them in U/V and/or "culling" them and/or ... whatever) and then (c) it modifies (using the nurbs normals) these VS any collection of attractors (in push/pull mode) and/or using a "global" distortion mode and/or random stuff and/or ... some other options.

Then it "combines" the modified points with the rest and does the new nurbs.

see 2 possible variations (out of a zillion) as regards "shifting" in U/V (using VERY LOW U/V "resolution" for speed).   

Attachments:

Thank you.

But is it possible to rebuild your algorithm using regular GH components (or may be add-ons)?

Theoretically maybe but it could require a lot of work and a rather very big collection of components (and it's not "algorithm" ... it's about 550 lines of code that do the job).

Here is some Direction .

Attachments:

Thank you!

Hi Shridhar:

Cool, Interesting script!. I have 2 questions.

1. How could you or I randomize the "wave pattern"  
so that the pattern would look like randomized waves, such as a water pool surface?

2. The Script does not behave well on a curved surface or cylinder, how can I improve that?

TIA!

If its random waves u r trying to do then u should use "Perlin noise" plugin or generate through Graph mapper and moving points u should use Vector not as "z" but as "surface normals" which is output from "surface divide" component.

I m guess ur talking about something like this. 

I have set move point direction to surface normals which should be applicable to Double curvature surfaces too..


Attachments:

Thanks! Shridhar, I was unaware "purlin" was a noise mapper

with respect to my 2nd question. I get some odd results wrapping your formula around a Brep instead of your original Surface. Are the UV segments somehow interfering?  

Attachments:

RSS

About

Translate

Search

Videos

  • Add Videos
  • View All

© 2024   Created by Scott Davidson.   Powered by

Badges  |  Report an Issue  |  Terms of Service