Grasshopper

algorithmic modeling for Rhino

Hi guys,

Anyone knows what is the proper approach to modeling this (attached)?

Any help is highly appreciated.

Thanks

Views: 7128

Attachments:

Replies to This Discussion

Moved from FAQ into Discussion.

You'll have to write down the algorithm you want to encode first. What variables do you need? What data is already available, what data needs to be computed? And so on and so forth.

Can I insert a 3D model, and have voxel pixelate it form me?

You could create an array of cubes and start shaping the form your after, by deleting cubes, till your happy with the resulting structure, this approach does not require GH.

In practical terms you really want to start with a geometric structure and have that structure replaced by cubes. (voxels)

The attached pic shows how I voxelized a portion of a Fractal, the resulting 3d mesh can be 3d printed, no GH required..

What you would like to do In Grasshopper is think, is have a script which accepts any geometric structure and have GH voxelize it?

Attachments:

Thank you guys for your replies.

I am designing an Architectural space that has a defined form which is not very regular. It is triangulated and has a lot of voids from within, it also has an auditorium and more space that are shaped from cubes. That is why even after I create the shape, I need to be able to pixelize it or turn it into array of equal cubes 50cm*50cm*50cm. I tried a rectangular grid with attraction point, I tried meshes but I still doesnt give me full control.

I have a C# that does this ("3d pixalize" brep/mesh skins so to speak). If someone else doesn't provide a solution using standard components I could post it here (but it's a bit complex to handle especially if you are not familiar with the language). There's a nasty GH/R bug as well around yielding some "3d pixels" omissions.

Thank you for writing back.

What I need is a regular array of cubes 50*50*50 cm. they cant overlap or be different in size, because sometimes they form architectural elements such as stairs, ceilings..etc

Well this C# thing captured it accepts any "3d pixel" x/y/z sizes - thus the "resolution" is fully under your command (see the blue Planes indicating the x/y/z user controlled divisions). For big numbers of "3d pixels" it's quite slow mind (tested up to 100K "pixels"). Note: As I said, the Split mesh R method  yields some omissions (depending on the topology, most notably with concave ones). The C# works in 3 passes > first z > then y > then x (inn order to control the millions of "3d pixels" that may occur).

Anyway notify if you want it - but I would strongly recommend to wait a bit for some solution using standard GH components: this C#  IS NOT easy to handle by any means.

BTW: the y/x/z division coordinate system is also user controllable meaning:

This is the massing of the intended model.

Attachments:

Hi ,

I am looking for logic/method to create multiple cubes inside a big box with sizes of 3,6,9,12 in relation to the attr points( with no overlaps and voids).pls let me know how i can acheive that.

This is trail reply to my previous query.

i tried a manual culling of the points at each lvl of boxing size and use the remaining as input.But its looks so complex,redundant  and sometime too intutive . please let me know how can i do more efficient

Attachments:

RSS

About

Translate

Search

Photos

  • Add Photos
  • View All

Videos

  • Add Videos
  • View All

© 2024   Created by Scott Davidson.   Powered by

Badges  |  Report an Issue  |  Terms of Service