Twisted Box Selection

Hi,

I want to select twisted boxes located on black regions. Can you help me, please?

Thanks in advance


twisted box.3dm (268.3 KB)
twisted box.gh (41.9 KB)

Several plugins - can you internalize your objects in question?
Best,
RC

I think you can see it now

twisted box.3dm (419.5 KB)
twisted box.gh (27.8 KB)

1 Like

One {first} quick way, because you’ve already identified the mesh faces on black regions, is to move one stack of twisted boxes to the locations (centroids) of said faces:
twisted box.gh (32.7 KB)

Thank you so much

Another {second} quick way is to find all stacked twisted boxes that ‘coincide’ with the mesh face centroids:
twisted box2.gh (33.0 KB)

Cheers!

These are so beneficial for my thesis. Thank you, again

@René_Corella Hello Rene,
I want to ask you another question similar to this example. I want to apply the same method to distribute my lattices based on surface color. I tried to use the same definition on my model, but an error happened. My lattices don’t touch and connect to each other. They are distributed improperly. How can I fix it? I ask for your help if possible.

Thanks in advance,

anticlastic lattice shell.gh (161.6 KB)

Hi, I don’t have millipede installed - could you internalize the objects and maybe show a screenshot of what you’re after, that always helps :slight_smile:

Sure! I believe you can see it now.
internalise anticlastic lattice shell.gh (184.2 KB)

1 Like

cool, thanks - because you already had culled mesh faces from color, I thought you could rely on that mesh to intersect it with MeshRay:

internalise anticlastic lattice shell.gh (190.4 KB)


*Note:
Reuploaded the file to get rid of the group parmeter you see in the image - not needed

*Note 2:
I used Cull Pattern so stuff not on black color went away - I guess you can use Dispatch instead.

Cheers

1 Like

Thank you for your reply. I have a problem with lattice cell distribution. Cells are located in the proper position but they don’t accurately connect to each other. Do you have an idea how to solve it?


rc_internalise anticlastic lattice shell (1).gh (126.2 KB)

If I understand correctly, you wish to morph your module lines to the ‘voxels’ so the lattice can (1) follow the shape and (2) make these lines actually connect for the purpose of using Multi-Pipe?

Like the image below?

If yes, I used box morph and then deleted the duplicate lines.
Result is one closed SubD.
Check here:
rc_internalise anticlastic lattice shell (1).gh (116.4 KB)

1 Like

It looks nice! Thank you so much.

1 Like

Dear Mr. Rene,

I know I take up a lot of your time. Sorry for that. However, I need to ask one last question. I encountered a research paper that had a similar intent to my work. According to the paper, they have divided the surface analyzed by Millipede into 5 zones by color. As you know, we can separate my model into two zones black and white. Can we further divide the colored surface into two more zones? That is, can we separate our voxels into white, gray, dark gray, and black zones? I need your help last time.

Thanks in advance,



for_mr.Rene.gh (215.6 KB)

1 Like

Hi Alim,

I think you’re now trying to cull mesh faces based on color zones, which means you can probably use Divide Domain for the matter.

  • Divide the initial color domain into 2 or more zones.
  • Test for inclusion of your color values within these zones.
  • Use Includes (Inc) to get a cull pattern of this test.
  • Cull Mesh faces with this pattern to ‘split’ the mesh into the color zones.

Something like this:

for_mr.Rene.gh (217.7 KB)

Thank you so much, Mr. Rene. According to your definition, I think there is a small mistake. Some of the mesh zones of index 0 and 2 are overlapping each other. The gray areas are selected by both indexes. How can I select these gray areas separately?



.

1 Like

Hey Alim, apologies - couldn’t come see this earlier.

You’re correct! I definitely misinterpreted your previous question about the mesh zones and then gave you a potato instead :slight_smile:

(I read ‘zones’ in your doodle and threw mesh faces into those buckets without considering the indices of the mesh faces).

Anyway, I think (if I’m getting it right this time haha) you might want to rely on something more like “color steps” instead. For this you’d have to find domains more than testing for value inclusion.

Check the attached revision, there’s a ‘playful’ way and a ‘strict’ way:
for_mr.Rene.gh (256.5 KB)


1 Like

Thank you very much, Mr. Rene. You helped me a lot. You are the best.

1 Like