Can someone explain to me how to make these surface lattice patterns?

[Jimbo_Jombo]

I’ve attempted a couple of methods but its never quite right. See in the reference image we have this continuous flowing curvature along the organic shape. I have tried slicing it and extracting curves on two different angles, but it doesn’t give this same kind of flowing form.

Any help would be huge! Thanks in advance.

Attached file:

Outsole Test Sample.3dm (927.2 KB)

[Japhy]

There are plugins, this is a new offering as well.

About the ShapeMap category

This is a group for the ShapeMap plugin: https://www.food4rhino.com/en/app/shapemap [image] About ShapeMap ShapeMap is a plugin for Grasshopper. It integrates multiple basic functions in Rhino and has undergone secondary development. This plugin allows users to map any curve texture designed on a plane onto the surfaces of complex 3D models, while minimizing or eliminating deformation of the texture curves on the 3D surfaces. With ShapeMap, designing textures on 3D model surfaces becomes much…

[jessesn (Jessesn Chen)]

@Jimbo_Jombo

I’ve attempted a couple of methods but its never quite right. See in the reference image we have this continuous flowing curvature along the organic shape. I have tried slicing it and extracting curves on two different angles, but it doesn’t give this same kind of flowing form.

Any help would be huge! Thanks in advance.

As @Japhy reminded, you can give it a try with ShapeMap and for your reference.

How to design this shoe?

Please give it a try with ShapeMap, SM4ShoeDesign.gh (19.7 KB) [SM4ShoeDesign]

[René_Corella]

@Jimbo_Jombo

but it doesn’t give this same kind of flowing form.

Not sure if you must copy that design or simply using as reference—just in case here’s a simple ‘contour’ alternative to keep you from trying to imitate; define your vector(s) and generate your own lines:
shape_contours.gh (12.8 KB)

*Edit:
Also, it seems you started with something related in your other thread?

I’m surprised nobody warned you about multiple threads for the same project. Anyway, for diagonal lattices (not the same as flow lines) also see these potential approaches, second link shows another nice example by @jessesn:

Uniformly orienting mesh faces by vertices

You were still missing the curves you’re morphing. But I think you got it Rather than quad-remeshing your SubD, I think using MeshFromSubD might be convenient to stay closer to the original smooth form? Just pay attention to the ‘D’ level so it’s not heavily subdivided prior to box-morphing the lines you’ll turn into multi-pipe. Alternatively, you could keep the mesh pretty coarse but make your line pattern more dense before you box-morph to find the right balance. Very easy to …

Difficulties in modeling shoe models

Hello everyone, how should we model this shoe model? It has the same mesh division inside and outside, which is very clever but also very difficult. Can you help me come up with ideas on how to model it. [05] [04] [03] [02]

[Jimbo_Jombo]

Sorry didn’t mean to do the multiple threads. I thought this first initial thread was to close to “please figure this out for me”.

I wanted to ask more directly and show my thinking and attempts instead of the broad “how to do this “ question I asked. But I will keep this in mind for next time and not post multiple times!

@jessesn @Japhy @René_Corella
Really appreciate all the help so far, will post back with any progress I’ve made on this!

[Jimbo_Jombo]

So I have been trying all afternoon unsuccessfully to do this, this is the closest I’ve gotten with the diagonalmesh tool.

But I can’t seem to separate the two line areas no matter what I try. I’ve followed a bunch of tutorials but they seem to fall apart at a certain point.

Tried to follow this tutorial but it’s not behaving in the same way in terms of those corners and creating those diagonal lines: https://www.youtube.com/watch?v=2SjR9jOxIrc&t=173s

Goal is I want the blue lines below to be different from the green lines. Different thickness or offset or height.

Thanks in advance!

Diagonalmesh Test.gh (22.3 KB)

Outsole Test Sample.3dm (1.1 MB)

[laurent_delrieu]

After many attempts with Grok, I remember it was in Kangaroo

[Tom_P]

what the WarpWeft component (never heard of it or used it) seams not to support is some kind of “distribution” fence / stop / boarder (black line / connection).
currently the flipping / mirrowing takes place at the cyan curves where it s not wanted.

it looks like it s quite straight forward to code..:

UPDATE:
… well gave it a try…
just using a meshbox with welded vertices - which - in its simple form - has the nescessary, proper face-Setup as @René_Corella also mentions below…

WrapWeft

as far as i see it will output “random seams”

…you can not influence where flipping takes place if necessary

custom script

with a custom script

i introduced the sketched “fences” from above. a recursive algorithm (c# script) will not pass over the black edges - so this is where “flipping” will happen.
… just a first run / proof of concept - i did not look at all special cases, n-gons, triangular faces, unwelded edges, … vertices with more then 4 faces…and so on…

demo

some simple ballon-like mesh…

good night sleep tight - cheers -tom

[laurent_delrieu]

@Tom_P

it looks like it s quite straight forward to code

I am really not sure of this, the topology here doesn’t help. I tried using Grok, I tested 5 strategies, none was as good as the Kangaroo tool. Perhaps it could be better to make some group and let the user decide.

I would like to see what you can output.

[René_Corella]

@Jimbo_Jombo

Goal is I want the blue lines below to be different from the green lines. Different thickness or offset or height.

I’m nowhere near the coding skills from Laurent or Tom. Nonetheless, common sense tells me you must distinguish that one goal is to identify and separate the lines, another is to have something like what’s on your sketch, provided it’s a literal indication.

If I’m allowed to interpret it,

Your desire is a “true” continuous diagonal grid across the entire mesh—this won’t be automatic as you’d expect through QuadRemesh + Diagonalize. I suspect that even if you provide guide curves (diagonal contours) your quad mesh edgeflow won’t be ideal. You might still have to go back to a method where you specify and generate the required ‘isolines’.

[Tom_P]

@laurent_delrieu

I would like to see what you can output.

see my updated post above…

[Tom_Newsom]

What will help a lot is lining up the 3-way vertices in the source mesh. For example, at the toe they’re offset by one:

Which means it’s hard to tidy up the seam in the diagonal grid:

Your inspiration images have nice neat seams at heel and toe.

You could then actually hard-disconnect the “patches” of the mesh to force the diagonlization to choose how the continuity flows (there are basically only two options - the inspiration image does the second one)

[laurent_delrieu]

@Tom_Newsom

lining up the 3-way vertices in the source mesh

Exact, here is the 14 (or more) zones, spheres represent the 3-way vertices. So this lead to many variations for the “Warp Weft”.

Indeed 15 zones