Point order starting at 0,0,0

Replies to This Discussion

Permalink Reply by Hyungsoo Kim on October 25, 2016 at 2:05am

Hi.

I don't understand the use of C# code + WB mesh fr lines, anyway check attached.

Attachments:

• CurvedModel2positiveGaussiancurv_re.gh, 165 KB

Permalink Reply by Sebastien de Wit on October 25, 2016 at 2:53am

Hi thanks for your quick reply,

My apology, I saw I asked the question in a way which is not exactly what I meant.

Your solution partly solves the problem but the renumbering needs to be done after the form finding with Kangaroo so a loft can be created through those points.

If I use your solution the result is:

This is logic because you use the Y coordinates which are all in line prior to form finding, but not after form finding. Is there a way to tackle this problem?

Yours sincerely

Permalink Reply by Sebastien de Wit on October 25, 2016 at 2:54am

And the file

Attachments:

• CurvedModel2positiveGaussiancurv_re2.gh, 23 KB

Permalink Reply by Hyungsoo Kim on October 25, 2016 at 3:14am

Check this.

If you can generate your initial mesh properly, then you don't have to worry about the order of points after simulation.

Attachments:

• Tri&Quad_Mesh_Tessellation.gh, 19 KB

Permalink Reply by Sebastien de Wit on October 25, 2016 at 3:50am

First of all: MAJOR thanks!! the algorithm is much easier particularly in the fist part instead of the scripting. I am not used to work with scripts anyway because I am a structural engineer and not that much of a programmer (yet).

I will look to your model carefully and hope this is what I was trying to achieve last weeks.

Thanks again!!

Permalink Reply by Sebastien de Wit on October 25, 2016 at 7:19am

A (familiar) question I suppose; the geometry obtained in the end is a mesh. I am able to get a single open Brep from it, but a single continuous surface is required for panelling (lunchbox). Is there a way to do this. It is a common question but a proper answer is hard to find. Thanks in advance!

Permalink Reply by Hyungsoo Kim on October 25, 2016 at 7:59am

I've already answered on that matter in your previous post.

Attachments:

• Tri&Quad_Mesh_Tessellation_rev1.gh, 25 KB

Permalink Reply by Sebastien de Wit on October 25, 2016 at 8:21am

Sorry, thanks anyway. Case closed ;)

Permalink Reply by Sebastien de Wit on October 27, 2016 at 8:03am

Dear Hyungsoo Kim,

I was working with your model and added a few things and it works fluently with the quad mesh.

But.. if I change it to triangle A,B or C as mesh type, the part you posted above for converting a mesh to a surface ain't working any more because the mesh it is not 4 sided.

I played a while with the integers etcetera, trying to figure it out, but after 8 hours on this topic I am now stuck. What is the trick to solve such problems. If it works on triangle mesh A it should also work for B and C.

Yours sincerely

Attachments:

• Model anticlastic triangle A mesh.gh, 178 KB

Permalink Reply by Hyungsoo Kim on October 27, 2016 at 8:24am

Hey, Sebastien.

First off,"Surface fr points" only works with regular square grid, not with tri_grid from lunchbox. That's why I used quad mesh after simulation.

Second, If you want another tri-grid tessellation using lunchbox after generating one single surface, there's no difference of converting final mesh to polysurface after simulation. 

Permalink Reply by Sebastien de Wit on October 27, 2016 at 8:53am

That is good to know that Surface fr. points only works with quads.

But for example, if I use your last model (3 posts above) and do the simulation with a Tri A mesh, the SrfGrid in the end (logically) turns red.

So if I want my Hex mesh on that inflated surface, The output of the solver should be Re-Meshed with quads first?

Permalink Reply by Hyungsoo Kim on October 27, 2016 at 9:12am

Be clear about what you want to do. I don't understand your process.

I've already gave you the lunchbox hexa grid sample after generating a single surface.