Connect Geometry to Point Matrice

Replies to This Discussion

Permalink Reply by Joseph Oster on June 6, 2017 at 10:56am

Not from screen shots alone.  Please post code?

'Orient' or 'Orient Direction' are probably your best bet, so the next question is how is that "moon sliver" surface to be oriented at each point of the spiral?

Permalink Reply by Jérémy Chaillou on June 6, 2017 at 11:17am

Thanks a lot 

ahah really nice question ! Didn't think about it yet. Is there a componant like magnet ? 

I put the GH file in this answer 

Attachments:

• CCLOGO.gh, 15 KB

Permalink Reply by Joseph Oster on June 6, 2017 at 12:11pm

You didn't internalize the points used to create the crescent moon shape so I faked them.  Maybe not the same as yours?  You can change that.  VERY IMPORTANT: You need to understand its relationship to the 'pA' and 'dA' inputs of 'Orient'.

The "magnet" in this case is a 'Vec2Pt (Vector Two Point)' from the center point to each point on the spiral (or from each point to the center).  The 'U (Unitize)' input is set to "True" because if not, it will scale the oriented shape (try it).  And try using the X and Z vectors instead of Y on the 'dA (Direction A)' input.

I reversed the order of your cull pattern so that the first point (the origin) would be culled; otherwise, there is a zero length vector error.

For more examples, type "orient spiral" in the sidebar "Search Grasshopper" feature.

Attachments:

• CCLOGO_2017Jun6a.gh, 17 KB

Permalink Reply by Jérémy Chaillou on June 13, 2017 at 10:43am

Hi Joseph, 

Many Thanks for your help and your time, I really appreciate. 

I tried and play with the model you gave me, it's nice but it's not exactly what i am looking for. I imagine that I am not doing the right thing to get it. 

My goal is to replicate a pine cone viewed from the top. So the base is a phylotaxis pattern but i have to create C shape object that look to the center (Thanks to your GH file), and now to create a growing/scale factor that make those C shape evolving. 

Here is the insparation Photo : 

Here is the closest result i get with the GH model : 

Do you have any idea how we can improve the shape of model ? On this bake, I did it with the zero length vector error. 

Once again thanks a lot for your time and your help. 

Best regards, 

Jérémy 

Permalink Reply by Joseph Oster on June 13, 2017 at 11:39am

I haven't thought this through, but here are some of the first ideas that sprang to mind...  The two 'Graph Mapper' components are optional, intended to modify the 'Series N (Step)' value and a scaling factor I added, in non-linear ways.

I'd be inclined to model a 3D pine cone instead of a 2D top view, to really capture what's going on, but I'm not that ambitious today.

By the way, my implementation of the crescent moon is NOT oriented symmetrically with the 'World XY' plane, which skews the result a bit - I suggest you fix that.  Oh, and if you look at the real pine cone, you see that the size of its "seeds"(?) vary from top to bottom (or outside to center), so it might help to scale these crescent moons in a similar way.

Attachments:

• CCLOGO_2017Jun13a.gh, 20 KB

Permalink Reply by Joseph Oster on June 13, 2017 at 12:22pm

Here is an easy (too easy?) linear way to get a 3D model from the 2D results.  It makes it easier to see that I didn't get the pine cone effect I'd hoped for, in side view.

P.S.  Here is a guy trying for a pine code in 3D, though I think his code was overly complex?

http://www.grasshopper3d.com/forum/topics/pine-cone-structure

P.P.S.  Here's a page that describes a 3D method worth looking at?

http://www.instructables.com/id/Pine-Cone-Lamp/

Attachments:

• CCLOGO_2017Jun13b.gh, 15 KB

Permalink Reply by Joseph Oster on June 13, 2017 at 12:39pm

One more - lovely!

http://www.studioseed.net/blog/software-blog/parametric-generative-...

Permalink Reply by Jérémy Chaillou on June 14, 2017 at 11:57am

Hi Joseph, 

Once again, many thanks for your help. 

It's a really good idea to add graph editor. Powerfull tool. 

I decomposed the geometry of the pine cone viewed from the top. 

Here is a schema of it : 

To achieve the gh model, I think we need to find a way to edit the scale (in X and in Y)  of the croissant moon (ruled surface on the model) with a graph editor. Do you have any idea how is it possible ? In the final render, croissant moon must not touch each other. 

Best regards, 

Jérémy Chaillou 

Permalink Reply by Joseph Oster on June 14, 2017 at 1:16pm

'Graph Mapper' isn't the only way to achieve adaptive parameters.  Here is a re-write of the "croissant moon" code using two 3-point arcs; it adapts to the distance from center of spiral...

I want to make something very clear though; this code is a hodge podge of different techniques and ideas that, collectively, make a poor GH model as far as I'm concerned.  Too many mysterious tweaks with unpredictable results.  I just don't have the time and attention to come up with a better model right now.  So take all this with a few grains of salt; don't be afraid to throw it away and create a better algorithm.

Attachments:

• CCLOGO_2017Jun14a.gh, 28 KB