Grasshopper

generative modeling for Rhino

select entire voronoi cell without cropping it

Hey Para-Campers,
I am back ready to finish my exercise. I started all over again cause I lost the thread! It is still the same idea I am working on. Here a little rough overview what I want to achieve.

With some support of DeDackelZucht and Danny Boyes, some forum members, I got the basic idea done. While enhancing the definition, I don't know how to select the entire voronoi cells inside my given geometry. All I get are croped cells cause of the boundary. Is it possible that the cell wouldn't be cropped but all the ones after the cells disappear?

Furthermore, how can I select the cells that are NOT affected by the attractor to perform different operations on them?!

And finally, how can I get a surface between the cells?

Sorry for all those questions but I am stucked. Maybe another way is easier!? I am open for new ideas and hopefully a lot of input! :)

cheers, Sebastian

Views: 534

Attachments:

Replies to This Discussion

Hi Sebastian, let's go!

1.- Easy cake, you just use the inside check to cull out cells whose original point is not inside the region. This is actually not a boolean operation, but just a list management operation:

2.- Just create another true/false pattern with whatever was affected:

3.- The surface will be tricky. The way you are trying is not bad, although so heavy operations are sometimes buggy. What do you want it for? Can't you do it another way?

Attachments:

The idea of 3. was using this surface partly as a pathway trough the landside. I think I can also do it in another way! At the moment everything is flat but the final result should be on a shaped surface. I think I will try to project the curves on the shaped surface an use the "surface split" component. But until this I have to define my regions where I want to have what, like economy center, hotels, parking area, ... see the following pic!

I hope I can post some more results in the evening!

Hello :)

Ok, so I worked a lot today. I defined another region where I want to put some buildings. I used the same process for choosing this specific area (inside outside check). It worked proper so far. For the buildings I applied a surface on the cells and I choosed to offset the boundarys of the single cells two times and so on, ... I've seen this before by Zaha Hadid (Kartal Kendik Masterplan Turkey 2006 CLICK ME) and later on in the forum by Nathan Miller and RWNB I rebuilded it and added a lot of stuff to get round corners and so on! here is a screenshot of where I got with this:

1.

2. Ok combining this definition with the other one works, ... but not always, that's my problem! :( Sometimes I need to shift the points so that the cells work properly. Is this "normal" or is there a way to optimise this progress? see next picture:

3. Furthermore I thought it might depend on the short segments of the voronoi cells, so I tried to implement a step that rebuilds the cells. Honestly, I am not sure if it really helped or slowed down the whole thing. You can find this approach in the attached definition.

4. Applying height to those buildings. I used the extrude component. But I would prefer, making a second surface over the whole area and apply something similar like we did in the camp with the jews memorial in Berlin.

Well that's it so far. I don't know how to handle the problems in point 2. 3.

Point 4. I guess I will do this in the end when everything else on the landside is done, ...

I appreciate any help,

have a nice evening,

Sebastian

Attachments:

1.- Awesome, looks quite good.

2.- The problem comes when some voronois, when doing offset, the offset doesn't close properly in the corners:

Either the component doesnt' work with too small bevel edges, or is buggy in these situations. This is probably somethign you are going to have to live with.

3.- I've also tried, but doesn't seem to work properly either...

4.- Find attached an example for that:

Attachments:

Ah cool, this is a nice approach for the 4th point! I worked on something similar today, I red a lot about that in the forum and I found another approach on that, I will post it. But I think yours is less heavy! :) You can apply the def on your previous posted 3dm file.

2. and 3. Yeah it is a pitty it doesn't work. I thought rebuilding the cells would solve the problem, ... well I will go on maybe I will find something.

Thanks for your help! Have a nice evening!

Sebastian

P.S. Tomorrow Evening, I will leave for some days. I have to move some parts of my appartement here in Paris. My Erasmus year is over^^  I will be back next week. Of course I try to go on with my work!

Attachments:

A couple notes:

1.- Beware that, although far fmore optimal to compute, the arithmetic average of a set of points IS NOT the same as the centroid... Doesn't really matter here, but just to puntualize.

2.- The project to mesh alternative is quite interesting, can see you are doing your homework! ;)

Hey,

Thanks for your notes! I have one more problem, I don't know how to solve it. I attached the def and a example picture.

Thanks for any hints and help! :)

Attachments:

Lucky I am around! Can you post the Rhino file as well? If too big, zip it (not rar please)

Here ya go! Thanks!

Attachments:

Just some booleaning... Just project the voronoi on top of the surface and split it with the curves:

Attachments:

Ah, damn it seemed to be an easy step, sry man! ok thanks again for your help! it worked perfectly! :)

Now, I will concetrate more on the "breaking ice part" Hope to post more results in the next days!

Wooow, looking cool!

• View All