Grasshopper

algorithmic modeling for Rhino

I have been trying to recreate this with the 'BoxCollide' component in kangaroo: http://spacesymmetrystructure.wordpress.com/2012/12/30/orthogonal-c...

The only way I've gotten it to work is to create a new 'BoxCollide' force component for each reaction possibility. I can't help but think that Daniel's definition would be as inelegant as mine is. It would be incredibly tedious with the number of boxes he is using in the video.

I know I'm missing something entirely obvious, but any help would be appreciated.

I've attached my testing definition. 

Views: 4091

Attachments:

Replies to This Discussion

Here's an example

Attachments:

Thank you Daniel!

Hi Daniel,

This definition is quite cool!!

I am wondering using BoxCollide is it possible to achieve this:

Don't know why when I use a MD slider to move boxes from their initial position the collision doesn't work.(see attached files).

Thanks you in advance :)

Cheers,

Jack

 

Attachments:

Hi Jack,

I've attached an updated version of your definition.

To be able to move one of the boxes you need to make its corners anchorpoints.

(If points are anchors you control where they go, if not then Kangaroo controls where they go)

Also, when 2 boxes are identical height and perfectly aligned they do not collide well because the vertices of one box never actually penetrate the other, but lie exactly on its surface.

You can get around this by shifting one of the boxes a tiny amount vertically.

Hope that helps,

Daniel

Attachments:

Hi Daniel,

Thanks for your definition and specific explaination!

That is awesome!!

I have play with it and some further queries:

01.Seems when I change the box orientation the collision will be weaken...how can I maintain the collision force if all boxes having different orientation?

02. Is there anyway I can swap the anchor pts from one box to another when kangaroo running, so each box is potentially can be the one we use to collide the others? 

I gave it a shoot however it doesn't work as I expact. It will choose the initial control box and set it back to the initial position...

I undersand I can feed both short box and tall box to be anchor pts but in that case there won't be any collision and interaction between them...

Any suggestions?

Once again, thanks for your help :)

Best Regards,

Jack

Attachments:

Hi Daniel,

I've been trying to use the demo to allow the boxes to change their proportion. It doesn't seem to allow that, the geometry disappears if the "VolumePreserveStrength" and "DimensionPreserveStrength" are set very low (approx 0.0001). Also doesn't seem to allow changing that strength if kangaroo simulation is running. Is this correct?

thanks,

Bob

Hi, guys!
does antbody know, how to collide elastic spheres ike these box?
I realy need help with this.

Hi Daniel,

I am trying to combine Box Collision with Springs or a Power Law. My goal is to find a arrangement of boxes that is as close as possible but without an intersection of the boxes.

I used your Demo script and added Springs and also tried Power Law. In both cases it's not working. Each goal seems to work without to take the other goal into account.

Can you tell me where my mistake is?

Michael

Attachments:

RSS

About

Translate

Search

Photos

  • Add Photos
  • View All

© 2021   Created by Scott Davidson.   Powered by

Badges  |  Report an Issue  |  Terms of Service