Grasshopper

algorithmic modeling for Rhino

Views: 883

Albums: planarfab

Comment

You need to be a member of Grasshopper to add comments!

Comment by iottx on May 22, 2014 at 1:45am

;) good questions, dear AB! sometimes grasshopper forces u to cheat. i hope u r not going to be disappointed. as for the right size of the initial quads: for me it was enough to use a numeric slider. once all neighbors overlap, its fine. otherwise just think, they can not be too big, as they will be trimmed, so maybe say: quad size is given by the host surface BBX diagonal length. for the trimming: as long as u have no concave parts on the surface, u can trim each cell with all others. try to work with planes instead of the planar cells. that saves performance. use plane-plane intersection, then u get edge lines, which then u just need to chamfer together.
u can find neighbors for each cell by i.e. (the cheapest option) neighbors are the 10 closest cells to a test cell. remember: u just need to limit the number of intersecting planes. this works for rather evenly distributed cells with little concave curvatures. i found useful a kind of searchlight strategy. think of a rotating cone. from all points currently in the cone, the closest is designated as neighbor. i know this is totally unscientific, but i hope i could help!
ps. lets see this http://www.grasshopper3d.com/photo/faceted-shell

Comment by AB on May 19, 2014 at 8:17am

iottx,

I know you put this together years ago and it may seem like ancient history to you, but I am wondering if you could describe your approach to sizing the planar surface prior to trimming.  I guess when I try to recreate this to learn from it... i am getting stuck with how to trim large numbers of planar surfaces with each other.

Are you finding the nearest neighbors and trimming a group at a time?  Is there any recursion being employed? I know you mentioned in another post that the solution was part manual, part automatic (or something to that effect) 

Any advice is really appreciated. Thanks

Comment by iottx on May 16, 2014 at 9:26am

hi. the recipe is the following: take a free-form  surface, distribute a bunch of planar surfaces over the UV domain of the host surface. the planar surfaces, tangential planes, like tiles, are then trimmed against each other. thus they form irregular polygons. its a simple 2d array over UV domain. compare it to slicing an apple with straight cuts or polishing a diamond. hope it helps, good luck and all the best! ps. i still have to try voronoi.. ;)

Comment by hem lui on May 7, 2014 at 11:47am

hi i am curious , do you get the volume out of 2d voronoi ?

Comment by iottx on April 28, 2010 at 4:38am
hi Casper, thank you so much for your compliment! looking forward to see your work in this forum!! ;)
Comment by Casper Le Fèvre on April 28, 2010 at 4:32am
Hi Titusz, what a beautiful structure, my compliments.

About

Translate

Search

Photos

  • Add Photos
  • View All

Videos

  • Add Videos
  • View All

© 2024   Created by Scott Davidson.   Powered by

Badges  |  Report an Issue  |  Terms of Service