Differential Growth in Curves

Hi all,

I m trying to do differential growing curves and couldn't come up with any ideas to get started.Any strategies will be helpful .

Thank u

Replies to This Discussion

Permalink Reply by Shridhar Mamidalaa on June 24, 2015 at 2:31am

I tried to place small circles on random points and do a Recursive 2d Hull.

Permalink Reply by Gia Djahaia on June 25, 2015 at 12:53am

Thanks, Daniel! It's really a lot of fun :)

Permalink Reply by Nick Tyrer on July 7, 2015 at 8:44am

Turns out Mediated Matter at MIT were way ahead of everybody on this idea. Action starts at 1:20

MUSHTARI from Mediated Matter Group on Vimeo.

Permalink Reply by David Stasiuk on July 7, 2015 at 9:05am

The guys that did the coding are at Deskriptiv, and they do amazing stuff!

Permalink Reply by Nick Tyrer on July 8, 2015 at 3:44am

Yes i love their work, its where i saw the video. Though i am really curious to the design process/workflow. For not being a part of the Mediated Matter team, they seem to have been totally pivotal on the last few projects.

But then again the whole unit is so savvy with its public image its hard to tell. Everything is always perfectly photographed/filmed/documented, always consistent style and curation. I would love to see behind the scenes, pre-polish. I think i've just realised i want to be part of M.M...

Permalink Reply by Kim hauer on July 7, 2015 at 11:02am

Very NICE! MIT also has a significantly larger budget. I think the female torso is a distraction. The women that would wear this chastity belt contraption hasn't been born yet :)

Permalink Reply by Vicente Soler on July 10, 2015 at 12:44pm

I got this differential growth thing working for curves and meshes as a C# scripting component. It's similar to Andy Lomas, Deskritiv, Nervous, etc. but by far not as nice (yet?).

I guess the main difference to the Kangaroo version posted here is that it dynamically adds springs/faces (splits the springs into two when they reach a certain length) and the radius of the collision force remains constant. I'll try to fix a few things and post the definition.

Permalink Reply by Thomas Helzle on July 10, 2015 at 12:47pm

AWESOME!

Permalink Reply by Shridhar Mamidalaa on July 10, 2015 at 12:52pm

Nicely done Vincente ....

Permalink Reply by David Stasiuk on July 10, 2015 at 1:00pm

Looks great!

Permalink Reply by Kim hauer on July 10, 2015 at 6:16pm

Yikes... I can wait, ......this is awesome!

Permalink Reply by Nik Willmore on July 11, 2015 at 4:43am

Is there any way to perturb it to equilibrate out the historical artifacts of so many parallel lines, by making parallel or just straight lines unhappy? I guess starting with a symmetrical seed curve would make such artifacts into an interesting feature that gives a uniform result that really does "naturally" conform to the boundaries manually. Were this 3D printed as a single filament the strength may be less where there are so many thin grooves along the same local direction, and it looks weird too. Mere randomness might help in nicely ruining the order.