generative modeling for Rhino
I have two 3d shapes (generated in grasshopper) that connect to corners of a triangle and square and want to tile it in the pattern below.
However, in the third image, you can see that they overlap when I try to connect them.
Is there a way to create a bounding "box" for these to avoid overlaps?
Would really appreciate if someone could give a few suggestions or have a look at my files attached!
Intersections between the shapes
The surfaces will never perfectly touch each other, but you can avoid the overlapping. The main problem is the triangle geometry. There is no clean way to make a morphbox from that geometry. At least i wasn't able to get one. But look at the files i've added and you'll understand why.
Thanks for that. The ones in the square boundary seem to work. And yes, the triangles do struggle to fit in. Do you think there is a way to make a bounding "box" shaped of an extruded triangle?
Could it be done in Rhino without cage editing it? 'Cause I'd still like to unroll these surfaces.
very nice solution danny...ra, there is no better solution for placing the objects. To unroll the surfaces you can use this (if you're using rhino 5).
The geometry still unrolls, but this looks good. Unfortunately, the labs I'm in are still using rhino 4 sr8. Hopefully we'll get the update soon.
Thanks a lot for that! Anyway, the going's good, I think. Just trying to get back into grasshopper again. It's been awhile! What have you been up to these days?
Some silly questions:
- In regards to the planes, when trying to find the 'closest point on curve' on the curve from the polygon center, how does it manage to pick a side when they should all be equally spaced? (Does that make sense?)
-Didn't quite understand the radius group. Why values 1/6 and 1/4 and sine?
If all things are equal does it matter which side? The only purpose is to align the X axis of a plane with one of the vertices I don't think it matters which one.
r = D/Sin(angle)
Where the angle PI/4 = 45 Degrees and PI/6 radians = 30 Degrees
Although I have just realised that the 4 pointed object is not bound with in a square so this method might not be completely correct.
It is a square, no worries
Thanks, I was just curious as to how it chose a point although equal distances.
I'm now trying to add an attractor to the whole definition. This is gonna be fun!
does the attractor distort the target curves? because this method wouldn't work if that was the case.
I don't want them to distrort the target curves. Probably just control how much the geometries open up. Yea, I can't seem to get it to work
Hmm... so I managed to simplify the variables to one slider. If I distributed the two geometries across a grid made up of coordinates, do you think that might work?