I realize I’ve landed myself way in over my head, and it need some expert input on how to solve it.
I’m trying to create a faded hexagonal pattern that has to move across double curved geometry, so I have been trying to follow youtube tutorials and forum suggestions ( like this one: Creating a circular, hexagonal fading pattern with paneling tools curve attractor ) but I can’t wrap my head around how to set it up, so that the pattern will follow the double curved surface and use a curve attractor.
The two images above pretty much shows what I’m going for, and so far I managed to dismantle a few algorithms from images online, this is the one that comes closest:
Which is… kinda okay. It’s not perfect as I cant seem to have it focus on a curve attractor element, only on a point defined by the multi dimensional slider, and that limits how far out the edges I can pull the hex pattern.
I need it to be focused on the center of the surface with larger hexes which then fades outwards.
The tricky bit is that I need a similar result on two other double curved surfaces. And nothing I can come up with creates anything useful.
Okay, after surfing a bit more I found a tutorial that put me a bit closer…
However I need some smooth surface to be left. I don’t want hexes all over the area, so they need some distance between them as seen in the result of the huge algorithm.
I’ve been crunching some more on how to do this and thought I had it nailed after a break.
However, though I’ve now managed to put in an isocurve to the definition to attract to, it now doesn’t solve my cull functions anymore… so… back to the drawing board for version 7!
In all your files above, the geometry is missing, because you forgot to internalize it! This is essential if you upload Grasshopper files without the corresponding Rhino document.
Ahh, I’m sorry about that.
I’ve been a bit tired after trying to get my head around it.
I’ve managed to find the attractor point, however I need to convert it to a curve, so that the pattern will follow that line in intensity.
In my latest attempt i managed to setup a curve attraction by using an iso curve, however, that breaks my cull parts, and thus doesnt produce a result.
I’ll return to this on Monday, so I’ll follow up when I manage to get to the next step.
Just got back down with this to have a look at it with a pair of fresh rested eyes and well set head.
I’ve had a look and a poke at your file, however it still follows the approach of my original attempt, where I use the md slider to make an attraction point.
I’m looking for how to use a curve as an attractor. My latest attempt was to make an iso curve along the middle of the surface, which then in turn would work as the attractor.
However when I try this, all my cull actions fall apart, and the pattern doesn’t solve.
This is going to work as a bit of a progress diary it seems, so I hope others can benefit from it!
In my v.9 definition I’ve managed to get the isocurve input to work, but the result comes out wonky as seen in this screenshot. It seems that the values are not adapted correctly to the surface area, which I’ll have to work out as well.
The attraction isn’t that good either, as the hexes seems to be more or less identical across the target surface, but I hope that might change if I manage to get the the attraction curve to work and not just an isocurve.
The newest version also has a curve element with the desired attration curve internalized.
Taking your approach with the projected curve onto the brep works fine in my definition, however it still makes some wonky geometry output like the screenshot I posted earlier.
I guess it’s something in the mapping and distribution I did earlier that I might have to redo completely.
no that’s great and works wonders - eventhough I can’t get it to work on other surfaces inserted, which is strange. The “split with other brep” action fails and I can’t figure out what makes that happen.
When I use the project onto brep section from your definition in my original definition, it still messes up the out put.
I would like to have the spacing between the hexes like my original definition:
Since the initial hexagonal cells produced by LunchBox are not planar, be aware that only the top and bottom face of the “rubis” are currently planar.
A workaround for this could potentially be to planarize the LunchBox hexagons with Kangaroo first, but that’s another topic and depends on your base surface.
