Basically what I want to make is a triangle mesh with fixed length, for example 48, 51, 54, 57 and 60 (or another set of length). This triangle mesh has to be fitted on any random given surface as close as possible. What I use now is LengthSnap (thanks to Daniel Piker). This snaps every triangle edge to the closest of these lengths. For example 49,3 snaps to 48. I started with a simple double curved surface. In the GH-file you can see the deviation in every meshpoint in relation to the surface. The highest deviation is 11.33, pretty high. So i was wondering if it is possible to minimize these deviations, maybe by using galapagos?
There’s no need to make a new forum topic each time you have a follow on to the original question. It makes it easier for others to follow if it is all in one place. (see earlier posts on this here and here).
You can add another goal pulling the points to the original surface like this: Discrete_triangle_mesh_deviation2.gh (30.8 KB)
This will be fighting against the edge length snapping, so you’ll need to slowly reduce the strength of the highlighted slider down to zero to guide it into place while still meeting the length constraints.
This brings the deviation down to 9.3
It might not be possible to get it much lower than this with the constraints you have, which are quite restrictive.
I managed to get it more down, just by rearranging some things. But for some reason the number is always different: 5.59, 5.55, 5.75, 5.63, 5.85,… Maybe I’m doing something wrong?
By changing the measurements of the surface a little bit you can find a mesh that has a lower deviation from the surface. However it takes a lot of time to look for the surface with the best results. Is it possible to use Galapagos for this? I tried it, but it isn’t working fully.
