I’m not sure why Isovist refuses to work with my carefully crafted trees.
Each point to analyse is in a separate path, and that path corresponds to an obstacle curve with the exact same path.
240811_PROG_Isovist and trees.gh (1.7 MB)
You can move the slider to see that the main branches coincide perfectly.
For each main branch, I have as many sub-branches as points, and the corresponding sub-branches for the curves used as obstacles.
All I get is this error message :
I have the same problem with Isovist sometimes . It seems it cant handle big lists.
my way to go is using a lines> polar array >intersect (probably you have think of this)
tried 3 lines to test the logic on your file and it works. I am not sure my old computer can handle 360 points for each layer to try.
Cheers
Hi Rezza,
Ah, this re-assures me. I still kind of understand how data matching works.
I hope this “home-made IsoVist” doesn’t explode my execution time…
Similarly, I might also try “MeshRay” since then I can use the same mesh for all points instead of wrestling with data trees of curves.
I made some tests with a single level :
If I keep my points in separate branches, and have a single branch for the obstacle cure, it works.
But if I use the same branching as the points for the obstacle curve, IsoVist fails.
Since the data maching doesn’t care about the branch indexes, but blindly matches branches in sequence, this means I can’t process multiple levels at once.
While GH adds useless levels of branching all the time to supposedly adapt to changing input data structures (but mostly making tree management a real chore), some components can’t manage level 2 trees…Go figure.
240811_PROG_Isovist and trees_02.gh (1.7 MB)
Well, my flaming new hot rod of a computer with 64 GB RAM choked with the Curve Intersector solution !
MeshRay, on the other hand, is super quick and frugal.
I extruded each loop, made it a mesh and then joined all the meshes into one.
Thinking of it, maybe I could connect somehow all my loops into one curve.
This would make a smaller tree that IsoVist could process…
