but these fields defined in python can not be used in “IsoMesh2” or the other nodes you provide in Isopod?
Ah I misunderstood your question, didn’t know this was what you intend to do. For that to work, the python class would have to inherit from the class Field defined in Isopod.dll (c#), not sure if that’s feasible.
this thread might help:
@Jakob_v_Schirmeister Here’s a basic example of a custom Isopod field defined in Python:
python_isopod_field_test.gh (9.9 KB)
import Isopod
class SphereSDF(Isopod.Field):
def __init__(self, center, radius):
self.center = center
self.radius = radius
def ValueAt(self, P):
return center.DistanceTo(P) - radius
a = SphereSDF(center, radius)
It works, though performance seems orders of magnitude worse than doing the same in C#
(I’m not really a Python user myself, so might be missing some better way of doing this)
2 Likes
Have you tried updating to a more current release of Rhino8? Does the problem still persist if you do?
This is one of my fav threads.
I’m starting some new projects soon and I will be attempting to learn to use electrolysis to turn plastic prints into metal parts.
Reason is cause the 3D metal printing technology is still lagging behind like 10 years.
So, I’ll be really liking to try some fancy geometry with grasshopper that might make this process more awsomer 
I compiled some pics I like to stare at here in this thread.
I’m hoping to do some electrolysis where the molecules will soak into the print as much as possible to create some nice strong metal parts from silly plastic prints 
So porosity I think is good, but I will also like to control it so that there’s well defined smooth surfaces where necessary as well – potentially the porosity would maybe be best in the internal sense rather than external.
Hence, the outer most layers I think I will attempt mostly smooth. Somehow internal porosity of course would need prep for electrolysis in order for most awesomer molecular surface energy actions. 
3 Likes
I still need to study implicits I think
I wanted to share this link to this hand geometry cause looks cool:
I’m liking to learn the art of electro-forming. And I think I would really like to try some of these amazing geometries I’m seeing grasshopper do 
I’m thinking I can get some mind blowing results if I can get this to work via electro-forming.
I’m planning on using some special conductive PLA, and I’m imagining that some particular geometries would work well with the molecular bonding. 
I’m tryna think of surface textures that would bond well but still form smooth electrolysis actions…
If this works, might not have to buy $100k metal printer lol. Plus metal printers have that silly shrinkage problem haha. 
“Electroforming rate is surface-area-dependent.” I definitely need to have good sense of this parameter somehow haha.
looks like nurbs being extracted from a mesh
I’m putting in some time to wrap my brain around some of this information, and I’m seeing this like for the first time and wait what this looks like reverse engineering, what’s happening rn 
isn’t that… the opposite? I don’t follow you.
2 Likes
I just did something similar where I extended a ribbon and then used a 3D convex hull.
I"m mostly just saying this looks similar to what you would see from a reverse engineering workflow where you’d see nurrbs surfaces being conformed to a mesh.
I don’t see why that’s hard to follow.
Maybe I’m missing something.
At any rate, this is cool for me to see cause just make my brain happy to see GH with things like this, even if I don’t understand it. 
2 Likes
@DanielPiker I’d still benefit from an improved IsoMesh1 component.
For this torso with a lot of hidden details inside, Isomesh2 takes 37 minutes to compute and Isomesh1 just 10 minutes. The Isomesh1 result shows an additional mesh.
The initial mesh is oriented so the bounding box volume is as small as possible. For Isomesh1 the object is moved into positive xyz quadrant.
2 Likes
Hi Daniel, Thanks for developing this powerful toolset! I really enjoy using it.
One thing that holds back the productive use is the long meshing time for thin walled parts. Your IsoMesh1 component using the Octree option would already be a good solution to get rid of large chunks of unfilled space in the given bounding box.
Unfortunately, the Octree algorithm often disregrads octree levels where there should be geometry created.
I attached a minimal example of this behavior. Does the algorithm maybe only check sign changes on the corner points?
IsoMesh1_Holes.gh (11.8 KB)
1 Like
@DanielPiker I am looking for ways to model thickness graded and cell size graded tpms lattices
if possible, also for strut and planar based ones also.
Eagerly looking forward to your responses
And then, are there any components which converts the brep body into a SDF implicit body within isopod?
The screenshot doesn’t show the relevant part… Can you upload the Grasshopper definition?
2 Likes
Hi sir, sorry for the delay.
The component worked after restarting grasshopper.
Thanks
1 Like
Are there any ways for peforming the above mentioned modulation in TPMS uidng isopod?
@martinsiegrist @DanielPiker
Use Mesh Brep for this