hello
i was trying to close the holes in this pattern to end with a single surface. is there any way i could select the interessed edges within a list to create a surface with a fragmented patch or a boundary surface?
Masharabiya.gh (126.4 KB)
Masharabiya.gh (126.4 KB)
I reuse your group of 6 lines. I used weave considering that lines were ordered but merge could be surely enough. Then I made 3 triangles, but you could also make a surface using 4Points surface
Masharabiya.gh (129.0 KB)
5 Likes
brilliant. if i can ask: what does the letter ‘Q’ mean in the panel?
thank you again sir.
I don’t know. If you look at mesh construct by default a mesh face is named like that.
Assuming ‘Q’ stands for quad, perhaps that ‘Q’ should be ‘T’?
1 Like
Here’s an alternate solution lofting naked edges that were selected using the Tree Item component. An interesting DataTree exercise.
240204a_Masharabiya.gh (131.1 KB)
-Kevin
3 Likes
Wow! that seems pretty articulated. I can’t even understand half of your logic 
Maybe one day I will be as good as you Kevin!
Yes it is that, but as David made a lot of Auto cast it works with other letters, I think it doesn’t use them. It works also with others brackets
Without Lunchbox I tried TriRemesh and am delightfully surprised that your solution seems to still work!
But looking very closely, it appears to me that the Srf4Pt surfaces you added to fill in the holes are coplanar with the star shapes I have colored? If that’s the case, then why create the star shapes at all? Or is that a question for @s138617 ?
Masharabiya_2024Feb4a.gh (69.6 KB)
P.S. To simplify it further… This should work with Lunchbox triangles too, eh?
Masharabiya_2024Feb4b.gh (67.2 KB)
This is interesting. Four slider controls produce a very wide range of patterns.
-
‘Length’ determines the size and number of faces produced by TriRemesh.
-
‘Height1’ and ‘Height2’ range from -100 to +100 and set the height of center points.
-
‘CntrPt’ (0 to 1) determines the “center” point position for the second phase. Area ‘C’ is used for the first phase where triangles have three similar edge lengths but that’s not the case in the second phase, so choosing the ‘CntrPt’ produces more interesting results.
Masharabiya_2024Feb4c.gh (93.4 KB)
P.S. Slider values are appropriate for this particular initial shape but must be adjusted, of course, for shapes of different sizes.
3 Likes
man! i didnt know you can achieve this level of complexity and randomness with only few nodes! thank you for sharing!
to answer your question
If that’s the case, then why create the star shapes at all? Or is that a question for @s138617 ?
long answer: I am trying to preserve the parametric side of the pattern based on an image mapper and I lack a lot of base concepts so I decided to deconstruct a definition i found on a book.
short answer: I am a newbie…
do you guys think it is possible to combine this pattern with a parametric surface that can be 3d printed? i mean: can I offset the surface inside based on the diameter of the nozzle of my 3d printer? Or is it more practical to simply bake it and offset inside rhino?
Definitely possible, @Birk_Binnard makes a lot of 3d printed vases similar to this using grasshopper. General process is to scale in XY axes to get inside surface, construct bottom, and close top. Search the forum, there are many examples.
-Kevin
1 Like
I did those things and found that this surface pattern is so complex it causes problems. 3099 faces including the top and two bottoms. No point really in having the pattern on the inside, it could just as well be smooth - but that’s not easy either since the mesh doesn’t have circular cross-sections.
Masharabiya_2024Feb4d.gh (116.2 KB)
2 Likes
jesus! It is 5 hours since i was trying. thanks for this amazing help. i will study this def.
it could just as well be smooth - but that’s not easy either since the mesh doesn’t have circular cross-sections.
can you link some reference to make me understand what circular cross-sections are in relation to build a smooth surface offset in the inside? (I just discovered there is a whole academic work on offset for planar meshes?!! - EXACT FACE-OFFSETTING FOR POLYGONAL MESHES
If this is the kind of design you’re looking for I can post the GH that makes it.
sample.gh (1.2 MB)
I’ve spent a lot of time on this too. That’s how we learn 
This has a smooth interior surface with a fillet (rounded) bottom seam. That’s where I spent most of my time… 
I created ten evenly spaced planes to intersect the patterned surface, then used the average distance from the center of the polyline vertices to create circles, which I lofted to get the inner surface. I used Scale NU to shrink the diameter and height (independently), though it might work better to use offset surface instead?
Masharabiya_2024Feb4e.gh (116.2 KB)
There are so many things one can do without really considering how practical it will be to 3D print stuff like this. I haven’t changed the surface pattern since I posted it ten hours ago, now in the gray group below. What I’ve been messing with is thickening the surface in various ways.
This effort uses offset surface instead of scaling, which I believe does a better job of matching smooth inside to patterned outside. At first I offset the “derived surface” (yellow group, now unused) but replaced that with the original surface instead, Offset surface introduced two things to deal with:
-
Bottom thickness is now accomplished by setting a ‘Bottom Thick’ slider value (teal group), which moves a plane that intersects the offset surface, creating a curve used to split it.
-
The top opening of the offset surface is slightly below the original plane, which required lofting and joining a very short cylinder so the flat top between inner and outer surface works properly.
Other things changed as a result. The pink group does the FilEdge radius between inside surface and bottom. The white group is the end point where the pieces are joined to form a “Closed Brep” solid.
This image shows the inner surface in red and the outer surface in transparent white to give an idea of thickness. In fact, though, some areas might be extremely thin (too thin?) due to the surface pattern.
Masharabiya_2024Feb4f.gh (120.5 KB)
