Why didn't you instance the boolean?
I also found it really odd that you didn't model quads? It really wouldn't have been any more difficult to model and would have meant you wouldn't have jumped through a load of hoops at the end trying to fixing the tri shading issues. If you'd started with a loop around the cut out you could have easily resolved to quads AND taught better topology in the process.
Glad there's a lattice modifier. Looks like Join is the Blender equivalent of Stitch in C4D
OK so I've seen a few questions from you thehomme,
It looks like you are a competent C4D user,
I do suggest though, you put aside your knowledge of C4D while learning Blender.
There are many differences, approaches, techniques, and I'm positive you will learn Blender more efficiently and enjoyably.
Hmmm - you're right. I don't know if it was naive but I kind of assumed it would simply be a matter of learning a different UI. I put a lot of time into understanding topology and C4D but it was just getting too much to pay for when RedShift was added. Checking out Blender is high cost in terms of time but low cost in terms of financial commitment. I'm just starting so I guess these are my growing pains. Will try and keep a lid on it.
I still don't know why the cut out wasn't instanced... surely that would have saved a load of time?
Why not instance the Boolean?
I think that's a totally valid question! In this case I think it would have taken more time. The Array modifier doesn't do curved arrays, so we'd have to go through setting up a curve and spacing it properly, or we'd need to use Geometry Nodes, both of which are slower than just placing it by hand. We could have instanced the object itself and put those instances into a collection and used that collection as the Boolean source, but we apply the modifier right away so that also just ends up being extra steps.
Why not use quads?
The triangles aren't actually the source of the shading issue in this case - it's the change in poly density around a smooth curve. With quads you'll still get pinching around the poles and anywhere where two loops are close together. I made the same shape with just quads before recording since I was also curious about that, and it ends up with roughly the same shading issues but takes more time and adds unnecessary complexity.
Edit: though Martin's quad solution in the other thread is quite clean and I would say is better overall, though the shading issue still stands so the end result is the same.
Like Adrian mentioned, there are always a variety of ways to accomplish the same thing, and each approach has inherent pros and cons, so as you're learning don't hesitate to try both and go with whatever makes the most sense to you, though do try to look at the process with fresh eyes.
Thanks all. I haven't seen all these replies till now. I will keep plugging away. The learning/adjustment curve is much steeper than I expected. *much*
