Is this method of constructing complex shapes practical? Seems like it offers very little freedom depending on what 3d primitives are available
graphic_content
I think the number of shapes that we can represent through this is very limited.
nouyang
I wonder if the final shape is being saved as a tree of boolean operations in real software? Then it will take a huge amount of space and time to render.
keenan
@rlpo It's actually quite common for, e.g., designing mechanical parts. See the Wikipedia article on solid modeling.
keenan
@nouyang Actually, a tree like this will take way less than a typical polygon mesh. Even a super complex model might use only a few thousand primitives (spheres, cylinders, etc.), and a tree has size linear in the number of nodes. Compare with polygon meshes and point clouds, which regularly use millions of elements.
Is this method of constructing complex shapes practical? Seems like it offers very little freedom depending on what 3d primitives are available
I think the number of shapes that we can represent through this is very limited.
I wonder if the final shape is being saved as a tree of boolean operations in real software? Then it will take a huge amount of space and time to render.
@rlpo It's actually quite common for, e.g., designing mechanical parts. See the Wikipedia article on solid modeling.
@nouyang Actually, a tree like this will take way less than a typical polygon mesh. Even a super complex model might use only a few thousand primitives (spheres, cylinders, etc.), and a tree has size linear in the number of nodes. Compare with polygon meshes and point clouds, which regularly use millions of elements.