fuck I capitalized the G
I'm die, thank you forever
I wasn't sure whether to do the <name of the thing where the anime drawings turn into cubes> cubes or just generic ground-pound, so it's sort of half way in between
A tip if you're using it is to animate both scale and position of the empty. And tweak the rotation/scale colour ramp curves in the nodes to fit your style as well :)
e.g. here i pushed the noise scale ramp further outwards, so the middle part is flat and the edges shoot up (bottom ramp). Then so the middle isn't too flat i added more rotation in the middle (top ramp) as well. There are lots of bits to play around with!
There's also this map-range node in the middle which i forgot to plug into the modifier inputs. This is what controls how much things are pushed downwards (arguably the most important part lmao), so definitely tweak that as well
I have no self-control
Since a couple people asked, it works with any way of chopping up a mesh, doesn't have to be cubes. I just happened to use a grid mesh with some distortion as the base, but you can use whatever.
• • •
Missing some Tweet in this thread? You can try to
force a refresh
Trying something a little different to my usual style with Lineart/Grease Pencil in #b3d. No strokes were (manually) drawn in the making of this image :^)
The shading is very minimal, it's just a bayer matrix dithering thing and some gradients slapped on stuff lol.
The inspiration is chapter 39 of shoujo shuumatsu ryokou. The one with the elevator AI, that (aside from the last few chapters) is probably the one that left the biggest impression on most people, me included.
I saw @iquilezles sdFbm article (iquilezles.org/www/articles/f…) and tried it in blender.
5 octaves of noise, 16 raymarching steps. That's all opengl can handle before hitting uniform limits with nodes.
As you can imagine, lots of nodes :D
The node editor lags to death, but viewport is nice and fast still (since unrolled loops run about the same in the end on the GPU side)
Note there's no drivers in the shader, as that *wouldn't* run anywhere near as fast!
Here's an overview of the nodes, with a little bit of annotation to show what each chunk of nodes is doing