commits · proc-nodes · pear-git-v3

docs: the README claimed 59 nodes, its own catalog listed 62, there are 122

4d65e35 by plankatron · 2026-09-10

scene: adopt the MeshOutput a published scene already carries

fcc56da by plankatron · 2026-09-09

mesh: normals from FACES, and turn the cage round on the way IN

c1edeac by plankatron · 2026-09-08

mesh: the two engines wind the other way, and a drawn surface needs normals

e051e5a by plankatron · 2026-09-08

mesh: take a MESH from the bridge, and subdivide FACES

6118a06 by plankatron · 2026-09-08

fields: do not materialise an output nobody reads

dfb575f by plankatron · 2026-09-08

exec: run what the output can reach -- 8.3 s -> 4.7 s, past Blender

7948510 by plankatron · 2026-09-08

fields: thread the noise nodes -- 74 s -> 8.3 s end to end

82f9985 by plankatron · 2026-09-08

sdf: sign the band from the flood, not from pseudonormals -- 9.6 s -> 3.6 s

9aa63da by plankatron · 2026-09-08

sdf+fields: a worklist dilation, a threaded Voronoi, and where the time really is

467e7ec by plankatron · 2026-09-08

sdf: compute the band, not the volume -- 53.7 s -> 12.4 s, same mesh

29ec6a5 by plankatron · 2026-09-08

graph: a node remembers the id the authoring tool gave it

b47f2a7 by plankatron · 2026-09-08

glsl: one graph, two products -- emit the density's cone, and rotate in it

f8461c1 by plankatron · 2026-09-08

density: the graph declares its own extent, beside its density

07e6964 by plankatron · 2026-09-08

density: a graph may declare its density and have no geometry at all

b66f3e9 by plankatron · 2026-09-08

glsl: the emitted density was not the graph's field -- four ways

c627c51 by plankatron · 2026-09-08

gate: the emitted DENSITY, against the CPU, on a real device

d2f27e8 by plankatron · 2026-09-08

gate: the moments accumulator was noisier than the difference it measured

3fd11ea by plankatron · 2026-09-08

proc-nodes: 3D simplex in the emitted library, and erosion becomes subaerial

efaea46 by plankatron · 2026-09-07

proc-nodes: a hypsometric curve, and the slope the filter already had

4ce4179 by plankatron · 2026-09-07

gate: the shape gpu_batch_gate could not test -- a producer outside the batch

903973d by plankatron · 2026-09-06

gpu: a batch that reads a producer outside itself must not run

ee4c16a by plankatron · 2026-09-06

demo: cage_graph_probe -- a real Blender cage through a real bridged graph

c9de802 by plankatron · 2026-09-06

gpu: in() must check emission too, and a failed shader now prints itself

e9c167b by plankatron · 2026-09-06

gpu: emitted means WRITTEN; mesh: a non-indexed surface is still triangles

7351194 by plankatron · 2026-09-06

gpu: the batch fuses the value cone, so a field reaches the GPU at all

37c71e1 by plankatron · 2026-09-06

gpu: a connected push-constant socket was one value for every vertex

00fb4ae by plankatron · 2026-09-06

volume: the SDF-grid nodes are Blender's, and the fill is threaded

240ed09 by plankatron · 2026-09-06

volume: Grid to Mesh never hung, and the field going in was inside out

57c76da by plankatron · 2026-09-06

volume: the SDF-grid family, and Grid to Mesh hangs

5484d5f by plankatron · 2026-09-06

fog: a height gradient and noise, and Position could not reach the GPU at all

4994e81 by plankatron · 2026-09-06

libdc: the extractor is shared, and reaching for it found the CPU volume nodes are stubs

a8cad1c by plankatron · 2026-09-06

fog: a volume graph renders as a FogVolume, and compiling one found two bugs

ac43b85 by plankatron · 2026-09-06

volume: a volume graph can compile now -- ProcVolume carries real GLSL

aac88dc by plankatron · 2026-09-06

volume: "no consumers" was wrong -- the importer factory is a consumer

7feabba by plankatron · 2026-09-06

volume: the flag has a reason, half the reason was stale, and the rule was written twice

b322740 by plankatron · 2026-09-06

gate: the erosion filter joins the parity gate, and the f-suffix hazard is measured

bccc60b by plankatron · 2026-09-06

gpu: pc.paramN meant two different things, and three nodes used both

92f5cc8 by plankatron · 2026-09-06

simplex: the node was computing a different function on CPU than on GPU -- fixed

e87e2f1 by plankatron · 2026-09-06

noise: the hand-written GLSL library is gone -- one implementation, everywhere

4501578 by plankatron · 2026-09-06

glsl: Wave, and Mix's colour and rotation -- the list is closed

dcaf7d0 by plankatron · 2026-09-06

voronoi: the fourth .inc, and the node -- four of five

cdee735 by plankatron · 2026-09-06

glsl: Noise Texture comes back on -- three of five, and the second Perlin can go

d556349 by plankatron · 2026-09-06

perlin: the third .inc, and a gate that stops taking the machine down with it

9717c7f by plankatron · 2026-09-06

glsl: Random Value, all four data types, and the node that started the audit

0c40cad by plankatron · 2026-09-05

glsl: White Noise comes back on, and it is the hash and nothing else

0346f6c by plankatron · 2026-09-05

hash: the layer becomes ONE text, compiled as both GLSL and C++

3dd5989 by plankatron · 2026-09-05

gate: the hash layer, compared exactly, and the second Jenkins measured at last

0540c94 by plankatron · 2026-09-05

cpu twin: a real density evaluator, and the scalar floor measured before anything is deleted

832dc9c by plankatron · 2026-09-05

docs: the CPU twin design, and the measurement that makes it possible

33a35c9 by plankatron · 2026-09-05

proc-nodes / commits