OverviewFilesCommitsIssuesDocsReleasesPeers

dc-terrain / commits

ref:
docs: split the README into docs/, and verify every claim against the bindings
a7d16f5 by plankatron · 2026-09-10
README: six meshers, and the one claim that was wrong about them
a16f152 by plankatron · 2026-09-10
docs: a README, and the drift the audit found writing it
9373143 by plankatron · 2026-09-10
sdf: the sign was already known, so stop computing it twice
6b678f0 by plankatron · 2026-09-08
bvh: an LBVH, and the measurement that says it was never the problem
7207aae by plankatron · 2026-09-08
bvh: parallel build, and two hypotheses that did not pay
839766c by plankatron · 2026-09-08
gate: assert the box's SHAPE, not one .blend's numbers
4b63073 by plankatron · 2026-09-08
generator: a proc-nodes graph, meshed at the size Blender says it is
b868717 by plankatron · 2026-09-08
qef/polish: the Newton step was missing a factor of voxel_size
01fdec3 by plankatron · 2026-09-08
geiss: keep the noise volumes as HALF, not float32
41486bd by plankatron · 2026-09-07
geiss: reachable from GDScript, and a fallback that cannot be mistaken for terrain
399713d by plankatron · 2026-09-07
geiss: the density as a dc-terrain generator, and a path for its volumes
adb27c6 by plankatron · 2026-09-07
gpu density: the SAME sampler bindings on the density_fill side
3ddf576 by plankatron · 2026-09-07
gpu pipeline: allow a generator to sample TEXTURES, not just buffers
a4408a7 by plankatron · 2026-09-07
dc-terrain: mirror the erosion mask on the CPU, and name the gap that is left
14fcd80 by plankatron · 2026-09-07
dc-terrain: a volumetric seabed, and the depth mask that gates it
ffac461 by plankatron · 2026-09-07
dc-terrain: measure the surf zone before building an ocean on it
589ebb7 by plankatron · 2026-09-07
dc-terrain: tell the material how coarse the mesh under it is
2c29b2e by plankatron · 2026-09-07
dc-terrain: wire the hypso curve, mask detail by slope, export the slope
bd097c5 by plankatron · 2026-09-07
dc: throttle the bake's GPU-wait warning
a85a915 by plankatron · 2026-09-06
dc: bake a ProcVolume to a mesh on the GPU, through the pipeline that already exists
f1cb2d7 by plankatron · 2026-09-06
dc: the scalar path is a map-load cost, not a per-frame one -- measured
0eb2676 by plankatron · 2026-09-06
dc: a compile failure was silent, and the built-in generator looks like terrain
99a8659 by plankatron · 2026-09-06
dc: the density wrapper resolved its graph per VOXEL, and nothing could see it
c9eb176 by plankatron · 2026-09-06
dc: spawn workers AFTER the GPU, and stop re-sorting a backlog nobody dispatches
da0e556 by plankatron · 2026-09-04
dc: instrument the GPU path, because "is the GPU running" had no answer
931d9f1 by plankatron · 2026-09-03
procnodes: take the heightfield GEOMETRY from the grid, or nothing samples it
dd47dc1 by plankatron · 2026-09-03
gpu: stop uploading an atlas nothing samples; fix two gate assumptions
707264b by plankatron · 2026-09-03
procnodes: read the tile grid, and collapse FOUR CPU samplers into one
60b8bc3 by plankatron · 2026-09-03
gpu: a content VERSION, because a stable pointer cannot express mutation
c3fdfe5 by plankatron · 2026-09-03
gate: prove a tile can arrive at runtime WITHOUT recompiling the shader
b2bbb4a by plankatron · 2026-09-03
gate: the tile grid's GPU addressing, against real tiles on a real device
4737292 by plankatron · 2026-09-03
Revert "atlas: emit HBLK_ATLAS_W + HBLK_USE_ATLAS so binding 14 can actually be read"
029cca8 by plankatron · 2026-09-03
Revert "atlas: announce the transport from the point of emission, once"
94bd5f8 by plankatron · 2026-09-03
atlas: announce the transport from the point of emission, once
047a215 by plankatron · 2026-09-03
atlas: emit HBLK_ATLAS_W + HBLK_USE_ATLAS so binding 14 can actually be read
f8103c4 by plankatron · 2026-09-03
atlas: free the textures in cleanup, and CONFIRM the path runs
2db9172 by plankatron · 2026-09-02
atlas: carry the heightfield as a texture too, binding 14 — RUNTIME UNVERIFIED
396da97 by plankatron · 2026-09-02
mesh: carry per-vertex CURVATURE, from the density grid to the material
6136d08 by plankatron · 2026-09-02
procnodes: the emitted sampler named a buffer that no longer exists
144d932 by plankatron · 2026-09-02
erosion: field_half_range property, wired to the wrapper and the CPU twin
c5d4001 by plankatron · 2026-09-02
erosion: pass erosion_scale to the CPU twin, so collision matches the shader
fdb47b5 by plankatron · 2026-09-02
erosion-minimal: a wrapper with nothing in it but the erosion
66f48f4 by plankatron · 2026-09-02
erosion: a real on/off toggle, from the terrain panel
ca60568 by plankatron · 2026-09-02
terrain: upload the whole mip pyramid, and stop emitting a shader that cannot compile
332df1c by plankatron · 2026-09-02
procnodes: the graph decomposition, parked for review
457b89c by plankatron · 2026-09-02
tests: emit the wrapper from the shared header, and name the one place the map's constants live
d1d252e by plankatron · 2026-08-31
tests: mirror map_dc_erosion.gd centring its band on zero
156b2e4 by plankatron · 2026-08-31
edits: one contract for both templates — the mesh stage applies them too
9a4cb84 by plankatron · 2026-08-31
procnodes: stop discarding lod_voxel_size — clamp the wrapper's octaves
4dc8188 by plankatron · 2026-08-31