Repository navigation
Commit 6f4aab1
fix(showcase): remodel mushroom-stump with a stepped felling cut, recessed peels, conforming moss and bipinnate ferns
The stump read oddly in review: the torn holding wood stood up in the
middle of a level cut as a 13 cm comb of splinters, the peeled patches
looked like planks pasted on the trunk, the moss was a flat green slab,
the ferns were the cartoon fronds of fallen-log, and the stump sat small
on a wide soil disc. It is rebuilt:
- the felling: the back cut is level; beyond the hinge chord (0.172 m
toward the fall, now to the camera's right) the undercut floor lies a
34 mm step lower. The top is bisected along both edges of the step face
so the step is its own strip of faces, and a low band of torn hinge
fibres (up to 29 mm, irregular tips, leaning toward the fall) runs along
the step's edge over the middle two thirds of the chord, its front face
running down the step to below the undercut floor. Past its ends the
step is a clean sawn face
- peels: the sapwood is the trunk's own surface 26 mm under the plate
crests, weathered grey-brown with vertical fibre streaks, fine checks
and bark-beetle galleries; the torn bark curls up to 6 mm proud in a
ragged lip showing rust-brown inner bark; 12 bark flakes lift off the
edges, roots set in the bark, undersides inner bark
- moss: four thin cushions (north flank, a small one high, two root
crotches) laid on the bark as built, furrows and all, ~10 mm thick,
thinning to a lobed rim tucked 2 mm under the bark with tongues along
the furrows, 19 satellite tufts, the rim colour thinning via Zone
- ferns: twelve bipinnate fronds (0.40-0.64 m) in two clumps either side
of the stump, the frond builder of fern-mossy-rock copied inline at
about half size (alternate pinnae, pinnules, pinna separation, sori)
- composition: soil disc 2.0x1.9 -> 1.7x1.6 m, roots, agarics, twigs,
pebbles and fern clumps pulled in; camera 3.7 -> 3.3 m; lathe 96 -> 128
sides; bark cross-cracks and end-grain glare toned down; key/wedge
relit so warmth sits inside the calibration spread
- coplanar hygiene fixed by construction, not bands: moss bases belly
into the stump by their own depth, tuft bases are cones, honey caps tip
a hashed way each, pinnae separated as in fern-mossy-rock
Budget changes:
- Base tris 34700-35800 -> 56150-57250 (measured 56710; LOD 0.5000/0.2200)
- Outer AABB -> (1.9351, 1.6283, 0.6888) m; face floors re-fit
(4441/1702/3080/1596/672/1820/4050/1758/1232/17884/384/210 measured)
- Cut level now fit on the back cut only (Notch attr): 0.0894 deg;
size 0.6210 / 0.4966 m
- New 25 felling: undercut step 0.028-0.041 (0.0346), hinge tallest fibre
0.018-0.034 (0.0286)
- New 26 peel: torn bark over sapwood 0.020-0.034 (0.0268, 0.0249), 12
flakes rooted 0.002-0.020 (0.0042-0.0150)
- New 27 moss conforms: greatest height over bark 0.006-0.016 (0.0104)
- Asset-quality remap 25 -> 28
Falsifiers (flag -> measured -> exit), all on the final code:
--skip-decimate LOD1 1.0000 9
--stray-vert loose_v 1 15
--lift-z zmin 0.05000 16
--perch-stump sealed 6/8 17
--float-caps bite -0.0156..-0.0100 18
--tilt-cut 4.9584 deg 19
--fat-stump 0.6706 m across 19
--arch-roots -0.0524 (lift 90 -> 120 mm) 20
--float-mushrooms -0.0249..-0.0032 21
--float-brackets -0.0164..-0.0131 22
--sunny-moss 0.4270 23
--float-cover 77/339 loose 24
--flat-felling step 0.0006 25
--tall-hinge tip 0.0358 (top +7.2 mm, tol 10) 25
--flush-peel -0.0015, -0.0011 26
--float-flakes -0.0108..0.0011 26
--slab-moss 0.0360 27
Every falsifier keeps 56710 tris and the default AABB except --tall-hinge
(tallest fibre sets the top; +7.2 mm inside BBOX_TOL). The lip's own stand
over the bark was tried as a sub-budget and dropped: its median read the
flare and lobes, not the lip (flush and default measured alike).
Evidence (live runs, local):
- .scratch/5.2/blender-5.2.1-windows-x64/blender.exe (5.2.1 LTS): two
default runs identical; all 17 falsifiers exit their codes;
tests/check_falsifier_targets.py --run --only mushroom-stump passes.
- .scratch/blender-4.5.11-windows-x64/blender.exe (4.5.11 LTS): same
sweep, same codes, identical measured lines (glTF +16 bytes).
- .scratch/blender-5.1.2-windows-x64/blender.exe (5.1.2): default exit 0,
identical measured lines.
- Framing: fill 0.709 x 0.833; margins l .122 r .169 b .039 t .128.
- Contact sheet: luma 0.2391 (band 0.2369-0.2428), stage 0.2534,
warmth +0.0994 (spread -0.144..+0.105; was +0.1456), sat 0.1565.
- Asset sheet: the stump now carries a readable felling step, recessed
peels and bipinnate ferns; not the least-designed object in the lineup.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>1 parent ff67ff1 commit 6f4aab1
11 files changed
Lines changed: 2728 additions & 544 deletions
File tree
- docs/gallery
- asset-sheets
- assets
- contact-sheets
- mushroom-stump
- showcase
- mushroom-stump
| Original file line number | Diff line number | Diff line change | |
|---|---|---|---|
| |||
170 | 170 | | |
171 | 171 | | |
172 | 172 | | |
173 | | - | |
| 173 | + | |
174 | 174 | | |
175 | 175 | | |
176 | 176 | | |
| |||
296 | 296 | | |
297 | 297 | | |
298 | 298 | | |
299 | | - | |
| 299 | + | |
300 | 300 | | |
301 | 301 | | |
302 | 302 | | |
| |||
0 commit comments