d14cf7fa6e
qa.sh --strict: 18 passed · 0 failed · 0 warn · 0 skipped. selfcheck 157,647/157,647, fingerprint 0x5f76e76 unmoved. No tag — John's playtest session rules the epoch. THE FIVE ASKS. [C] wardrobe wired via exported WARDROBE_BASE + stockBaseFor(shop), preloaded inside the existing STOCK_REAL gate: ON under ?stock=real (52 pack items, 69 garment ids on rendered meshes, 1 atlas, exactly 2 requests) and CHEAPER (59->55 draws, 111->100 textures); OFF is what ships by default, A/B'd against a reverted control tree — default arm identical on every counter (160 URLs same hash, 131/18,018 street, 59/28,505 interior). [D] three gates wired (r41_shots.py writes to TMPDIR so gate runs never rewrite committed shots). [E] clips_verify wired — 46 clips / 6 groups / 3,498,124 B. [B's bookmark finding] CONFIRMED AND WIDER: street_noon, shopfront_detail, crossroads_busy AND market_square all give 3 distinct hashes across 3 boots; only night_neon is stable. RULING: no gate pins a screenshot byte-hash on any bookmark — everything R41 asserts is a counter. Filed to B: one await document.fonts.ready in buildings.js. [C's drawSweep] CONFIRMED INDEPENDENTLY with F's own sweep (never calls C's): GLB-on worst 123 @ dept/auto. The phantom control (one stale room left in scene) reads 197 — +74 on every reading — decomposing R39/R40's '188' as 116 real + ~72 phantom. NEW GATES, controls demonstrated. r41_denylist.mjs (ruling 3): 738 files / 197.3 MiB scanned by path AND bytes, 7 banned names, 0 hits; the control plants a banned manifest row every run -> RED on 2 names -> removed -> GREEN. r41_integration.py: wardrobe both arms · interior <=350 on both instruments · ?noassets=1 zero across six fetch classes over a 3-shop walk (control fetches 5/6) · ?clips=0 142 URLs and ?classic=1 125 URLs with zero R41 cargo. BUDGETS. Street: noon 282 · NIGHT 291/120,093 — margin 9, INTACT · classic 269 byte-exact · ?clips=0 NIGHT 291, delta 0. No lane spent a street draw. Interior: true pre-R41 116, R41 123 @ dept/auto, margin 227; over 33 real shops worst 110; pub 49 quiet / 77 gig night. FILED BACK: flags_check's _enter_record_shop went RED on 8 smokes because §41.4's bins put every one of the first six record-shop counters inside 2.6 m — widened to the whole open set, verified not a game defect. smoke_djdance read _actions[0] and D's per-instance clip swapping made the mixer multi-action, so a healthy dancer was called a stall — now follows the clip by name. D's r41_shots.py browse arm is deterministically red here (re-enters the winner after a ~3 s/candidate scan, by which time the occupant has left) — wired WARN-level with the reason at the call site, fix handed to D. F's own r39_transmission.py had an unmeasured 30 s goto default that flaked once under a full strict run; pinned to 90 s. 11 proof frames in docs/shots/laneF_r41/ with sidecars: street_postures (5 rigs, 3 clips playing) against its ?clips=0 control (1 clip) · street_lean · browse_interior · pub_furnished + the honest kit before/after pair · record_dj_booth · opshop_wardrobe on/off · credits_panel with ODbL on screen. Every humanoid frame carries its R10 human-sized line. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> |
||
|---|---|---|
| .. | ||
| before | ||
| laneA | ||
| laneB | ||
| laneB_r4 | ||
| laneB_r37_kerb | ||
| laneB_r38 | ||
| laneB_r39 | ||
| laneB_r40 | ||
| laneB_r41 | ||
| laneC | ||
| laneD | ||
| laneE | ||
| laneF | ||
| laneF_r39 | ||
| laneF_r40 | ||
| laneF_r41 | ||
| references | ||
| release_v3 | ||
| release_v3_1 | ||
| release_v4 | ||
| release_v5 | ||
| v1_tour | ||
| v2_tour | ||
| v4_alpha | ||
| v4_beta | ||
| v7_alpha | ||
| v7_beta | ||
| v7_tour | ||
| contact.html | ||
| README.md | ||
| v2_contact.html | ||
docs/shots — visual regression + devlog reference tree
Owned by Lane F (each lane also drops its own acceptance screenshots here).
docs/shots/
laneA/ … laneE/ … per-lane acceptance screenshots (each lane owns its dir)
v1_tour/ … F: the v1 beauty shots (tools/shots.py) — frozen devlog set
v2_tour/ … F: the v2 tour (tools/v2_tour.py) — the CURRENT canonical set (see below)
before/ … F: last known-good capture, for side-by-side regression
references/ … real-world reference photos named per shot
v2_contact.html … generated contact sheet for the v2 tour
v2 tour — stable filenames (do not rename; the v3 site links these)
cd web && python3 -m http.server 8130 then tools/.venv/bin/python tools/v2_tour.py
→ 1280×720 PNGs in v2_tour/ + docs/shots/v2_contact.html.
| file | what it documents |
|---|---|
street_noon.png |
main strip at midday |
crossroads_busy.png |
busiest intersection, peds about |
shopfront_detail.png |
verandah + sign + window up close |
patronage_door.png |
streamed peds at an open shop's door (Lane D patronage) |
rain_verandah.png |
rain outside a lit ?winmap video-shop window |
rain_street.png |
wet strip, rain down the street |
tram_stop.png |
the v2 tram paused at a shelter (?tram=1) |
night_crowd.png |
VIDEO REGAL glowing amid CLOSED neighbours at night — the §3.5 hours story (clean-night boot). NB: no visible street crowd — see the note below. |
browsers.png |
an occupied shop's interior life — a browser at the rack + the keeper at the counter, both human-sized (R10 giant-rig fix). figures human-sized vs door/fittings: yes. |
dig_real.png |
riffling a real-cover record bin (?dig=1&stock=real) |
katoomba_main.png |
plan-agnostic main strip (?plansrc=osm&town=katoomba) |
Note on night_crowd (Lane F, R10)
Re-framed to the tighter, readable "one lit shop in a closed town" composition. It does not show a
standing street crowd, and that is not a framing bug: Lane D's night patronage parks ducked-in patrons
hidden inside the shop (they become the interior browsers seen in browsers.png), and seed 20261990's
video shop sits on a low-traffic corner, so none gather visibly outside (tested: 0 patronage / 0 visible
at night). The town's night life is the interior presence + the neon-vs-CLOSED mood. A visible outdoor
night queue would need a new Lane-D mechanic (loiter-outside-the-openLate-shop) — parked for v3. The
~20 fps in night captures is a headless-chromium software-GL artifact (the scene is light, ~128
draws / 65k tris); on real hardware the night scene is not a perf concern.
Regression workflow
Before a risky change, snapshot the good state into before/, then re-run v2_tour.py and eyeball
v2_contact.html. Process rule (R10): any shot containing a humanoid gets an explicit
"figures human-sized relative to doors/fittings: yes" line in the progress note — actually look.