vr.vehicle_view "first_person" (the default) has the director treat every
seat as first person, which hides the player's body, and puts the eyes at
the player's head in the seat (a seat's camera marker is where the chase
camera sits). The view turns with the vehicle's yaw only, the horizon
kept level; a passenger seat facing aside keeps its turn from the
vehicle. The player's own gun is put away in driver and gunner seats.
vr.vehicle_steering "stick" (the default) leaves a driver's facing to the
game, so the right stick steers (and pitches the Banshee) as on a flat
screen while the head only looks; "head" and "hand" steer where you look
or point. Gunners aim with the head relative to the vehicle, so a turret
follows the head without turning the view.
vr.diag_drive_seconds seats the player as the nearest vehicle's driver,
for unattended checks (after cheat_all_vehicles). The headset went
offline before the first vehicle test ran; the build is warning-clean.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Recover transparent geometry submit (+1,285 meaningful code, zero exact losses); reconcile COMMON ownership and Q10 85+EXT1 credit, actions/main data, and separately disclosed denominator corrections. Preserve canonical-only repairs and all held research. C09 is ordinary disclosed cleanup, not a claim of proven scaffolding intent. Clean rebuild, 1,663 tests, parks, audits, provider links and independent EXT1 check pass. Local reconciliation only.
Import only independently reviewed production changes. Preserve canonical physics and virtual-keyboard repairs; keep B3 and all unresolved holds out.
Record exactly four approved canary debits (6987 meaningful bytes), six Halo admissions and two revocations. Keep vendor ioinit/storage work separate. Full build, 8252-row sweep, 1318 tests, park/admission/fake audits and 621-object binding pass within approved debit bounds.
Do not publish donor research dumps or private assets. README and unrelated user research remain untouched. See frontier_publication_reconciliation_20260927.md for complete evidence limits and accounting.
Integrate verified source and data progress; exclude unproven COMMON ownership, held UB/asm forms, and address-named data credit. Stable sweep: 23 exact functions gained, zero lost.
- _hs_compile_and_evaluate: calls the real static hs_compile_source_error
helper with no source file instead of a hand-written copy (January
inlines the helper here, leaving its dead loop's alignment pad), reuses
the expression parameter; formerly recorded as unreachable
- _inverse_kinematics_adjust_matrices: named squared upper-arm length and
consistent basis-row locals (park reopen)
- _actor_stimulus_prop_acknowledged: MAX clamps, DATUM_INDEX_TO_ABSOLUTE_INDEX,
January's ternary polarity (park reopen)
- _lrar_allocate: found-flag loop allocating inside the fitting arm, as the
sibling _lra_allocate (park reopen)
- _prop_add: reads the dead field back and binds the actor lookup
- _director_script_camera: real director_set_camera helper call and the
admitted bare director_get validation (park reopen)
Stable diff: 0 regressions; active parks 213 -> 209.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Wave w2 landing, adversarially verified. park reopens: _director_update_controls. Full ninja build and per-function stable_verdicts sweep: 0 regressions. 1 park(s) removed.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
