Add controllable lab walking checkpoint with collision and chase camera

This commit is contained in:
UnixNight committed 2026-10-23 23:29:40 -04:00
1 parent 37fa18094f
commit 58424120bb
26 files changed
+1848 -110

No files matched your search

+15 -7
View File
@@ -30,17 +30,25 @@ avoids a Python runtime dependency in native packages. Mesh draw ranges omit nul
materials following the original renderer, while retaining their geometry. Default
normal atlas and morph transforms are supported by binary/duplicate-asset evidence.
Physics arithmetic fragments are independent of a full controller; no invented
collision/task system has been silently inserted. The main program remains a
scene investigation runtime; future simulation must be a separate stateful module
with recovered timing, input, collision, triggers and progression.
`Game/LabSimulation` supplies the no-gadget walking/jump checkpoint, using recovered
speed/yaw/impulse/airborne/camera formulas. `FixedStepClock` supplies a deliberate
60Hz port policy independent of rendering; original invocation cadence remains
unverified. SDL events feed held keyboard/mouse state; focus/minimize clear input
and pause simulation, while C pauses for inspection and R resets the start area.
`SceneCollision` sweeps/slides a vertical capsule against retained original scene
triangles, including invisible proxies, and clips the chase camera with a sphere
sweep. Solver/masks/radius/pivot are explicit approximations. The physical position
is feet; a separate original-root height feeds camera math and standing mesh minimum
Y feeds visual placement. GAM/ASE data is preserved. There is no task/trigger engine.
Application/PreviewCharacterPlacement and PreviewAnimationLibrary keep inspection
policies separate from gameplay fragments. Ground contact is a model-only offset
policies available alongside the walking checkpoint. Inspector ground contact is a model-only offset
derived once from visible surfaces and the standing bind pose, with raw source
placement preserved. Clip selection displays actual still/moving state and does
not simulate walking/jumping actions. These policies need replacement/integration
as original controller, actor transforms and state behavior are recovered.
not simulate walking/jumping actions. In walking mode, original idle/run/backpedal/
jump/fall clips follow simulation state. Looping, timing, transitions and the visual
floor-derived pivot still require original comparison and further recovery.
Subsystem implementation and validated ownership are documented in
`infrastructure.md`, `analysis/graphics.md` and `rendering.md` as they become