Match original collision rounding through populated dynamic reclip

This commit is contained in:
UnixNight committed 2026-11-07 08:26:17 -05:00
1 parent 135ff3642a
commit bf876e9c3c
22 files changed
+701 -14

No files matched your search

+6
View File
@@ -143,3 +143,9 @@ Empty-grid tests close that finite ownership contract, not nonempty shared-world
behavior, real actor responses or broad live admission. The source kernel itself
has since been recovered in collision-response/body reports; the older unresolved
wording above describes this report's original implementation checkpoint.
The [populated follow-up](../checkpoint/dynamic-reclip.md) adds14 exact full
observations and12 kernel-step matches, with source-store corrections in shared
edge/ray arithmetic. This closes the populated synthetic reclip contract without
replacing the live injected adapter. Real callbacks, live getter timing and
retained physical/helper order remain the admission boundary.
+7
View File
@@ -5,6 +5,13 @@ SDL3 for platform windowing, input and audio. Linux x64 and Windows x64 are
equally required. Source is `src/GamePort/`; root executable csproj allows
`dotnet build` and `dotnet run` from GamePort. No runtime network dependencies.
Shared collision kernels retain recovered float-store boundaries in edge
projection and ray normalization/hit publication. `StaticCollisionContext` keeps
nearest/last polygon, passes, pre-geometry node visits and actual primitive tests
separate. Dynamic pair/reclip composition remains a test fixture until real
actor callbacks and retained physical order are recovered. See
checkpoint/dynamic-reclip.md; no generic live box solver is implied.
`Scooter` is a separate registered C3DJEEP in VR05/LEV1, owning body, wheels,
steering/impact/jump clocks and camera state. Selection summons at the companion;
strict contact transfers raw control. PlayerSceneBinding runs body, player/input,
+7
View File
@@ -1,5 +1,12 @@
# Core architecture coverage map
2026-11-07 populated follow-up: [dynamic reclip](dynamic-reclip.md) adds14 full
original/C# comparisons and12 kernel-step matches (S/P/M). Real synthetic static
walls/ceiling now exercise shared nearest, retries,370/374 order and primitive
mode. Kernel arithmetic corrections follow captured divergences and original
float-store boundaries; no numerical tolerance was added. Actual actor callbacks,
post-callback flag/radius ownership and retained helper order still gate live pairs.
2026-11-07: [dynamic/shared context](dynamic-context.md) adds17 complete original/
C# test-composition cases and12 captured kernel argument matches (S/P/M).
Hit/last-polygon retention, old static callback and final local-root publication
+5
View File
@@ -137,6 +137,11 @@ cleanup, child deletion, sweep execution or final retained registry order.
## Remaining admission gates and next chunk
Follow-up: [populated reclip](dynamic-reclip.md) now executes real synthetic static
walls/ceiling and matches14 full observations/12 kernel steps after directly
evidenced rounding corrections. The first gate below is closed for that bounded
composition; real actors, ordered-body provenance and reentrancy remain separate.
1. Exercise static geometry after a dynamic hit, rather than only empty-grid
reclip. Compare shared nearest/last ownership across two real primitive
candidates and ordered bodies, including face/edge ties and flag transitions.
+153
View File
@@ -0,0 +1,153 @@
# Populated dynamic reclip and recovered rounding boundaries
2026-11-07. Continues [dynamic context](dynamic-context.md). The complete original
box query now runs against a populated synthetic static grid. Fourteen full
observations and12 diagnostic kernel steps match the C# composition exactly as
parsed float32 values. This checkpoint corrects directly evidenced shared math;
it does not admit dynamic pairs or new actors to live gameplay.
## Evidence and runtime boundary
Canonical EXE SHA-256:
`f5281076e1a2492f2d61f87aae3f4544b6d0f535ebe33f9fbb6f78710fe1aea8`.
The generator reuses the prior11 guarded bodies/102 relocations, including
47A5E0,47B4B0,47BBB0,47BD80,469880 and469B70. Body hashes, PE offsets and copied
OMT2.dll identity are retained in `artifacts/dynamic-reclip/native/source.json`.
REA evidence is reused, not regenerated for an unchanged question:
| Evidence | Source |
|---|---|
|`ev_0d64cc58fa8f5573fc8e61e0f2c20e36288099212bd2257d3a3121361fd6c904`|Complete original instruction listing; decisive float stores/call order.|
|`ev_76827fca086933b7c48a9036a5a4a0a0f86d714935de6373eca4a3530fd231c5`|47A5E0 mode2 pseudocode, retained under dynamic-context/rea.|
|`ev_f39128dafc2a3a20b5768613ce22b3c3168fe8653908837d92ad9a69ef5f2ad8`|47B4B0 static traversal/retry; see collision-grid-traversal.md.|
|`ev_5e39f2668d04b1fdd2c91b298a28413327dc3879e6b8bbffe0a14bc1dcbc1128`|4695B0 nearest-edge pseudocode; assembly resolves intermediate stores.|
`tools/binary/dynamic_reclip.py` generates a bounded harness containing copies of
the existing probe sources without their mains. The new executable calls copied
source bodies with synthetic box/sphere getters, point transforms, callbacks and
grid nodes. It never invokes the original entry point. An instrumented private
mapped47BD80 counts primitive mode and captures12 steps only for the skew case;
the hook restores original entry bytes for each real kernel invocation. Original
files and on-disk copied EXE remain unpatched. This is Wine32 component execution,
not full original gameplay, live actor callbacks or native Windows verification.
Final execution is under `artifacts/dynamic-reclip/probe-verified/`, with no
network or Original-directory mount. An earlier overlapping prefix preparation
produced no observations; a separate completed prefix then exceeded the12-second
startup bound while still running. The45-second bound completed successfully.
Incomplete attempts and intermediate comparisons are preserved separately and
are not counted as source observations. One initial ceiling triangle missed the
intended path; the final fixture covers it and proves a real370 callback.
## Shared state through populated static geometry
The box is identity-oriented with cube(−50,−50,−10)..(50,50,10), radius40;
sphere radius10, previous(0,0,50), current(0,0,15). The dynamic face query normally
gives rootZ20, face40, edge10 and hitZ10. A separate old polygon99 is seeded.
| Fixture | Original result / dependency |
|---|---|
|Static wallZ35|Nearer static face wins: rootZ45, face15, hitZ35, last0;2 passes/4 cell-list visits.|
|WallZ10 at equal dynamic face|No replacement: rootZ20, face40, old polygon99 survives.|
|Far wallZ0|No replacement; old polygon99 still supplies the terminal callback.|
|Walls35→30 and30→35|Both end atZ45/face15; winning polygon is0 or1 according to its identity.2 passes/8 visits.|
|Root-mutating callbacks|Final local root still overwrites the mutation, now toZ45.|
|Box response inhibited|Only dynamic callback; no static visits, passes or setter.|
|Static geometry disabled|2 node visits, no static kernel calls; retained old-polygon callback still occurs.|
|Sphere42C enabled|12 dynamic kernel calls in mode0,4 static kernel calls in mode1, terminal callbacks/setter in mode0.|
|Old ceiling polygon|No accepted static face, but retained normal−Y invokes370 rather than374.|
|Real ceilingY35|Root(0,25,23.7499962), face43.7501259, last1;370 follows dynamic374, then314;2 passes/12 visits.|
|Repeated query without context reset|Nearest limits reset per dynamic query; last polygon and all callback/pass/visit/kernel counters persist across calls.|
The real ceiling fixture starts currentY30 and uses a triangle covering positiveZ.
The skew fixture instead starts previousY−30/currentY0 and registers the original
wall/ceiling triangles without claiming that the ceiling accepts. Its dynamic
edge→face sequence is the numerical diagnostic below.
The callback event string records virtual slot, polygon identity and global mode.
Dynamic callbacks use374/null. Static callback370/374 selects from the **retained
last polygon's normal**, and final314 follows callbacks. Every observation compares
contact, rootXYZW, face/edge/hitXYZW, last identity, callback/setter/pass/visit counts,
global mode, mode0/1 kernel counts and complete event order. Sequential queries use
the same synthetic actor; they do not prove registry order across real actors.
## First divergences and supported corrections
Text offsets are VA−400000; the source tools check actual PE mapping.
1. **Visited nodes and primitive tests have different owners.** At47B6A8..47B6B9
(file7B6A8),695D70 increments before4F69E5 is checked. The first managed
comparison reported0 visits versus original2 with geometry disabled.
`StaticCollisionContext.NodeVisits` now counts before that gate, while
`TriangleTests` continues counting only dispatched primitives. This is a
diagnostic-state correction, not a new collision response.
2. **Segment projection X retains its product.** The skew case's first differing
kernel is zero-based6: input(0,0,15,1) matches, but source hitX is
`8.68490133e-7` and managed hitX was0.4699A0..4699DE (file699A0) retains
directionX×projection until adding originX; Y/Z products have intermediate
float stores. `OriginalSphereResponse.NearestOnSegment` now follows those
separate boundaries.47C194..47C1D0 likewise retains X/Z products through the
final edge-root subtraction. This makes the complete step6 output match.
3. **Ray normalization and hit publication retain intermediates.** The next
mismatch is step7's face hit.469BE5..469C09 (file69BE5) uses OMT's returned
quick-magnitude quotient directly for reciprocal normalization; a float return
inserted an extra rounding.469DF0..469E13 (file69DF0) stores hit coordinates
after multiply+add. `OriginalBodySupport` now retains the magnitude intermediate
as double for these ray callers, and delays coordinate rounding until that
final store. Its public float QuickMagnitude result and caller-owned nearest
reduction remain float. All14 final observations and12 kernel steps now match.
Intermediate failure logs and experiment snapshots remain local, including
`managed-initial.log`, `managed-after-visits.log`, `projection-experiment.log`,
`ray-experiment.log` and `wide-experiment.log`. No epsilon or near-zero clamp was
added. The comparator parses native9-digit and managed shortest decimal spellings
back to float32; distinct values remain failures. Double intermediates reproduce
these fixtures, not every x87 extended-precision or table-boundary case.
## Implementation and validation
Runtime changes are confined to the shared support/ray and sphere kernels plus
the additional static visit counter. `tests/GamePort.DynamicReclipTests` keeps
fixed numeric references and all12 kernel steps for artifact-free normal runs;
`--native` checks the complete generated observation file, and `--dump` supports
future differential tracing. Python tests guard instruction boundaries, safe
generation and exact-float comparison semantics. Both probe source and expected
data are test fixtures, not new actor implementations.
`scripts/test.sh` exits0 with348 Python+2 HUD Python checks, all managed projects,
36 Vulkan/render checks with zero validation errors, audio lifecycle and539
original hashes. Debug build has zero warnings/errors. Native Linux SDL Carl
control (first neutral input/300 idle ticks) and VR04 jetpack steering/resize
also pass with zero validation errors. Windows x64 self-contained cross-publish
succeeds to a separate artifact directory. Existing packaged DLLs and BUILD-INFO
remain identical to924ad2f; no new package or Windows execution is claimed.
Reproduction after **waiting for disposable prefix preparation to complete**:
```sh
python3 tools/binary/dynamic_reclip.py ../Original --output artifacts/dynamic-reclip/native
i686-w64-mingw32-gcc -O0 -Wall -Wextra -Wno-unused-function -o artifacts/dynamic-reclip/native/probe.exe artifacts/dynamic-reclip/native/dynamic_reclip_probe.c
python3 tools/recon/reference_capture.py --asset-probe --probe-executable artifacts/dynamic-reclip/native/probe.exe --probe-success DYNAMIC_RECLIP_COMPLETE --seconds 45 --output artifacts/dynamic-reclip/probe-verified
python3 tools/binary/dynamic_reclip.py ../Original --output artifacts/dynamic-reclip/native --capture artifacts/dynamic-reclip/probe-verified
source scripts/env.sh
dotnet run --project tests/GamePort.DynamicReclipTests -- --native artifacts/dynamic-reclip/native/observations.json
```
For a complete field/step difference report, run `--dump` to a local file, then
pass that file as `dynamic_reclip.py --managed <file>` with the same output path.
The generated comparison explicitly records every mismatch; it does not turn
nearby values into parity. Final platform/suite receipts are in progress.md and
compatibility.md; local identities/results are in artifacts/dynamic-reclip/.
## Remaining gate and next chunk
Populated-grid composition now supports a bounded shared-context response. Before
live admission, recover actual Jimmy/VR04 actor callback mutations, live flag/
radius reads after callbacks, constructor-only helper re-entry/cleanup and real
physical registration order across LOAD/RETURN. Reentrant global-mode and grid
mutation, extreme coordinates/degenerate inputs and general x87 boundary behavior
remain open. Preserve lab/outside capsule and the existing VR04 static-only actor
set. Next focus on real actor callback ownership, then implement the bounded
response adapter against those interfaces. No movement tuning, angle reset,
gadget feature or unsupported dynamic actor is introduced here.
+6
View File
@@ -43,6 +43,12 @@ mutations/children/order remain open. Next extend to nonempty static reclip and
ordered-body context, then real actor callbacks and retained helper cleanup.
Keep the current live solver and924ad2f packages while those gates remain.
Populated follow-up: [dynamic reclip](dynamic-reclip.md) now matches14 complete
observations and12 kernel steps. Direct source-store corrections remove two
rounding divergences without admitting new actors. Next prioritize real Jimmy/
VR04 callback mutations and live getter timing, then constructor-only helper
cleanup/order across LOAD/RETURN; integrate the bounded response only afterward.
## Objective
Establish the original ownership and timing of player state across gameplay,
+16
View File
@@ -1,5 +1,21 @@
# Compatibility and validation
2026-11-07 populated-reclip correction:14 complete original/C# observations and12
kernel steps match exact float32 values after source-store arithmetic corrections.
Initial node-visit, edge-projection and ray-hit failures are retained. No epsilon
or near-zero reset is added. `scripts/test.sh` exits0:348 Python checks,2 HUD
Python checks, all managed projects,36 Vulkan/render checks with zero validation
errors, dummy-device audio lifecycle and539 unchanged original files. Debug build
has zero warnings/errors. Native Linux1920×1080 SDL Carl-control and VR04 jetpack/
resize replays also pass with zero validation errors; first neutral input and300
idle ticks stay upright. Windows x64 self-contained cross-publish succeeds to a
separate artifact directory; no new Windows execution is claimed. Existing Linux/
Windows packaged DLLs and BUILD-INFO remain the verified924ad2f baseline.
Dynamic composition still uses synthetic actors/callbacks/grid and is not live
pair admission. Double intermediates reproduce these cases, not all x87 inputs.
See [populated reclip](checkpoint/dynamic-reclip.md), with final logs/identities
under artifacts/dynamic-reclip/validation.json and its referenced artifacts.
2026-11-07 dynamic-context checkpoint:17 copied-original mode2/static reclip
observations match the C# test composition in full, including XYZW, shared
nearest/hit/old polygon, callback/setter/pass counts and mode. All12 instrumented
+10
View File
@@ -412,3 +412,13 @@ assets and generated packages stay excluded.
Keep source topology and kernel boundary distinct. Empty-grid/synthetic actor
results do not admit live scene collision. Guard default Jeep deletion while
leaving cleanup-mutated active lifetime open. See checkpoint/dynamic-context.md.
58. **Follow evidenced float-store boundaries instead of accepting nearby roots.**
Populated reclip exposed an edge projection mismatch at triangle6, then a ray
hit mismatch at7. Delayed X projection/edge-root products, retained quick-
magnitude normalization and delayed hit-coordinate stores close14 complete
float32 comparisons and12 kernel steps without epsilon or near-zero clamps.
Keep public float magnitude/caller nearest state and source per-coordinate
stores distinct; double intermediates do not establish general x87 identity.
Count static node visits before the geometry gate, separately from actual
primitive tests. Real actor callbacks/lifetime still gate live integration.
See checkpoint/dynamic-reclip.md.
+25
View File
@@ -7,6 +7,31 @@ Next-work plan: [recover interactions, then implement verified changes](checkpoi
It defines the Carl source/differential gate, the bounded implementation step,
movement/contact/animation recovery and the end-to-end platform checkpoint.
## Populated reclip and source rounding
[Populated dynamic reclip](checkpoint/dynamic-reclip.md) extends the copied
original query to14 cases with actual synthetic static triangles and12 captured
kernel steps. All fields match exactly as float32, including wall/ceiling retry,
old-polygon callbacks, disabled-geometry visits and per-primitive lower-edge mode.
A skew edge first diverged at triangle6; original instruction stores support
delayed X projection/edge publication and ray normalization/hit-point rounding.
Those shared kernels now follow the evidenced boundaries. The public float
quick-magnitude result is preserved; double intermediates are not a universal
x87 oracle. Node visits are now counted separately from primitive tests.
No new dynamic actor or scene admission is enabled. Actual Jimmy/VR04 callback
mutations, post-callback live flag/radius reads and retained helper cleanup/order
remain next. Native source execution uses synthetic actors/grid under Wine32;
whole-original and native Windows gameplay remain separate. Detailed verification
receipts and limits are recorded in compatibility.md and the linked report.
Verification complete: scripts/test.sh exits0 with348 Python+2 HUD checks, all
managed projects and36 Vulkan/render checks with zero validation errors; audio
lifecycle and539 original hashes pass. Native Linux Carl neutral/300-idle and
VR04 SDL steering/resize pass with zero validation errors. Windows x64 cross-
publish succeeds; no new Windows runtime claim. Existing924ad2f packages remain
unchanged while the verified source correction is committed separately.
## Dynamic response and shared context
[Dynamic context](checkpoint/dynamic-context.md) adds17 complete original/C#
+5
View File
@@ -5,6 +5,11 @@ offsets, virtual addresses, API symbols or observed behavior. Decompiled output
evidence, not original source. Original gameplay has not yet been observed successfully; isolated original DLL
probes and modern runtime checks are recorded separately.
- [Populated dynamic reclip](checkpoint/dynamic-reclip.md):14 full observations
and12 captured kernel steps. Original instruction stores explain delayed edge
projection, ray normalization and hit-coordinate rounding. Shared kernels now
reproduce those cases exactly; node visits are distinct from primitive tests.
Actor callbacks/live getter timing and retained helper order remain open.
- [Dynamic box/shared context](checkpoint/dynamic-context.md): full47A5E0 plus
static reclip/kernel execution,17 complete C# composition comparisons and12
captured kernel triangle/normal arguments. Nearest reset, retained hit/last
+8 -2
View File
@@ -86,8 +86,14 @@ Execution order and acceptance criteria: [next work plan](checkpoint/next-work-p
Preserve old polygon/hit, post-callback publication and kernel winding.
Guard Jeep default teardown eligibility separately from full cleanup. See
checkpoint/dynamic-context.md; no live pair admission is enabled.
- [ ] Extend dynamic reclip to nonempty static geometry/ordered bodies, real
actor callbacks and cleanup-mutated helper lifetime before live integration.
- [ ] Recover ordered-body context provenance, real actor callbacks and cleanup-
mutated helper lifetime before live integration. The populated fixtures below
close the bounded nonempty reclip contract.
- [x] Match populated static reclip in14 complete original/C# cases and12 kernel
steps. Correct directly evidenced projection, ray normalization and coordinate
float stores; separate pre-gate node visits from primitive tests. See
checkpoint/dynamic-reclip.md. Sequential same-actor queries do not establish
real physical ordering, callback mutation contracts or helper lifetime.
- [x] Expand original lifecycle coverage:96 Carl contact,8 composed input/
controller/skip-prefix,48 task/objective and140 base/STOP cases; recover complete