Slice 2 v1 (`892019b`) attempted to close issue #96 by removing the
PhysicsEngine.cs L622 per-tick CP seed. v1 build/test green, CP-write
count dropped 91% in scen3 re-capture, BUT user happy-test surfaced
a regression: BSP step_up at the last step of stairs failed because
sub-step 1's AdjustOffset had no ContactPlane to compute the lift
direction.
Slice 2 v2 (`f8d669b`) reverted the seed removal + added a no-op-if-
unchanged guard inside CollisionInfo.SetContactPlane. The guard
early-returns when called with values matching current ci state.
Outcome:
- #96 PARTIALLY ADDRESSED, scope updated in ISSUES.md to "accepted as
documented retail divergence." The seed is load-bearing for step_up;
closing #96 fully would require deeper refactor (AdjustOffset
fallback to body.ContactPlane). Guard is benign improvement.
- Slice 2 v2 verification capture (scen3_inn_2nd_floor_slice2v2/
acdream.log) committed as evidence — 226,464 cp-writes from L624
seed confirms guard doesn't trigger for fresh-ci-per-tick pattern.
- Slice 2 v1 verification capture (scen3_inn_2nd_floor_slice2/
acdream.log) also committed — confirms v1 actually reduced cp-writes
(2,690 total) but the step_up regression made it unshippable.
NEW M1.5 BLOCKER FILED — issue #98: cellar ascent stuck at last step.
Evidence in slice2v2 capture's cell-transit chain:
0xA9B4014B → 0xA9B4014A → 0xA9B4013F → 0xA9B4014A → 0xA9B4014B → ...
(Z stable ~96.4; CellId ping-pongs every tick)
This is Finding 3 family (cell-resolver hysteresis missing) — same
root cause as #90 workaround + scen4 sling-out. Retail oracle:
CObjCell::find_cell_list Position-variant at
acclient_2013_pseudo_c.txt:308742-308783.
NEXT — A6.P3 slice 3:
- Port retail's cell-array hysteresis into ResolveCellId +
CheckBuildingTransit.
- Closes#98 (cellar-up), possibly #97 (phantom collisions same
instability family), enables #90 workaround removal.
Documents updated:
- ISSUES.md — #96 scope updated, #98 filed
- docs/plans/2026-04-11-roadmap.md — A6.P3 slice 2 marked SHIPPED,
slice 3 scope added
- CLAUDE.md — Currently-working-toward block updated to slice 3
Test suite: 1148 pass + 8 pre-existing fail (baseline maintained).
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Unexpected slice 1 win: the synthesis-strip + Mechanism B (LKCP
restore) fix didn't just close Finding 2 (CP-write blowup) — it also
unblocked stair-walking, which A6.P2 had categorized as Finding 1+3
territory expected to need separate fixes. User reports walking up
and down the inn stairs multiple times in acdream post-fix.
Shape shift in tag distribution:
Tag Pre-fix (FAIL) Post-fix (SUCCESS) Signal
---- ------------- ------------------ ------
indoor-walkable 859 0 synthesis gone
push-back-cell 1478 879 (-40%) multi-cell relaxed
push-back 51 345 (+577%) real step-up firing
push-back-disp 4156 6055 (+46%) real traversal
cp-write 33969 57846 L622 seed (slice 2)
Pre-fix: synthesis firing while physics hammers BSP trying to resolve
stair-step (failure mode). Post-fix: real BSP queries succeeding, real
step-up + step-down landing. Same shape as retail's stair-climb
(retail scen2: BP2 step_up=188, push-back-disp dominates).
A6.P2 Finding 1 (dispatcher entry frequency mismatch) hypothesis was
"likely secondary effect of Finding 2 — may close as side effect of
the fix." Confirmed empirically: dispatcher activity now matches
retail-like shape without explicit Finding 1 work.
Remaining (slice 2 territory):
- L622 per-tick PhysicsEngine.ResolveWithTransition seed fires 99.3%
of remaining cp-writes; retail's equivalent fires zero times on
flat-floor walks. Gate this seed to close the remaining CP-write
gap.
- Phantom collisions + occasional fall-through on 2nd floor reported
by user during happy-testing. New issue to file.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Code-review feedback on commit 6b4be7f:
- Section 1: strip stale [309NNN] inline annotations (off by 2-8
lines from actual file content; the 0052c1xx address comments
are the reliable anchor); address comments already present in the
decomp output are now used as inline anchors instead
- Section 2: validate_transition function header is at file line
272547 (was: 272538, inside the preceding check_collisions
function). Address 0050aa70 + LKCP-block range 272565-272583
were already correct. References section updated to match.
- Section 5: add note that SetContactPlane re-latches LKCP fields
(no-op when LKCP is the source, but non-obvious side effect)
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Pre-fix research note grounding the indoor CP-retention refactor in
retail's exact LKCP-restore pattern (acclient_2013_pseudo_c.txt:272565-272582)
and CEnvCell::find_env_collisions tiny shape (line 309573).
Key findings:
- find_env_collisions writes NO ContactPlane — only BSP Path 6 does (Mech A)
- validate_transition Collided/Slid/Adjusted branch calls set_contact_plane
from LKCP when proximity guard passes (global_curr_center, not global_sphere)
- Our ValidateTransition is missing the SetContactPlane call in that branch
(sets Contact/OnWalkable flags only) — this is the gap Task 4 closes
- Proximity sphere should be GlobalCurrCenter[0] not GlobalSphere[0]
- Exact insertion point: TransitionTypes.cs ~line 2849, inside the
'radius + EPSILON > |angle|' proximity-guard branch
Output of this note drives the per-transition Mechanism B insertion
point selection in Task 4 + the slice-1 acceptance shape.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
A6.P1 (cdb probe spike) + A6.P2 (analysis report) both SHIPPED this
session. Updated:
docs/plans/2026-04-11-roadmap.md — M1.5 phase block now shows A6.P1
+ A6.P2 SHIPPED with commit refs; A6.P3 entry expanded with the
Finding-2-first sequencing recommendation from A6.P2; A6.P4 entry
notes the original "Holtburg Sewer end-to-end" acceptance walk is
unreachable (sewer doesn't exist).
docs/plans/2026-05-12-milestones.md — M1.5 demo scenario split into
building/cellar half (achievable post-A6.P3) + dungeon half (blocked
on issue #95 visibility blowup; promote to post-M1.5 if #95 isn't
fixed in scope). Issue list updated: added #95 + indoor sling-out
(new from scen4); marked stairs/2nd-floor/cellar as characterized by
A6.P2 Finding 2 family.
CLAUDE.md — Currently-working-toward block now points at A6.P3 as
the active phase. A6.P1 + A6.P2 ship noted with the findings doc
pointer. Demo-scenario note updated to reflect the sewer + #95
reality. Issues-in-scope updated.
Also includes a 1-line trailing-prompt addition to scen3 + scen4
retail.log files (cdb wrote one more `0:000>` after the kill that
landed after the original capture commits).
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Replaces the A6.P1 stub with the analysis pass over 5 paired captures
(scen1-5). Scen6-9 (sewer-specific) cancelled because the Holtburg Sewer
doesn't exist on this ACE server and any substitute dungeon hits issue
#95 (portal-graph visibility blowup) on entry.
Four findings ready for A6.P3 sequencing:
Finding 1 — Dispatcher entry frequency mismatch (4x to 281x fewer in
acdream). Likely secondary effect; may close as side-effect
of Finding 2 fix.
Finding 2 — ContactPlane resynthesis blowup. 250x to INFINITE more CP
writes in acdream. Strongest single signal; scen3 shows
retail wrote CP zero times during a flat 2nd-floor walk
while acdream wrote 86,748 field updates. Primary M1.5
root cause. HIGH severity.
Finding 3 — Indoor cell-resolver sling-out (scen4). Resolver flings
+Acdream across landblock boundary; CheckBuildingTransit
fires 5,495 times during the sling while indoor BSP is
barely queried. Same family as the M1.5 cell-tracking
ping-pong hypothesis. HIGH severity.
Finding 4 — Portal-graph visibility blowup (scen5 incidental). Filed
as issue #95; not strictly A6 scope but documented here so
A6.P3 sequencing knows about it.
Tables 1+2 (per-site push-back delta + path-frequency diff) deferred to
A6.P1.5: the v4 cdb probe captures function entry only, not exit values.
Adding paired exit BPs is ~1 hour of cdb scripting work but not needed
unless A6.P3 fixes fail to close the symptoms.
Table 3 (CP lifecycle) fully populated — geometric mean CP-write ratio
across 4 finite scenarios is ~1,470x; median ~2,200x.
Table 4 (sub-step state mutations) partially populated with proxy
metrics (per-tag firing rates).
M1.5 symptom coverage matrix: every in-scope physics symptom maps to
at least one finding. Acceptance per spec §4.7 met.
A6.P3 sequencing recommendation: Finding 2 first (highest-confidence
single-cause; may close Finding 1 as side effect), re-run captures, then
Finding 3. Issue #95 handled separately outside A6.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Substituted "Holtburg Sewer" portal with Town Network Portal — no
sewer entry exists in this world (user-verified). Town Network is
also an outdoor->indoor portal transition with the same physics
signature.
Both clients walked to the portal, entered, walked 2 m inside.
Retail: clean traversal. Acdream: also clean (no failure mode).
Retail (decoded, 23,890 raw / 9,769 BP lines):
BP1 transitional_insert: 13,863
BP4 find_collisions: 9,552
BP5 adjust_sphere: 97
BP6 check_walkable: 55
BP7 set_contact_plane: 65 (moderate, portal threshold + indoor)
BP2 step_up: 1
Acdream (31,914 lines, no failure):
[cp-write]: 20,956 (vs retail BP7 = 65 — ~322x ratio)
[cell-cache]: 9,642 (Holtburg landblock streaming)
[check-bldg]: 740
[push-back-disp]: 34 (flat-ground walking)
[push-back]: 1
[cell-transit]: 12 (CLEAN traversal, no thrashing)
cell-transit event chain — captures the portal entry signature:
0x00000000 -> 0xA9B30030 (login teleport)
0xA9B30030 -> 0xA9B40029 -> 0xA9B40021 -> 0xA9B40019 ->
0xA9B40011 -> 0xA9B40012 -> 0xA9B4000A -> 0xA9B4000B ->
0xA9B40003 (walked across Holtburg, all reason=resolver)
0xA9B40003 -> 0x00070143 reason=teleport (PORTAL ENTRY)
scen5 is the "control" — both clients reached their target, no
visible failure. The CP-write blowup persists as the only A6.P2
divergence. Useful baseline for separating "indoor physics broken
during walking" (scen2, scen3, scen4) from "indoor physics okay
when portal-delivered" (scen5).
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Asymmetric pair (scenario-level, not protocol-failure):
- Retail: user walked UP out of the cellar (ascent of 2 cellar
steps + exit through doorway) — captures ascent + indoor-to-
outdoor transition.
- Acdream: user teleported INTO the cellar, walked a few meters,
the resolver flung +Acdream OUTSIDE the cottage entirely
(landblock prefix changed A9B4 -> A9B3 mid-walk) — captures
a real indoor physics failure that's not a stair issue per se.
Both traces are valuable to A6.P2 even though they don't match
walk-for-walk.
Retail (decoded, 22,536 raw / 12,875 decoded BP lines):
BP1 transitional_insert: 9,402
BP4 find_collisions: 12,596 (ended in mem-access error
@ hit#12596 - cdb hit a null
transition arg, dropped to
interactive prompt; worth a
note for A6.P2 retail edge)
BP5 adjust_sphere: 136
BP6 check_walkable: 128
BP2 step_up: 13 (2-step cellar = 13 vs scen2
4-step inn = 188; non-linear)
BP7 set_contact_plane: 3 (Finding 2 holds)
Acdream (42,001 lines, ended with sling-out):
[cp-write]: 35,624
[check-bldg]: 5,495 (CheckBuildingTransit fired
constantly trying to re-resolve
which building +Acdream was in)
[cell-cache]: 540
[push-back-disp]: 82 (very few dispatcher hits)
[push-back]: 1 (almost no sphere-adjustment)
[indoor-bsp]: 2 (indoor BSP barely queried!)
[cell-transit]: 3 (3 transit events captured the sling:
0xA9B40148 -> 0xA9B40029 -> 0xA9B30030
all reason=resolver)
Sling-out signature: indoor BSP never engaged (only 2 indoor-bsp
hits), but the cell resolver fired 3 transit events crossing a
landblock boundary, with check-bldg thrashing in between. This is
distinct from scen2's stair-attempt pattern (which hammered the
BSP); scen4 shows the resolver pushing the character out of indoor
space entirely without triggering the indoor BSP collision path.
A6.P2 fix surface: investigate why ResolveCellId / CheckBuildingTransit
push a player from indoor cell 0xA9B40148 to outdoor cell 0xA9B30030
through routine walking. Likely the same family as the M1.5 hypothesis:
indoor cell membership isn't sticky (the ping-pong bug from the
2026-05-20 A4 handoff in a different guise).
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
User reached the 2nd floor in acdream via ACE @teleport (stair-physics
unblocked separately by scen2). Retail walked normally to 2nd floor.
Both clients performed the same walk: forward 3 m, sidestep 1 m,
walk back. Flat-floor scenario, no stairs, no transitions.
Retail (decoded, 21,337 lines):
BP1 transitional_insert: 10,217 hits
BP4 find_collisions: 10,636 hits
BP5 adjust_sphere: 113 hits
BP6 check_walkable: 113 hits threshold=0.6642
BP2 step_up: 0 hits (no stairs)
BP3 set_collide: 0 hits (no walls)
BP7 set_contact_plane: 0 hits (KEY: zero CP updates)
Acdream (93,558 lines):
[cp-write]: 86,748 (vs retail BP7 = 0 — INFINITE ratio)
[push-back-disp]: 2,752
[push-back]: 320
[push-back-cell]: 550
[other-cells]: 550
[indoor-bsp]: 1,061
[indoor-walkable]: 707
KEY FINDING for A6.P2: scen3 is the strongest CP-write blowup
evidence yet. On a flat 2nd-floor walk where retail's
set_contact_plane fires ZERO times across the entire scenario,
acdream rewrites the contact plane 86,748 times. This is the
exact pattern Finding 2 hypothesized (M1.5 design spec §1.2):
acdream resynthesizes CP every frame instead of retaining it
through the documented retention mechanisms (LKCP-restore,
Path-6 land write, post-OK step-down probe).
scen3 pair confirms CP-write blowup isn't stair-specific — it
fires equally for ordinary flat-floor walking inside any indoor
cell. A6.P3 fix surface: same as Finding 2 — stop resynthesizing
CP per frame.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Original acdream capture (a9a427f) was a doorway-walk because acdream's
indoor stair physics doesn't work. For A6.P2 to characterize the
divergence we need the FAILURE captured, not a substitute walk.
User re-attempted the inn stairs in acdream (whatever it produces:
bumping, sliding, stuck). Failure signature is dramatic vs door-walk:
Tag | door-walk | stair-attempt | ratio
----------------+-----------+---------------+------
push-back-disp | 1,141 | 4,156 | 3.6x
push-back-cell | 87 | 1,478 | 17x
other-cells | 87 | 1,478 | 17x
indoor-bsp | 343 | 1,286 | 3.7x
indoor-walkable | 227 | 859 | 3.8x
cp-write | 70,244 | 33,969 | 0.5x (!)
The 17x explosion on push-back-cell / other-cells says acdream's
CheckOtherCells loop fires constantly when physics can't resolve a
stair-step — the indoor BSP query fails, then the multi-cell
fallback fails, then the next tick repeats. The cp-write DROP
(half the door-walk volume) is the inverse signal: when no ground
plane resolves, no CP gets written. Both are A6.P2 fix-surface
indicators.
Now scen2 pair = retail successfully climbs (BP2 step_up=188) vs
acdream tries and fails (push-back-cell explosion).
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Python tool that decodes the retail.log hex-bits float fields produced
by a6-probe.cdb v4 into IEEE 754 single-precision values. Required
because cdb's .printf %f doesn't reliably format floats from dwo()
reads — v4 works around this by emitting 32-bit hex, this script
reinterprets via struct.unpack('<f', struct.pack('<I', value)).
Verified against scen1 retail.log:
BP6 threshold_h=0x3F2A0751 → threshold=0.6642 (= FloorZ exactly)
BP5 hit#1 Nz_h=0x3F800000 → Nz=1.0 (ground normal)
9,517 float fields decoded across 9,331 lines.
Output written next to input as .decoded.log. Format matches
acdream-side [push-back] probe (4-decimal floats), so A6.P2
analysis can compare line-for-line.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Acdream-side capture for the Holtburg inn doorway walk, paired with
the v4 retail capture committed at 180b4a5. 84,130 lines total.
Probe line distribution (~30 sec session, ~2 sec actual walk):
[push-back] (adjust_sphere): 8 hits — vs retail BP5 12 hits
[push-back-disp] (dispatch): 295 — vs retail BP4 5818 (!)
[push-back-cell] (other_cells): 5 — vs retail's check_other_cells
[indoor-bsp]: 26
[cell-transit]: 30 (cell ID changes)
[cp-write]: 73,304 (per-field writes) — vs retail BP7 18 fn calls (!)
[cell-cache]: 540
Two major divergences already visible from this single scenario:
1. DISPATCH FREQUENCY: retail's BSPTREE::find_collisions fires 20×
more than acdream's BSPQuery.FindCollisions. Could reflect either
different physics tick rate, different sub-step cadence, or
different call paths into the dispatcher.
2. CONTACTPLANE LIFECYCLE: acdream writes CP fields 73,304 times
in 30 seconds (~2,400/sec). Retail calls set_contact_plane 18
times (~0.6/sec). Even with a 6× field-write multiplier per
set_contact_plane call, that's ~100 actual CP updates in retail
vs ~12K in acdream — 100-1000× more frequent in acdream. This
directly confirms the spec's hypothesis that FindEnvCollisions
indoor branch is rewriting CP every frame (sub-step?) instead
of retaining it across frames. Same family as the
TryFindIndoorWalkablePlane workaround.
Per-call shape comparison (BP5 hit#1):
Retail: plane=(0,0,1) d=-0.0, sphere=(0.0046,10.31,-0.27) r=0.48,
mvmt=(0,-0,-0.75), winterp=1.0
Acdream: plane=(0,0,1) d=-0.0, sphere=(-0.43,11.02,0.46) r=0.48,
mvmt=Z-down, winterp 1.0→0.96 (small adjust applied)
Identical operation SHAPE (ground plane + vertical step-down probe
+ same radius). XY positions differ because walks were independent.
Scenario 1 complete. Remaining 8 scenarios deferred per user
direction. Python hex→float decoder + A6.P1 handoff doc to follow.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
v4 cdb probe captured paired field data for the Holtburg inn
doorway walk. 13,552 BP hits in ~2 sec of walking. Distribution:
- BP1 transitional_insert: 7,686 (sub-step loop)
- BP4 find_collisions: 5,818 (per cell per sub-step)
- BP5 adjust_sphere_to_plane: 12 (the over-correction suspect)
- BP6 check_walkable: 12
- BP7 set_contact_plane: 18
Smoking-gun verification:
BP6 threshold_h=0x3F2A0751 ≈ 0.664 = PhysicsGlobals.FloorZ
BP5 plane normal = (0,0,1), movement = (0,-0,-0.75) — classic
step-down probe against the ground polygon
BP5 sphere radius = 0x3EF5C28F ≈ 0.480 m — player foot sphere
All hex-bits floats decode cleanly via Python struct.unpack('<f').
Decoder script TBD as part of the handoff.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
v3 with @@c++(*(float*)..) STILL produced 0.000000 across the board.
Conclusion: cdb's .printf %f is unreliable for our use case — possibly
doesn't handle the float-to-double promotion in varargs the way C
printf does, or has a deeper limitation we don't have time to debug.
Pivoting to: print all floats as 32-bit hex bits via %08X, reinterpret
in the Python analysis pipeline via struct.unpack('<f', bytes.fromhex(...))
to recover IEEE 754 single-precision values.
This bypasses cdb's float formatting entirely. Integer reads (which
work — substeps, insertType, collide flag, isWater) stay as %d.
The smoking gun: BP6's check_walkable threshold should be 0.0871556997
(cos 85°) per the decomp call site at acclient_2013_pseudo_c.txt:273202.
v4's BP6 should output threshold_h=0x3DB283D7. If it does, the
infrastructure is sound and we can proceed to all 9 scenarios.
v3 capture preserved as retail-v3-cpp-zero-floats.log audit trail.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
v2 dry-run produced correct hit counts but all %f field values
printed as 0.000000 — including BP6 threshold which the decomp says
must be 0.0871556997f (cos 85°). Root cause: cdb's MASM evaluator
returns dwo(addr) as a 32-bit integer; .printf %f expects a 64-bit
double; passing the integer to %f produces formatted-zero garbage.
Fix: switch all float-reading expressions to @@c++(*(float*)addr).
The C++ evaluator dereferences memory as a float pointer, returning
a proper float that .printf %f formats correctly. Integer reads (%d)
still use MASM dwo() — that works.
For double-indirect (pointer args), the form is
@@c++(*(float*)(*(unsigned int*)(@esp+N)+offset))
which reads the pointer at [esp+N], adds the offset, and treats the
result as a float pointer.
v2 capture preserved as retail-v2-zero-floats.log audit trail.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Dry-run of scenario 1 (retail-v1-broken-offsets.log preserved as
audit trail) surfaced three issues with the v1 cdb script:
1. STACK-ARG OFFSETS WRONG: BP actions used arbitrary registers
(@edx, @edi) to read function args, but __thiscall puts non-this
args on the stack ([esp+N] after the return address). All 12 BP5
"adjust_sphere" hits printed Nx=0.0 Ny=0.0 ... — fields not read.
Fixed by writing a type dumper (a6-types-dump.cdb + runner) that
uses cdb's `dt` command against the loaded PDB to get authoritative
struct offsets. v2 probe script (to be written next) will use
double-indirect reads (dwo(poi(@esp+N)+offset)) with correct
offsets from the dump.
2. TEE-OBJECT UTF-16 ENCODING: PowerShell's default Tee-Object writes
UTF-16 LE with BOM, making logs unparseable by grep without
conversion. Runner now uses Out-File -Encoding ASCII. Sacrifices
live console echo; use `Get-Content -Tail 50 -Wait` in a separate
shell if live monitoring is needed.
3. BP6 SYMBOL NOT FOUND: `acclient!CTransition::validate_walkable`
doesn't exist in the PDB. Decomp at line 272811 has
`CTransition::check_walkable` — likely the actual name. To be
verified + fixed in v2.
The BP hit-count distribution from v1 is still meaningful diagnostic
data (14,318 transitional_insert + 16,558 find_collisions + 40
set_contact_plane + 12 adjust_sphere + 1 step_up + 1 set_collide in
a 2-second walk through the inn doorway). Preserved as a baseline
sanity-check the v2 distribution can be diffed against.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
User swapped in the correct Sept 2013 EoR build acclient.exe.
GUID {9e847e2f-777c-4bd9-886c-22256bb87f32}, linker UTC
2013-09-06T00:17:56 — exact match for refs/acclient.pdb.
T15 captures are unblocked.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Creates the 9 per-scenario capture directories (gitkeep stubs) and
the findings doc stub at docs/research/2026-05-21-a6-cdb-capture-findings.md.
A6.P1 fills the capture log slots (Task 15, user-driven); A6.P2
fills the analysis tables and findings section.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Audit trail for the A6.P1 capture session: the retail binary at
C:\Turbine\Asheron's Call\acclient.exe is the 2015-06-12 build
(GUID {08e25c14-e2a1-46d5-b056-92b2e43a7234}), not the Sept 2013
EoR build that pairs with refs/acclient.pdb
(expected GUID {9e847e2f-777c-4bd9-886c-22256bb87f32}).
BP-driven A6 captures cannot proceed until the matching binary is
installed. User needs acclient.exe v11.4186 (linker timestamp
2013-09-06) to match our PDB.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
O-T1 audit (REPORT-ONLY) maps acdream's transitive closure on WorldBuilder:
33 files / ~7.7K LOC across Chorizite.OpenGLSDLBackend (28 files) and
WorldBuilder.Shared (5 files). Verdict on O-Q1 (thread-model): SAFE —
adapters run render-thread only; no worker-thread access to WB code.
Spec amendments incorporated via brainstorm:
- O-D7: Refactor ObjectMeshManager to take DatCollection directly (not
via adapter). T4 safety check — fall back to thin adapter if call-site
count >20.
- O-D8: Drop LandSurfaceManager, EnvCellRenderManager, PortalRenderManager,
TerrainRenderManager from the extract list — audit confirmed not reachable
(we have our own ports or never used them).
- O-D9: Promote 3 internal types in Chorizite to public on extraction
(EmbeddedResourceReader, TextureFormatExtensions, BufferUsageExtensions).
- O-D10: Strip [MemoryPackable] from TerrainEntry (we don't serialize).
- O-D11: Namespace AcDream.Core.Rendering.Wb.* for extracted code.
- O-D12: Drop ResolveId + [indoor-upload] NULL_RESULT diagnostic block.
Task breakdown: T6 (EnvCell/portal) eliminated; T5 (stateless helpers)
shrinks to 0.5d; T4 (mesh + refactor) grows to 2.5d. Net effort estimate
holds at ~7.75d.
All originally-open spec questions are now closed (Q1/Q2/Q3/Q4) or
deferred to T3 with an explicit verify step (Q5: SixLabors.ImageSharp
reachability).
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Self-contained prompt for a fresh Claude Code session. The next session
reads it once, has all the context it needs, produces the WB-usage
closure audit at docs/research/2026-05-21-phase-o-t1-wb-audit.md,
and stops before any extraction. Investigation-only (the /investigate
skill applies). User reviews the audit before T2 begins.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Continued indoor testing through 2026-05-20 surfaced a deep family
of physics + lighting bugs that span buildings AND dungeons. Today's
session shipped 5 surgical fixes (A4 + #89 + #90 + #91 + #92) that
close the user-visible "walls walk through at Holtburg inn" symptom,
but #90 specifically is a CLAUDE.md-rules workaround (sphere-overlap
stickiness on top of point-only cell containment) added without prior
approval. The underlying issue (BSP push-back distance probably
diverges from retail) hasn't been measured. Plus the umbrella #83
(indoor multi-Z walking) has been open since 2026-05-19 with multiple
aborted fix attempts; plus indoor lighting (#80 + #81 + new #93 +
#94) has been deferred as "M7 polish" but is actually part of the
same indoor-experience problem.
Promoting to a milestone of its own forces the work to be central,
retail-anchored, and complete — not another whack-a-mole patch.
Milestone M1.5 — "Indoor world feels right":
Demo: enter Holtburg Sewer through the in-town portal, navigate
through 5-7 rooms with stairs + a multi-Z chamber, exit back to
town. Walls block. Stairs work. Items block. Lighting reads
correctly. Cell transitions smooth.
Phases:
A6 — Indoor physics fidelity (cdb-driven)
A7 — Indoor lighting fidelity (RenderDoc + retail-decomp driven)
Issues in scope: #80, #81, #83, #88, #90 (workaround removal),
#93 (new lighting umbrella), #94 (held-item spotlight),
+ TryFindIndoorWalkablePlane synthesis removal.
M2 ("Kill a drudge") deferred until M1.5 lands.
This commit updates:
- docs/plans/2026-05-12-milestones.md (M1.5 block inserted, M2 moved
to deferred status)
- docs/plans/2026-04-11-roadmap.md (A6 + A7 sub-pieces detailed)
- CLAUDE.md (Currently working toward updated to M1.5, M2 paragraph
marked deferred, M1.5 baseline shipped paragraph added)
- docs/ISSUES.md (#80, #81, #83, #88, #90 tagged M1.5 scope;
new #93 indoor lighting umbrella + #94 held-item spotlight filed)
- docs/research/2026-05-21-open-items-pickup-prompt.md (landscape
table reorganized around M1.5 phases)
A6 + A7 specs to be drafted in the next session(s).
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Phase A4 (multi-cell BSP iteration) ships in three commits (e6369e2,
493c5e5, 691493e — with revert 3add110 + reapply during visual
verification that proved A4 is not the cause of the issue surfaced).
1139 + 8 baseline maintained. 10 new unit tests pass. Wires retail's
CTransition::check_other_cells (acclient_2013_pseudo_c.txt:272717-272798)
into Transition.FindEnvCollisions.
Visual verification at the Holtburg inn vestibule surfaced a separate,
pre-existing M2 blocker (filed as #90): CellId ping-pongs between
outdoor 0xA9B40022 and indoor 0xA9B40164 on every wall push-back
because the push-back exits the indoor CellBSP volume, causing the
resolver to flip back to outdoor and bypass walls on outdoor ticks.
Indoor BSP results (Collided/Adjusted/Slid all firing) prove walls ARE
detected when the player is indoor; the aggregate "walls walk through"
appearance comes from CellId classification instability, not from
collision detection.
Bug reproduces fully with A4 reverted (launch-revert2.log captured 18
cell-id flips between 0xA9B40022 ↔ 0xA9B40164, 11 inside=True
building-transit events, 61 indoor-bsp queries firing the full
result distribution). A4 is correct and tested but dormant in
practice until #90 is fixed.
Updates:
- docs/research/2026-05-20-phase-a4-shipped-cell-pingpong-finding.md (new)
- docs/plans/2026-04-11-roadmap.md (A4 shipped row added)
- CLAUDE.md (Indoor walking Phase A4 paragraph + next-step pointer
to #90 with retail oracle anchor at acclient_2013_pseudo_c.txt:308742-308783)
- docs/ISSUES.md (#90 filed, HIGH severity, M2-blocker)
- docs/research/2026-05-21-open-items-pickup-prompt.md (landscape
table updated — A4 closed, #90 promoted to top blocker)
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
After the 2026-05-21 session merged A1/A1.5/A1.6/A1.7 to main, six
discrete items remain. This doc maps them as a landscape rather than
single-phase:
- Collision (M2 critical path): A4 multi-cell BSP iteration → verify
stairs → A2 PHSP inversion → A3 synthesis removal
- Rendering (M7 polish): indoor lighting + spotlight-projection bugs
The recommended order is A4 first (biggest user payoff, unblocks A3),
then stairs verification, A2 + A3 paired, lighting in a separate
session. A3 must NOT ship before A4 — that's the Bug A regression
from 2026-05-20.
Includes a pasteable session-start prompt that the user can box into
a fresh Claude Code session.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
After merging the 2026-05-21 session into main (56d2b5e), update the
pickup prompt so the next session starts from main and creates its
own worktree for the A4 work — not the now-merged lucid-goldberg
branch.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Captures everything that shipped in the session — A1, A1.5, A1.6,
A1.7 plus the walk-miss probe spike — and what's still open:
- A4 (multi-cell BSP iteration) — the next big architectural fix,
closes the "walls walk-through-able in vestibule cells" gap
- A2 (PHSP inversion) — small fix, but only meaningful paired with A3
- A3 (synthesis removal) — needs A4 in place first to avoid
reverting back to Bug A's free-fall regression
- Lighting bugs (indoor lighting + spotlight projection) — M7 polish,
separate session
Includes per-fix commit SHAs, code anchors, retail decomp anchors,
probe + launch reference, anti-patterns, and a fresh-session pickup
prompt for boxing into Claude Code.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Three docs from the indoor walk-miss probe spike landed in commits
27c7284..a2e7a87:
- Spec: design of the [walk-miss] + [floor-polys] diagnostic emissions
with the H1/H2/H3 disambiguation matrix.
- Plan: 3-task TDD implementation plan (flag, aggregator, emissions).
- Findings: live-capture analysis showing H3 (walkable_hits_sphere /
adjust_sphere_to_plane synthesis rejection) is the dominant defect.
817 of 876 ground-contact misses (93%) cluster at dz~0.48 m, while
the 7 HITs all sit at dz~0.46 m — a 2 cm boundary between working
and broken that points at the sphere-overlap math, not the probe
distance. H1 (multi-cell iteration missing) is real but only 3%
of misses, secondary. H2 (probe distance) ruled out.
Next step: line-by-line decomp comparison of FindWalkableInternal /
walkable_hits_sphere / adjust_sphere_to_plane against retail at
acclient_2013_pseudo_c.txt:322032 / :323006 / :326793, then design
the fix in a follow-up session.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Companion to the Bug A wrong-scope handoff (35c266a). Provides the
boxed copy-paste prompt for a fresh session + quick reference for the
user and the helper:
- Branch state + KEEP/REMOVE recommendation
- Anti-patterns to avoid (don't repeat Bug A, validate risks with
probe data, stop at three failed verifications)
- Code anchors for Mechanisms A/B/C in our code
- Retail decomp anchors for the doorway investigation
- Probe + diagnostic env var menu
- 5-scenario visual verification list
- Launch command with UTF-8 conversion step
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Bug B (indoor BSP world-origin fix) shipped today at de8ffde.
Bug A (delete per-frame walkable-plane synthesis) attempted and
reverted at 0a7ce8f. Real bug is deeper than scoped:
Indoor cell floor polys don't cover the player's full XY range when
crossing thresholds (doorways). Step-down probes miss past the floor
edge, Mechanism C (post-OK step-down) can't catch the player,
ContactPlane invalidates, gravity pulls them through the void.
We have all three retail CP retention mechanisms (A, B, C). The
defect is geometry, not retention. Either dat-decoder missing some
floor polys, or cell-transition timing too late, or some retail
mechanism we haven't traced.
Handoff includes:
- State of every commit on this branch + KEEP/REMOVE recommendation
- Bug B evidence and recommendation to ship to main
- Bug A failure analysis with probe data
- Mechanisms A/B/C location in our code vs retail decomp anchors
- 5 prioritized investigation targets for fresh session
- Anti-patterns to avoid (don't repeat Bug A approach)
- Lessons learned (probe-first discipline, risk-as-falsification,
3-fails-in-a-session stop signal, Matrix4x4.Decompose idiom,
binary-timestamp paranoia)
Recommendation: merge Bug B alone, leave the rest for fresh session.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Self-contained fresh-session prompt that points at the BSP-port
shipped-handoff, summarizes the foundation work to keep vs delete,
notes the retail decomp anchors for CTransition::transitional_insert /
last_known_contact_plane, and includes the session-lesson reminder:
probe-first, design-second.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Foundation work (6 commits ff548b9..f845b22) landed but visual
verification 2026-05-19 FAILED to fix the user-reported indoor bugs.
Documenting the deeper diagnosis + the next phase target without
reverting the foundation work.
What landed (kept):
- BSPQuery.FindWalkableInternal gained ref ushort hitPolyId (Task 1).
- New public BSPQuery.FindWalkableSphere wrapper over the existing
retail-faithful walkable finder (Task 2).
- Transition.TryFindIndoorWalkablePlane refactored through it,
PointInPolygonXY deleted (Task 3).
- [indoor-walkable] runtime-toggleable probe (Task 4).
- 5 new tests + 9 updated existing tests, all green; build clean.
What didn't fix: cellar descent FAIL, 2nd-floor walking FAIL
(intermittent falling-stuck), single-floor cottage REGRESSION (was
stable, now intermittent falling-stuck), phantom collisions PERSIST.
Probe evidence: 1443 MISS / 2 HIT over 1445 calls. Smoking gun:
foot-sphere-tangent-to-floor case fails PolygonHitsSpherePrecise's
|dist| > radius - epsilon check by ~0.0002. The BSP walker is
correct; the caller (TryFindIndoorWalkablePlane) is misusing it.
Root cause (deeper than originally diagnosed): TryFindIndoorWalkablePlane
exists only as a Phase 2 commit eb0f772 stop-gap. Retail doesn't
synthesize a ContactPlane per frame — retail RETAINS the previous
frame's plane when the BSP says no collision. Retail's find_walkable
only runs inside step_sphere_down (a sweep), never as a standing-still
query.
Next phase target: port retail's ContactPlane retention so the
resolver retains state across frames. Likely eliminates the per-frame
TryFindIndoorWalkablePlane call entirely. Foundation work (BSP walker
+ probe + tests) remains useful regardless.
ISSUES #83 remains OPEN with the deeper diagnosis.
Roadmap header updated to reflect partial-ship status.
Handoff at docs/research/2026-05-19-indoor-walkable-plane-bsp-port-shipped-handoff.md.
Spec: docs/superpowers/specs/2026-05-19-indoor-walkable-plane-bsp-port-design.md
Plan: docs/superpowers/plans/2026-05-19-indoor-walkable-plane-bsp-port.md
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
User explicitly redirected the next-phase track: no M2 (kill-a-drudge), stay
on indoor walking issues, collision, physics, and dungeons. Update the
pickup prompt to reflect this:
- Drop M2 from the candidate list entirely.
- Add #83 (walking up stairs) as the recommended next phase — pure
indoor/physics, unblocks both multi-floor cottages AND dungeons.
- Add a "dungeon stress test" candidate (Path B) — verify Phase 2's
portal traversal works on multi-cell indoor spaces.
- Move indoor lighting from "recommended" to Path E with a note that it
depends on stairs (#83) landing first to be testable.
- Update the helper section with concrete file pointers per path.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
After merging Phase 1 + Phase 2 to main at 1af49b7, file a pickup prompt
that orients a fresh Claude Code session: what shipped, what's open, the
four ranked candidate next phases (indoor lighting / M2 / #88 vibration /
triage), and key file pointers for whichever path is chosen.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Closes ISSUES.md #87 + #85 + the remaining wall-pass-through portion of
#84 (fully closes#84). Portal-graph cell traversal replaces Phase D's
AABB containment. Walking through doors promotes/demotes CellId correctly
via portal traversal; walls block from inside indoor cells; indoor walkable
plane is synthesized from the cell's floor poly so the resolver tracks
walkability correctly during indoor movement.
Files two new issues: #88 (indoor static objects vibrate — pre-existing,
spotted during Phase 2 testing) and #89 (BSPQuery.SphereIntersectsCellBsp
— follow-up to make CheckBuildingTransit retail-faithful; currently uses
radius-less PointInsideCellBsp as a documented approximation).
ISSUES.md: #87, #85, #84 moved to DONE. #88 + #89 filed.
Roadmap: Indoor walking Phase 2 added to shipped table.
CLAUDE.md: recent-phase paragraph updated to reflect Phase 2 shipped.
New handoff: docs/research/2026-05-19-indoor-walking-phase2-shipped-handoff.md
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Cluster A's investigation pinned #86 (picker) as structural and closed
it (Phase B). #84 and #85 both pinned on missing indoor cell tracking;
Phase D promoted CellId via AABB containment which un-stuck the
spawn-in-building case (closes#84 partially) but proved too tight for
threshold/doorway cells to keep CellId indoor during normal walking.
The proper fix is retail's portal-based cell traversal; filed as a
new ISSUES.md issue (see body) for the follow-up phase. Phase E
diagnostic infrastructure ([cell-cache] + extended [indoor-bsp]) stays
in place as scaffolding for that work.
ISSUES.md: #86 → Recently closed. #84 status updated to PARTIAL with
resolution paragraph. #85 status update note added. New issue #87 filed
for portal-based indoor cell tracking.
Roadmap: Cluster A added to Recently shipped with partial-ship note.
Forward entry added for the portal-traversal follow-up under Phase G.
CLAUDE.md: current-phase paragraph updated to reflect Cluster A partial
ship. Next phase deferred to Claude's choice in a future session.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Phase 2's one-line WB patch (Setup-prefix guard at ObjectMeshManager.cs:1230)
fixed the symptom but is structurally a band-aid. CLAUDE.md's
no-workarounds rule says we should retire it.
The proper fix is switching our EnvCell rendering from the
general-purpose PrepareMeshDataAsync entry point (which iterates
static-object parts + emitters we don't need + triggers the buggy
TryGet<Setup> call) to WB's narrower PrepareEnvCellGeomMeshDataAsync
API at ObjectMeshManager.cs:386. That function only builds cell
room mesh — which is the only thing we use WB for at the cell
level. Static objects are already hydrated separately, particle
scripts already run via our own EntityScriptActivator.
#87 is the issue tracking that refactor. When it lands the WB fork
returns to pristine state (no acdream-specific commits on the
acdream branch for this file).
Handoff doc updated to flag the patch as a known band-aid pending #87.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
After Phase 1+2 (indoor cell rendering — missing floors fixed),
9 follow-up issues (#78-#86 in docs/ISSUES.md) need their own
phases. This handoff doc gives the next session everything it needs
to start cold: probe infrastructure status, issue cluster groupings,
suggested phase order, and the verification approach that worked
for Phase 1+2.
Companion prompt file is a self-contained kickoff that can be pasted
into a fresh Claude Code session to start work on the cluster.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
User visually confirmed floors render in Holtburg Inn after the WB
TryGet<Setup> guard. Probe re-capture: 0 [wb-error] lines (was 385),
0 NULL_RESULT (was 55), Holtburg 123/123 cells complete (was 97/123).
Documents the nine pre-existing indoor bugs the user observed during
verification (see-through floor, indoor collision, stairs, walls,
clicking, indoor lighting artifacts, stabs-don't-react-to-atmospheric-
lighting, slope terrain lighting). All pre-existing; filed for follow-up
phases via docs/ISSUES.md.
Phase 2 complete.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Phase 2 diagnostic chain identified the EXACT cause of 26/123 Holtburg
cells silently failing in WB's PrepareEnvCellMeshData:
ArgumentOutOfRangeException thrown from Setup.Unpack inside
DatReaderWriter when WB calls TryGet<Setup>(stab.Id, ...) on a stab id
whose prefix is GfxObj (0x01xxxxxx), not Setup (0x02xxxxxx).
DatReaderWriter finds the file in Portal's tree (GfxObjs and Setups
share tree-lookups), attempts to parse GfxObj bytes as Setup format,
throws OOR. Exception bubbles to PrepareMeshData's outer try/catch
which silently swallows + returns null. Entire cell fails to upload.
This commit lands the diagnostic infrastructure that surfaced the bug:
- WbMeshAdapter: replaced NullLogger<ObjectMeshManager> with a small
Console-backed ConsoleErrorLogger<T> private class. Filters to
LogLevel.Error+. WB's existing _logger.LogError(ex, ...) at the
swallow site now writes [wb-error] lines with type + message + top 5
stack frames. Bridges WB's intentional log point to acdream's console.
- WbMeshAdapter: extended [indoor-upload] NULL_RESULT probe with
reader-divergence diagnostic (ourCellDb.TryGet, wbResolveId.Count,
wbSelectedType, wbDbIsPortal, wbDbTryGet<EnvCell>, hadRenderData).
Made it possible to rule out cache-hits and reader-divergence as
causes before identifying the real one.
- Cause report at docs/research/2026-05-19-indoor-cell-rendering-cause.md
documents the full chain: 55 ArgumentOutOfRangeException stack traces
captured in one launch, all from PrepareEnvCellMeshData line 1223.
The fix itself (1-line guard at WB's TryGet<Setup> call site) is applied
to references/WorldBuilder/.../ObjectMeshManager.cs — which is a git
submodule. Will be committed separately to the WB submodule after
visual verification.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Captured at Holtburg landblock 0xA9B4 with ACDREAM_PROBE_INDOOR_ALL=1.
Result: 123 EnvCells in Holtburg get [indoor-upload] requested but
ONLY 97 get a matching [indoor-upload] completed. 26 cells silently
fail in WB's PrepareEnvCellMeshData / PrepareMeshData. The first
interior cell 0xA9B40100 — likely the inn entry or another major
building anchor — is among the failures, exactly matching the
user's "floor missing" symptom.
Other hypotheses ruled out:
- H2 (empty batches): completed cells have cellGeomVerts=14-86.
- H3 (cull bug): walk probe confirms cells pass all visibility filters.
- H4 (double-spawn): partCount values match expected SetupParts.
- H5 (transform double-apply): xform probe shows composedT==meshRefT;
no double-apply.
- H6 (MeshRefs structure): lookup probe shows isSetup=True and
partsHit≈partCount for uploaded cells.
Phase 2 plan: wrap PrepareMeshDataAsync with our own catch-and-log
in WbMeshAdapter so the swallowed exception (most likely cause of
the 26 silent failures, per WB ObjectMeshManager.cs:589) becomes
visible. Once we know the actual failure reason, target the fix.
Also flags IsEnvCellId false-positives on GfxObj IDs whose lower 24
bits ≥ 0x0100 — tightening recommended in Phase 2.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Indoor cells rendered "almost black" because the hardcoded ambient at
GameWindow.cs:8342-8345 was an early-2026 guess (0.10, 0.09, 0.08 — half
retail brightness, warm-tinted) rather than the retail value. The named
retail decomp (acclient.pdb, Sept 2013 EoR build) shows
CellManager::ChangePosition @ 0x004559B0 calls
SmartBox::SetWorldAmbientLight(0.2f, 0xFFFFFFFF) whenever the player's
CObjCell::seen_outside flag is 0 — a flat 0.20 white floor, not a
dungeon-tone warm color.
Investigation also confirmed:
- EnvCell.dat does NOT carry inline lights — CEnvCell::UnPack reads
numVisibleCells where Binary Ninja's heuristic decomp inferred
"num_lights". Retail's CObjCell.light_list is populated at runtime via
add_light() calls from neighbouring cell light registrations + per-cell
static-object Setup.Lights, NOT from the dat byte stream.
- Setup.Lights from indoor static objects (entity.SourceGfxObjOrSetupId
prefix 0x02xxxxxx) DO flow through LightInfoLoader.Load (line 5765)
and reach LightManager via LightingHookSink. The wire is intact; the
per-frame Tick + UBO upload chain (line 6865-6867) is intact.
- Retail's particle system does NOT emit lights from particles themselves.
The light comes from the owning Setup's LightInfo records.
Pre-existing failures in DispatcherToMovementIntegrationTests, BSPStepUpTests,
and MotionInterpreterTests are on the branch already and unrelated to this
change (verified by stashing + retesting).
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Self-contained handoff doc for a follow-up focused session that fixes#77.
Captures: the two bugs (NPC at walking range never auto-walks; pickup at
walking range overshoots and snaps back), the trace evidence from Step 2's
verification run (cmd=0x0005 speed=-1.84 from ACE), four ranked
root-cause hypotheses (H1 missing BeginServerAutoWalk fire / H2
walk-run-threshold misclassification / H3 negative ForwardSpeed
sign-interpretation / H4 arrival predicate firing too early), the
reproduction recipe with ACDREAM_PROBE_AUTOWALK=1, acceptance criteria,
and the "don't workaround, fix root cause" guardrail.
The auto-walk diagnostic infrastructure already exists from Phase B.6
work — the next session just turns it on and reads the trace.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
User chose fundamentals-first after M1 landed: indoor walking is
untested (M1 only verified outdoor + doorway), indoor lighting is
broken, camera clips into walls indoors. Better to fix indoor
foundations before stacking combat on top.
Three sub-phases proposed for the new M2:
1. Camera correctness (~1 day)
2. Indoor collision audit (~1 week)
3. Indoor lighting basics (~1-2 weeks)
Combat ("Kill a drudge") slides to M3.
Next session opens with superpowers:brainstorming to scope the
demo scenario + agree on sub-phase boundaries + update the
milestones doc + CLAUDE.md to formalize the reorder.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Captures M1 landing, Phase B.6 architectural details, new rules
(no-workarounds, no-demo-videos, graceful-shutdown), M2 next steps,
and test baselines for fresh-session pickup.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
