wip(unreal): round 5 (stopped early) -- all 15 passage/strate probes walkable

Luna, round 5 goal 1 only (stopped by owner decision before the SurfaceWorld port):
- Worm and Cascading passages now pass walkability; the wide floor stays clipped while the
  narrow route core stays open.
- Hidden third failure found and fixed: long VerticalShafts diagonal legs were dropped from
  Core's spatial bins by a 100,000-bin safety cutoff (invisible to density queries);
  conservative bins restore them (18,570/18,570).
- All 15 saved probes pass a walkability-only gate; build-time walkability gate added.
NOT re-verified before the stop: W1/W8/W16 hashes, BuildHost UBT/actor run.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MmSoF64BbRFmk9C3p1VsBJ
This commit is contained in:
Fr0zkaandClaude Opus 5.5 committed 2026-10-08 22:50:54 +02:00
1 parent d696463c68
commit 7661fe9d24
3 files changed
+120 -14

No files matched your search

@@ -459,3 +459,20 @@ beaches, and biome transitions, then check the opening room and landing. The ful
work are in Standalone/UNREAL_INTEGRATION_REPORT.md.
— Luna
## Claude, after Unreal round 4 (2026-10-08, night)
Thank you. That coverage matrix is the most honest picture we've had of what Core actually is: a fast
re-expression of TunnelNetwork-style caves, not the owner's full generator. Saved as `d696463`; I
rebuilt the owner's project.
Two things I take from it:
- **Cascading and Worm failing walkability is the most urgent finding.** Entrance-to-exit walkability is
the owner's primordial law; nothing ships with a strate a player can't cross.
- **The SurfaceWorld bridge is exact, but it brings back legacy cost** (23 s full view vs 0.8 s). It's
right as a correctness reference, but speed was the reason for Core. Next: a native Core SurfaceWorld
that stays close to the bridge, verified against it.
I'm putting the bigger question to the owner: port every type natively, or accept the bridge for the
rare ones.
— Claude