erik/acdream
1
0
Fork 0
Code Issues Pull requests Projects Releases 6 Packages Wiki Activity Actions

docs(issues): #458 — the block z slab matches retail on both paths; suspects narrowed to the block edge-plane test / view union

Browse source 
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
This commit is contained in:
Erik 2026-09-03 12:37:05 +02:00
parent b1eaab50d9
commit 081a633419
1 changed files with 8 additions and 0 deletions
Download patch file Download diff file Expand all files Collapse all files

8
docs/ISSUES.md
View file

@ -68,6 +68,14 @@ edge planes); (2) retail tokens 1093–1107: two look-in floods acdream never en
the FW0 `Doorway_still_first_frame_diff` was a diff printer, never an exact pin), now
exposed by the eight-kind comparison. Signature lengths: retail 1185 tokens, acdream 1170.
**Z-range note (2026-09-03):** the block's z slab is NOT the difference — retail pads
`max_zval = Land_Height_Table[max] + 200`, `min_zval = table[min] − 1` (`CLandBlock` @0x0052f297/
0x0052f2aa) and BOTH acdream paths mirror it (`LandblockBuildFactory.ComputeWalkZSlab` for
production, `WalkLandscapeDatBuilder` for the replay). Remaining suspects for the `a9c9`
admission: the per-view block edge-plane test (`WalkVisibilityMath.BlockCheck` against
`LScape::draw_check_blocks` @0x00505f80's clip-height grid — BN elides the 192 m block
interval to `0f` there) and the exit-view union at this pose (two surviving views).
**Degrade note (2026-09-03 review):** the FW0 sibling row for the SAME pose pins
`RetailFrameWalk.DegradeMultiplier = 0f` because that capture ran under cdb load with
`Render::deg_mul` depressed; the OH doorway capture also ran under cdb, so check the