From 14820991cab04f1a513db6e0a9405a884079d94b Mon Sep 17 00:00:00 2001 From: Jan Petykiewicz Date: Sun, 19 Apr 2026 10:59:42 -0700 Subject: [PATCH] Narrow tier2 kind-8 availability path --- ...ect-kind8-tier2-named-availability-note.md | 82 +++++++++++++++++++ docs/rehost-queue.md | 14 ++++ 2 files changed, 96 insertions(+) create mode 100644 artifacts/exports/rt3-1.06/runtime-effect-kind8-tier2-named-availability-note.md diff --git a/artifacts/exports/rt3-1.06/runtime-effect-kind8-tier2-named-availability-note.md b/artifacts/exports/rt3-1.06/runtime-effect-kind8-tier2-named-availability-note.md new file mode 100644 index 0000000..f747e9e --- /dev/null +++ b/artifacts/exports/rt3-1.06/runtime-effect-kind8-tier2-named-availability-note.md @@ -0,0 +1,82 @@ +# Runtime Effect Kind-8 Tier2 Named Availability Note + +This note records the current named-candidate availability comparison between the strongest +title-overlap pair: + +- `Louisiana.gmp` +- `Dutchlantis.gmp` + +The comparison uses the checked `rt3_105_save_name_table_probe` surface from `inspect-smp`. + +## Immediate Result + +`Warehouse05` is **not** the differentiator. + +Observed values: + +- `Warehouse05` + - `Louisiana.gmp = 1` + - `Dutchlantis.gmp = 1` +- `Port01` + - `Louisiana.gmp = 1` + - `Dutchlantis.gmp = 1` +- `Furniture Factory` + - `Louisiana.gmp = 1` + - `Dutchlantis.gmp = 1` +- `FarmGrain` + - `Louisiana.gmp = 1` + - `Dutchlantis.gmp = 1` +- `Logging Camp` + - `Louisiana.gmp = 1` + - `Dutchlantis.gmp = 1` + +So the strongest current `Louisiana -> Dutchlantis` title overlap does not also come with a unique +`Warehouse05` availability toggle that would directly explain the one-row +`Add Building Warehouse05` dispatch cluster in `Louisiana.gmp`. + +## Full Named-Availability Differences + +The current comparison shows `18` differing names: + +- `AluminumMill`: `Louisiana = 0`, `Dutchlantis = 1` +- `AutoPlant`: `Louisiana = 0`, `Dutchlantis = 1` +- `Bauxite Mine`: `Louisiana = 0`, `Dutchlantis = 1` +- `Chemical Plant`: `Louisiana = 0`, `Dutchlantis = 1` +- `Farm Corn`: `Louisiana = 1`, `Dutchlantis = 0` +- `FarmCotton`: `Louisiana = 1`, `Dutchlantis = 0` +- `FarmRice`: `Louisiana = 1`, `Dutchlantis = 0` +- `FarmSheep`: `Louisiana = 0`, `Dutchlantis = 1` +- `FarmSugar`: `Louisiana = 1`, `Dutchlantis = 0` +- `Fertilizer Factory`: `Louisiana = 0`, `Dutchlantis = 1` +- `Iron Mine`: `Louisiana = 0`, `Dutchlantis = 1` +- `Nuclear Power Plant`: `Louisiana = 0`, `Dutchlantis = 1` +- `Oil Refinery`: `Louisiana = 1`, `Dutchlantis = 0` +- `Plastics Factory`: `Louisiana = 0`, `Dutchlantis = 1` +- `Recycling Plant`: `Louisiana = 0`, `Dutchlantis = 1` +- `Steel Mill`: `Louisiana = 0`, `Dutchlantis = 1` +- `Tool And Die`: `Louisiana = 0`, `Dutchlantis = 1` +- `Uranium Mine`: `Louisiana = 0`, `Dutchlantis = 1` + +## Current Implication + +The Tier 2 candidate/world-state rebuild strip remains plausible in general because it owns named +candidate availability seeding and refresh, but the current evidence no longer supports the narrow +claim that `Louisiana.gmp` gets its `Add Building Warehouse05` row from a unique +`Warehouse05` availability difference relative to `Dutchlantis.gmp`. + +Two deeper grounded notes tighten that further: + +- `structure_candidate_collection_rebuild_runtime_records_from_scenario_state` `0x00412d70` + does **not** consult the scenario-side recipe-book name at `[state+0x0fe8]`; it formats the + rebuilt candidate display stem from one of two fixed built-in roots `0x005c93d8` or + `0x005c93e8` plus the current ordinal. +- `scenario_state_upsert_named_candidate_availability_record_and_refresh_runtime_filters` + `0x00434f20` now reads safely as a real boolean availability override writer rather than a wider + mode enum. + +So the next Tier 2 recovery question is now tighter still: + +- whether the relevant difference is in broader candidate-state rebuild sequencing, latch refresh, + or cargo-membership/runtime-record reconstruction under + `0x00437737 / 0x00412c10 / 0x00412bd0 / 0x00412d70 / 0x00412fb0`, + rather than in the direct named availability bit for `Warehouse05` itself. diff --git a/docs/rehost-queue.md b/docs/rehost-queue.md index 8d5f9d1..3f1f39a 100644 --- a/docs/rehost-queue.md +++ b/docs/rehost-queue.md @@ -481,6 +481,20 @@ Working rule: port/warehouse recipe rebuild `0x00435630`, and only then the later world bringup / event-service neighborhood `0x004384d0 / 0x00443a50 / 0x00432f40`. + - the strongest same-stem title pair has a Tier 2 availability note now too: + `artifacts/exports/rt3-1.06/runtime-effect-kind8-tier2-named-availability-note.md` + shows `Louisiana.gmp` and `Dutchlantis.gmp` share the direct named-availability bit for + `Warehouse05` (`1/1`), as well as `Port01`, `Furniture Factory`, `FarmGrain`, and + `Logging Camp`; their current `18` named-availability differences fall elsewhere + (`Bauxite Mine`, `AluminumMill`, `Farm Corn`, `FarmCotton`, `FarmRice`, `FarmSugar`, etc.). + The same note now folds in two deeper grounded constraints too: + `0x00412d70` does not consult the scenario-side recipe-book name at `[state+0x0fe8]`, and + `0x00434f20` writes only a boolean availability override bit. So the current Tier 2 question + is no longer “is `Warehouse05` simply toggled differently?” or “is a recipe-book display name + leaking through?” but + “which broader candidate-state rebuild or latch sequencing difference above + `0x00437737 / 0x00412c10 / 0x00412bd0 / 0x00412d70 / 0x00412fb0` leads to the shipped + `Add Building Warehouse05` row in `Louisiana.gmp`?” kinds”; it is the smaller set of scenario-specific records where that sweep explicitly writes `[event+0x7ef]` itself or a still-later owner does. - two explicit trigger-kind materializations are now grounded inside that retagger: