docs(meta): D-255 — the ladder floor is one subtile per pixel (T-1213)

Measured during T-1213, through the production canvas builder on Ferrath: at
Quarter and below, morphology collapses to ONE zone and vegetation to ONE
class. The uniform frames in the 2026-08-06 descent ladder were those rungs
drawing exactly what they contain.

The cause is arithmetic, not a missing feature. At the uniform 2x2 px display
ratio a 3440x1440 window gives 540 gridunits on the short axis, so:

  Quarter  512 m  ->  0.948 m/gridunit  ->  0.474 m/px   ~1 subtile per pixel
  Block    128 m  ->  0.237 m/gridunit                    4 gridunits per voxel
  Chunk     64 m  ->  0.119 m/gridunit                    8 gridunits per voxel

Block and Chunk magnify beneath the finest datum that can exist, so they can
only ever draw one voxel larger. Quarter lands within 5% of one subtile per
pixel and becomes the floor.

Stated as a rule so it survives the constants moving: the deepest Atlas rung
is the one at which a screen pixel shows one subtile. It is derived from the
data model rather than chosen, and it moves automatically if the subtile does.

WHAT THIS IS NOT. Chunk remains the 64 m stream/derive unit of D-243 and stays
vital — it is what Phase 5 derives first-person walkable content on, and
D-012's load-around-the-player is expressed in chunks. Block remains the 128 m
generator planning unit. Only Atlas VIEWABILITY is retired; the containment
ladder is untouched. This record governs what the map draws, not what the
generator builds.

The justification is the Atlas's purpose (Jeroen): it exists to give the player
information, and a rung earns its place by answering a question the rung above
cannot. Once a pixel is a subtile there is no finer datum to answer with.

The resulting Global -> Region -> District -> Quarter steps at ~93x -> 100x ->
4x. That unevenness is NOT from this change — the rungs removed were 4x and 2x
steps carrying no information — it is D-243's one non-power-of-2 rung, and
T-1218 already exists to re-balance it. A compensating rung above Region was
considered and declined here; it belongs with that ticket.

CLAUDE.md's cascade line updated in the same commit.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
This commit is contained in:
2026-08-07 14:05:41 +02:00
co-authored by Claude Opus 5
parent 6beb4bbea1
commit b429f633e6
4 changed files with 95 additions and 1 deletions
+22
View File
@@ -2298,6 +2298,28 @@ Technical foundation decisions that constrain implementation: engine, client-ser
**Status, honestly:** Region is confirmed working. **Global still reads wrong and this amendment does not fix it** — it is correctly *sized* now but still rides the orbital derive, so it carries no hydrology. Its repair is [D-258](#d-258)'s rung 0.5, not further work on this record.
- **Amendment (2026-08-07, Jeroen's ruling — THE LADDER FLOOR: the deepest rung is one subtile per pixel; Block and Chunk stop being viewable rungs):** The rung set narrows from five fixed rungs to **three** — Region → District → **Quarter** — beneath the Global opener. Ruled on measurement, during T-1213.
**The rule, stated once:** *the Atlas's deepest rung is the one at which a screen pixel shows one subtile.* Below that a rung is not showing more, only bigger. This replaces "chunk (64 m) is the deepest Atlas rung" from the 2026-07-13 D-226 narrowing.
**The arithmetic.** The uniform display ratio is 2×2 px per gridunit (item 4 above), and a 3440×1440 window yields 540 gridunits on the short axis. With `VOXEL_M = 1` and the subtile at 2⁻¹ = 0.5 m:
| Rung | Extent | m/gridunit | m/pixel | Against the finest datum |
|------|--------|-----------|---------|--------------------------|
| **Quarter** | 512 m | **0.948** | **0.474** | ≈1 voxel per gridunit, **≈1 subtile per pixel** |
| Block | 128 m | 0.237 | 0.119 | 4 gridunits per voxel |
| Chunk | 64 m | 0.119 | 0.059 | 8 gridunits per voxel |
Quarter lands within 5% of the rule. Block and Chunk magnify *beneath the finest thing that can exist*, so they can only ever draw one voxel larger. Measured on Ferrath through the production canvas builder (`t1213_ladder_field_variance`): at Quarter and below, morphology collapses to **1 zone** and vegetation to **1 class** — the uniform frames in the 2026-08-06 descent ladder were those two rungs rendering exactly what they contain, which is nothing new.
**What this is NOT.** **Chunk remains the 64 m stream/derive unit of [D-243](#d-243), untouched, and stays vital** — it is the unit Phase 5 derives first-person walkable content on, and [D-012](#d-012)'s load/unload-around-the-player is expressed in chunks. Only its *viewability as an Atlas map rung* is retired. Block likewise remains the generator's 128 m planning unit. Nothing in the containment ladder changes; this record governs what the **map** draws, not what the **generator** builds.
**Why the floor is principled rather than a tuning choice.** The Atlas exists to give the player **information** about a body (Jeroen: *"the goal here is to provide the user information, while implementing the cascaded generators to make that possible in the first place"*). A rung earns its place by answering a question the rung above cannot — continents, provinces, a district's terrain, a quarter's ground. Once a pixel is a subtile, every further step answers nothing new, because no finer datum exists to answer with. The floor is therefore derived from the data model, not from taste, and it moves automatically if the subtile ever does.
**On the resulting ladder shape.** Global → Region → District → Quarter steps at roughly 93× → 100× → 4×. The unevenness is **not** caused by this amendment (the rungs removed were 4× and 2× steps carrying no information); it is [D-243](#d-243)'s single non-power-of-2 rung, and [T-1218](../../.pql) already exists to re-balance it (region → 262,144, district → 4,096, making that step 64×). A compensating rung above Region was considered and **declined here** — it belongs with that ticket's re-balancing, not as an ad-hoc addition.
**Consequential edits:** CLAUDE.md's Phase-4 cascade line (which named chunk as the deepest Atlas rung) is updated with this record. [D-226](#d-226)(d)'s prohibition is unaffected and now has margin — it caps Atlas granularity at settlement/quarter, and Quarter is exactly that cap.
---
### D-256: Canonical sampling convention — one absolute-metre derive core; the batch layer is a survey raster