fix(rendering): port retail shared alpha list

Queue translucent world GfxObj batches and scene particles in one stable far-to-near stream using transformed DAT sort centers, then drain it at retail's landscape and final-world boundaries. Preserve authored blend, cull, lighting, opacity, and adjacent-only batching so particles behind lifestones are composited through the crystal instead of overpainting it.

Release build succeeds and all 5,914 tests pass with five intentional skips.

Co-authored-by: OpenAI Codex <codex@openai.com>

docs/architecture/acdream-architecture.md

@@ -220,6 +220,8 @@ src/
       TerrainModernRenderer.cs -> mandatory bindless+MDI terrain path
       Wb/WbDrawDispatcher.cs  -> ordinary live/static entity draw dispatch
       Wb/EnvCellRenderer.cs   -> indoor cell-shell draw path
+      ParticleRenderer.cs     -> DAT particle billboard/full-mesh dispatch
+      RetailAlphaQueue.cs     -> shared delayed world alpha ordering/flush owner
       TextureCache.cs         -> done
       ChaseCamera.cs          -> done
       FlyCamera.cs            -> done
@@ -512,6 +514,10 @@ matches retail's `CPhysicsObj::DoObjDescChangesFromDefault` behavior.
    └── Read current entity mesh refs, draw
        TerrainModernRenderer + WbDrawDispatcher + EnvCellRenderer
        (frustum cull, translucency pass, portal visibility, etc.)
+       World Wb translucents + Scene particles submit to RetailAlphaQueue;
+       RetailPViewRenderer flushes the landscape scope before the conditional
+       depth clear and GameWindow flushes the final scope before private
+       viewports/UI. Sky/private viewport particles remain immediate.
        Projectiles remain ordinary live-entity draws; there is no global or
        projectile-specific render pass.
 

docs/architecture/retail-divergence-register.md

@@ -129,7 +129,7 @@ AP-94..AP-112 for the confirmed retail-UI completion gaps.
 | ~~AP-18~~ | **RETIRED 2026-07-10 — faithful retail radar port.** Exact `RGBAColor_Radar*` floats were recovered from named static data and `gmRadarUI::GetBlipColor` was re-ported with `_blipColor` overrides plus portal/vendor/attackable-creature/admin/PK/PKLite/free-PK/fellowship precedence. The old implementation was not merely hue-tuned: it also had wrong portal/vendor colors and an incomplete dispatch matrix. | `src/AcDream.Core/Ui/RadarBlipColors.cs` + radar classification tests | — | — | `gmRadarUI::GetBlipColor` 0x004d76f0; static RGBA initializers at named decomp pc:1089736-1089804 |
 | AP-19 | `PortalSideEpsilon` 0.01 (≈1 cm) instead of retail F_EPSILON ≈ 0.0002 — a documented render-root-lag tolerance, NOT a retail constant. DO-NOT-RETRY: T2 (BR-4) tried the retail value; CornerFloodReplay refuted it | `src/AcDream.App/Rendering/PortalVisibilityBuilder.cs:49` | Retail's tight epsilon only works with eye-exact swept curr_cell tracking; our viewer cell lags the eye by up to ~1 cm at pressed corners. Tighten after the #108-membership family + cdstW near-clip pin land | A 1 cm misclassification band at portal planes can flood or cull a portal the eye hasn't crossed — one-frame leaks / grey flashes at knife-edge doorway/corner positions | F_EPSILON @0x007c8c70; `PView::InitCell` 0x005a4b70 |
 | AP-20 | Sub-pixel view-polygon vertex merge fixed at 1080p-reference NDC units (2/1080); retail merges at ~1 actual screen pixel | `src/AcDream.App/Rendering/PortalProjection.cs:179` | Unit approximation whose coarseness only strengthens convergence — the merge is the flood's fixpoint floor (replaced MaxReprocessPerCell=16) | At 4K+ a legitimately visible 1–2 px sliver aperture collapses to degenerate and rejects — a thin/distant doorway stops admitting its flood slightly earlier than retail | `Render::copy_view` 0x0054dfc0 |
-| AP-21 | Entity translucency: two-pass alpha-test (N.5 Decision 2, invented 0.95/0.05 thresholds); AlphaBlend + Additive + InvAlpha all composite under (SrcAlpha, 1−SrcAlpha) — retail applies per-surface D3D blend incl. true additive. EnvCellRenderer + ParticleBatcher DO switch to additive; divergence confined to GfxObj/Setup entities via WbDrawDispatcher | `src/AcDream.App/Rendering/Wb/WbDrawDispatcher.cs:1563` (+ `Shaders/mesh_modern.frag:10`; #52 amendment removed the α≥0.95 discard) | Matches original WB's model; keeps the bindless MDI pipeline at two indirect draws; spec §6 documents the falsifiable fallback — a third indirect call with `glBlendFunc(SrcAlpha, One)` (~30 min) on a magic-content regression | Additive glow/magic entity surfaces composite darker / occlude instead of brightening — the predicted regression once spell VFX density increases; α<0.05 discard drops faint fringes retail blends | SurfaceType.Additive → D3DBLEND_ONE per-surface routing |
+| AP-21 | Entity translucency retains the invented α<0.05 fragment discard. World GfxObj/Setup instances now apply their DAT AlphaBlend/Additive/InvAlpha factors through the retail shared alpha queue, but sealed off-screen WbDrawDispatcher consumers (paperdoll/UI Studio) retain the old immediate normal-alpha pass for all three kinds | `src/AcDream.App/Rendering/Shaders/mesh_modern.frag`; `src/AcDream.App/Rendering/Wb/WbDrawDispatcher.cs` (`DrawDeferredAlphaBatch` versus immediate Phase 8) | World presentation needed exact per-surface blend for spell/particle density and translucent-object intersections. Off-screen object previews are isolated render targets and have not shown an authored additive entity surface that justifies splitting their compact immediate pass | A faint world fringe below 5% alpha is discarded; a hypothetical additive/inverse-alpha paperdoll or UI Studio entity composites darker than retail inside that private viewport | `D3DPolyRender::SetSurface`; `D3DPolyRender::RenderMeshSubset`; SurfaceType.Additive → D3DBLEND_ONE |
 | AP-22 | Invented `setup.Radius` cylinder (height = Height or Radius×2) for shapeless live entities; shape + height formula not from the retail shape walk | `src/AcDream.App/Rendering/GameWindow.cs:3250` | ShadowShapeBuilder (faithful walk) only emits CylSphere/Sphere/Part-BSP; the legacy cylinder preserves prior behavior so rare decorative props don't lose collision | Those props collide with an invented footprint (especially the Radius×2 height guess) — slides/blocks at non-retail distances | `find_obj_collisions` → `CPartArray::FindObjCollisions` pc:286236 |
 | AP-23 | Invented per-type use-radius heuristic (3 m creatures / 2 m doors-lifestones-portals-corpses / 0.6 m rest) for close-range gating + the speculative local TurnToObject/MoveToObject install through the player's MoveToManager (R4-V5; retail's client use flow issues the same manager calls, §9a/§9b). **R5-V3 narrowed it: the speculative install now threads the target's REAL setup radius/height (`GetSetupCylinder`, same as the wire mt-6 route) and the player's own radius is real — only the use-radius BUCKETS remain invented** (retail reads the object's UseRadius property) | `src/AcDream.App/Rendering/GameWindow.cs` (`InstallSpeculativeTurnToTarget`) | ACE broadcasts nothing actionable on the close branch (WithinUseRadius shortcut); the true radius arrives only on the far MoveToObject branch — a local stand-in is required | A target whose real UseRadius differs from the bucket misjudges the gate — Use/PickUp deferred for a completion that never comes, or fires early into a server "too far" | ACE Player_Move.cs:66; wire MoveToObject (type 6) carries the true radius; `CPhysicsObj::TurnToObject/MoveToObject` callers §9a/§9b |
 | ~~AP-24~~ | **RETIRED 2026-07-11** — matching v11.4186 x86 disassembly recovered `ATTACK_POWERUP_TIME=1.0` seconds and `DUAL_WIELD_POWERUP_TIME=0.8` seconds from the operands loaded by `GetPowerBarLevel`; jump and combat now share those constants. | `src/AcDream.Core/Combat/CombatModel.cs`; `src/AcDream.App/Input/PlayerMovementController.cs`; `src/AcDream.App/Combat/CombatAttackController.cs` | — | — | `ClientCombatSystem::GetPowerBarLevel @ 0x0056ADE0`; static data `0x007CEFC8/0x007CEFD0` |
@@ -141,7 +141,7 @@ AP-94..AP-112 for the confirmed retail-UI completion gaps.
 | AP-31 | Scenery placement drift + the 0xA9B1 road-edge tree — WB-upstream divergences from retail, ACCEPTED (**#49/#50**, 2026-05-11) | `src/AcDream.Core/World/SceneryGenerator.cs` (via `WbSceneryAdapter`) | Piecemeal patching against WB upstream is net-negative (the `e279c46` road-check attempt over-suppressed scenery elsewhere, reverted `677a726`); visible impact = a handful of trees a few meters off | The same WB-upstream class could hide a *larger* placement divergence elsewhere; revisit only via a coherent ACME-style per-vertex filter port | `CLandBlock::get_land_scenes`; ACME GameScene.cs:1074 per-vertex road filter |
 | AP-32 | Cell shells DRAW +0.02 m above the dat EnvCell origin (`ShellDrawLiftZ`, z-fight vs coplanar terrain); retail draws at the origin verbatim. Split invariant: PHYSICS + visibility graph UNLIFTED (f35cb8b, **#119**-residual), every DRAW-space consumer of portal/cell geometry LIFTED (OutsideView color gate via `Build(drawLiftZ)`, seal/punch fans — **#130**) | `src/AcDream.App/Rendering/GameWindow.cs:5604` (const at `PortalVisibilityBuilder.ShellDrawLiftZ`) | Shell floors coplanar with terrain z-fight in our z-buffered frame; the 2 cm lift is the documented stand-in | A new draw-space consumer of portal/cell polygons that forgets the lift re-opens a 2 cm seam at horizontal aperture edges (the #130 top-edge strip, ~7 px at 2.4 m); a visibility consumer that picks up the LIFTED transform re-opens the #119-residual horizontal-portal side-cull | retail draws cell geometry at the dat EnvCell origin (no lift) |
 | AP-33 | Interior-root look-in cells (**#124** sub-pass) draw their statics + DYNAMICS + emitters WHOLE — no per-part/per-object viewcone check; retail viewconeCheck's each vs the installed view (the **#131** portal closure: a server object in a look-in cell drew nowhere — dynamics-last culls cells absent from the main cone, and post-seal it z-fails anyway) | `src/AcDream.App/Rendering/RetailPViewRenderer.cs` (`DrawBuildingLookIns`) | The main viewcone has no entries for look-in cells; over-include is the safe direction (z-correct, repainted outside apertures by the root's shells); look-in cell counts are small (~1-3 cells) | A few wasted draws on content outside the doorway region (repainted); no under-draw direction remains | `viewconeCheck` 0x0054c250; nested `DrawCells` objects pc:432878 |
-| AP-34 | Landscape-stage alpha deferral is a TWO-PHASE slice split (statics-early / dynamics+particles+weather-late around the **#124** look-ins) + outdoor-root attached scene emitters moved to the post-frame pass, not retail's single deferred alpha flush. Residual: building exteriors' / outside-stage dynamics' own translucent MESH batches still draw within their stage draw call (before later stage content) | `src/AcDream.App/Rendering/RetailPViewRenderer.cs` (`DrawLandscapeThroughOutsideView` late loop) + `GameWindow` post-frame Scene pass | The MDI dispatcher draws translucency inside each Draw call; a faithful FlushAlphaList port needs a global deferred alpha list across all landscape draws — the split covers the user-visible cases (#131 portal swirl, #132 candle flame indoors + outdoors) | Translucent landscape content drawn early and screen-overlapped by content drawn later in the stage gets overpainted (no depth self-protection) — the portal-swirl/candle-flame class re-appears in the residual configurations | `D3DPolyRender::FlushAlphaList` (DrawCells pc:432722) |
+| AP-34 | The world now shares one delayed alpha queue across Wb GfxObj/Setup entities and scene particles and drains it at retail's landscape/final boundaries. Residual: the modern reconstruction uses one stable scope-global CYpt sort rather than retail's per-`CPartCell` `CShadowPart` sort followed by cell traversal; `EnvCellRenderer` transparent shell batches also remain immediate and outside this queue | `src/AcDream.App/Rendering/RetailAlphaQueue.cs`; `RetailPViewRenderer.cs` (`FlushLandscapeAlpha`); `Rendering/Wb/WbDrawDispatcher.cs`; `ParticleRenderer.cs` | The mandatory modern renderer no longer owns retail `CPartCell` shadow lists. The shared queue restores the material consequence that motivated the port—particles and ordinary translucent parts can interleave—without rebuilding a second scene graph; stable sequence retains authored order on equal CYpt | Transparent objects from different cells can exchange order at a narrow overlap compared with retail cell traversal; an alpha-blended EnvCell shell cannot interleave with a particle or Wb entity, so those rare overlaps can still overpaint differently | `RenderDeviceD3D::DrawObjCellForDummies` 0x005A0760; `CShadowPart::insertion_sort` 0x006B5130; `D3DPolyRender::FlushAlphaList` 0x0059D2E0; `PView::DrawCells` 0x005A4840 |
 | AP-36 | Dungeon streaming gate triggers on the player's CURRENT cell being a sealed EnvCell (`CurrCell.IsEnv && !SeenOutside`), an approximation of ACE's full landblock `IsDungeon` (all-heights-zero + NumCells>0 + Buildings.Count==0). The retail BEHAVIOR (a dungeon loads no adjacent landblocks) is faithful — only the runtime TRIGGER is the cheap cell predicate instead of classifying the center landblock. **#135 pre-collapse:** at login/teleport the same collapse is triggered EARLY (the instant the streaming center is recentered onto the spawn/dest cell) via `IsSealedDungeonCell` reading the EnvCell **dat** `SeenOutside` flag — because the physics `CurrCell` is null until placement, which waits for hydration; without the early trigger the full 25×25 ocean-grid window loads then unloads (the ~30 s login FPS ramp). **#215 cell identity:** the pre-collapse/recenter decision compares the player's current `Position.objcell_id` landblock with the received destination `objcell_id`; it never reconstructs the source from XYZ because dungeon frame origins may be negative. **#145/#138 teleport-hold suppression:** during a teleport arrival HOLD the player is unplaced, so `CurrCell` is the frozen SOURCE cell, not the destination; the gate is suppressed for the hold (`DungeonStreamingGate.Compute(isTeleportHold:true)` → not-inside-dungeon) so a teleport OUT of a dungeon follows the destination (the PortalSpace observer pin) and `ExitDungeonExpand`s, instead of re-pinning streaming onto the source dungeon (which left the outdoor destination un-hydrated → 600-frame readiness timeout → force-snap to ocean — the #145 "second teleport does nothing" + #138 incomplete-world) | `src/AcDream.App/Streaming/TeleportLandblockTransition.cs` (source/destination cell-ID classification) + `src/AcDream.App/Streaming/DungeonStreamingGate.cs` (`Compute` — per-frame predicate + teleport-hold suppression, called from `GameWindow.OnUpdate` ~:7401) + `GameWindow:IsSealedDungeonCell` + `:OnLiveEntitySpawnedLocked`/`:OnLivePositionUpdated` (login/teleport pre-collapse hooks) + `src/AcDream.App/Streaming/StreamingController.cs` (collapse/expand/`PreCollapseToDungeon`) | The predicate is already computed for sun/sky gating (playerInsideCell) and exactly matches for sealed dungeons vs windowed building interiors (SeenOutside=true → not gated); no landblock re-classification needed. The dat-flag read is the same `EnvCellFlags.SeenOutside` the hydrated `ObjCell.SeenOutside` is built from (`EnvCell.cs:72`/`PhysicsDataCache.cs:224`), so the pre-collapse decision matches the eventual per-frame gate exactly. The cell-ID comparison matches retail's complete `Position` flow. | A dungeon cell that reports SeenOutside (an entrance cell open to the surface) briefly un-collapses and re-streams the window; a hypothetical windowless building back-room (IsEnv && !SeenOutside but HasBuildings) would wrongly collapse its outdoor neighbors; a sealed-dungeon entrance cell that is itself SeenOutside is simply MISSED by the early trigger and falls back to the existing late collapse (no worse than before #135) | ACE `LandblockManager.GetAdjacentIDs` (dungeons→empty) Landblock.cs:577-582; `IsDungeon` Landblock.cs:1264-1277; retail `SmartBox::TeleportPlayer` 0x00453910 |
 | AP-43 | Per-object torch (point/spot) lighting AND sun are both gated on the OBJECT's own cell via the same `IndoorObjectReceivesTorches(ParentCellId)` predicate (`(id & 0xFFFF) >= 0x0100`): indoor objects (EnvCell-parented) get torches + NO sun; outdoor objects get the SUN + ambient + NO torches. This is the faithful per-draw port of retail's `useSunlight` gate — `DrawMeshInternal` (0x0059f398) calls `minimize_object_lighting` only `if (Render::useSunlight == 0)`, and `PView::DrawCells` (0x005a4840) calls `useSunlightSet(1)` (0x005a485a) for the outdoor stage and `useSunlightSet(0)` (0x005a49f3) for the interior-cell stage. **#142 (2026-06-20):** the sun gate is now PER-INSTANCE in the shader (binding=6 `instanceIndoor[]` flag in `mesh_modern.vert`, filled by `AppendCurrentLightSet`) — it was previously a per-FRAME global keyed on the PLAYER cell (`UpdateSunFromSky`). The per-frame global is retained for sealed dungeons (correctly kills the sun frame-wide when no sky is visible). **Residual:** the `ebp_2` second seen-outside test in `CellManager::ChangePosition` (0x004559B0) is unaudited — unclear whether it changes the ambient/sun regime for a subset of cells. No observed behavioral impact in tested cells. | `src/AcDream.App/Rendering/Wb/WbDrawDispatcher.cs` (`IndoorObjectReceivesTorches`, `ComputeEntityLightSet`, `AppendCurrentLightSet`, `_instIndoorSsbo`/`_indoorData`/`InstanceGroup.IndoorFlags`); `src/AcDream.App/Rendering/Shaders/mesh_modern.vert` (binding=6 `instanceIndoor[]` gate on sun loop); per-frame sun `src/AcDream.App/Rendering/GameWindow.cs:10421` (`UpdateSunFromSky`, unchanged) | Torches: outdoor objects never torch-lit (exact retail). Sun: indoor objects (furniture, NPCs, player in a windowed building) never sun-lit (exact retail per-stage). Ambient: per-player-cell regime unchanged (exact retail `ChangePosition`). | The `ebp_2` unaudited test in `ChangePosition` could affect a narrow class of cells (entrance cells? sub-cells with special flags?) — no symptom observed; audit it if a lighting edge case arises in an unusual cell type | `useSunlight` gate `DrawMeshInternal` 0x0059f398; `useSunlightSet` 0x0054d450; per-stage `PView::DrawCells` 0x005a4840 (`useSunlightSet(1)` 0x005a485a / `useSunlightSet(0)` 0x005a49f3); `minimize_object_lighting` 0x0054d480; `CellManager::ChangePosition` 0x004559B0 (ambient + seen_outside) |
 | AP-35 | Point/spot lights are now PER-VERTEX Gouraud (`pointContribution` ~line 153 of `mesh_modern.vert`) matching retail's `SetStaticLightingVertexColors` bake path. Half-Lambert wrap (`(1/1.5)·(N·D + 0.5·d)`) AND norm distance attenuation (`distsq>1 ? distsq·d : d`) ARE ported (A7 Fix A, `aa94ced`). Point-light sum clamped to [0,1] on its own accumulator before adding ambient+sun (A7 Fix D D-1, mirrors retail's per-vertex bake clamp). CPU oracle: `src/AcDream.Core/Lighting/LightBake.cs`, locked by `tests/AcDream.Core.Tests/Lighting/LightBakeConformanceTests.cs`. **Residual (two parts):** (a) acdream lights in-shader each frame (per-frame GPU evaluate); retail bakes into the vertex buffer ONCE — an architecture/performance difference; the wrap + norm + clamp formula is the same, but bake-once is cheaper for static geometry; (b) acdream's `SelectForObject` keeps only the 8 NEAREST reaching point/spot lights per object/cell (`MaxLightsPerObject=8`, see AP-16), whereas retail's bake sums ALL reaching static lights per vertex — a surface reached by >8 point lights is dimmer in acdream than retail's bake result (rare in practice; a room has a handful of torches) | `src/AcDream.App/Rendering/Shaders/mesh_modern.vert` (`pointContribution` ~line 153; wrap ~line 163; norm ~line 167; point-sum clamp line 210) | Per-vertex Gouraud + wrap + norm + clamp all match retail. The two residuals are: (a) per-frame GPU vs bake-once — architecture/perf only; (b) 8-light cap dimming when >8 lights reach one surface — rare. `LightInfoLoader.cs:81` folds static_light_factor 1.3 into Range | (a) A new frame-time consumer bypassing `accumulateLights` would need to replicate the wrap + norm formula; per-frame GPU re-evaluate has higher per-frame cost than bake for static geometry. (b) A densely lit scene (>8 torches reaching one wall) renders dimmer than retail — see AP-16 for the 8-cap ownership | `calc_point_light` 0x0059c8b0 (line 0x0059c9a2 ramp; 0x0059c925 wrap); `SetStaticLightingVertexColors` 0x0059cfe0; static_light_factor 0x00820e24 |

docs/architecture/worldbuilder-inventory.md

@@ -130,6 +130,25 @@ anchors: `CPhysicsPart::GetMaxDegradeDistance` `0x0050D510`,
 `CPhysicsObj::ShouldDrawParticles` `0x0050FE60`, and
 `ParticleEmitter::UpdateParticles` `0x0051D180`.
 
+**Retail shared world-alpha seam (2026-07-18).** WorldBuilder's editor
+renderers classify translucent mesh batches correctly, but they have no live
+retail `CPartCell`/`CShadowPart` list and render particles in a separate
+batcher. `src/AcDream.App/Rendering/RetailAlphaQueue.cs` is therefore an
+acdream-owned runtime seam above the extracted mesh pipeline. During the main
+world frame, `WbDrawDispatcher` and `ParticleRenderer` submit transparent
+GfxObj subsets and scene particles into one stable far-to-near stream keyed by
+the transformed DAT `SortCenter`; only adjacent compatible entries may batch.
+`RetailPViewRenderer` drains the landscape scope before the optional depth
+clear, and `GameWindow` drains the final scope before private viewports/UI.
+Sky and sealed off-screen render targets remain independent. No DAT reader,
+mesh decoder, or second scene graph was introduced. Retail anchors:
+`CPhysicsPart::UpdateViewerDistance` `0x0050E030`,
+`RenderDeviceD3D::DrawObjCellForDummies` `0x005A0760`,
+`CShadowPart::insertion_sort` `0x006B5130`,
+`D3DPolyRender::AddMeshToAlphaList` `0x0059C230`, and
+`D3DPolyRender::FlushAlphaList` `0x0059D2E0`. The modern per-cell-order and
+EnvCell-shell residual is tracked explicitly as AP-34.
+
 **Retail portal-space viewport adapter (2026-07-15).**
 `src/AcDream.App/Rendering/PortalTunnelPresentation.cs` uses the extracted
 Setup/GfxObj mesh pipeline and mandatory `WbDrawDispatcher` for retail's