docs(gfx): correct 'immediate-mode slot-0' -> engine is RETAINED (native-verified)

The animation-slice note mis-called the SC0000 opening 'immediate-mode slot-0
blits', inferred from our own gfx oracle (which mis-reported slot 0). Verified from
native code + raw bytecode: draw-texture (0x1fb) -> gfx_object_bind_draw@0x47e870
binds a RETAINED object by handle (stores the slot INDEX, a live per-frame ref, not
a snapshot). The opening is a sleep-paced sequence of retained objects with distinct
handles + per-object working slots (CG loader: handle=CG_array[G[0x62450]] INIT2
array, slot=G[0x62452]). Our VM collapses the paced sequence -> only the final state
shows -> needs frame-pacing (scene-coroutine/sleep), not this alpha channel.

engine-re.md: new 'opening render path is RETAINED' subsection. Ghidra: annotated
gfx_op_0x1fb_draw_bind (gfx_object_bind_draw already documented retained).

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>

docs/engine-re.md

@@ -351,6 +351,33 @@ both annotated) and grepping the SC0000 opening settles the animation model and
   draw-worker we deliberately don't model. Determine it empirically from the animating channel + screenshot, not
   by RE'ing the surface layer.
 
+##### The opening render path is RETAINED, not immediate-mode (2026-07-08, ground-truth correction)
+
+A working note in the animation slice mis-called the SC0000 opening a set of "immediate-mode slot-0 blits." That
+was **wrong**, and it came from trusting our own `Age.Cli gfx` oracle (which executes our VM and mis-labeled the
+CG draws as "slot 0"). Verified against native code + the raw bytecode:
+
+- **`draw-texture` (op `0x1fb`, handler `gfx_op_0x1fb_draw_bind`@`0x422510`) is a RETAINED bind, not a blit.** It
+  writes cmd-type `0x11` and calls **`gfx_object_bind_draw`@`0x47e870`**, which on the object keyed by `handle`
+  (operand 1) sets: `flag|=1` (visible), `obj+4 = source SLOT index`, `obj+8..0x14 = source rect`,
+  `obj+0x24/28/2c = position`. Its plate comment (prior RE) already states the key fact: the object stores the
+  **slot INDEX — a live ref to `surface[slot]`, resolved each frame at render — NOT a texture snapshot.** Objects
+  persist and are composited each frame; this is exactly the surfaces+objects model in "The full gfx render
+  model" above.
+- **The SC0000 opening is a retained scene of distinct objects, `sleep`-paced.** Raw bytecode: fixed-handle UI
+  objects (`0xcf08` slot 3 full-screen, `0xc350` slot 0xe, `0xe678` slot 0xd — a 400×30 element re-bound 20+
+  times), an animated sprite (`draw-texture (G[0x62457]) (G[0x62452]) … (G[0x62498]) (G[0x62499])`, computed
+  position), and the CG loader (`SC0000` @`0x126e1`/`0x12970`): `set-texture G[0x62424] → slot G[0x62452]`,
+  `get-texture-size`, centre it, then `draw-texture (handle = CG_array[G[0x62450]] = INIT2 array G[0x62455..])
+  slot G[0x62452] …`. `sleep 0x64/0x3e8/0x2ee` sits between steps. So different draws use **different handles and
+  per-object working slots** — not one slot-0 canvas.
+- **Why our port still doesn't animate the opening (conclusion unchanged, mechanism corrected):** we execute the
+  whole load/draw/`sleep` sequence **instantly** — no `sleep` timing, no per-frame present — so we only ever see
+  the *final* retained state; the intermediate `AE*` frames (`AE001D→AE002B→AE003B`, surface swaps on the working
+  slot between paced frames) never get a frame to display. The fix is **frame-pacing** (scene-coroutine / `sleep
+  0xc8`), a separate subsystem from the transform/alpha channel. **Lesson: never characterise the engine's render
+  mechanism from our own VM's oracle output — use native code + raw bytecode.**
+
 ### The render drift's SECOND half: missing system-boot state (2026-07-07, resolved)
 
 Implementing the gfx ops (above) was necessary but not sufficient — a cold single-scene run of SC0000 still

docs/phase-a-slice-plan.md

@@ -206,8 +206,10 @@ full-screen **event-CG layer (`EV052*`) renders correctly** as the opening plays
 2. **No alpha/blend.** `AE*` full-screen fade/flash effects draw **opaque and instant** (a static grey/white
    sheet over the CG) instead of alpha-animating. No chromakey either (sprites would show green boxes —
    moot until they position).
-3. **Slot model is an approximation.** We use one `TextureRect` per slot, replace-on-draw; the game
-   actually **blits onto slot 0 as an immediate-mode canvas** (everything composites into slot 0).
+3. **Slot model is an approximation.** We use one `TextureRect` per slot, replace-on-draw. (⚠ Earlier this
+   line claimed "the game blits onto slot 0 as an immediate-mode canvas" — **DISPROVEN 2026-07-08**: the engine
+   is RETAINED — `draw-texture` binds a retained object by handle (`gfx_object_bind_draw`), objects have distinct
+   handles + per-object source slots, composited each frame. See engine-re.md "opening render path is RETAINED".)
 4. AGF is **pre-converted to BMP offline** (`convert_agf.py --scene`); a runtime C# AGF decoder is deferred.
 
 **Next chunk — graphics-geometry/blend subsystem:** implement `0x208` (host returns the slot's real image
@@ -361,15 +363,20 @@ opening page 2 is pixel-identical to baseline at settle 3 and 300; Godot selftes
 **Tracker delta (`scene_opcode_coverage.py SC0000`):** GAP 68→**64** ops (836→**741** instrs), impl 49→**53**,
 correctly-handled 60→**65/129 (50.4%)**.
 
-**⚠ Honest scope — the opening `AE*` explosion does NOT visibly animate from this chunk.** The `gfx` oracle
-(`Age.Cli gfx --boot SC0000.BIN`) shows the opening's `AE*` effect is `AE001D → AE002B → AE003B` blitted
-**immediate-mode onto slot 0** full-screen, each overdrawn by the next `EV*`/`AE*` — a **frame sequence** the
-real game paces one-per-tick via `sleep`/the render loop. Our VM executes the whole burst instantly, so the
-compositor only ever sees the final slot-0 state; the transient `AE*` frames never show. This per-object alpha
-channel is the correct foundation for **retained-sprite** animation (character fades/scales via the handle
-system), but the opening explosion needs **frame-pacing** — modeling the scene-coroutine / `sleep 0xc8` timing
-(`0x7b`/`0x140`/`0xc8`, still GAP) so the burst is *not* collapsed. That is the clearly-scoped next chunk for the
-visible opening animation, and a genuinely different subsystem than the transform/alpha channel landed here.
+**⚠ Honest scope — the opening `AE*` explosion does NOT visibly animate from this chunk.** The opening's `AE*`
+effect (`AE001D → AE002B → AE003B`) is a **`sleep`-paced sequence of RETAINED-object loads/draws** — verified
+2026-07-08 from native code + the raw bytecode (NOT our `gfx` oracle, which mis-reported these as "slot 0"; see
+engine-re.md "opening render path is RETAINED"). `draw-texture` binds a retained object by handle
+(`gfx_object_bind_draw`@`0x47e870`); the SC0000 CG loader supplies `handle = CG_array[G[0x62450]]` (the INIT2
+handle array `G[0x62455..]`) and a per-object working slot `G[0x62452]`, with `sleep` (`0x64`/`0x3e8`/`0x2ee`)
+between steps. Our VM executes the whole load/draw/`sleep` burst **instantly** (no timing, no per-frame present),
+so we only ever see the *final* retained state; the intermediate `AE*` frames never get a frame to display. This
+per-object alpha channel is the correct foundation for **retained-sprite** animation (character fades/scales via
+the handle system), but the opening explosion needs **frame-pacing** — modeling the scene-coroutine / `sleep 0xc8`
+timing (`0x7b`/`0x140`/`0xc8`, still GAP) so the burst is *not* collapsed. That is the clearly-scoped next chunk
+for the visible opening animation, and a genuinely different subsystem than the transform/alpha channel landed
+here. (**Correction:** an earlier draft of this note called it "immediate-mode slot-0 blits" — wrong; the engine
+is retained, per the native `draw-texture → gfx_object_bind_draw` bind.)
 
 ---