diff --git a/restoration/source410/BT/MECH.CPP b/restoration/source410/BT/MECH.CPP
index 150d462c..aefc563c 100644
--- a/restoration/source410/BT/MECH.CPP
+++ b/restoration/source410/BT/MECH.CPP
@@ -302,7 +302,12 @@ Mech::Mech(
 	//
 	duckRequestLatch		= 0;
 	duckPhaseRequest		= 0;
-	duckDemand				= 0.0f;
+	mobilityScale			= 1.0f;	// [T1] held at "legs healthy" until the
+									// +0x7ac leg-roster scan lands; the
+									// binary re-derives it every frame
+	defaultStateRequests	= 0;	// the streamed mode-request counters
+	gimpLeftRequests		= 0;	// (binary ctor @004a1674 zeroes
+	gimpRightRequests		= 0;	// +0x334/+0x338/+0x33c)
 	collisionVolumeState.Initialize(2);
 	collisionVolumeState.SetLevel(1);
 	standingVolumeHeight	= 0.0f;
@@ -1260,6 +1265,16 @@ void
 	currentBodySpeed = 0.0f;
 	bodyTargetSpeed  = 0.0f;
 
+	//
+	// The binary Reset (@0049fb74): the movement-mode alarm drops to
+	// DefaultState (FUN_0041bbd8(mech+0x2c, 0)) and the streamed mode
+	// request counters clear with it (+0x334/+0x338/+0x33c = 0).
+	//
+	SetSimulationState(0);
+	defaultStateRequests = 0;
+	gimpLeftRequests     = 0;
+	gimpRightRequests    = 0;
+
 	statusAlarm.SetLevel(0);
 	deathTransitionDone = 0;
 
@@ -1658,99 +1673,76 @@ void
 	// reconstructed.
 	//
 	//
-	// STAGED (the master performance's job, mech4 @004a9b5c -- never
-	// decompiled): promote the damage model's limp signal (statusAlarm level
-	// 3/4, MECHDMG's leg-half write to mech+0x2c) into the movement mode the
-	// gait machinery keys on.  The binary's damage code READS mode 3/4 to
-	// decide "already limping", so the promotion demonstrably exists; the
-	// master perf is the only unaccounted-for writer.  Guarded so a death
-	// (mode 5+) is never overwritten.
+	// BT_FORCE_DUCK=1: one authentic duck press every ~4s on its own
+	// clock (independent of the one-shot button harness below) -- the
+	// DuckRequest message, plus the held cell and analog demand every
+	// frame.  Press one ducks, the next (from the hold) stands: the
+	// full-machine cycle soak.
 	//
 	{
-		unsigned status = statusAlarm.GetLevel();
-		int mode = MovementMode();
-		if ((status == 3 || status == 4) && mode < 3)
+		static int s_forceDuck = -1;
+		static Scalar s_duckClock = 0.0f;
+		if (s_forceDuck < 0)
 		{
-			SetSimulationState(status);
+			s_forceDuck = (getenv("BT_FORCE_DUCK") != NULL) ? 1 : 0;
 		}
-	}
-
-	int limping;
-	{
-		int mode = MovementMode();
-		limping = (mode == 3 || mode == 4) && hasGimpClips;
-	}
-
-	//
-	// The master performance clears the leg reset latch at the top of every
-	// frame (@0x4a9bff); the wind-down inside the advancer may set it again,
-	// and the turn-in-place dispatcher below honours it -- so a walk that
-	// wound down THIS frame cannot re-enter as a turn until the next.
-	//
-	legResetLatch = 0;
-
-		//
-		// BT_FORCE_DUCK=1: one authentic duck press every ~4s on its own
-		// clock (independent of the one-shot button harness below) -- the
-		// DuckRequest message, plus the held cell and analog demand every
-		// frame.  Press one ducks, the next (from the hold) stands: the
-		// full-machine cycle soak.
-		//
+		if (s_forceDuck)
 		{
-			static int s_forceDuck = -1;
-			static Scalar s_duckClock = 0.0f;
-			if (s_forceDuck < 0)
+			MechControlsMapper
+				*pulse_mapper = (MechControlsMapper *)subsystemArray[0];
+			if (pulse_mapper != NULL)
 			{
-				s_forceDuck = (getenv("BT_FORCE_DUCK") != NULL) ? 1 : 0;
+				pulse_mapper->duckCommand = 1;
 			}
-			if (s_forceDuck)
+			mobilityScale = 1.0f;	// hold the legs-healthy gate open
+			s_duckClock += time_slice;
+			if (s_duckClock >= 4.0f)
 			{
-				MechControlsMapper
-					*pulse_mapper = (MechControlsMapper *)subsystemArray[0];
-				if (pulse_mapper != NULL)
-				{
-					pulse_mapper->duckCommand = 1;
-				}
-				duckDemand = 1.0f;
-				s_duckClock += time_slice;
-				if (s_duckClock >= 4.0f)
-				{
-					s_duckClock = 0.0f;
-					ReceiverDataMessageOf<int>
-						duck_press(
-							Mech::DuckRequestMessageID,
-							sizeof(ReceiverDataMessageOf<int>),
-							1);
-					Dispatch(&duck_press);
-				}
+				s_duckClock = 0.0f;
+				ReceiverDataMessageOf<int>
+					duck_press(
+						Mech::DuckRequestMessageID,
+						sizeof(ReceiverDataMessageOf<int>),
+						1);
+				Dispatch(&duck_press);
 			}
 		}
-
-
-	if (limping)
-	{
-		AdvanceLegAnimationGimp(time_slice);
-	}
-	else
-	{
-		AdvanceLegAnimation(time_slice);
 	}
 
 	//
-	// The gyro's joint writes (binary master-perf tail @0x4aaf74/83, AFTER
-	// the animation pass): the idle sway onto the EyeJoint, and the
-	// integrated eye offset + body tip onto 'jointeye' -- the joint the
-	// cockpit eyepoint rides.  Gated off during the death clips exactly as
-	// the tail is.
+	// THE MODE-REQUEST COUNTERS (master perf @004a9b5c -- the function IS
+	// in the BT411 decomp after all, part_013.c:3059; the old "never
+	// decompiled" note was wrong).  Three STREAMED request counters
+	// (+0x334/+0x338/+0x33c); while one is positive the master perf
+	// re-asserts the movement-mode alarm every frame, tracing the
+	// authentic state name to the debug stream.  They are fed only by the
+	// network stream appliers (creation record +0x1c0.., update records);
+	// the LOCAL damage path writes the mode directly (MECHDMG,
+	// @0x49c8fb/@0x49c926 half-leg limps, @0x49c88f/@0x49c8c5/@0x49c83c
+	// kills) -- the staged statusAlarm->mode "promotion" that lived here
+	// was a misconstruction and is retired.
 	//
-	if (gyroSubsystem != NULL && MovementMode() < 5)
+	if (defaultStateRequests > 0)
 	{
-		((Gyroscope *)gyroSubsystem)->WriteEyeJoint();
-		((Gyroscope *)gyroSubsystem)->WriteMechJoint();
+		DEBUG_STREAM << "DefaultState" << endl << flush;
+		SetSimulationState(0);
+	}
+	if (gimpLeftRequests > 0)
+	{
+		DEBUG_STREAM << "GimpLeft" << endl << flush;
+		SetSimulationState(3);
+	}
+	if (gimpRightRequests > 0)
+	{
+		DEBUG_STREAM << "GimpRight" << endl << flush;
+		SetSimulationState(4);
 	}
 
 	//
-	// THE DUCK DRIVER (master perf @004a9f61..@004aa155, disasm-decoded).
+	// THE DUCK DRIVER (master perf @004a9f61..@004aa155; 5.3.121 confirmed
+	// line-by-line against the decomp C, part_013.c:3059, and MOVED here --
+	// the binary runs it right after the mode-request counters, BEFORE the
+	// leg advance, so a duck arm is seen by the same frame's advance).
 	//
 	// Phase pick, every frame: not standing (raw state != 0) or duck button
 	// released -> no request.  Standing with the button held: from gait
@@ -1792,8 +1784,12 @@ void
 				leg_state = (int)legStateAlarm.GetLevel();
 			if (leg_state == 0 || leg_state == 4)
 			{
+				//
+				// The binary gate here is |mobilityScale| > 1e-4: "the
+				// legs still work" (see MECH.HPP) -- NOT a crouch analog.
+				//
 				duckPhaseRequest =
-					(duckDemand <= 0.0001f && duckDemand >= -0.0001f) ? 0 : 1;
+					(mobilityScale <= 0.0001f && mobilityScale >= -0.0001f) ? 0 : 1;
 			}
 			else if (leg_state == 1)
 			{
@@ -1893,9 +1889,55 @@ void
 		collisionVolumeState.SetLevel(collisionVolumeState.GetLevel());
 	}
 
+	int limping;
+	{
+		int mode = MovementMode();
+		limping = (mode == 3 || mode == 4) && hasGimpClips;
+	}
+
+	//
+	// The master performance clears the leg reset latch at the top of every
+	// frame (@0x4a9bff); the wind-down inside the advancer may set it again,
+	// and the turn-in-place dispatcher below honours it -- so a walk that
+	// wound down THIS frame cannot re-enter as a turn until the next.
+	//
+	legResetLatch = 0;
+
+	//
+	// DEATH-GUARDED (master perf: FUN_0049fb54 right before the pick) -- a
+	// mech in the death transition (modes 2/9) does not advance the leg
+	// gait; the crash clips run on the body machine's clock.  (The binary
+	// leaves its advance-distance stack slot stale in that case; we skip.)
+	//
+	if (!Mech::InDeathTransition(this))
+	{
+		if (limping)
+		{
+			AdvanceLegAnimationGimp(time_slice);
+		}
+		else
+		{
+			AdvanceLegAnimation(time_slice);
+		}
+	}
+
+	//
+	// The gyro's joint writes (binary master-perf tail @0x4aaf74/83, AFTER
+	// the animation pass): the idle sway onto the EyeJoint, and the
+	// integrated eye offset + body tip onto 'jointeye' -- the joint the
+	// cockpit eyepoint rides.  Gated off during the death clips exactly as
+	// the tail is.
+	//
+	if (gyroSubsystem != NULL && MovementMode() < 5)
+	{
+		((Gyroscope *)gyroSubsystem)->WriteEyeJoint();
+		((Gyroscope *)gyroSubsystem)->WriteMechJoint();
+	}
+
 	//
 	// THE TURN-IN-PLACE DISPATCHER (master perf @0x4aa505-0x4aa588, decoded
-	// from raw disasm -- Ghidra never decompiled the function).  From
+	// from raw disasm; 5.3.121: confirmed against the decomp C, which DOES
+	// carry the master perf at part_013.c:3059).  From
 	// Standing, arm the trn clip when the mech is TURNING (the binary tests
 	// |angularVelocity| > 1e-4; at a standstill the turn rate is exactly
 	// walkingTurnRate, so the operand is turnDemand * walkingTurnRate), the
diff --git a/restoration/source410/BT/MECH.HPP b/restoration/source410/BT/MECH.HPP
index da129c52..d264dc67 100644
--- a/restoration/source410/BT/MECH.HPP
+++ b/restoration/source410/BT/MECH.HPP
@@ -673,11 +673,19 @@
 			TakeDamageMessageHandler(TakeDamageMessage *message);
 
 		//
-		// Damage-side death flag: the body graphic alarm (statusAlarm,
-		// binary @0x714) at level >= 9 = the death/fall state.
+		// Damage-side death flag: the MOVEMENT MODE (the entity mode alarm,
+		// binary mech+0x2c, level cell +0x40 -- entity-base machinery, the
+		// engine seeds it from the class descriptor @0041bdf0) at level >= 9
+		// = killed.  5.3.121 CORRECTION: the old claim that the damage code
+		// wrote the @0x714 body-graphic alarm was a mis-anchor; the binary
+		// TakeDamage kill writes are FUN_0041bbd8(mech+0x2c, 9) at
+		// @0x49c88f/@0x49c8c5/@0x49c83c.  Death-SEQUENCE membership is the
+		// narrower Mech::InDeathTransition (modes 2/9, @0049fb54); the fall
+		// variants 5-8 sit between (armed as crash clips by the body
+		// machine) -- their writer is still undecoded.
 		//
 		Logical
-			IsMechDestroyed()				{ Check(this); return statusAlarm.GetLevel() >= 9; }
+			IsMechDestroyed()				{ Check(this); return (int)GetSimulationState() >= 9; }
 
 		//
 		// The live torso twist (the cylinder table's rotate-with-torso rows
@@ -914,7 +922,13 @@
 		//       recomputed every frame from the raw simulation state, the
 		//       mapper's duck button cell, the leg gait state and the duck
 		//       analog demand (deadzone 1e-4 @0x4ab16c).
-		//   duckDemand (+0x79c)         the analog crouch input.
+		//   mobilityScale (+0x79c)      NOT an analog input (5.3.121, from
+		//       the master perf C): recomputed each frame as the MAX leg
+		//       effectiveness (item +0x31c) over the +0x7ac leg roster,
+		//       it scales the throttle command, zeroes the turn command
+		//       when ~0, and the duck-down pick reads it as the "legs
+		//       still work" gate.  Our leg-roster scan is not
+		//       reconstructed yet, so the cell HOLDS 1.0 -- [T1] staging.
 		//   collisionVolumeState (+0x4c4)  the volume alarm: level 1 =
 		//       STANDING, 0 = DUCKED.  On an edge the collision template's
 		//       maxY (BoxedSolid +0xc -- the box TOP; the ground probe's
@@ -926,11 +940,24 @@
 		//
 		int						duckRequestLatch;
 		int						duckPhaseRequest;
-		Scalar					duckDemand;
+		Scalar					mobilityScale;
 		AlarmIndicator			collisionVolumeState;
 		Scalar					standingVolumeHeight;
 		Scalar					duckedVolumeHeight;
 
+		//
+		// The streamed movement-mode request counters (binary +0x334
+		// DefaultState / +0x338 GimpLeft / +0x33c GimpRight -- the
+		// authentic state names, traced by the master perf as it consumes
+		// them).  They are fed ONLY over the wire (the creation-stream
+		// ctor @004b3778 reads them at record +0x1c0.., the update-record
+		// appliers likewise); the LOCAL damage path writes the mode alarm
+		// directly.  Zeroed by the ctor (@004a1674) and Reset (@0049fb74).
+		//
+		int						defaultStateRequests;
+		int						gimpLeftRequests;
+		int						gimpRightRequests;
+
 		void
 			DuckRequestMessageHandler(Receiver::Message *message);
 		void
diff --git a/restoration/source410/BT/MECH4.NOTES.md b/restoration/source410/BT/MECH4.NOTES.md
index d88d1114..f82b9a6a 100644
--- a/restoration/source410/BT/MECH4.NOTES.md
+++ b/restoration/source410/BT/MECH4.NOTES.md
@@ -77,10 +77,21 @@ per contact by `Mover::ProcessCollisionList`. The decomp
   `ret`), hidden inside the manifest's mis-sized @004a9770 entry (that fn
   is really the stream-builder's 6.6KB name→SharedData strcmp resolver,
   and Ghidra swallowed everything to @004ab188 behind it).  Registration
-  proven from the ctor decomp: `(flags & 0xC) == 4` → replicant perf
-  @004ab9d8 (also hidden, inside @004ab430's claimed range); else master
-  @004a9b5c.  Recipe: extract CODE bytes (va 0x1000 → file 0xe00),
-  `ndisasm -b 32 -o 0x4a9b5c`.
+  proven from the ctor decomp: `(flags & 0xC) == 4` → replicant perf,
+  else master @004a9b5c.  Recipe: extract CODE bytes (va 0x1000 → file
+  0xe00), `ndisasm -b 32 -o 0x4a9b5c`.  **5.3.121 corrections: the DECOMP
+  has both perfs in plain C (master = part_013.c:3059, and @004ab430 IS
+  the complete replicant perf — my "@004ab9d8 hidden inside it" note was
+  its CONSTANT POOL @004ab9c8.. right after the body).  Replicant perf
+  read (not yet landed): zero +0x1dc, IntegrateMotion(dt, flavor 1);
+  ret 1 (fresher update present) → quaternion slerp +0x10c toward the
+  update target +0x26c with t = dt/(update age + dt) + matching velocity
+  lerps (+0x1d0 vs +0x2a4, +0x1f4 vs +0x298); either path applies the
+  +0x598 error-offset vector smoothed over the @004ab9cc window while
+  the +0x5a4 latch holds, then position ← the DR anchor +0x260; without
+  flags&0x4000 it re-resolves its room via the world chain (@0040e36c on
+  the +0x2e8 box) → +0x2f0; same ground clamp; `FUN_00421b6c`; same five
+  ring publishes + acceleration tail; NO update-record writer.**
 
   **The ordered call skeleton (111 calls)** — blocks in execution order,
   matching our Simulate's structure and locating every staged gap:
@@ -106,9 +117,11 @@ per contact by `Mover::ProcessCollisionList`. The decomp
      awaits a BT_MAP_LOG audit on a duck-bound RES.
   4. verify-guarded state read (mech4 accessors + Fail 0x40385c)
   5. gyro feed a (vec pair → gyro_b 0x4b2de4, gyro_a 0x4b2d8c)
-  6. death arm: InDeathTransition 0x49fb54 → the 1.9KB mech2 advancer
-     0x4a71f4 under it (death-clip arming)
-  7. **status→mode LIMP PROMOTION** (0x4a5028 + SetAnim/Mark/SetLevel)
+  6. death GUARD on the advance dispatch: InDeathTransition 0x49fb54;
+     0x4a71f4 = AdvanceLegAnimationGimp (modes 3/4 + 0x580), 0x4a5028 =
+     AdvanceLegAnimation -- NOT a death advancer (5.3.121 correction)
+  7. ~~status→mode LIMP PROMOTION~~ -> the STREAMED MODE-REQUEST COUNTERS
+     (see the 5.3.121 section; landed)
   8. update-record block (0x421bac/0x40e5f0/0x4227e0/0x41db7c/0x422bac +
      Point3D copies) — the dead-reckon/update writer
   9. collision → gyro crunch (gyro feeds + alarms)
@@ -130,3 +143,142 @@ per contact by `Mover::ProcessCollisionList`. The decomp
   (its replicant branch writes it with a vector subtract), so
   `motionEventName` in MECH.HPP is mistyped as CString — fix when this
   lands.
+
+
+## 5.3.121 -- THE DECOMP HAS THE MASTER PERF (ledger correction + full C read)
+
+**`part_013.c:3059` = `@004a9b5c file=bt/mech4.cpp` -- the ENTIRE master
+performance is in the BT411 decomp as plain C.**  Every "never decompiled /
+raw-disasm-only" note above is superseded; the asm read (scratchpad
+masterperf.asm) and the C agree line for line where both were done.  Rule
+for the rest of this file: read the C first, keep ndisasm for byte-level
+checks.  (`@004ab1c8` IntegrateMotion and `@004ab188` -- its dead-reckon
+position fold -- are decompiled right below it; the replicant-perf interior
+sits inside `@004ab430` from part_013:3804 with the SAME five ring
+publishes at :3948.)
+
+### The movement-mode carrier (the big one)
+
+* **mech+0x2c is the entity MOVEMENT-MODE AlarmIndicator; +0x40 is its
+  LEVEL cell.**  Everything we had read as "raw simulationState at +0x40"
+  is this alarm's level.  It is entity-BASE machinery: the engine seeds it
+  from the class descriptor's state count (`@0041bdf0`,
+  `FUN_0041b9ec(entity+0x2c, *(desc+0xc))`).
+* Mode roster so far: 0 DefaultState, 2/9 death transition (InDeathTransition
+  `@0049fb54` == 2||9 exactly), 3 GimpLeft, 4 GimpRight, 5-8 the four fall
+  variants (the body machine `@004a5678` head arms body clips 0x1c-0x1f from
+  them, one-shot latched by +0x650), 9 killed.  WHO WRITES 2 and 5-8 is
+  still undecoded (the death FSM brick).  Mode 1: seen only as "not
+  standing" in duck gates; identity open.
+* **The damage path writes the mode DIRECTLY**: TakeDamage `@0049c690`
+  calls `FUN_0041bbd8(mech+0x2c, ...)` -- half-gone right leg -> 4
+  (`@0x49c8fb`), left -> 3 (`@0x49c926`), leg destroyed while already
+  limping -> 9 (`@0x49c88f`/`@0x49c8c5`), vital zone -> 9 (`@0x49c83c`),
+  gated by the REAL `InDeathTransition`.  The old model (statusAlarm@0x714
+  as damage carrier + a Simulate "promotion") was a misconstruction --
+  RETIRED in 5.3.121; MECHDMG now writes the mode, and the @0x714 alarm is
+  ONLY the body-graphic clip selector (level = body clip index 0..0x20,
+  0x3c-stride name table at 0x50cfe8 "StandingAnimation"...).
+* **Block 7 truth**: not a promotion.  Three STREAMED request counters --
+  +0x334 DefaultState / +0x338 GimpLeft / +0x33c GimpRight (authentic
+  names, traced to the debug stream as consumed) -- re-assert the mode
+  alarm every frame they are positive.  Writers: the creation-stream ctor
+  `@004b3778` (record +0x1c0/1c4/1c8) and the update-record appliers;
+  zeroed by ctor `@004a1674` and Reset `@0049fb74` (which also drops the
+  mode alarm to 0 and zeroes +0x328/+0x32c/+0x330 -- those three cells
+  still unidentified).  LANDED with the authentic traces; inert until the
+  MP stream feed exists.
+
+### Corrected field identities (from the C)
+
+* **+0x79c = the MOBILITY SCALE**, not a duck analog: zeroed each frame,
+  then MAX of item+0x31c over the +0x7ac leg roster (count via the
+  iterator); throttle command (mapper+0x128) *= it; |scale| <= 1e-4 also
+  ZEROES the turn command (mapper+0x12c).  The duck-down pick reads it as
+  the "legs still work" gate.  Our tree HOLDS it at 1.0 ([T1]) until the
+  leg-roster scan is reconstructed.
+* **+0x3f4 = the REVERSE/BRAKE flag** (the "airborne re-pick" note was
+  wrong): set when speedCmd < 0 while +0x348 (current speed) > 0, cleared
+  otherwise; +0x344 (the accel rate the body advancers slew by) picks
+  +0x5b8 forward / +0x5bc reverse accel (sentinel -1.1f @0x4ab174 = whole
+  system disabled; IntegrateMotion re-derives the pick at its head).
+  Entering reverse: MarkUpdate(0x100), gyro kick (0,0,+/-rand)@0.4 (50%
+  sign flip via 0x408050 rand + 0.5f compare @0x4ab170).  While reversing
+  above the walk threshold (+0x534 < +0x348): pitch kick (0,1,0)@0.2 on a
+  0.4s countdown (+0x5c4) -- the braking shudder.  NOT YET LANDED.
+* +0x654 = the legResetLatch (saved at frame top, cleared, compared at
+  frame end -> MarkUpdate(8) on change).  +0x584 = squat-capable,
+  +0x588 = turn-clip-capable, +0x580 = gimp-clip-capable.
+* +0x574/+0x578 = walk/run turn rates; turn attenuation: speed >= +0x538
+  -> t=(speed-+0x534)/(+0x34c-+0x534), lerp A->B, t>1 -> B/t^2; clamp >=0;
+  x mapper turn cmd (+0x12c); ZERO while fallen (+0x650) or legState
+  1/2/3; result -> +0x1d4.
+* +0x348 = current speed, +0x34c = max run speed, +0x350 = min(?) speed
+  clamp in the shortfall term, +0x7a0 = real max speed (5-notch throttle
+  quantizer: cmd = round(5*cmd/+0x7a0 + 0.5)/5 * +0x7a0... exact rounding
+  via ftol of +0.5f), +0x6b4/+0x6b8 = body-channel demand/speed pair.
+* Perf head: `FUN_0049fa1c` = the combat-effectiveness census -> +0x414:
+  walks roster slots 2.., classID 0xBBE (generator) charge fraction
+  +0x12c/+0x128, 0xBC1 count with state != 1 && +0x210 != 4 (working
+  coolant units), RTTI-gated weapon count (excluding launchers with
+  +0x364 == 7); +0x414 = 1 when weapons < +0x448 min, no coolant, charge
+  below threshold, or limping with the duck cell 0.  Consumer unknown.
+* Perf head one-shot: +0x7a4 set && +0x7a8 clear -> sends the
+  **RealMaxSpeed message (0x15, 0x20 bytes)** carrying +0x7a0, then
+  latches +0x7a8 -- the network share of the computed real max speed
+  (pairs with the 0x15 handler @0049f604, still to land).
+* Link plug at +0x418: first bound entity's distance -> +0x400, presence
+  -> +0x3fc (identity of the link: open).
+* Mission-list y-compare -> +0x394 (the old "verify-guarded read" block):
+  compares *(+0x67c) against a mission-chain entry's position y when
+  legState not 0/1 -- identity open (in-water / off-pad flag?).
+
+### Block 8 (move/collide) -- decoded, NOT yet landed
+
+Ground clamp `FUN_0040e5f0(+0x2f0, pos)` (penetration -> pos.y and the
++0x2e8 box y/z); save old basis/pos; `FUN_004227e0(this,0)` volume;
++0x44c = 0; `FUN_00422bac(this, vol, dt, &oldpos, &probe)` = THE MOVE.
+Outcomes on probe[1] (local_1c8): == sentinel @0x4ab180 -> restore
+pos+basis, gyro kick along normal @0.4 + pitch @0.2, CollisionState alarm
+(+0x450, 3 levels) -> 0;  <= 0 -> alarm 0;  else REAL HIT: partner
+material *(partner+0x24) -> +0x4a4, world->body normal -> +0x4a8, closing
+speed -(n . v) -> +0x4b4 (the CollisionMaterialType / CollisionNormal /
+CollisionSpeed attribute cells), alarm <- +0x44c, STOP at old pos
+(basis <- identity const @0x4e0f74), then **collision damage = a
+SELF-DISPATCHED TakeDamage message (id 0x12, 100 bytes, damage type
+DAT_0050b9ac, attacker = own entity id +0x184)** via vtbl+0xc; slow
+(speed^2 <= @0x4ab184) -> legs to 0 + legResetLatch=1 (+ MarkUpdate(1|
+0x40) if +0x464 == 1), fast -> stagger clip 0x20 + MarkUpdate(1|0x20).
+After: +0x464 changed over frame -> MarkUpdate(0x80); vel.y = pos.y -
+oldpos.y (raw delta); `FUN_00421b6c`; IntegrateMotion(dt, 0); dirty
+thresholds (+0x768 pos-xz^2, +0x76c turn-delta, +0x770 heading-delta ->
+MarkUpdate(0x10)/(1); mapper cmd != +0x6b4 -> MarkUpdate(4)).
+
+### The tail (decoded, partially ours already)
+
+Instability accumulator +0x3f0, THREE terms then clamp 1.0:
+(|accelvec +0x82c| / +0x784 * +0x788)^2  +  speed-shortfall
+(clamped cmd vs +0x348, relative) * +0x78c  +  (legState==4 ? +0x798 : 0);
+-> gyro+0x3a8 when gyro exists.  Gyro churn `FUN_004b2eac(gyro, dt)` only
+when legState != 0 && +0x57c == 0, then `FUN_004b34ec(gyro)` always --
+both gated !fallen(+0x650).  Torso::UpdateJoints(+0x438, &+0x360) gated
+!fallen.  `FUN_0049fe80(this)` (a second roster census -- damage-side;
+decode parked).  Five RING BUFFERS (FUN_0043ade4 = ring push, 0x43ae47/
+0x43ae0b = ring reduces): +0x7e0<-vel.z, +0x7ec<-vel.y, +0x7f8<-vel.x,
++0x804<-turn rate, +0x810<-dt; smoothed values -> acceleration vector
++0x1dc (delta/sum-dt vs prevs +0x81c..+0x828) -> copied to +0x82c (the
+instability input).  `FUN_00421b2c` opens AND closes the frame (called
+twice); the mobility iterator is destroyed at the very end
+(`FUN_004a4ffb`).
+
+### Landed in 5.3.121
+
+MECHDMG kill/limp writes retargeted to the mode (SetSimulationState);
+IsMechDestroyed reads the mode (>= 9); the staged promotion replaced by
+the authentic counter block (+ ctor/Reset zeroes + Reset mode drop); the
+duck driver moved to its authentic slot (counters -> duck -> limp standup
+-> volume swap -> advance) and its gate renamed mobilityScale; the
+advance dispatch death-guarded.  SIDE EFFECT: `Mech::InDeathTransition`
+and the camera director's cut-away are now LIVE (mode 9 is finally
+written on kills) -- the director path needs a rig eyeball when a pod is
+back.
diff --git a/restoration/source410/BT/MECHDMG.CPP b/restoration/source410/BT/MECHDMG.CPP
index a0f43a3c..fa2415d1 100644
--- a/restoration/source410/BT/MECHDMG.CPP
+++ b/restoration/source410/BT/MECHDMG.CPP
@@ -404,7 +404,12 @@ void
 			//
 			if (damageLevel >= StructureMax)
 			{
-				mech->statusAlarm.SetLevel(9);
+				//
+				// Binary @0x49c88f/@0x49c8c5: the kill is a DIRECT write of
+				// the movement-mode alarm (mech+0x2c) to level 9 -- 5.3.121
+				// carrier correction, formerly mis-routed via statusAlarm.
+				//
+				mech->SetSimulationState(9);
 				if (getenv("BT_MECH_LOG"))
 				{
 					DEBUG_STREAM << "[death] LEG zone " << damageZoneIndex
@@ -413,16 +418,23 @@ void
 				}
 			}
 		}
-		else if (!mech->IsMechDestroyed())
+		else if (!Mech::InDeathTransition(mech))
 		{
-			// half-gone leg -> the limp gait (right 4, left 3)
+			//
+			// Binary @0x49c8fb/@0x49c926: a half-gone leg writes the
+			// movement mode DIRECTLY (right -> GimpRight 4, left ->
+			// GimpLeft 3 -- the authentic state names from the master
+			// perf's request traces).  The gate is the authentic
+			// FUN_0049fb54 (InDeathTransition), not the staged
+			// "not already destroyed".
+			//
 			if (rightLeg != 0 && damageLevel >= LegHalfStructure)
 			{
-				mech->statusAlarm.SetLevel(4);
+				mech->SetSimulationState(4);
 			}
 			else if (leftLeg != 0 && damageLevel >= LegHalfStructure)
 			{
-				mech->statusAlarm.SetLevel(3);
+				mech->SetSimulationState(3);
 			}
 		}
 	}
@@ -432,7 +444,7 @@ void
 		// Vital zone destroyed -> mech death.
 		//
 		SetDamageZoneState(BurningState);
-		mech->statusAlarm.SetLevel(9);
+		mech->SetSimulationState(9);	// binary @0x49c83c: mode alarm, level 9
 		if (getenv("BT_MECH_LOG"))
 		{
 			DEBUG_STREAM << "[death] VITAL zone " << damageZoneIndex
@@ -666,7 +678,7 @@ void
 		{
 			if (victim->IsVitalSubsystem())
 			{
-				((Mech *)GetOwningSimulation())->statusAlarm.SetLevel(9);
+				((Mech *)GetOwningSimulation())->SetSimulationState(9);
 			}
 			victim->ForceCriticalFailure();
 			if (getenv("BT_MECH_LOG"))
diff --git a/restoration/source410/BT/MECHDMG.NOTES.md b/restoration/source410/BT/MECHDMG.NOTES.md
index 27a25eff..ae9593e4 100644
--- a/restoration/source410/BT/MECHDMG.NOTES.md
+++ b/restoration/source410/BT/MECHDMG.NOTES.md
@@ -74,11 +74,14 @@ mapper) installs post-ctor, slot 1 (voltage bus) can be empty, and
   roll the special-damage short (Energy hit + a Generator crit pick →
   `Generator::ForceShort()`, novice-exempt), and drive failure state: a
   destroyed non-leg zone runs `RecurseSegmentTable`; a destroyed LEG or
-  VITAL zone is the mech kill (statusAlarm level 9 = the @0x714 body-graphic
-  death level; `Mech::IsMechDestroyed()`); half-gone legs set the gimp
-  graphic (right 4 / left 3). The binary's MovementMode(3||4)/IsDisabled
-  gates on the leg branches are STAGED as "not already destroyed" until the
-  gait FSM exists.
+  VITAL zone is the mech kill; half-gone legs set the limp. **5.3.121
+  CARRIER CORRECTION**: every one of those writes goes to the MOVEMENT-MODE
+  alarm at mech+0x2c (level cell +0x40) -- `FUN_0041bbd8(mech+0x2c, ...)`:
+  right-leg-half → 4 GimpRight @0x49c8fb, left → 3 GimpLeft @0x49c926, leg
+  destroyed while limping → 9 @0x49c88f/@0x49c8c5, vital → 9 @0x49c83c --
+  NOT the @0x714 body-graphic alarm (that one's level is the body CLIP
+  index; damage never touches it).  The leg branch gate is the real
+  `InDeathTransition` (@0049fb54, modes 2||9), landed to match.
 - **`LODDamageRouter` (@0049c40c)**: same-attacker clustering — reuse the
   previous child inside [0.25s, 10s] (hysteresis 0.33 — a binary DOUBLE,
   bytes 1f85eb51b81ed53f); new attacker or stale window re-rolls
diff --git a/restoration/source410/BT/MISSILE.NOTES.md b/restoration/source410/BT/MISSILE.NOTES.md
index 20c78a00..3914a3e5 100644
--- a/restoration/source410/BT/MISSILE.NOTES.md
+++ b/restoration/source410/BT/MISSILE.NOTES.md
@@ -92,3 +92,24 @@ app task, one delete per frame, not mid-walk).
 - Cluster SPLASH + explosion entity (ClassID 0x5C @004be078), world-geometry
   collision (@0042291c), target-velocity intercept lead, Missile
   WriteUpdateRecord (tag 0x78) for MP replication.
+
+## OPEN DEFECT (found 5.3.121, PRE-EXISTING): lifecycle kill-cycle crash
+
+`lifecycle2.conf` (KILLTEST.EGG, BT_SPAWN_ENEMY + BT_FORCE_FIRE +
+BT_FORCE_ZONE=10) now dies ~seconds after the first SRM salvos launch,
+BEFORE any [death]: `Exception 0E ... illegal address 000000FC`, faulting
+instruction `mov eax,[eax+0xfc]` after `mov eax,[ebx+0x128]; mov eax,[eax]`
+with ebx = THE ENEMY MECH (matches the [enemy] spawn address in OUT.TXT).
+Map resolves the EIP into `Missile::MoveAndCollide` (0x4b1b54-base) -- the
+seeker/thruster work is folded in there, but our LeadTarget NULL-guards
+targetEntity, so the exact deref is not yet identified (mech+0x128 in OUR
+layout -> a pointer whose first dword is NULL -> +0xfc).
+
+**PROVEN PRE-EXISTING by a control run**: the identical crash (same 0xFC
+address, EIP shifted 0xCC by the relink) reproduces on a build with all
+5.3.121 changes stashed.  The rig last ran green at 5.3.25 -- the break
+landed somewhere in the gait/renderer/duck arc (5.3.26..5.3.120) and was
+never noticed because nothing re-ran the kill-cycle conf.  Bisect hint:
+the [msl] LAUNCH lines print, so the launch path is fine; the fault is in
+the per-frame chase or the mutual-fire path against the enemy mech.
+Two-agreeing-runs done (both builds crash identically, deterministic).