Myomers::MyomersSimulation (@004b8d18) reconstructed, five blocks in the
binary's order, and it answers a question left open since 5.3.94:
runSpeedMax (+0x7a0) IS WRITTEN BY THE MYOMERS. Each myomer's
RegisterMaxOutput (@004b8ef0) raises the mech's run-cycle cap to its
full-throttle AvailableOutput -- so the mech's true top speed is a
property of its MUSCLE ASSEMBLY, and it degrades live as heat crosses the
degradation band and as the myomer's own zone accumulates damage. The
1e9 ctor sentinel now stands only until the first myomer registers.
The drive computation (@004b8ac0): base = the mech base speed
(reverseStrideLength, mech+0x34C) scaled by the gear ratio -- input voltage
over the RECOMMENDED gear -- with a quadratic thermal bleed across the
degradation band, zero past failure, and the subsystem's wear taken off the
top. speedEffect republishes per frame as a 0..1 fraction of full drive:
input clamped to the SELECTED gear then the top gear, through
AvailableOutput, normalised by the base speed.
AND THE MUSCLES HEAL: while a myomer has NO VOLTAGE and its zone is not
destroyed, it hands its own zone a NEGATIVE explosive tick (-0.011,
0xbc343958) -- unloaded muscle knits itself back together. The donor's
history records this block as the fix for "torso twist dead after respawn"
(the Torso mirrors this subsystem's electrical level; with the base machine
never running, a myomer that lost power in the death window sat at
NoVoltage forever) -- and our runs respawn now, so the fix is live where it
matters.
Registered on the master (the sibling gate); the base electrical machine
runs FIRST and unconditionally, per the donor's hard-learned ordering note.
Not carried over: the port's BT_FORCE_SEEK / repair-test bench scaffolding.
Live run clean; the gait's speed sequence is unchanged (7.31972, 26.6726,
...) -- the muscle cap sits above the demanded 26.9 on a healthy Mad Cat,
exactly as it should.
Stub census: 18 across 12 files.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
CydandClaude Fable 5
02b78eadc7