110ecec2ca3edc135e260e0f3ee07e7945cbf4ec
8
Commits
| Author | SHA1 | Message | Date | |
|---|---|---|---|---|
  CydandClaude Fable 5
|
110ecec2ca |
BT410 Phase 5.3.18: flying missiles -- spawn/home/detonate LIVE, teardown open
The Missile entity family is reconstructed: Projectile : Mover gets a real make-path ctor; Missile hosts the Seeker + MissleThruster roster (the AUTHENTIC surviving SEEKER.HPP / MISTHRST.HPP interfaces -- the 1995 misspelling included) and flies the binary's guided integrator (@004bef78): age/lifetime, seeker re-lead with the decoded loft constants (200/50/300/0.1), thruster burn, steering (gain 4.0, deadband 1e-4), and the proximity fuse delivering the cluster damage ONCE with the launcher as inflictor. Live-verified end-to-end: LAUNCH -> flight telemetry (thruster visibly accelerating) -> DETONATE on the target ~0.9s out. OPEN: the first ~Missile off the death row completes, then the frame page-faults with EIP in the heap -- the first entity/subsystem teardown the reconstructed sim has ever run. The spawn path is therefore OPT-IN (BT_MISSILE_FLIGHT=1); default SRM delivery stays the 5.3.18a instant-hit cluster (verified 1:1, zero exceptions). Forensics + engine truths (SearchList crashes on non-directory ids; MakeMessages need FAMILY resource ids; explosionModelFile is a model-list INDEX) in MISSILE.NOTES.md. Also fixed on the way: the Subsystem NAME ctor left damageZone UNINITIALIZED (latent garbage for every name-built subsystem). Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> |
||
|
CydandClaude Fable 5
|
76cfc64353 |
BT410 Phase 5.3.16: targeting + damage -- two-mech fights are LIVE
The simulation is now a FIGHT. Mech target slot (+0x388), authentic Loaded->Firing gate (viewFireEnable && HasActiveTarget), UpdateTargeting range refresh, and SendDamage via the engine TakeDamageMessage. MECHDMG.CPP is BORN against the AUTHENTIC surviving CODE/BT/BT/MECHDMG.HPP: the Mech ctor's Pass-3 hull zone fill constructs Mech__DamageZone per zone (the Entity base only allocates the raw pointer array -- unfilled slots were crash #1; a base-class fill parses each record short and skews the stream -- crash #2). Streamed ctor parses the full BT tail (flags / Scalar-weighted criticals with the Master+Dynamic plug gate / LOD redirect table) and normalizes the armor economy: scale[type]=1/(raw x armorPoints), legs halved -- BT411-verified @0049ce50. Live two-mech verification (BT_SPAWN_ENEMY harness in DropZoneReply): 20 named hull zones stream on both mechs; PPC lands 11.77/hit (12 x chargeRatio^2, type 4), ER-M laser 3.43 (type 3), SRM6 salvos 5.83 x 6 rounds on independent zones (type 2, one message per missile -- burstCount is NOT in the zone formula); repeat hits climb linearly, leg zones absorb at half scale, zones reach 1.0 Burning/Destroyed. No-target and novice regressions hold (weapons load but never fire without a target). Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> |
||
|
CydandClaude Fable 5
|
b772e716fe |
BT410 Phase 5.3.15: the electrical charge model -- real charge, closed feedback loop
The emitter charge cycle is now the authentic electrical model end to end.
- WIRE-VERIFIED resource fix: the Emitter block is graphicLength /
dischargeTime / seekVoltage[5] (FRACTIONS of the generator's rated voltage,
-1 sentinel) / seekVoltageRecommendedIndex. The old two-field guess read
seekVoltage[3]=0.99 as "dischargeTime 0.99s" -- the true PPC discharge is
0.2s, and the calibration reads seekV={6000,7000,8000,9900} x rated 10000,
recommended gear 2: the documented curve exactly.
- Ctor calibration (@004bb120): energyTotal=(dmg+heat)x1e7; EC pins
E=0.5V^2EC to energyTotal at the recommended gear; voltageScale calibrates
the exponential charge to reach seekV[rec] in the authored RechargeRate
seconds cold; damageFraction = the damage share.
- TrackSeekVoltage (@004ba838) + ChargeTimeScale (@004b0d50): the charge
integrates from the generator through the heat-stretched time scale, and
the I2R loss heats the GENERATOR -- verified GeneratorA at 477.9K charging
its PPC (BT411 predicted ~+480K) while the avionics generator idles at 77.
Hot generators charge slower: the heat/firepower feedback loop is CLOSED.
- Emitter::ComputeOutputVoltage override (now virtual, as in the binary
vtable): dial = level/seekV[gear], 0.01 snap, over-1 clamp.
- Sub-stepped Loading (1/60s slices -- the BT411 weapon-brick fix) + the
documented-divergence overcharge rescue. VERIFIED: the PPC loads at level
~7920, inside the binary's exact [7920,8080] snap window.
- FireWeapon energy algebra (@004bace8): damage/heat = energy shares x
chargeRatio^2. VERIFIED dmg=11.78 / heat=1.079e8 at ratio 0.9906. The
heatCost x 1e7 partial is retired -- heat now comes from the energy.
Zero Fail; jams/shutdowns regressions stay live under spam.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
|
||
|
CydandClaude Fable 5
|
57507cb15c |
BT410 Phase 5.3.10: power/heat wave -- the thermal + electrical economy is LIVE
Weapons dump firing heat into their own sinks, sinks conduct through the Condenser bank into the central HeatSink, temperatures drive the degradation/ failure alarms, and every powered subsystem tracks its generator through the electrical state machine. Verified: the authentic fire-discipline game -- a PPC spikes 77->709K per shot, relaxes to 441K across its 5s reload, and sustained fire climbs 441->674->838->960K, brushing the authored 1000K degradation threshold. All calibration values match the BT411 audit exactly (central bank 1.39e6, PPC sink 174000, thresholds 77/1000/2000). - HEAT: HeatSinkSimulation (@004ad924 absorb->T->load->conduct->alarm), ConductHeat/ComputeHeatFlow (@004ad8ac/@004ad9ec two-body equilibrium relaxation; flow==0 exactly at uniform T), UpdateHeatLoad (15-sample filter); heat-state enum + accessors; installed by the HeatSink ctor. - WIRE-VERIFIED: HeatSink resource gains linkedSinkIndex -- THE missing ancestry int that shifted every descendant block +1 (voltageSourceIndex had been reading the linked-sink index: "power sources" appeared to be Condensers). Conduction topology wired from it at construction: weapons->Condensers1-6->central; Generators->Condensers; Reservoir->central. - MECHWEAP resource: authentic pip tail (pipPosition int + pipColor 3 floats + pipExtendedRange int) per the BT411 verified overlay; with linkedSinkIndex this closes the whole +3 alignment mystery (ProjectileWeapon pad deleted). True bhk1 reads: PPC recharge 5.0s (not 1.0), discharge 0.99s, range 900. - POWERSUB: ctor resolves voltageSourceIndex -> GeneratorA-D (wire-verified), taps via Generator::TapVoltageSource (-1 when full); PoweredSubsystemSimulation (@004b0bd0) electrical FSM (Starting/NoVoltage/Shorted/GeneratorOff/Ready); GeneratorSimulation partial (start/short-recovery timers). - Weapons: power step at sim head; Loading recharge gated on electrical Ready (authentic @4bbdf5); FireWeapon dumps firing heat. HEAT UNITS: the chain is 1e7-native with TWO authoring conventions -- energy weapons store small (PPC=11, x1e7 from the energy algebra), ballistics store native (SRM6=5.06e7, dumped raw; double-scaling it was the bring-up runaway-temperature bug). - SENSOR: authentic gating (@004b1c4c) -- power step, electrical-Ready gate, heat-state switch (Degradation x0.5 / Failure 0). Verified Ready + 100%. Deferred: coolant depletion/venting, the novice HeatModelOff gate, the charge model (TrackSeekVoltage/voltage sag/I2R), weapon gate 1 + jam roll, the central sink's forward-linked drain. Zero Fail across fire + neutral runs. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> |
||
|
CydandClaude Fable 5
|
ab7191dc38 |
BT410 Phase 5.3.9: ballistic fire path -- SRM launchers fire salvos + run dry
The SRM6 MissileLaunchers now run the authentic ballistic fire cycle against their linked AmmoBins, wire-verified end to end. - PROJWEAP ProjectileWeaponSimulation (PARTIAL, binary @004bbd04 -- fully recovered in the BT411 RE): one trigger-edge sample; the dry-bin gate pins NoAmmo(7) every frame; Loaded(2) -- the AMMO PULL LIVES IN THE CALLER (a denied shot does the 4->2 denial blip with NO ammo pulled -- early-returning gates from FireWeapon while the caller cycles the alarm is the 1995 "denied shot fakes a full firing cycle" defect class); granted -> Firing -> FireWeapon -> Loading (or NoAmmo on the emptying pull), recoil=rechargeRate; Loading bleeds recoil -> Loaded when a round is chambered, dial animates; Jammed(5) held; NoAmmo(7) = the roach-motel (re-asserts unconditionally). Deferred: electrical step, gate 1, eject, jam roll, HasActiveTarget. - AMMOBIN: FeedAmmo() + GetAmmoState() (1 round-ready / 2 EMPTY). - ammoBinLink wired from the resource's ammoBinIndex (the bin's roster slot). - WIRE-VERIFIED RESOURCE ALIGNMENT: a raw-stream dump pinned the ProjectileWeapon field block +3 ints past our staged struct (the MechWeapon resource ancestry runs 3 ints short -- the pip-family fields). Fixed with resourceAlignPad[3]; phantom muzzleVelocity/missileCount tail dropped. Confirmed on-wire: ammoBinIndex=27/29 (the exact bin slots), tracerInterval=1, minTOF=0.5, minVolt%=0.3, minJam=0.05, missileCount=6 (an SRM6!). - MECHWEAP: weapon-state enum extended to the full 1995 set (DryTrigger 1, Jammed 5, TriggerDuringJam 6, NoAmmo 7; count 8). - MISLANCH FireWeapon PARTIAL (@004bcc60: heat + salvo spawn only -- spawn needs the entity/targeting waves). VERIFIED headlessly (BT_FORCE_FIRE): bins resolve with the authored 24 rounds; SRM6s fire 6-missile salvos at the authored 1s cadence, rounds 24->0; after the 24th salvo the launcher goes DRY and stays silent across ~3,000 subsequent fire events while the energy weapons keep cycling. Zero Fail. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> |
||
|
CydandClaude Fable 5
|
896bbbed4d |
BT410 Phase 5.3.8: weapon fire path -- authentic attribute table + fire state machine
The energy weapons now run a live fire cycle, and the REAL trigger wiring is in place: the 1995 MechWeapon attribute table (binary @0x511890) is published with PINNED IDs, so the streamed per-mech fire-button mappings bind TriggerState (0x13) -> fireImpulse and the controls push writes the trigger into the weapon. - MECHWEAP: attribute enum + table (PercentDone 0x12, TriggerState 0x13, DistanceToTarget 0x14, TargetWithinRange 0x15, WeaponRange 0x16, EstimatedReadyTime 0x1A, RearFiring 0x1B, WeaponState 0x1C). Pads bridge the chain gap 2..0x11 -- LOAD-BEARING: AttributeIndexSet::Build leaves uncovered gap slots as uninitialized garbage, so every ID up to the max must be covered (they shrink to the authentic 0x0F..0x11 when the parent attribute waves land; SENSOR.HPP pins PoweredSubsystem::NextAttributeID at 0x0F). - MECHWEAP: fire-machine members (fireImpulse/previousFireImpulse, rechargeLevel, rangeToTarget, effectiveRange, estimatedReadyTime, recoil, weaponAlarm 0/2/3/4); CheckFireEdge (@004b9608, BIT-PATTERN int compare -- TriggerState carries raw ControlsButton ints whose release value is a float NaN; IEEE compare would latch the detector shut); ComputeOutputVoltage (@004b9c9c recharge dial). - EMITTER: EmitterSimulation (PARTIAL @004baa88) -- Firing/Loaded/Loading FSM on the weapon alarm, viewFireEnable gate at Loaded->Firing, recoil-decay recharge over the authored RechargeRate seconds; FireWeapon (PARTIAL @004bace8) discharge bookkeeping; ChargeLevel re-pinned to the authentic 0x1D; index re-chained to MechWeapon's. Deferred: electrical charge model (TrackSeekVoltage), heat dump, HasActiveTarget gate, beam/damage submission. - BT_FORCE_FIRE dev hook: trigger pulse whenever Loaded (auto-fire cadence). VERIFIED headlessly: forward view -- PPC_1/2 + ERMLaser_1 cycle FIRED->LOADED at the authored 0.6s/1s cadence, rear lasers silent; BT_FORCE_LOOK=3 -- ONLY the rear lasers (ERMLaser_2/3) fire. The view-fire x fire-FSM interlock holds both directions. Zero Fail. (SRM6 ballistic FSM = a later increment.) Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> |
||
|
CydandClaude Fable 5
|
d422dfffc0 |
BT410 Phase 5.3.7: weapon view-fire gating + build header-stamp hardening
Completes the weapon-side half of the look commit: - MECHWEAP: real rearFiring/viewFireEnable members (binary @0x334/@0x3E0) + accessors. Ctor resolves REAR-FIRING authentically (@004b99a8 tail): the mount SEGMENT site name is tested for the 'b' (back) marker -- only the back gun ports (sitelbgunport/siterbgunport) carry it. Spawn = forward-armed. - MECH CommitLookState: re-arms every roster weapon per view (forward = non-rear, LOOK-BACK = rear-mounted, side/down = none) and flips the HUD reticle pip group with the authentic Reticle::Front/RearFiringWeaponsOn flags (BT411's raw "bits 1/2" resolved to their 1995 names). - MECHWEAP: defined MechWeapon::MessageHandlers (0-entry set inheriting the Subsystem chain) -- it was declared-but-undefined, and the surviving CODE PPC.CPP binds the inherited name into PPC::DefaultData; the zero-filled common block was a latent NULL deref (same trap as Emitter::AttributeIndex). Discovery: the TEST.EGG mech genuinely mounts TWO REAR LASERS (ERMLaser_2/3 on the b-ports). Verified on a full-clean baseline (engine 181 + bt 42, 0 fail): look-behind arms exactly the rear lasers, disarms the other five, pipMask=0xfffffffe; neutral run holds forward-armed; zero Fail. build410.sh HARDENED against the stale-obj layout-skew trap this wave exposed: cc() skipped recompiles on obj-vs-.cpp timestamps only, so the MECHWEAP.HPP member additions left emitter.obj on the OLD MechWeapon layout -- its chargeLevel=0.0f write landed exactly on the new rearFiring (a silent cross-TU struct skew, heisenbug-grade). cc() now honors header stamps: a source410 engine-header edit invalidates every bucket, a BT/BT_L4 header edit additionally invalidates the game buckets. CODE/ is immutable, no stamp needed. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> |
||
|
CydandClaude Fable 5
|
d9b34ddc63 |
source410: THE WHOLE 4.10 TREE COMPILES - 11/11 under the fleet's BC++ 4.52
All 10 surviving original TUs plus the reconstructed BTL4APP.CPP pilot build clean with the authentic OPT.MAK flags (compile410.sh --sweep). First time the 4.10 BattleTech source has compiled since 1996. - layout_probe.cpp: the period compiler measures 1995 layouts from the surviving headers; validated 3/3 against binary alloc sizes (BTMission 0xFC, BTL4ModeManager 0xC, BTRegistry 0x10). Base boundary: Simulation 0xD0 / Entity 0x1C4 / Mover 0x300 / JointedMover 0x328 => Mech-own region 0x328-0x854 (MECH-LAYOUT.md staging worksheet). - 13 staged headers close the family ([T3] reserved[]-parked layouts; interfaces PROVEN by the surviving consumers): MECH, MECHSUB, HEAT, POWERSUB, MECHWEAP, EMITTER, MECHMPPR, BTPLAYER (BTPlayer__MakeMessage = Player's 8 args + roleName/teamName [T1 via BTREG]), PROJTILE, MISSILE, BTL4MPPR, BTL4VID + the round-2 BTCNSL/BTSCNRL. - compile410.sh: PCH gate retired (bt.hpp/mungal4.hpp full include sets), DPL SDK roots on the include line. - BTL4APP.CPP: appmgr include + GetApplicationManager()->GetFrameRate() + ResourceDescription:: scoping; Fail() held at its recorded line 400. - Evidence ledgers: STAGED-HEADERS.NOTES.md, MECH-LAYOUT.md, BACKFILLS, BTCNSL (binary-recovered console wire IDs), BTSCNRL. Remaining for a linkable BTL4OPT: the ~40 missing TU bodies (917-function manifest) - the header skeleton they grow into now exists and compiles. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> |