Entry 304

Kyvia's second fight ported: ccBossKyvia02, its two stages and thunderbolts; event 211 done

2026-09-29battlevolumes

Filescrates/piney-battle/src/boss/kyvia.rscrates/piney-battle/src/boss/kyvia/thunder.rscrates/piney-battle/src/boss/kyvia/core.rscrates/piney-battle/src/boss/kyvia/gomora.rscrates/piney-battle/src/boss/kyvia/tests.rscrates/piney-battle/src/boss.rscrates/piney-battle/examples/battle_probe/kyvia.rscrates/piney-battle/examples/battle_probe/boss.rscrates/piney-gen/src/manifest.rscrates/piney-data/src/tables/combat.rscrates/piney-data/src/pack.rscrates/piney-world/src/combat/boss.rscrates/piney-game/src/fx.rstools/test_battle_kyvia_rs.pydocs/engine/boss-kyvia-second.mddocs/engine/boss-kyvia.mdplans/outbreak-story.md

Outbreak's event 211 stopped at field 10's entry type=7 code=13. OUT bossFunc (gcmn 0x00616d60) entry 13 is 0x004d5870, a task that makes new ccBossKyvia02(2) (0x29740 bytes, the string "Kyvia02") and runs its Main. So row 13 is Kyvia's second fight, not another boss: the plan's "Cubia, the second Kyvia" was half right. Cubia is the English bossTbl's name for Kyvia (rows 12 and 13, "Cubia Core" for 32-39).

Finding the code

OUT's sidecar names only the constructor. The class's functions follow Infection's order (its syms list ccBossKyvia02's 23); the prologues between 0x004d5830 and 0x004de150, the jal targets into them and the vtable (0x00385560: +0x18, +0x1c Main 0x004d7300, +0x8c, +0x90, +0x94) place each, and the same for OUT's ccBossKyvia01 (Main 0x004d0780). The disc's Move and IsMove are 0x00419c80 and 0x00419d10 (called from CheckDiscMove), KyviaGenerator 0x0061d0d0 (0x690 below KyviaDmgGenerator, as in Mutation).

Diffing OUT's ccBossKyvia01 against OUT's ccBossKyvia02 (the same compiler) with member offsets mapped between the layouts gave the change list: Main, Think's switch but act 6, SwitchDeadCheck, SetBPos, Slave, EntrySlave and DeadKyviaVoice are the same code; the rest differ in constants and a few steps; ThunderboltAtk, ThunderDmg and ThunderSound are new, with ccBossEffThunderbolt (OUT gcmn 0x00477410-0x00477ef0). OUT's layout is Infection's with a byte at +0x29350 (the cinema flag), the next three pointers 4 on and the rest 0x10. Details are in docs/engine/boss-kyvia-second.md.

First, the first fight on Outbreak

The harness took a fight (PINEY_KYVIA_FIGHT) and Outbreak's unnamed addresses. Outbreak's copy of the first fight agreed with the port on its first three cases, so OUT's core and gomoras at level 1 are Mutation's.

The port

kyvia::Fight (First, Second) picks the numbers where the code differs; kyvia::new(cx, fight). kyvia::thunder holds act 6 and the bolt, whose Draw draws ccRand each frame (a stroke: ccRand, two ccRandF(pi) unused, the turn, three offsets; two more a segment), so the kyvia manager pass runs it as it runs the meteors. Found on the way:

• The ride's camera looks at the disc itself in the second fight (SetFreeCamPosView(eye, DiscPos)), 150 over it in the first.
• DmgGp[k] past four reads DeadGp and then CamRotX: with SwitchHp a seventh, the fifth smoke fills DeadGp and there is no sixth.
• SetCoreBaseParam at level 2 adds 3000 to both rows' maxHP each call while the core is dead. The port added it once, to the HP only, so a later change of attribute would read the row without it. Now Core::row_hp carries it, and the harness restores bossTbl for each case (the game's table had kept the first case's 3000).
• The gomoras' lists by level and death: AllGomoraList_2 (two steps) in the manifest (kyvia_gomora_lists_2, also kyvia02_anims), ListStepUp(n) taking step n where it is not null. DATA_VERSION 19 to 20.

Checks

PINEY_VOLUME=outbreak python3 tools/test_battle_kyvia_rs.py bulk: the first case after the port agreed; the next failed only on bossTbl's kept HP (above). After that, 100 cases (320,004 frames, seed 9500) and, on the final code, 30 more (85,575 frames, seed 9600) agree: the body's acts 0-6 and 14, thunders 0-2, the core's two deaths. Mutation's first fight, 25 cases, still agrees. Cases of the second fight run up to 6000 frames with hits to 1200 so the core dies twice.

PINEY_SURVEY_ONLY=211 PINEY_SURVEY_GOD=1 PINEY_SURVEY_FRAMES=150000 cargo test --release -p piney-game outbreak_story_survey -- --ignored --nocapture: story 211: done. The fight starts near frame 11,650; the first core (2000 HP) is down near 16,500, the second (5000) near 22,000. The pilot needed nothing new: Kyvia's parts were already its focus.