Skip to content

feat(port): Brute plays its seven own moves and tires - #117

Merged
sp00ktober merged 2 commits into
mainfrom
port/brute-own-moves
Oct 22, 2026
Merged

sp00ktober merged 2 commits into
mainfrom
port/brute-own-moves

Conversation

@sp00ktober

@sp00ktober sp00ktober commented Oct 21, 2026 •

Copy link
Copy Markdown
Contributor

Stacked on #116. Brute plays its seven moves the Tigrex has no counterpart for as own moves derived from the donor ("own"), and tires (fatigue = true).

  • mhfu-port splice: 22 routines, +1649 bytes, worst run 93 of 999; own moves in slots 1-9 (p3_7_13 6, p3_7_14 7, p3_7_15 9, ...); fatigue 28 pair costs and 3 own moves; base_13's tired picker is back.

  • [moves.charge] goes: it painted the (7,13) clip on em75's leave-the-area pair (2,8), which phase 3 used to carry (7,13), so the monster sometimes flew off instead of attacking.

  • Left out with a warning: donor record 26 (the rush's hind feet) has no Tigrex record; [brain.attacks] "26" = <id> with a port-defined [[attack]]/[[hitbox]] would add it.

  • Live (lane 3, the stack built into the PRX, port and brain from the stick, 1280 s of combat):

    • the stick container ran (5113 of 5113 script dispatches inside it);
    • all 57 enter 16 k played the right own move 0.07 s later, chain steps followed, and the script picked again after each (91 starts, none cut by a pair change);
    • attacks spawned at the donor's frames (record 7 at 136, 8 at 138, 15 at 86, 4 at 76) and landed; mhfu hits: 69 spawns, 23 hits, 981 damage in 180 s;
    • (7,13) turned in clip frames 10-40 only (89 of 89 inside, 0 of 794 outside);
    • fatigue: every stamina change equal to the derived cost, none while enraged, two complete tired spells of exactly 2250 frames (a third still running when the window closed), then 500; base_3 took the tired picker 20 of 20 times while tired and the other 149 of 149 otherwise.
  • Fixed after the run: the rush's stop-and-slam step ended on the rush's wall after 2 frames; a derived step now ends at a wall only on a looping clip.

  • Found, not from this stack: a second quest without a cold boot latched em_vhook onto a Popo (the registry poll saw no new Tigrex pointer); fixed in fix(framework): a second quest re-reports its monsters, and em_vhook latches only a big one #121 (on main).

Tests: the splice through the expanded manifest (own enters, if_gauge_a_low, Built.ok), the CLI lines, walls per step.

🤖 Generated with Claude Code

sp00ktober and others added 2 commits October 21, 2026 22:58
The seven donor pairs with no Tigrex counterpart are derived own moves (`own`), and fatigue is on,
so the tired picker returns. `[moves.charge]` goes: it painted the (7,13) clip on em75's
leave-the-area pair (2,8), which phase 3 used to carry (7,13) and which flew the monster off.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Live (lane 3), 12 of 19 Brute rushes (7,14) ended on a wall and the chain's stop-and-slam step,
whose clip does not travel, then ended on the same wall after 2 frames, so its two attacks
played in 6 of 19. The donor ends no move at a wall: a run (a looping clip) stops there, a
one-shot step plays out.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
@sp00ktober
sp00ktober changed the base branch from port/fatigue to main October 22, 2026 21:58
@sp00ktober
sp00ktober merged commit 9338fc1 into main Oct 22, 2026
5 checks passed
@sp00ktober
sp00ktober deleted the port/brute-own-moves branch October 22, 2026 21:58
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant