grom reports `/cast [noslamclip] Slam; [slamclip] Heroic Strike` never picking
Slam. Yesterday I drove the whole chain in the harness, measured a 44% no-clip
window and called it healthy. The 44% came from a cast time I supplied:
s.set_spell_tooltip(1464, … cast_time: Some("1.5 sec cast") …)
**OctoWoW's Slam is a 2.5 s cast.** `Spell.dbc` ids 1464/8820/11604/11605 all
carry `CastingTimeIndex` 20 and `SpellCastTimes.dbc` row 20 is 2500 ms, where
vanilla's Slam is 1500. SuperCleveRoidMacros' own
`DEFAULT_SLAM_CAST = 2.5 -- Turtle WoW` agrees independently. Against his 2.70 s
weapon the window is `(2.70 - 2.50) / 2.70` = 7.4%, 0.2 s wide — not something a
hand can hit, which is the whole report.
The same DBC gives every Slam rank `spellLevel` 54, and he is level 14. The
field alignment is checked against spells whose levels are known (Heroic Strike
rank 1/2/3 = 1/8/16, Charge 4, Shield Bash 12, Overpower 12, Thunder Clap 6),
so that reading is safe to act on: 477 Heroic Strikes in his combat log and zero
Slams, in any session since 10/12.
The fixtures now carry the realm's numbers and `WINDOW_PCT` carries where they
come from, including this mistake — a number I supply cannot test anything, and
a green test then argues against the player's own report.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01SgstFWtn1RxEJ5kVrpgYwn