V2 Minimum Lovable Launch (MLL) — P1 De-Scope Ladder

Status: planning baseline, adopted 2026-06-12. Fills the "no minimum-lovable-launch cut-line inside P1" gap identified by V1_V7_PLAN_SET_AUDIT_2026-06-12.md §6.2.

P1 as defined today (V2/V2_TODOS.md:63-66: §2-§94 + §138) is a single monolithic launch target. This document defines three nested, individually shippable tiers inside P1 so that schedule, cert, or netcode pressure has a pre-agreed answer other than "slip" or "panic-cut." Tier selection is a formal decision with criteria (§4), recorded as an ADR per the phase-cut convention (V2/V2_TODOS.md:73-75, V2/docs/decisions/phase-cut-mll-<tier>.md).

Owner-roles: EP = Executive Producer (tier decision), GD = Game Director (lovability bar), TD = Technical Director (netcode/cert inputs), PD = Production Director (velocity inputs), RM = Release Manager (cert calendar), FIN = Finance Director (pricing implications).

Production arithmetic referenced throughout comes from content-production-model.md (same date); risk triggers from program-risk-register.md.

1. Design rule for every tier#

Each tier must be a coherent fighting game a genre fan would call complete, not a demo:

Every shipped ruleset meets its full per-ruleset feel spec (V2/V2_features.md:431 ff.) — tiers cut which rulesets ship, never how well a shipped ruleset plays.
Per-fighter content floor is never reduced: ≥ 4 specials, ≥ 2 supers, ≥ 2 finishers, ≥ 4 alt costumes, ≥ 3 taunts, full signature library (V2/V2_features.md:1023-1025, :1049-1066). A smaller roster of complete fighters beats a larger roster of hollow ones.
Rollback netcode quality bar is never reduced: ≤ 8 frames at ≤ 100 ms RTT (V2/V2_features.md:212-214) for every online mode that ships.
Compliance, accessibility, parental controls, privacy (§53, §68, §80, §94) and launch-readiness gates (§138) ship at every tier — they scale with content volume but are never deferred.
"Coming soon" panels are banned at every tier (V2/V2_features.md:220-221); a deferred mode simply does not appear in the front end.

2. The ladder#

MLL-0 — "Core Duel" (survival tier)#

The smallest product that is still a lovable, ranked-viable fighting game.

Content:

24 fighters = 2 per archetype × 12 archetypes (archetype list per V2LaunchRosterData.cpp; selection drawn from the 42 originals at :567-608, partner-IP per rights-strategy.md §6 gate).
12 stages: 6 MK-flavor (pits/hellish realms) + 6 SF-flavor (urban/dojo/cathedral) — exceeds the per-ruleset ≥ 5 floor of V2/V2_features.md:1030-1031 for both shipped rulesets.
2 rulesets: MK ruleset + SF ruleset, each at full feel spec including MK finishers (§16) and SF super/Drive mechanics (§17 subset).
Cosmetics: ≥ 200 wardrobe items (pro-rated from the ≥ 500 floor at V2/V2_features.md:1046, scaled by roster ratio 24/60).

Modes: ranked 1v1, casual 1v1, private rooms + server browser (§66), local versus + couch (§91 core), training (§60), combat trials + tutorial curriculum (§61), arcade ladder with per-fighter arcade endings (§19 + the ending slice of §73), tower (basic survival/time-attack ladders from §21), replay theater + spectate (§62).

Online envelope: rollback 1v1 only. No tag, no client-server modes.

Narrative: no cinematic Story mode; per-fighter arcade endings (60 s cinematic + 5 panels + 30 s narration, V2/V2_features.md:1070-1071) carry the lore. 24 endings ≈ 24 finished cinematic minutes — affordable (content-production-model.md §4).

Locales: UI/subtitles in all 18 launch locales (V2/V2_features.md:2445); full VO in 4 locales (EN, JA, FR, LATAM-ES) instead of ≥ 8 — requires a features-promise amendment via the same ADR.

Creator suite: color/edit-level fighter customization only; CAF / CAW / CAS / CAA / CAE / CAM deferred. This breaks the product-promise bullet at V2/V2_features.md:230-232, which is exactly why MLL-0 requires an ADR and a messaging reset (§5).

TODOS sections deferred by MLL-0 (all move to post-launch updates, in this order of return): §11 (WWE system), §12 (UFC system), §13 (SC weapons), §14 (Def Jam), §18 (stipulation library beyond the two shipped rulesets), §20 (Story), §22 (Krypt), §23 (Battle Hub — replaced by §48/§66 lobbies), §24 (World Tour), §25 (Tekken Force), §26 (Devil Within), §27-§35 (Universe, MyCAREER, MyGM, MyFACTION, MyRISE, Royal Rumble, UFC PPV, Chronicles, Def Jam Story), §36 (Crew), §37-§40 (creator suite), §43/§44/§45 scale with shipped content, §51/§79 (esports beyond in-game brackets), §64 (tag), §69 (bonus modes), §74 (keeps Versus/Time-Attack/Score-Attack; defers Wager/Boss), §76 (companion app), §81-§83 (reduced to shipped stages/fighters), §85 (community surfaces minimum), §90 (AI director), §92 (co-op), §93 (marathon/event modes).

TODOS sections that ship in full at MLL-0: §1-§10, §15 (for the 12 stages), §16, §17 (shipped-ruleset supers), §19, §21 (basic), §41, §42 (12 stages), §46-§50, §53-§58, §60-§63, §65-§68, §70, §71, §72 (24 fighters), §75, §78, §80, §84 (account progression core), §86-§89, §94, §138.

MLL-1 — "Tag & Steel" (pressure tier)#

MLL-0 plus the third lineage and team play.

36 fighters (3 per archetype), 18 stages (+6 Tekken-flavor stages with stage breaks/multi-floor, meeting V2/V2_features.md:1032).
3rd ruleset: Tekken ruleset at full feel spec (§9 Tekken layer, §15 stage breaks).
2v2 tag with rollback (§64 + the 2v2 envelope of V2/V2_features.md:212-215) — only if gate G1 (§4) passed for 2v2.
Adds: Crew basics (§36), CAF + CAM (the two highest-retention creator surfaces; CAW/CAS/CAA/CAE still deferred), Battle-Hub-lite social lobby (§23 reduced), Story Act 1 (~45 cinematic minutes) + side stories for the 12 most-played fighters, Krypt (§22), full VO 6 locales.
Defers everything else MLL-0 defers.

MLL-2 — full P1 (plan of record)#

Everything in §2-§94 + §138: ≥ 60 fighters, ≥ 30 stages, all 7 rulesets, full mode/creator/story slate, ≥ 8 VO locales. The only note added here: per-fighter Side Stories at 60 fighters are the single largest content line in P1 (~720 cinematic minutes, content-production-model.md §4) and are the designated first internal cut within MLL-2 — i.e., MLL-2a ships side stories for 16 launch fighters and back-fills seasonally — before the program drops a whole tier.

3. Tier cost envelopes (from `content-production-model.md`)#

Tier	Fighters	Stages	Roster effort	Stage effort	Cinematics	Indicative total content effort
MLL-0	24	12	432 pm	90 pm	24 finished min ≈ 24 pm (endings only)	~546 pm ≈ 46 person-years
MLL-1	36	18	648 pm	135 pm	~225 min ≈ 225 pm (36 endings + Act 1 + 12 side stories)	~1,008 pm ≈ 84 person-years
MLL-2	60	30	1,080 pm	225 pm	930 min ≈ 930 pm uncut / 402 min ≈ 402 pm as MLL-2a	~1,707-2,235 pm ≈ 142-186 person-years

(pm = person-months; systems/engineering effort excluded — it is largely tier-invariant for whatever ships, see market-rationale.md §5.)

4. Tier-selection criteria and decision gates#

Definitions: GM = gold-master date. A "completed fighter" = passes the per-fighter signature checklist (V2/V2_features.md:1064-1066) and the QA cert pass. Steady-state roster velocity at planned staffing ≈ 2.5 completed fighters/month (planning assumption adopted 2026-06-12, derived in content-production-model.md §5: 45 roster FTE ÷ 18 pm/fighter).

Gate	When	Question	Outcome rule
G1 — netcode	GM − 18 mo	Has rollback milestone RB-VS1 (2-fighter vertical slice, 8-frame rollback @ 100 ms RTT — `program-risk-register.md` R-01) passed? Has the 2v2 extension passed?	RB-VS1 red → freeze roster production at 8 fighters, all tiers slip; 1v1 green but 2v2 red → tag is cut from the target tier (MLL-1 ships without tag, MLL-2 drops to MLL-1 content + extra rulesets).
G2 — content velocity	GM − 12 mo	Completed-fighter count N. Capacity adds ≈ 2.5/mo × 8 mo = 20 more by GM − 4 (content-complete).	N ≥ 40 → MLL-2 viable. 24 ≤ N < 40 → MLL-1 (needs 36 by GM − 4; 24 + 20 = 44 ✓). N < 24 → MLL-0 (needs 24; if N < 8, GM moves — below 8 even MLL-0 fails).
G3 — cert & licensing	GM − 9 mo	(a) Rating-board status for shipped gore content per region; (b) `V2/legal/platform-cert-bans.json` next review (currently 2026-08-11) — any Aje/Web3 surface still banned on a target platform; (c) partner-IP licenses executed per `rights-strategy.md` §6.	(a) red in a launch region → regional content adjustment, not tier change; (b) any ban active → Aje surfaces compile-time-excluded from console/Apple builds (already the plan, R-03), zero tier impact; (c) no license → fallback roster names per `rights-strategy.md` §5 — zero tier impact because tiers were defined on originals.
G4 — lock	Beta (GM − 6 mo)	Tier locked by EP + GD + TD sign-off, recorded in `V2/docs/decisions/phase-cut-mll-<tier>.md`.	After G4 the tier may only move down, never up; content finished beyond the locked tier banks for post-launch drops.

Date pressure is handled by the gates, not ad hoc: an immovable GM date makes G2's arithmetic the binding constraint; a movable GM date means the EP may instead slip GM by up to 2 quarters once (and only once) to hold a higher tier — a second slip forces the tier drop.

5. Pricing and messaging implications per tier#

(Consistent with editions at V2/V2_features.md:1503-1506 and PPP at :1755-1761; details in market-rationale.md §6.)

MLL-2: $69.99 Standard / $99.99 Deluxe / $119.99 Ultimate as specced.
MLL-1: $59.99 Standard; deferred rulesets/modes announced with dated (not "soon") free-update roadmap; Deluxe bundles Year-1 fighters.
MLL-0: $49.99 Standard (planning assumption adopted 2026-06-12: 24-fighter/2-ruleset content parity with genre standard editions does not support $69.99 against SF6/T8/MK1 comparables), with a paid upgrade path that credits owners when MLL-1/2 content lands. Marketing repositions from "seven lineages, one game" to "the duel core, expanding" — BMD owns the reframe; the seven-lineage promise is only spoken once MLL-2 content is dated.
All tiers: deferred-content roadmap is published with quarters, and every deferred TODOS section keeps its [P1] phase tag with an mll-deferred annotation in the ADR — deferral is a release decision, not a backlog deletion.