Darkness Falls — Class Identity & Distinctiveness (Initiative)

Status: ADOPTED initiative design home (owner 2026-07-31, human_question 25288) — living home for class distinctiveness. Act I/signature contract: act-1-class-signatures.md (implementation + proof). Live class sheets and formulas remain authoritative in index.md § Class System and progression.md. This doc owns problem framing, lever toolbox, and phased rollout; it does not claim design authority over live formulas until each ship item lands (live sheets stay in index.md and progression.md).

Adoption note: Phase 1 clarity and Phase 2 felt identity (L1 signatures for all 8 classes, automatic playstyle) are largely local_proved on the Act I pass. Remaining work is Phase 3 (deeper caster divergence / Star Map lens / class-mastery multi-character meta) plus prod_proved deploy of Act I signatures.

Origin: Player Dragonfly (cleric): "there's not really a way to raise wisdom or cleric based stuff in the star map — there's paths for agility, strength, constitution, and intelligence, but no wisdom path." Owner directive (2026-07-17): "this brings up a great concern beyond just cleric where we really should think about how to actually make classes feel distinct and alive and truly beneficial. We should expand to make the choice of class feel meaningful in play style and also make people want to make multiple characters." This doc is the design home for that initiative; the cleric/wisdom gap is its first case study.

The problem — why classes don't feel distinct today

Five compounding causes, all verifiable in the current specs:

1. The archetype collapse. 4 of the 8 classes are intentional stat-identical variants of the original 4 — same primary stat, same resource, same per-level growth — differentiated only by which paragon paths unlock at level 50 (index.md § Class System):

   Original	Archetype twin	Shared	Diverge only at
   Fighter (STR/Stamina)	Barbarian	STR, Stamina, +1 move innate	L50 paragon
   Rogue (AGI/Energy)	Ranger	AGI, Energy	L50 paragon
   Sorcerer (INT/Mana)	Necromancer	INT, Mana	L50 paragon
   Cleric (INT/Mana)	Paladin	INT, Mana, +shop discount innate	L50 paragon

   For the first ~50 levels a Barbarian is a Fighter, a Paladin is a Cleric.

2. Only three primary stats, so the two casters are the same character. Growth is STR, AGI, or INT (progression.md § Stat Growth). Sorcerer, Necromancer, Cleric, and Paladin are ALL "INT / Mana." There is no Wisdom, no Faith, no Spirit — so a cleric and a sorcerer have identical stat sheets. This is the direct, correct root of Dragonfly's "no wisdom path" observation: the game has no wisdom to give a path to.

3. The Star Map has no class identity. Its four paths are combat-generic — North Defense, South Offense, East Elemental, West Utility (star-map.md). A cleric walking in sees no healing / faith / support branch; the only "caster" nodes are shared INT/elemental nodes every caster uses. Nothing on the tree says "this part is for you."

4. Identity is back-loaded to L40/L50. The real class fantasy only arrives at the Promotion (level 40, and only for the original 4 base classes) and Paragon (level 50) tiers. Levels 1–40 — the whole early game most players actually experience — are where classes feel most interchangeable.

5. Innate differences exist but are invisible. Each class has a small innate exploration bonus (index.md § Class Innate Exploration Bonuses: Rogue traps, Cleric/Paladin shop discount, Sorcerer/Necromancer avoidance, etc.), but they're passive percentages the player never feels moment-to-moment, so they don't read as identity.

The multi-character disincentive

The owner also wants players to want multiple characters. Two current facts work against that:

• The roster cap is 3 (MAX_CHARACTERS_PER_ACCOUNT, characters.md) — fine.
• But account-wide power is shared: Trophy Room bonuses, cosmetics, and the stash all live on the account and apply to every character (characters.md, trophy-room.md). Combined with barely-different classes, a second character is "the same game with a different name and a redundant grind," not a fresh experience. There is no class-specific unlock, achievement, or account meta that a new class earns you.

Design goals

1. Felt identity from level 1, not level 40 — every class plays differently in the first hour.
2. Each of the 8 classes has a one-line fantasy a player can name, and a signature lever that delivers it in play.
3. The two casters diverge — a cleric is a support/healer, a sorcerer is a nuker; they should not share a stat sheet.
4. De-collapse the archetypes — Barbarian ≠ Fighter in feel, even before paragon.
5. A concrete reason to roll a second character — class-specific progression the account tracks.

Case study #1 — the Cleric / Wisdom gap (owner decision 2026-07-17: FOLD IN)

Per the owner, we are not inventing a Wisdom stat in isolation — the cleric gap is the first worked example of the broader initiative. Options considered:

Option	Verdict
A. Add a real Wisdom stat + a Star Map faith path	Deferred. Threads a 4th stat through every gear roll, node, and formula — a large change to make for one class. Revisit as a Phase 3 lever if the initiative decides casters need a dedicated stat.
B. Keep INT, give the Cleric a distinct SUPPORT identity + surface it (chosen direction)	Cleric stays INT-scaled, but gains a felt healer/support identity (see levers) AND the Star Map / class copy clearly names INT as the caster stat so a cleric player can find their scaling. Ships value without a stat-model rewrite.
C. Do nothing / tell the player "it's INT"	Rejected — it answers the letter of the question but leaves the real "clerics feel like sorcerers" problem the owner flagged.

Immediate low-risk clarity fix (can ship independent of the big pass): the base-class selection prompt already carries a one-clause playstyle hint per class (progression.md changelog 2026-05-21); extend that so a cleric is told "clerics channel Intelligence into healing and support," and add an INT-caster note to the Star Map entrance readout. This directly closes Dragonfly's confusion while the deeper identity work is designed.

The lever toolbox (how to make classes distinct)

1. Signature mechanic at level 1. Give each class one defining verb from the start (e.g. Cleric: a heal/ward; Rogue: a stealth/opener; Barbarian: a rage/lifesteal pulse; Ranger: a pet/mark). Pull the identity forward even if the numbers still scale into promotions.

2. Make the resource matter. Mana / Stamina / Energy are near-cosmetic today. Differentiate how each is spent/regened so an INT class and a STR class play on a different rhythm.

3. Diverge the casters via secondary/tertiary stat + ability scaling. Even inside "INT," a Cleric can scale healing/support off a support secondary and a Sorcerer scale burst off a damage secondary — distinct sheets without a new primary stat.

4. Class-flavored Star Map. Either a small per-class node cluster near center, or a "class lens" that tags the tree's relevant nodes ("Support" nodes highlighted for a cleric), so every class sees a path that is theirs. Longer-term: a dedicated faith/support branch (pairs with Option A).

5. De-collapse the archetypes early. Give Barbarian/Ranger/Necromancer/Paladin a distinct L1 innate or signature mechanic, not just different L50 paragon paths.

6. Per-class fantasy (target one-liners):

   Class	Fantasy	Candidate signature lever
   Fighter	Reliable frontline anchor	Guard/taunt, highest sustain
   Barbarian	Reckless berserker	Rage: damage up as HP drops + lifesteal
   Rogue	Precision assassin	Stealth opener, crit burst
   Ranger	Hunter with a companion	Pet/beast + mark-target
   Sorcerer	Elemental artillery	Big single-target/AOE nukes
   Necromancer	Attritional death-caster	Summons / drain / DoT
   Cleric	Support healer	Heal/ward, buff, anti-undead
   Paladin	Holy bruiser	Melee + self-sustain smite

Multi-character incentives

• Class mastery meta: the account tracks per-class progress; leveling a new class unlocks a cosmetic, title, or small account-wide perk — a reason to roll class #2/#3.
• Class-specific achievements/trophies distinct from the shared boss trophies.
• Seasonal nudge: a season objective or reward for playing a class you haven't before (ties into seasons.md).
• Roster synergy (longer-term): party/warband bonuses when characters of complementary classes group, giving a reason to keep a second class ready.

Rollout phases

• Phase 1 — clarity (now, low risk): cleric INT-scaling copy in class-select + Star Map; extend the playstyle hints per class. Closes Dragonfly's confusion immediately.
• Phase 2 — felt identity (medium): L1 signature mechanic per class + resource differentiation + de-collapse the archetypes' early game.
• Phase 3 — deep systems (larger): caster stat divergence (or a real Wisdom/Faith stat), class-lensed Star Map, and the class-mastery multi-character meta.

Open questions for the owner

1. Stat model: keep 3 primary stats and diverge casters via secondaries/ability scaling (lever 3), or commit to a 4th caster stat (Wisdom/Faith) later (Option A)?
2. Scope of Phase 2: all 8 classes at once, or pilot the two most-collapsed pairs first (Cleric/Sorcerer casters; Fighter/Barbarian melee)?
3. Multi-character meta: is a class-mastery account track the right carrot, or do you prefer season-driven "try a new class" incentives?

Refs

• index.md § Class System — the 8 classes, 12 promotions, 24 paragon paths, innate bonuses
• progression.md § Stat Growth, § Promotion Classes, § Paragon Paths
• star-map.md — the 4 combat-themed paths (no class identity, no wisdom)
• characters.md — roster cap 3, account-shared Trophy/cosmetics/stash
• combat.md — per-class weapon prefs + audio identity (existing differentiation to build on)
• Origin: Dragonfly cleric feedback + owner class-distinctiveness directive, 2026-07-17