The Banked Fire
A deep dive into Shroud of the Avatar's Fire school, what twenty-five years of fire design since Ultima Online can teach it, and twenty-four concrete proposals.
Fire is not a weak school. By our own balance audit it is the strongest of the four elemental schools, and the recent Chaos Magic review used Fire's structure as the benchmark every school should meet. This document argues something different: that Fire is shallow where it should be deep, that most of its depth already exists in the data but is invisible at the moment of play, and that twenty-four targeted changes, almost all of them data changes, would make it the school you master rather than the school you repeat.
Fire is the school the rest of the magic catalog gets measured against. The 2026 balance audit ranked it the strongest elemental school, roughly twice Water's sustained output, and the recent Chaos rework used Fire's seven actives, five passives, and ward as the structural benchmark every school should meet. Nothing on this page argues that Fire is weak, and nothing on it asks for more damage. The argument is that Fire is shallow where it should be deep: the rotation reduces to casting the same damage-over-time pile in any order, the mechanics that could reward skill (a combo state wired into six spells but reachable in practice only through one rare artifact, a burn-speed tradeoff buried in the math, resistance debuffs nobody can see) are invisible at the moment of play, four of its seven spells participate in no combo, and against the heavily resistant creatures where an elemental school should matter most the audit measured its output at effectively zero. This proposal spends no power budget and stays inside the balance program's bands. It takes the numbers Fire already has and makes them visible, buildable, and worth mastering.
Measured against the genre (the full tour is Section 2), our gaps cluster in three places:
Twenty-four proposals follow. Twenty of them are pure client data work: spell asset edits, skill table cells, channel entries, and test pins, with every number routed through the balance program's reviewed changeset process. None adds server-side systems. The keystone is a pair: S12 makes burning a visible, stackable state, and S13 adds Combust, the spell that spends it.
| # | Proposal | Tier | Size | Gate | One-line pitch |
|---|---|---|---|---|---|
| S1 | Give fire its voice | 1 | S | none | audio for the silent charge and projectiles |
| S2 | Endings for the fields | 1 | S | none | wire up the finished expiry visuals |
| S3 | Meteor Shower consistency | 1 | S | changeset | a Lunar spell stops scaling off Fire training |
| S4 | Right summon animation | 1 | S | none | the fire elemental arrives by fire |
| S5 | Honest cleanse rules | 1 | S | none | burns are cleansable; the tooltips finally say so |
| S6 | One burn, one rhythm | 1 | S | changeset | burn timing aligned across all four spells |
| S7 | Water Warding earns its place | 1 | S | changeset | the ward gets a real prerequisite |
| S8 | Say what the skills do | 1 | S | none | tooltip truth pass across the school |
| S9 | Pin the school down | 1 | S | none | the test net everything else stands on |
| S10 | Combos for the combo-dead four | 2 | M | changeset | every fire glyph combines with something |
| S11 | Fireboost lives | 2 | M | changeset | the school's own heat, earnable in the school |
| S12 | Burning becomes visible | 2 | L | shared data | burn pips on the target; the keystone state |
| S13 | Combust, the missing spell | 2 | L | changeset | detonate your burns at the moment you choose |
| S14 | Ignite joins its siblings | 2 | M | changeset | one weapon enchant slot, a real choice |
| S15 | Immolation gets its own slot | 2 | M | changeset | the aura stops costing you your shield |
| S16 | A pyromancer's passive | 2 | M | changeset | conditional mastery in the Sun Worship mold |
| S17 | A fire stance | 2 | L | changeset | hold the heat, pay the price |
| S18 | Resistance that resists | 2 | L | changeset | one honest lane for fire defense |
| S19 | A tax with counterplay | 2 | M | changeset | the specialization's weakness becomes situational |
| S20 | One knob, one job | 2 | M | changeset | area and burn pacing stop sharing a stat |
| S21 | Burns cook through bone | 3 | L | changeset | fire becomes the answer to the restless dead |
| S22 | Ember patches | 3 | XL | changeset | kills while burning leave scorched ground, strictly capped |
| S23 | The fire-typed spine | 3 | L | changeset | fire spells outside the tree join the system |
| S24 | Flames that grow with you | 3 | L | none | presentation that scales with mastery, plus a tint option |
Fire is twelve skills: seven castable spells and five passives, two branches meeting at Ring of Fire and crowned by the elemental summon. It is a complete school by any structural measure, and the recent Chaos Magic review explicitly held it up as the shape other schools should reach: a full active kit that works from level one, a passive for every dimension the school cares about, and a ward. The balance program's audit backs the structure up with output: the strongest sustained damage of the four elemental schools, with the most damage-dealing spells of any magic school.
Every proposal on this page plugs into the same pipeline. A glyph is played from the hand, targeting and validation run (range, focus, reagents, the fizzle roll), the cast winds up with its charge effects, and the spell's effects apply: a direct hit, a burn over time, a field, a buff. The burn is the school's signature: every one of the seven spells applies one. What no step of this pipeline does today is remember, show, or reward that fact.
flowchart LR A["Glyph played\nfrom the hand"] --> B["Targeting + validation\nrange, focus, reagents, fizzle"] B --> C["Cast\ncharge effects, cast time"] C --> D["Effects apply\nhit + burn + field + buff"] D --> E["Burn ticks on the target\ninvisible today"] style D stroke:#c9a55a,stroke-width:2px style E stroke:#e0703c,stroke-width:2px
Credit where due, because this page builds on it rather than re-proposing it: the Frost Weapon and Shock Weapon enchants shipped this month and proved the school-extension recipe end to end; the balance program has a working audit, published target bands, and a reviewed changeset process for landing numbers; the combat HUD is moving to UI Toolkit window by window; and the Chaos Magic review established the pattern of treating a school as a design surface. Fire inherits all of that machinery for free.
Fire magic's genre history starts where Shroud's own lineage starts. In Ultima Online, fire was never the biggest number; it was the school of position and timing. Fire Field turned corridors and doorways into weapons, ticking interrupts on anyone who stood their ground. The Explosion spell carried a two-and-a-half-second fuse, and the entire skill ceiling of UO PvP lived inside that window: land Energy Bolt at the exact moment the fuse resolved and the two hits struck as one, faster than any healer could answer. Defense was a choice with a price: Resisting Spells consumed a full skill slot that could have been offense, and players trained it by deliberately standing in fire. Flamestrike, the pillar of flame, was the spell that announced a grandmaster before the damage even registered.
Twenty-five years later the genre has converged on a grammar UO only gestured at, and it has three parts. First: burning is a state, not just damage. Guild Wars 2 shows burn stacks as a number every player can read on the enemy's bar; Path of Exile made "your biggest hit burns" a build archetype with a whole stat lane that only touches the burn; ESO unified every flame source, staff, spell, or weapon glyph, into one standardized Burning status that all downstream effects key off. Second: fire is the build-and-spend school. WoW's fire mage turns two crits into Heating Up, then Hot Streak, then a free instant Pyroblast, and the screen itself changes temperature as the loop charges; FFXIV's Black Mage carries an entire job identity on one heat gauge where holding maximum fire is a wasting asset that costs resources to sustain. Third: counterplay is legible. Magicka made wet-beats-fire a deterministic table any player could learn in a tutorial; UO priced resistance in build slots; PoE sells a flask that douses the burn and grants immunity for four seconds.
And one game marks the outer boundary: Divinity Original Sin 2 built the maximal fire model, where every surface burns, oil ignites, water steams, and fights are terrain-state puzzles. It is glorious in a four-actor turn-based party game, and its own designers pulled back from it in their next game. The lesson this page takes is the road-not-taken lesson: fire states belong on characters, where they replicate cleanly and read instantly, not painted on the ground of a scene with dozens of concurrent actors.
Ultima Online: fire fields as choke-point control, the Explosion fuse as hand-timed detonation, Resisting Spells as defense with an opportunity cost, Flamestrike as the prestige visual. WoW: the two-stage proc ladder (Heating Up, then Hot Streak, then a free Pyroblast) with a guaranteed-crit converter that turns luck into a decision, and a rolling Ignite that made every crit feed one growing burn. FFXIV: the Astral Fire gauge, proof one heat meter can carry an entire class, with heat that halts your resource regen while you hold it and coolant you bank in the cold phase. Path of Exile: ignite as a first-class ailment scaled from the hit that applied it, burn spreading as an opt-in modifier rather than terrain, and Righteous Fire as the self-burn archetype. GW2: burning that stacks in intensity with a visible counter, and the combo-field grammar where any player's finisher through any player's fire field produces a deterministic bonus. D:OS2: Spontaneous Combustion, the cleanest consume-the-burn detonator in the genre, inside the terrain model this page declines. Magicka: the small, public, exceptionless interaction table as the whole counterplay design. ESO: standardized Burning fed equally by spells and weapon glyphs, ground DoTs with capped budgets, and Wall of Fire hitting burning enemies harder, the bounded amplify-the-burn pattern.
| Capability | UO | WoW | FFXIV | PoE | GW2 | D:OS2 | Magicka | ESO | SotA today | SotA proposed |
|---|---|---|---|---|---|---|---|---|---|---|
| Fire damage-type identity | ● | ● | ● | ● | ◔ | ● | ● | ● | ● | ● |
| Burn as a visible stacking state | ○ | ● | ○ | ● | ● | ◔ | ◔ | ◔ | ◔ hidden | ● S12 |
| Build-and-spend heat loop | ○ | ● | ● | ◔ | ◔ | ○ | ○ | ◔ | ○ unreachable | ● S11 |
| Detonation / consume-the-burn | ◔ fuse | ◔ | ◔ | ◔ | ○ | ● | ○ | ◔ | ○ | ● S13 |
| In-school spell combos | ● | ● | ● | ● | ● | ● | ● | ● | ○ zero | ● S10 |
| Cross-school combo grammar | ● | ◔ | ○ | ◔ | ● | ● | ● | ● | ◔ three | ● S10/S23 |
| Ground fields / zone control | ● | ● | ◔ | ● | ● | ● | ● | ● | ◔ two fields | ◔ by design |
| Fire spread between targets | ○ | ● | ○ | ● | ○ | ● | ◔ | ○ | ○ | ◔ S22 |
| Weapon ignite integration | ◔ | ◔ | ○ | ● | ◔ | ● | ● | ● | ◔ 1 of 3 | ● S14 |
| Vs-burning hooks | ● | ◔ | ○ | ● | ◔ | ◔ | ○ | ◔ | ○ | ◔ S13/S16 |
| Legible defensive counterplay | ● | ◔ | ○ | ● | ● | ● | ● | ● | ◔ muddled | ● S5/S18/S19 |
| Anti-resist niche in hard PvE | ◔ | ○ | ◔ | ● | ○ | ○ | ◔ | ◔ | ○ flatlines | ● S21 |
| Risk-cost fire | ◔ | ○ | ● | ● | ◔ | ● | ● | ○ | ◔ flat tax | ● S15/S17/S19 |
| Feedback tiers with investment | ● | ● | ● | ◔ | ◔ | ◔ | ◔ | ◔ | ○ silent spells | ● S1/S24 |
● native · ◔ partial or conditional · ○ absent. "SotA today" = the live Fire school on the migration branch, 2026-07. Row definitions and per-game evidence in the research notes accompanying this proposal.
Several fire spells are quiet in ways players notice: the shared charging effect most of the school uses makes no sound at all, and both fire projectiles fly silently. We wire crackle, roar, and impact audio into the effects that lack it, reusing sounds the game already ships. The most aggressive school in the game should be audible before it is visible.
Seen in: every action game since Doom; ESO and WoW both treat fire audio tiers as core class feedback.
The shared charge effect prefab used by most of the school carries no audio sources today, and the Fire Arrow and Flame Fist travel effects are also silent; Ring of Fire and the Fireball impact are properly voiced, so the fix is additive AudioSource wiring on existing prefabs with existing clips, then a mix review. No new prefabs, so load cost is unchanged. A small EditMode test pins that every fire spell's charge, travel, and impact effects carry at least one audio source so this cannot regress.
Immolation and Ring of Fire vanish without a trace when they expire. Finished end-of-effect visuals for both already exist in the game files; they are simply not connected. We wire them up so fires gutter out instead of blinking off, and give the fire ward's activation its idle visual back the same way.
Seen in: GW2 and FFXIV, where every timed effect telegraphs its own expiry.
Several completed fire effect prefabs have no references from any spell today (aura endings, the ring's burn-out, a ward flash). Connecting them is an end-effect and lasting-effect field assignment on the existing spell assets. Verify the orphaned prefabs are still in the shipped bundle set before referencing; if any dropped out, they re-enter through the normal effect-group registration. Pins: a wiring test asserting the field assignments resolve.
Meteor Shower is a Lunar school spell, but its casting range currently grows with Fire school training instead of its own. We correct it so the spell reads from the school it belongs to. A consistency fix, not a rework: the spell keeps its role, and Lunar mages stop needing a detour into Fire to reach further.
Seen in: standard cross-school hygiene; WoW and ESO patch notes are full of exactly this class of correction.
One range expression on the Meteor Shower spell asset, edited through the reviewed balance-changeset process with the before and after recorded. Ships in the same changeset batch as S6 and S7. The pin suite (S9) captures current behavior first so the change is a deliberate diff, not a drift.
Summoning a Fire Elemental currently plays the earth-magic casting animation. We switch it to the fire cast so calling your elemental looks like what it is.
Seen in: baseline presentation polish everywhere.
One animation state name on the summon spell asset. A data-contract test asserts the fire summon uses a fire cast state so a future asset pass cannot regress it.
Cleansing magic can strip burning off a target today, but nothing in any tooltip says so. We make the rule explicit and visible: cleanse tooltips state that burns can be removed, burn tooltips state that they are cleansable, and the cleanse list itself gets a tidying pass so what it removes is exactly what it says it removes. Counterplay only counts if players can discover it.
Seen in: GW2's condition-cleanse clarity; Magicka's douse-the-burn grammar as the readable extreme.
Tooltip strings on the cleansing spells and the burn effects, plus a de-duplication pass on the shared cleanse ailment list (one entry appears twice; the tidy-up is behavior-preserving and pinned by a new list-integrity test). Whether burns should remain cleansable at all is a design question S12 makes more interesting; if that answer ever changes, it ships as its own balance changeset. Keep the cross-school rule consistent with the recent Chaos work, which made its damage-over-time effects cleansable on purpose.
A burn timing correction was applied to Fire Arrow earlier this year but never carried to Fireball, Immolation, or Ring of Fire, so the same burn behaves differently depending on which spell lit it. We bring all four into line so burning has one rhythm across the school. This matters twice over once burns become a visible state (S12): the meter should never tick to a different drum per spell.
Seen in: ESO's standardized Burning status, one behavior regardless of source.
The Fire Arrow fix aligned every scaled burn effect on the spell to a single tick length; the same alignment applies to the other three spells' burn expressions, through the changeset process, sized against the balance bands since it changes effective damage pacing on three spells. The existing Fire Arrow duration test is the template; S9 extends it across the school first so the diff is explicit.
Water Warding is Fire's ward, and unlike every other ward in the game it unlocks the instant you buy its prerequisite, at level zero. We give it a real prerequisite level so the tree reads consistently, and nobody discovers the quirk on a wiki instead of in the game. Existing trained wards are untouched.
Seen in: every skill tree with consistent gating.
One prerequisite cell in the skill table, through the changeset process's cell-level edit. Grandfather rule: already-learned skills are never retroactively locked, which the learn flow already guarantees since prerequisites gate learning, not keeping. S9's structural audit pins the whole fire block (prerequisite graph acyclic, card levels sane, ward gating consistent) so tree edits become reviewable diffs.
A truth pass on fire's own words. Fire Speed grows your spells' reach and their area, but its name and description only mention speed. Six spells carry hidden bonus text tied to a state almost nobody can reach (see S11). The specialization's real trade-offs are nowhere on its tooltip. We rewrite the school's strings so every skill says what it actually does, in plain language.
Seen in: WoW's long campaign of tooltip literalism; PoE's "the tooltip is the contract" culture.
Skill name, description, and execution strings in the localization source for the school, re-exported through the translation pipeline (coordinate with the Chaos program's localization groundwork so the export path is exercised once, carefully). S11 and S12 rewrite some of the same strings again; batch the passes. The internal reference doc gets the same corrections, including its summoning combo erratum.
Not player-visible, and listed anyway because everything else depends on it: an automated test net that pins every fire spell formula, every combo table entry, and the whole fire skill block before anything on this page moves. Exactly one fire spell has such a guard today. After this, any change to the school is a reviewable diff against a known baseline, and nothing regresses silently.
Seen in: the recent Frost and Shock weapon enchant work, which shipped with exactly this kind of pin test from day one.
EditMode tests following the two established patterns: expression pinning with the re-tokenize guard (stored formula tokens must match a fresh parse), and data-contract pinning (school membership, tree rows, combo table entries, effect structure). Evaluate every fire expression at low and at grandmaster inputs so dead terms and division hazards surface. Pure data reads, no play mode, runs in the standard editor test pass.
Fireball, Immolation, Ring of Fire, and Ignite Weapon cannot combine with any other glyph. That cuts half the school out of the deck game entirely. We give each at least one combo so stacking fire glyphs always has somewhere to go. Working sketches: Backdraft (Fireball + Gust: a wider blast that shoves), Cinder Strike (Flame Fist + Immolation: a melee hit that shares your aura's burn), Firestorm (Ring of Fire + Fireball: one larger field replacing both), and an Ignite Weapon pairing extending the Flaming Bash family. Final pairs are a design pass; the point is the coverage.
Seen in: GW2's fire fields and finishers, the genre's cleanest combo grammar; UO's Explosion plus Energy Bolt timing as the ancestor.
| Glyph | Today | Proposed |
|---|---|---|
| Flame Fist | Flaming Bash · Smoldering Shiv | + Cinder Strike |
| Fire Arrow | Fiery Decay | |
| Fireball | none | Backdraft · Firestorm |
| Immolation | none | Cinder Strike |
| Ring of Fire | none | Firestorm |
| Ignite Weapon | none | + shield pairing |
| Summon Fire Elemental | ice swap |
Combo recipes are data on the result glyph's asset; the combination table builds itself from those entries at load, and all three existing fire combos are already cross-school pairs, so new pairs need no school table changes. Each product is a new spell asset composed from existing effect building blocks, sized through the changeset process. Combo products double as clean heat hooks for S11. Pinned by a combo-table test asserting exactly the expected fire entries exist.
Six fire spells already carry bonus behavior when you have a state called Fireboost: longer burns, stronger flames, at a higher focus price. In practice only one rare artifact family can grant it, so for nearly every fire mage that text is a bonus that never happens. We add ways to earn Fireboost inside the school itself: committed play banks it, and the spells that already know how to spend it finally get to. This is the heat engine the rest of the page builds on.
Seen in: WoW's Heating Up into Hot Streak, FFXIV's Astral Fire, PoE's ignite investment: build and spend is the modern fire grammar.
The Fireboost value is already a replicated character stat with a player-facing name, read by six spells' formulas; what is missing is any in-school grant. Grants are timed self-buff sub-effects on existing spells (candidates: Immolation while its aura burns, combo products from S10, Combust from S13), all plain data on the spell assets, following the recipe the Frost and Shock enchants proved. The existing six consumer terms were written when the stat was rare, so they get re-based in the same changeset that adds the grants (the current focus surcharge would be punishing at sustained stacks). Because it is caster-side replicated state, every player sees the same result.
Being on fire becomes a real, visible condition instead of invisible ticking damage. Burning targets show flame pips under their nameplate, one per burn, and your own buff bar shows what you have applied. Fire spells can then check for the state and react to it. Every fire spell already applies a burn; this makes that shared behavior the school's identity instead of its hidden bookkeeping.
Seen in: GW2's stacking Burning readout, PoE's first-class ignite, ESO's standardized Burning status.
Burn sub-effects on the fire spells additionally apply a burn-stack counter value on the target, capped at the same stack limit burns already obey, cleared when the last burn expires or is cleansed. The counter is target-side replicated state, so any spell formula may read it and every player sees the same count. UI is a UI Toolkit element on the new combat HUD's nameplate and buff bar; no legacy UI work. This is the item on the page that most needs live multi-client verification (two clients, caster and observer, matching pips), and it is specified with that checklist attached. S6 lands first so the state it displays has one rhythm.
A new fire spell that detonates the burns on your target: their remaining burn time is traded for one immediate hit, scaled by how many burns were consumed. Stack your fires, then choose the moment it all goes up. Fire gains a real decision where it previously had only repetition, and the payoff is loud, bright, and instant. Detonation is a flash, not a field: nothing lingers on the ground.
Seen in: WoW's Combustion, PoE's detonate-the-ailment builds, Magicka's steam-blast payoffs.
Ships by the exact recipe the Frost and Shock enchants proved: a new skill table row, a new spell asset registered with the rune set, a school mapping entry, an icon, and localized strings, with no server logic. A reserved skill identifier already exists for the school, so even the identifier is free. Mechanics: the hit expression scales with the target's burn count (readable replicated state once S12 lands), followed by a fire-typed damage-over-time removal, the same removal family cleansing spells use. Damage sizing is wholly owned by the balance gate; the design constraint is expected value roughly equal to letting the burns run, so Combust is a timing choice, not a damage buff. Icon batches with S16 and S17.
Frost Weapon and Shock Weapon shipped this month as proper weapon enchant spells: one exclusive slot, distinct utility, deliberate trade-offs. Ignite Weapon, the original, still follows its own older rules and stacks freely beside them. We put all three on the same exclusive footing so choosing your weapon's element is a real choice, and we bring Ignite's numbers to the same reviewed standard.
Seen in: ESO's flame weapon glyphs; our own Frost and Shock enchants are the in-house precedent.
Effect exclusivity rides the existing channel system: one new channel string shared by all three enchant spells, a one-line data addition, the same mechanism that already keeps shield spells exclusive. Magnitude parity (Ignite currently front-loads a longer burn where the new pair deal a single measured hit) is a changeset decision for the balance gate, not this page. Gear-side elemental enchant coverage for bows and shields is the weapons proposal's W2 and stays there; this card touches only the spell rail. Extends the existing enchant pin test with the shared-channel assertion.
Immolation currently counts as a shield spell, so running the school's signature burning aura costs you your actual defensive shield. We give it its own aura slot, and we let the game recognize the state: while you are immolating, other fire effects can know it and build on it. The aura stops being a defensive tax and starts being a stance the school notices.
Seen in: PoE's Righteous Fire as the aura-that-defines-the-build archetype.
Two data changes: the aura moves from the shield exclusivity channel to a new fire-aura channel, and it raises an "immolation active" flag for its duration using a fire stat that already exists in the data with no gameplay expression, so nothing new is invented. The flag is caster-side replicated state, safe for any formula to read (an S16 passive gate, a Flame Fist rider). Freeing the shield slot is power-positive by definition, so the changeset decides what Immolation pays in exchange. The existing aura field's per-pulse behavior is untouched.
A passive ability package for committed fire mages that only functions under conditions you build toward, in the mold of the Sun and Moon worship passives: for example, a bonus that lives only while your aura burns, or only while you wield a caster weapon. Commitment becomes a build choice you can feel, not just a bigger number.
Seen in: conditional mastery passives across the genre; our own Sun Worship package is the in-house precedent.
The game ships a formula-driven, equipment-gateable passive package system that exactly four abilities use today, none of them fire. One new package asset attached to a fire passive's reward slot, gated on S15's aura flag or an equipment class. Pure data plus one asset; magnitudes through the changeset. Because the primitive is rarely exercised, a data-contract test evaluates the package's gates at low and grandmaster inputs.
A stance you enter and leave. While it burns, your fire spells change character (faster burns, hotter hits, a stated cost the entire time you hold it) and only one stance can be active at once. Where S11's heat is earned moment to moment, the stance is a promise you make for the whole fight.
Seen in: FFXIV's Astral Fire, the genre's proof a held fire stance can carry a class; GW2's attunements as the multi-stance frame.
A long self-buff spell on its own exclusive channel (the S14 and S15 precedent), whose effects are stat bundles the formulas already read. Needs one new skill identifier appended to the shared skill list, mirrored on both sides of the shared-code boundary per standing process, plus row, asset, icon, and strings. The stance's trade-off sizing is wholly a balance-gate matter. Sequenced after the channel precedent lands so exclusivity is proven infrastructure by the time this ships.
Defensive investment against fire currently reads stronger on paper than it plays: the stat sheet says resistance, but what the number mostly does is quietly dilute the attacker's attunement advantage rather than subtract damage. We pick one clear, honest lane for what fire resistance means, make the character sheet say that, and make Fireproof worth its training time. This card is the page's heaviest balance item and it is deliberately a decision, not a foregone answer.
Seen in: UO's Resisting Spells as resistance-you-choose; ESO's flat, legible mitigation lines.
Fire defense today flows through the attunement comparison inside the damage formula rather than through the flat percentage mitigation lane that other damage reductions use, and gear or training labeled as fire resistance feeds the former. Two honest options: route the resistance stat into the flat mitigation lane (bigger feel change, truest to the label), or keep the dilution lane and rename what the sheet displays (smaller change, honest words). Either way the aggregation lives in shared client-side character math, changes through the changeset process, and re-runs the school audit afterward because it re-tunes every fire matchup. The armor page owns gear-side defensive display and aggregation; coordinate wording there. Prerequisite for S21.
Fire Specialization currently makes you permanently more vulnerable to water, a flat penalty stamped on your sheet no matter what you are doing. We keep the elemental price but make it situational: the vulnerability applies while you are actually running hot (aura up, stance held), so the weakness comes from choices in the fight, and a water mage's counterplay has a moment to aim at.
Seen in: Magicka's deterministic wet-beats-fire grammar; FFXIV's fire stance carrying its own cost.
The specialization's innate vulnerability restructures from an always-on reward line into a conditional term gated on the S15 aura flag or S17 stance state, both replicated caster-side. This is the same formula family as a previous cross-client sync fix, so it carries the live two-client verification checklist explicitly. Net power is held neutral at the changeset gate: weaker tax while cold, the full price while hot.
One training stat currently controls both how wide your fire spells reach and how fast your burns tick, so investing in area quietly changes your damage rhythm as a side effect. We separate area from pacing so each investment does one understandable thing, and the tooltip for each says exactly that.
Seen in: PoE's strict one-stat-one-meaning affix discipline.
The area stat appears in the burn tick-length denominators across four spells; the restructure moves pacing onto the burn-duration stat or a constant, spell by spell, through the changeset process, holding grandmaster output within the established parity band. Lands after S6 and ideally after S12 so the visible burn state is stable while its math is re-plumbed. S9's pins make each step a reviewed diff.
Skeletons are supposed to fear fire, and the game's data already marks them vulnerable to it. Yet high-tier undead carry blanket magic resistance so heavy that the balance audit measured fire's sustained output against them at effectively zero, vulnerability and all. We let burn damage cook through part of that blanket: direct hits still respect the wall, but a fire kept burning on an undead target does honest work. Fire becomes the school you bring against the restless dead, which is exactly the fantasy.
Seen in: the anti-undead fire niche across the genre, from D&D turn-and-burn tradition to WoW's Holy-and-fire undead lore.
Mechanically: damage-over-time ticks weight blanket magic resistance lower than direct hits do, in the shared damage aggregation, with creature fire vulnerability unchanged and still amplifying. Scope is deliberately DoT-only so the counterplay is slow, visible pressure, not a burst bypass, and no other school's direct casts change. Sequenced explicitly behind the balance program's undead-resistance wave and re-sized against its outcome, and behind S18's lane decision because the two touch the same aggregation. Fills the community's long-standing ask for active tools against heavily resistant enemies.
When an enemy dies while burning, it can leave a small patch of embers that scorches anything standing in it for a few seconds. Fire fights start leaving marks on the battlefield, and chaining kills through your own embers becomes fire's crowd-clearing signature. Small, short, and strictly capped: embers are punctuation, not territory.
Seen in: PoE's proliferation fantasy delivered at MMO-safe scope; D:OS2's surfaces as the maximal version this deliberately is not.
An on-death trigger while burning spawns a short-lived, small-radius field reusing the Ring of Fire field template and its existing visuals, so no new effect prefabs. The caps are the design, not a tunable afterthought: hard concurrent-patch limit per scene, short duration, modest pulse rate, hostile-only targeting at first, because persistent fields carry a real performance cost and this page keeps that budget fixed. Damage attribution follows the same ownership rules existing field spells use. Ships last, after the burn state and its rules are proven.
Your fire training should matter to every fire-typed spell you cast, not only the seven in the tree. Searing Ray, Fire Flies, Corpse Explosion, and the fire face of Chaos Bolt join the burning system: their burns light the same visible state, count toward the same detonations, and read the same heat. A fire specialist feels it across their whole deck, and hybrid builds get honest cross-school texture.
Seen in: ESO's any-flame-source-can-proc-Burning unification.
Once S12 exists, joining the system is pure data on each spell's burn rider: apply the same stack counter the fire school applies. Verify each candidate already reads fire damage and attunement terms so membership is consistent. The Chaos rework's planned retyping of Chaos Bolt composes cleanly with this (their change adds a chaos identity on top of the elemental faces; this change makes the fire face light the shared state). Widens S12's multi-client verification surface, same checklist.
Your Fireball at grandmaster looks and sounds identical to your first cast. We scale the presentation of fire spells with real investment (fuller flame, deeper roar, brighter detonation) inside the game's existing effects budget, and add an accessibility-reviewed per-school spell tint option so players who need to distinguish schools at a glance, or simply want their fire blue, can have it.
Seen in: UO's Flamestrike as the prestige visual; every modern action RPG's tiered cast feedback. Both halves are standing community wishlist items.
Presentation only, so none of the formula-replication constraints apply: existing fire effect prefabs gain scale, emission, and tint parameters driven by locally known caster stats, with no new prefabs so load cost is unchanged, and parameters clamped to authored bounds with a low-end quality fallback. The tint option is a client setting applied at effect spawn. This is an art-time item more than an engineering item, and it runs as a parallel track; S1 and S2 lay its audio and wiring substrate.
The danger with twenty-four proposals is a school with twenty-four systems. The composition rule: one buildable resource (heat, S11), one visible state (burning, S12), one payoff verb (detonate, S13), and everything else is either polish, a combo, or a rule those three make legible. Here is the loop in one fight, as the combat HUD would show it:
| Wave | Ships | Depends on | Why this order |
|---|---|---|---|
| 0 | S9 pin suite, S8 truth pass | — | Pin everything before anything moves; the strings pass exercises the localization path once, carefully. |
| 1 | S1, S2, S4, S5 anytime; S3 + S6 + S7 as one changeset batch | S9 | Each is a standalone patch-note item; the changeset batch is the school's hygiene pass. |
| 2 | S12 burn state, then S11 heat and S13 Combust; S14 then S15 channels; S10 combos | S6, S9 | The identity core. Burn state first so heat and Combust have something to read; channels prove exclusivity infrastructure. |
| 3 | S18 lane decision, then S21 undead counterplay; S19, S20 | S12, S15; balance program's undead wave | The structural items, each re-audited; S21 explicitly re-sizes against the undead resistance wave's outcome. |
| 4 | S16, S17, S23; S22 last; S24 as a parallel art track | channel precedent, S13 proven | The swings, once the loop is proven in the school; embers ship only after burn rules are stable. Icons for S13, S16, S17 batch as one pass. |
flowchart TD S9["S9 pin suite"] --> CS["S3+S6+S7 changeset batch"] S9 --> S12["S12 burn state"] S6["S6 tick parity"] --> S12 S12 --> S13["S13 Combust"] S12 --> S23["S23 fire-typed spine"] S12 --> S22["S22 ember patches"] S11["S11 heat"] --> S13 S10["S10 combos"] --> S11 S14["S14 enchant channel"] --> S15["S15 aura slot"] S15 --> S16["S16 conditional passive"] S15 --> S19["S19 situational tax"] S14 --> S17["S17 fire stance"] S18["S18 resist lane"] --> S21["S21 burns cook through bone"]
| Proposal class | What carries it |
|---|---|
| Audio, visuals, animation (S1, S2, S4, S24) | Existing effect prefabs and asset fields; no new load cost |
| Formula and tree tuning (S3, S6, S7, S20) | Spell data expressions and skill table cells, through the reviewed balance-changeset process |
| New spells and combos (S10, S13, S17) | The same data recipe the Frost and Shock weapon enchants shipped with this month |
| Exclusivity and stances (S14, S15, S17) | The existing effect-channel system; one data entry per channel |
| Heat and burn state (S11, S12, S23) | Replicated character stats the spell formulas already know how to read; one shared-data addition for the burn counter |
| Rules (S5, S18, S19, S21) | Shared character math and effect lists, changeset-gated, re-audited after |
No proposal on this page adds server-side systems or new network operations, and none moves an ounce of tuning outside the balance program's process and bands.
Every card names its EditMode pin (S9 is the net under all of them): expression pinning with the re-tokenize guard, data-contract pinning for tree rows and combo tables, list-integrity checks for the cleanse pass, and bounds checks for the presentation scaling. The two multi-client items, S12 and its S23 extension, additionally carry a live two-client verification checklist because state that every player must agree on cannot be proven by unit tests alone.
UO: UOGuide (Fire Field, Explosion timing, Resisting Spells, Corpse Skin), Ultima Codex, era-server compendia on the Explosion + Energy Bolt dump. WoW: warcraft.wiki.gg (Hot Streak, Ignite, Combustion histories), Icy Veins and Method rotation guides, the Blizzard forums' ignite-munching fix thread, Wowhead (Fire Blast, Sun King's Blessing). FFXIV: the official Black Mage job guide, Console Games Wiki (Elemental Gauge, Astral Fire, Umbral Ice, Despair, Enochian), 7.2 patch coverage. PoE: poewiki and poedb (Ignite, Proliferation, Combustion, Flame Surge, Scorch), build guides for Righteous Fire and Explosive Arrow, PoE 2 ignite rebase notes. GW2: the official wiki (Burning, Combo, Fire field, Attunement, Effect stacking), the 2015 intensity-stacking combat update post. D:OS2: Fextralife wiki (Environmental Effects, Necrofire, Torturer, Spontaneous Combustion, Elemental Affinity), designer commentary on pulling back from surfaces in BG3. Magicka: Carl's Guides and the tutorial archives (element queue, opposition pairs, status table), Arrowhead on friendly fire as a pillar. ESO: UESP and ESO-Hub (Burning, status-effect proc table, Glyph of Flame, Wall of Elements, Combustion passive, Engulfing Flames), with Diablo 4's damage-vs-burning affix as the named itemization fallback. Full URL list in the research notes accompanying this proposal.
Prepared 2026-07-27 against the unity-6.5 migration branch. Companion markdown: docs/plans/2026-07-27-fire-magic-improvements-proposal.md. Twenty-four proposals: every one of them client work, most of it data; the heaviest items are decisions, not systems.
No account needed. Every proposal above also has its own Discussion box for talking about that one alone; this thread is for the plan as a whole. Comments are plain text, held to basic decency, and may be trimmed.