Dualblade: no fixed rotation, and that's the point — the priority tree that replaces one

✍️ Alice・愛麗沙

⚠️ Confirmed live on Taiwan’s server today — Dualblade’s own build page (/meta/dualblade) draws real traffic, and “dualblade” variants are this site’s single largest class-name search family on the Chinese side. As with every class build page on this site, the specific gear/rune data is Korean community research (a maintained class spreadsheet, cross-checked against two independent forum sources four months apart), not an official Nexon publication — Nexon publishes drop tables and skill text, not build guides. Confidence is marked per fact.

The mechanic that makes this class different: there is no rotation

Every other melee class on this site’s build pages has a combo string. Dualblade doesn’t, and the person who compiled the Korean reference sheet says so directly: “there is no fixed damage rotation, because Gale Force resource generates randomly.” That’s not a gap in the research — it’s the class design. So instead of a button sequence, Dualblade runs on a priority list, evaluated continuously:

1. Spend skill 4 the instant it hits 4 stacks of Gale Force. This is the main damage skill; holding it for a “clean” opener costs more than it saves.
2. Keep skill 1 cycling. After the Laceration+ rune modifies it, skill 1 becomes your ranged strike-and-stagger source — use it every time it’s off cooldown.
3. Fire the ultimate whenever everything else is on cooldown. It doesn’t compete with your resource, so it’s a filler, not a finisher to save up for.
4. Never wait for a “perfect” opening. Randomized resource means expected-value play: use what’s ready now, and let the accumulated pattern be your rotation rather than planning one.

In practice the read is “keep rhythm,” not “recall a combo” — your only live decisions are how many Gale Force stacks does skill 4 have and which cooldown comes back first. Two secondary habits matter: work your movement skills into the middle of combat rather than saving them purely as an escape (Plains Wanderer’s +12% damage buff only triggers after you’ve used a movement skill), and land Break-condition hits deliberately — Rock Giant’s burst-stack mechanic is powered by them, and the Laceration+-modified skill 1 is your most reliable Break source, which is the other half of why it sits at priority #2.

Full loadout: BIS, mid-tier, and free-to-play

The two irreplaceable runes are Laceration+ (turns skill 1 into a ranged, cooldown-shortened strike) and Rush+ (using skill 5 resets skill 4’s cooldown and adds a chance at bonus Gale Force resource) — everything else on this table can be substituted down a grade, these two can’t, because they change how the class works, not just how hard it hits. If you’re not yet geared for the full set, the sheet’s advice is blunt: gear the condition-free runes first (Stability, Sniper, Enlightenment — anything that triggers without a setup requirement) rather than chasing conditional runes like Rock Giant’s Break trigger or Full Moon’s Break-window cooldown reset before you can reliably land Breaks in the first place.

Thousand Swords note (same caveat as every other class page on this site that uses it): the rune’s community-inferred “sword class” roster is Warrior, Greatsword, Swordmaster, Frostmage, Rogue, Dualblade — six classes, +4% attack each at class level 40, +24% at all six maxed. Warrior isn’t live on TW yet, so the practical ceiling today is five classes = +20%, not 24%. One board-tested detail worth knowing: this bonus doesn’t move the visible combat-power number — don’t assume it isn’t applying just because the on-screen figure didn’t change.

The seven signature runes

Class-only runes live in the accessory slots (necklace, two rings) and rewrite specific skill behaviour. One labelling note before the table: the catalogued base grade of each rune (below) is not always the grade the BIS loadout equips — several of these runes have a +-suffixed evolved form that reaches legendary rarity, one grade above (in one case, Laceration, two grades above) its catalogued base. Both figures are real; they describe different points on the same rune’s upgrade path, not a contradiction.

Third-ring choice is a content call, not a power ranking: Swift+ for single-target boss fights (the steadiest general-purpose pick), Reinforce+ when the fight has a clear, telegraphed burst window worth timing skill 4 around, Whirlwind+ for trash-clear and multi-target content — and it’s a genuine trade-off, not a strict upgrade, since Whirlwind loses value the moment the fight is single-target again.

Pre-BIS priority, if resources are limited: Rush+ and Laceration+ first (they change how the class plays, not just its numbers), then the two BIS armour runes (Mana Torrent, Rock Giant), then Thousand Swords last — its stacks come from levelling other classes to 40, which enhancement materials can’t accelerate, and those same class levels feed the “Versatile” title’s 675 cumulative-level threshold, so it’s not wasted effort either way.

Imprinting runs through the universal, officially-published rune-imprint success table — gear vs. the emblem slot, which never fails:

Skill remodel: one standard build, three situational variants

The Korean reference sheet’s 1-5-6 remodel is corroborated twice, four months apart, by independent sources — about as solid as this kind of community data gets. 156 (Laceration on skill 1, Rush on skill 5, the third ring’s pick on skill 6) is the default; the sheet also documents three matched variants for specific ring-3 picks — 136 (a straight weapon-swap phrasing of the same idea), 356 (Whirlwind builds, which also requires swapping Mana Torrent for Cold-Blooded Duelist in the hat slot), and 456 (Reinforce builds). Skill-level-30 priority is skill 1 → skill 5 → the ultimate, consistent across every source checked, with the ultimate ranked third for the same reason as most classes on this site: it doesn’t feed the class’s core resource loop, so it’s the one skill you can afford to level last.

Cards and gems

Attack cards are the flat priority; Dualblade’s main stat is Agility and its sub-stat is Luck, and — unusually for a DPS class — the Korean sheet recommends taking Luck cards alongside the main stat rather than skipping them, since Dualblade converts Luck into crit rate and follow-up-hit rate at a rate worth the card slot. Gem reforge tags: Combo (damage), Movement (damage) ahead of everything else, Follow-up (damage) second — except on a Whirlwind build, which reforges for Interference (damage) instead. One board-tested caution: the Heliodore special effect only counts one copy per loadout — socketing a second is wasted currency, not double the bonus.

Auto-battle: the honest answer is no

Unlike Frostmage or Greatsword — the two classes this site’s community consistently recommends for AFK play — Dualblade is built around the exact thing an unattended AI can’t do: hold skill 4 until it hits 4 stacks rather than firing it the moment it’s off cooldown. Randomized resource generation compounds the problem, since a scripted rotation can’t adapt to a resource bar that doesn’t refill on a fixed schedule. This class has never appeared on this community’s AFK-recommendation list, and there’s no build variant that changes that.

The practical workaround costs nothing on this server: class-switching is free, and every class banks its level permanently. Level an AFK-friendly class for unattended farming, switch back to Dualblade for content you’re actively playing — and the AFK class’s levels feed the same “Versatile” cumulative-level title track either way, so nothing is wasted by splitting time between two classes.

Where it actually ranks

• Daily / auto-clear content: A-tier. The reasoning is specific: sustaining Gale Force stacks is close to pure button-mashing, which an auto-battle AI genuinely can’t mess up — ranked the top auto-battle pick among the game’s “rogue-adjacent” classes (Dualblade > Rogue > Fighter, per the same source), even though it doesn’t make the top AFK recommendation for unattended-content-clearing overall (see above — A-tier for the daily grind isn’t the same claim as being AFK-viable for content that needs the priority tree run correctly).
• Team raids: B-tier. Reinforce’s stage-2 skill can fire repeatedly regardless of cooldown, which the source specifically calls useful for “coping with a boss’s Break-demand mechanic” — a sustained-damage profile that fits the raid axis reasonably well.
• Abyss: B-tier, same rank as Swordmaster, Fighter and Greatsword — Twisted Blade’s damage-over-time is an area effect, which one player log calls “looks very strong in deep-dungeon auto or abyss content.” Worth stating plainly: this class gets talked about in terms of fun more than raw power in Korean community discussion — one representative quote translates roughly to “it’s the #1 for enjoyment… if you just want to have fun regardless of what anyone else says, Dualblade, no question.” That’s a real signal, not a knock: a tier letter doesn’t capture it, so this page isn’t going to pretend a single grade does either.

What would overturn this page

Two concrete, dated tripwires from Korea’s own patch history, since TW tends to follow KR’s schedule roughly a year behind: Korea’s Season 1 (2025-09-25) gave every class two new “reforge-a-skill” accessory runes, and Dualblade’s is called Cheonchim+ (Taiwan’s eventual localised name unconfirmed) — the current “Laceration + Rush + pick-a-third” structure gets restructured the moment TW reaches that content. Korea’s Season 2 (2025-06-25 on KR’s own calendar) goes further: the endgame loadout becomes a three-piece Cheonchim set built around a “Cheonchim → basic attack → Cheonchim → basic attack” weave, and Thousand Swords drops out of the mainline KR S2 build entirely. Everything on this page describes TW’s Season 0 through Season 1 window — this page gets rewritten when TW’s own version catches up, not before.

Written in English from Korean community class-spreadsheet data, alongside this site’s Traditional Chinese original. The searchable rune/gear/skill database behind the tables above lives at /en/meta/dualblade/ and /en/builder/; cross-class comparisons are at /en/tierlist/. This site’s enhancement-priority reasoning (which system to feed gold into first) is not yet translated into English — the Traditional Chinese original has the full argument.