Dungeons · Published 2026-09-02 · Updated 2026-09-14 TW live

Why does everyone party up for wards, holes and bosses?

✍️ Alice・愛麗沙

Quick answer

Partying isn’t etiquette — it’s line items on your stat sheet. Per KR community measurement (never officially published, medium confidence): the party bonus grows with member count and caps at four — attack power, drop rate, plus one stat line that varies by class. Separately, smash cheer and multi-hit cheer switch on the moment you’re in any party, regardless of size. One condition governs all of it: party members must be in the same area. So for content where grouping is even possible — summon wards, black holes, field bosses — a four-member party out-earns a solo run on more than clear speed: the drop-rate bonus is free expected value, and that is the real reason the recruitment channel never sleeps. An eight-member raid gets the same party bonus as a four-member party; it does not stack higher.

The numbers first

Layer one: the bonuses you get purely for being in a party. No trigger, no upkeep — in the party, in the same area, they’re on:

BonusValueCondition
Smash cheer / multi-hit cheer+150 eachApplies the moment you are in a party, any size
Attack power+1033Scales with party size, caps at 4 members (larger parties get the same)
Drop rate+10%Scales with party size, caps at 4
Class-matched statup to +387Varies by class, also scales with size

Only the bonuses granted by the party STATE itself — the job-line synergy skills are a separate table. All figures are KR community measurements (dcinside enban#949207, read directly 2026-09-01); never officially published. Medium confidence.

Three readings of this table. First, the drop-rate row is the headline. An attack bonus only buys speed for someone who could already clear the content; a drop-rate bonus is one extra payout every ten runs in expectation — a direct production difference for anyone farming emblem runes or key boxes. Second, the stat row varies by class — a class whose matched stat gets pulled up collects the cheer buffs and the stat line at once, so the table understates their gain. Third, four is the ceiling: an eight-member raid party receives exactly what a four-member party does. “Bigger raid, bigger bonus” is a myth.

Layer two: the six job lines each carry a synergy skill — and these are not passives. Each fires on its trigger and then benefits the whole party:

Job lineTriggerParty-wide effect
Rogue lineOn finishing attackParty-wide pursuit +80
Warrior lineWhen taking a hitParty-wide attack +40
Archer lineOn pre-emptive attackParty-wide crit +80
Mage lineOn AoE breakParty-wide skill power +80
Healer lineWhen healing an allyParty-wide recovery up (value unpublished)
Bard lineOn campfire shareParty-wide bonus HP for 3 min (value unpublished)

The six job-line party synergy skills — not always-on passives: each fires only when its trigger condition is met, then applies party-wide. KR-community summary, unofficial, medium confidence.

The synergy table also explains a lobby habit veterans have: caring about line composition. A party where someone reliably lands pre-emptive attacks and someone else opens AoE breaks keeps these lines lighting up in rotation — the damage rhythm is a different game from four people soloing side by side.

Who measured these numbers — the provenance matters

Every figure above comes from a KR community measurement (this site fetched and archived the original posts); the developer has never published party-bonus numbers. Worth spelling out, because the Taiwan board circulated the same figures attributed to “an official QA an AI dug up.” We went looking for that QA. It does not exist. The numbers were right; the provenance was invented.

Why be pedantic about this? Because “officially published” and “community-measured” fail differently. An official number that’s wrong is the publisher’s problem; a community number can drift silently after any patch, with no announcement owed to anyone. Hanging an official label on a community figure lends it a warranty it never had — which is exactly why this site tiers the source of every number it prints. Use the number; file it under the right name.

Converted into a weekly schedule

Put the bonuses back into an actual play week and the priority sorts itself.

Ominous summon wards are the textbook beneficiary. The official timetable is fixed — spawns hourly, starts summoning after three minutes, gone from the hunting ground after seven. That’s a timed fight: the party attack bonus converts directly into “cleared or not,” with the drop bonus layered on top. Recruiting costs nothing (the crowd is already standing at the ward) and both bonus lines pay out.

Black holes come second. Exploration entries have daily and weekly caps — every entry is a finite resource — so the same entry spent in a party has strictly higher expected output than one spent solo. This calculation has nothing to do with whether you can clear it alone; it’s about not wasting caps. (Added 9/14: the abyss-bound black hole Taiwan added on 9/9 — 通往深淵的黑色坑洞 — is the exception. Update log 3541097 states its entries are “counted separately from existing black-hole exploration counts, with no entry limit,” so there the only reason to party is the bonus itself; the cap-saving argument does not apply.)

Field bosses depend on the hour. At peak times everyone tags for credit anyway and the marginal value of a formal party is small; off-peak, when the damage threshold actually needs meeting, the bonus matters again.

And the honest inverse: solo-stable dailies (daily quests, gathering routes) don’t deserve a forced party. The same-area condition means the moment your party members scatter to their own routes, the bonus isn’t even active. Party value is always chained to “actually hitting the same thing on the same map.”

What would overturn this article

Every number here is community-measured and unverified on TW. The easiest row to test is drop rate: run the same hole the same number of times solo versus four-stack and the direction shows up in the tally. If field results disagree with the table — or if the developer ever actually publishes party-bonus numbers (welcome!) — this article changes the same day, with the correction dated. The blanks in the synergy table (exact recovery and bonus-HP values) were blank in the source compilation too; they stay blank here. An empty cell beats an invented one.