Zenless Zone Zero 3.3 "Journey Back to the Sky" launches on October 21 with Phoenix (Fire, Anomaly) and Verina (Wind, Anomaly) in phase one, then Severian (Wind, Attack) on November 11. This guide breaks down the Polychrome and Dennie bill, the pull verdict by account type, Drive Disc and W-Engine priority, three team setups and the Shiyu Defense loop — every number carries its source grade.
| Item | Answer | Basis |
|---|---|---|
| Cost to guarantee Phoenix | 180 pulls worst case, 28,800 Polychrome | Any S-rank within 90 pulls, 50% featured (in-game banner rules) |
| Cost to max one S-rank | 3.105 million Dennies + 300 Senior Investigator Logs | Same-tier S-rank milestone (third-party) |
| Free income per patch | 50–56 pulls (8,000–9,000 Polychrome) | 42-day renewable tally (third-party) |
| Worth pulling? | Yes if you lack an anomaly slot, no if you need a DPS | Official role card |
| Signature weapon priority | After the character, 25,600 Polychrome worst case | 75% UP, 80-pull hard pity |
Applicable version: 3.3 "Journey Back to the Sky" (launches 2026-10-21) | Live client 3.2.0 | Verified 2026-10-18
ZZZ 3.3 at a Glance: October 21 Launch and Two Banner Phases
The 3.3 start date is not guesswork. The official 3.2 patch notes state that version runs 42 days and closes at 2026/10/21 06:00 UTC+8 . The official preview published 2026-10-09 confirms the version name, "Journey Back to the Sky", along with both new S-rank agents.
First half: Phoenix and Verina
Phoenix is tagged as a Fire Anomaly agent whose kit lets team-wide anomaly damage from every attribute roll crits, and who can re-enter the field with restored HP after being defeated . Verina is a Wind Anomaly agent built around high Wind anomaly damage and a brand-new anomaly type called Turbulence .
Two anomaly agents in one phase is a readable signal. The banner is selling a second anomaly leg, not a main-DPS replacement.
Second half: Severian on November 11
Severian sits in phase two, with an official banner window of November 11 at 12:00 through December 1 at 14:59 . His role is a strong Wind damage dealer on the Attack side.
HoYoverse has not published a separate end time for phase one, so the only way to bound it is backwards from phase two: before 11:59 on November 11. Treat the minute-level boundary as a community estimate — official phases often close a few hours away from what the next phase implies.
How the two phases hand off
Pity does not reset between banners. Pulls you bank in phase one carry into phase two .
That single rule kills the "stop halfway" plan for 3.3. Either you arrive with 180 pulls or you spend none.
How Much Phoenix Actually Costs: 28,800 Polychrome and 3.105 Million Dennies
ZZZ 3.3 Phoenix team planning starts with money. The bill splits into two halves: currency for the pull, materials for the build. Both can be pinned down, though only the first is arithmetic — HoYoverse publishes Phoenix's role but not her ascension material names, so the material side leans on same-tier S-rank milestones.
The Polychrome math behind 180 pulls
Featured banners guarantee an S-rank within 90 pulls, and that S-rank is the featured agent 50% of the time. The rule is not "90 pulls gets you Phoenix"; it is "90 pulls gets you an S-rank, and half the time it is Phoenix." Lose the 50/50 and the next S-rank is guaranteed to be the featured agent .
Worst case is therefore 180 pulls, and at 160 Polychrome per pull that is 28,800. The number is easy to remember and easy to check against your balance.
An endgame account regenerates roughly 8,000–9,000 Polychrome per 42-day patch, or 50–56 pulls . At 8,500 that is about 142 days to save 28,800 — four and a half months. Free-to-play pacing is one character every two patches, not one per patch.
The material bill and the 11-day Dennie gap
Material costs follow the same-tier S-rank milestone, and the four blocks stack like this :
| Block | Materials | Dennies |
|---|---|---|
| Agent 1→60 | 300 Senior Investigator Logs, 4/32/30 promotion seals (66 total) | 800,000 |
| W-Engine 1→60 | 4/32/30 components, 200 Energy Modules | 400,000 |
| One skill 1→12 | 5/15/50 chips, 1 Hamster Cage Pass | 500,000 |
| Core skill A→F | 60 high-dimensional data, 9 weekly boss materials | 405,000 |
Dennies total 3.105 million. At 1,200 Dennies per battery charge and 240 charge per day, that is 2,588 charge, or 10.8 days — round it to 11 .
Eleven days covers Dennies only. Seals, chips and high-dimensional data stack on top, and core skill D→F weekly materials are capped by the weekly reset — three runs a week, so nine materials take three weeks. That part cannot be rushed, only scheduled.
Where should you farm Phoenix's promotion seals?
Promotion seals come from Combat Simulation — Agent Promotion, picking the training branch that matches her class. A run costs anywhere from 20 to 100 charge . Maxing out needs 66 seals; at 20 charge per run and a median drop of 8, that is 8 to 9 runs, or 160 to 180 charge.
The daily cap is simply battery charge: 240, with no separate attempt limit. Split it roughly 70/30 — seals and chips first, Drive Discs second. Discs are a probability sink, seals are a guaranteed return, so clear the guaranteed part on day one.
Is Phoenix Worth Pulling? Three Account Types, Three Answers
Verdict first: pull if you are missing a Fire anomaly slot, skip if your Fire anomaly core is already built, and wait for November 11 if the hole in your roster is front-loaded damage. An anomaly slot stacks inside a team; it does not create one from nothing.
| Your roster | Verdict | Why |
|---|---|---|
| Fire main DPS, no anomaly slot | Pull | She feeds the whole team's anomaly damage — exactly the missing link |
| Mature Fire anomaly team | Skip | Gains stack, they do not replace; marginal value under a new slot |
| Anomaly covered, needs an Attack unit | Wait for 11/11 | Severian is Wind Attack, a better role fit |
| Under 180 pulls saved | Keep saving | Pity survives the banner; stopping halfway is the worst outcome |
Accounts missing an anomaly slot
Pull, and you can field her the same day. Phoenix does not ask you to rebuild the team: drop her into an existing Fire lineup and the anomaly damage line goes from a fixed number to a number that can crit. The gear bar is low, too — a finished 4+2 Disc set is enough to slot in.
Accounts with a working Fire anomaly core
Hold. Your anomaly application is already handled, and what Phoenix adds is a crit roll on damage you were already dealing. The same 180 pulls buy more structural change on the November 11 Attack banner.
What if you're short of 180 pulls
Save. This is arithmetic, not caution: 100 pulls in with no Phoenix leaves you with no Polychrome, no agent, and a pity count that rolls into the next banner anyway. Go in with 180 or with zero.
Phoenix's Kit Explained: Why Team-Wide Anomaly Damage Can Crit
The load-bearing line in the official description is "team-wide anomaly damage of every attribute has a chance to crit" . The subject is the whole team and every attribute — not a single Fire-only stat stick.
Where the crit rule actually stops
Anomaly damage normally sits outside the crit calculation entirely. Phoenix's kit opens one specific door: a skill-provided proc chance routes that damage through a crit roll.
The distinction matters in practice: stacking crit rate does not raise the proc frequency, and stacking crit damage is what scales each proc. HoYoverse has not published the proc rate, and community estimates put it in the double digits as a percentage — wait for hand-tested numbers before you re-roll substats around it.
The revive: safety net or rotation crutch
The second official clause is that she can re-enter the field with restored HP after being defeated. In Shiyu Defense that is worth real time: one mistake does not force a floor restart, which matters most on long high-floor fights.
It is not invulnerability. The revive runs on an unannounced cooldown or charge limit, so do not plan a rotation around it — if you are leaning on it, your loop has already broken.
Phoenix Drive Discs and W-Engines: 4+2, Main Stats and Weapon Priority
Drive Discs use six slots, standard build is a 4-piece plus a 2-piece, and each disc caps at +15 enhancement . Phoenix's stat priority is not the double-crit pattern a direct-damage agent uses — she scales on anomaly damage and Anomaly Proficiency.
Slot 4, 5 and 6
Slot 4 takes ATK%, slot 5 takes Fire DMG bonus, and slot 6 is a choice between Anomaly Proficiency and Energy Regen depending on team energy. If the team already has an energy provider, slot 6 goes to Anomaly Proficiency; if not, run Energy Regen first and smooth out the Ultimate cadence.
Substat targets and when to stop
Substat priority: ATK% > Anomaly Proficiency > Crit DMG > PEN. Crit DMG only pays out on a proc, which is why it sits behind Anomaly Proficiency — raise the proc count and the anomaly base first, then scale.
A workable stopping point: 40%+ cumulative ATK% and 30%+ Anomaly Proficiency from substats. Past that, the marginal gain is smaller than two more disc levels on your main DPS.
The three 3.3 W-Engines ranked
Three W-Engines ship alongside her :
| W-Engine | Rarity | Class | Standing |
|---|---|---|---|
| Nirvana Plume | S | Anomaly | First choice, class and kit both line up |
| Convict Cage | S | Attack | Second, no anomaly synergy — raw stats only |
| Candle Trick | A | Defense | Transitional, free, use it while you farm |
The W-Engine banner runs 75% UP with an 80-pull hard pity, so the worst case is 160 pulls, or 25,600 Polychrome . That is nearly the price of a second character. Order of operations: character, then a finished Disc set, then the weapon.
Phoenix Team Comps: Three Setups and Their Rotation Order
Zenless Zone Zero team comps for anomaly lineups divide labor into "who spreads, who breaks, who closes", a different design logic from a direct-damage team built around "who stays on field". Get the order wrong and the crit window lands on an empty anomaly bar.
| Team | Lineup | Rotation |
|---|---|---|
| Dual anomaly | Phoenix + Verina + stunner | Verina spreads → stunner breaks → Phoenix closes |
| Fire flex | Phoenix + Fire main DPS + stunner | Main DPS cuts HP → stunner opens window → Phoenix scales anomaly |
| Safety net | Phoenix + Defense + Support | Defense shields → Support buffs → Phoenix revives as backup |
Dual-anomaly: who spreads, who breaks, who closes
Verina fills the Turbulence bar, the stunner fills the daze bar, and Phoenix closes inside the window. The three steps do not swap: spread first, break second, and only a full daze bar creates the window where the crit actually pays.
Fire-team flex: where to cut the health bar
In the flex setup, let the main DPS push the enemy below roughly 70% HP before the stunner opens the window. Cut too early and the window is wasted; cut too late and the anomaly damage misses the damage buff.
Why two DPS in one window is a trap
Two damage dealers stacked into one entry window do not double the output — they slow the anomaly spread. Against the same enemy, a double-DPS setup cleared 15 to 25 seconds slower than the standard lineup .
The trade-off is simple: one window fits one full combo string. Split it between two agents and neither finishes. Run one damage dealer plus two function slots and make the window deeper, not wider.
Shiyu Defense in Practice: The 20-Second Loop
Shiyu Defense scores floors on clear time, so loop consistency beats a single big burst. The Phoenix loop has four steps, and landing it under 20 seconds counts as clean.
The four-step rotation
- Bring Verina in and fill the anomaly bar — do not swap early.
- Swap to the stunner and fill the daze bar; start lining up the swap at the halfway mark.
- The moment the window opens, bring Phoenix in: enhanced attacks first, Ultimate second.
- Let the window close, recover energy, and wait for the next break instead of forcing damage.
Three mistakes that cost you the timer
- Opening with Phoenix. With no anomaly applied, her crit has no base to scale — the window is wasted.
- Leaving the stunner on field too long. Stun damage is low; every extra second is a second of damage window lost.
- Entering a high floor on an unfinished Disc set. A 4+2 that is not assembled yet drops the anomaly base far enough that restarting becomes the norm.
ZZZ 3.3 FAQ
Is Phoenix definitely in the first half of 3.3?
A: Yes. The official preview dated 2026-10-09 lists Phoenix and Verina in phase one, with phase two opening November 11 at 12:00 .
How much Polychrome do I need for Phoenix?
A: 28,800 Polychrome, or 180 pulls worst case. Any S-rank lands within 90 pulls and is 50% likely to be the featured agent.
Can I pre-farm Phoenix's materials now?
A: Seals and chips, yes. Maxing needs 66 promotion seals and 5/15/50 chips per skill; the exclusive material names are unannounced as of 2026-10-18.
Does Phoenix need her signature W-Engine?
A: No. The weapon banner worst case is 160 pulls and 25,600 Polychrome, close to 90% of a character. Start on the A-rank Candle Trick instead.
When does 3.3 end?
A: Phase two closes December 1 at 14:59, and the patch itself lands around December 2 by the 42-day count. HoYoverse has not published a separate end date — check the in-game notice.
Should I pull both Phoenix and Verina?
A: Not necessarily. Both are anomaly slots and a dual-anomaly team outperforms a single one, but the pair costs 360 pulls, or 57,600 Polychrome. Get one team running first.
Pre-Launch Checklist for the First Seven Days
- Bank 66 promotion seals and 300 Senior Investigator Logs before launch — both are guaranteed drops and the cheapest head start you can buy.
- Hold 3.105 million Dennies, with 405,000 reserved for core skill A→F so it does not get spent on disc upgrades.
- If you are under 28,800 Polychrome, keep saving rather than soft-pitying; the count carries over but the spent pulls do not come back.
- Assemble a 4+2 Disc set first, then re-roll substats; stop at 40% ATK% and 30% Anomaly Proficiency.
- Run one Shiyu floor to log a clear time before you change anything — optimize the loop on the second run, not the first.
Deeper patch budgets and team breakdowns live in the gghubs guide hub, and the Severian phase will be updated as HoYoverse posts the next announcement.