The first dev stream since the mint sold out ran with Adam, Craig, Spencer and Matt on stage — and it delivered on three fronts at once. The team held its own post-mortem on the free mint, dropped an enforcement number nearly triple the one we published a day earlier, and finally opened up the most undershared system in the game: battles. There are dates, too — softer than a countdown, firmer than “Q4.”
The standing disclaimer got extra weight this time: the game is in beta prototyping, ideas are changing right up until launch, and the team asked — only half joking — not to be haunted by screenshots a year from now. Everything below is what was said on September 17, reported as such.
The mint, graded by the people who ran it
Nobody on stage pretended it was smooth. The mint was heavily oversubscribed, the traffic hit every Ronin endpoint at once, and the middle waves turned into exactly the click-race chaos we graded C+ before it happened — wallets reporting “not allowed,” balances showing no RON, pages refusing to load, and one team member refreshing in a panic from a restaurant. Players with prepared setups — extra gas ready, alternative wallets like Rabby — had a visibly easier time.
Two takeaways matter going forward. First, this was the team’s first Ronin mint, and Ronin has since recommended its raffle system for drops with tens of thousands of simultaneous minters — it spreads the load across days instead of a single stampede. Second, the team repeated that future in-game drops will run through in-game systems, the way Sunflower Land does its big distributions, precisely because off-chain tech is smoother at this scale. Read those together and the Genesis mint starts to look like the exception, not the pattern.
On the why-free question, the answer was strategy, not charity: the drop earned the studio nothing by design. The goal was eyeballs on Ronin — hype, new players, a flag planted in the ecosystem — with upfront revenue deliberately traded for longevity. The contracts themselves came through clean: exactly 10,000 minted, not one extra, with Craig noting the deliberately bare-bones NFT contracts carried extra code written just for the Ronin launchpad, future-proofed for every Yakkamon still to come.
973 Bad Eggs — and better Legendary odds for everyone else
The enforcement update was the news of the night. The Bad Egg count now stands at 973 — read off the live filter on stream, after the host first ballparked it at 900-plus. That’s nearly triple the 365 we reported on completion day, and close to one minted egg in ten. For scale, the team said a typical NFT free mint sees this sit around 3%.
Then came the sentence that changes the math for everyone who played fair: the Legendaries and Rares will not be distributed among Bad Eggs. The 68 Legendaries and 50 Rares land in clean eggs only — so every flagged egg quietly improved the reveal odds of the other 9,027. Multi-accounting didn’t just fail; it backfired into a gift for the honest majority.
The review is essentially done. The team described the latest pass as one of the last markings, with monitoring continuing — a few more flags can still land, but no big wave is expected. That’s also exactly why the advice hasn’t changed: don’t trade until Ronin Market opens on September 21. And when it does, Bad Eggs will be right there on the marketplace — they’re NFTs, nobody can confiscate them — visibly labeled, trading like any other sealed egg to a buyer who doesn’t check the Status trait first.
The team also credited the community’s own chain-watchers — including players running home-built, AI-assisted forensics — as a real part of the enforcement effort. On a public blockchain, it is getting genuinely hard to hide.
Battles, finally: an auto battler where the plan is 70% of the fight
Matt — the design intern the team credited for the combat system back in August, now in his final semester of college — gave the first proper public tour. The team says it has been through roughly 40 iterations of the battle system, and the style is now cemented even as details keep moving.
The shape: an auto battler. You pick your squad — three or four Yakkamon, the count still settling — place them in the scene, and they fight on their own. The stated target is a 70/30 split: 70% of a battle is decided by planning — who you bring and where you put them — and 30% by what you do live, like throwing a potion to keep a key monster standing. Under the hood it runs turn-based, but fast enough to read as real time. The August framing of a lane-based 3v3 has loosened into something freer about placement; the planning-first spirit survived intact.
What makes it click is conditional skills. Every Yakkamon has a normal attack, and skills fire when their trigger condition hits — the team’s favorite example is a little seed Yakkamon who puts enemies to sleep when he dies. Chain reactions, where one skill’s effect trips another, are a deliberate design goal, and setting one up in advance and watching it cascade is described as the most satisfying moment in the game right now.
The internal meta already exists: AROC, “the little rock guy,” parked up front as a tank while fragile hitters work from behind him — three separate team members named the same lineup. Consumable battle items are coming to the live-input side: health potions, plus an experimental area-of-effect “blizzard pot” that Craig and Matt literally finished coding the morning of the stream (their own caveat: it may not survive to launch).
Two hard answers for roster planners. Battle stats are four — Health, Attack, Defense, Speed — not Pokémon’s six, with room left for non-battle stats later. And identical buffs do not stack, exactly as the August ruling said: two of the same creature, or two different creatures with the same ability, give you the effect once. The team declined every deep stat question on purpose — balancing is named as the big remaining hurdle, and numbers will move through beta.
Why battles can’t sprawl: Yakkamon is a farming, automation, hunting, collecting and battling game at once. Fights have to be short enough not to eat the rest of the loop, strategic enough to reward the planning — a balance the team says the current build finally hits. Full system detail: The Battle System →
After crunch week, “it really feels like we have a game”
The team ran a deliberate crunch week with one goal: every piece of early-access content in the build, polish or not. Spencer’s verdict — the moment the separate parts stopped feeling separate — was the emotional center of the stream. Where the build stands:
- 30 Yakkamon are in the game — confirmed as the early-access roster. Our public roster stands at 23 revealed, so seven are still faces nobody outside the studio has seen.
- 36 levels are built out, and the Legendaries are already in the world — “bopping around,” in Spencer’s words.
- Hunting is the least settled system — areas exist, but the mechanics are still being prototyped, which matches everything published so far.
- Mobile testing is a dedicated focus right now, and animation polish is the between-tasks job.
- Onboarding is the obsession: the team called the first few clicks make-or-break for keeping new players, and that’s where the fine-tuning effort is going.
Craig’s first week on Yakkamon gameplay code — after months on the trainer dashboard, deposits and contracts — also produced the stream’s accidental reveal: a Yakkamon Training Center, a way for monsters to train themselves up, is in the works. He showed a teaser image and then stopped himself, several leaks too late.
The dates: beta in October, early access aimed at mid-to-late November
Asked directly when the game arrives, the team went further than the official “Q4” for the first time — with the explicit caveat that nothing is final until it’s close:
| When | What |
|---|---|
| September 21 | Ronin Market opens — sealed-egg trading goes live |
| October (internal aim) | Beta testing begins |
| Mid-to-late November (best case) | Early access launches — officially still Q4 until confirmed |
| Early access + ~6 weeks | The “six weeks of gameplay” runway ends and the first hunt begins |
| ~6–8 weeks after early access | Chapter Zero — the additional content drop |
| Early next year | The full competition arena |
Two process commitments came with it: the team will give roughly two weeks’ notice before early access opens, and the leaderboard finalizes a few days before that date — so nobody gets ambushed by the lock. On the first hunt, expectations were managed carefully: early access buys a head start on setting up your farm and systems, not an exclusive window on Legendaries and Rares — the gates open wider for pre-registered trainers when the hunt itself begins. The full schedule →
Gyms, regions that never reset, and monsters that put themselves to bed
Gyms are further along than anyone knew. An internal prototype already works: the host showed his own boulder-decorated, rock-themed gym and invited challengers. The plan is to ship a version of this before the arena — decorate your gym, visit other trainers’, battle them — initially outside the core reward loop, but with leaderboards and status attached. It doubles as the team’s live laboratory for win rates and the matchmaking system Bryn has wanted since his Gods Unchained days.
Expansion learned Sunflower Land’s hardest lesson. Growing your land works much like SFL — pay the cost, land gets bigger — with one big difference the team stated flatly: lands never reset. You build your first land to its finish, move on to a new region, and keep full access to everything behind you. Players hating land resets was named, out loud, as the SFL mistake not being repeated — consistent with how the Regions post already described it.
The automation half stays true to the pitch. Assign a Yakkamon to a job and it manages itself: when stamina runs out, it walks to a bed, sleeps, and gets back to work — after battles too. You can override manually, but the intended role is overseer, not micromanager. The detail with teeth: too few beds and tired Yakkamon queue for the mattress — rest capacity is a real resource to build around.
Quick hits from the Q&A
- No Yakkamon shortcut in Sunflower Land’s map — but crossovers between the two games are planned, in both directions.
- Dedicated teams, with rotation. Yakkamon and SFL each keep their own people, rotated over time to avoid burnout; Craig (technical direction) and Bryn (game design) lead across both.
- Buffs, stats, deep mechanics — deferred to beta and early access on purpose. Specific numbers given now “can be guaranteed” to change.
- Wallet metadata caching (will a Bad Egg show as bad in MetaMask?) — the team is making sure Ronin Market itself shows the right images and descriptions, but couldn’t vouch for every wallet’s cache. Check the marketplace page, not your wallet.