What it is
MiPixArt Survival is a drawing game with a lit fuse. A picture appears, you repaint it pixel by pixel, and a spark burns around the border toward a bomb. Cross the pass line and you survive. Keep painting for two and three stars, if you dare.
There's a 72-level campaign that runs from tiny 8x8 icons up to a 20x20 castle finale, a DAILY mode where the whole world gets the same picture and you build a streak, and an ENDLESS mode with one life, a shrinking fuse, and a proper arcade hi-score table with three-letter initials. Four difficulties, from a trace-and-fill mode built for small kids through to two hearts and no mercy.
Where it came from
MiPixArt started as a free drawing toy and became a paid app on three stores. Survival is what happened when I asked a different question: what if the drawing had stakes? Same pixel canvas, same blocky characters, but now there's a timer with a personality. Wrong pixels make the fuse flare. A streak of correct pixels douses it. Every mistake and every hot streak is something you can hear.
It reuses the engine and the hard-won store plumbing from MiPixArt, so most of the build time went into what makes it a game: level design, difficulty tuning, the daily puzzle, and making the fuse feel fair.
The pricing experiment
This one matters to me more than the game itself. MiPixArt is pay-upfront: A$4.99 before you see anything. That works, but it filters out everyone who won't buy an app blind.
Survival flips it. The download is free, the DAILY is free every day, and there's a single one-time unlock for the full game at A$4.99. No ads, no subscription, no accounts. The free daily puzzle is the hook: it's the same picture for everyone, there's a share card built for the family group chat, and streaks give you a reason to come back tomorrow. If the game earns a place in your day, the unlock is sitting there.
Wiring that up meant my first in-app purchase: RevenueCat handling the entitlement, purchase and restore flows tested on real devices on both platforms. That's a whole layer of moving parts that pay-upfront apps never have to think about.
The store grind, round two
Shipping the second app was faster than the first, but not free. Apple and Google both wanted the full paperwork again: age ratings, privacy declarations, screenshots at the right sizes, review notes. By now most of that runs through scripts I built against the App Store Connect API, so a release is closer to one command than an afternoon of form-filling. Both stores got the submission on the same day.
Status
Submitted to the App Store and Google Play, currently in review. The web version at survival.flaux.com.au is live now and free to try.
The bigger lesson
The first app teaches you the pipeline. The second app is where the pipeline pays for itself. Everything painful about shipping MiPixArt, signing keys, store metadata, review queues, became a checklist this time, and a lot of it became code. If you're one person shipping apps with AI, the real asset isn't any single app. It's the growing pile of things you never have to figure out again.