I Had Never Opened Unity. Four Days Later We Shipped a Game.

2026-07-28

On Thursday I did not know where Unity keeps its play button. On Sunday we submitted Count Downcula to GMTK Game Jam 2026. This is what happened in between.

I want to be precise about the starting point, because it's the whole reason this post is worth writing.

I had never opened Unity. Not "hadn't used it in a while." Not "know the concepts, rusty on the editor." Never. I opened it for the first time on a Thursday, and by Sunday night my name was on a shipped game alongside nine other people.

That gap — between total beginner and shipped — is the thing I keep turning over. It's much smaller than I thought, and the reason has almost nothing to do with how fast I learned Unity.

The theme was "Count Down"

GMTK gives you a theme and a weekend. Ours was Count Down, and the trap with a theme like that is obvious in hindsight: it's so easy to bolt a timer onto a normal game and call it done. A countdown in the corner. A bomb. A level that ends in sixty seconds. Theme satisfied, technically.

We spent day one refusing to do that.

The idea we landed on takes the theme exactly one step further, and that step is everything:

Your health bar is your clock. It drains every second you're alive.

Say it out loud and you can feel what it does to the game. Every action game you've ever played has trained one instinct into you: when you're hurt, back off. Create space. Wait for an opening. Heal. Be patient — patience is free.

Here, patience is the thing that kills you. Backing off is bleeding. Waiting is dying on schedule. The safest-feeling move in your muscle memory is the one draining you the fastest.

So how do you get time back? You take it.

You parry. You break the enemy's posture. You execute — and the execution is what pays your clock back. Then, whatever HP you have left when the enemy drops becomes the currency you spend on upgrades before the next one walks in.

Kill fast, stay rich. Hesitate, and the clock finishes the boss's job for it.

That's the entire game in one mechanic, and I love that you don't need a tutorial to explain it. The dread does the teaching.

Day one: we argued instead of building

The instinct on the first morning of a 72-hour jam is to open the editor immediately. Time is bleeding — very on-theme — and every hour spent talking feels like an hour not spent making.

We talked anyway. We ran the brainstorm through Apple's Challenge Based Learning framework, which is basically a structured way of refusing to fall in love with your first idea. You take the theme, pull it apart into the actual challenge underneath it, and only then start proposing solutions.

It felt slow. It was the highest-leverage day of the jam.

Because the thing we walked out of day one with wasn't just the health-bar-as-clock idea. It was one written-down decision that saved us: one state machine, agreed on by everyone, on day one.

Architecture pays you back fastest under pressure

This is the lesson I did not expect to learn at a game jam, and it's the one I'll carry into every project after it.

Ten people. Four programmers, six artists and animators. Three days. Under those conditions, the bottleneck is never how fast any individual can work — it's how often two people's work collides.

We agreed on a single state machine on day one and wrote it down. That one artifact is what let ten people build in parallel without stepping on each other. Everyone knew what a state was, who owned which one, and where their piece plugged in. No one had to stop and ask permission, and no one merged two versions of the same idea.

I used to think architecture was a luxury — the thing you do when you have time. A jam inverted that for me completely. Architecture is what you do precisely because you don't have time. It's not overhead you pay; it's parallelism you buy.

Day two: six artists, one style

The other collision risk was visual. Six people drawing at once should have produced a game that looks like six games.

We locked the asset list on day two — decided what existed before anyone drew it — and held the line on a shared style. Six artists worked simultaneously and it still reads as one thing. That's a coordination win more than an art win, and it came from the same principle as the state machine: decide together early, then work alone fast.

Full credit to them, because that's the part players actually see.

"Done" beats "clever"

Here's where being new was almost an advantage.

A jam deadline is a very honest teacher. You stop asking whether the code is elegant and start asking whether it ships. There's no room to be precious. I wrote things over that weekend that I would flag in a code review on Monday, and shipping them was the correct call, because a beautiful abstraction inside an unsubmitted game is worth exactly zero.

I don't think this generalizes to everything — most software isn't a jam, and "just ship it" is how you end up maintaining a swamp. But it was clarifying to spend three days in the extreme version of that tradeoff, where the deadline strips away every argument that isn't does this exist yet.

On being the newest person in the room

I was the one who'd never opened the engine. There were nine people around me who knew more than I did about basically every part of this.

That's a fine place to be — as long as you're the one asking the questions.

The failure mode isn't being a beginner. The failure mode is being a beginner who's quiet about it, burning four hours on something a teammate would have unblocked in four minutes, because asking felt like admitting something. Nobody on that team cared that I didn't know Unity. They cared whether the thing I owned worked by Sunday.

Four days is not enough time to become good at Unity. It's plenty of time to become useful on a Unity team. Those are different skills, and only one of them was actually required.

What we shipped

Count Downcula — built with Unity 6 (2D URP), free, about ten minutes long.

Made at the Apple Developer Academy @ UC Surabaya — Game Development Club.

Programming

Art & Animation

👉 Play it here — it's free and takes about ten minutes. We'd genuinely love feedback on the page.

If you've been putting off a jam because you don't know the engine yet: that was me on Thursday. The engine is the smallest problem you'll have.