Eon and his dad made a game together for the Playdate — a tiny handheld console with a crank and just a couple of buttons. In this clip they do the part that comes after building: they play-test it. That means you sit down with the thing you made and try to break it, on purpose, to find out what really works.

You only get so many buttons

Right away they hit a classic game-design problem. Eon works out the controls out loud: "the jump right is A... B is the shooting." Then the catch — "so, not enough buttons." Dad names the lesson exactly: "When you design a game you have to really think about the interaction."

A Playdate has only an A button and a B button (plus the crank). If jumping, shooting, and moving all need inputs, you have to decide which action each button does — and sometimes one button has to do two jobs. That trade-off is real game design, and they ran straight into it.

Easy level, then "a lot harder"

Next they test the difficulty. Level one is gentle: "Level one is really easy. You got a couple of jumps." Then level two arrives and the tone changes fast — "Oh jeez, this is a lot harder." Eon loses a life, then another, shouting "come on, come on" at the screen.

That jump from easy to hard is the difficulty curve, and feeling it while you play is how designers tune it. Too easy and players get bored; too hard, too fast, and they quit. The only way to know is to play it and watch where you start to struggle — which is exactly what they're doing.

Why testing is its own job

Here's the big idea: building the game and testing the game are two different jobs. You can think your controls make sense and your levels feel fair — but until someone actually plays, you don't really know. Every "oh, I died" and every "not enough buttons" is information. It tells the builders what to change before anyone else plays it.

That loop — build, test, find a problem, fix it, test again — is how almost all real software gets made, not just games.

Try it

Next time you play any game, be a play-tester for five minutes:

  • Count the buttons. What does each one do? Could you control it with fewer?
  • Rate the curve. Where did it go from easy to hard? Was that jump fair?
  • Write one fix. If you could change one thing to make it better, what would it be — and why?

Then try the hardest one of all: make a tiny game of your own (even on paper), hand it to someone, and watch them play without helping. What they get stuck on is your next thing to fix.