How We Built Catch the Cup
In short
Catch the Cup started as a prototype with power-ups, currency and multiple characters. Almost all of it was cut. The final game is one tap and one score, and the majority of development time went into input response and difficulty pacing rather than features.
The first version of Catch the Cup had power-ups. It had a coin currency, three unlockable characters, and a shop. It was playable in about a week and it was not good.
The prototype problem
Everything worked. Nothing was broken. And playing it produced no particular desire to play it again — the most damning result a prototype can give you, because there is no bug to fix.
What we eventually worked out: every system we had added was competing with the tap. A player deciding whether to spend coins is a player not building rhythm. A power-up that trivialises an obstacle removes the moment the game exists to create.
The cutting
So we removed things, in roughly this order:
- 1.Power-ups. They made the hard parts easy, which is the wrong direction for a skill game.
- 2.Currency and the shop. Both existed to support the power-ups, so they had no reason to remain.
- 3.Character selection. Cosmetic variety in a game where you look at obstacles, not at yourself.
- 4.Level structure. Discrete levels fought with the endless format. One of them had to go.
What survived: run, tap, avoid, score. That is the entire game.
Two weeks on one tap
With nothing left to hide behind, the tap had to be perfect. That meant measuring the full path from finger contact to visible response and removing every avoidable millisecond — touch handling, animation start, frame timing.
The gap between a tap that feels connected and one that feels slightly detached is small enough that most players cannot name it, and large enough that they can feel it. Sub-100 milliseconds is the threshold where the response stops feeling like a reaction to your input and starts feeling like part of it.
Getting the difficulty steps right
The original curve raised speed smoothly with distance. It tested badly in a way that was hard to articulate — players said it felt 'flat', which is not a bug report you can act on directly.
Switching to stepped increases fixed it. Hold a speed until the player is competent, then step up. The step is briefly uncomfortable, then they adapt, and the adaptation is the moment that feels like improvement. A smooth ramp never gives them that moment.
What we would do differently
- Cut earlier. Every feature we removed was a feature we had built. Building then deleting is expensive; a stricter first spec would have been cheaper.
- Test with strangers sooner. People who know you will not tell you a game is boring.
- Instrument the failure screen from day one. It is the most-viewed screen in the product and we treated it as an afterthought for too long.
The game got better every single time we removed something from it.
Catch the Cup is out now on the App Store and Google Play. It is small, it is free, and it is exactly as complicated as it needs to be.
Play the result
Catch the Cup — free on iOS and Android.
Frequently asked questions
How long did Catch the Cup take to make?
The playable prototype took days. Making it feel right took months — most of that spent on input latency, obstacle spacing and the pacing of difficulty increases.
What engine is Catch the Cup built on?
It is a native mobile build targeting iOS and Android, deliberately kept small so it downloads quickly and runs on older phones.