Narrowing My Options

I think the saying is true: less is more.

I thought I needed to build something big. Something original, something deep, something that felt like a real game. That’s where my mind went immediately.

Where It Didn’t Add Up

Before I started building anything, I looked at the current submissions for Vibe Jam. Then I went to the official site and started playing some of them: https://vibej.am/. That’s when things started to feel different. Most of the games were simple, but not in a bad way.

They didn’t have grand stories, mind-blowing graphics, or massive systems. They loaded quickly, ran smoothly on my laptop (which is pretty weak), and within a few seconds I could understand what I was supposed to do.

That forced me to look at my own idea more honestly. I realized I was trying to make something impressive before I had even made something playable.

The Shift

That was the first real shift for me. My idea didn’t need to be revolutionary or prove everything I could imagine. It just needed to work. Something simple, clear, and playable would already be enough. That realization sounds obvious in hindsight, but it completely changed how I approached the project.

Breaking My Own Thinking

Instead of opening a game engine and trying to build immediately, I did something different. I ran my ideas through ChatGPT, not to generate the game itself, but to simplify my thinking. At the time, I had this massive idea in my head. It was closer to a full Final Fantasy-style game with multiple characters across different countries, investigation segments, combat systems, and a final boss tying everything together. It pulled from Final Fantasy, Kingdom Hearts, and The World Ends With You.

It made sense in my head, but it was completely unrealistic for a game jam. So I asked a simpler question: what can I extract from this?

From Too Big to Usable

It came back with smaller, more focused ideas. Here were a few:

• Echo Snap - React to audio cues by selecting the correct visual match. Speed and distortion increase over time.
• Can You Hear It? - Identify sounds or words buried in noise. Noise, accents, and speed increase difficulty.
• Tone Sprint - Quickly distinguish between very similar sounds (tones, pitches, phonemes). Designed to break “I know it but can’t hear it.”
• One Strike - Attack only at the perfect moment based on a cue. Mistiming = damage.
• Fake Out - Enemies telegraph attacks—but sometimes lie. You must learn which signals are real.
• Don’t Fight - Conflict situations appear → you choose to escalate or de-escalate. Avoiding fights leads to better outcomes.
• Voices in Paradise - Peaceful world with hidden manipulative messages. Player must interpret meaning from distorted “kind” voices.
• One Day - Live a single day via choices. Your ending depends on how you balance actions.
• As One - An action rpg where you randomly control one of three characters to take down a gauntlet of enemies.

Out of all of these, one of them stood out to me: As One, an action RPG where you control one of three characters randomly in combat.

For the first time, the project felt smaller without feeling empty. It felt like something I could actually start.

Choosing Something I Could Actually Make

I chose As One, not because it was the most impressive idea, but because it felt possible. At that point, that mattered more than anything else. I still wanted the game to have some kind of emotional weight, so I kept a loose backstory in my head. Nothing fully written out, just enough context to guide decisions as I built.

The idea itself came from something bigger I had imagined before. Instead of forcing that entire vision into this project, I took one piece from it and focused on that.

The Structure

The structure I landed on was simple. The opening screen shows a city in turmoil, and the player is dropped straight into a battle controlling one of three characters: a police officer, a child with a tiger in an attack stance, and a man. Each character changes the environment, with the officer in the city, the child in a paradise-like setting, and the man in an abandoned house. Once I started actually making the game, some ideas morphed due to the limitations of building with AI but the general feeling stayed.

Each round, the player fights one of a few enemy types. Each character has two or three actions, and the inactive characters act as assists. The round ends when the player wins or dies, and the game ends when all characters die, all enemies are defeated, or time runs out. It wasn’t complex, but it was enough to begin.

Letting It Stay Incomplete

There were still ideas I wanted to explore, like having the environments affect gameplay in different ways. But I knew that if I tried to solve everything at once, I would probably never start. So I let the idea stay incomplete.

That felt uncomfortable at first, but it turned out to be necessary.

Learning How to Start

At that point, I realized I didn’t need more ideas. I needed to build. I watched a video on how to vibe code a game (https://youtu.be/yKyjcbQiar4?si=RQpjazyGVzT-GTeg), not to follow it exactly, but to understand how to get something on the screen. Then I started gathering assets, testing things, and trying to make the project real in the smallest way possible.

Nothing polished. Just something that existed.

Where This Goes

I didn't know if my game would be good. I didn't know if I would finish it. But I did know I made an important shift. I went from trying to build something big to building something I could actually start.

And that’s enough.