Starting Something

I'm making a game.

I've been circling this idea for a while now. It started as a document, then it became a design vault, then a set of decisions, and today it became a real project with a real engine and real folders waiting for real code. The working title is project_void. It's a psychological JRPG, turn-based, 2D pixel art, built for PC. That's about as much as I'm going to say about what it is for now.

What I will talk about is the how.

Why Godot

I landed on Godot 4 after a longer deliberation than I'd like to admit. RPG Maker MZ was the first serious contender because it's purpose-built for the genre, but the combat system I've designed is too structurally weird for RPG Maker's assumptions. I'd have been fighting the engine more than building the game. Unity was briefly considered and quickly dismissed. Godot is open source, has no licensing surprises, and gives me the control I need without the overhead I don't.

The twist: I'm writing it in C#, not GDScript. I'm primarily a JavaScript developer, and C# is new territory for me. But the syntax overlap with JS is real, the strong typing will keep a complex project manageable, and honestly, I just wanted to learn something new. Sometimes the best time to pick up a language is when you've got something worth building in it.

The Toolchain

This is the bit that genuinely excites me. The development workflow I've set up is probably overkill for a solo dev, but it feels right:

  • Obsidian for the design vault. Every design decision is logged, numbered, and reasoned. If I come back to this project after a break, the vault tells me exactly where I left off and why every choice was made.
  • Godot 4.6 with .NET as the engine.
  • Claude as a development partner. Not writing the game for me, but acting as a documentation system, architectural sounding board, and increasingly, a direct interface into the Godot editor via MCP.
  • A Godot MCP plugin that lets the AI tooling talk directly to the editor. Read project state, manipulate scenes, diagnose issues, run the game. It took a few attempts to get the connection working, but it's live now.

The design vault alone has 25 documented design decisions, a full combat system specification, narrative foundations, and character breakdowns. None of that is code yet. But it means when I start writing C#, I'm implementing a solved design rather than designing in the editor. That distinction matters.

Why This Game

I'm not going to talk about the story. Not yet, maybe not ever in these posts. What I will say is that this game is personal. It's not a portfolio piece or a commercial play. It's something I need to make. The themes are ones I've carried around for a long time, and turning them into a game is the first time they've felt like they have somewhere productive to go.

That's enough for now. The project has a repo, a folder structure, an engine, and a toolchain. Next step: placeholder assets and a character moving on screen.

Let's see where this goes.