Time

When I travel, I like to take my time: observing, discovering cultures, people, flavours. With my tools, it’s the opposite. I can’t stand waiting on them.

In my last years in the AAA industry, waiting 30 to 45 minutes to recompile a project, then another 10 minutes to test a change, was part of my daily routine. Whether it was an in-house engine at Ubisoft or Unreal Engine, it was the same story. I understand why these tools are so big, and I find the technology they bring fascinating, but I think we underestimate what that time really costs.

For me, one thing is clear: the loop between an idea and testing it in the game has to be extremely short. Iterating should feel smooth and simple.

Under 5 seconds

The first time I launched Godot on my Mac (M1, 2020), the editor opened in less than a second. In my opinion, that’s the minimum an editor should offer for an empty project, whatever the technology behind it.

Four weeks later, with a prototype of about thirty files, opening the project, changing a few lines of C# and getting into the game takes me less than 5 seconds. Last weekend, I exported a build so my brother could try an early version on Windows. The whole game weighed 194 MB. Within an hour, he was able to give me his first impressions. That’s fantastic!

What I compared before choosing

Before deciding, I compared Godot with Unreal and Unity:

  • The type of game. Probably the most important point: a clean, systemic strategy game doesn’t need cutting-edge graphics technology.
  • Cost. Godot is free and open source (MIT license), with no royalties, no matter how much the game earns. Unreal takes 5% of revenue beyond the first US$1 million. Unity is free under US$200,000 in revenue, then costs about US$2,300 per seat per year.
  • Stable rules. A private company can suddenly change direction, its level of support or its terms of use. With an MIT license, those changes are much harder to impose along the way.
  • A proven track record. Slay the Spire 2 moved from Unity to Godot and sold more than 3 million copies in its first week of Early Access. Brotato, Backpack Battles and Dome Keeper are also successful games made with Godot.
  • Languages. C# for the game logic, and C++ (through GDExtension) when a part needs more performance.
  • Architecture. In Godot, everything is a node, and scenes nest inside other scenes. One concept, all the way up. It’s simple, and simple is good.

The trade-offs I accept

Godot isn’t for everyone, and I know it:

  • Consoles. Console support isn’t built into the engine, because console tools are under non-disclosure agreements. It will take more work if I expand to consoles.
  • Web. A Godot 4 project in C# can’t be published for the browser yet, so no playable web demo for now.
  • Optimization. Some parts of the C# code will take more work to optimize.
  • Ecosystem. There are fewer ready-made tools and resources than with Unity or Unreal.

The fun of playing comes first

My priority is a strategy game with balanced, clean and polished mechanics. Balance isn’t found on the first try. It’s found by testing, tweaking and testing again, hundreds of times. Every minute saved in that loop is one more test.

I don’t like waiting on my tools. I don’t want you waiting on my game.

The rest remains to be proven in the coming months. I’ll keep you posted.