Sunken cost-driven development - Or: why would you ever build a game engine?

A question I get asked a lot is why I decided to build a game engine. All in all, there’s already great commercial engines out there, that, as long as you don’t earn any money, cost nothing.

Unity and Unreal Engine are powerhouses with an abundance of tools and resources built on top of them.

In addition, if commercial engines are not your cup of tea and you wish to go open source, there’s a plethora of options there as well. Godot being the most cited one, but far from the only option.

However, one thing I have experienced when trying to build 3D applications or experiences using these engines that go far beyond a game, the abstraction these engines provide start to work against you, as they make a lot of assumptions on how you want to architect your software, and how to render things on the screen.

At a certain point you’ll be implementing your own custom UI framework on top of the one the engine provides, or you’ll implement a different way the scene should be represented. Want to import assets on runtime into the engine? Now you have assets that are tied to a specific engine version. (Unity’s AssetBundles).

So the game engine becomes a glorified level editor and renderer.

How hard could it be to just write that yourself as well?

Well, as it turns out, really hard. If I would have known how much I would be dealing with platform specific peculiarities or the wondrous world of transpiling shader code, and a seemingly infinite amount of unknown-unknowns, I don’t think I would have embarked on this journey.

But I’m glad I did, because sunken cost fallacy is an amazing thing, and having already sunken 3 months in this process (rookie numbers, I know), I don’t want to quit now.

My reasons have solidified a bit throughout this process and can be summed up by the following points:

• I like having control over software, and don’t want to be subject to the whims of publicly traded companies that are flying close to the sun.
• I want to understand how it all works. It’s not the desire to learn new things per se, but it is the curious drive of wanting to understand fully how the world works.
• In order to truly innovate in a highly technical field as a designer / creative person, you need to deeply understand the technology you’re working with, otherwise almost all decisions will be made from a engineering perspective, rather than from a user perspective. You’ll end up with hard to use software, that at most looks good.
• Contemporary game engines are overcomplicated, incoherently designed, cobbled together systems. Unreal Engine is marvelous in what it has been able to achieve in terms of technological advancement. But I have not yet seen a piece of software that is so obtuse and hard to understand.
• There’s so much wrong with software I encounter on a daily basis. I need a way to stay hopeful and optimistic, through “tending my own digital garden”. No Electron or web-based “developer time is more expensive than compute time”-type software such as Microsoft Teams. Just simple, elegant reliable, coherent, creative software.