Since I was a kid, I’ve always wanted to work in the gaming industry. Yet 15 years into my professional career, I’ve worked almost exclusively on the web. Be it for my job or for side projects, I love the web and I enjoy what I’m doing and learning on a daily basis. But I keep wondering if and how I could also get into gaming some day.

I figured that publishing a little game is something I should do. I have a couple of posts in mind that I’d like to write on this topic, but first: Let me tell you how I’ve “failed” in the past.

What usually happens when I “want to make a game”

I remember printing out hundreds of pages of DirectX C++ API reference manuals at home—sorry mom! In 2010, I tried making a Tower Defense game for Global Game Jam using Web APIs, but performance was way too bad, and WebGL didn’t even exist back then. Over and over, my brain had somehow tricked me into trying to write a game engine instead of making a game.

Therefore, choosing elements that make a game

• 3D graphics

• Third person camera

• Genre: RPG, Open World

• Graphics style: Placing emojis on an open world map landscape

• Story: Well… Something ridiculous! I’m starting to really enjoy writing, and I haven’t written fiction since school, so this is really exciting for me! Coming up with infinitely complex silliness is something I do on a daily basis.

Leveraging open-source Deep Learning projects

I absolutely think I can use publicly available generative Deep Learning tech for a couple of elements in the game:

• Synthesizing text to speech for character voices. I wonder if intonation is a thing yet.

• Use a combo of GAN (e.g. thispersondoesnotexist.com) and style transfer for character avatars.

• Hell, maybe even throw some random GPT-3 in for infinite lore!

Thing is, video games are only going to get bigger and more interesting in the future. And I would like to be part of that. So might as well start by making a game :)