Considering other game engines...

A few weeks ago, we decided to continue this game and try to release it in 2026. Although we did not communicate much about this, we already wrote the whole story and made quite a few decisions about how the game will evolve.

Until last week, it was nearly taken for granted that we would use the Godot game engine, but I came up with the idea that this would be a nice opportunity for me to (at least) try another game engine (I've been using Godot for more than 10 years, and only touched Unity and Unreal to test Material Maker exports).

So I started trying to port some of the main features from our Little Shadow gamejam game, and spent 2 to 4 hours a day discovering Unreal for a week.

Day 1: Basic character movement

I don't like huge tutorials that teach you how to create a full game. I prefer a small video that explains a tiny, very specific task. I found one that explains how to create a first person or third person character and started following it. I also started to figure out blueprints and the overall structure of an Unreal Engine project. I finally ended up with a very basic Shadow character that I could control and follow with the camera. The controls were nowhere near what is expected, but at least there was progress.

Day 2: more character movement

The day after, I kept following the tutorial  and started digging more into Blueprints to add animations for our little shadow character.

Day 3: Sound and character movement

The third day was dedicated to adding sound and updating the character movement.

I spent most of the time trying to integrate the Wwise plugin (not add sound to the project, just load the project with the Wwise plugin enabled). Wwise itself is quite easy to use in Unreal (but I think it's still a bit easier with Godot since you can do nearly everything using the Wwise global).

I also fixed character movement so it behaves like our game jam game.

Day 4: First interactable object

On day 4, I started adding basic interaction with objects. This first test is extremely simple (interacting with cubes will rotate them by 45 degrees), but it was an opportunity to give a first try to the widgets system).
Blueprints are really easy and elegant, but there are way too many nodes when you're discovering them. I'm pretty sure this will be a really weird idea for seasoned Unreal users, but a "Nodes I already used" filter would help a lot.

Day 5: More interactable objects

The next day, I took the time to update the interactable objects so we have switches that control lamps.With a bit more experience with blueprints, this was pretty easy.

I started using timelines (they seem to be similar to Godot's Tweens) to animate the switches, and they are pretty convenient.

Day 6: The menu

On day 6, I started adding a very simple menu to dig into Unreal's UI features. This was pretty easy and I replicated all logic that's used to control the audio bus volumes (this uses Wwise game parameters).

I also started adding music, but failed to find how to control Wwise switches (didn't want to investigate further, though).

Day 7: Detecting light

The main mechanic in the Little Shadow game is that you must avoid light if you want to avoid being slowed down and injured. I used a very simple raycast approach and calculate light manually. This probably would have to be fine tuned, but at least this basic implementation works.

Conclusion

I had lots of fun discovering Unreal and trying to port Little Shadow's main mechanics. Unreal seems way easier than I suspected, but I only scratched the surface for now.

Will we use Unreal for Little Shadow? Nothing is decided for now, and there's still a lot to be investigated before making a decision. I will not push this experiment further for now, because I will be busy with Material Maker in December. But I'll probably get back to trying stuff with Unreal once 1.5 is released.