Welcome to This week in Game Engines! This is a digest on gamedev tech news and articles from the week before.
Godot number of contributions and contributors increased quite a lot over the years since Godot got open-sourced back in 2014. From the blog post:
"The pool of contributors has grown as well and it is no longer feasible for new contributors to slowly learn the vision of the engine by chatting with the engine maintainers. Accordingly, the Board of the Godot Foundation has decided that it is finally time to write down a high-level statement of what our vision for the engine is so that it is easily accessible to all."
Per statement, this doesn't mean Godot changed anything on what it is/aiming for. It just makes the target mission of the Engine clearer for every contributor, specially newer ones. The vision itself is a small document you can read here.
Coincidentally, Godot Foundation also updated contribution policies this week, thanks to the short of code reviewers compared to the wave of AI-generated Pull Requests every open-source project is getting. Main point is similar to the general AI discussion regarding AI-generated PRs: contributions to an Open-Source project is also a way to "train" new contributors to become full-fledged maintainers, and AI is making the process much more difficult (and demoralizing) for most open-source maintainers out there, so a line is drawn to define how people should contribute to the project, and to make PR filtering process easier.
On the AI note, Castle Engine (my favorite Pascal Game Engine) published a post on the process of reviewing the engine codebase using Claude Code. The main motivation is nicely explained here:
"AI reviews of commits/PRs by Claude are useful in my experience. [...] I have to emphasize that by useful I mean that they contain useful bits. Of course they are also filled with nonsense — AI misunderstanding what the application/engine around it does, misunderstanding differences between FPC/Delphi or particular platforms, suggesting bad approaches (always with confidence), suggesting “defensive” coding techniques that would actually hide problems instead of keeping the flow validated…
But if you’re prepared + capable of filtering out the nonsense, you’re left with useful observations how to improve the code. AI reviews do sometimes note a real improvement or even real bug-fix that my own eyes didn’t catch — that’s why I use it. It’s a special case of “many eyeballs make all the bugs shallow” rule, here AI is another set of eyeballs."
This is obviously different from the new-contributor problem that Godot is trying to solve, but it's a good perspective on the usefulness of AI as an auxiliary tool by an experienced contributor (and in this case, creator).
A nice blog post from Defold website about the last 6 months of development, starting from version 1.12.0 all the way to 1.13.0. Some of the highlights:
Overall, a great blog post to understand a bit what Defold is and what's aiming for. Specially if you are interested in making web games, Defold might be worth taking a look.
(I know, everyone is being overwhelmed by AI related news. I'll try to filter some here only when it's relevant for gamedev, or when I think it's an actually good article).
If you have any suggestions, send me feedback on Mastodon! And if you want to add a new game engine to the website, consider suggesting a new Game Engine.