Displaying posts by Christopher Allan Webber
Catching up with Godot: An interview with Juan Linietsky
byon December 13, 2016
Godot is a multi-platform engine for 2D and 3D games. It provides common functionality for graphics, audio, physics, and more, so game developers can build on top of a solid foundation and spend more time working on the unique parts of their game. Godot is a Conservancy member project. (Become a Conservancy Supporter!)
Christopher Allan Webber, co-maintainer of GNU MediaGoblin and a Conservancy volunteer, recently sat down to learn more about Godot with one of the core developers, Juan Linietsky. Recently Linietsky’s been working on a new graphics renderer for Godot with funding from a Mozilla Open Source Support grant that Conservancy helps administer for the team.
CW: Tell us a bit about your project. What’s Godot, and why might a potentially-interested user pick it up?
JL: To explain why Godot is important, I believe it’s enough to explain some numbers about the software industry. Globally, the software industry moves around 400 billion dollars in revenue, out of which 100 billion are from the video game industry. Games are a huge industry that keeps growing.
Game development happens mostly with proprietary software, and almost half of it is done with a single proprietary tool (Unity 3D). For being such a huge industry, there are no comparable open source alternatives like Apache or Linux that game developers can use instead of proprietary solutions, even if they would be very welcome.
Godot is a project that aims to change that, by being a comparable open source offering. It’s growing very quickly in community and has a large and dedicated group of users and developers.
CW: So you’re a Conservancy member project! What was the decision like to join under the Conservancy umbrella?
JL: Godot is a project that has many users, institutions, companies and other non profit organizations interested in funding its development. As we the developers are only individuals, we need to make sure our donors and sponsors have a strong promise that their donations will be used for the benefit of the project. Conservancy is an institution with a flawless track record in this regard.
CW: Godot has its own internal language, I believe partly to control issues like garbage collection pauses. But Godot recently started adding support for C♯, which is garbage collected. Could you talk about your history of game scripting support, its motivations, and where you see it going in the future?
JL: The reasons why we ended up using our own scripting language are not obvious, but are very strong. Popular languages like Python and Lua were not designed with real-time applications in mind. The garbage collector can cause random stalls, and support for multithreading is not so efficient because they use giant locks. On top of that, the complexity of binding them to Godot was huge and very error-prone.
GDscript is a language that “just works” and Godot users love that. That said, we’re adding C♯ support due to popular demand, and the fact that Mono was released under the permissive MIT license a few months ago.
CW: How would you describe Godot’s community, and your relationship with it? Is there a direction you’d like to go towards stewarding community interactions?
JL: The Godot community is awesome, they are very helpful with each other (and if they are not, we make sure they are). We try to help everyone, no matter how much (or little) experience they have, and always encourage users to help each other.
CW: If someone is interested in getting into game engine development, with Godot or otherwise, where should they start and what sorts of resources are available to them?
JL: There is some documentation on the C++ side of things, and the code in general is well-organized. The best approach to starting with development is to implement your own feature (make sure main devs would like to merge it first) and then do a pull request.
CW: How long until Godot overtakes Unity? ;) More seriously, do you see Godot as an alternative to proprietary engines like Unity or the Unreal Engine? Or do you see Godot as serving a different niche?
JL: Our development is very focused on the following premises:
Godot must be very easy to use, simple and to the point, even by compromising flexibility if necessary.
It must be easy to make a 2D or 3D game look great with little effort.
So, in pratice, we aim for an engine that is easier to use than Unity, and that looks better than Unreal, and we are very close to getting there in the upcoming Godot 3.0.
CW: What kind of overlap do you have or hope to have with related libre software media production projects and communities like Blender?
JL: Many Godot users create their 3D art on Blender, and we’d like to have a better relationship with the team. We just heard they are working on a PBR viewport, so we will work hard to allow users exporting their 3D content as faithfully as possible to Godot.
CW: Thanks for the interview! Anything else you’d like readers to know about Godot before we wrap up?
JL: We always listen to our users, their problems and wishes even though we can’t always take immediate action. Currently we are focused on shipping a 3.0, wich will come with a completely new renderer. Godot 3.1 will improve some peripheral areas like the physics engine and the audio engine. Afterwards, I dream of a time where no new large features need to be implemented, so we can polish and improve as much as possible everything that is there!