Ask HN: Favorite Game Engine?
Given Unity's recent announcement what's your favorite engine/framework to work in/with and why? It can be one you made, niche, or super popular.
My favorite is Panda3D [0] because I can use it from Python and C++ and that it lets you code your game how you want while being batteries included at the same time. It was used for some of Disney's MMOs and themepark rides too so it's production ready. Overall it's a lot of fun to work in and I'm also always discovering cool new features like the very good async support it has.
44 comments
[ 3.0 ms ] story [ 165 ms ] threadBut maybe not real game engines, as you need to do quite a bit of work by yourself. I guess it depends what your definition is of a game engine.
I tried Unity and didn't like it. I don't like drag & drop interfaces, but I understand the appeal.
---
[0]: https://love2d.org
[1]: https://lovr.org
nowadays I just do my own stuff but it's kind of interesting that nobody's really made an open-source Game Maker clone, to my knowledge.
For Desktop/Mobile I'd use Godot or MonoGame with C#.
For silly stuff or really short games and/or prototypes, Pico-8 is hard to beat.
And when the low-level itch starts, Raylib with C.
https://github.com/blitz-research/blitz3d
I've always wanted to try Haxe, because of my Flash background: https://haxe.org/use-cases/games/
Kinda reminds me of a lightweight C++ Unity with some weird tidbits.
Godot and Bevy have a lot going for them:
* They provide a basic structure (nodes in godot, ECS in bevy)
* They provide some built in objects/libraries that you have the _choice_ to utilize
* They don't bog down your entire system when you are just making a simple toy.
I really want to learn Unreal, but after spending 2 nights trying to get direct control over cursor locking I decided the other fancy features weren't worth destroying my iteration time.
In the past LibGDX and Game Maker 8 were how I started programming in the first place. They feel pretty dated now though.
Visual tools have their own quirks and special workflows that creators got to delve into for quite a long time in order to master. Code on the other way is a universal interface that coders already feel confident with and lets ideas turn into playab le prototypes real quick.
That said: phaserjs, love2d, pixi.js
But can the average person with 10 years of web dev experience finish a sellable game within a year doing code-first?
Unity plan pricing and packaging updates - https://news.ycombinator.com/item?id=37481344 - Sept 2023 (272 comments)
https://gist.github.com/raysan5/909dc6cf33ed40223eb0dfe625c0...
My personal preferences:
Godot
OpenBOR (I really like these beat'em up stuff)
Gamemaker 1.4
Unreal seems to produce the most impressive visuals.
Bevy seems to make it easy to produce WebGL. As a hobbyist and a Rustacean, this is my choice. But I have no aim to release anything commercial with a timeline. I’d go Unreal then.
The engine that has been instrumental to it is DragonRuby Game Toolkit [0]. It was easy to set up and use. And I developed a simple game [1] following a really accessible book [2].
It uses Ruby which is an easy language to pick up- especially as someone who is experienced in Python.
I really loved the experience and talked with the author to turn the book into a livecoding tutorial series.
[0]: https://dragonruby.org/toolkit/game
[1]: https://github.com/ritog/dragon-game
[1a]: https://ritog.itch.io/silly-dragon-target-game
[2]: https://dragonridersunite.itch.io/dragonruby-book
So did I cleverly use all these freely built components to make the greatest XR experience ever? Nah, I decided to re-invent the wheel and do everything myself :). I do enjoy the challenge and freedom to build my own ultimate tool though and I know it's not likely others will use it, but I did release it under GPL3 anyways. It has a wiki (slightly outdated atm :)), starter script and some examples too.
It's pretty fun as both a learning/practice experience and actually be able to use it for various builds. Makes it even easier to prototype XR ideas which I love. I am currently wrapping v0.3 which has added tons of improvements, although it's not exactly ready for release, if anyone feels like checking it out I added some links as ref...
* AUXL (A-Frame UX Library) v0.3 repo --> https://github.com/Minty-Crisp/AUXL
* AUXL Grid Collision/No Clip v0.3 Demo -> https://auxlibrary.com/
* AUXL Physics & Powers v0.3 Demo -> https://minty-crisp.github.io/AUXL/physics.html
* Minty's XRcade (mini game arcade built with AUXL) -> https://mintyxr.com
I've also been a bit frustrated by the lack of updates or communication from the team in general (and their develop Github branch hasn't had any activity since May except for a flurry of pull request merges 2 days ago) that I'm considering looking into how hard it would be to port my game to Godot (was considering Unity before the announcement today).
I'm suspecting hard enough I might want to just stick with Monogame until this game is out there, though, and maybe make my next game using Godot.
I also like Phaser.js for web games.
If I can devote some time, I'd love to check out Defold. I've heard good things about it.
A cursory youtube search brought up this random video with a wide variety of examples: https://www.youtube.com/watch?v=UjnOdDzSxWs
There are also several versions of random level generators for Doom engine maps out there: https://obsidian-level-maker.github.io/
I can't remember the names but there are mods and games out there integrating the random level generators.
Of course the maps are still BSP and therefore somewhat constrained but the level of creativity in the mod community is pretty amazing.
[0] https://libgdx.com/