Conversation

Notices

1. Embed this notice
   Talon (talon@dragonscave.space)'s status on Friday, 15-Mar-2024 01:30:42 JST Talon

   Been working on and off, very slowly, on a game engine that should do most of the things I want an engine to do. One of the tips I've heard and can definitely echo is to work on a game first, and engine second. Make the engine work with your game, not the other way around.
   For the longest time I've been working in a relatively OOP style because that's how I learned, and that's how things made sense. But it always ended in a mess. Whenever I wanted to reuse code I had to untangle it and rewrite stuff or copy a bunch of other stuff I would never need. So i decided I'm going to try an entity component system.
   And like I dunno... Over the last year my brain slowly started rewiring my thought patterns and now it makes sense to me intuitively. You don't even have to be an ECS purist to get massive benefits out of it. Even just going in an ECS direction helps a lot.
   While making a game, you will still end up depending on other entities and systems and what not. Games are complex. You can't stop dependencies. But the difference is that if you do want to break off a bit of code and reuse it, it is much more obvious where to do that.
   Like while developing this thing, I switched the project I'm developing it for (or maybe with?) several times. The game changed. But I was always able to drag most of the code with me into this new game idea and keep iterating on it. And I'm fairly sure this would have been a lot harder if there wasn't this very compartmentalizable idea of components and systems.
   Again I'm not an ECS or die kind of person. Especially if you're working alone, it's preferable to take some shortcuts. But even if you only loosely structure your projects in this way things seem much nicer to just... I don't know... find your way around? Reuse?
   Just like any paradigm, you have to apply it in a way that makes sense to you. I've taken shortcuts. It's not pure ECS either. But even so things make sense, things fit together nicely, and there's no mess that immediately makes me want to tear the whole thing down. Which is more or less the case with literally all of my other games. Which is also why I never update them. Always learning, always growing, it's a pretty nice feeling sometimes.

   • Embed this notice
     Talon (talon@dragonscave.space)'s status on Friday, 15-Mar-2024 01:30:39 JST Talon

     in reply to

     Like for example tools to wysiwyg a world. You could just drop into an engine with an editor and just do that. And make some objects. And attach scripts and what not. You even have other tools to help like modelling software and whatever else. None of that's accessible. So we kinda reinvent the wheel very terribly to make this work. And a lot of the time this wheel gets reinvented per game. Most of the time we don't even wysiwyg a world at all. It's all a big mess.

   • Embed this notice
     Talon (talon@dragonscave.space)'s status on Friday, 15-Mar-2024 01:30:40 JST Talon

     in reply to

     I don't even want to make an engine if I'm completely honest. I want to make games. I've wasted so much time on this. But accessibility concerns... it's always the same story, really.

     GreenSkyOverMe (Monika) repeated this.
   • Embed this notice
     Talon (talon@dragonscave.space)'s status on Friday, 15-Mar-2024 01:30:41 JST Talon

     in reply to

     Maybe this will eventually lead me to actually release a new game. That'd be nice wouldn't it? I'd like that. Unless that Godot accessibility branch gets merged first. Then I'll spend another little while learning a new thing.

   • Embed this notice
     GreenSkyOverMe (Monika) (greenskyoverme@ohai.social)'s status on Friday, 15-Mar-2024 01:33:52 JST GreenSkyOverMe (Monika)

     in reply to

     @talon Huh, never heard of entity component system before ... gotta read about it