If I was able to time travel to 10 years ago and show myself a RUclips gamedev video, this would be the one. So much information packed in here, absolutely fantastic work.
Unity has one big killer feature that no custom engine could ever replicate it can upload to VRChat that's it that's the only reason why I use it GREAT VID 🎉I enjoyed it so much!
Great vid. I feel very similarly about pre-built engines. Decent general-purpose tooling which will get something on the screen quicker, but afterwards you're mostly generating content anyways. Create your own workflow! Embrace true efficiency customized to yourself.
Great video, Cass! I lolled at "laptops people pull out of ancient Egyptian tombs." I think you summed up the right attitude for performance in games. Sometimes when I read discussions of performance that are divorced from larger projects or practical goals I start to wonder what it's all for and what some of this optimization work could be going toward instead (better game feel, fixing bugs, adding features, etc.). And yeah, msbuild is definitely the worst part of C#.
Very nice. I don't think you can be an elitist without actually condescending to others. Whatever suits someone's needs to get their project done, commercial engine or no, is perfectly valid. I think people getting upset about it is ridiculous - from either side. That's not to say being proud of going on your own is bad, it's a hard thing to do. I know Unity has its controversies, though using Godot and GameMaker I haven't personally felt snake-oiled (if that is grammatically sensical lol). I can often tell a game is a Unity game, though, which I think is a big negative. I'm nearly finished with a menu-game-ish project in GameMaker and ended up using very little of its features, leaving me wondering how things would be going outside of an engine. But I feel like I'm so far from doing something like this as a programmer, maybe I could've used something lightweight non-proprietary like raylib because an engine is just... a lot, though you've made it make sense. To be realistic though, I'm doing fairly complicated art and doing it alone (and music) while simultaneously learning this kind of thing is, again, a lot for me and probably most people like me. What you said about your own tools precisely fitting your own workflow is certainly a big appeal.
Really nice video, thanks for focusing on the work rather than pure entertainment. I used to do Godot Game Dev professionally for a few years, but got kind of sick of all the magic, black boxes and lack of control. I'm currently trying out "engine-less" game dev in Odin + Raylib, so I'm interested in why starting out with an Entity Component System gets you faster to where you want to be. Wouldn't that be a minimal engine game loop, as opposed to a framework which gives you full control on the execution? Why not have plain old structured programming with a game loop that takes input, process ball, paddle movement & animation , and render it, instead of trying to fit that in some ordered Systems. ( I might be really wrong in my understanding, sorry if that is the case)
I would say that my approach isn't faster starting out, per se, but it helps me maintain my pace better than less structured programming methods. Game development is a marathon, not a sprint; I find having some kind of disciplined method for organizing my code helps dampen the impact of the whole "the last 20% of the game takes 80% of development time" thing.
Hey I'm trying to start real game development, what would be the best platform/game-engine to start on (2d games)? Only experience I really have is roblox studio and Luau along with a tiny bit of python and java knowledge from school, but all of which is on engines that do a lot of the work for you in terms of setting up the world/development space. Also you mentioned how you learn new languages pretty well, is there a strategy like reading documentation first or just jumping in, looking up some of the basics and then only looking up things that would be necessary for (a) certain project(s) until you get a good grasp of the language? Finally I was wondering how long it should take to learn a new language (understanding a majority of functions) given that recommended languages to learn before said language have already been learned (I hope what you understand what I'm asking)? And is there a good order of languages to learn?
Really recommend checking out love2d! It uses lua so hopefully it'll be familiar to you and it's really easy to get started with. As far as learning new languages goes, I usually do Project Euler or Advent of Code puzzles in new languages until I feel like I get it.
This is super inspirational and I hope it encourages more people to experiment outside of the common engines!
If I was able to time travel to 10 years ago and show myself a RUclips gamedev video, this would be the one. So much information packed in here, absolutely fantastic work.
THIS ONLY HAS 311 VIEWS?!? WHY? I thought I'd stubbled across another popular Sebastian Lague-a-like channel
this is rly well made for such a small channel
Definitely going to try incorporating the collision Relation into my own project, looks very clean. Great vid!
Unity has one big killer feature that no custom engine could ever replicate
it can upload to VRChat
that's it
that's the only reason why I use it
GREAT VID 🎉I enjoyed it so much!
Great vid. I feel very similarly about pre-built engines. Decent general-purpose tooling which will get something on the screen quicker, but afterwards you're mostly generating content anyways.
Create your own workflow! Embrace true efficiency customized to yourself.
Great video, Cass!
I lolled at "laptops people pull out of ancient Egyptian tombs." I think you summed up the right attitude for performance in games. Sometimes when I read discussions of performance that are divorced from larger projects or practical goals I start to wonder what it's all for and what some of this optimization work could be going toward instead (better game feel, fixing bugs, adding features, etc.).
And yeah, msbuild is definitely the worst part of C#.
A confession: I stole the ancient egyptian tombs line from Evan.
Very nice. I don't think you can be an elitist without actually condescending to others. Whatever suits someone's needs to get their project done, commercial engine or no, is perfectly valid. I think people getting upset about it is ridiculous - from either side. That's not to say being proud of going on your own is bad, it's a hard thing to do. I know Unity has its controversies, though using Godot and GameMaker I haven't personally felt snake-oiled (if that is grammatically sensical lol). I can often tell a game is a Unity game, though, which I think is a big negative.
I'm nearly finished with a menu-game-ish project in GameMaker and ended up using very little of its features, leaving me wondering how things would be going outside of an engine. But I feel like I'm so far from doing something like this as a programmer, maybe I could've used something lightweight non-proprietary like raylib because an engine is just... a lot, though you've made it make sense. To be realistic though, I'm doing fairly complicated art and doing it alone (and music) while simultaneously learning this kind of thing is, again, a lot for me and probably most people like me. What you said about your own tools precisely fitting your own workflow is certainly a big appeal.
this video radicalized me
Really nice video, thanks for focusing on the work rather than pure entertainment.
I used to do Godot Game Dev professionally for a few years, but got kind of sick of all the magic, black boxes and lack of control.
I'm currently trying out "engine-less" game dev in Odin + Raylib, so I'm interested in why starting out with an Entity Component System gets you faster to where you want to be.
Wouldn't that be a minimal engine game loop, as opposed to a framework which gives you full control on the execution?
Why not have plain old structured programming with a game loop that takes input, process ball, paddle movement & animation , and render it, instead of trying to fit that in some ordered Systems.
( I might be really wrong in my understanding, sorry if that is the case)
I would say that my approach isn't faster starting out, per se, but it helps me maintain my pace better than less structured programming methods. Game development is a marathon, not a sprint; I find having some kind of disciplined method for organizing my code helps dampen the impact of the whole "the last 20% of the game takes 80% of development time" thing.
Hey I'm trying to start real game development, what would be the best platform/game-engine to start on (2d games)? Only experience I really have is roblox studio and Luau along with a tiny bit of python and java knowledge from school, but all of which is on engines that do a lot of the work for you in terms of setting up the world/development space. Also you mentioned how you learn new languages pretty well, is there a strategy like reading documentation first or just jumping in, looking up some of the basics and then only looking up things that would be necessary for (a) certain project(s) until you get a good grasp of the language? Finally I was wondering how long it should take to learn a new language (understanding a majority of functions) given that recommended languages to learn before said language have already been learned (I hope what you understand what I'm asking)? And is there a good order of languages to learn?
Really recommend checking out love2d! It uses lua so hopefully it'll be familiar to you and it's really easy to get started with. As far as learning new languages goes, I usually do Project Euler or Advent of Code puzzles in new languages until I feel like I get it.