The reason I stopped making games... The tooling! It takes way too long to open Unreal Engine, or to do anything within its UI, and Unity gives me "bad vibes" (technical term) any time I open it. I know these days there are other engines, but I started my programming journey by learning game engines, and perhaps I should have learned programming properly before jumping into game programming.
You should try Godot. Open source and extremely lightweight compared to Unity and Unreal. It's pleasure to use, none of the baggage of the other engines. I'm a patron contributor. Fantastic for hobbyist tinkering and also viable for full blown commercial games.
Yeah Godot editor is snappy and smooth, unlike Unity and Unreal.
However Godot's 3D and physics implementation leaves a lot to be desired. Not sure if Godot 4 solves those issues.
When I was moving away from Unity I looked at Godot and really liked it. It's nowhere near as performant or feature rich as other engines and it's future is undecided but it was fun to tinker with and get things moving. I hope to return to it in the future.
The lack of easy testing made me not use it. There were a couple testing plugins users had made which could kind of do unit and integration testing but they were super finicky.
3.4~ as they were preparing early 4.0 releases was when I was experimenting with godot.
I should add that I was specifically trying to use c# ecosystem for godot at the time and their plugins for testing. Since I ended up with Unreal I'm refreshed enough in c++ that I might check Godot 4.0 out this weekend again.
Unity had unreal are both excellent million $$ packages you can use for free(ish) that come with integrated editors and asset pipelines, massive plugin/package ecosystems and even one click playable templates.
What more do you want?
It takes too long to open the editor? Cry me a river! What do you want, a lean editor / hit code reloading / functional code only / truely free? Look at bevy or Godot.
I mean come on, this is the golden age of indie development for games, where literally anyone can make a game in like a days effort following a YouTube tutorial.
The tooling is bad. Pppohhfhfhfhfhttt.
This is like some saying they don’t use the internet because the speeds are too slow when they don’t have a free optic fibre literally from the provider to their door step.
The tooling is great, don't get me wrong. But as soon as you need to do anything that slightly deviates the norm in Unreal engine, you're constantly butted up against broken features, or terrible performance. I can't count the number of times we had to fix a feature or some silly performance problem that was broken in the engine itself.
The latest of which was Iris (to be fair, it is in an early state, but it is a major improvement to ue4's replication), there are some baffling design choices that deliberately made it way harder to make custom serializers for networking.
It's not all sunshine and rainbows. It mostly depends on if you walk the well trodden path or not.
> I mean come on, this is the golden age of indie development for games, where literally anyone can make a game in like a days effort following a YouTube tutorial.
The tooling is a rabbit hole. Unity gives me the creeps and unreal (afaik) still doesn't work on Linux. However:
Löve is super fun, lua is cool but at some point it gets tiring to get more advanced things working. For 'simple' things you can't go wrong.
Haxe is amazing (I learned flash after it was dead o.O) several different frameworks to choose from, many plugins kinda dead but still big community. Compiled to everything you need, some Frameworks even have console targets.
However this coming from someone who hasn't released a real game yet.
This brought some flashbacks : I worked with a Linux developer on a small binaural game prototype over 8 weeks. He spent those eight weeks comparing tools and beginning to hash something out in assembly which he didn't finish. We made a custom C# wrapper for the spatial sound in " Unity " instead.
Unity used to provide a UnityHub.AppImage (used for licensing and version management) that with enough duct tape worked on non-ubuntu distros. Now they maintain their apt repo to ship UnityHub and anything other than Ubuntu 20.04 is not viable for getting things done.
Unreal just provides the source code and you compile it yourself, and there are community solutions for different distros.
Does Unreal actually work “fine” on M1 Silicon? I very recently installed the latest version on my Mac Mini and it seemed to perform pretty poorly, even with a mostly empty scene.
3D is hard. Large studios have gotten over the growing pains, but there's still no equivalent of "Flash" in terms of those tight, fast iteration times that made flash games explode.
Same, I'm waiting for someone to make a game engine for programmers. Most seem aimed at designers. I don't want low code no code tools... I want to get into code asap.
I had the most fun on this using löve (love2d), it is pretty sweet. Just a text editor of your choice and you are good to go - you can also connect it to an ide for debugging if needed.
I am not a super fan of C# but if you are more into that there is also Monogame/FNA, but I haven't used those - I did use XNA back in the day and they seem very similar.
" Unity " and C# is a better way to bridge a prior understanding of programming with making games. You just need to get over it somehow. There's a lot of great games made with each engine, and all are slow to start up but you're not there for that. You're there for the 80% those engines do so you can spend more time on making the parts that are fun. If you don't start - you'll never make it.
Yeah, honestly too many people try to pick up something like a game engine or a web framework without knowing the underlying language. It's a recipe for disaster.
The game engine of Blender faced limited support due to its requirement for games to adhere to Blender's license (GPL) when generating platform binaries. Although there were workarounds available, such as externalizing blend files or using the binary as a loader, these solutions were not practical for most commercial games. It would have posed a significant challenge, particularly if developers intended to port their games to platforms that did not permit such restrictions. Consequently, this limitation likely deterred widespread adoption of the feature, with its usage being primarily limited to prototyping purposes.
When I was 13-14 there was a brief period where I was making levels for Jedi Knight: Dark Forces 2. I had a “clan” encouraging me to make us our own level and it was simple enough that at 13 I could listen to some Disturbed on my headphone while learning how to use the tooling and follow tutorials to make stuff and get things up and running. It had a basic scripting language (cog files) that you’d use to make elevators move and things like that. You’d attach these cog files to in game objects and it was a very good intro to games and level design. I think that was definitely my foot in the door for thinking like a coder.
Unfortunately Unreal Tournament became extremely popular right after this and thirteen year old me couldn’t wrap his head around C++, and didn’t have anyone to help me, so my learning journey ended for about 12 years or so until I learned to code in my mid 20s. I’ve still never picked up video game design again, despite consuming them a decent amount. Stuff just got too complicated.
Battle.net also used to be great for learning your way around computers. Designed and released my first "game" via Starcraft's Map Editor, first "server" was a Battle.net chat bot hosted in my closet... First peek into how complex software works was map hacking Diablo 2... etc etc.
I have such fond memories of that game, specifically the single player. It was the first time I ever felt truly engrossed by a game storyline. I even loved the FMVs which felt like the perfect blend of CG and live action back then - I later learned that the saber duels were the first captured on film since Return of the Jedi.
Such a shame that it was passed up for porting to other platforms. Would love to play it on my Switch like I can the followup.
I was also fairly obsessed with modding this game, which is strange because I'm not really a massive Star Wars fan.
The cog language (more or less C syntax) was far more powerful than the base game really made use of. I think the grappling hook mods really got me started though, the engine wasn't super modern but few games really sold the vertigo of wide open cliffs and pits like this one did. I like tinkering with it far more than I did playing it. Cog seemed almost bolted on and the base cogs really only were sort of complicated for weapon handling and scripted level events. But I remember making laser trip mines (and the ability to shoot the mines with a blaster), a crude rope for the grappling hook, reloading weapons, etc. By the time I moved on I was implementing A* pathfinding for a replacement AI written in cog. I'd switched to Mysteries of the Sith just to get access to more cog verbs.
Dark Forces 2, cog files! I haven't thought of any of this in years!
We're very similar, it was my introduction to game editing - I was never a fan of DOOM, Quake or Unreal/Tournament. But, I played around so much with JKEdit, and even continued with Jedi Outcast. It was by far the most fun I had with modding a game until Morrowind came out years later.
I really wish that modern games were as accessible, at least at a surface level - modding has fallen to the sidelines in recent years - I can't even think of the last game that came with an editor outside of a Bethesda game.
I developed/published mobile games for awhile during and immediately after college. I think my games saw somewhere in the ballpark of 500,000 downloads and I made "OK" income from them for 1 - 2 years (roughly on par with a minimum wage job in the US).
I briefly considered going full-time into solo game dev. But once I got my first "real" job offer after college, I quickly let go of that idea.
Many years later, as a SaaS founder now, I have a strong itch to get back into game development. One of my dream goals is to some day exit my company and then comfortably/leisurely "retire" into a career as a solo game dev.
That is definitely not the world's worst plan. I've started building a tiny, rubbish Pong clone in PixiJS for my son to play with me, and it is pretty intoxicating.
Its the " unknown - unknowns " that'll get you every time. From somebody who has an interest in game design , development and worked on a few released titles. Those games just don't look to be any fun ? Its a mistake that a lot of independent developers make : trying to make games , that have the semblance of classic games. Part of this comes from the learning process as well - it takes a lot more to create 3d content , rig it , animate it so on.
Though my best advice is to look within yourself , and your own experience for the situations and narratives that resonate. Look at somebody like " Lucas Pope " who made " Papers Please " " Return of the Obra Dinn " after leaving a full time developer position at " Naughty Dog ". It seems Austin works with AI - assisted code , why not turn that into a game ?
The world is already full of people trying to emulate Tarantino and Miyamoto. Be your own thing , be the next thing and your situation - process is all part of that. Just need to change the mindset a bit and turn your limitations into a different kind of expression. Then there's a lot of hard work !
Interesting how you advice is to not to emulate classic games while the author came to the opposite conclusion:
>Chasing novelty for the sake of novelty is silly. Gameplay mechanics often don't work out the way you plan in your mind. I find that bottom-up design is much more effective.
But I guess that's the thing about game design. There is no one true blue answer, and it ultimately comes down to how you approach problems in general. I'm guessing some are just really good at deeply analyzing a game and polishing up their own variation of that game, while others are better off learning concepts and then drawing their own experiences together rather than following a tight template.
(Note: this ofc can vary even in the same person, from genre to genre. I'm sure some fighting game experts can design their own fighting game down to the number of frames per attack per character. but also can't figure out a basic game loop for an FPS.)
I created a handful of games at university, each taking a few months, and I'd say the biggest lesson was how little programming was involved. Programming new functionality would only take a few hours, but then I had to spend days on creating models, music, map design, environment, resources, menus, HUD..
Granted, we were tied by not being allowed to purchase assets due to copyright but even finding useable free assets took most of the dev time. I'm curious how AI will improve this but it's definitely something for programmers to have in mind when starting out. It looks so easy in tutorials/courses when they have an asset pack ready to go and can build the core game mechanics in a few hours. The reality is a little different.
I do little game programming on the side. I don't program that much anymore on my day job so these little 1-2 kLOC projects with a quick feedback loop are quite enjoyable.
Nothing complicated, really. Wolf3d-style, puzzles, etc. Because these are mostly made to have some computer-related fun together with my daughter, they don't really need that boring kind of polish a real game requires. So it's pure fun.
After making a few stupid 3d game prototypes I've come to a conclusion that pure libSDL tinkering gets old really fast so I wanted an engine, and Quake 1 suits my purpose perfectly: scriptable through a simple language (QuakeC), a bunch of level editors available(Trenchbroom is a dream), as well as a small and welcoming community. And this is probably the last broadly used 3d engine that a semi-decent programmer can understand at all levels.
I even made an Emacs mode for QuakeC editing. This is part of the fun!
I work in professional spaces as an engine/graphics programmer for some 7 years now. But I've also been doing stuff on the side. Nothing serious tho, so no links to show you.
Thing is, the kind of game I want requires some competent art and at this point I think it's better for me to learn that. So while (very very slowly) planning my game and learning knowledge professional, I want to do some classes in art for a few years, and hopefully by that time I can start to make a game that will truly wow people. The kind that makes you say "wow, this was made by 1 person?". That's my goal. Don't care if it takes 2 months or 10 years, but I feel like I already lost so much time.
(note: I'm not opposed to help. Far from it. But I have nothing to offer and all my friends aren't really doing that kinda thing. I imagine the art I use may not even be final art. Just enough to show I'm serious and maybe attract an artist who actually knows what they are doing).
Unfortunately, that is indeed the case. Think action adventure game with a moderately sized over-world and a few dozen types of enemies (I DON'T want it to be open world, but from a technical POV it will probably have similar challenges).
I certainly can greybox some of this, but a lot of the feel will depend heavily on some good VFX, and likely some minor environmental deformation. I definitely can't put all that off into late development.
I will inevitably have to pay for a few artists, but Ideally I'd wouldn't be asking for concepting more so than polish and giving a cohesive direction to it all.
I did for a bit, but it's a bit tricky, in the sense that differently from other types of products, games have a very strong separation between the artistic and the engineering aspects. This has very significant consequences.
If one works alone on a game, they may find that they keep solving technical problems, without actually producing a game, but rather, producing an engine. If one works in a team, and nobody takes a creative lead, the same thing happens - technical advancements in the game, but no content.
In the right conditions (either a balanced team, or a single dev who enjoys both aspects), game development is fantastic - from the engineering perspective, game development is a very challenging (in an interesting way) field. But I think it's not easy to find the right context.
It was, in fact, a childhood dream of mine to write a game like this. However, the little GW-BASIC programming I knew and the very limited access to computers I had during my childhood was insufficient to write anything more sophisticated than simple text-based adventure programs. This game is written using plain HTML, CSS, and JavaScript. It uses HTML5 Canvas for graphics and Web Audio for sound. Although 25 years too late, writing this little game fulfilled that childhood dream of mine!
Worked in AAA, left to keep it a hobby. Great experience, but not the career for me.
I'm very excited for AI enabling people with cool ideas to achieve them much faster - game dev is far too technically labor intensive for a lot of creative types, it'll be fun to see how AI-enabled tooling allows more creators to realize their visions.
At a stretch, I'm hoping it'll produce the same explosion of individuated expression as widespread literacy did with the novel.
Yeah. And no, I'm not linking them for anyone for two reasons:
-I value the tiny shred of anonymity I have here so I can continue to express stupid opinions without being too self conscious about it.
-Honestly they just aren't very good, at least not in any traditional sense. I find that I am not passionate enough about most of my ideas to expend the time that seems necessary to flesh them out into a coherent and worthwhile experience.
I've tried to make games, but what I ended up was making game engines because that was more appealing to me. My penchant for huge overambitious projects is good sometimes, but not always.
I do, but only as a hobby. I could have taken it as a career choice but the fun is in pursuing my own ideas, designing my own characters, design the cinematics, the dialogue, their equipment and how they use it...and of course the code and visuals to make it happen. The idea of working on someone else's design seems atrocious.
I've had short stints but I overthink / over-engineer and lose interest very quickly, lol.
A good platform that forces you to think small is pico-8, it's got all the tooling built in but it's on a very small screen and you're limited to something like 8K tokens of source code. But some of the games they've built in it are great, like Celeste.
I've tried a platformer (from a tutorial) and a Zelda-like game, but got stuck at some point so I should make something even smaller / simpler. Maybe that tunnel game mentioned in the OP, I used to love games like that. Oh, and the first game I made from a tutorial was like the 'helicopter game', single button, randomly generated levels, that kinda thing. Can be cobbled together in a matter of hours.
On the side using Unity as a family activity.
The kids love the official demo projects, and I'm slowly teaching them code.
My dreams of opening a wife and husband studio are dead dead dead after seeing all the failed hopes and dreams of indie developers on reddit, discord, etc.
Yeah, unless you get some good backing out of the gate it’s only something you can do on the side. And maybe hope for a banger like Project Zomboid time to shine.
Yup, for fun. Building a game from scratch almost from from the bottom. Starting with a CPU, sound-chip and graphics-chip (using an FPGA). On top of that a Forth system and on top of that a platform game. Lots of fun and a great learning experience.
Yes – it's both the most fun and frustrating hobby. :) Nothing gets me in that focus zone like working on a level or cl
oding a new gameplay system. I think it's the thrill of tweaking, testing, tweaking, over and over again, which eventually produces a very satisfying product.
I'm currently working on a Godot project (highly recommend the engine for hobbyists).
Thank you. I am making a new sprite scaler adventure in the same sense, but it's a bit bigger area with more things to explore and all, I hope to have at least something of it released this Sunday or maybe in the next week.
Absolutely, currently remaking and old arcade game (SAMI) with a team that has skills beyond engineering, because engineering and a bit of guitar is about all I have time for :)
It's a great way to learn a lot, and it has definitely helped me with how I approach problem solving and software design in my day job.
I do agree with this sentence on the article <<The meta stuff takes as much effort, if not more, than the core game. I mean the menus, instructions, release materials, API integrations for advertisements and high scoreboards, and so on.>> as the menu you can see in my game (which is very unpolished) took as much as the game itself.
I tried to when I was at school, but there was never enough time. In retrospect I should have severely scaled back the game ideas. Also, this was the DOS era, so getting better graphics/audio was always a preoccupation. But then who would be the artist, etc. Also, most of this happened before I had an internet connection, so not much to work with.
That's why the game engines of today are really awesome.
Certainly sounds interesting, I use to play no games except for Nexuswars.com, a browser based game with some d&d mechanics. The 15 minutes a day keeps it very controlled and manageable (unlike games that want to keep you in)
