[–]▶ No.955083>>955147 >>955152 >>955200 >>955252 >>955284 >>955285 >>955512 >>956073 >>956198 >>956397 [Watch Thread][Show All Posts]
I'm so fucking tired of every engine I try is not perfect.
>unity and unreal
BLOAT to the max propietary garbage, literally spyware.
>gamemaker or rpg maker
literally toys and propietary.
>godot
can't even use the latest version on my toaster nor I can get the third party modules to work.
>libgdx
can't even get working the wizard on my linux
>love2D
trash language made by thirld worlders.
Every fucking engine is garbage, bloated or using third party meme shit like Net framework or Oracle garbage.
And I'm sick of doing something and hoping there's not some magical bug in the engine.
I just wanna make games but every fucking retards keep telling me that raw Opengl and SDL are just impossible to use and that I shouldn't make my own engine and that is hard as fuck.
At this point I'm so frustrated by this.
▶ No.955085>>955087 >>955674
What exactly do you need the engine for? Physics equations are well known, including optics. The API's are well documented. Do enginges really need to provide anything more than handling physics and passing to the hardware? And if that's all they do, why can't you do that yourself with less?
▶ No.955087>>955089 >>955091 >>955093 >>955099 >>955147 >>955676
>>955085
because everyone in the game industry now keeps telling me that using raw opengl and SDL is impossible and is a waste of time.
▶ No.955089>>955098
>>955087
If you want to waste your own time doing raw OpenGL and SDL, please go ahead. The general purpose engines exist for people who are interested in game development who aren't too concerned about engine development. Remember that game engines solve a significant set of game problems and that's why they exist.
▶ No.955091>>955098
>>955087
Echoing what the other anon told you, the reason for this advice is that in the gaming industry time is a premium. You need to finish development fast and move on to the next game ASAP. There's not much money to be maid in reinventing the wheel.
That said, if you don't have a commercial interest, and graphics programming is your main focus, then by all means create your own engine for the learning experience. Basic linear algebra and knowledge of trig will get you quite far. As you get more advanced, you'll need to know more about numerical computation.
▶ No.955093
>>955087
>everyone is telling me that they need a job so you can't do it yourself
Come on anon.
▶ No.955096>>955098
>godot
Then use the older version as it's still supported and getting updates.
And we can't help you much if you don't tell us what kind of game you're trying to make.
▶ No.955097
>leddit spacing """game dev"""
▶ No.955098>>955120
>>955089
>>955091
>>955096
These are guys who think that once you develop your own personal engine you can't use it for any other projects and that's why you need to use their engines that they are paid to shill for.
▶ No.955099>>955100
>>955087
It's certainly not impossible, but if what you want to do is sufficiently complex, you'll probably end up creating a game engine on top of OpenGL and SDL anyway. The question is whether that's the best use of your time, and whether you can create something superior to existing game engines.
Also, you're complaining about bloat, but all but the simplest games demand complex behavior from the computer. Coaxing complex behavior out of the machine requires a lot of lines of code, at some level. Either you're writing all of those lines of code, or you're leveraging the work of someone who has.
What is it you want to do, anyway?
▶ No.955100>>955108 >>955165
>>955099
I don't know if games really require the accuracy of integral and derivative calculus. Shit, Carmack used an entirely inefficient equation to calculate light refractions but it didn't have any effect on gameplay.
▶ No.955108>>955122
>>955100
They don't
The physics can be all kinds of wrong but as long as they "feel good" nobody will complain.
▶ No.955120>>955122
>>955098
That's a strawman argument, nobody says that you cannot repurpose your purpose built game engine.
▶ No.955122
>>955108
>The physics can be all kinds of wrong but as long as they "feel good" nobody will complain.
No shit. Hence the Carmack reference for Quake. The only problem is when a bullet should land a hit and the math doesn't add up to it, or definitely when the bullet lands a hit and the real math says it shouldn't.
>>955120
Yes they do.
▶ No.955147
>>955083 (OP)
>>955087
>making your own engine is hard
You're taking advice from the same people who use shit like Unity. It's only a problem if you want muh AAA grafix feel or advanced physics.
▶ No.955152
>>955083 (OP)
>literally toys and propietary
If you exclude the proprietary part (and I'll bet you're not using a Free as in Freedom environment anyway), they can still be used with interesting results.
Take Hotline Miami as an example. It was made with GameMaker and sold roughly 500k copies [citation needed]. It made millions.
▶ No.955154>>955231 >>955281 >>955478 >>955479 >>955493 >>956198
Dont waste your time listening to anyone who says you can't make an engine. It's more time but its totally doable. Don't worry about people saying that making your own engine is "hard". It is only "hard" because those people are "soft" and don't want to challenge themselves. Yes it is a hard task to do but you will also become a hard programmer in the process. I am also in the /agdg/ thread and saw your original post here: >>>/v/15291975
I also have the problem with newer engines not supporting older computers, for example most engine require OpenGL 3.3 and my laptop (T61) has OpenGL 2.0. So since sometimes I only have my laptop for weeks at a time, it's not reasonable for me to use modern bloated engines. I also don't agree with the way they work. So I write my own. Just go for it, ask questions on /agdg/ threads and maybe some people can help, but a lot of it is self-learning.
The only thing that you might notice about this path is that you might enjoy making engines more than the games themselves. Nobody (that I know) thinks they want to make a game engine first. But they incidentally make a game engine to make a game, and find it more interesting to work on.
You might also notice that on /agdg/ most of the people who finished a game also did an "unncessary" amount of work. For example speebot which is made on its own engine / map editor, hellbreaker which has it's own map editor, red sky which has it's own engine / map editor. Yet the vast majority people who have an advantage like the unity engine can't produce a finished game, giving the engine programmers / non-engine programmers a 50/50 split in terms of finished games. What this is saying is that, the people who are capable of finishing a game are so productive, that even though they don't have a premade engine, they still outproduce developers who are using a premade engine. People who have the mindset to complete an engine are that much more productive than people who have a massive advantage over them.
▶ No.955164>>955170 >>955174
If you aren't capable of using an engine designed for people who barely understand programming like Unity you have absolutely zero chance of writing a decent engine yourself. Don't let nodevs here convince you you're ready for something way above your skill level. Focus. Complete a project, ignore technology.
▶ No.955165
>>955100
>Carmack used an entirely inefficient equation to calculate light refractions but it didn't have any effect on gameplay.
Carmack used an incredibly good and fast approximation to do that, he didn't just throw in any random equation.
You need to know math and computers well to come up with that kind of trick.
More in general, you need derivatives and integrals for an incredible amount of things, most notably you can use them to avoid having fast objects clipping through other objects.
▶ No.955170>>955172 >>955173
>>955164
I have no idea how to use Unity besides copypasting tutorials, had trouble finding my bearings in Gamemaker, and felt that Godot was clunky and awkward.
Making my engine in comparison is much easier because I actually understand the platform I'm working on and can make it work however I want.
>Focus. Complete a project, ignore technology.
Sounds like something a marketing team came up with. You can't complete a project without understanding the technology.
▶ No.955172>>955178
>>955170
>I have no idea how to use Unity besides copypasting tutorials
Try following the tutorials and understanding what they do at each step, instead.
You need to learn, not to produce a working prototype immediately: copypaste is only good for the latter.
>Making my engine in comparison is much easier because I actually understand the platform I'm working on and can make it work however I want.
That's not how any of this works, dumbass.
Code doesn't just work "however I want", bugs will happen and good luck fixing those if you can't even deal with Unity.
>You can't complete a project without understanding the technology.
What is abstraction.
▶ No.955173>>955178
>>955170
I eagerly await the nothing that you will produce.
>You can't complete a project without understanding the technology.
Explain Undertale.
You can fuck around pretending to be Carmack all you want, but all you actually have to do is make something fun.
▶ No.955174>>955175
>>955164
Okay smartass.
Try programming Ladder, after all, it was designed with nocoders in mind.
▶ No.955175>>955176
>>955174
No idea what that is. I'm actually a professional game developer by the way.
▶ No.955176>>955182
▶ No.955178>>955179 >>955183 >>955184 >>955289
>>955172
>Code doesn't just work "however I want", bugs will happen and good luck fixing those if you can't even deal with Unity
I'm getting the feeling that you don't actually program at all. Code in fact works EXACTLY how I want, it's not a person or a living entity, it's a bunch of instructions that I gave the computer. Finding and fixing bugs is very easy when I understand all of the software. Meanwhile if you get a bug you'll be googling for Unity answers. You'll also be googling for Unity answers when you want to use a part of the engine in a way you haven't before, while for me it's only a matter of doing it.
>>955173
>sales/""fun"" > good software
Give me 1 reason to continue talking to you.
▶ No.955179>>955181
>>955178
>I'm getting the feeling that you don't actually program at all.
A bold move, coming from the "halp I can't figure out Unity" guy.
>Code in fact works EXACTLY how I want
No it doesn't, bugs exist for a reason.
>it's not a person or a living entity, it's a bunch of instructions that I gave the computer.
And that is exactly why it doesn't always work as you want: code has no knowledge of what you want, so if you write something that doesn't actually mean what you think it means you will not be warned.
In a similar fashion, some code that might not be logically needed might be needed in practice, see object deletion and resource deallocation, and you might forget to write such code.
>Finding and fixing bugs is very easy when I understand all of the software.
You probably don't even understand the full specs of your language of choice, for example do you know how C and Java differ in optimizations for loops that never terminate?
>Meanwhile if you get a bug you'll be googling for Unity answers. You'll also be googling for Unity answers when you want to use a part of the engine in a way you haven't before, while for me it's only a matter of doing it.
Consider the following: if the big team of smart and well paid people that made Unity couldn't avoid that, why do you expect to do better than them?
>Give me 1 reason to continue talking to you.
Not the same anon, but you seriously need to deflate your ego and learn stuff, so refusing feedback will only hurt you.
▶ No.955181>>955187
>>955179
You sound like you're projecting all your insecurities and boogiemen into me.
>coming from the "halp I can't figure out Unity" guy
Using someone else's bloated OOP shit software is not the same skill as programming something from scratch yourself. The latter however is more relevant to programming ability, so I wouldn't talk shit to others about it if I were you.
If you make something yourself, it's going to be significantly easier to understand thoroughly than if you buy some black box that changes every couple months and learn it from tutorials.
>You probably don't even understand the full specs of your language of choice
Nobody understands the 100% full specs of their language, besides you have the same problem even if you use and engine. But if you're using it to script Unity, you probably understand less of it than someone who made an engine with it.
>if the big team of smart and well paid people that made Unity couldn't avoid that, why do you expect to do better than them
So you're telling me I'm better off than smart and well paid people since I don't have this problem?
▶ No.955182
>>955176
Blueprints are the gamedev version of that kind of visual logic. imo it's more effort than it's worth. Unity doesn't have them by default for code, but they have something like it for the animators.
▶ No.955183>>955185
>>955178
You can't have either sales or fun if you never get beyond a git repo displaying a triangle. I know your type, you never accomplish anything.
▶ No.955184>>955185
>>955178
>it's easier to write an engine from scratch than read a manual
▶ No.955185>>955192
>>955183
Where's your game?
>>955184
Who said that?
▶ No.955187>>955193
>>955181
>You sound like you're projecting all your insecurities and boogiemen into me.
Nigger you're the one that started with "I'm getting the feeling that you don't actually program at all.", kill yourself.
>So you're telling me I'm better off than smart and well paid people since I don't have this problem?
I'm saying you are too delusional to notice you have the same problem, if not worse.
As above, since you don't seem interested in anything that might hurt your ego the only solution left is killing yourself.
▶ No.955192
>>955185
The last game I worked on is 95%+ rated on steam with several million sales.
▶ No.955193>>955198
>>955187
>Nigger you're the one that started with "I'm getting the feeling that you don't actually program at all."
That's because you say shit like "good luck fixing bugs if you can't use Unity". It sounds like you haven't done programming beyond scripting in a game engine or using similar frameworks.
>I'm saying you are too delusional to notice you have the same problem, if not worse.
As for this thing, I think there's a communication problem. What I'm saying is that I don't need to google for anything for how to use my engine or make new things. What I believe you're saying is that I do need to google for shit sometimes, which is true if I want to implement a new technology that's too complicated to come up with myself, for example efficient path finding or something.
▶ No.955198
>>955193
>That's because you say shit like "good luck fixing bugs if you can't use Unity"
If you cannot manage to understand simple Unity tutorials (very easy programming task), you stand no chance of finding and fixing bugs on your own in an entire game engine (very hard programming task).
You do not need to have ever heard of Unity to be a god tier programmer, but if you tried and failed at Unity you are not a decent programmer for sure.
It's not a hard concept to understand, so kys.
>What I'm saying is that I don't need to google for anything for how to use my engine or make new things.
> What I believe you're saying is that I do need to google for shit sometimes
Do you serously think devs that use Unity need to google shit at all steps, or even look it up in the docs?
They don't, because they are familiar with their tools.
Your entire point is a retarded tautologism, of course if you create the tools yourself by the time you're done you will be familiar with them.
Thinking that familiarizing yourself with a game engine will take longer than making a new one from scratch is unbelievably stupid, so kys.
▶ No.955200
>>955083 (OP)
write your own in sfml and suffer greatly upon learning its fundamental design flaws once you try writing UI components
▶ No.955226
I ended up accepting that everything sucks, and went with Unity. I don't have the time to save the world, I just want to get my work done. The only alternatives would have been to not do the things I want to do and waste my time on building tools instead, or to do nothing at all.
▶ No.955231
>>955154
So in basic terms
>enginefag
>I need to make an engine or map editor or my own game maker
>oh right I finished this one now, I'm not stopping now that I got results
>oh right I finished the game holy cow
>premadefag
>Stackoverflow [unsolved] I have problem using my bloated engine that I don't know how it works and my problem is it broke
>what do I do now?
>posts/brags on forums about their gamedev setup and occasionally answer newfag questions to "make gamedev feel so pro" but literally haven't finished anything not even the sprites.
▶ No.955233>>955245
nibba juz do sum webgl + html5 + js game so you don't have to go through all the bull. maybe try to focus a lot more on the "app" thing
▶ No.955245
▶ No.955252>>955273
>>955083 (OP)
It's the same with any IDE I have seen, not that all those options don't likely have their use but it always kills my drive to do anything if the setup and config take longer than installing Funtoo.
▶ No.955273>>955282 >>955420
>>955252
IDEs are a total meme. Setting up a simple editor is almost always the better choice.
>b-but muh 10000000 LOC project
IDEs won't save you here either
▶ No.955281>>955291
>>955154
>for example most engine require OpenGL 3.3 and my laptop (T61) has OpenGL 2.0.
I'm currently working on my engine because I want to try a few things related to animations.
And the reason there's a sudden clear cut is the way OpenGL operate between these versions. In the 2.0 it's the classic CPU pushing the geometry and matrices and let the graphic card draw, in 3.3 and above, you setup memory zone containing data on the graphic card, define memory structure and let shaders do the rest of the work.
So supporting both versions require a significant amount of work.
On my end, it took some time to have a general working model loading+animation+instanciation running, but now I can get onto the actual game, then "slap the graphic call on it" I'm sure a lot of things will go wrong.
▶ No.955282>>955368
>>955273
>IDEs are a total meme. Setting up a simple editor is almost always the better choice.
This. Use a programmable text editor and a build system. It's amazing how far just GNU Make alone can take you.
▶ No.955284
>>955083 (OP)
>I want a powerful engine with modern features
<but I want it for freely licensed
<but it cant be bloated
you are an idiot
▶ No.955285
>>955083 (OP)
>very engine I try is not perfect
No shit: the perfect engine is the one that fits your project perfectly and unless you're extremely lucky you'll have to make it yourself.
▶ No.955289>>955294
>>955178
I'm getting the feeling YOU don't program at all because everything you say is shit that sounds good in theory but falls apart the second you actually write shit.
▶ No.955291
>>955281
I support both OpenGL and Vulkan. The OpenGL code is only 700 lines while the vulkan code is 3000. They look exactly the same right now... it's not that hard to support older OpenGL since it's not very many lines to add. Also VBO has been in OpenGL for a long time, for example even the oldest card that I test my engine on, an ATI Radeon Mobility 9000, which only supports OpenGL 1.3, still supports ARB_vertex_buffer_object. But I don't require it still because I don't need to require it.
Really it's not so unreasonable to only require OpenGL 1.1 if you can. But i'm sure you're doing something more complicated than I am. But most things can be done on it, you just have to embrace the limitations.
▶ No.955294
>>955289
>sounds good in theory but falls apart the second you actually write shit
Do give a practical example.
▶ No.955310
▶ No.955317
>Game Maker
You can probably do a lot with that if you write some engine-ish code on top of it to handle stuff it ordinarily wouldn't or not as efficiently.
>RPG Maker
Yeah, don't. This one is for non-devs like English majors and street musicians desperately looking to win the lottery and leave poverty.
That said, another concept to look into is using one of those engines as a prototyping platform then when you've found something you really want to do, do it from scratch.
▶ No.955368
>>955282
>Use a programmable text editor and a build system
I suppose that sounds pretty good and I have seen people do that but so far I haven't found any details about not only converting the editor but also to hook it into a build system
▶ No.955420
>>955273
The vim plugin (there are two, I forget which is the good one) for Visual Studio is great and you get the benefit of VS's ability to find references and refactor. It's the best of both worlds and nothing is even close to its quality in pure open sores. I had several people on the team using it and loving it on the last Unity game I worked on.
▶ No.955478
>>955154
How did you figure out programming an engine in C and where could i learn to do that?
▶ No.955479
>>955154
Also nice job on the sigma engine man it's looking really cool, wish i could do something like that
▶ No.955493
>>955154
bit = (leaf - (byte << 3));
wat. Why not,
bit = leaf & 7;
▶ No.955512>>955519 >>955521 >>955522 >>955526
>>955083 (OP)
>INB4 the entirety of /tech/ descends on me.
What's wrong with proprietary software if it does what you need it to do?
This is like the MP4 vs WEBM issue, where Google wanted an open source option on a format that's coming to the end of it's patent (and will effectively BECOME open source) because "muh sheckles".
Which despite that I really only see WEBM used on image boards be it a Chan or Booru, you download a video from Youtube or most other sites it's in MP4.
I can understand not wanting to get involved with oppressive proprietary software/hardware schemes, but throwing a shitfit over proprietary purely because it's proprietary is stupid.
Practically everything we use came from something that was proprietary at one point though was popular enough to became THE standard.
▶ No.955519
>>955512
what a retard lmao
▶ No.955521>>955524 >>955985
>>955512
I don't remember ever getting anything other than a webm out of youtube.
▶ No.955522>>955985
>>955512
>if it works then there's no need for improvements
This is why we're still using jpg when formats with significantly better compression exist.
▶ No.955524>>955985
>>955521
they serve .mp4, .webm, .3gp and .m4a
▶ No.955526>>955985
>>955512
>What's wrong with proprietary software
▶ No.955674
>>955085
I almost replied with "why do you need an IDE?" to make a point, but then I remembered this is /tech/ and you'd probably tell me I should be coding in notepad.
▶ No.955676>>955716
>>955087
It's not impossible but it definitely is a waste of time.
▶ No.955678
pro tip: don't put effort into making games, nobody else does and they still make lots of money
▶ No.955687
Try ENIGMA, it's made new non-retards
▶ No.955698
>get laughed out of /agdg/ for being a meme spacing retard
>comes to /tech/ for validation
Every time.
▶ No.955716>>955812
>>955676
That's what they said to Terry when he built Temple. Now look at him!
▶ No.955812
>>955716
Nobody has ever said that to Terry. Terry was absolutely clear about what he wanted with TempleOS and he invested many years of his life into making it happen. He authored something unique that wasn't covered by the existing systems.
The OP of this thread is a real shithead because he complains about so-called "bloat" while at the same is too incompetent to understand why people choose to work this way.
▶ No.955985
>>955521
>>955524
So it get's even better, not only do they STILL have MP4 which they made WEBM to replace, they also have yet two more formats.
>>955522
Welcome to the real world people don't drop everything and immediately replace it because there's a version 1.1 that may have qualities the person doesn't like.
>>955526
Answer the question Mr. I've never installed an OS.
▶ No.956048
engines are a meme. games don't even have engines. any that claim to are full of shitty rubberstamp code and bugs they couldn't fix because it would require not treating the engine's shitty "abstractions" as a black box. even if engines had good interfaces it still wouldn't matter because they're all shit and would be full of internal bugs
▶ No.956073>>956221
>>955083 (OP)
>unreal
>BLOAT to the max propietary garbage, literally spyware.
Why is Unreal spyware?
▶ No.956100
>get laughed out of /agdg/ for being a meme spacing retard
>comes to /tech/ for validation
Every time.
▶ No.956198
>>955154
I finished a 2D godot game, just haven't put it in /agdg/
Started another game, 3D in godot. Probably won't post it on /agdg/ either, don't see the point, in fact it's got some negatives associating the finished game with 8ch, but I have gotten some great pointers from there and I do try to give back from time to time, /agdg/ and the game boards here are better than most other communities when it comes to getting advice.
Using godot engine makes me feel like I own my game, couldn't do it in unity feeling like the game wasn't mine. I tried using unity, super convenient, but it was a bit buggy in my linux install and the ui looks like what I imagine the asshole of an anthro iphone would look like.
>>955083 (OP)
Make your own game engine is good advice, it's something I want to do at some point, another solution would be to use an old 3d engine, like doom or eduke32, practice your level design or modelling now as those will be transferrable skills/ideas. Also, if you want to add a story to a game later, write it in twinejs now and it'll be easier to add it to your game later. Or work on your graphic editing skills now, build asset pack for your game first, while you look for the solution that'll work for you, and you'll be ready with assets and design by the time you find/make your engine.
▶ No.956221>>956232
>>956073
Both Unreal and Unity have deeply integrated support for telemetry. Unreal's is literally drag+drop and helps you datamine players and send it to your choice of datamining services. Unity's support goes a step beyond as they host their own datamining service you can subscribe to and run search queries over all the data on your players you've stolen.
Both companies advertise this as a way to do things like identify where in a level your players have walked in circles for 30 minutes, but it's like advertising a hitachi magic wand as a massager - we all know who the intended audience is.
I've worked on a game where they were using Unity's telemetry to capture your browsing history. This is normal now.
▶ No.956232>>956237
>>956221
>installs proprietary software
>gets spyware'd
wow didn't expect that
▶ No.956237>>956241 >>957250
>>956232
I don't think people understand the scale. Almost every program is spying on you now, even open sores like chromium and systemd.
▶ No.956241
>>956237
>muh chromium
>muh systemd
opinion discarded :^)
▶ No.956397
>>955083 (OP)
>unity and unreal
WHY ISN'T THIS COMMERCIAL SOFTWARE USED BY MASSIVE STUDIOS CATERING TO MY SOLO NEET ASS?
>gamemaker or rpg maker
I'M GOING TO COMPLAIN ABOUT THESE BEING TOYS EVEN THOUGH I CAN'T OPERATE ANYTHING MORE COMPLEX THAN GUI WIDGETS
>godot
I'M TOO RETARDED TO SET UP MY DEVELOPMENT ENVIRONMENT
>libgdx
I NEED TO USE A WIZARD TO INSTALL A DEVELOPMENT LIBRARY BECAUSE USING A PACKAGE MANAGER IS FUCKING ROCKET SCIENCE
>love2D
I RATE DEVELOPMENT TOOLS BASED ON HOW COOL THEIR DEVELOPERS ARE
TL;DR I'M A FUCKING IDIOT AND I'M GOING TO KILL MYSELF NOW TO MAKE THE WORLD A BETTER PLACE
▶ No.957250