What are some Jow Forums approved game engines?

>PixiJS

>his game is so unoriginal that it can fit into an engine
never going to make it

someone please shoot gamebyro in the head and end its misery, please.

>unironically this. universal engines are a dead end. you want to leverage the power and innovation of a truly unique engine tailored exactly to your needs.

Exactly like the car industry, or any industry for that matter. Custom make from the ground up rather than upgrading old parts and iterations. You're unique mechanics or crafting system wont work right in something like Unity.

name one game made with godot

>no programming experience
Game Maker
>programming experience, 2D engine
Game Maker, Godot
>programming experience, 3D engine
Unity, Godot
>from scratch 2D
OpenGL/DX11
>from scratch 3D
DX12
>Shit you shouldn't touch
Unreal, Vulkan

why would you want to use an engine?

all these libraries available are so high level already theyre almost an engine themselves..

you really think something like vulkan or metal or opengl is too "low level"? lol if you cant be arsed to even learn that wtf are you doing designing games..

fmod, sdl, etc, etc so many libraries that already deal with assets and sound and input and whatever..

only tool you need is a designer for levels, blender or something like it are good enough

itch.io/games/made-with-godot
godotengine.org/showcase

All of them unironically look like Flash games.

>the developer making shitty art is the engine's fault

/thread

Godot
Löve

Does ZDoom count? At least when it comes to a retro-style FPS?

Attached: 1427951801792.gif (75x75, 19K)

oops forgot, also in shit you shouldn't touch is Gnm, Gnmx and NVN

Lua is the worst programming language ever created.

I've started learning OpenGL a few months ago so I am not really knowledgeable in graphics programming. Isn't Vulkan supposed to be the next big thing where you can program closer to the hardware and is cross platform like OpenGL?

What is your opinion on CryEngine?

None.
Use SDL or SFML like a white man.

Godot shills are the most horrible fucktards I know when it comes to gamedev.
It's complete crap. The fucking design is just shit. The godot scripting language is horrible.
It also runs like crap, even though the graphics are shit.
It's genuinely terrible.

I made my own 3D engine. Sure it took time, but its amazing.
If you don't make it yourself, or at least glue together open-source parts, then I will disapprove.

>Isn't Vulkan supposed to be the next big thing where you can program closer to the hardware and is cross platform like OpenGL
It is, but its also shit.
Can't be helped that they fucking ruined the only chance we ever had at getting a good graphics api.

none of that matters because making shitty demos in godot takes 1/1000th of the time that it took you to make your special snowflake engine

What's so bad about it?

Why exactly is it shit?

It's not but the added complexity of supposedly low-overhead API is certainly not justified when making an indie game. Even tons of AAA titles regularly ship with both DX11 modes (OpenGL style), and DX12(Vulkan style), with basically identical performance.

You'd have to have thousands of different objects, with different shaders to hit the perf ceiling on even a low-end desktop PC.

Complex, undocumented, buggy, and the company behind it is struggling with bankruptcy.

make games, not game engines

Then you're not doing engineering, you're making craftworks, or artworks depending on what kind of pretension you wish to display.

Attached: marshall-m-fredericks.jpg (338x450, 33K)

Yeah, but making a real game is faster in your own engine, and the product WILL be better.
Not that retards like you godot fuckers would understand.

your shitty minecraft clone isn't a real game and it definitely won't be a real game in 10 years when you finish it

>If you don't make it yourself, or at least glue together open-source parts, then I will disapprove.
How would making your own engine alone be more performant than using an engine like Unity that has an entire building of people working on it?

Stay jelly fucktard.
You low skill losers will never accomplish anything significant in your lives, so you can't imagine that others will.
This is also why you try to get 'help' for your retarded shit-programs with crap like godot, or other premade bullshit.
Because deep down you know, you couldn't achieve it by yourself even if you tried.
The worst part is, at least unity or unreal have some actual advantages, it is possible to make decent games with those engines, while the crap you shill to other faggots on here (godot) is just utter junk.

sounds like you're just mad you couldn't figure out how to hack godot to get better performance

I remember following the example code on the Vulkan repo to render a triangle and it was a thousand lines of shit I didn't understand just to render a fucking triangle. Shit's wack.

Learn about game-programming, and you will know.
Basically its much easier to optimize shit, if you know the use case 100%, which you do know if you make a game.
Unity and Unreal are engines for every scenario, and they are pretty good at being performant for a lot of different circumstances, but they are still at a major disadvantage.

Why would I even try to?
The fucking system architecture was built by retards.
The 3D graphics api is a complete joke.
The main scripting language is tremendously slow.
There is literally no upside to godot.
No single feature is well thought out or performant.
The only good thing about it is open source.
But who cares about open source if your game is going to end up being shit?

Meanwhile with my own engine, I don't have any of these problems.
It's fast. Easy to use. Easy to build special tooling for different use-cases. It's extensible.
And best of all, I actually own it and can use it forever for all kinds of different programs without having to worry about shit.

If that's the case, wouldn't they just release plugins for various sets of 3D use cases?

>admits to not being able to figure out how to use or improve any of those things in godot
you'll understand when you're older pajeet, get back to fixing those bugs in your backlog

Vulkan is to OpenGL as DX12 is to DX11
You CAN get higher performance out of it, but it's much harder because Vulkan was designed for people who know OpenGL in and out already

It doesn't work like that.
There are many individual things, which have to be adapted.
Shaders, assets, loading, networking, entity logic, AI, and more
All of this stuff can be adapted greatly to the game you make, and thus can be better optimized and easier to use.
Especially when it comes to the point where you want custom tooling for various things, is when you just want your own stuff.

>guy who makes his own engine instead of using premade crap
>pajeet
You must be new.

>bragging on an anime forum about completing a 2nd year CS exercise
>not a pajeet

dial down your ego for a second and tell us about your engine.
does it have skinned meshes and proper materials with normal/specular maps? what about shadows?

If it is just a 2nd year CS Exercise, why use the piece of shit called godot then?
If its so easy to make your own 3D engine, why are you salty?
I'm not bragging, I'm saying everyone should make their own engine, BECAUSE it isn't very hard. It's just some work. Needs to be done, but isn't impossible. In the end you will be smarter, and your game will be better.
The thing is that most programmers are so fucking retarded that they need handholding by some shitty ass garbage like godot, instead of going the route I recommend, which is 'do it yourself'.
And then fuckers like you come out of nowhere and claim bullshit about me, because they can't internally stomach that there are people who like to make their own shit, instead of using prefabricated garbage.
And really worst of all is, that people recommend godot.
They have never used that shit for anything worthwhile, and they just spam that shit in every single fucking thread and I don't know why.
Guys like you recommend it, but can't defend it. No actual arguments why godot is supposedly better than the alternatives. Probably because deep down they actually know that godot is pretty shitty.

>using code written by someone else is handholding
write your own OS then faggot
>No actual arguments why godot is supposedly better than the alternatives.
already explained, because it takes less time
>Probably because deep down they actually know that godot is pretty shitty.
if you've written your own engine before then it should take you a small amount of time to understand godot internals and improve it for your own needs

If you want any sort of coherent graphics you need an “actual” engine, not just a for loop drawing sprites or models. Well beyond year 2 CS project.

Skinned meshes, with a really tight and fast loop, meaning I can animate a SHITLOAD of models at the same time, even on crap hardware.
No specular/normal maps because it doesn't fit the art-style of the game I develop. I'm doing something custom to add depth though.
I have basic shadows, but I'm still pondering if I can do something prettier instead. Sure, I could just stick with good old shadow maps and be done with it, but I think I might want something softer looking.
Maybe sending something like shadow probes to the ground could be a good fit. But it's still just experimenting around.
The next thing I'll be working on is a light pre-baking solution. Probably I'll try some ray-tracing, and see how it looks. Should work, since you can make that look pretty soft.
Have to be careful though; I don't want to bloat graphics memory with shadow-maps, since I'm targeting the switch, and want decent fps on there.
It's really a balancing act.

I like you. You get it. What's your favourite programming language?

>write your own OS then faggot
Wish I could.
>already explained, because it takes less time
Godot takes less time than Unity? News to me, faggot.
And sure, getting a shitty demo up and running takes less time with something premade, but you inexperienced faggots are completely missing the point.
With your own engine, it is TRIVIAL to make all the tools that go with it, meaning for your game, you can actually work on the game much more quickly later.
So much stuff that you can automate easily. So many things that just go a lot faster and better if your tools fit their use.
For example in my level-editor I can pre-visualize so much stuff about the game, its a real help.
> then it should take you a small amount of time to understand godot internals and improve it for your own needs
It took me a small amount of time to realize I would have to rewrite practically everything to make it decent, and would be better off just writing my own stuff.
Which is exactly what I did.

interesting. from what i know, you're not gonna get around pre baked shadows on a larger world. how long did it take you to write the core engine? it's c++ i assume

Quake 2

Clickteam Fusion!!!

Dude that sounds cool as shit. I've always been a, 'maybe one day' kinda dreamer about getting into programming but you make it sound like once the ball is rolling it doesn't stop. Could you recommend an absolute beginner to how you got to where you are?

>unity
not open source
>it is TRIVIAL to make all the tools that go with it
except now you have a bunch more external code that takes resources to maintain
>i don't know how to make small modifications and think i have to rewrite everything
typical pajeet

Its written in C99.
I don't really know what I would consider the core of the engine.
Maybe the moment I could run around with a dude exported from blender, colliding with stuff, can be considered the core, which took me around half a year or more.
If you don't write your own physics like I did, it would probably be faster.
Also, since I wasn't very experienced, it took quite a while to figure out how I want stuff to happen.
The main part of the graphics loop was rewritten two times or so, and basically everything about my structure changed, but I guess that's to be expected.

About pre-baking shadows, I don't think I can do without them.
The graphics just lack a lot of depth without them.
Maybe having low-rez, blurry shadows, and adding noise to them in the shader for some sharpness and detail could be a good trick.

Don't tell that to me, but to the guy who claimed making your own engine is easy.
My point is, people should make their own engines, and if people would more often write their games from scratch, they wouldn't suck as much as they do.

> once the ball is rolling it doesn't stop
It really doesn't stop. I get better all the fucking time, and it feels great. Being able to just make your own shit is really good.
The only thing that you have to keep in mind though, is don't be just content with mediocrity.
If you care about making good stuff, then you will get better automatically.
As for getting a beginner started, just learn how to do stuff with C.
Basic exercises at first, just to get a grasp on how code works. This will take quite a while, until it clicks, and writing most day to day stuff becomes obvious.
Then study and understand what others are doing and how, read code by other people. Don't be afraid to just study someones code, even if it seems impenetrable at first.
A good project to study for example is box2D, especially the lite version. You don't always need to understand every little detail, but need a good overview on what people are doing and why they are doing it.
A good video-series/stream for learning games and general programming is handmade hero. The MOBA-hero-name-generator episode which he did just for fun recently is probably excellent for a beginner. Other episodes can be quite advanced.
Also If you don't understand something at the start, just stick with it. repetition is a key element to learning it.
Be sure to actually find stuff to program that takes longer than just half-an hour. Learn how to keep doing stuff, whatever it is.
If it is relevant to your interests even better.

holy shit why would you write your own physics engine when there is bullet physics?
shadows is what put me off the most from making my own engine. there's just no solution that's really statisfying unless you can raytrace.
Got some screenshots to share?

Incredible that someone who wants to make games would say this in 2019

Build Engine

Attached: Blood_logo.jpg (291x350, 19K)

>not open source
Irrelevant if you want to make a good game.
Unreal is open source for example, so why not use that?

>except now you have a bunch more external code that takes resources to maintain
It doesn't take "more ressources to maintain", you goddamn fucking shitter. You basically share most of your code between your level editor and your game, thats the beauty of it.
I can start my level editor, while I play the game for example, change around stuff in the level, and just continue where I left off. No restart or anything! Can godot do that? Don't think so.
But please continue telling me how much I suck at programming because I made my own engine and didn't use fucking godot.

what do you mean? the part about the physics engine or about shadows? elaborate pls

Bullet was just annoying to use.
They even know that fact themselves, which is why they wanted to make a C-api, but since everyone is too busy it seems abandoned, or at least really horribly documented.
Seems everything the dues focusing on now is making that python wrapper.
I can kind of recommend bounce github.com/irlanrobson/bounce ,which from my look at it seemed pretty nice.I think some stuff wasn't very performant and might need some slight moification, but generally it is really fast, and has the same api as box2D, which is quite good for a C++ api.
My main motivation for writing my own game physics was that I didn't really want to use C++ and that most physics engines have a "physics-world" concept, which makes the stuff inside hard to reuse for other stuff.
Since I target slow as shit devices lie the switch and would ideally still like to have 60fps, I wanted to use the fact that I don't need complicated rigid body physics for the game I'm doing, to make the implementation faster.
For example for (big) bullets in a game, copying them inside the physics world, and back out, is unnecessary. You'd want custom physics for bullets anyway, since adding different sphere (or capsule depending on the speed) colliders every frame is just not very good, and you can do a lot better by having a tight SIMD loop just for bullets.
Also physics engines kind of lack control over the 'resolution' phase imo, which I wanted to have to make my character movement feel really tight. Character movement and collision between characters is something that greatly benefits from having your own game physics, since having it be completely physics based always feels like shit.

There is no one size fits it all solution for shadows sadly.
It's a complicated topic that really needs dedication to get right.
But I don't think you need great shadows for good looking games.
Take a look at some ps2 and gamecube games. They looked amazing without fancy shadows.
Simple stuff was good enough, and for an indie studio, having Gamecube tier graphics is still 1000x better than indie-pixel-shit or the Unity-shaded ugly messes that we see too often.
Even just a drop shadow under your character can be good enough!

>open source is irrelevant
are you the same person saying this while linking open source libraries in this same fucking thread?
>unreal
source available, not open source
>Can godot do that?
yes and you're a moron
>But please continue telling me how much I suck at programming because I made my own engine and didn't use fucking godot.
you don't suck at programming because of that, you suck at programming because you're a brainlet pajeet

upvoted

very much agree. everything around bullet physics sucks, especially the documentation. but the engine itself is the best open source one there is. it has the most stable and versatile constraint system. that is essential if you want good ragdoll physics. see gif attached.
well, i have no idea what kind of game youre making but sounds reasonable then. same with graphics.
bounce looks great. will try it out thanks.

Attached: ragdoll.gif (596x658, 1.65M)

...but im using unity ;_;

So open source is your only argument for godot?
Too bad that doesn't matter for most people making games, since they just want to make quality stuff, that runs well and is well supported.
These things are more important than open source for people developing. Quality is what matters.
And you know yourself, that godot is not even at 10% of the quality of commercial solutions. Which is why you don't even try to argument against any points I make.
I am linking open-source projects because, in comparison to godot, they are high quality.
If there would exist a good quality open-source complete engine solution, then I would recommend it. I can't recommend godot, because I know it's crap. Everyone recommending it doesn't know shit.
I would rather tell people to just use Unity, if you care about making your own game. At least Unity has become somewhat optimizable since mike acton has been working there.
That unreal is not free as in freedom doesn't matter, since you are allowed to modify the source, and release your game with the modified code, which might be the only advantage to free software when it comes to the quality of your game, which is the defining factor for my recommendation.

But the only thing I can truly recommend is, just bite the bullet, and make your own game from scratch.
If you can do it, it will result in the highest quality game, and even if your game ends up making you no money, you can still get an industry job, since you can show people that you have the knowledge necessary to maek game.

>Too bad that doesn't matter for most people making games
Yeah, that's bad.

I'unno, maybe C#? It allows you to write shit code fast (and even that will run pretty decently), but if you want to get into performance autism mode,
it has structs, pointers, SIMD intrinsics and
fast native interop.Besides that, there's a lot of existing libs for it, and will get you an actual job.

I like Go as well, for similar reasons, but for gamedev the support is terrible, plus the perf of calling native code is third-world tier.

Please don't use the standard Unity shaders, they just don't look good.
Especially the screen-space effects look obnoxious and tasteless.
But you can do decent shading with Unity, you just have to actually do the work and make it look good.
Most indies don't care, and just use the default stuff mostly.
I have the feeling what most indie studios need is some dude who actually knows how to graphics.
Too bad the best graphics dudes all work in AAA companies, using their magic-powers for crap games.

This

Attached: 1550459218604.webm (700x392, 556K)

>Too bad that doesn't matter for most people making games
most people making games seem to be "reinvent the wheel for no reason" faggots just like yourself
> godot is not even at 10% of the quality of commercial solutions
and your special snowflake engine is not even 1% of the quality, point?
>Quality is what matters.
not when you can't even use the code due to licensing restrictions
>That unreal is not free as in freedom doesn't matter, since you are allowed to modify the source, and release your game with the modified code
that's not the only thing that that matters
>But the only thing I can truly recommend is, just bite the bullet, and make your own game from scratch.
i made these long ago and i am telling you now, it doesn't matter

are you saying you would have to write your own shaders for unity or do you know of any premade ones? they introduced the new shader graph feature, which is basically just a unreal rip off. is that a good starting point?

And more replies completely missing the point, by this fucktard.
Godot is shit. Making games with it is a pain, and every feature of godot is crap and slow.
All of these are 100% facts.
For mostly everyone these facts are pretty obvious. The results speak for themselves.

i won't argue any of that, however unreal and unity are even worse, and your special snowflake engine is 1000x worse, deal with it

Shader graphs are probably a good start.
Just remember, pbr is a horrible fit for most art-styles, and the standard screen-space effects in Unity look bad.

Good thing you can just use C#

like bloom and reflections?
let's say you was aiming for realistic graphics. what could you do to improve them. the default shaders provide most of the things you would need.

> however unreal and unity are even worse,
No it isn't. Unreal and Unity are quality products, with which many successful games have been made over the years.
They have great features, good support, good documentation, are easy to use and learn, and are generally fast and well thought out.
You must be really retarded to think that Unity and Unreal are worse than Godot...
But anyway, since Unreal/Unity are one-size-fits all solutions, their models aren't fit for every kind of game, and you can do a lot to make your game better and the development process more enjoyable by employing your own tech.
Sure it takes time, but considering the entirety of your professional career, it is worth to spend the time, and dig really deeply into the topic.
And at the end, the resulting game will be better.
I'm not the only person who thinks this way. This is not my original opinion. Many industry veterans think exactly the same.
And with this I'm out. I will only talk about technical stuff from now on.

The only reason why people think Unity is bad is because it's accessible. Since it's so accessible, it's so easy for some inexperienced schmuck to make shitty games out of it.

The depth-blur is especially awful with Unity. It just looks cheesy.
For realistic graphics the Unreal standard shaders are pretty alright.
In general you have to judge the situation, and adapt things to how you think they look good.
Just keep in mind that this is an area that you have to keep a close eye on, since it affects the perceived value of your game greatly. Less is probably more most of the time.

>They have great features, good support, good documentation, are easy to use and learn, and are generally fast and well thought out.
disagree with all of that, they're bloated garbage
>I will only talk about technical stuff from now on.
you haven't been talking about technical stuff this entire time, your entire argument is
>unreal and unity are good because someone else used them once
which is wrong, you are just here to shit on something that it sounds like you tried to use for an hour and never bothered learning properly because you were too invested in your own shitty engine

How do you learn to write shaders? Seems like an obscure profession to get into from my research.

The first shaders you write are out of necessity, and from then on you just learn little by little.
Book of Shaders is a good intro, shadertoy has many fun demos.

Godot does the same Game Maker does, but Godot does it for free.
Unity if you are a good developer with fewer art skills.
Unreal Engine if you are a 3D artist with fewer programming skills.
Unity and Unreal are becoming similar though.

Custom engine in C++ & SDL.

Drawing and 3D modeling, unwrapping, texturing, rigging and animating is a lot of job.
Trying to reinvent the wheel by making your own game engine is retarded and you'll end up doing nothing.

blood, sweat and tears

Is SDL considered enginedev?

based

Attached: C6CKSvOWgAUizC5.jpg (1200x675, 219K)