How's your gamedev going, vee?

How's your gamedev going, vee?

Also. Unity or Godot?

Attached: 54829110-af7ad400-4c83-11e9-8e07-3b3aba4c498c.png (1366x728, 1.05M)

Other urls found in this thread:

youtube.com/watch?v=YTy2KWZJbd8&t=20s
youtube.com/watch?v=2W2xZ0vN-P4
twitter.com/NSFWRedditGif

going alright bro

Attached: Untitled.jpg (1280x800, 85K)

Has Godot had a released successful game yet?

Godot is a pretty trash engine. The only good part is that it doesn't crash every 5 minutes like unity and ue4 do on some platforms.
As for unity vs ue4, ue4 is strictly superior. Infinitely faster and less buggy.

Is there a reason it wouldn't be UE?

Godot for 2D
UE4 for 3D

You shouldn't use a premade engine for 2D. 2D is too simple and easy that it will take you more time to learn the engine and convert your assets to the engine's format/pipeline than to make an engine and editor from scratch. Yet the latter will be tailored to your exact game, which isn't the case of a general-purpose engine.

I don't understand why people ask if an engine released a good game, look at the features of the engine and make up your own mind if its a good solution for your project.

many game engines from the 1990s were built in a couple of months and didn't have basic features like a GUI or support for multiple file formats

Attached: science_woof_barkbark.jpg (670x496, 37K)

Having an editor which handles all the files from images to audio, can do pathfinding, physics, packing of projects, camera handling, multiplatform support, fast and easy level edition and so on makes it the far superior choice vs building things from scratch.

You can't tell from just looking at a feature list because sometimes a feature is advertised using wording that suggest the feature is there, working, and complete, when actually it's a borked mess. Another aspect is code quality and API. If it's too convoluted to attach a node to a bone's node (as in, you have to patch the engine to allow that to even happen but it's spaghetti code galore and takes forever to figure out what to patch and where), you're not going to be using the engine for serious games.

Tiled + opengl + openal + box2d. Camera handling is 1 line of code, multiplat support is called SDL2. All of this takes less than an hour to setup, which is orders of magnitude less than you'd take learning a new engine.
I've never heard of an image loading solution that doesn't handle virtually any format, and news flash: you only need one format to be handled for your game.
If you have trouble doing pathfinding in 2D, you're a lost cause and you're not going to be making any game, engine or no engine.

>orders of magnitude less than you'd take learning a new engine.
So how long does it take you to build an editor where you can drag and drop your content in seconds like with Godot to see if the level will work without having to run things?
>Tiled + opengl + openal + box2d
And one doesn't have to learn how those work?
>If you have trouble doing pathfinding in 2D
It's not about difficulty but how much time one has to invest to get there.

Real good, finished a bunch of the base game assets and am beginning work on the creatures.

Attached: gamesamp.png (1897x929, 2.86M)

Looks cool.

>So how long does it take you to build an editor where you can drag and drop your content in seconds like with Godot to see if the level will work without having to run things?
If you didn't used tiled (for what purpose?), it would take about an extra hour assuming you've never done that before. Otherwise it should take around 15 minutes. It's as simple as choosing a key for the mode toggle, adding a pause flag to your game, binding e.g. scrollwheel to spawnable selection (left-click to place, right-click to remove), and adding save + load.
>And one doesn't have to learn how those work?
They're very straightforward because they're not a monolithic whole that requires several layers of abstractions. Because of that they're much faster to learn, and can be learned as-you-go, whereas that's impossible for an engine. This as-you-go paradigm is what really kills it.
>It's not about difficulty but how much time one has to invest to get there.
copy-pasting a* from some stackoverflow pajeet isn't exactly years of hard work.

Unity is better, but just use Godot so that'll continue to improve.
It's not like Blender could outdo maya etc from the start either.

In my opinion godot is not having any signs of going in the right direction and doesn't offer anything interesting over the competition. Blender always was in the right direction even from the start and improved (again in the right direction) significantly with each update. It also happened to be the first serious open-source solution.

The fact that your imagine looks like nonsense makes me want to say Unity right away.

Whatever you choose just make sure it has the available needs for you without having to wait for new releases. The hiccups of the engine updating can be a pain in the ass sometimes.

Hey user, I've been learning C++ and JS for a little over a week. Can you hook me up with good tutorials for what you're describing? Trying to make a 2D SRPG. Much appreciation.

What is the "right direction" for godot to go?
I see he's starting to get vulkan working.

Yeah but Blender is great software and free/open source.

I vastly prefer UE4.
More powerful, creator friendly, free assets, etc.

I can't understand.
Ok, right, i can make a game, but it looks like manure. How do i make a cool character? I've tried pixel art but, i can't even draw on paper?
I wanted to implement a clothing system but i can't even draw the human!

Advice? How would i go about a Clothing System? How do i find a human and animals? Pheraps drawing clothes is easier.

Thanks user

2D is hard if you're not already an artist because it takes actual talent to make sprites. Try learning 3D rendering.

nice. kinda looks like Stronghold which was a pretty comfy game.

C++ support and access to the source code is a huge plus. If you have a part of the code that needs lots of performance, using C++ is very nice (look up some godot performance comparisons). And if you need a custom tool integrated to the engine, you can just build it yourself and it will be there, conveniently.

The only thing I'm missing is a better renderer. Hopefully, godot 4 and vulkan will do something about that.

youtube.com/watch?v=YTy2KWZJbd8&t=20s

heres my latest update. I'm watching Vinny play Ion Fury while I think of ideas for the first level.

Attached: images.jpg (218x231, 8K)

The idea is to make a minimal viable game with simple sprites and find an artist to complement you. If you make some cool videos and concept, sometimes you can get some patreon cash.

If you can't make cool character art then either learn it or go for a different approach.

3D? Ok, but that will take AGES, like, how i'm supposed to make a character + make the skeleton + make the animations, textures, shaders
That's a lot more mork?
It's like, you can have a pretty 2d easy, a fast 3d is worse than minecraft

To exemplify, check this out. It leaves unity eating dust.

Attached: data_chart.embed_.jpg (800x600, 36K)

In my opinion, he should start by fixing the defaults, then fix the engine capabilities, then fix the api in that order. The api is pretty bad right now and a pain to work with, whatever system you're using to control the engine. But the engine capabilities (lack of support for file formats, asset pipeline in general, performance, choice of shadow techniques, rendering techniques (e.g. forward+)...) are currently lacking in some ways that are only evident when you try making a real game.
Vulkan is not the right direction given the engine's current shortcomings, especially since vulkan adds a shitton of work due to its low-level nature (that means that the vulkan-related parts of the engine will steal even more dev-time from the team when they should be focusing elsewhere, with no tangible benefits because opengl is fast enough for doom 4 and no doom 4 is being made in godot at this point).
The defaults being bad means it takes a while to setup before you can even see your assets properly in the engine. Example of bad defaults include the crap lightning shader and shadows.

Another aspect that would tremendously help is if there were some decent base assets to prototype with. It's definitely within the purview of an engine on the scale of godot to provide that.

Because there's a legitimate concern when the most an engine is used and popular for is as a teaching tool because it's lightweight.

If much more experienced programmers have nothing of worth to show, then what chance do I have. Where's the documentation? Where's the support base?

It's not worth the effort to learn it without a proof of concept.

>C++ support and access to the source code is a huge plus
Agreed, but then there's UE4. CDNative is also a pain to use in practice (have you checked it out yourself?).

>voxel data
What if I don't want voxels

I say it from experience myself.

Is my game and I do it all in Blender.

>(have you checked it out yourself?).
Yes, it really is a pain. The lack of documentation and good tutorials most of all. But once it's all set up, it's not that bad. Compile, run.

I think that is mostly testing nested loops. So any game that have many nested loops would benefit in the same way. Of course, C++ is faster for just about anything else as well.

Thanks user.

For characters and fauna it's not so hard because there's makehuman and speedtree. But I completely disagree with the other guy, 2D is infinitely easier. In both cases though, you can find spritesheets and "draw within the lines" to make your unique donut steel base assets and do like the other guy suggested and get a real artist in the future, if ever.
You can also find lots of free assets (usually of mediocre quality) such as on opengameart.org

>wasting a couple of months/years of your life on something you will lose interest in and never finish

HIYAAA

I keep writing down ideas.
They're very nice ideas.
I need someone to beat me with a stick until I get my ass in blender and model everything for these ideas.

Attached: 1558916100226.jpg (1024x1024, 456K)

Mine comes and goes in phases. Currently on week 2 of a binge of work like 10hours a day after work. Then soon I'm sure I'll do nothing for 3 months.

>good tutorials most of all
By the way, I did find an introductory one, if anyone's interested.
youtube.com/watch?v=2W2xZ0vN-P4

No

My binges are a lot shorter, maybe 4 hours a day for a week, then 2-4 weeks of nothing. doesn't help I've never textured anything, I just want to make essentially a walking simulator anyway

Attached: 1549814926778.png (564x542, 108K)

Godot chads rise up!

Attached: 1559487229932.jpg (1280x720, 112K)

Anons, what's the best engine for a (2D) point and click game?

Adventure Game Studio

Texturing is dumb as shit, but I suck at like everything still. Luckily I render Isometric so my stuff that's shitty in 3D comes out decent looking in 2D lmao.

Godot is regarded as one of the best 2d engines. Try the tutorial.

Pic related, I did.

Attached: Screenshot_20190817_144109.png (1920x1080, 201K)

Your own. Failing that, literally any engine ever because pnc is so trivial it's impossible to mess it up. Only easier thing is a VN, and beyond that you're not even a game anymore.

All these pictures of Godot make me thing of like it being a game engine for the Linux user. The UI is ugly as shit and seems dumb and cheap.

It works well and it makes sense. I don't understand your complaints.

>Your own
I don't understand why anyone would ever say that. It's like you want people to suffer for nothing.

What suffering? You get to just barely learn the bare minimum while gaining the tools to eventually do more than pnc. Bonus points: if you have a unique idea (unique inventory system, object combination, whatever), you can implement it whereas no engine would support that without you learning to do it yourself which will be an excruciating experience.

>using godot
If you're interested in getting employed, then using Unity or Unreal Engine 4 makes more sense, since they're way more common engines. Almost no companies use Godot (apart from microscopic literal who indie devs that only distribute their games on itch.io).

The point is you don't really need a engine for PnC. "suffer" lmao stfu

Presentation in software matters. It just looks like cheap messy shit.

Fucked around with billboarding in godot, made a couple enemies you can shoot. Then realised that there's an entire fps tutorial in the documentation so I'm gonna run through that next.

>getting employed
Wrong site, mate.

yes, a homosexual furry shmup/vn/adventure mashup. it's going ok. stuck on character design for a while though

Attached: Photoshop_2019-08-16_21-05-52.png (638x503, 37K)

I'm using gdevelop and it is a shit engine.
People upload open sourced templates on itch.io instead of github or gitlab.
I have to edit some sprites so it can follow the cursor correctly which is retarded since the engine has no way to edit where it believes the front of the sprite is.
I got combat to work correctly and after fixing the bug I can finally start making the actual game.

>It's as simple as choosing a key for the mode toggle, adding a pause flag to your game, binding e.g. scrollwheel to spawnable selection (left-click to place, right-click to remove), and adding save + load.
And then the requirements grow with rotating tiles in the editor, having multi layered tiles, changing settings of the enemy ai in the editor, needing to change ui element positions and so on. Either one expands the editor functionality again and again or keeps working inefficient. While with Godot one can just concentrate on the game instead of those basics.
>can be learned as-you-go, whereas that's impossible for an engine
But that's exactly how one uses the engine. Implement whats known, look up new things to learn them and continue.

I'm not the "Tiled + opengl + openal + box2d" user.
If you want to use Godot just go to their site and click the learn button. It's rather easy to start with it.

Still trying to learn the ropes.
Doing the Unity tutorials, actually.
Recently made a terrible remake of the first Mario Kart Level using the presets on the Kart Tutorial.

Also downloaded Unreal and will try it later on.
I'm gonna make it, bruhs.
We are all gonna male it.

Don't start with modeling. Start with some spheres and cubes, see if they work, then add the models.

>And then the requirements grow with rotating tiles in the editor, having multi layered tiles, changing settings of the enemy ai in the editor, needing to change ui element positions and so on.
All of these are game-specific and take literally seconds to add, each. Your game might not need it so you aren't wasting time on bloating your editor. Or your game might need something the engine doesn't have and so you're adding the feature either way. When you do need to add it, it's game-specific so it always works better.
Just because you're so stubborn and obtuse I'll also repeat it again: it literally takes seconds to add each of the features you mentioned. Seconds!
>But that's exactly how one uses the engine. Implement whats known, look up new things to learn them and continue.
Not really. You have to learn the API and/or class hierarchy and what function is available. You also have to figure out the memory ownership model where applicable (is everything a shared pointer? Do you have ownership and therefore must delete objects created by factories? How does that interact with components?). For example, one engine requires you to create a CharacterController class with a bunch of relevant functions (which are error("not implemented") by default so they must be overridden), plus a Character class, plus a CharacterState class and tie them together correctly. Another might just need you to setup a bunch of callbacks but need a specific invocation sequence to initialize the required components.
Most engines have proprietary asset formats and won't let you load normal formats like fbx or png, for example, and require using a tool shipped with it for the conversion (which makes the asset pipeline fucking awful and obtuse to work with) or worse, writing some xml by hand to describe the asset/shader pipeline.
Given your posts so far I would bet you'd never tried either, let alone both. You should though.

Quick question
What do you want in a top down shooter?

Fun

A melee weapon to swing and deflect certain bullets

Well, there it is...

Attached: Jim-export.png (219x197, 1K)

This is a terrible chart. You can't compare voxel generation in C++ native (godot or not) to the c# wrapper of Unity. Something as intensive as voxel generation needs to be delegated to Unity IL2CPP for it to be fair. The other aspect is that Unity's job creation manager would likely help performance gains depending on the application too. Godot didn't do anything special here, just used something closer to the hardware.

Kek.

Also voxels r dum

>it literally takes seconds to add each of the features you mentioned. Seconds!
Ok, show your custom editor for your game and it's code.
>You have to learn the API and/or class hierarchy and what function is available.
Learning the API is faster than implementing whats behind it.
Why do you use tiled, openai, box2d if you have to learn their api? You could as well just implement it yourself and have them tailored to your game if that's so much better.
>requires you to create a CharacterController class with a bunch of relevant functions (which are error("not implemented") by default so they must be overridden)
>Most engines have proprietary asset formats and won't let you load normal formats like fbx or png
So you have never used Godot. It loads assets like PNG files just from a folder. Replace them with new ones and Godot reloads them automatically. If you don't like the default import settings you can change them but there is no special required format.

rate

Attached: background.png (512x288, 14K)

Oh, you were just a godot shill all along. Nothing to see here. What a waste of time. Can't fix brain damage after all.

Looks cool, but why such few colors?

makes it 16 bit

Generated garbage/10
You can make it work by blobbing the shapes more and removing the noise.

You could say exactly the same about 3d, like just once in your life you will suffer then you save a file that does the dirty work for you and voila

1. Unity
2. close to 7 months
3. I'm starting to lose my mind

What even is this? Looks pretty bad.

Wait I fucking remember using this.

at least you figured it out
I am testing that so I can paint the background irl then convert it into pixel art.
It is easier and faster to paint than make a good pixel art piece.

My original post was about recommending Godot for 2D and Ue4 for 3D. With that in mind did you expect I would take Unity as reference for the arguments?

Attached: city.png (224x126, 11K)

Funny this thread comes up. I've literally been flip-flopping between Godot and Unity all day for a Metroidvania I wanna work on.

compared to what? :^)

How do you do projectile ballistics? I hear that you have to use calculus and what not.

Compared to all other engines?

>trying to make a dungeon crawler with a friend helping every now and then
>they're so advanced in the curb when it comes to programming a lot of methods they bring up to me to make it more seamless overwhelm me
I'm definitely rushing in headfirst, maybe too eagerly, especially since I'm relatively new to programming, but it does feel good when solving issues.

On that note would you guys recommend Godot or Unity more? I keep seeing people champion Godot online but I've seen a lot of anons here shit on it endlessly.

Look at it this way as another user in this thread already mentioned. Name one game released in Godot that's made or a name for itself.

unity and unreal look way worse

Lol bullshit, I'm sure Godot has customization but either every user in this thread is fucking shitty with setting up their work screens or the engine looks like shit prove me wrong.

Godot is utterly featureless.

Godot is just 5 years old and only since the last few years it's well rounded. Thats not a long time to make an impact on top game releases considering top indie games will have likely 3-5+ years dev time.
But the engine has all capabilities needed and a lot of advantages for 2D games.

how to learn coding

Saying it's only 5 years old doesn't change the fact that it looks like shit.

Look for online courses. Try to code things and solve issues.

Deponia PS4 port

Got any recommendations?

I like Godot for 2D
But if you're going 3D, go Unity

I don't understand the obsession with the engines, I don't really what tools people use, I care about the product. Titanfall 2 still uses Source and that shit is crippling to work with.

Not that user. I've started recently and found a few good starting tutorials.

I started with JavaScript using khanacademy.

Moved to C++ with freecodecamp's youtube video.

Now I'm starting my 2D game on godot since that seems to be general recommendation.

Good luck user.

C++ is better for user input than JS imo, but I'm still super new so

Doesn't Khan Academy only have videos? I like going at my own pace and videos are usually incredibly slow.

The Respawn guys had engine developers who could mold Source into what they want.
99% of people here couldn't make a textured triangle in OpenGL. They're going to be completely reliant on the base engine features.

As such, they're afraid if they choose the wrong engine they'll hit a wall when they try adding a feature and then realizing the engine doesn't do said feature well.

>Saying it's only 5 years old doesn't change the fact that it looks like shit.
Who cares about the optic if it's efficient? The default Unity look is way worse to me but it doesn't really matter as they are just a tool.

They have videos and a built in JS for you to mess around with. I was able to build a sloppy version of pong in it between videos.

Whatever works for you, I`m using Unity and I`m hating, specially the UI features, is extremely slow, and overly designed.
Check multiple engines, look for one that works for you.

Tribalism and nodevs.

Also user, consider w3school if you want text based tutorials.

Godot is pretty easy to use. I'm a fan. Pic related (Alphaman on itch.io)

Attached: Alphaman_chefBotStageGameplay.gif (620x435, 2.01M)

>they're afraid if they choose the wrong engine they'll hit a wall
Depending on the project one will have to spend years with the engine. It's nice having something decent if one has to work every day for such a long time with it.

Because people think the tools make the game not the person and they all suck.

Not really. 2D games have no meaningful performance requirement because modern hardware can bruteforce even the most retardedly unoptimized software rendering solution at full 60 fps, for example. The various libraries that exist for 3D tend to be a lot of trouble to get working together as well, compared to those for 2D, and there exists no satisfying 3D level editor that's system agnostic and easily available. Moreover, asset creation is much harder and also necessary to verify that the system is working.
While in 2D you don't need to care, in 3D you have to do scene management at some level if nothing else for alpha sorting (but realistically you need to keep at least one, if not several, ordering for query efficiency, and without it you'll see crap performance even on the most modest games).
It is still true that it's viable to take libraries together to make your own engine and it is still fairly easy to make a simple level editor, but it's way more complicated to make a good level editor because you have to deal with constraints (e.g. you might want to force object placements to stick to the ground to make placement not a pain in the ass, while in 2d it really doesn't matter; you kindof need free rotation and free placement tools that's axis aware, there are significantly more properties to modify as you're not going to do bump mapping in 2d, etc.).
The performance point is enough reason to go with at least a competent renderer and put things around it... except there is no good standalone 3d renderer (ogre3d 2 is promising but has 0 documentation, shit asset pipeline and no plugins (modules that interop other libraries with it), osg and ogre3d

Because you need a large team of professional employees OR a decade, plus source access in both cases, to be able to use a crap engine to make the game you want. Neither of these are viable for a solo dev. Of course in practice most engines are more than capable of doing whatever crap game people SHOULD be making as first games in these threads. But that's another topic.

Still working on the game ?

How's UE4?
Also how can I do cheap mocap? Is that method with multiple xbox kinects still a thing?

Attached: 1558445282323.jpg (768x1024, 67K)

If by cheap you mean just a couple grand it's out there and it's great. Anything simpler than that and there's no reason not to just rig whatever you're trying to animate in your animating platform of choice.

Not entirely true, my isometric game gets shit for stats if I don't properly optimize my tiles ahead of time. If I just put them in the game and render the full alpha image and draw calls and batches and doing chunk etc. yadda yadda it's terrible performance .