game programming general

I keep getting bored with 2D sprite games in html5 or pygame, I keep getting overwhelmed with OpenGL. So lets talk about game engines and platforms that finally get you making complex games. Unity and Unreal I think are the most popular. There are game engines for html5 and mobile, whatever language or platform there is a game engine for it.

Unity looks good to me, and I hear there's plenty of tutorials and documentation. But I would also choose Unity to be pragmatic: I eventually want to slap the project onto github and then say on a resume that I have experience with C# and visual studio. I'm still not sure if this is such a great plan, though. I don't exactly like microsoft, which probably goes without saying when I'm on lainchan, but I really do like the idea of using Unity to kill two birds with one stone. It's supposed to be a passion project that's fun to keep up and simultaneously good for raking in resume xp points. I also work on a toaster in terms of graphics, so I'm probably stuck with Unity 2D for now.

I've worked with both Unity and Unreal and I can tell you that Unreal blows Unity out of the fuarrrking water. The amount of work required to get something is way lower on Unreal and the blueprints, even when I personally don't like it, it's great for people who are just getting into vidya development, and their business model is way less intrussive than Unity's. You do need a good rig to work on it though. And you gotta pay good attention to optimization and stuff.

Ren'Py is pretty great for making visual novels. If you know Python and can draw hentai then you're only a story away from eroge money. I figure writing is the easiest part, since we've all been doing it for most of our lives. I haven't messed too much with 3D engines, since I don't know how to make 3D models or animation and it's a lot harder to make something look nice in 3D than in 2D. Every time I get into Unity, I start doing a tutorial and then realize I need to backtrack and learn some Blender. Speaking of Blender, does anyone know what the Blender Game Engine is like? I've never messed with it.

>>2299 >Speaking of Blender, does anyone know what the Blender Game Engine is like? I've never messed with it. Blender Game engine uses graphical logic blocks that you wire together, unlike something like Scratch where the blocks are stacked on top of each other the fact that i uses wires allows you to connect the blocks in any order which gives you a lot more freedom. Unity has a plugin you can buy called PlayMaker that allows you to do visual scripting with logic blocks just like Blender Game Engine.

>>2299 I can draw okay-ish and do all the basic soykaf in python, but I don't like 99% of porn anymore. Is there any money in doing fuarrrked up horror games like Saya no Uta or do you have to get lucky? Do you have to PRETEND it's typical harem porn just to get noticed?

>>2323 I was joking about it being that easy. I have no idea. Also, Saya no Uta actually is a porn game.

...so, what is the nearest BSD/MIT-licensed equivalent of unity/unreal, in the sense of a full-featured engine/platform/level editor? I understand the technology will necessarily lag a decade behind. I tried playing with torque3D but it crashes whenever it feels like it.

>>2338 What comes to my mind is Godot. https://en.wikipedia.org/wiki/Godot_(game_engine)

>>2338 Godot would indeed be the unity equivalent. As for unreal you might be interested in torque 3d, but its a pain in the ass to build. http://torque3d.org

>>2326 That game is a mesmerizing paradox. The story/context defeats the point of the porno, and just leaves you in shock. Anyway, I guess it won't be easy. You have to scream it off the rooftops that you have a new visual novel, do viral marketing soykaf, and hope someone is crazy enough to give you money. Of course, I could claim to have created something even more messed up than Saya no Uta but I don't know if I can live up to those expectations.

>>2340 >Godot games are created either in C++ or by using its own scripting language, GDScript Well soykaf, I don't know jack about C++

>>2343 What about GDScript? https://godot.readthedocs.io/en/stable/learning/scripting/gdscript/gdscript_basics.html https://www.reddit.com/r/godot/comments/5ffzm8/using_godot_with_gdscript_only_no_c/

>>2343 Godot 3 will introduce more the ability to load code written in arbitrary languages.

>>2344 This might be okay if it's as straightforward as Python- >>2345 For real? That sounds pretty amazing. Just pick whatever the hell you want.

>>2345 C# is a short list of arbitrary languages that will come with Godot SDK or so I heared. Please correct me if it's old news.

>>2294 Is Unity good for making games for iOS and Android? Or is it better business-wise to use Apples devkit, since the iTunes store wants 30% anyway?

>>2342 >That game is a mesmerizing paradox. The story/context defeats the point of the porno, and just leaves you in shock. You're so pure. >Anyway, I guess it won't be easy. You have to scream it off the rooftops that you have a new visual novel, do viral marketing soykaf, and hope someone is crazy enough to give you money. I think things like greenlight make it easier.

>>2372 Beware the alien, the mutant, the heretic. >I think things like greenlight make it easier. Isn't steam getting swamped in games regardless? By the way, does kickstarter still work for anything vidya or is it mostly used to pitch internet-of-crap toasters to normals now?

>>2372 >I think things like greenlight make it easier. I thought greenlight was on its way out.

When I picked up programming I wasn't interested on game development but I already had a background in art and music so one day I released that if I mixed one thing with the other video games will come out. Sadly I chose C as my first language and it has been a struggle without snuggles

What's a good way to make little browser games? Suppose you want to make another homestuck clone except have it not suck as much as usual, by sprinkling some minigames among the non-interactive pages. Ideally, something that can be played by the user without needing flash. Do you have to become javascript god?

>>2408 There are a ton of books out on HTML5 game programming which basically means Javascript programming using the HTML5 API. There are lots of game engines for making web games and you can even make 3D games with WebGL.

>>2408 Construct makes simple HTML5 based games

Has anyone experience with game maker studio 2? I used #1 to teach a kid programming, it was quite nice, especially since the basis version is free. V2 costs quite a bit of money and while the steam reviews are generally positive, I first wanted to ask around how it really is.

Does anyone want to work on a PC Engine game with me? I've been working through the tutorials on ObeyBrew but want to have someone to work with.

>>2625 Sure; we can at least discuss it some. Meet me in the #programming channel in the IRC.

>>2286 >>2286 >>2293 >>2294 I'm developing a mobile game (for android for now) and I'm using unity (I think it's the easiest to put an android game together in.). I do prefer Unreal and have messed around with it and was going to make a game in it for pc but got distracted/lost motivation (and a harddrive failed (might bbe able to recover stufF) which had the project and assets on it that I modeled myself). Unity is alright for game dev. Right now I need to optimize the lights so that there isn't as much lag as I have hundreds of lights. Trying to figure out how though.

>>2306 So I could connect things instead of drag and dropping? I don't see a point aside from that.

>>2631 Posted but didn't hear anything, let me know if you start going through the tutorials

>>2635 Yes of course you can connect things, there are a few books out on Blender Game Engine, you should take a look at them. You can make games almost as complex as regular programming language code games. There is an open source game made with the BGE called "Yo Frankie" https://apricot.blender.org/ , you can open the .blend file of the game in Blender and look at how the logic blocks are wired together. Also the Unreal Engine has a visual scripting system called Blueprint. The pic in the OP shows Blueprint being used in the Unreal editor.

>>2636 I'd not expected you to show up so quickly. I noticed you in my logs, but not until after you'd left. Syncing schedules, at least initially, can be tricky and so you should simply install an IRC client and lurk as it's called. Anyways, I should be fine to discuss this with you in, say, eleven hours from the time of this post, so let's aim for that, if it's convenient for you.

>>2649 I'm gonna be gone all weekend honestly, but I'll be idling all next week

>>2293 It's still not perfect, but Microsoft has done quite a bit of work lately to put C#/.NET into the open source community with Roslyn, .NET Core, and VS Code.

Hello Lainons. It's been a year since I learn to make a game. Because i want to create one (even is it can be a flop) However for several months I find myself facing a problem. I need to have for the future servers that manage several hundred thousand players and have a huge map (At least 10 x that of GTA 5 in comparison). The problem is that there is only unity and unreal engine that meet these criteria. Unity is less buggy than unreal but unity is not advanced for multiplayer on a large scale, you have to go through photon but it's not great either. Since I will have to spend a lot of time to do this game, I need to choose the best between unreal and unity otherwise I should change and waste a lot of time. Does anyone have an opinion on the question? (Sorry to post it here, but personally here is the best community about a lot of things, so I think it's better than the opinion of a gamesvideo.com guy or others. Thank you in advance.

>>3104 Unreal is about 10x more advanced than Unity. Unity is really just for mobile/browser games. Also you most likely will fail: It sounds like you are going for an MMO. Turn back! MMOs is where even big AAA studios go to die. Doing one alone is simply flat out impossible. Make a game. Then add multiplayer, that will already be challenging enough. If you need to ask if Unreal or Unity is better, you are not part of a game studio that could stem the successful development of an MMO, otherwise you would have people whose only job it is to analyze the different features of different engines and come to a conclusion which one to pcik.

>>3104 First of all, the engine you use has nothing to do with the server: you can write it in any language, as long as it speaks what the client is sending it. Call of Duty's servers are written mostly in Erlang! I would advise you to use Unity, it has much more resources for learning. Regardless, you should just pick one, thinking about stuff like this is almost always nothing but procrastination. >several hundred thousand players and have a huge map Please stop. Not to be rude, but I seriously doubt you're gonna be able to develop an MMO of that size all by yourself, even if it's a massive randomly generated map, and even if you do manage I doubt several hundred thousand players are gonna play it. Start out with a simpler concept. Making games is one of the most difficult tasks you can take on as a programmer, even without factoring in stuff like multiplayer.

For those with experience with unity and blender: should I rig and animate in blender solely or unity easier for it?

>>3147 not really a programming question, you should try /vis/

Which engine is the best to create fighting games on?

>>3161 VBA+Excel evidently: http://mashable.com/2014/08/27/microsoft-excel-games/ https://www.vertex42.com/ExcelTemplates/mine-blaster-game.html

>>3160 well really what else is there to game dev using unity/unreal finding that most of the work I need to do is assets

Has anyone tried making their own engine? I have this idea of an engine that is optimized for integrated graphics and I want to know how viable could the project be and what hardship may I find.

>>3799 Youre going into a world of pain if you want to make your own engine. There are many reasons why there are not only prebuilt specific game engines but also generic engines too.

>>3799 Do it. Relevant to the OP, there's been some awesome work on CEPL: https://www.youtube.com/watch?v=XEtlxJsPR40

>>3808 But what will make it so hard?

>>3836 There is no way you can even get close to anything that is already out there, forget about even getting close to anything of that kind. You have to start from scratch, its just not worth the effort.

>>3853 I know I have to start from scratch, I was thinking of studying already existing engines and see how to replicate what they do but as I said the engine I imagine is optimized for integrated graphics so very elaborated visual stuff is out of the plans already, unless I'm overlooking something more complex than graphics.

Anyone here tried Xenko ? Right now I'm messing around with it, though I'm still trying to understand how to use this engine.

>>3799 I'm actually really interested in this sort of thing as well, but have no idea where to even begin.

>>3928 you havent even specified what kind of game engine you want to make. if you just want to make a simple 2D game engine I would suggest checking out all the HTML5 game engines, many of which are open source. If you want to make a 3D engine, this is very difficult, it requires you to learn some linear algebra and to learn either OpenGL or DirectX

>>3929 I understand your concern better now Yes, I want to make a 3d engine and I know that a smart art direction like fewer polygons and smaller textures may make the game more accessible to lower end machines but I want more optimization.

I actually forgot basic trigonometry. Any good sources to help me get up to speed? I was about to add some simple collision detection to my game and I was rather surprised at how long it took. Also, for a tile / grid based game, what kind of broad phase do you usually use?

>>5996 >simple collision detection Are you just trying to check if two 2D shapes overlap? If so, you don't necessarily need trigonometry.

>>3928 First of all, if you want to make an actual game then use an existing engine. If you want to learn how the engines work and aren't afraid of entering the aforementioned world of pain, then by all means start making an engine. Start by opening a window and then draw something in that window. Triangle and then a cube, and then a texture mapped cube that you can move/spin around is a classic in 3d graphics. That'll teach you the fundamentals of 3d graphics with OpenGL/DX, and then you can start implementing more of the few million things that the game engine needs, like asset/model loading, advanced shaders, lighting, scenes, culling etc.. And this is all in the graphics side, there's still all the audio, UI, editors that you need to implement. It's a lot of work, but very rewarding and interesting work especially if you're into 3d graphics. But don't even try to make an optimized engine because you're probably not going to achieve that. Squeezing the max performance out of the game engines is whole another realm of nasty black magic that's very dependent on the platform itself.

>>3799 You can always try starting with this.

>>6137 How relevant is the "graphics" part in that book still? It's 20 years old already, surely things must've changed a lot.

>>3799 Hey! I've been developing my own engine (not finished yet, currently I've got no time). It's in opengl3 and can be used to build apps for PC, android (ndk) and web (emscripten). I plan to rewrite it from scratch starting around march this year. My goal is to have builds minimized by their size. What I don't like in huge libraries/engines (take Unity3d or Qt5 for example) is that even simple hello world is more than 10Mb of data. I want to have some kind of project manager where you can check which modules you use in current project. Every module is distinct .dll/.o depending on target architecture. This way you din't have to import texture loading libraries and classes if you don't want textures etc.

>>3799 Also idea of optimizing for integrated GPU seems cool. But how do you want to optimize it?

>>6144 If you have access to the platform and understand your code and its requirements it's very easy to play around with the settings and figure out what works best for a specific platform. Reading up on the architecture of the GPU would also help.

>>6153 But you are speaking on Opengl level or lower like Vulcan or CUDA?

>>3799 It depends on what you are going to achieve with that. Also it depends on how complex it's going to be. I made my own 2D game engine and editor for it in one summer, and even created a small demo-game with it during the September. But it was simple enough, mostly built on Qt library. It's really good thing to understand OOP, make some problems and successfully solve them, start to understand why design patters were invented etc. But I think you shouldn't have plans to make some profit with it, it's a really utopia. But if you want just make games you should choose between Unity and Unreal (or any other game engine).

I've been playing with löve2d, it's pretty nice, like a better pygame. There are more genuinely helpful libraries available (I use HUMP heavily), and since it uses LuaJIT you can do way more before you start having to worry about lag.

>>6437 >löve2d >HUMP >ANAL I forgot about this.

>>6437 It's nice but I prefer pyglet and SFML

Godot 3.0 is out. https://godotengine.org/article/godot-3-0-released It has great new features that for certain make it the best foss general purpose game engine out there. Its an absolute breeze to work with and powerful too.

>>6601 GDNative is pretty cool, allows for binding of any language to Godot's API. I'm thinking about messing around with golang in godot, should be a bit of fun.

>>6601 How's the work flow if I want to work 90% in code? Unity and Unreal bothered me with how poor the programming work flow is.