Amateur Gamedev General ~ /agdg/ + /vm/

Selection year: 2017
The screencap is from last year's form. Assuming it's like any other government form, you'll have until March 2nd of this year to apply for 2018's benefits.

I'm having a problem conceptualizing how I would go about solving this problem:

So my problem is that I can't figure out how to go about testing for valid connections between the anchor-points. How do I go about making a test case for improper attachment? Or this perhaps the wrong approach to it. First pic is a good connection, 2nd is what I'm trying to avoid (among others, as more complex objects are made by attaching them together)

They're supposed to snap together when the mouse button is released when a valid connection is found and highlighted like that.

>>>Holla Forums855436
>>>Holla Forums854951
Just how desperate for a pity party are you?

just remaking old nes and snes games, but I really doubt somebody will give a fuck.

I do give a fuck! I was making my own game, but I got unmotivated. I can't do music or very much drawing. But I'm pretty good at programming. If I can help someone, that would be great!

>>>/trash/
stay there

is not that, even if I were able to code and music and art, it will take years to do a game.

sure the guy that made stardew valley and minecraft, but it took them several years.
compare that to music where a masterpiece took vivaldi three weeks to compose, or how you can finish a digital painting in less than a week.
a game is simply too much work and you need hundreds or dozens of highly skilled workers.

is just a pipe dream to think a single guy can compete with a professional team.

No, that depends on what kind of game you're making. And if you have a working prototype, more people will be willing to help.

What kind of games are you remaking? I'd love to make something like the modern Battle City, lol.

I've always taugh maybe I could specialize into art and remake the assets of such games in HD and hope someone code the game.

but I am not sure, should I give a try to that?

Ubisoft did something like that with HOMM3 and it was pretty successful. They even did a deceptive marketing campaign. If people have good memories from the game, they will jump onto it.

I would love to remake the first super mario brothers sprites in HD.

...

Scrapping Unity and switching to MonoGame was the best decision I've ever made. So glad I don't have to deal with dumb "engine" issues that are out of my control. Fuck that bloatware. I recommend all other anons start programming from scratch

How useful are Accelerated C++ (Andrew Koenig and Barbara E. Moo) and Beginning C++ through Game Programming (By THE Michael Dawson) for learning the language?

I'm not exactly a complete beginner, but I'm taking the rest of the year out of Uni (Programming course) for personal reasons, so i want to try and learn as much as I can in the meantime. If there are other tomes that are highly regarded then those would be nice to know of too.

What, you think you'll make yourdream game this early? Do people participate in a pro boxing championship before even entering the amateur league? Step up your game nigger first do some games, grow, make some bux, perhaps start a company and one day make whatever dream games you want, but until then it's hard work and eating shit. There is no shortcut.

they are shit IMO, you only learn from case specific stuff.

Just go with the regular Uni books and don't forget to get gud at math and you'll be making shit in 6 months

you can't outwork something that took entire teams of professionals several years to make.

is not possible.

You're right, you'll likely never create a game with will compete or eclipse a release from AAA studio. At least not by yourself.

Who cares? If you're doing this to make a lot of money or get famous you've got the wrong hobby.

that's not the issue, the issue is that at least in music you can compete with the genius by yourself, it will be hard, but is doable for a single guy.

or literature, or art.
is not possible with games.

but videogames are art

seriously though how did you think that fat blob Notch made his billion?

The true art of the game is on the design of its gameplay, all else is but complimentary enhancements

Well you can't start a rock band by yourself if you only know how to sing. People like that become solo singers. Instead setting your goals too high, just make something that you can make alone.

So don't. Make something different. Your perfect game only exists in your mind, so make that, and if others like it too, that's bloody phenomenal, but otherwise, fuck them, they have shit taste.

what the fuck does this even mean

do you think that AAA games are "genius"?

Also you realize that throughout history there are examples of composers, writers, and artists that were unknown when they were alive until someone rediscovered their work years after their death.

If you want to make something "artistic" then do it. But if you want to be recognized as an artistic genius, then make an "art" game about overthrowing the patriarchy while being a trans-woman of color.

You're not getting it.
In music somebody could finish a piece in a couple hours that could make millions.
Or in art you could finish a painting in a couple hours.

In games you can't even finish a prototype in one day less alone finish a single game.

Is about focusing on a single skill that being mediocre at several.

And that's what makes video game development so enjoyable. It takes so much work to accomplish that it's impossible to get bored. You can write a song, sure, and it's exciting, but you'll probably get fucking sick of it after doing 20 in rapid succession. Same with 2d graphics, 3d graphics, writing books, or programming of any sort. But combine all of it and you will not only never be bored, but you'll also nearly never run out of work. If you're a musician and get tired of making music, you're fucked and have to force yourself to do what you don't want. But if you're making games and don't wanna write music? There's probably about thirty features you still need to code, plenty of NPCs you need to draw/model and write dialogue for.

Stop being a whiny faggot and just like make game. And if it's "too hard", well, go do something else with your time, no once's forcing you to stay.

Technically possible, but by the same logic someone could buy a $1 lottery ticket and make millions. Such an outcome is not typical. And if you think that music only takes a few hours to make millions you are naive. Even modern pop music has teams of people that work around the clock with expensive tools and equipment. And modern albums still take weeks to months to come out. I'm sure there's some exception where somebody slaps music together in an hour and makes millions, but again WHO CARES

Again, this is simply naive and unrealistic. Most of the great painters spent literal YEARS on a single painting.

Sure it is possible to slap a painting together in a few hours. It is also possible to make money off of such a painting. Maybe even a lot of money.

I'm not sure what it is you're complaining about. Are you mad that people can make a lot of money without (seemingly) a lot of effort? If you really think that, then you should take up music or painting and start making millions with the pieces you create in a few hours.

absolutely false. Why do you think we have game jams and ludum dares where people make a complete game in 48 hours? It's possible. Most of the time it results in a subpar game, but sometimes not.

I think you're just just frustrating and its lead to you being whiny and lashing out.

Bottom line: you are not going to be regarded as a genius-level artist by making games. Making games won't make you a millionaire. If any of these hard truths are deal breakers to you, then get out of game dev now and save yourself the anguish.

Yes, you're right, so stop whining and develop artistic and musical skills. If you don't want to, then pay someone else to do it. That's how it works

I just want to make some money, not millions, is hard to be 29 and be a failure.

On topic:

I need some advice for Godot (3.0 RC 1)

I've got a lot of Floor and Wall tiles, all as separate images.

What's the best way to get them as sprite nodes in Godot so I can create a tileset out of them? I don't want to have to manually add all of them (there's about 248 total tiles).

There used to be an addon for this type of thing I think, called Tileset helper. But it doesn't support 3.0 yet.

Consider getting a job, then. Doing gamedev as a job is like doing youtube as a job - you can't rely on it unless you already have a following, and even then you shouldn't rely on it because that following will drop you in 15 minutes for the next guy.

I just want a fucking game engine where you code everything and they don't force you shitty editors.

I guess libgdx is what I want, but they still don't have engine features like quadtrees and shit.

Game dev is not the industry to try and make money in. You're better off just trying to win the lottery. People make games because they love to, otherwise they'd be software engineers.

Here's a secret:
code it yourself

I'm sure you can still select lots of images when making a tilemap.

You're being incredibly insulting to musicfags and how hard they work to make things sound good, and artists and how long they can slave over a painting. Neither of which take into account the probable tens of thousands of hours they've put into either of those crafts.

Your attitude stinks and is the kind of self-deprecating "woe is me" bullshit that I can't stand. You want people to agree with you and lament that gee guys it really is too hard I guess I'm validating in giving up well fuck that. People that want to make a compelling game on their own work damned hard at it and from the types of posts you seem to make I struggle to believe you've tried at all.

Depends on the idea, most normalfag game ideas invole stupidly complex systems, huge game worlds, and so on
Basically whatever they like from a game they want to throw into some AAA 3D trash FPS
My dream project is 2D, and doesn't involve anything overly complex, just some comfy life sim stuff
Look at Everyday Life user too, he's making good progress on his dream project.

Make games or make money, you can't have both
If you want to make a quick buck, try drawing furfag shit or making vapourware, You can't live off indie vidya.

The only way I've found to make a tileset is by manually adding sprite nodes to a scene and exporting that scene as a tileset. I wish there was an open to open up each image as a separate sprite node

docs.godotengine.org/en/latest/learning/features/2d/using_tilemaps.html

Good work, follow your dream fam, remember we are all here to support you.

You're setting yourself up for disappointment by comparing things that are very different. Video games have always been primarily a team effort. Music, art and literature have always been primarily a solitary effort.
Not to mention you're ignoring the amount of failure all these artists went through before they got their winners. It's even ignoring how many game ideas are trashed by AAA studios before they publish a game.
Your reasoning is shit.

If your reason for wanting to make video games is because you feel bad about yourself. Then you're doing it for the wrong reason.

Did you even read the entire post? Reread:
If you can't make your dream game alone than you should be prepared to get a team at one point.

Then what's shit like flappy bird? If you want to make big bucks through luck and no effort then try mobile games.

Stop being such a weakwilled faggot, jesus christ. You should never depend on luck either. Not surprising you're a failure with 29 if you always had this kind of cuck mindset.

How has nobody pointed out how fucking idiotic this statement was

Pride.

What are gamejams?

Taking up any form of art won't help you with feeling like a failure, it might even make it worse because you'll fell like your life depends on whether or not you make a hit game and you'll end up like Tetradev.

Stop chastising yourself for your past mistakes and wasted opportunities over and over again. Stop dreaming of fame, you're a failure already so accept that. By that I mean if you're life is shit then there's no point in expecting sudden change overnight. Instead try to find ways to make your day a little more meaningful and bearable. Reward yourself and appreciate the smallest good things you do, instead of feeling like you've never done good enough. Find something that motivates you to improve your life, read self-help books or some shit, I don't know what works for you. And don't fucking treat gamedev or whatever form of art as a gateway to dream life, it's a hobby meant to be fun, not a life-changing lottery.

The way I would probably start is by having each anchorpoint have a value that dictates which part of its side its on(ie like a coordinate system) and then when two anchorpoints are linked you can math out which ones are allowed for the second point. So if the top left point is 0,0 and its linked with another 0,0 then the point to the right(0,1) can only link with another 0,1 and not a 1,1(which would be the bottom right aka on another face's anchorpoint.
I dont know how much sense this makes. I qould make a visual which would make more sense but I'm on my phone and I'm just trying to visualize in my head.

Help please. I have lost almost all my drive for game making. Took a few months off (hoping a fresh take would do the trick), but still have zero passion for anything. Not like I hit a roadblock or anything negative that would ruin my time, just completely lost the will to work on it.

Someone please suggest ways to get myself out of this garbage mindset. How do you guys hype yourselves up to work on stuff when you otherwise wouldn't want to?

Schedule an hour of time to work on the game, and then during that hour work on the game and don't do anything else.

Are you finding yourself doing other things instead? If so, try to consciously avoid them or restrict access to them. Start by doing something, no matter how small. Tackle that small, insignificant bug or something trivial that you put off. Anything to gain back some momentum. Beyond that are you working on a game you're looking forward to play and are invested in seeing get done? Are you unsatisfied somehow with how things are going? And be honest with yourself.

Play video games until I feel like developing again.

Nah I get what you mean.

In my opinion simply starting and the first few minutes is the hardest phase. Just open a program you can work in right now no matter how much you don't feel like it, since once you started working it's easier to keep going. Another thing that could help you is Music, listen to some high energy music like dnb or rock to get pumped up. Other than that, stand up right now and do jumping jacks or something similar warming up will turn your body into "work" mode and your mind will be more eager to do something. It's also very important to never say "ok i'll start tomorrow" because you will say that everyday from then on, don't procrastinate, it's a dev killing trap.

The problem is I don't think that scales very well with both the object scale and its shape. If I had a triangle or something. It's gotta be more robust

Like you want to connect a triangle to a square kinda thing?
If that's the case maybe you should instead check for overlapping colliders when two blocks are connected and reject the link if two blocks are overlapping. It looks like you are using Unity and it has built in functionality for checking overlapping colliders. That might be simplest way to do it in a general case that will work for different shapes.

How many of you have a side project to your game? I have the main game I develop but sometimes after slamming my head against issues in the main game I feel like having a side project I can work on. However I feel like I'm just going to start treating the main project like work and the side project like the actually fun one, which won't end well (eventually I'll abandon the main one and the side project will become work).

Correction, game dev is the industry where you can make a ton of money, but people here aren't part of the industry and being in the industry is soul crushing work.

Sure, but even in the industry, the hours-to-pay ratio is incredibly bad. For programmers anyway.

I don't have a specific side project, but I do side stuff for fun. Right now, I'm editing an overlook of the budget phone I got, as there's only one review on Youtube, and it's in Russian. Editing videos in general is a really fun, one-off two-day side project for me, and helps me think over my gamedev work in my mind.

Will try that next. I remember hearing good things about keeping yourself to a personal schedule. Hope I can actually keep to one though.

Hmm. I'll have to think on this one more. Thanks for the advice user.

I tried that, doesn't work all that well for me as I just start wanting to play vidya instead.

Music sounds like a good idea, and never thought of using exercise to get warmed up. Any suggestions for music to go with vidya work? Also on this bit:
I am guilty of this. Far too guilty of this.

Thanks for the help guys. I don't usually check these threads, but I think I might drop by more often in the future.

Stop posting on the X-Com thread.

Do not view game development as a chore and don't force yourself into it even if you don't really want to, because you think you have to. It's important to realize that you develop games because it's fun. That's when the motivation will just come in on its own.

Joke's on you, I never have. I just saved the photo when it was in the catalog.

I wish i had a good idea for a new project.
is a meme, at least when talking about fleshed out full on concepts, not one-line pitches.

Gets me thinking, is there anyone here who works in the industry as a developer or some sort of artist? What are you doing exactly?

I just wish I had ideas, beyond Sokoban or Asteroid clones. I just want to make something simple that I can put on some app store and break the ice.

rate the manbutt

9/11 boypussy

Stop posting varg

also I gotta ask. does anyone here know how to into java and would be willing to work on a Holla Forums mod for wurm unlimited for us?

i do
and i don't want to do any more java

cool beans you'll start on monday

The lock on camera still needs a lot of work, but it'll do for now. Doesn't seem to be much reference material on them out there.

Onward to Interactables and NPCs

Cool stuff. Are you using blueprints?

And, unrelated, but did KH ever give a specific answer to who the disembodied voice is, or is it just a narrative thing? It would make sense for Sora's to be Ventus, but it doesn't speak like him.

the n64 logo has 64 faces and 64 vertex

your stuff always gets me wet, once i finish what i'm working on i will make an autism hearts clone
featuring dante from the devil may cry series

girls can't agdg, only cuteboys

Neat

No surprises here.

I know this has probably been asked a million times before, but is there somewhere I can go for decent beginner 3D modelling tutorials? I've been hopping from youtube video to youtube video trying to find just one decent video about anything, but all the videos are 40+ minutes long with some retard that seems like he doesn't even know what he's doing droning on and saying "uhh." Actually why is every type of tutorial for 3D modelling either done by a retard or done like they're trying to teach a retard?

tbh

Made with 99.9% blueprints. I had to dig a bit into C++ to create an instrumental component for all my custom flow control nodes. Unreal really could stand to improve their own flow control library.

Always good to hear that others are interested in my practice work.

IIRC Sora kind of talks like this to Ventus when he gives part of his heart to him to keep him alive. I'd wager that there is more going on with Sora then we are currently told. 5 bucks says he is somehow kingdom hearts incarnate.

I may be able yo help, but there is one caveat. Do you speak German?

Poste mal den Link, bin auch auf der Suche nach nem guten Tutorial

I only speak English, but thanks anyway.

Too bad the EU is falling in on itself
Thats a lawsuit

I dunno that Broforce game gets away with having characters like Rambro and Indiana Brones in its game and didn't get sued.

Of course nobody will care. Nobody cares about anything one makes unless it has good marketing. Take something such as Monopoly for instance. Not a really good game, but it blew up in popularity because of marketing: an iconic mascot, a low price-point(not free, as free games gain absolutely zero traction, popularity, or attention), and the right time for release all combined to make Monopoly tremendously successful, so much so that we still talk about it today.

Ist kein Link, sondern ein Buch.
amazon.de/gp/aw/d/3836244543/
When I wanted to learn Blender, all the online and YouTube tutorials pissed me off. Most were obviously made by people who taught themselves.
>We'll cover this in a future video that I'll never upload, because I'm going to loose interest soon
And don't even get me started on their fucking around in the menus, because they apparently don't know the basic hotkeys of an application that is build around using hotkeys.
I bought the first version of that book for 50€ and don't regret it one bit. It was written by a guy who knows Blender inside out, it's well written and structured, it's complete and it doesn't waste any pages on fucking around.
If you put in the time, 1-2 weeks with that book will make you know Blender better than most of the idiots doing tutorials on YouTube. The rest will be up to practice and experience.
It's not a book on how to create art though. It's a book on how to use the tool.

Sorry. Last time I checked it wasn't translated into English.

Danke, user

wew

I've started work on a texture browser for my map editor. This will be a lot more advanced than the previous texture browser in the first map editor. Right now I have to finish clipping filenames and then I will be able to work on having multiple textures displaying in a grid like pattern. Then I can work on scrolling up and down.

...

Here it is displaying a line of textures, it will make the images all lie on the same lower plane. Right now it doesn't have multiple lines but I can open multiple texture files and have them displayed, which is cool.

Are you doing like 1 texture per file?
My game takes advantage of a 512x512 sprite sheet and I'm deathly afraid of using larger or multiple textures for some reason

My friend went to school as a programmer, later left the vidya industry a few years later though as there were better paying + more stable jobs in other industries. He sounded like he hated it tbh, and said most of his coding friends either went Indie or followed him into other non-vidya stuff.

The thing is software developers/engineers can make much better money than game devs, by spending lower effort developing business to business software.

Breaking into the game dev industry proper is difficult. Work in the industry proper (and even indie work due to the current landscape) is soul crushing.
Doing gamedev is best as a hobby.
If you can, do the same, but if you want to do game dev, save up, pay off debts, and try to get zero worries so you can focus on the work you love.

yeah I am, it's true that it's not the best way of doing it, since for example something like CONST2 has the obvious problem of requiring ARB_texture_non_power_of_two at best and ARB_texture_rectangle at the very least. It can't be mipmapped on a card that supports ARB_texture_rectangle and not ARB_texture_power_of_two. So, it would make a lot of sense to just pack all of those special textures into true power-of-two textures and access them that way. I'm going to have to deal with that eventually, just not right now.

If I packed a few textures into one big texture sheet, and then had a separate file labeling where they all were, that would make a lot of sense. I could also check for mirrored textures and then label them so that I can use IBM_texture_mirrored_repeat on them, saving more memory by letting me cut them in half. That whole system i'm thinking of will take a while, and I don't know how much it is worth to implement, so for right now I will just use one texture per file and think about it later.

Now also, isn't a texture atlas basically just a dynamic texture surface that can be drawn to, with entries being added or removed as they are loaded?

It's just a texture that has multiple images on it. Instead of using one texture per image, you use one texture with multiple images on it and then crop it accordingly to access the image inside of the texture that you want. It's really useful for stuff like fonts. It's not dynamic but you can add and remove things to it if you want. This however is worse than just storing a bunch of textures and then adding and removing new ones as they are needed. To do it, you would need to keep an uncompressed copy of the atlas in RAM and then to update the atlas you would need to copy that into texture memory which is a slow operation, it's a lot better to just manage a bunch of little textures if you need it to be that dynamic. Ideally you want to set your textures once per level or whatever.

The issue with using a texture atlas though, is that you cant tile textures that are on a texture atlas if you load the entire atlas as one texture on the GPU. Also it has the issue of being bad for mip-mapping because of textures bleeding together. I actually didn't think of those problems, so perhaps a texture atlas is a bad idea for what I am doing, which will heavily rely on tiling textures across a quad. So, I will probably just manually tile textures like CONST2 to get power of two textures.

What shitty tool are you using that doesn't allow for that?

I think he means that while the whole image can be tiled, if you go OOBs on the texture coordinates, you can't tile subimages from it.

It'd be cool if I could figure that out in XNA/FNA for a future project actually

Graphics cards are simply not designed in such a way- API's like OpenGL reflect this by not allowing the programmer to do it.

So let's say I have a texture atlas and I want an image that is at UV (0.0f,0.0f) to UV (0.2f,0.2f) in the texture. The way tiling works is that you just set your UV's to values higher than 1, so then it repeats the texture like that. So, giving a quad UV (0,0) to UV (2,2) would tile it so that I have four copies of the texture. But of course, if I give the quad anything higher than UV (0,0) to UV (.2,.2), it will just show me the other parts of the texture atlas, and tile the entire texture atlas. You can of course just fix this and achieve the same visual result by having more quads but that is not the same as tiling the texture across one surface.

Give all connectors on the same side of a object a value that they share, that is different from the value of the connectors on the other sides.
When you check if a connection is valid

>dependencies are fine for the most part, but every single node has lost it's properties
Aw fuck. I knew this was gonna be painful, but this is ridiculous.

game dev diary:
2 years in and i still have no idea what the fuck im doing
here's some crap i jammed together from store assets

Make a tactical fps cyberpunk game about being in a virtual world hunting evil hackers.

Also, give it a green filter.

There's no reason you can't modify the basic shaders (that you have full access to) to cause it to wrap around on the tile you've got, without adding extra quads.

sounds like a really bad bug
the kind that would make you search the issues or write one if it needs it

And now you realize why bigger companies don't upgrade to newer 3rd party engines often.

why are you doing and editor in a rpgmaker game?

that is some real wild shit right there
nintendo illuminati confirmed

hownew.ru

Is Sigma the one who keeps making those shitty doom clones? Basically retro/pretentious "I made a minecraft clone" stage everyone goes through.

...

I'm not dogging him, I also want to make a doom clone sometime

>someone makes ONE Doom clone, then starts working on his next engine

YOU MAKE ONE DOOM CLONE AND YOU'RE KNOWN AS THE DOOM CLONE GUY FOR THE REST OF YOUR LIFE. DOESN'T MATTER HOW MANY ENGINES YOU MAKE, IT DOESN'T MATTER, ALL PEOPLE CARE ABOUT IS THAT ONE TIME YOU GOT WASTED AND REMADE DOOM HOLY SHIT

t. rustdev

Let's cleanse the palette with something that always motivates me, personally

Yeah, I forgot about shaders when I was writing that post. Even then, some cards don't have shaders, and other cards don't have shaders that are advanced enough for me to want to use them for that too, for example my RMB9K card's shader has 8 instructions per pass and 2 passes total. So, i'm more concerned with seeing if I can fit normal mapping onto that shader, anything else is a waste of instructions.

See pic related for what I am doing- I am building my own game engine. It has been in development for slightly over one year. I know that it doesn't look like much for such a long time, but I have been very busy with school, and I also have been learning Vulkan. The engine has an OpenGL 1.1 and Vulkan Renderer, so it can take advantage of old and new hardware.

My last game was similar to doom, but I released it a year ago and haven't finished a game since. It's not that the goal was retro minecraft graphics, but that I was programming the engine and tools and wasn't able to program an engine that had better graphics. Besides it has slightly better graphics than doom, I have lightmaps and true-3D in that engine. Really bad performance though.

My next game might get called a quake clone because it will use 3D models, who knows how it will turn out.

I'm just giving you a hard time, we all like to see your progress

I spent the evening lurking the other thread wondering why the fuck it got so quiet.
I am not a clever man. Have some frames of the Doomslayer I've been working on

...

Question boyos,

So my "game" is a 2d grid based dungeon crawl and I'm implementing very basic physics / collision detection.

Basically when a GObject status changes the physics component of that object triggers and it sees what direction its going and then resolves the collision with whatever is the square it's moving to (its passed a pointer to the world class).

Right now I have a map of active entities that exists in said world class. There's no easy way to tell if there's an object in (x,y) when pinging the world.

So I decided to go ahead and add a container to the world that is a two dim int array that holds an integer (entity id) if an entity is there otherwise it holds a -1.

So far so good but then I realized I wanted to have multiple things on one tile.

Is a three dimensional array of integers really the best structure for this type of thing?

I just feel like that is a lot of wasted space since most of the map will probably be empty tiles. I had planned on using this container as my second layer for drawing to the screen as well but my map works better (since there's no wasted spaces).

And now to the autism question:
Is having a world function that iterates through a map of entities that then pings each entity for it's x and y and adds them to an array to be returned better than the 3dim int array (this would also require moving these values around whenever something moves)?

And yes I do realize this goes against the whole just do whatever is able to be implemented quickly and "just like make game".

Instead of holding an array of every single position, which is a giant waste of memory, why not store x,y points for each entity? Now of course this takes up much less space, and is better than the arrays, but the superior option is a spatial partitioning tree like a quadtree which will allow you to search and store enemy positions efficiently.

...

NVM. I had a hard time reading the other post because it's formatted like shit. I didn't realize the other user literally wanted to make an array that holds every single grid space in his entire game. Holy shit.

This is my texture browser tiling the textures correctly. It has a bug in it where it thinks that the max_height_per_line is actually the max height in the entire array of textures. This isn't a big deal though since I am not using textures that are too big right now.

I've seen this exact same thought process so many times on here. user wants to make 2D game, and then immediately thinks up some ridiculous brute force approach involving 2D arrays.

okay I need a quick help here
I'm trying to create a new Label node in godot but for the life of me the editor does not like it
what is the fucking deal?

I'm pretty you need .new() for the objects and .instance() for packed scenes, right?

cool, just wrote my first needed code for bitflag operations in godot, still lacks test

# toggles bitwise at positionfunc set_type(bit): if bit & T_SUM: type ^= bit;# checks bitwise at positionfunc get_type(bit): if bit & T_SUM: return (bit & type) != 0;

pardon my ignorance, I think .new is right, the editor was just messing up for some reason
maybe because the script isn't loaded as scene

Putting ideas for some enemies down on paper for the new stage now that I have the geometry and new falling and teetering platforms all working with the pendulums. I'm trying to decide what I want to do with the demo day though now. I could put a really lengthy stages 1 thru 4 demo out, but I've done public tests for all those stages except 4. I'm also considering maybe putting a boss rush type of demo out since a lot of people struggled to get to bosses 2 and 3. I don't know, I personally didn't find the stages too hard but I have toned them down a bit.

do a combo mode like savant - ascent

I just looked that up and damn thats some great art. I wish I could do it like that. What do you mean "combo" mode though? Like as in it has both?

the game is about going for as long as you can while getting alot of points via hitting enemies in a succession to get higher combos

and the colors are just the normals aren't they?
I never realized that before

Hm. I'm not sure how that would work with my design, since I'm following a more traditional Castlevania/Ninja Gaiden style design philosophy where enemy and obstacle placement is very methodical with specific intended methods to beat them. So I don't generate anything, its all placed out. I could try and come up with something like that but it would probably feel weird.

(me)
nevermind they're not

This is why you /agdg/

for every good GDC talk there are 10 gender study pink haired brown whales talking about their experiences in being a diversity hire instead of talking about vidya

something like the first mario game then, Mario Bros

objects you use .new(),
for scenes you use .instance()
do not rely on the editor, it is garbage.

Is developing for Android hard? I don't see why more people aren't taking advantage of the market if it isn't. Why aren't people copy pasting and tweaking games for mobile? They already copy paste other mobile games all the time so why don't you just reskin and tweak a good game from pc to mobile?

I was trying to actually draw a human for once instead of a robot. I think the body itself turned out fine but for some goddamn reason I can't make a proper looking neutral face. She either looks horrified or smug as shit, as pictured.

guess i'll just have to keep trying.

is this some sort of sign?

Because you don't have M&KB on mobile and the vast majority of PC games would play like shit without them. Even strategy games like dungeon keeper would have to be completely different. Also you can probably just rip the source code/assets or find the code the original makers copied off the internet, and copy that. Plus a lot of android games are basically simplified flash games (keyword simplified because even PC flash games probably have 2-3 buttons you need)

I'm porting my game to Android alongside the PC version, as the game, nor the controls are too complex and it could work with little hassle. (it's an RPG-maker styled game) Though the Love2D Android compiler is a bit iffy. It works fine for the most part, but there's something with Elseif's specifically that's buggy. I haven't been able to reproduce the problem but a piece of code with quite a few elseifs worked on PC but not Android, so had to rewrite it all.

That, and I'm focusing mostly on main mechanics of the game itself now, and once that's done, I'll implement stuff like actual mobile controls, as now the only control set is the Accelerometer for movement, and that's purely because Love2D automatically maps the accelerometer as a joystick, so all it took a few sensitivity adjustments.

Considering you can use a variety if engines for both platforms, that's kind of a difficult question to answer. For the most part, no.
Because the PC and console ("core") market and the mobile market are two different markets with little overlap. Mobile games don't sell very well on PC and "core" games don't do well on mobile, either. Mostly because the mobile sector consists of casual gamers that need something to play for ten minutes while on the train. Look at the games that dominate the app stores. They're all really, really simple, instant-gratification games that any semi-talented Unity developer can fart out in a week or two. (That's, incidentally, how they're developed. You can read as much on the Unity forums.)
You won't find a sizable consumer base of gamers there, because those people haven't spent hundreds of dollars on their expensive phones in order to game on them.
You'll find the gamers, the enthusiasts, the ones who care about complex games on the platforms where hundreds, even thousands of dollars are spent on dedicated gaming gear.
If you make games for the mobile sector, you're making games for people who play games because they don't have anything better to do at the moment.
If you make games for PC/console, you're making games for people who will make time to play.
Because it's damn near impossible for the most part. Touchscreens are ridiculously limited.

Menu autism continues

lol posted this in the Deus Ex thread by mistake

Looking real good man. How far do you figure you're out from release anyways? You have such a large base done it seems.

we'll see how much can be done before demo day, I'll be able to judge then. There still is a heap of things to polish and add. Like itemization, which will still be a ton of "placeholder +50" stats on 02/02, unfortunately. **expect
new skills though**
I was actually hoping to have most of the game/base done in 2017 but it wasn't the case, with more and less obvious factors contributing to delays in development. We seem to be back on track, though so wish us luck.

Moral of the story: Don't go to game events. They're just for getting cool with the San Francisco indie circle jerk although his games might have just been shit

Not sure about 2px outlines.

I watched that talk, it seems geared toward party/arcade games.
The real lesson is to get your game shilled by a big youtuber.

That seems more like a matter of luck since probably every dev in the industry is shilling to them.

made progress, babby's first random map gen

I'm not pleased with the results so I'll probably spend a lot of time later making this not shit

It's probably true though. Long time posters on various Holla Forumss typically only watch E3 for the lulz and cringe factor. They don't go to these events, and normally the games that are showcased there are not interesting in the least.

What's true? That his game is shit, it's only for indie circle jerks, or you can make connections?

Interesting idea. I'll write it down and see how it comes out in practice. Thank you user.

I'm no expert on pixel art, but the placement of the eyes and nose might be a little off, and the outline around the eyes probably doesn't help. With the eyes boxed in with the black border they immediately get my attention and the very next thing I see is the nose and mouth which suffer because of the lack of definition around them; not to mention the large blocks of skin tone around them. If I were you I'd look at other examples of pixel art and see how others approach the problem of representing a good looking face.

Personally I prefer it with the eye-liner (assuming it's eye liner)

weird nose and belt line tbh

Good, I like doing things programmatically. That'll be good for when I actually pick up Godot

It's like I'm really on Holla Forums

I quite like your tileset. I assume you're using whatever one you could find, but it still has a nice quality to it, for some reason. Are you randomly rotating or flipping the drawn tiles, or do you just pick a set of random ones to use?

Which sort of mapgen algorithm are you using? I've seen some that use polygonal bounding and it went way over my head

anyone wanna collab with an artist to remake urban champion?

the idea was that it was supposed to be eye liner, that's specifically what the bottom line was. here's the other attempt to make a neutral face. It looks perfectly fine off the face but the second I put it on the face it turns into that.

I quite like 2px outlines, I would try out some 1px ones but all my shit is in 2px and I don't want to have to throw my work away.

so where's the game?

hint: the best random maps I've ever seen pre-built blocks to them, so you can have sensible touches and details to each block, but they overall structure is still random
doing actual full random map gen seems pretty complicated and specially hard to look good

isn't this the thread where we talk about shit we'll never finish?

For most anons, yes. But will you give up and let your game die? Or will you become a yesdev and finish what you started?

Fuck my ass.

lube or spit?

I found time and motivation to reanimate the standing kick series and add new followups, since I noticed that I accidentially drew the regular kick too large during testing and the charged regular kick just looked bad.

Aside from that, scripting the player moveset is ~90% done. After I finish that, it's mostly just fixing bugs that shouldn't be in a release version and adding a title/option screen. Judging by current progress, I'll only be able to release a training mode-tier tech demo with a finished player, but only a sandbag enemy for next demo day, since I might not be able to add basic enemy behavior and player hitstuns in time. Not sure if that's up to my or demo day's release standards, but I do want to release something to see whether there's any issues with distributing the game at the very least, maybe just not as an official entry.

it's my dream project, of course i'll finish it. I just want it to be good.

I'm
why not make the sprites in blender?

It will take you much less effort and will look better.

trust me, you will quit at some point.

I have a passion for pixel art, 2D animation and such, so I can channel my autism into it, one animation at a time, and have fun drawing and learning. I'd quit if I didn't use sprite animations.

I'm following a roguelike tutorial that was built for Godot. I found this project on github ( github.com/YeOldeDM/godot-roguelike-tutorial) that contains all the source code and graphics, and it also contains mini projects of each step broken down iteratively. The problem was it was built for an earlier version of Godot and I wanted to use 3.0 for C#. So it was completely broken. I'm in the process of digging through the code and refactoring it, learning as I go.

As for cycling the tilesets, I have several tileset familys (sets of walls and floors that go together) and right now I randomly choose a family when starting the scene.

I'm not using a formal named algorithm for map gen at the moment. It's just a shitty brute force method of generating a bunch of rectangles in random spots on a map and connecting them with corridors.

...

There's no rule against releasing demos whenever you want.

agdg police will get on your ass

Still working on these swords. I like the tracking well enough but if two swords are following from the same side for long enough they have this tendency to start overlapping and becoming one unit. Still trying to find a lightweight solution to this, maybe I just have to randomize the tracking position a little bit. It's more of a minor issue anyways since they're too dangerous of an enemy to have multiples of in most situations, it gets quite ridiculous when trying to platform with 2 or 3 chasing you,.

Is it possible for a corridor to connect two rooms but also jam through a third room? Or are you doing spatial stuff? Can your corridors do HVH corners or what not or are they just a single corner?

And they're right to go into indie. Game programming is like carpentry. Doing it independently is so much better once you find your market and build a reputation. Which isn't that difficult if you actually take the time to make something that is good.
Fun Fact: Every indie who talks about failure and how hard it is to stand out in the industry never made anything good, only just barely competent. The stories of how hard it is to become an indie success never come from the people who make games like Darkest Dungeon, Nidhogg or Rogue Legacy. It comes from guys who make mobile games in saturated genres or 16-bit "retro" platformers.

It's more likely that his game was just shit. Cuphead was almost completely unknown until it showed up for all of 10 seconds in Microsoft's Indie Reel and the fact that it looked rad as fuck is what generated it's audience.

Rather than moving them constantly until they reach their attack distance, make their moves segmented, so they move 30px, then stop, then 30px, then stop. You'll have to play with the distance moved and time stopped to get something that feels good. Also, make them spin for a 1/4 second before attacking.

Do you mean like they pause during their normal movement every now and then? I feel like they might be too easy to escape by simply ignoring them if they did that. I suppose I could have them speed up if the player is further away though. I'll give that a shot, it does sound like a good way to get them to stop overlapping since it'll offset their movement. As for the spinning before attacking thing, that sounds pretty cool. Right now the tell is that they stop moving and the eye on the sword narrows.

The problem is they move towards you linearly and consistently. I'd suggest that once they get within a certain range of the player, they pause, record the player's position, then move really fast towards that position while in the attack state, then pause again after they reach that location.
This way dodging one is no problem but getting sandwiched between two like in your clip will be very stressful.

Look up CV COTMs Bloody Swords

They can also float over walls

I actually have them do just that when they attack. They record the position they want to hit, then move towards it. They also wait there for a little bit until their attack animation finishes. However, I just implemented 's solution using a really short timer and it works great. They pause every so often and I have a variable thats tuned by their distance to the player that accelerates them if they're too far away.

I just tried having them pass through and that looks a lot better and is a lot more fair since they have less of a chance of hitting you twice.

Thanks for the suggestions guys it cleaned them up a lot. The behavior is a lot more crisp.

Boys I might be retarded here.

I have forward declared the World Class in my GameObject class (which is a base class for my Player)

In my player class I include the World Class (so I can call the resolveCollision function). Im getting those errors "undefined type" but I'm not sure why.

I do include the Player.h in the World.h but I thought the compiler was smart enough to not see this as a circular dependancy? I've played around with removing the include of the Player.h in the World.h class but then I have to forward declare my different components for their constructors and that still doesn't work so I'm at a loss.

Just to clarify the line its complaining about is the "world.resolveCollision(&object);" line even though it says line 77 it means line 74 (I have been making some changes to try and figure this out).

Never mind, it was a circular dependancy. I split my code from my Player.h file into a .cpp and now it works. Compiler / Linker treats them differently

Huh that is weird. I just looked through it and I can't seem to see what it is either. While I'm familiar with the stuff you're doing from some school projects, I don't use em enough to be an expert though. Maybe someone else can help, but this is a little beyond my ability to answer lol. But a quick question ,why in the second image do you have "class World;" above your class definition for GameObject?

But it looks like you found out right as I posted anyways. At least you were able to fix it.

Never mind I'm retarded, you talked about it in your post.

...

...

...

what can you do, at least they are not as bad as TED talks. I have a rule of mainly watching veteran devs who I'm sure are have a lot of interesting experience to share, even if they are cucked and/or simply rambling about their careers. Barton's channel is way better for that than most GDC talks.

Oh, and you know what? There are thigns way, way worse than SJW devs telling you about your priviledge.
Incompetent devs.
Just watch this piece of shit. Unbearable.

...

Hey, what are the names of the libraries to include when you compile your ogre3d game ? I can't find shit.

Nvm found it.

yeah fuck off

right now it's only a single corner. Corridors are only built to connect the last two rooms together.

God fuck someone end me, animating these big sprites is pain.
It's gonna be worth it at the end but oh god.

Can someone explain how to get this to work?

answers.unity.com/questions/700211/inputgetkeyup-for-axis-input.html

If I set my axis as "Key Or MouseButton" then I don't get inputs for the axis anymore so how the fuck is this supposed to work?
Am I being fucked with?

If I change it to a key or mousebutton, does my axis become a button of the same number of the axis? How do I find out what button it is if not?

Why though

lol okay retard

what do you wanna do? Sounds kinds strange, especially that getbuttonup only works in the frame the button gets released

Steamspy says it has sold almost 700k copies but maybe this talk is before they sold their soul and did tie ins with shit like Rick and Morty.

I don't disagree that the editor can be finicky at times, but it got a decent completion
and besides, the neat thing about it is that its wholly based only on itself, its super lightweight

I need to know if the user releases the button, but also use that button as if it were constantly being held down in another situation

I know I can make a separate input class to do it myself but I wondered if there was an easier way such as the one described in the post.

A friend of mine said that his game did see an increase in demo downloads after having a booth at a few events. He got interviewed a few times as well, but didn't make any connections despite meeting a few relatively big names.
I think it definitely does help spread the word if people want to play your game and see others having fun with it. If it's trash then it's obviously not going to help.

Unrelated, but I really dislike GDC talks and the GDC in general. Every time I see someone link a talk it's something like "Your game sucks" or "You will fail because", held by someone hugely successful like Vlambeer. It's like indirectly discouraging potential competitors from trying to get into making games. Not to mention the whole thing is a San Fran circlejerk.

I really think the industry needs a smaller version of GDC. I'm optimistic this will eventually happen now that indie games have gotten to a point where a lot of devs can actually create quality content consistently.

The important is moving away from the US honestly

Also this.

I love that slide.
Like, it starts off giving it a logical reason as to why Marxism would be useful. You read it and you go, alright I might not agree with the ideology but determinism could be useful in game design. That's probably what the guy means.

Then you read the next point and your brain doesn't quite grasp it and it leaves you confused.
Then the guy goes back to the original point as if the second part wasn't even mentioned.

Really special

Spreading it out is a good idea, both as a different event and in a different country. Who knows if or when that'll happen, though.
Germany has Gamescom but I think that's starting to become too big and isn't exactly in the best country. We have next to no domestic indie devs here.

Yeah, this kind of shit as well. Shoving politics where it doesn't belong.

I have no tolerance for cringe and don't think I could bear retards who ruined one of my favourite game series talking like they have authority.

I think the "X things I hate about your game pitch" was good, because it gives you a perspective of what publishers are looking for so it's worth digging for the diamonds.

You're not wrong. The majority of the talks are by out of touch San Francisco leftists or incompetent hacks who literally only have a job because their major studio controls most of the market share.

Gamescom is more of an european E3 I think.

I don't watch GDC talks anymore, but taking that one as an example I think a title like "how to pitch to a publisher" would be more effective in the long run. You can still warn beginners of potential mistakes but you ultimately want to encourage more people to try. It's easier than ever to become successful as a small developer, so the negativity those hackfrauds force is really bothering me.

At this point it is, yeah.

TBH I prefer the harshness. It's the one place where you wont be treated like a special snowflake and tell you how it is. Honestly, they probably word it like that because they get so many trust fund kids and cucks that they have to be bold and to the point.

the fucking piece of shit looks old enough to have lived in the cold war and know what communism IS.

can't you use input.getaxis instead and check whether the value is 0 or not?

The problem isn't getting the value the problem was that I needed to know if the button was released or not.

I found a workaround anyways

shouldn't you be using input.getbuttonup instead of getkeyup tho?

That's what I was doing

your shit is way too close up. I know you spent a lot of time on your sprites but you need to make it smaller, or at least increase the number of visible cells.

can't really test this now but have you tried creating a new input and calling it something different but having the positive button remain the same as the axis button?

Yes, but I have no way of knowing if the positive button is the same number as the axis is.
Like, I'm using axis 10, but I can't find any info to confirm if it's called joystick button 10 or something else

do you have any debug logs set up?

wiki.unity3d.com/index.php?title=Xbox360Controller
Right Trigger then? I see no reason why it can't be set for two separate button inputs, one set to axis and other to key press, as said before and make both use axis 10

This did not work.

What do you niggers think of having a seperate thread specifically for demo days? Specifically for non-agdg-anons to see the shit and provide feedback. We could submit our demos to the regular thread and then one user could make the "Demo Day" thread with all the demos linked in the OP.

post screenshot of input in inspector and code. I really wonder what might be the problem

that's what was done the last time. It was even stickied. The only problem was that there was irrelevant /agdg/ discussion happening as well.

Well shit. That's cool I was awol then pls no bully

We did a separate thread last time. I think we should still all post our games individually though. I'm sure most of us need to explain what we'd like people to test, what requirements there are, controls and so on. Having an anchor post or something, like they do in draw threads, would help make them easier to find.

The Anchor Post as second post last time kinda worked, but people kept posting their demos without quoting it because it seems not everyone here is well versed in imageboard etiquette.

time to catch up with the backlog then
8agdg.wikidot.com/agdg/ thread archive

web.archive.org/web/20171222135052/https://8ch.net/v/res/13774137.html

demos were good

Woo, I might be not useless for a change.

why

they were not shown in Diablo, I think

But that's why I post this stuff, so you can give me suggestions like that.

I'll have to remind myself how decibels work (they are logarithmic, right?) to have the numerical values for the sound. I'll add them, though,

Showing the numbers just leads to autistic min-maxing. I'm glad there are none.

Autistic min-maxing of brightness, contrast, gamma, audio effects, and UI scale?

You probably shouldn't use decibels for sound, because that would imply how loud the sound is coming from the speakers, which depends more on the OS and speakers themselves. Just a 0-10 or 0-100 scale for the majority of them is probably good enough (or even -10 to 10, with 0 being default).

That mouth placement was triggering me.

wow people actually post on this chan?

calle captcha is gone so 4chan is now dead forever

was it your base intent with your game to have the view be so zoomed in? because it looks extremely claustrophobic and uncomfortable to play

No, they don't. Fuck off

Wew. I really hope he doesn't post here.

I think the biggest problem is the HUD taking up 1/5th of the vertical screen space. Changing that might make it less claustrophobic.

That too, I asked if it's his intent because it looks almost like the gimmick he would use to market it

"This game is uncomfortably zoomed in, prepare 2 die"

I've been thinking about either getting rid of the top thing altogether or increasing the resolution a bit. I like having it there but It does tend to have this effect of squishing things down. Makes it hard to plan vertical challenges.

Personally I'd put it at the bottom. Blocky HUDs like that generally feel pretty "heavy". If it's at the top it feels like it's looming and about to fall onto you, but at the bottom it would feel like your character is walking ontop of them. You have a lot of unused space there anyway.

Seems kinda unnatural to look down at your health though. Almost every game of this type ever has that information at the top.

It's more natural to me than looking up.

...

You get a grip!

I think bottom HUDs are make more sense in handheld games since you are naturally looking down at your screen.

lmao. Well I guess it probably depends ok n what games you played the most. I just remembered stuff like Doom and Diablo had the UI at the bottom while Ninja Gaiden and Mario have it at the top. I have too much of a one track mind when it comes to this stuff. I'll play around with things. I think a resolution change or a shift in the camera vertically will take care of it but I'll play around with it.

so they froze godot on 2.1.4 stable while making the move for the features changes of godot 3
don't they have a more recent version 2 with fixed stuff that doesn't fiddle with 3's features?

If you're mimicking older style HUDs you have to remember they were made for 4:3 screens. Vertical screen space is at a premium for 16:9.

That's true. I think I'll try removing the black background and accomadating some more transparency by shrinking some things down a bit.

rate my topology

For the people who were discussing my UI with me:
what do you guys think of this mockup? Better? I haven't done any reshaping or redrawing here, all I did was remove the background and reposition a few things. I also shrunk some elements like the player and boss life down.

yet this is what Unity uses in its volume mixers… what can you do. I'll do some experiments and give it a percentage value.

in-editor

Decibels are logarithmic, +10 decibels is about double the perceptive loudness, and -10 decibels is about half the perceptive loudness. You just need to convert from Unity's linear taper to something more manageable:
wiki.unity3d.com/index.php?title=Loudness

thanks, bro. exactly what I'd be looking for.

I seriously hope you guys use bitmasks for your multiple boolean values instead of like 20 boolean variables.

you should be cleaver with your important stats so that it's easy to tell what it is as well as out of the way enough not to get in the way of the player enjoyment tbh

anyone on godot 3 care to share his experience?
are at any real major changes, does it a lot of improvements interface/documentation wise, regarding old stuff?

Pretty huge change to the 3d rendering, it uses pbr now. Lots of improvements to gdscript, visual scripting (lol who cares), a ui facelift, nine patch support for tilesets, and 2d soft shadows are some of the features. I like it a lot.

You can't take those 19 bytes you saved to hell with you, grandpa

you need to fix that shit, user

not the quarter of gui on screen?

That's probably what's causing it, among other things

>tfw I won't be ready for demo day for the next two years

Always user. Almost every struct in my codebase has a "flags" variable for shit like this.

I want to make an online game. I have some general concepts down when it comes to multiplayer, but I want to implement it all myself because Unity's online is slow, expensive, and easily hacked. Any good reading material out there? pls respont

You know that you have to show us your penis first for that to happen.

be nice, user, or else I'll ask what dilation is

This is /agdg/, not AGDQ.

What's dilation?

msdn.microsoft.com/en-us/library/windows/desktop/ms737629(v=vs.85).aspx

STOP

Your wish is my command

2.1.4 was released in August, so it's not too old. After 3 stable comes out, they're planning to put out 2.2 for compatibility/final tweaks. If you're impatient, look for unofficial builds or compile yourself.

...

user NO

You know you're an awful person.

I don't need to dilate, so I'm relatively not that awful.

...

You should check out the second one.

fuck you. post wholesome talks

Thanks, definitely gonna watch this. Alexander Brandon is a legend.

I am far too afraid to open this.

You need bravery to /agdg/

It's worse than you think
dont open user

Tranny "vaginas" are nothing but a permanently open wound, and in order for it not to heal up, they have to dilate it regularly by sticking things in it and using anti-healing drugs. They are just some images of that beautiful and natural process in action.

this begins my journey into a 3D MMORPG

Brilliant. Love the guy

Yes, I am 100% not opening that now.
I think I'll just keep playing with some different resolutions.

What's so bad about this in particular? I understand that in a lot of old games it doesn't do that since the sprites are so small and the resolutions are high enough that the camera tends to cover the entire vertical space. In Castlevania 4, anytime the camera has to move vertically it does this hard "shift" instead of smoothly moving to maintain the whole "see the ground under you" thing. I like the smooth thing better so I'm curious whats better about the other method.
Regardless, I beefed up the resolution a little bit and shifted some camera control elements so that now the direct ground under you is visible when jumping. I also have modified the UI a bit along with this to make it feel less cramped. What do you think of this?

Hard shifts are better because having the camera move all the way up and all the way down for a simple jump is really fucking annoying. I don't think I've ever seen a good platformer that did that.

Did you make those graphics all by hand? What technique did you use?

Yeah, I do all the work right now by myself(code, sound, graphics). Only thing I can't really do even at the most basic level is music, still looking for the right guy to do this.

Doesn't Symphony of the Night behave the same way in any room where the vertical space is high enough that the camera has to move? And Super Metroid and those types of games? They might use some kind of threshold of movement before moving the camera though, I can't remember.

I forgot to answer your question though. I use asesprite. Before that I just tortured myself with GIMP. I would say about 80 percent of what I've done is done in GIMP using the pencil tool and all the new stuff is done in asesprite. I use BFXR for most sounds, though some are sampled using some equipment I borrow from a friend.

Thanks user

this really shouldn't be this slow

It's a good tool. I have heard that the developer is a toolbag but I don't know about it or care, it's a really good program.

I wasnt implying anything or being sarcastic if you thought that, it looks pretty good

At least it's free if you compile it yourself, which isn't very hard at all.

So I went looking for a nice C framework, and found something called raylib. I liked it at first, because at a glance it kind of reminds me of Monogame. But upon further inspection it looks like it's even more abstracted than Monogame. To the point it feels like a cute walled-in playground for children. I wish I had a better understanding of OpenGL.

I'm almost there!

Post some code. What exactly are you doing to raytrace that? Is it in software? Is it with a GPU shader?

Older versions of OpenGL are much easier to work with, and also more powerful than abstracted 2D libraries. It's a good middle ground between ease of use and power.

i am shooting 8 rays at once and my performance has not changed

i am using embree on my cpu

No That's not what I meant, I'm just covering myself because last time I brought the program up there was this guy who went on a rant about the creator.

I've heard this before, and thought about trying to learn Legacy OpenGL 1.1, but is that a good idea? Should I actually try something like OpenGL 2.1?

Id delete that and start again. There are so many errors, you use a 3 point to double up a loop when you could just move the loops above it down, theres a six point pole, what were you thinking?

Topology isnt exact but use some reference holy shit, even if you have a personal style of topology that you know works its still a good idea to use reference.

...

Throw them away and try again.

Not bad advice.

UR GAYME SUC DIK DUUUUDE now lis'in tuo muh gayme ideya ookies? do liok dis thig whare yous craft munstars alruite?

Not bad idea.

BUT dem liok yous gotus tu hite ur munstars wuth stick dat liok yous put stowns inum dat drip fram uher munstars

...

OpenGL 1.1 is the simplest to learn and has all of the features that a beginner needs. If you use OpenGL 2, you will have access to programmable shaders, and all of the other extensions adopted into the standard between OpenGL 1.1 and 2.1. You probably won't take advantage of the features of OpenGL 2 beyond the programmable shaders at first, if you are starting out, so it's really more of a question of whether you want to learn how to write shaders right away or not.

I'm pretty bad at math. Would (((college))) help?

everything past school doesn't teach you dick. you're just there to pass exams

Only if you major in math and stay with it for more than a year.

How deep does it go? Around here you have to take the whole physics chain to graduate with a bachelor's

yeah…. what was that about?

see or just click it faggot

Chances are that if your field just forces you to take some supplementary courses, like physics or computer sciences or something you can get away with not learning too much, and forgetting most things after you somehow pass. If you go full math you either can't avoid learning math thoroughly or you give up and quit. If you don't like math, you'll probably quit. You could also try to work your way through a textbook, but that needs major discipline since you need to actually work on and solve the exercises to get it, or simply try to properly do whatever supplementary math courses you might get in your major. Or you minor in math.
The thing is, all of this shit isn't really necessary for gamedev. All you need for gamedev is basic vector math and being able to implement other people's algorithms. It just helps sometimes.

it was deleted, that is why I asked
but from the comment I think I've already seen it

Can't speak about US colleges but the EU CS equivalent teaches and demands a decent amount of math, from the basics to mathematical logic, linear algebra and so on. Even if you forget everything, it gets you used to reading complicated definitions and formulas. Pic related is a random page from discrete maths from our first semester, but before starting at university i would've been immediately repulsed by the sight of it.

Shouldn't it be "Zyklen" and not "Zykeln" right in the text under Permutationen?

Just like watch Khan Academy.

Aseprite is good but the creator is a bit of an sjw

Seconding this. CS requires shockingly little math education at most schools and making games requures even less than that(for the most part). There are some areas of gamedeving that can get really math intensive, especially if you're trying to do something new and dont have many resources to work with. In that situation having a string understanding of math can really pull through for you. But moatly gone are the days where you needed a strong understanding just to get started.

Yeah, you're right

This was the ref, got anything better?

original content please do not link me to any examples of people doing the same thing but better

Aight boyos I finally into'd collisions.

Basically I'm doing a small force calculation based on GObject weight (which i plan on modifying l8r on with items).

Now I just need to fix rendering of objects that are occupying same tile to swap the draw call between the two of them so you can see what item or whatever might be underneath.

Fug wrong pic

Are you documenting the developmemt of this in more detail anymore, like on a blog? Id be interested in reading more about this since it looks like you're song your own engine. Its kinda beyond me and Id like to learn more about it.

That topology would work, it just seems stupidly inefficient at many steps, I only have a phone for a few more weeks, I can't provide any reference or good crit til then, try look at models actually used in real videogames, preferably more recent ones for good topology.

Neat.
Alex is a cool guy. His stuff back in the demoscene's pretty cool, too; went by the name Siren.

making a porcelain bowl?

or bitflags, and
because I didn't get any (you)

can I ask you godots for examples on how you handle loading stuff (game soft data and etc) and storing them and using it your games?

pic somewhat related, but as an example of what I was thinking of implementing on my own,
a intermediary node/class to handle the loading process, from data files to node/classes in memory, doing whatever it needs, checks and operations
storing those is another question, something that can be easily reference from most of the game, either a singleton or the like
got any hints, suggestions, use cases?

That dress is going to be a pain to animate around.

I haven't been documenting much outside my code comments but when I'm done I plan on releasing the code for anyone to take a look at, bully, or use.

The only library I'm really using is SFML which I started using after making my way like halfway through pic related.

I'm not opposed to sharing knowledge of the stuff I know about at least, but I'm still learnin and a lot of the anons here know way more and better ways to do things.

why can't things just work

Ded

Just learn modern OpenGL, it's really not that complex and it's way more interesting

The way I handle Animations is I keep an int rect array that I loop through / update the sprite on whatever sprite sheet I'm using.

pastebin.com/mf1TVvrb

Here's my code if you're interested.

What's a good resource for sound effects?

Right now, I could use placeholders, but the sites that host them are shaky at best, and either need a login, have limited downloads, or stuffed with shit people made in SFXR/BFXR (that's fine, but if I'm going to use those sounds, I might as well make them myself)

FL Studio seems likes overkill, and I don't know if SFXR is suitable to make what I want. What do?

Freesound maybe?

...

So if I'm trying to innovate on procedural destruction of in-game objects (Similar to UE4's destructible mesh system) and level geometry, should I invest more time studying math?

So… SFML is kinda neat

You can hit the ground running a lot faster than SDL with it.

Not much to show off, I've just been experimenting with a few SFML stuff and getting familiar with it.

It looks like the .NET version has everything as a struct, so I have to remember to be careful when passing shit around so I don't copy a fuckton of large records around accidentally.

I made a simple timing thing that measures the last 300 updates, and figures out the min/max/avg frame times. I wanna make that a little more robust, but not actually adjust the timing with this class, only measure it.

An unbound fps of ~2000 is more or less normal, right?

Well I won't pretend to know enough about anything like that to suggest it, but based on what I do know my instincts tell me that yes you should at least be pretty good at it if you want to do something new that's not just new on a surface, visual level but is rather new from a technical standpoint.

There exists some relatively cheap but good sampling equipment that can be had from places like Amazon and you can use free programs like Audacity to manipulate them for basic stuff. Depends on the fidelity you want though. If you're going to be downsampling a lot then this method is pretty good and can get you good results, but it depends on what kinds of sounds you need to make and how much you know about sound engineering. What exactly is it you're trying to do anyways?

Huh well I'll look forward to it then. I just really like reading about people's though processes as they're doing this thing. That one robot platformer game that a guy posted around here that got released on steam and was made in a custom engine had a website with tons of details, it was fascinating.

Can someone please explain how custom signals work in Godot? Specifically asking about C#

Is it possible to declare a signal and handle in one Node's script, and emit the signal from a different node's script?

I'm able to get it to work if I declare the signal, connect the handler, and emit the signal all in the same script, but that seems redundant and not useful. I'd like to be able to use this to pass info between different nodes

added ability to switch between my shitty mess(Rendertape) and embree.
the performance difference is negligible, so i'm going to work on an OpenCL renderer
also it's really finicky

Simple sounds for my game. I'm too cheap for a foley artist, but at the same time, don't want to ship with free sounds, so somewhere in the middle I'll end up making some stuff I think

godotengine.org/qa/19389/how-to-make-a-custom-signal-using-c#

According to this, it might still be a pending thing to be added, as of Nov last year

That's the entire point of signals. Might not work that way in C#, though.

hooktube.com/watch?v=l0BkQxF7X3E

brought spheres back under rendertape, still broken under embree

I found out that when you create a signal via AddUserSignal, it creates the signal only locally to the object. So you need to take that into consideration when connecting signal handlers. I think I've got a handle on it now

we're all gonna make it brah

i'll get this eventually

I remember trying out with custom signals at some point, it obviously needs to emmited from itself on the same script, but you should be able to connect to somewhere else

I mean like what kind of game? What kind of sounds are you needing? High quality realism or arcadey sounds? That kind of thing.

spheres almost back in embree

what is your rendering code like? Is it a GPU shader or CPU software rendering?

software

Also, have a fantastic SO answer about cleaning up unmanaged C# resources
stackoverflow.com/questions/538060/proper-use-of-the-idisposable-interface

signals also has bindings, which is just an empty array by default, with bindings you can pass useful information from one node to another.

Does anyone have any resources on the mid-level parts of game networking? I'm comfortable with higher level networking concepts such client side prediction and I'm fine with the UDP library I'm using, but I don't know of simple ways of doing the stuff in between.

One problem I need to solve is clients and servers sharing their settings and info. For example, a client's name and profile picture. In some cases, this data may be very large. Some parts of the data will also be updated after a connection, meaning it isn't necessarily all sent at a single time. For example, if the client decides to change their name during a match. Additionally, because this data might be too large to fit in an MTU, it should be fragmented into separate packets. In the initial send when the client joins, I'd like to have a progress bar showing speed and progress. The Unity's UDP library supports fragmentation, but there is no way of knowing speed or progress when sending the data. Basically, I want Source's "Sending client info" box, ideally with a way of limiting the sending bandwidth so it doesn't screw with the match in progress.

Another problem I have is sharing other data that doesn't necessarily fit into a "send input, receive state" paradigm and isn't necessary to update with every server tick. An example of this might be the name of an object or something mundane. Basically what I want is Unity's SyncVars or Source's NetworkVars. Some problems I face here are: 1. variables that belong to a specific entity, 2. variables that don't belong to a specific entity and more a specific client (might be something that fits in more with the above paragraph), 3. variables that are client authoritative, 4. keeping all this shit updated for everyone, 5. identifying each variable uniquely across the network.

I could do this all poorly but would prefer knowing 'proper' ways to deal with this stuff.

From what I understand, streams are more or less an array of byte[] but of an undetermined size.

I know that's not quite what you're asking, but it greatly simplified how I thought about files, streams, and IO, so I imagine reading packets is fairly similar

I fucking love player sprites that change depending on the weapon you're using.

so this is something I don't get about Holla Forums right
/agdg/ is like one of the fastest threads until it goes over bump limit, then it goes ded
whats the fucking deal, is majority here just shitposters?

The slide to page thirteen is a sacred time for reflection and progress, in preparation for the eventually death and rebirth of the thread.

I imagine that when it's on the frontpage anons expect to receive thoughts and suggestions from nodevs as well as get some exposure for their projects. I read over here that cuckchan is even worse, with the threads only getting good around demo days.

so its like the Holla Forums conundrum

lemme also ask in the calm before our demise
what does /agdg/ like to hear on the background to help concentration?

for me there are some few that really get me going:

Skullgirls OST and other jazzy stuff works for me. Sometimes I listen to asmr as well. Rob Zombie if I want to get pumped up and get shit done.

weird tracker shit
i'll just list what i have currently open in OpenMPT:
Scorpik - My Styles
Vesuri - Luminenssi
AceMan - Polar Gleams
Necros - Isotoxin
Filip Kupsa - BB

Good stuff, although out of these only MCMXC AD and Tubular Bells get me in a mood to work.

thanks for the recommendations
Oh yeah. when I need to get in a transe, I play stuff from good, old Staylight, as mentioned by Alex

I have this in my car - the version I ripped from .umx.

Other than that, stuff from Matt Uelmen (rather obvious considering what I'm making) and Mark Morgan really helps me concentrate. Good ol' Neotokyo OST and SCDA soundtrack by McCann are also great for that.

When I'm not feeling picky, I just play whatever from NewRetroWave or LuigiDonatello and let the playlist continue.

Oh. And for some unfathomable reason I have spent countless hours working with embed related on repeat. It just works on me, despite not being the best thing in the world.

I cant wait for the demo day. I'm already ready, just cleaning things up and tuning the difficulty. I think people will like the changes I've made in response to feedback. Itll be a pretty long demo too.

Okay somebody gotta help me with this shit, I need some spiritual advice with this shit
developing in godot basically what I'm doing is text/menu based navigation system game
my idea was to have basically the majority of "content" side of thing written in data files (you know the less hardcoded stuff the better idea)
and this is the basis of the game design of a lot big content games where you don't exactly know where stuff is gonna go (eg civilization/rimworld)(and this stuff also enables easy modding)
so I just needed a loading process, right? during that make all the valid checks, create nodes and stuff, add hierarchy, not that big a deal right?

the problem is that basically with that process I lose basically of that godot convenience, because its so heavily focused on node handling and scene editing
so my alternative would to just make everything as a scene resource by hand and end the insanity, I can just use inherited and instanced scenes right?
well kind of not, this part also puzzles me, because even if I'm gonna make a ton of menus almost exactly the same, they will all have different attributes, connect to different things and whatnot
please /agdg/, this is giving me mind knots to decide, give me some advice

I blew away what I had from December and jumped to SFML,let's see if I can make something in a week 😂👌

Bob Ross when modeling and texturing, Aphex Twin and Boards of Canada sheen programming.

I mean I don't know anything about Godot or how it handles things like this(i wish I did, next project i do is gonna be 100% in Godot, seems awesome) but I imagine that sticking to the engines intended behavior(sounds like the acene resources udea you describe) is going to be the less headache inducing option. My general thought process is unless its necessary for performance reasons then don't stray too far from the tools you're being provided.
But thats just my inuition, take it with a heap of salt. I know there are godot guys in these threads so just ask again in the new thread when it gets made.

thanks user, god bless ya
godot is pretty good to be honest, there are some caveats though
the differentiation between a node, class and resource, for example, I was having such a hard trying to figure why I wasn't able to just instance some custom class
but it turns out some of the core node function can only be accessed if node is already part of the SceneTree
the thing is does allow some really cool and complex instance and moving of those things around,
but it does have a somewhat lacking documentation (where most contributors rather improve on code than documenting),
so if you stray too far from what its expecting you to do it can get bad

In my case I'm really more worried on how make this "dynamic" content without too much constraint,
I'm really wondering about how much people developing here really puts on soft code

Page 13!

NEW THREAD:

Well it depends a lot on the type of game youre making whether or not you do a lot of softcoding or not. I'm sure most people use softcoding for menu options like graphics and controls and that sort of thing and save em in a config file. But like in my case its just a simple platformer, there's not really a need to use external files to hold information because there's nothing really to hold.

Fucking rip in piss

The fuck happened here

this is the thread now:

Apparently clicking "delete post" on a post you make in your own thread, deletes your entire thread. So, I tried to delete one of my posts on the thread and it nuked the entire thing.

You sure you didn't delete the OP on accident?

Yeah, i'm sure that it was a post I made.

NEW THREAD FOR REAL