/agdg/ - Amateur Game Dev General

>what do compsci people do at their job
go to meetings to figure out what features we need, talk about how to implement things, implement things, and fix bugs. it's pretty straight-forward.

what kind of programming you do depends if you're a Java engineer, UI programmer, web developer, network engineer, tools developer, etc.

Do some interesting math stuff in games.
I love stuff like that.
Or shader stuff.

I maintain and update code. Basically, we have a git repo of our entire web application, and it's my job to add new features to it without breaking everything else. It's alright I guess.

My day consists of sitting quietly at my desk listening to music, messaging my work friend on the internal Skype For Business client, and working on feature requests in the current development cycle. I have very few meetings and no one really cares if I'm MIA for a long lunch (I usually take 1-1.5 hour lunches and have a beer during them) as long as I get my work done.

IIRC GraphicGale's free version isn't complete, like, you can't animate stuff and such. I wouldn't recommend it over Aseprite considering googling "aseprite download" gives you easy and legit download links.

>look at how good the game look and tell me
Do you have autism? I have a friend with Asperger and you remind me of him somehow.

I really like architectural stuff, which I think is underrepresented. I'd prefer to stick to that kind of topic if possible.

I could talk about serialization, coroutines, threading, debug console, stuff like that, I suppose.

Co-routines or threading would be nice as I'm currently looking into them right now.

Yo I reread your latest one and tried to implement that kind of thing when I'm rewriting my roguelike and I'm basically really confused on why it would be better to structure my code this way. With pure ECS I can rely on the systems to do the heavy lifting in an almost "blind" manner -- I make one call per system to work on the components necessary. But the way your new article says to do things, I'd be calling each system multiple times for each "tree" in the forest, which seems to me like it kind of muddies the flow of the game loop a little? You did mention that it takes more game loop code to do, but it seems like the overall idea is instead of entities being abstract compilations of components, they're set up as "trees" which are basically set-in-stone components, and then optional modifier components, with logic still removed from them and kept in systems.

I didn't understand the last paragraph or two of the article. What did making a class called "props" which is an array of a few different components accomplish exactly?

I'm jealous of ur smarts

By myself in the office today (aka my garage) because my right hand man got a temp job working for Bob Katters Australian Party for a few weeks. So I come to you guys for replacement bantz.

A serialization or save/load thing would probably be the most useful to people - the pseudocode is pretty portable. This is a really neat set of code in Unreal that uses the same function to both save and load by overloading the

>Working on blender
>Feel like hot shit, making stuff from tutorials
>Thinking of game ideas and marking them down
>Talk to friend who also has been working in blender
'Yeah, I've been working with blender for a few years now, it's really labyrinthine, I can barely make anything still'

Is it really that tough once you stop with tutorials? So far 3D work has seemed rather simple, as long as you have a hang on what you want.

reposting my hanging pole mimic for the necromancer game (I just did the model and rigged it, I don't want to start working on the necro game and end up never working on hellball again).

>IIRC GraphicGale's free version isn't complete, like, you can't animate stuff and such
You wot m8 the only thing you can't do is save .gif and .ico. You can even animate anything then save as a spritesheet, or even as multiple files then import on some other program and make a gif, that is if you really want a gif because I don't see why anyone would use gifs in a videogame in this day and age.

it's a lot harder to make something when someone isn't telling you exactly what to do. it makes sense when following a tutorial but as soon as the guide is gone, you feel lost and don't know where to start/what to do next. coding is the same way.

there's also a lot of steps to 3d - modeling, texturing, animating - each is its own skill and quite a bit different from the others, and it's time-consuming to put them all together.

Merge the two games

I'm planning to make a bunch of my games connected by lore somehow

make the comic book path

same universe with diferent dimensions n shit

Yeah, I know. I suppose hearing that a year in, someone still hadn't progressed got to me a bit. It's hard learning that something your doing as a hobby is so intensive, but over time, I'll at least make some sort of progress.

>But the way your new article says to do things, I'd be calling each system multiple times for each "tree" in the forest, which seems to me like it kind of muddies the flow of the game loop a little?
No, the flow isn't messed up. You'd still do all the calls to system A, then all the calls to system B, and so on (you also could do otherwise if you wanted). It's a bit more code to put in the loop, yeah, but that's the tradeoff for being able to specialize and add a bit more structure. You could even do it by having an interface for each system, where the implemented method just calls the system (which is still a static/free function) with the components, if they're available.

>What did making a class called "props" which is an array of a few different components accomplish exactly?
That's a long story, but you can get the gist from something like this:
blog.molecular-matters.com/2011/11/03/adventures-in-data-oriented-design-part-1-mesh-data-3/
tl;dr it generally allows the CPU to use memory much more efficiently. You can benefit from this even in C# if your components are structs and you store them in arrays like that. But it's also not worth it to make everything follow this organization (because memory usage and iterations can skyrocket), so it's nice to be able to do it only for specific entities where you know it won't be wasteful, like I showed.

Any MusicBros here? If you want to make music for a game for free I need music for my sewers level.

Thanks friendo. So am I accurate in my summary of your proposed system being basically:

>Mandated Components
>Variable Components ("either")
>Optional Modifying Components

What game?

post your game

Basically, yeah. Instead of saying "every entity can have any component!" you split it up into buckets, where each bucket can require a component type, forbid a component type, or do something in between.

I'll try this out and let you know the results in a few days if you care. If you don't care, I'm still gonna try it out.

No, please do, I want to see how it works for you.

Have a reference you'd like it to sound like, sewer levels are a pretty broad thing and can be anything from drip drop ambient to Grant Kirkhope-core

Eh. At least it's better than pedo anime shit.

Radium has been our working title for a long while. Its a top down twin stick with some puzzles combat, platforming and exploration.

If this guy is around, I want to ask:

Why are u making a fighting game?

I'm part of the FGC (I follow EVO, follow fighting games scene, can name pro fighting game players, go to my local scene, start playing fighters in 2005, have SRK account and so on).

I just noticed you said you don't want to make a SF clone, so what's the point?

Fighting games are basically street fighter clones.

>Block button
garbage.

The standart is back to block.

block button eliminates crossups, mixups, 50/50 setups.

the reason why MK is garbage and considered a party casual fighter for normies is because of the block button.

Also your game seem to have awfull and slow reaction times.

I ask again:
Why do you want to make my favorite genre of videogames if you don't have at least a decade playing them?

But, user, you can win street fighter by button mashing.

strawpoll.me/10365817
strawpoll.me/10365817
strawpoll.me/10365817

all me

Honestly, I might need an idea guy soon. I'm not sure where to go with my game after the mechanics are implemented

>button mashing
if you're a begginer sure.
I can easily stomp you if you spam normals by simply spaming fireballs or choosing long range normals.

>I'm part of the FGC
>Street Fighter

LOVE2D user here. I Just wrote a pathfinding script. I'm hoping someone with Lua experience could take a look and critique it.

Anything I'm doing wrong? Anything I could do to make it run faster?

I've actually just rewritten it to be threaded so it doesn't slow down the main game thread, but I don't have that on github yet. Other than running in a separate thread, the code is pretty much the same.

www.github.com/danielpower/zStar

I don't play turd fighter.
It's boring.

I play SNK fighters.

What is the "correct" player collider in Unity to use for 3d platformers? The default cylinder is garbage.

>those dcss altars
>trog and xom

you have insanely good taste

Huh, my uni doesn't even separate them into different majors.

You probably don't need to make it threaded unless you're going to have many agents doing long pathing every frame.
It'll be a pain later if you have things like moving obstacles, inter-thread communication is a major hassle.

Wait, misread the comment. Didn't see word "game" before programming. Disregard.

I actually just finished the multithreaded implementation. It took me a few days to wrap my head around multithreading, but it seems to be working pretty well.

I start a new thread, pass it the few variables it needs. Then when it's finished the calculation, it sends a callback and gives the object the result.

I can see how it might cause issues though, because the result doesn't come immediately. For example, if my CPU is maxed, I can move around, and the pathfinding takes a moment to catch up to the player.

I think I'm going to stick to the single-threaded implementation until performance becomes an issue. But I'm still glad I put the time into learning it. It was interesting.

Yeah, I would recommend against ever doing multithreading stuff that can takes longer than a frame. Kick off some parallel jobs and wait for them all them to complete, yes.

Best way to do this is not work with threads directly and instead just use them to implement a job queue.

why arent there any proper engines for fighting games? MUGEN and 2DFM2016 are the closest we have... wtf?

Where is the "I am an artist that doesn't need anybody because engines make gamedev so easy nowadays that programmers are literally deprecated" option?

No such thing as a cylinder collider in unity. The capsule collider is the most commonly used as collisions are reliable smooth and simple.

alright first, sorry for my REALY bad english.

>Why are u making a fighting game?
because i love playing fighting games
im not realy good at then, not a scrub by any means but not ONLINE tier good,

>Fighting games are basically street fighter clones.
I know. the ones that break that formula of are the ones that i like the most. i love street fighter but every time i try a new fighting game i like i playing the same game with diferent characters.

stuff like guilty gear, tekken or even fight night round feels diferent

fight night is not realy ' fighting game" is a boxe simulator. but i aways wanted to play a arcade version of it, in the ways of street fighter and the such, thats were the first idea to my game came from.

>The standart is back to block.
the first thing i did was back block, and is still on the game
but a friend conviced me to add a block button because fuck it. the back buttons on the gamepad were unused anyway., so one blocks and the other one dodges like in that webm with the mirrors

>the reason why MK is garbage and considered a party casual fighter for normies is because of the block button.
the reason i think MK is shit is because every single combo is pre programed to the game.
the physics are too heavy and the moves dont actualy connect, you cannot do a free combo because that character only has like 3 combos that connect to expecific powers in a expecific way.
that feels "clunky"

>Also your game seem to have awfull and slow reaction times.
because i dont know how to use the unity animator thing very well
im still working on it.

the game "feels" completly diferent playing on a gamepad against someone else.
if you have a friend that likes fighting games, i would love if both of you tried the game.

I wish artists weren't all such entitled cunts.

>Fighting games are basically street fighter clones.
Is this bait?

This

Because there isn't an audience for it. It's better for an engine to just have lots of tools that will fit many game genres, rather than specializing.
The audience for fighting games just play Street Fighter, or Melee, or whatever else. They don't tend to be people that want to make their own fighting games, and they don't buy or play indie fighting games. They stick with one game for sometimes 15+ years.

RPGs and Visual Novels have specialized engines because those two genres tend to have more people that want to make their own stories. And the audience doesn't play just one game, they play a lot, they're always buying new ones.

I wish programmers weren't so poor

It's this your first game?

If so, maybe you need to learn to make othe types of games, get some experience.

if you want, you should try making a fighter on libgdx or monogame with kof or SF sprites, as practice and you'll gain deep knowledge on how they work.

It seems you're simply trying to tweak that awfull unity fighting game addon.

It looks nice, but it's shit.

make smaller games first.

I've started to make games in 2010 in game maker but now finally I can make my first fighting game, basically an urban champion remake.

If you make something new the FGC will cry it's not a fighting game, like smash or pokken or power stone.

If you're making a game for casuals, don't market it as a fighting game, focus on the single player rather.

the only people who play fighting games rather than smash or LoL are the FGC.

I think as long as you see this as a learning experience and don't get frustrated when it isn't THE BEST FIGHTING GAME EVER when you're just starting out learning to make this kind of game, learning Unity, and so on, you'll be fine.

I'd remove the running-in-place guy and the Jesus impersonator from the background, though.

It isn't fiction

he could make the animations and render them as sprites in blender and import them on 2D fighter maker or mugen if he's too scared to code.

It will be better than that shitty unity fighting game addon.

I didn't know there was a shitty Unity fighting game addon. I use Game Maker.

well shit nigga, we're more neet than most artists, what the fuck do you want from us.

why don't you use mugen?

Because I'm not , I'm , and I'm not making a fighting game.

oh well.

In any case, he's better using 2D fighter maker or mugen if he wants to make a fighting game.

its not a addon, i codded it from zero. what realy fucked the project were the 3d animations timming.

>It's this your first game?
not realy, i made some stuff before, like a DMC clone , a Castlevania clone and, my personal favorite a C# adventure game.

just testing the shaders for the last 2 characters on the game

its a small enough project, only 4 characters.

Good idea or bad idea: flak cannons now inflict friendly fire, but stop shooting when an ally gets too close to their target

Will also make it so they can fuck up and keep shooting anyways, and I'll slow down their reaction time a bit too

those dust clouds are annoying

Am I on crack or is that Advance Wars art?

They'll be actual explosions eventually, plus there won't be a dozen of them spammed in one place. They'll also probably be smaller since right now they're impossible to avoid even if you see them coming

The background tiles, yeah. I'm still learning to do pixel art so I'm using them as placeholders

...

They appear to be explosions.

Sounds cool. You could have it so you fly in front of an enemy jet which causes them to fly into a wall of flak.

>smooth and simple
How can it be smooth if it can't handle slight slopes without stair-stepping and ground detection flipping out?

>No just ideas guy option
C'mon now, give us some pure love. That's my idea by the way, where do I contact the strawpoll heads for cash?

Play with your friction values and physics materials, most people will swap out materials or change values if there is player input or not. Your issue is not with the physics of the capsule collider but more so whatever system you are using to handle player physics.

>RPG is too big of a project for right now
>don't like Shmups or puzzle games
>did all the programming for a platformer, made about half a level, then lost interest
I don't know what to do, /agdg/. I've done the Pong clone and the Tetris clone. I just want a project I'll actually feel excited about.

>feel excited about
Spotted your problem.

Don't worry, though. Everything in life will let you down under such a criteria. Humans have an amazing capacity to get bored by even the most exciting things.

Suggestion: KYS

pick an old game you really liked and make it better/3d/etc.

You know not all RPGs have to be Witcher Scrolls.
Not all SHMUPS are 2hu shit
Puzzle games have a dozen sub-genres.
>cam't even name other genres

Sounds like you have a case of shit taste

You should go cure that by playing more video games

>that suggestion
Why? This is literally the one thing in life with which I'm unsatisfied.
>better
That's an idea.
>3d
I don't really want to.
>etc.
Now you're just talking crazy.

I can name other genres, but I was naming the supposedly easiest genres (platformers and shmups) and the one I like most (RPGs).

I've actually never played any of the Witcher games, and I've only played about an hour of Daggerfall in the TES series. My favorite RPGs are the Ultima series, Secret of Evermore, Avernum, and Ogre Battle: March of the Black Queen.

That said, if you want to recommend some more games for me to play, feel free.

Pong RPG

What do you do when you've been trying to implement a feature for almost 2 weeks but gotten nowhere and kind of want to drop everything?

we farmin now

when I did that I dropped everything, quit for 4 years, then started again on a completely different project

Implement something else. Come back to it later with a fresh perspective.

The number of times I've spent hours trying to get something working, give up for the day and finish it in 5 minutes when I start again the next day is amazing.

Need more frames in your animation. And better art. 1-pixel limbs look so fucking bad, how did it ever catch on?

Revolver360 and Sine Mora to change your opinion about SHMUPs.

From Dust for puzzle
Tales Of series, the first one is still great.
Nox
Stronghold

But all those games you just made are great project options and are relatively programmer-art friendly as well.

Not seeing the issue here

I'm not an artist, and I can't afford an artist, so this is how its going.

we tried to tell him that the action animations were pretty stupid. he didn't listen.

Yes or no: Do you think your game is going to be better than Advance Wars?

Made a sonar shader
i.gyazo.com/95a25502fbed0f1d0caff15d5dd52fb0.mp4

fug yeah.
Make it so diggy spots next to each other will connect.