Amateur Gamedev General ~ /agdg/ + /vm/

POST PROGRESS EDITION

Resources:
>>>/agdg/
>>>/vm/

Other urls found in this thread:

datanon.itch.io/s4
mega.nz/#!1cIEhbiY!jISu8ZvnHY9Pgees0fCUvzP7Hc_O_OCSW6LZ-1GeRA0
ika.neocities.org/sigma_engine.html
youtube.com/watch?v=cUlSbZ_NTjs
sfml-dev.org/documentation/2.4.0/structsf_1_1BlendMode.php
pastebin.com/bXG5y1uR
adammil.net/blog/v125_Roguelike_Vision_Algorithms.html
pastebin.com/xLPsf2ax
pastebin.com/M7UGinH0
strawpoll.me/11220074
poal.me/lqpudv
cs.umd.edu/Outreach/hsContest16/questions/problems.pdf
reactos.org/es/node/219
p.trumpshare.com/nakj/dvn4.zip
mega.nz/#!ABJ2ib6Y!klgapAAkm_Rh2aTY2enrT13xqsVHmseH0MAQ7kd9NUk
gitgud.io/greenman/rts.git
opengameart.org/content/bone-texture-pack
twitter.com/SFWRedditGifs

datanon.itch.io/s4
pls no bully for >phonegame

help im retarded

i'm an ideas guy

move somewhere with a high minimum wage, and then use that money to hire an army of indians

two years later, kill yourself

I can connect them to shit, sure. But I'm too retarded to know what to do after that or what to really connect them to.

spoiler your pornographic images next time

Is he supposed to be a centaur or something?

Make enough money to hire a programmer(s if you're already a rich mother fucker). Learn how to actually take your ideas and break them down into individual game mechanics and component. Consult with the dev team you've assembled and establish reasonable deadlines. You don't have to adhere to them early on. But once you go public with your game, that's when you need to tighten the belt a bit.
Once you have all that in your mind just get to it and keep a steady pace. This is a heavily abridged version but if you follow these instructions you'll probably make something decent.

the guy from the right is a centaur, the guy from the left is a Faun

The idea is some type of mage that can stop stuff for one turn. If the enemy has a buff, you can push it so he can't use it until after his next turn. It could also be used to stop debuffs quicker or make the enemies debuffs last one turn longer. Later on abilities would be elemental weakness time increase and giving a teammate a pause in time to do two turns in one. Though the problem is I feel it's too powerful. As for my game I debating whether to have it be one fairy just going on a quest that could have multiple abilities, or multiple fairies with singular abilities that fight each other because they each have a piece necessary for the goal. The enemies would change as with the singular fairy would have enemies along the road where are the multiple fairies would have enemies based on allies to the other fairies. If I do multiple fairies I would probably have stages be all the other fairies you chose not to play as.

Dude unspoiler that shit, its not like you are posting porn.

...

this is more like idea guy general

Oh yeah? I'll have you know that my Unity prototype is already shaping up to be the Next Big Thing, here's screenshot proof

Good for you, I hope you make loadsamoney from your unity project.

Alright, I finally can get a new build of this out there.

I've spent 3 or 4 days working on bug fixes, I forget how long.

Anyway, all the bugs that got complained about should be fixed. The game isnt so broken now.

You can download the newest build (0.7.3) here: mega.nz/#!1cIEhbiY!jISu8ZvnHY9Pgees0fCUvzP7Hc_O_OCSW6LZ-1GeRA0

My development site: ika.neocities.org/sigma_engine.html

reminds me of this

but with a lot less effort

I'll show off progress when I get past this current roadblock.

I was being facetious. I'm never gonna finish anything.

Hell, who am I kidding? I'm never gonna start anything.

So what are you actually making with your engine?


So was I.

Not videogames. Fuck off gook.

well its a game where you walk around rooms and shoot people that try and shoot you

and you have to find mice and collect them to get to the secret levels

There are about 13 diffrent enemies that try and stop you, and they are all diffrent in some way.

The levels look real industrial. I think you're in some kind of industrial setting? I didnt think up a story, but there are a lot of pictures of concrete walls online to use as textures.

I think there is going to be a boss you have to kill or something.

So its a fully-3d FPS game kind of like doom except unlike doom I can have areas over areas like bridges (which doom didnt have). So there are cool things I can do with level design, if I could make myself make creative levels…

Anyway its just a doom-like FPS game, I guess. I want it to have a story through title cards or something, so that its more interesting.

The tldr is that im not sure what im doing. Im just putting things together so that it resembles a game but I dont have any real direction or idea of what the finished project will look like.

well i made the first step

You should team up with that guy making the space pirate game and make a version that makes you cash money, just like charge dick all for it.

So I'll ask here again in the hopes I'll get lucky. Anyone used the UE4 AI Perception Component?

Got a problem where the AI will detect characters properly, but will always lose sight of grenades the exact moment they spot it, meaning they won't know that it's there to avoid it. No idea what's causing it as I've tried it with a few different size meshes and scales and the problem persists.

I haven't worked on mine in months other than a few tweaks to some models every once in a while, haven't even helped a friend with testing his game either even though I said I would. I've been working a lot and reading manga instead.

I dunno, guess I'm just not feeling it.

And that is what? Downloading the engine?

It's worth noting that The Completionist is a fucking centaur on the left there.

The engine he is using is much better than the one I made, so he would never be able to make something like his game on my engine. Its fundamentally flawed under the hood in so many ways.

Maybe with my next engine I will team up with someone, I don't know. I want to build a new engine that fixes all the mistakes the one I am using now makes, once I finish my game. A new engine will be able to have lots of new features and will just be easier to make games on.

My level editor does not have slopes built in and it crashes all the time. Sometimes when it crashes it gives me "corrupted heap" errors. I don't want anyone else to have to use it

POST THE CUTE WHITE HAIRED GIRL WITH THE BIG TITS NOW

NOW

There is more than one girl with that description.

...

I hope for your sake you make a good game from your engine.

Are you worried about me? What do you think would happen to me if my game wasn't good?

The engine part is the most fun anyway. Actually my last game was really bad, it ran on an engine that was kind of like wolfenstien-3D but in openGL. Maybe this game wont be shit because its in real 3D but who knows.


Embed related

Not at all but you look like you've improved a whole bunch, it's always good to see someone doing something outside of shitposting and baiting on this website.

That looks really bad.

its because I decided to make all the graphics myself and didnt really play test it with anyone.

Embed related is the full play-through. Even though it looks bad, I can enjoy playing it, but probably only because I spent a bit of time making it. I found its fun if you can get into playing it, but I dont know anyone who got past the second level… a big problem with it is that i made the levels in order, and your first attempt at level design is probably your worst attempt.

I am going to bed so I wont reply until tomorrow.

not too bad, i wish it was more forgiving

Anyone in here use RPG Maker or GameMaker Studio?

That's for later, right now it scales up with your combo, but once I finish up more of the basics I'll script the levels a bit better and change the scoring system. Pic semi-related, the new menu, half done.

Whoops my post got eaten. Reposting.

Been replacing placeholder assets bit by bit. So here's a new logo for the space pirate game.
The old version was made out of a whole ton of GIS'd images shooped together, the new version was drawn by a friend.

It looks like arse when shrunk down for in-game, I need to figure out how to clean it up for the menu.

Just redraw the lines in the new resolution. That should be fine as long as you don't go overboard and make them too thick.

Just resizing stuff usually isn't the best option, they need cleanup afterwards but you probably know that already.

What's the story behind Sharkboy anyway?

Particles?

Particles.

RPG Maker just werks. I'm too lazy to get out of my comfort zone.

A bit sluggish, but you are making a good progress progress.

Needs more particles

youtube.com/watch?v=cUlSbZ_NTjs

Particles.

If you mean why I made him, it's because pretty much as soon as I'd decided "this is going to be a space pirate game", I knew I'd needed a crew.
Also, every pirate story needs a shark.

Unless you mean in-universe, in which case I'm only on, like, the second draft of him. Either way, the very tl;dr of the matter is that he's technically the kidnapped son of nobility.
Raised to be an elegant and proper prince, was taught a lot of cooking and literature and shit. Learned a lot about his ancestors being warriors back in the day. Found that far more interesting.
Couple accidents with some outlaws later, the Crimson Skull (the protagonist) saves his life and brings him along on her ship. He decides to live up to his ancestors and become a fighter.
He's an arrogant brat, but is super-excited to be shooting things in space and is completely subservient to the protagonist.
He also does a lot of cooking for everyone. He's very good at it.

Had the "stable" beta build of my game tested by several people on the interwebs, already fixing 2 critical and 7 serious bugs one day later.
Kill me, Pete

I use GM:S to make prototypes really quickly sometimes.


That needs LESS particles, holy shit.

anyone here got experience modding mount and blade? specifically with the module system?
i got some questions since some of the stuff in the forum tutorials seems to be outdated
if i can't find anyone here i'll probably just make a forum post

Sounds cute.

Anyway I'm super hyped for this game. I actually didn't look much into Doom modding aside from playing a few levelpacks. This got me exited and looking into it I actually learned about DemonSteele and all around it. It's really fun and I'm considering trying out Doom modding myself with a buddy. Thanks for broadening my horizon.

sage for blogpost

How do I get myself to narrow down the scope of my game so I can finally start making it. I want to turn it into so many different things but I know thats stupid shit

ok someone tell me if this is possible with sfml
I wanna use pic related as a texture so it just a transparent circle with semitransparent grey around it
I wanna put this texture in my game and i want the alpha to blend in such a way that circle remains entirely transparent
I want to use this for fog of war
So I cover the entire screen in semitransparent grey
Then I put this texture over units and want to circles to remain transparent.
So the fog of war would be like an extra layer over the rest of the game and the circles around the units would essentially make that circular part of the layer entirely transparent
So i thought i could use BlendMultiply or BlendAlpha but neither of those are working
Can someone tell me how this stuff works
I don't mind digging around the sfml source code and changing shit.

sfml-dev.org/documentation/2.4.0/structsf_1_1BlendMode.php
You want an alpha blendmode with the subtract equation instead of an add one.
By the way, your pic doesn't look like a circle to me. Wrong image?

weren't you going to be doing line of sight?

I've decided to make a top down 2D skyrim clone.

What do u think of top down 2D skyrim clone.


Make character, make enemy characters, make inventory system, make friendly non player characters, make dialogue system, voila you just made a perfectly decent game.

yeah afterwards ill just add blocked regions ontop of whats already their

does blendmode even work with textures?

yeah for some reason it didn't get copied properly o well you get the point anyway

My gut tells me that this can't possibly be a good way to go about implementing fog of war.

It's by far the best and most efficient way i can think of.

Why the hell is gamemaker html5 port so absolutely fucking shitty. Keep in mind, I use the pirated 1.4 version, but still. Shit framerate, buggy, some rooms don't even load, no loading screen, and it didn't work on chroma.

But if you're still going to have to calculate which side/angle your dudes are relative to vision obstructions for a final reobscuring step where you build up sane blocked off regions…. isn't that step already a proper fog of war system without the circles around dudes?

What do you guys think?

I've been working on this all summer. I'm trying to realease this for android but building this to work properly has been a nightmare.

it's been done before

Do you have an example of that?

not him, but it reminded me of solar flux, though a simpler version. looks like a neat project though

doesn't seem to be working
heres the code
sf::VertexArray test(sf::PrimitiveType::Quads); sf::BlendMode blendMode; blendMode.alphaSrcFactor = sf::BlendMode::Factor::One; blendMode.alphaDstFactor = sf::BlendMode::Factor::One; blendMode.alphaEquation = sf::BlendMode::Equation::Subtract; sf::RenderStates state(blendMode); test.append(sf::Vertex({0, 0}, sf::Color(100, 100, 100, 100))); test.append(sf::Vertex({100, 0}, sf::Color(100, 100, 100, 100))); test.append(sf::Vertex({100, 100}, sf::Color(100, 100, 100, 100))); test.append(sf::Vertex({0, 100}, sf::Color(100, 100, 100, 100))); test.append(sf::Vertex({50, 50}, sf::Color(100, 100, 100, 100))); test.append(sf::Vertex({200, 50}, sf::Color(100, 100, 100, 100))); test.append(sf::Vertex({200, 200}, sf::Color(100, 100, 100, 100))); test.append(sf::Vertex({50, 200}, sf::Color(100, 100, 100, 100))); window_.clear(sf::Color::White); window_.draw(test, state); window_.display();


no i need the circles since they don't have unlimited vision range

Looks like one of those flash games I played ages ago, except with less juice.

...

Yeah. I got this idea from a flash game i played years ago. Now I've made this (in python) to try and learn how to develop an application.

You could've done it in gamemaker in 3 days.

Yeah, but the purpose of this was to learn python.

How were you planning on calculating the reobscurening then if ranges don't have to be figured there?

You're subtract-blending 100 from white twice. You're supposed to have a "layer" (image the size of your screen) filled with gray at some medium alpha, then subtract-blend your circles onto that layer, then draw that layer over your game like you'd draw anything else.

go through every unit and check if there is an object that is in vision range
Create a convex polygon based on the object and the position of the unit
Draw the polygon with semitransparent black over everything without blending.
the only issue here is if you have several units near the same object with over lapping vision bing cut off.
Not sure now to fix this yet
I guess i have to check if the extra vision blocking polygons are inside the vision range of another unit.
Man this shit is gonna be difficult
Maybe ill forego the line of sight stuff altogether.

I am?
I guess i don't know what source and destination are in this context
I figure one rectangle get draw first with Alpha 100 then the other one gets drawn partially on top of it also with alpha subtracting its 100 so you get 0
What do I need to change?

something like pic related right? where that second pass would have everybody overwriting everybody elses vision if there's a wall between.

I need char portraits for lewd jam. How do I learn to draw cute anime girls fast?

yeah exactly

Pick one

I'm just guessing here while reading some documentation user, but as far as I see from your code they're both drawn with the blendmode active. So the only way that made sense to me was assuming it;s subtracting from white.


I assume he means learn "to draw fast", not learn "to draw" fast.

I was specific for a reason. I can draw kinda ok, I'm no rembrant but I do lifedrawing once a week and keep a sketchbook. I just have problems replicating that style.

I pick fast.

Learn to draw a perfect 3D ball, attach the 3D ball to a 3D human figure frame in various poses, then study loomis and fill the blanks.

whatever i try it just seems to be adding

this seems like it should work right?

window_.clear(sf::Color::White);

sf::VertexArray test(sf::PrimitiveType::Quads);

sf::BlendMode blendMode;

sf::RenderStates state(blendMode);

test.append(sf::Vertex({0, 0}, sf::Color(100, 100, 100, 100)));
test.append(sf::Vertex({900, 0}, sf::Color(100, 100, 100, 100)));
test.append(sf::Vertex({900, 900}, sf::Color(100, 100, 100, 100)));
test.append(sf::Vertex({0, 900}, sf::Color(100, 100, 100, 100)));

blendMode.alphaSrcFactor = sf::BlendMode::Factor::One;
blendMode.alphaDstFactor = sf::BlendMode::Factor::One;
blendMode.alphaEquation = sf::BlendMode::Equation::Subtract;

state = sf::RenderStates(blendMode);

window_.draw(test);

test.clear();

test.append(sf::Vertex({50, 50}, sf::Color(100, 100, 100, 100)));
test.append(sf::Vertex({200, 50}, sf::Color(100, 100, 100, 100)));
test.append(sf::Vertex({200, 200}, sf::Color(100, 100, 100, 100)));
test.append(sf::Vertex({50, 200}, sf::Color(100, 100, 100, 100)));

window_.draw(test, state);

window_.display();

yeah, that's why I was figuring you would need to be accounting for ranges etc in that step anyway.

My thought would be to do most of the number crunching during the map loading to sort out visibility ranges/obstacles for each map tile or piece. Then units could look up what's visible from where it's standing to do some kind of alpha overlay like you were thinking, or to tick flags in visible squares to let your drawing function know what it's okay to put on screen.

Not sure what the real RTSs do though. Or you could just ditch LOS and save yourself several days.


steal and trace and kill yourself

...

I did it baby

heres the code if anyone cares pastebin.com/bXG5y1uR

I'm not being an on the nose edgelord am I?

just so long as your protag it isn't the worst pixelshit in history

now just make it work with walls

That's the most confusing portrayal of loss I've seen yet

yeah, probably should have used more precise math, it's kind of chunky going around corners

Force yourself, my dude. You can do it.
Or failing that, work on the lesser parts of dev. Stuff like writing, or gathering music, or making sound effects. Even if it's nowhere near as important as coding stuff, it's still progress forwards.

Thanks. Give it a shot. If you decide to make something, I'd be interested in checking it out.

Might be a while though. I'm currently busy trying to learn how to draw Iris so I can make lewds of her.

doing god's work there, keep it up

Good job user, sorry I couldn't be more helpful.


That's a side character, user.

feels bad what do

oh I misread, sorry
but he's still kawaii

why not?

So when I made these pics over a year ago, I had in mind a Castlevania-styled game with weight combat like Dark Souls (eg, learn enemy patterns, don't just button spam, attacks have a long windup, etc). Enemies would be strategically placed, and even a giant rat could pose a danger later if you let it hit you more than a few times.

The problem was that it really wanted to be atmospheric, but that really affected the level design, mainly verticality. If you were slow and clunky, you couldn't really jump and move around a lot, which meant levels were mostly going to be boring horizontal areas with staircases. I didn't figure out how to fix that before starting to work on it.

Another problem is the size of the character sprite. I based the level size on La Mulana's sizes (resulting in the distinct layers" of the one level mockup); this is fine since it's consistent, but like I said it kind of feels weird if jumping and movement isn't a major focus. Movement more akin to the classic Prince of Persia would probably be more suitable, but I don't know if that's fun for people.

One other problem I had was with the lighting. At the time I had a set palette I was using, which is why there's a lot of busy textures in the tiles (besides them being 8x8). The backgrounds were actually the same tileset, but blended to 50% black. I had a tile based light system at the time, but the problem was that if you had more than a few light sources at a time, it would really chug. Think Terraria, but not really optimized. Not sure how to fix that.

Can you show the initial sketches of the main character?

post progress i guess
i got a working custom faction in m&b, with a new town which has it's own scenes
and i sort of got dialogue working
now if i can just get the game to not randomly crash while editing scenes, that'd be great

I've seen this so many times. Where is this from?

some game that huanita was reviewing positively

Sword & Sworcery, the epitome of hipster pixelshit.

You first. The white area is the visible area, right? Double check your code because it appears to fuck up where i've highlighted it in red. Look at brogue's code for a well done example of grid-based, line-of-sight visibility.

Okay cool. Had ideas for either an RPG Maker Horror game or a bullet hell shooter. I'll test out some things with each one and see how it goes.

This guy's implementation is great

adammil.net/blog/v125_Roguelike_Vision_Algorithms.html

Fuck off tumblrlard.

Right, so I never programed anything but I really want to make a game from scratch. My question is simple, where to begin? What engine to use? What language to learn? Must I learn how to draw? And finally, just how do I make the first step, what is the first thing to actually make?

The first thing you make is a clone of a simple game like pong. For that you are probably going to use Gamemaker or Godot and learn the respective programming languages for each.

hello triangle in openGL

You would need to learn how to program before using openGL, user.

Workout strategy for strengthening your code muscles:
- Complete one section of a book/online programming guide
- Make a shitty program that incorporates that section + some of the previous ones that you still haven't mastered.
- Only if you successfully implemented the program, subsequently jerk off to anime in order to create a Pavlovian association between programming and orgasms.
- Use this newfound energy to start the next sections.
- Repeat Ad Nauseaum until you literally become John Carmack.

I think it's satire, user.

I hate myself

Ok

We should have some of the artfaggier /agdg/ anons draw us a new equivalent to Motavio or something.

...

How about not making him a gay mexican this time?

Sounds good to me. Hell, who says it has to be a human and not something like a robot. Robots work hard, and probably think you should get back to work on your videogame.

Just a pic of what projects are still alive for perspective

K&R C or whatever for the basics

then start working on making programs, you wont be able to make games at first. Make an RPG in the console or something, so it feels like you're making progress. Eventually you'll know how to program well enough to start making games.

You said from scratch, but you mentioned engines. Really from scratch depends on how low level you're going to go. You need to be using the operating system (which you use a library to interact with) as a base for your game at least, and then a graphics library if you don't know (or don't feel like) doing all the graphics math yourself.

Some things like SFML do a lot of stuff for you.

That's a damned high casualty rate, but at least some of them actually released too.

How much from scratch are we talking about? Using a prebaked engine is far from what I would call "from scratch" but I digress, I'll give you answers.
If you want to work alone you need to learn to code at least a little and draw (assuming you want originar artwork). I recently wrote up some simple guides on coding and pixel art cause it comes up often in these threads.
how to learn to code - pastebin.com/xLPsf2ax
pixel art - pastebin.com/M7UGinH0

That completely depends on what you want to do. If I was talking "from scratch" then I'd say none(you literally don't have to use one, it's not necessary). But if you want it LOVE or Gamemaker are good for 2D games. I'd start with 2D as it's simpler.

Do you actually want to learn to code or just enough to like make a simple game and don't care about being skillful or creating marketable skill? If it's the former then follow my guide otherwise go with Lua or Python or the gamemaker lingo.

If you want to use your own artwork then yes. If you're ok with hiring an artist or using free artwork from places like opengameart.org then you can do that.

I'm not sure what you mean by this. As in what game to make first? What part of a game to do first? Generally start programming simpler programs/games first as you learn. If it's about what part of the game to make first well first you make a window and then create a game loop which does three things check input->update game state->draw stuff based on state then repeat until player quits.

Progress on AI. It pretty much does what it did before but better.

Wasn't sure how to word this exactly but I think the point comes across.

strawpoll.me/11220074

user please, we don't need all this tracking and social fuckery on the page

Sorry

poal.me/lqpudv

Please, no memeguides.

Carbonated beverage

Ooh ooh, I wanted to do something like this for a supermarket sim, how are you doing this low poly shit

uhh s-staying under 1000 tris?

it's a 12 sided cylinder fam, I don't really know what you mean

I have some sort of idea I want to fuck with. I'm going to be brutally honest here, i'm probably not going to get anything major done out of it, not even a game release or alpha/betas. Just something I want to try to see if it's possible. Now to get to the point:

I want to try a 2D turn based rpg where the environment you are in actually matters. When entering battle, you'll come across a certain set of enemies in different combination on different elevations. In order to beat them, you'll need to position your party appropriately(as you all start on the bottom floor) to face appropriate enemies. Rangers will need to deal with flying enemies or enemies that are really far back, Mages will need to provide support whether its for healing or additional damage, Melee will need to take out any ground forces immediately before they swarm everyone, and Tanks have to make sure no powerful enemies approach other party members.

There's a LOT more to this, but it's just a gist. What can I try this in?

I have ideas, just like everyone else, but they're probably shit. I can do some sprite work maybe, but I have no patience for a 3d modeler. I got stories that I can flesh out in a matter of minutes and I can give a character life like it's nothing, but I don't know who can't do that or who needs these skills.
I really came here to find out how that necromancer game is going.

It looks akward because the red hero is far smaller than the 8x8 grid that vision was being calculated on, and because it's drawing lines from the upper left corner of the drawn tiles.

It's really just drawing issues and I was more interested in how awkward it would be to do all the vision for each section of a map during loading or map editting so that thousands of units wouldn't be redoing the same math every frame or every step.


cool shit

you're a fucking retard

I also didn't recommend it as a sole resource. I recommend a book on basic algorithms to go with it and recommended to keep solving problems from HS competitions.

cs.umd.edu/Outreach/hsContest16/questions/problems.pdf

Here I found some simple problems in english (I'll have to add them to pastebin) and the first ones seem to be solvable after reading chapter one of K&R. Like the first one only needs you to know how to feed a file into stdin( using cat and a pipe if you're in linux ), read standard input, use variables, an if statement, a while cycle, simple arithmetic and printf. You get those tools literally in the first chapter of K&R, then it's just figuring out how to use them to solve the problem at hand. Which is how you actually learn to code properly, by creative use of tools at hand to solve problems, not handholding or ontology fuckery.

I feel like you fail to grasp the concept of a begginer just getting into programming, and the difficulties that he'd find by using K&R, it's made even more clear by the fact that you actually reccomend linux for thsoe newcomers.

Your guide's well-suited for novice programmers yeah, but don't tout it around as a "learn how to code for those who don't know shit" guide because it really isn't, though.

I've been planning a tactical rpg. Would godot be good for this?

I feel like you fail to grasp how retarded windows is for any coding outside VS. Setting up in windows made me want to slit my wrists and I have years of coding behind my belt so don't think you can lie to me about how linux is hard to use (even every decent uni with an intro to programming program recommends using linux, and I know cause I've been in couple really good ones). The popular distros are as user friendly as mac or windows these days and the C compiler is literally installed by default on most distros. You just have to follow K&R and copy the commands from the book into the terminal to get your "hello world" to compile. and run. The first chapter is literally aimed at noobs to quickly get them compiling code and running programs. I have the book open right here next to my keyboard so I know it's there, it teaches step by step how to get your code compiling and running in unix like system, you're a fucking retard, kindly kys.

"As user friendly as" is all well and good, but setting up another OS on your computer is always going to be far more hassle than not.

Something like love2d and notepad++ is literally just unzipping 2 files and pasting the proper run command into np++.

You see? That's exactly what I'm talking about. You cannot understand anything about begginers, your advice is good but you cannot get it through your head that 99% of people don't want to install linux just so that they can learn how to program.
You take every criticism as a personal insult as well, which is why you're resorting to smugposting and ad hominem at the end of each response.

You don't even have to dual boot. Just use Virtualbox.

Ah, just jumped in on this conversation. I see the more full context now.

I love linux, but I think making that the first step is a bad choice for people just looking to dip their feet in. Just install Python; it comes with it's own (very shitty) IDE.

If you want to get people to develop on Linux, just let them feel the pains of developing on windows. Of course, they have to also simultaneously be dipping their toe in on linux for this the have the greatest effect.

Package managment is great and the command line utilities make your whole OS your development environment. But you don't realize these things in full until you experience life without them.

I think it's something to do with how the place moves. If I let the player naturally fall down, the collision with the ground works, but if I were to press the down key he would go straight through the wall. Does Godot not automatically do this shit for me? I don't even know why this is happening.

Wow so hard. All I have to do is press F5 to test code. Done.

I don't want to install an OS, make sure it even compiles correctly, has all the low level drivers, etc., Windows just works out of the box and is feature complete. You don't want it harvesting your data? Make a user profile for devving and unplug your PC from the internet.

Also, fuck setting up a sterile coding environment, managing environment variables, and arranging all sorts of build paths and automation. It just works.

If people want to make games, point them to the simplest tool, like Game Maker, which is tailored to what they want to do.

nope, also centaur

is his front right knee raising from the water

hmm, you're right.
image boards might have conditioned me to expect penises everywhere

It's just that the image contained nude women and the men were also naked. Anybody that isn't autistic would see a penis in that.

...

I've explained why C, I've explained he can go the route of Lua/Python/whatever and why it can be bad, I've explained the guide is for people who actually want to learn to code well and make it a marketable skill not just like make simple 2d game in gamemaker for the rest of their lives.


>I want to learn to code properly, a skill which takes immense amount of time and dedication to gain but I won't spend

...

what's it called?

Xenobiology by InCase

have fun

...

It's never appealed to me so I never gave it much thought. You're right, it probably has caught up with a lot of modern conventions, but I just like to keep things simple

came up with this in another thread

"if yanderedev can do it,imagine what you can accomplish"*

* attempt to write good code,read a book on good code,avoid feature creep, be open to criticism

what's the best way to quickly check if an object is equal to another in a collection of objects, using Java?

I've been trying to work with a HashMap, but for some reason HashMap.containsKey(Object) is acting unreliably for me and I can't really figure out why. Is there another data structure I can use to quickly do these comparisons? I don't want to use an array and have to iterate through it every single time

Are you checking for same instance or equivalent objects?
Does HashMap use == or .equal?

de/buff extension,turn delays/swapping/haste have all existed in other games and mmos, so its not impossible to balance into your game, check out ffx or some jrpgs


any reason you didnt just make a doom mod?


what engine?


Also, shoutout to all the agdg devs, you're the only hope that fellow anons have left.

So basically fuck 4/agdg/, god damn mess of shitposting. Anyhow, posting progress. Started work on the second weapon in my phonegame, once it's done and the logic for switching is all there I'll be implementing the inventory screen.

as far as I know it uses .equal, which I made sure to overload in my object to make it work properly. I don't know, I don't really want to have to delve into the inner workings of HashMap to find the problem, I had hoped there was a similar data structure that was less finicky

looking for equivalent, btw

Are you literally using an Object type, or a generic T? I know generics were a key part of C#, and that Java also supports them, but I don't know how "first class language feature" they are over there.

In any case, every object instance should have a GetHashCode() method, for working with this data structure, even value and reference types. One instance should have the same hash code regardless of how it's used, as long as its own data doesn't change.

Well this sounds like your own fault, to be honest.
I believe that == checks for reference equality, and .equals checks for true object value equality… if you've changed the logic somehow or overlooked something, it sounds like its manifesting here.

If you wanna paste the relevant code I could take a look at it.

sure, why not. This is the object I'm storing in the Hashmap
public class BlockEx { public Block block; public int meta; public BlockEx(Block block) { this.block = block; meta = -1; } public BlockEx(Block block, int meta) { this.block = block; this.meta = meta; } @Override public boolean equals(Object other) { if(other instanceof BlockEx) { BlockEx that = (BlockEx) other; if(this.block == that.block) { if(that.meta == -1 || this.meta == that.meta || this.meta == -1) { return true; } } } return false; }}

I made something similar with 3 others for ios using cocos2d back in like 2011 or so,keep up the good work, are you trying to go commercial or just releasing a free game?

Well,
(this.block == that.block)

Looks like that you should make a .equals for Block and use that.

and this is the relevant part of the code that does the comparisons
public BlockEx getConvertedBlock(BlockEx inBlock) { BlockEx wildcard = new BlockEx(inBlock.block); BlockEx output = null; if(blockMap.containsKey(inBlock)) { output = (BlockEx)blockMap.get(inBlock); } else if(blockMap.containsKey(wildcard)) { output = (BlockEx)blockMap.get(wildcard); }

Unless they are supposed to share the same INSTANCE of block.

But right now it will fail unless those blocks are the same instance. Which would then make the subsequent conditional unecessary.

Probably free on app store at this rate. We'll see.

the block would be the same instance, every block only has one instance

Oh Java, I forgot how much I hated you

been considering learning Scala just to escape this…

Sure, I suppose. To be honest, I'm surprised you're interested, given I'm bad at art and my friends aren't used to drawing anime girls.
But here you go.

Included the initial drawings, subsequent iterations, and the art we used as design inspiration. My comments (at the time) normal, friend's comments (at the time) in italics.

If I recall correctly, he said this is more an experiment on creating an engine and practicing that rather than making a game.

If it interests either of you, I got it working. I had to override the hashCode() method in order to get the Hashmap to check the right hash for matching objects. So far so good.

Also caught and fixed an NPE in the process, wew

working on some new content, some new levels to play through.

Good job user

Make sure you thoroughly test your objects that arent simple data containers. Especially if they define custom operators that language runtimes are aware of

there is very little of that in Java, as operators are not override-able. Just a handful of methods in the base Object class can be considered honorary operators. hashCode() being one of them, so thanks for mentioning it. I had noticed that my equals() method was never being called so I had a hunch the hash was inconsistent, but had no idea that Object had a hashCode() method, I just assumed HashMap calculated its own hashes.

Anyways, that works now. I never planned on this project getting as big as it did, now that it has I wish I had setup unit tests up to this point.

bretty gud

So here's a video showing off the bullshit I'm getting from Unreal at the moment with their AI perception system. It clearly detects the player, but it will not successfully see anything else for more than a single frame, despite the debugger tracing to their position.

The bodies I throw around are blueprint pawns with a skeletal mesh. Because it's a pawn, it has the same AI perception detection as a character does.

I throw a grenade in the webm too which is a C++ actor with an AI Perception Stimuli Comp. Pawns can clearly sense it with both sight and sound, but again it drops the sense immediately as soon as it's found.

Logs captured in the webm indicate all of this. I've tried 5 dozen different combinations of meshes, capsules, pawns, perception stimuli comps and it won't successfully sense anything except the games characters. Even setting the grenades or tests to derive from character doesn't help (uses the exact same stimuli sense as a pawn).

Anyone have any idea why this would be?

I managed to run your game on muh ebin meme OS.
The mouse didn't work at all but maybe that's the VM being a fag. still, Doom works
vid_mode and vid_restart didn't work either

I'm not saying that you shouldn't, just that it's a bad example.

The mouse is using the windows API for raw mouse input, so it might have trouble with that. I know someone on linux had trouble with mouse input on wine too. I'm real glad that I got it to work on your OS, I just need to add the option for a mouse input system that might be more friendly to people running the application through a compatibility layer. Maybe I should add keyboard look… After all, the game does not care what y-position the bullets hit, kind of like in doom. I looked it up and it seems that react doesn't use that kind of mouse input, so I will add keyboard look into the next build as a quick fix. reactos.org/es/node/219

vid_mode can only be set in config.cfg. That's because the game requires you to run vid_restart to change resolutions (which has been broken since the texture system was added, and that was in version 0.2.0, which was released in 3/6/16, just over six months ago…) and so you cant do that because I never fixed it. I will fix it soon, though.

This is why in the readme, in the console guide, I wrote this entry for vid_restart:

"vid_restart:

Restarts openGL context. Causes graphical errors. Do not use. Incomplete implementation."

If you want to know how all the cvars work, you should read the guide in the readme.

Pic related is a nice picture of how vid_restart broke the game.

Are you using a behavior tree? It sounds like trouble with loops.

Alright fuck the particle effects I'm just gonna use sprites on the weapon animation itself

Hey, do you have a tutorial for fps sprites ?

No. It has nothing to do with AI behaviour, or blueprints, or code at all. It is purely on the AI perception system. The actors are being perceived by the system and instantly dropped because they weren't successfully sensed without any perceivable reason, except for the player character and other AI characters (Which just use the default Pawn sensory input, like my test ragdoll character)

das sum nice work tbh fam

u makin sum hexen type ish?

I use Gmod to pose my weapons n shit, then scale them down to 320x240 and do further cleaning up and editing using GIMP.

It's a doom mod, think G&G mashed with Magic Sword, the D&D beat em ups, and Doom and you have my project

I was going to take a day off to play videogames then saw your post and asked myself "Do I really want to be like this faggot?" and the answer was not so i worked on my game instead

The reason this won't be an issue with the code itself is because the code is the exact same for detecting ANY type of actors. It has to have something to do with the way the AI perception system either gains or loses sight of an object. The only info I've found trawling forums and answerhub is that the 'pivot' or base location of the actor is used to determine whether it can be seen or not. That is where the green debug trace points to. I've confirmed that in the engine code, yet despite tracing to the right spot it loses sight anyway. The code used to gain or lose sight is here:

// used to check perceived updates. Called by delegate.void ABaseAIController::OnSense(TArray TestActors){ for (int32 i = 0; i < TestActors.Num(); i++) { AActor* TestPawn = Cast(TestActors[i]); if (TestPawn) { const FActorPerceptionInfo* PawnInfo = AIPerceptionComp->GetActorInfo(*TestPawn); if (PawnInfo && PawnInfo->LastSensedStimuli.Num() > 0) { for (int32 j = 0; j < PawnInfo->LastSensedStimuli.Num(); j++) { if (PawnInfo->LastSensedStimuli[j].WasSuccessfullySensed()) { //////// SIGHT //////// if (PawnInfo->LastSensedStimuli[j].Type == SightConfig->GetSenseID()) { UE_LOG(LogGeneral, Log, TEXT("sensed pawn %s"), *TestPawn->GetName()); } } else { UE_LOG(LogGeneral, Log, TEXT("lost pawn sense")); } } } } }}

reminds me of this game

It is performing as expected then, it just doesn't have anything meaningful to attach itself to, so it flops the boolean.
Put it into a behavior tree and it should be working.

dat shit sound ryt playa

It has nothing to do with behaviour trees you fucking nigger. I don't use behaviour trees, I have written my own AI system.

I am not expecting the character to react because I have disabled the AI code. All it is doing is logging whether it has got sight of an actor or lost sight of an actor. If you see the attached webm you'll see that it instantly loses sight the frame it detects anything, EXCEPT for the player character at the start, indicated by the log saying "sensed pawn BaseCharacterPawn_C_0", and then only saying "lost pawn sense" when the character is out of line of sight. While whenever ANY other pawn is in sight, it will say "sensed pawn (pawn name)" and then immediately "lost pawn sense", due to it not being a successful sense for no perceivable reason, as the debugger traces to the correct location every frame that it's within sight, indicating a successful sense.

are you using _fixed_process(delta): ?

If the mouse works on XP, it's probably not worth fixing for React alone.

Well, I took the time to add in support for keyboard look, so you can play it in reactOS even though the mouse doesnt work. It really amps up the difficulty of the game though.

Here is a special build for you: p.trumpshare.com/nakj/dvn4.zip

I have been trying to get a response about the same problem for a long time now. It's been hours just wasting time browsing the same threads and getting frustrated at my own lack of the ability to be a competent programmer. It seems as tough Godot can not detect that I have stopped pressing the key. There is no "is_action_released" so I assumed that "is_get_pressed" will just work once and detect it again. But I can not set the velocity back to it's default (going nowehre) value. My code is like the following:

if (action is pressed):
set_velocity(0,1)
else:
set_velocity(0,0)

But the velocity does not change with the else, although I know it's going there because I did a test where it printed stuff. It just doesn't set the velocity. I don't know why. Either that or the velocity is changed instantly. I have this running in a process function so it's running over and over, and it should be setting the velocity when I have a key pressed, and when I don't have it, it should set the velocity to 0. Sorry for retarded post.

Thank you

kill me

set them to false each frame
like this
func _fixed_process(delta):#Input manager #movement var left = false if Input.is_action_pressed("movl"): left = true

set what to false?

thanks
you're a big guy

sorry that I can't be clearer.
create a variable that you so you can set to true or false
then check for that variable instead of input, remember to put it to false each frame and before any code that can affect it

I don't remember if you have to declare them each frame, I only know that I used to have that problem but now I don't

i tried htis, but it does not work. it just makes it to where i can no longer move:


func _fixed_process(delta):
var up = false
if up == true:
set_axis_velocity(Vector2(0,1))


if Input.is_action_pressed("player_up"):
up = true

I believe axis_velocity is cumulative.

Try set_linear_velocity

Er, scratch cumulative. More like it doesn't override your current velocity but simply adds a velocity along the new axis.

So in your original code you are setting the x-axis, and in the else setting no axis at all, so it was just ignoring it.

I replaced all the set_axis_velocity in my old stuff with that and now it works

thank you my man, have this image

Poor bastard.

blow it out your *ss

...

Pillow Shading: The Game

It has this 90's freeware vibe. I hope you will be able to score some catchy chiptunes for soundtrack.

Guy making a hopefully-not-furfag action platformer here. Currently mostly polishing up movement physics, but also trying to get the character sprites complete enough to throw on.

So far, aeronautics are feeling fairly nice. They're turning out differently than I'd expected, but I prefer this result. However, this preference has led me to reevaluate how I lay out maps and upgrades.

I'll probably show some shit once I have actual graphics placed on it. Right now it's just rectangles moving around.

I don't believe you

i made a helmet
now i just have to figure out how to make it not shit
i kind of assumed m&b at least did some volumetric shading or something, but i guess i need to make a normal map or something. that or i somehow fucked up the material when putting it in the brf files of the module

Im happy for them

apparently the issue is the shader
but i can't for the life of me figure out if any of the shaders have something like smooth shading
when i look at all the other ingame models, i'm pretty sure they have even less polys than my helmet, yet i don't see them as clearly as i do on mine

The mesh itself is the problem. You need to shade it smooth, then use a normal map (if that engine supports it). Alternatively you need to use smoothing groups in your mesh.

alright, in blender how do i smooth it?

There are various ways (please bear in mind that using smoothing groups will increase your polycount, because it creates some doubled edges).
The preferred way is to use auto smooth or the edge split modifier, because of how flexible these are.

If you want to use the edge split modifier instead, you disable "auto smooth" and add the edge split modifier instead. When you have "sharp edges" enabled, they behave exactly the same, except that the modifier allows you to disable the automatic splitting at certain angles. Don't apply the modifier manually and just let it sit there. It will do its magic when you export your model.

auto smooth worked
now i just have to git gud at actually modelling/texturing the thing
it would help if my tablet wasn't gathering dust in another city

and for the sake of not looking like complete shit i'm gonna add one of the ingame helmets underneath
at least now it sort of resembles pic related, even though i haven't bothered to uv map the second helmet

fuck, meant this pic

Nanosteam isn't dead.

Whatever happened to Broquest, was that not an /agdg/ game?

just retarded and too lazy to install SDKs and learn them

Notice how the reference pic has a flat brim and almost concave profile between the lighter and darker parts? Yours doesn't quite do that.

Fix that and the scale, it's too wide

The reason it looks so off in my opinion is
if you want it to fit in but you also want it to look somewhat good you need to change the texture from being flat to looking more worn or otherwise like it exists and has design to it like it was crafted and didn't just spawn from the ether. Make it look like it came out of Dork Sauce 2 basically, that's the closest thing I can think of to skyrim's lights and textures.
on that subject, skyrim and dook soos 2 have a lot in common visually when I think about it, and both of them when I look at it make me feel physically drained for some reason; everything in das2 looks like it's made of stone and everything in skyrim looks like it's trying to be realistic but also trying to have fantasy elements. Am I just wrong or is this kinda on the same wavelength as you guys?

well you're pretty wrong considering this isn't skyrim

is that an elderscroll's game at all? I guess I don't know shit then, but from what I have seen of skyrim and really oblivion too the worlds are all just like looking at gross gray fantasy.

it's mount and blade, ya dummy

that's mountain blade but the flat texture does look awkward in it too.
also light part too saturated, dark part too bright.

tbh i'm pretty shit at art, especially the digital kind. i still don't know how to make/bake normal maps
is there any program for converting png or whatever to dds? for whatever reason mount and blade only uses that file format, and i don't wanna have to use online shit for this everytime
i really hate texture painting. this model is supposed to already be uv mapped correctly and still when i paint on it in blender i get random color bleeding on some edges. i wish my toaster could run substance painter

The artist went on to be the artist for Breeding Season, then fucked over that team a while ago and announced his own clone of it.

Wasn't that that furry porn game and didn't he get absurd amounts of money already, yet still decided to fuck the others over?

Yup.


Show us dat UV map bby.

...

Well I guess I got brain problems

The first thing you should do is change those colors to be a bit more faded maybe. And texture it to look like fabric, maybe a cotton or something. I don't really know how you'd do that, I just know how fabric is supposed to look.

yea, the issue is that i'm trying to make this with a cartoon as a reference

Top kek.


What format are those in, I had no idea Mountain Blade was so easy to mod, normally you have to jump through hoops to get assets into games.

Try editing the mesh a bit.

it just looks like you're coloring it so that it looks rightish in blender, but what looks like dingy yellow/dark green in your blender turns into ninja turtles in m&b

Try making the texture have black or dark lines run down over it's length, maybe? And make your shit look like really really faded in your progrum

all the meshes are in obj
however, they're static meshes
you use OpenBRF to export game files, but it can only export models as skinned ones with some weird format that only it uses
so technically, i just export/import shit as static meshes. helmets are boneless by default
for the torso item, once i import the mesh into OpenBRF, i take another armour piece from the modules, copy it's "frame" and paste the rigging onto the new armour
and somehow that magically rigs it
then you need to save whatever file you imported the new mesh into, as a BRF file, specifically to your module, otherwise you're gonna be editing the base game files, and you don't wanna do that
then you need to edit the module.ini file in your module folder and tell it to load the BRF file
you also need to tell it to load textures
at least in this case i'm not making a new mesh, so i can just copy the exact rigging from the original
after that you need to download the module system for mount and blade, or morgh's editor and create the item, and then give the item to the player
and the annoying thing is, that if you're using the module system instead of morgh's editor (since it has more options than morgh), you have to randomly guess at what part of the file you should put the new stuff without it getting fucked because of some weird parsing. Or sometimes, i directly need to copy text from one file to the other in order to reference something, otherwise it might give me an error that it's not defined
i don't wanna edit the mesh though, considering that i have no actual control over the rigging, i'm afraid i'd accidentally fuck shit up
OpenBRF has some way of working with the skeletons, but i can't really even use it if i wanted to, since the program is bugged for me and doesn't preview any of the models, as you can see in pic 1
in this case, i've added 2 textures (helmet and robe), 2 meshes (helmet and robe), i made 2 materials for each, i saved both to a brf file in the module's resources folder
i set the module.ini file to read from the brf file and to load textures from the module's texture folder
i went into the module system files, created 2 new items for the helmet and robe
then i gave said items to the player (he has them at the start of the game, as well as whatever the game gives him from his bio)
i also copied the swadian knight and made a new troop with his stats but the new items
and i gave 5 of those to the player's party at the start of the game
then i ran build.bat, and hopefully if i don't get some errors because of random bullshit, things work out fine

i tried that, but i'm working without my tablet and shit is fucking unsteady, so it ends up looking like shit
i might try again by drawing with lines rather than like a paintbrush

forgot to also post references sorry

I have a flawless plan for generating cheap advertising for my future game


Anyways, I've found out that you can import adobe illustrator files in 3ds max to make 3D text in any font you want, though I just traced this one.

renamed my custom city to Omashu, custom faction to Earth Kingdom, it's king to Earth King Bumi
i also accidentally had to rename some scenes because of this, so now i need to remake the scenes for Omashu, currently they're just empty fields/rooms
i also need to actually make a culture for the earth kingdom, right now it's just using swadia's, which is why it has swadian guards in the castle and garrison

those two don't necessarily work because of each other

i dont know how to make videogames but i have lots of good inventive and great ideas in my head

Start with modding and level design.

That shit is great for learning how to make vidya

Wouldn't it be better to picture the earth kingdom as rhodoks? No Calvary but mass amounts of people. If you ever get to making the maps, make sure to make the earth kingdom fukhueg but hard to traverse by an invading force.

tbh as far as i remember none of the nations had any cavalry, aside from some ostrich horse thing
besides, i'm gonna replace the troops anyways, so it doesn't really matter

no im doing guitar and drawing

Wind nation had bisons, but they kinda died off.


I fapped to Overwatch, doesn't mean I'll ever buy it.

elephants in other mods ended up looking very clunky, i wouldn't bother making giant mounts
some horse-sized lizard is probably possible though

I can't refute that logic, it's just crazy enough to work.

Sounds like a perfect plan. Fucking do it.

Why not just add the porn into the game from the beginning?

That's what I'm planning to do.

Because then you're labeled a shitlord and your game's a porn game. People want porn OF games, since they're attracted to the characters, but don't want the game to BE porn because then it's automatically seen as nothing more than a porn game.

alright, so i got a culture going
currently the guard is the top tier troop, recruit is the tier 1
everything else between them is still the swadian ones
for the recruit, i took the leather vest and retextured it, and apparently there's no way i can export a texture from mount and blade's files. you can only import them. and yet you can somehow magically export models. go figure. if i could export the original texture, i'd just tint it green and reimport it
i gave the recruits pitchforks. is that a decent tier 1 troop weapon or is it too weak?
i also got swadia to declare war on the earth kingdom from the start of the game

Well, I for one want good porn games so I'll just go ahead with it.

also, for the heck of it i gave all the earthbenders 600 in throwing and 2 bags of stones
i think they still do less damage than arrows/bolts, but they're pretty fast so it gets balanced out
i might have to make a duplicate stone item that does more damage
in fact if i ever do bending, that's probably how i'll do the other bending arts

oh wait, i found the base game textures in another folder
guess i'll just tint/recolor those

Needs a way to charge shots worms-style.

What game should I make, /agdg/?

The one you've always wanted to play.

Alternatively, something simpler to pave the way to the former.

...

alright, now that i think about it, air thrower does sound a bit dumb

No shut up, why would it be the thing that you thing, which is air blower? It's gotta be the dumb thing that no one would think of.

I'll definitely try. I'm total shit at music and art but I'm doing my best.

...

behold!
now i just need to figure out how to get the hat to not look like shit

YES
it would be like that one mod with the wizards and magic except every unit on the field would be a wizard

It's almost time to make some stuff.

...

I saw that, except

This means it uses the normalfag definition of CP which means no loli, shota, or cub characters.

...

I don't know why this gets me so hard, but really thinking about it hurts my sides
Rocks are like minature flails in terms of how they would damage plate, so cushioned mail would make it mostly ineffective, but fuck plate mail with that shit. Chain mail would be cracking apart, it'd be fucking nuts
Also, I found a kind of texture that could help you; with the templar mod I have for Das1 I the fabric wasn't a flat white, it's colored in a way that I thought originally would help but didn't think about how it would work. basically, every tile of the fabric needs to be slid down in a very rough gradient to a darker shade, so each tile has its own set of light and dark, making the clothes look more like a fabric that shades properly, or something. Something about this texture turns this into something that fools the eye into thinking of a padded fabric, like this

furfaggotry

I like the character's design.
I hope you can find a decent artfag to help you.

Have a shitty drawing , and good luck

Also
Patrician taste

does anyone if there is someplace where i can get all the half life 2 models, textures and animations
Specifically im looking for the rebel and combine soldiers fully textured and animated
if it helps i own half life 2
also i don't necessarily need the i just need some way to turn them into a spritesheet so some kind of viewer would also work

if you want to use a spritesheet, you could also get the WoW model viewer and record animated models from there
no clue about HL2 though

I've been trying to draw her for days and it always ended up looking like shit, pic related. So I took a break and practiced eyes. They gave me the most trouble since I tried emulating the feel of the last one's in and large and cute.

I wanted to give it another try later on.

A new map system I've built appears to be slower than the old one it was supposed to replace, and created more problems than it solved, and I haven't even gotten halfway through implementing it yet. I'm at a point where I legitimately don't know how I'm supposed to build it.

I don't even care if it's fast, I just want it not to cause lag and freezes on a high end computer which it currently seems to be doing.

Best part is that I wouldn't even know where to begin to start looking for a hint about what I need to do differently. I can't ask for help because people aren't willing to help unless I change everything about my workflow and tools until I hate programming more than I do already and am forced to grow a beard and stop going to shower and buy a keyboard designed for writing raw binary.

It's times like these when I feel like I should quit game dev and go back to just being an artist. I already gave up trying to get into a school, not like I would want to lose my job for that anyway. Or maybe I should just forget about making a playable game and keep going with the assumption that it will be nothing more than a laggy proof of concept for my game idea, so maybe someone competent will get inspired to redo it from scratch properly.

Post code

I assume you're looking for CS grad-tier code. Here's some.

DID SOMEONE SAY EARTH THROWING
the new rocks do twice the damage, and aren't purchasable. that's about it

Yandev we don't want you on Holla Forums, get out.

You're overestimating my skills.

There's being self-demeaning and then there's just being a faggot
You're being the second one right now

Don't start shitposting if you can't deal with being shitposted back.

That's not what I meant.

...

Are you using javascript?

Or python.
Seriously, why'd you make a language without a switch?

Depends on whether you want to blame all my problems on the language instead of my programming skills or the lack thereof.

Do you know what a switch is y/n?
Did you make that code just to piss us off y/n?

no idea what that language is but use goddamn arrays
don't make a separate thing for each direction, use an array and define out a which direction relates to their index if it makes it easier

#define N 0#define S 1#define W 2#define E 3#define NW 4#define NE 5#define SW 6#define SE 7var rel[] = /*all that shit you posted, but using the defines.. */for(int i = 0; i < rel.size; i++){ //OO langauges have all this crap that automatically tells you the size of each array, right? if(rel[i].block) rel[i] = world.block.arrToObj(rel[i].block);}

git revert

Holy mother of god, is that supposed to be a person?

I mean, I'm not clear on how the language works but I'm not sure why you would do it this way instead of using a nested loop.

Dude.
Duude.
//shitty pseudocode//index order (numerical)//036//147//258for(i = -1; i

Decide for yourself.

...

This is basically what the function "world.block.arrToObj" seeks to prevent, it turns the map data into an object that's human readable and easy to deal with, instead of forcing me to make a million checks and interact with nested arrays every time I want to interact with the map. That's what I did with my previous system and holy fuck it was impossible to work with when it got complicated, because you'd be looking at 50 instances of shit like arr[3][8] in the same place and wondering what the fuck the whole thing is doing.


It was just 4 directions initially and I never bothered to turn it into a loop when I added more. I tend to just do whatever until I get it working, and then make it better when it's good. But I never got this working well.

This is why you use enums/defines so you can see 50 instances of shit like arr[E][SW] instead. That or change the method of printing to make it more readable.

you need to learn how to keep track of your loops. you're probably making it more complicated than it needs to be.

That probably wouldn't help considering what I'm doing. Not sure if I want to get into the whys.

Then change the print method into something that makes sense. You turned seemingly functional code that was hard to debug into less functional code that you find easier to read. That's not a good trade at all.

What do you mean by "functional"?

I think it is considering the previous system was utterly incomprehensible if I hadn't looked at it for a few days. I'd rather take code that I can work with than code that is faster, so long as the tradeoff doesn't fuck the whole thing up.

But that's exactly what it did?

I rewrote quite literally everything, the entire system works differently now. That is just a tiny part of the whole.

Based on what kind of things changed, I highly highly doubt that's the one causing problems.

Napkin sketch to protest removal of glasses.

Write comments and write formatted output functions.
Didn't it?

Except the problems started after the rewrite, which means the rewrite likely caused them.

You're right, it can look pretty cute.

Perhaps the glasses are part of her disguise but she doesn't wear them while on the ship?

What I'm saying is that there's other parts that are much more likely to have fucked things up than the data translation function. That function is nothing compared to the other stuff that's happening.

I'm also voting to keep the glasses.

They could fog up in a steamy engine room or such for flair.

Making a shotgun for an arena shooter.
The model isn't final by any means, it's for painting over to get some juicy concept art.

I'm planning to finish a demo in Unity in 1 month. How possible would that be if it will only have 1 weapon, 1 enemy type and a single map with some basic decorations, but everything will be very polished?

Then post everything, retard, Do you want help or do you just want to whine that "my code is bad and I feel bad?"

When does you computer lag and freeze? When the map is loaded? When the map is edited? Every single frame? We have no idea what is going on with your shit other than you don't appear to be using very efficient code. I've done fractals as an array of pixels at full screen size before at almost real-time (10-15fps), and I've programmed countless grid-based board games and roguelike stuff. Half the people in thread probably have as well and could spot your problems in a second if you'd just post your code and how it works/what you're doing with it.

Or don't, since these snarky responses make you sound like a huge faggot and I can't guarantee anyone wants to actually help you.

Odds are you're problem can be solved by making the map an array of ints using part of it as a type identifier and the rest as bit flags for shit like passable, etc.

I remember in the last thread you said you wanted her hair messy because she has permanent helmet hair or something too. You're overthinking it. Make her hair however you want it to look, or better yet, how it would look best. Same with the glasses. Don't come up with bullshit like "but she can't wear glasses under the helmet" well then the fucking helmet has vision correcting optics built into it along with the nightvision and zoom and whatever the fuck else. Don't make sacrifices that will alienate 99% of your player base just to the appease that autistic 1% who will go "oh… I get it, she's like X because Y. Ah ha, very clever of the dev, hrmm hrmm I see" or some shit. Do what looks and plays best.

I think you're confusing me with someone else.

I'm voting against glasess because of how cute she looks here

Oops, my bad, didn't follow the chain of replies right.


see

I'd need to make a github page or something to house "everything". I doubt anyone would actually care enough to read all of it in the first place.

I wasn't looking for solutions, I was done looking for solutions to big problems online long time ago because I tend to get everything but the answers I'm looking for. I've tried looking for help way too many times throughout the years, and I'm significantly more likely to be told not to use Javascript and to install linux and VIM and to scrap my plans and do something else than to actually be told how I should approach the problems in the code and which parts could be designed better. I don't care if this sounds like another "snarky comment" but that's the reality of what I've experienced.

I guess I'm just ranting out of frustration because I was so hyped to work on my game when I started on the rewrite.

When you're running around and the chunks and blocks are dynamically loaded in and unloaded. A whole bunch of stuff happens in that process so I can't just post a function and ask "what's wrong with this?" because there's no single function to look at.

I want to do something much more complicated that can be accomplished with a typical simple array of ints. I want to be able to store arbitrary data in each block, I don't know how you'd ideally do that so I'm just storing it in an array and using the aforementioned translation function to turn it into something readable.

So if I want a tree, it could for example look like: [id, age, woodRemaining]. A chair could be [id, direction]. A chest could be [id, direction, [list, of, items]]. The first value, id, can be used to recognize what the rest of the values are. If a value like woodRemaining is not in the array, a default value for that variable is used instead, so I don't have to store a direction and damage value to every single block.

The ID is where I get information that determines how the chunk collision boxes are constructed (e.g. solid, platform, liquid, lowered friction) and how the blocks are rendered (e.g. simple sprite, block with connecting edges, something unique).

I give up on life, I can't deal with this kinda shit anymore

Have every function report the difference between its start and end times in that process. Make sure they all run sequentially. Then you can nail down where the longest-running code is. Then do the same thing with each step of the function, and you can find out what's taking forever to run.
Store all the extraneous data under a container member object, which always has a field that can tell the program what type of data is in it (like a tree, or a chair). I'm not 100% fluent in JS, but I know that you can do this in other languages using polymorphism.

You underestimate how bored and willing to help out people can be, especially if they can berate and belittle you at the same time. It really looks like you're just doing too many function calls per square of your map. What size maps or what size is the play area at any given time? Are chunks being loaded too fast? Like, minecraft with a fighter jet mod? This block of code in is it called per tile, or just the one the player is on (rel.me)? Because if it's every tile, you have a huge redundancy when you check all 8 tiles around a tile (9 if you include the tile); you have 5/6 redundant checks out of every 8/9 per tile. Have you ever done an implementation of conway's game of life?

also, post or tell us about the functions world.translateCoords() and world.block.arrToObj().

Not directed at you so much, but for once, I'd really like a dev in one of these threads to do just that: post fucking everything. Instead all we get are guys showing minor cosmetic changes every few days then getting pissed when people nitpick how it looks, since we should all assume the gameplay is tight as fight despite never being shown off… Honestly, if you posted a complete design doc, compilable code, along with everything you've done so far, your game would practically get made for you.

I'd post more of what I've got but I don't do graphics at all so I don't do cosmetic changes, and what I have so far is not playable to my standards. These days I'm fighting more with UE4 than I am actually making progress. Will probably even post a GDD at some point, and definitely playable demos when I get around to it.

Instead of working on my game, today I worked on an application that does undefined behavior until it crashes

Should be fuck. Need to stop drinking.


I guess you're right. A lot of what people show off are nothing more than tech demos and prototypes of planned features. Where's that shitter with the white hair bitch with the big feet? I still don't have any idea what that is supposed to be about.

I'll probably post a video this weekend, actually. There's a couple things I want to patch-up and implement before I record any footage.

It'll still be rectangles however. Gonna take a bit longer than that to have the graphics ready to throw onto the characters.

Ahah, thanks very much.


Personally, I love the glasses. I think they're cute, and really cement the intended direction of the character.

I agree, actually.
I tend to think about stuff that I think would be fun/cool/cute first, check if it could work in-game, and THEN justify it afterwards.
I like cute and unkempt nerd girls, so I decide to go with that and then afterwards say "it's because she spends all her time in the suit".

The problem with the glasses is more a question of sprite size.
There are only so many pixels available on a face, and glasses would take up a lot of it, so I'm quite worried about if there'll be enough room to actually show the glasses.
Very bad pic related, whipped up for a quick demonstration.

Of course, this is me speaking entirely out of my ass. An experienced artist might say "Yeah, it's no problem, just gotta not be shit about it" or "I can do sprites at a higher resolution to where it's not an issue".

I actually didn't know you were doing a low-res sprite for the character, I figured you were doing first-person for the gameplay (obviously you can't see the protag), and then something like a VN would for cutscenes and dialogue since you keep showing illustrations of the characters.

I'm gonna make a fucking game about the golden age of piracy

Honestly, that's because illustrations is all I have.

A VN mode could be very easy to code (and in fact I pretty much already have all of the code that it'd need), but in-gameplay there's also camera movement that pans across environments, mirrors, security cameras, and etc that show the player.

I could just remove all of those real easily, but as much as I like the glasses (and believe me, I really do. I already pointed out but it's probably my favorite take of Iris yet) that's a lot of things to nuke just to justify a bit of character design.

Got any more art of her suit?

I noticed she has a small notch in the visor of her helmet.

There's this, drawn by Drawfrag, which I think has been my default "go to" art to post.
He put in a scar on the visor as sort of a take on the mandatory pirate eye damage, but on a helmet instead. It looks cool.

The lineart is actually surprisingly passable, see if you can't get a drawfag who can shade metal and cloth and shit.

start profiling/benchmarking your shit faggot.
Make things more modular (in the future), so you can unit test sections of your code, and then optimize shit based on the data you've gathered.

added in a new prop to the game

sampled ur video/audio for the /mu/ ARF collaborative song btw


dude weed lmao

want the file? just be warned, this thing is designed to read in random memory and spit it out on the screen- so it will crash, and it will have problems, because its completely reliant on undefined behavior. It'll look and sound different on your machine because you have other things in your computers RAM.

mega.nz/#!ABJ2ib6Y!klgapAAkm_Rh2aTY2enrT13xqsVHmseH0MAQ7kd9NUk

You'd expect those chunks to hurt more. A rock of that size might weigh something between 150-200 pounds. Looks like they're flying at a high double digit feet per second at the very least.

That's a thousand joules easy. That's .44 magnum levels of fuck being thrown at their heads, except with all of the penetrative and stopping power turned into skull-flattening, neck-twisting energy. And in relentless volleys.


why demon is spongebob

no m8 i just got bored and ran your youtube video through clipconverter as an .mp3 file

thanks for the file though, DL'ing it for later use

Right. Now to start on the actual attack

its not the video file, but its the program that was being recorded in the video, and the source code so that changes can be made.

nevermind, you knew that , i'm just being autistic.

Posting Progress or something

datanon.itch.io/s4
0.0.2 -> 0.0.3
- Made scanline shader look better
- Fixed misaligned pixels in menu due to upscaling
- Added heat mechanic to regular asteroids, the weaker blaster must heat them up before they explode
- Added new visual effect to heated asteroids
- Made an attempt to improve spin algorithm on satellites
- Darkened trash bags on earth level
- Fixed a shitload of minor bugs
- Added new sounds for button pressing (not all buttons yet, just a test sound for a few) and the Sine gun

So basically I've gotta go in one of a few directions now. I can either start really implementing the inventory/ship systems, make more levels and enemies, or refine the gameplay and scoring system. I'm not really sure what I'm going to do next, but maybe I'll just flip 2 coins for it. I kinda wish I was better at art, but getting better at it is kinda the point of this whole thing so whatever.

can you tell me why your asteroids look like drawstring bags? what is that little bit hanging off them? it's very distracting when they rotate. I understand you don't want to make them perfectly round, but it looks like a part isn't even attached. whats the deal?

Those are trash bags user, from people in the future ditching all their shit in orbit illegally. I agree about the draw string though, I should recolor it.

alright here is everything gitgud.io/greenman/rts.git
ignore the design-doc-rts.txt im now working with Rebellion-design-doc.txt

I'm going to a gamejam over the weekend, dubs decide what game we will make.

MMORPG

i'm at a crossroads, can't decide. 8-drection sprites, more (16?) direction sprites, or low poly 3d.
I love 3d sprite games, but I don't like the facing ambiguity, not seeing player's look direction in multiplayer, too-close interactions, high-low angle problems.

Good low poly 3D is hard to pull off, sprites are relatively easy as you can go as hgh poly as you want and it doesn't have to animate properly, just be posed into frames and you can edit things manually. So try making low poly 3D, post some stuff, and if it's shit we'll tell you to make sprites.

well this just looks stupid
i should have just tried to make the thing on my own rather than slapping ingame stuff with poorly done meshes of my own

...

have zero idea for a game, wanna help ?

There were several attempts. All of them croaked. Most promising one was the one that tried to be sengoku rance instead of a jrpg, but the artist got conscripted or something.

...

does this look like the face of mercy?

i don't think i'm gonna get it any better than this
as usual i fucked up the UVs which is why the triangle thing above the face is like that, but i really don't want to mess with the UVs anymore, considering that the texture is gonna look like shit regardless

the best

also maskless version, since the masked one looks fucking stupid
the troops randomly choose one or the other (although for some reason they seem to be picking the masked one more often)

part of the reason for that was that i accidentally put the eyes unevenly away from the middle

There's a reason cinematic/hd remakes desaturate everything and add grunge and extra detail onto their redesigns. Because clean and simplistic and highly saturated things look like toys in a realistic looking context.

it looks really funny
have you considered just increasing the polycount
Im not sure how this stuff works in mount and blade have u tried googling it?

the main issue is that i made certain polys too small in comparison to others in the UV map
anyways, i took the texture from the swadian knight armour, pasted and warped it a lot and now the UV issues are less noticable, + the color is more appropriate
now i just need to figure out how to get the mask to not look retarded

#1 way to make it look better is to match the colours with the scheme in mount and blade. You need to turn down the saturation a lot.

Put some other sort of texture on the mask rather than leaving it flat. Might be difficult coming from the source material being anime and shit but maybe give the mask a grainy sort of bone texture to it instead of being blank and white. Try to think of the materials the outfits are made of, and texture them as such.

this isn't exactly what i had in mind

actually making it brighter seems to help
i think i just need to make the mesh a bit more cylindrical to work with the texture

The fire throwers helms are made out of bone it looks like, and since that is actually better because hot stuff on metal on your face would not be good, it should look like bone. Also skeleton asthetic.

Starting to look more like a sinister clown than a fearsome warrior.

i can't help it
shit just looks silly one way or another
worst case scenario i'll get an actual skull texture in there, at least it will look slightly terrifying

Personally, I like the one with the large round eyes. It's simple enough to look ominous. Maybe a little detail shading to make it seem like the holes are actually holes, and make the red accents a little less Wacko the Clown flavored.

As in, keep the general design, shift the positioning to make it not quite hug the edges of the helmet and keep it symmetrical, and small changes in shading and texture to make it look like it's an actual mask. Right now it looks like a like a piece of copy paper worn in a clown hat and not really a physical thing.

...

Rest in peace, user. We will miss you.

Looks like user got kidded once again anyway.


Take a good look at how they did the eyes and mouthholes here, it's more than just a black stripe. Emulate that for your round eyes, and try using a bone texture like other anons suggested. Hell, have a link to some free ones opengameart.org/content/bone-texture-pack

i think i'm just gonna go and make everything cartoony, my lack of artistic talent is a bit too evident
at least then shit will be consistent

is there any way i can just turn that into a brush or something?

...