/agdg/ - Amateur Game Dev General

Is just using the time to seed my RNG a bad idea?

Got a bunch of minor things done, not sure whether I should update my art now or try to build the first sequence of the game using my placeholders. I should probably test out a few things like boss battles and puzzle mechanics before moving on.

Using RNG at all is a bad idea. Crutch of poor game design.

Go fuck yourself Keith Burgun. Auro was shit AND it had RNG.

autists actually believe this

RNG is fun

Depends if you want the randomness to be reproducible.

Any idea how I could have 360 degree aiming for 2D without having the separated from the body retarded looking rotating arms and weapon?

Dozens of unique sprites.

draw (360 x (number of walking frames + 1)) sprites

depends on how precise the time is and how you're converting the time into percentages

16 sprites or 8 then flipped for the other side

should be accurate looking enough, or you could do 32 / 16

Idea guy here
A game where you have a limited amount of frames to beat it. If you wan't to play with 60 fps you have to play very good but you can switch to 5 fps and have more time

>depends on how precise the time is
time_t time (time_t *arg)

>and how you're converting the time into percentages
UH time to out myself as an RNGlet: I don't know what that means or why I'd want to do that.

Copy Abuse.

rotating arms but with 2d mesh deformation to make it less separated.
also with lowres 2d pixelshit separated parts may blend together just fine.

I'm currently trying to learn Unity (doing the project tutorials)
Should I properly learn C# or is just contextually using it or looking for examples enough?
My only programming knowledge is a Python introductory course I took on college years ago and some basic stuff I've done on R for work.

Just learn the basics of C# and you're good.
Like simply understanding what functions are and how to use them is basically enough for a shit game

There was a similar question last thread if you want some more opinions on that

In Unity, C#, how can I make it so, that a child collider does not trigger scripts for the parent collider?

I have the player getting destroyed by contact, and a child collider triggers this script, although I don't want it to trigger it.

I read that I should just add a rigidbody to the child, but if I do this, my child just disappears.

Just check what objects are colliding in your kill script and don't react if one of the objects is your child object.

The problem is I have the destroy by contact script on the parent, so that I easily can call the explosion effect of it.

How should I be able to ask there if the parent collider collides or the child collider?

You can get away with very little knowledge about C#. However, once your project grows bigger and performance becomes an issue it's good to "properly" learn C# to avoid getting dragged down by inefficient coding.

bool valid = true;

foreach(ContactPoint contact in collision.contacts)
{
if(contact.thisCollider.gameObject == childGameObject || contact.otherCollider.gameObject == childGameObject)
{
valid = false;

break;
}
}

if(valid == true)
KillMe();

Thanks m8s
I think I'll learn some C# then

>collision.contacts
>childGameObject
What should I wrte instead of this? I get errors there.

Should I make an 'overworld' for my open-world top-down game?
Or should I keep the game in one scale all the time?

collision is whatever you've named your variable in

OnCollisionEnter(Collision whatever you called this)

childGameObject is obviously a linked GameObject

Does not work. They still give errors.

youtube.com/watch?v=9JhsUFuqbCM
what was the meme about this song again

that made me lol

Yes. And the overworld should be a sidescrolling platformer

Got memed on really hard last time I asked this, but is anyone seriously looking for an artist?

this isn't that bad an idea! I'll take it

You're giving me flashbacks user

unity

It's a programmer issue, not a code issue. If you can't figure out what childGameObject is, you need to stop working in Unity, hit the command line and learn how to program.

Good luck with your project etc etc
If you're actually serious, post a portfolio or you'll just get memed again

Stop acting like a retard. Your "game" looks like shit anyway, so stfu.
Your solution was retarded anyway. I already found a better one, asshole. Go fuck yourself retard.

>want to work on game
>have to work on uni stuff instead
[angry dev noises]

...

kek

Getting your GPA / degree classification as high as humanly possible comes first user.

>want to work on a game
>have to work on uni stuff
>working on a game is my uni stuff

>listening to youtube.com/watch?v=a_FAF_v87Qw while planning my cyberpunk dark souls game

feels good man

>going to uni when you could be making a game
unless it's free, going to uni is a mistake

invest your student loans into your game instead of being yet another victim of the degree mill

you have no future as a wageslave

>he fell for the shiny paper meme

unless youre very good and youre willing to work for free, and hand in all your assets 1 hour into the dev timeline, no one here will bother

strawpoll.me/14614889

Too late, I fucked this semester up because I had uninteresting classes.

I can't imagine having to do gamedev group work with retards (because you always end up with dumb people in group work). It sounds worse than hell.

>not Vangelis's Original Blade Runner OST.
Disappointing. You're that guy who did some 3D models ?

I have €200 left at the end of the month (My family's poor so I get studentbux, rent is super cheap), it's a lot more viable than staying in my shitty hometown and working shitty jobs.
I also get a lot more free time to work on my game, so hopefully I'll make it before I get out, because I sure as hell don't want to work an actual job other than gamedev.

Depends on the game entirely

>visiting russain tracking sites

Are there any good open source C# games?
Trying to learn Monogame but there are no tuts for the genre I'm actually interested in making and I have no idea where to start.

>I had uninteresting classes.
I know that feel

How do I store a code generated object in a way that allows it and its variables to be accessed and modified? I can make an array and store it there, but I don't know how to access its variables.

when will we get a new boogeyman?

the real answer is both

fun should take priority but you should also have some idea what you're designing. a forest will have different design elements than a city.

t. vladimir putin

>funfunfunfufnfunfunfunfunfunfunf
>unfun
everyone voting fun, what good is fun if it doesn't have a function?

>code in English
>comments in Japanese

hmmmmmm

what genre?

fun has the function of gameplay
the other option has the function of visuals
you're making a game, not a movie right ?
if the game isn't realist anyway, you can get away with both

You're asking for a simple answer to a complex question. Generally though you would decide on a theme for the area first and then create a layout that is fun whilst being cohesive with the theme to the degree that the tone of the game demands.

have two arrays:
the first arrray stores every label of a variable
the second array stores every value of the corresponding label

proof that flash is still king
youtube.com/watch?v=8_KBjd0iaCU

obj_array[0] = my_object
obj_array[0].my_property = my_value

That's disheartening. I am willing to work for free though, on a non-commercial project. Actually that would be ideal for me right now

just learn an engine should only take a year or two if youre not retarded like most people here

>year or two
This is too long

you gonna give me a hint what you mean by that or should I just ignore this post?

that guy was just memeing you but if you don't post a portfolio there is a 95% chance you are just memeing so everyone will ignore you.

>fun whilst being cohesive with the theme to the degree that the tone of the game demands
Nani?

why not both?

Strategy
or more specifically, artificial life
Want to eventually make something like the Sims but would be happy to have something like SimEarth

youtube.com/watch?v=m4wHnaujiao

my shitty breakout is 99% complete, pls rate.

ok, well is there another place an artist can find a game to work on that isn't so memey?

OK now I figured out what i need and remade everything right.

Before that I had a floor that spawns the green cursor at a location of hit location of mouse pointer hit from camera and tried to use UE4 navmesh and built - in AI for navigation, it was complete horseshit.

Then I tried instanced meshes and stuff, turns out it I cant reference an individual element of an instanced mesh.

Then I figured out and made it so:
1 a "floor" object that spawns a specified by X and Y number of static components to itself (tiles) and puts each one into an array of references
2 every component spawns a green "cursor" on top of itself when pointed at by mouse
3 cursor checks for collisions and turns red when is blocked, spawns a tower when not blocked
4 having an array of references I can check each one of them for collisions (towers and obstacles) and put them in boolean array of walkable tiles (true/untrue = walkable/blocked)

Next step is either getting very smart and implementing a path finding algorithm or going full brainlet and spawning an invisible man who will run very fast from the creep spawner to the player base and recording each tile he walked on as a way point for the creeps.

>Having to switch keyboard layout between the comments and code

Should the pivot of a 2d platformer character be at the bottom or center of the sprite?

feet

At the bottom in the center of gravity.

bottom of center

black shell media

>Find game with programmer art
>Ask dev if he needs an artist

if a game has programmer art it's a failed project to begin with

(you)

That's what I'm trying to do but all I get is memed on

Thanks, I try. But it's true.

Finally set things up so that the enemies can take their turns. Pretty sure I can make them "think" based on basic temperaments as well. Not sure what to do now.