/agdg/ - Amateur Game Development General

I added muzzleflash

post the gamedev equivalent of a baby picture, your first pong

your game lacks any kind of oomph
it's like battling sloths

Sometimes the bugs are kinda funny

>it's dark souls except too lazy to animate a walk cycle dev again

please let the player use enemies as mops

It's a feature. First strike after impaling throws them as a projectile.

I'll keep that in mind.
I've just started working on combat so I'll start adding new stuff and hopefully some of them will be the hook.

battle royale but 2d

you mean PUBG 2d?

of course. anything else is just replicating the experience for which PUBG is known

I've reported you to PlayerUnknown and he says he's already considering legal action for this post.

...

Play my game, /agdg/.
drive.google.com/file/d/0B7-Rg7c5ieLBMHpvaEVaOTZEME0/view?usp=sharing

----[ Recap ]----
Game: Cash Sample
Dev: SlowDancer
Tools: RPG Maker MV, Photofiltre, Cubase 8, Internet, Coke
Web: lemmyrodul.com
Progress:
- No concrete progress for the last 5 days
+ Got some good concepts gameplay wise

...

(Me)

get rid of your trip, fag

Because the people working the frames want to be paid as much as the guys on screen so they've always kept a lot of code behind gaps in access. You have to make a multi-threaded system that loops between active states to get something looking better than sprite work. Thats to just to run them you have to really work skeletal work into the build of a model to make proper looking animations out of it.

Hey anons, uploaded a quick build of my game for a competition over the weekend. There's no story mode but the core gameplay is in.
I'd appreciate any feedback if you have some spare time to play it.

aaaleee.itch.io/international-bounty-hunter

are walking simulators memes?
I want to make a trippy walking sim and maybe then throw some gameplay mechanics if it turns out alright.

>I want to make a game but I don't have any gameplay idea
yeah it's a meme

You shouldn't care what people think and JLMG.

Well, I kind a 3d placeholder working.
I fudged the math rather than doing a real billboard matrix so it doesn't look great, but since I'm trying to move the content all to 3D models anyway I'm not too worried.
I just want an excuse to keep in a couple sprites because I suck at modeling for now, and maybe for some super cool effects in the future.
Pathfinding also works! Exciting. I'm probably going to expand some win/loss conditions to a round with some more exciting traps and then I'll write a new roadmap.

----[ Recap ]----
Game: Mech game blabla
Dev: Tomo Ge
Tools: UE4, Blender, GIMP, and other various broken tools
Web:
Progress:
+ Fixed missiles collision, again, hopefully for the last time.
+ Theorical knowledge
- Worked on a model for a week, still no model
- No real progress, no time to do real progress this week either
- So little progress I forgot what was the other thing I fixed
- HDD about to fail, let's hope it holds up until I get a new one.

They're a meme because they have no visuals.
LSD dream emulator has a big reputation because the visuals are kind of unique. Also valid for Worlds.

thanks user

exactly, I want to base the game on visuals more than anything. I have some great ideas for gameplay which I could eventually implement, but the main aspect would be always visuals/sounds/exploration. I just don't know if games today can survive based solely on that.

walk simulator is an adventure game from developer that's too lazy to make puzzles

>puzzles
but then how will journalists play your game ?

if journalists can play your game you are doing it wrong

put on screen instructions on what button presses and place them where to press those buttons

We can agree on that.

Is hardwaredev something hobbyists can do? Is there a middle ground between professional electrical engineering and playing legos with off-the-shelf parts?

are you really taking enginedev this far

----[ Recap ]----
Game: Horror JRPG
Dev: LoneBuckshot
Tools: Unity
Web: twitter.com/LoneBuckshot & lonebuckshot.tumblr.com/
Progress:
+ Polishing code in preparation for save/load states
+ Random encounters that move on the maps
+ Debugging battle system
+ Polishing battle UI
+ Adding more enemies
+ Adding more battle attack effects

LSD dream emulator "works" because it's backed by 10 years of actual dream journals, and comes from a pre-indie era.

If you want to make a "trippy walking sim" today, you'd better be able to back it up with either unique visuals or a unique story. This means actual skill at either art, shader tech, or writing. Be careful not to fall into the overused filters and shader techniques, which you can usually recognize by being included stock with the engine. Alternatively, emulate an aesthetic that hasn't been translated to games yet.

how am i supposed to sex this one

Don't concern yourself with such things, user. Just answer the question.

Why don't you start with playing with legos and see if what you're doing requires you to learn more?

It's not the learning that concerns me, but equipment.

----[ Recap ]----
Game: Sudokube
Dev: Blank
Tools: Unity
Web: @PhilBlankGames
Progress:
+ Steam Leaderboards work now
+ Minor bugfixes
- I really want to move on to a new project already

Does this look okay?

Guys how to make lifelike NPCs?
I want to make a game where NPCs aren't single-purpose(like shopkeepers/questgivers and shit), they'll interact with the world the same way as the player and they will "play" the game autonomously, like the player. How do I make its AI? I have a feel that regular finite state machines will not cut it, they would be too robotic and preprogrammed, I want them to surprise me doing things that I didn't code explicitily. What tech would I have to use? Machine learning maybe? I have no idea, please point me to some directions, this is my dream game.

R a d i a n t A I

goal oriented AI

Made plots a lot nicer!

GOAP
Read up on FEAR.

transform from a piles of blocks into a block stack

By the way you are talking about it, it seems you are biting a way bigger bullet than you can take, man.

welcome to CYBER HELL

started something for spooktober

I like the animation in arms and legs, but the thing in the chest does not look right.

This . That said, you could make a "career" out of experimenting with AI. Keep the rest of the systems in your game constant and relatively simple (sounds like just generic RPG shit desu), and start building up to your dream game with different AI techniques. Also, you can make smaller games along the way that focus on single techniques in small contained games.

What's the next gaming gold rush?

We got minecraft, Moba, dark souls, and battle royale so far. Are indies forever stuck in platformers, survival crafters, metroidvanias, shmups, and turn based RPG?

What is this? Some kind of artificial life simulator?
This tickles my autism

You forget roguelites. The next big one is XCOM-likes, since Nintendo made one. But you're too late to get in on that. You're better off finding an under-explored genre and making a game in it, hoping that it's the next hit genre.

VR porn games. Trust me, I come from the future.

ah, yes, tennis, my favourite game

kek
guess you're right
gameplay needs tweaking

Maybe not strictly XCOM-likes but I could see CRPGs taking off, what with divinity being so popular.

>What's the next gaming gold rush?
If anyone here knew, they wouldn't tell you.

I mean just add pretty patterns, bombs and it could be pretty good on its own.
I'd do some spooky synthwave for you if I had the time to.

Why? There are maybe 10 devs here who are competent enough to execute a genre game in time for the rush. None of them are going to do it just because one user gave them a hot tip.

Exactly! Try it out TODAY!
youtube.com/watch?v=7tdKj3784Lo
bela.itch.io/blob

Gonna switch from cute to spooky for a while, since it's October and all. There will be more cuties in the game, but the time for calcium has come.

Any competition is too much when you're trying to be the next RoR or Minecraft.

----[ Recap ]----
Game: Gunsun Zeros
Dev: GearSting
Tools: UE4, Blender
Web: gearsting.tumblr.com
Progress:
+ added health bars to the enemies
+ weapons now have their own charge bars so they charge even when swapped out
+ I bit of ui animation
+ fixed the robots' aim going off slightly when they walk
+ fixed the robots not strafing around obstacles properly
- this is for last week as well
- wage slaving really cut into my days

It's going to be my game, mark my words.
No I'm not posting it I don't want to feel ashamed when it's not the case

The game I want has a pretty big scope, in the sense that NPCs can live a life, not just do some actions in a contained system. Something like Mount&Blade, X3: Albion Prelude and Dwarf Fortress where they can roam around the map, interact with the economy, make armies/fleet, conquer territory, marry, etc.

I want to make something where NPCs are relatively unique with personality traits(something like this civilization.wikia.com/wiki/Genghis_Khan_(Civ5)#AI_Traits) that the AI takes as parameter to generate the NPCs behavior.

Of course I don't want to start trying to make all of it in one go, a very simplified prototype that shows the principles working would motivate me to keep expanding on it, but I have no idea where to start learning how to make this kind of emergent AI.

"playing" the game with the player would be too much, but games like rune factory and harvest moon do have better NPCs, even if their behavior is simple and only follow a preprogrammed schedule, you can mix a lot of approaches, GOAP would be overkill, and it's close to impossible to test it, it can come up with the craziest and weirdest ways of solving problems, if you can abstract things a bit, behavior trees are the best, they're predictable (as in easy to debug), and are the standard for the industry, utility AI is good for simulating feelings and priorities too, you can have a behavior tree that is controlled by the utility AI.
for example, you can do a simple tree like this:
if hungry > get food
the "get food" action is actually a complex tree, it can check if the npc has food on him, or in his house, if not it will check if it has money and get it from the store, if not it will hunt or something, all of this can be executed with simple actions (move, get...), but from simple actions you can create something that is a lot more complex, you just need the right abstraction of the game world.

I think what you're looking for is Utility AI

>NPCs can live a life
>roam around the map
>marry
>personality traits
literally sims

Nice Bokube clone

looks fun, i like deflecting bullets

I don't think that analogy holds for both cases. There are two kinds of games that are relevant here: the founder of the genre, and the followers.

The founder usually makes a huge profit, enough for a solo dev or team to live off of, but not necessarily the most of any game.
Then the first few follower games also get some significant cash. When you're looking for the next, hit genre, you're aiming to be one of the first followers, NOT the founder

RoR was a folower. Minecraft was the founder. All the other voxel games (and Terraria) were followers.

Even if you KNEW in 2008 that Minecraft would be a hit, you wouldn't necessarily have been able to execute on it. It's really hard convincing people to be successful with something that's new. But you sure as hell could have made Terraria, because "Minecraft but 2d" is easy to understand once people know what Minecraft is.

>not just do some actions in a contained system
I don't think you understand how simulations work.

What about now?

>RoR was a folower
To what ?

First time dev here, I've been trying to make Pong in GM and even though I've reached what I think is a good result, I'm not satisfied
Fuck on collision events, gonna do that shit myself

Rate my GUI layout. In my opinion, the icons in the inventory itself looks a bit large, and maybe they should be in a list format to display names as well. And what do you think about dropping items on the ground, probably through drag and drop out of the boxes? Do the boxes have to be smaller to make it easy? I'll add some details to the screen later if needed and tweak the colors, gotta get the layout down first.

Roguelites. You could also call it the founder to RoR-likes, but that's not a large enough genre that the followers are also hits.

If you aren't satisfied with your pong then you will never be satisfied with GM, it's basically Compromises: The Engine.

What I meant is that it wouldn't be an AI just for "here, shoot up some dudes", but a broader scope, like in Dwarf Fortress for example.

>it can come up with the craziest and weirdest ways of solving problems
This is what I am looking for, emergent behavior, not just following a state machine

I'll take a look on utility AI and GOAP and see if it fits what I want

what?

GAMEMAKER IS NOT THE ENGINE YOU WANT IF YOU WANT AN ACTUALLY TECHNICALLY SOUND GAME WITH CLEAN CODE
IT'S FOR GETTING SHIT DONE AS FAST AS POSSIBLE AND NOTHING ELSE

>It doesn't matter what engine you use!
>Oh btw X engine is shit, don't use it
Can I get some consensus here?

It does matter, people who say otherwise are engine social justice warriors.

>What I meant is that it wouldn't be an AI just for "here, shoot up some dudes", but a broader scope, like in Dwarf Fortress for example.
So use GOAP

Hrmmm....things may be getting a bit out of hand here

It's almost like there are different people in this general.
It matters (I mean why the fuck would you use GMS for 3D, or UE4 for 2D ?), but not as much as people pretend it does.

The consensus is and always has been that GM is shit BUT it gets the job done.