/agdg/ - Amateur Game Dev General

Just like make game edition

> Next Demo Day 19
itch.io/jam/agdg-demo-day-19

> Play Demo Day 18
itch.io/jam/agdg-demo-day-18

> Helpful links
Website: tools.aggydaggy.com
Weekly Recap: recap.agdg.io
AGDG Steam Games: homph.com/steam
Fanart and stuff: drive.google.com/drive/folders/0B6j4pcv3V-vfb3hKSlhRRzlLbFE
New Threads: Archive: boards.fireden.net/vg/search/subject/agdg
AGDG Logo: pastebin.com/iafqz627

Previous Thread: Previous Demo Days: pastebin.com/74btH1aJ
Previous Jams: pastebin.com/mU021G8w

> Engines
GameMaker: yoyogames.com/gamemaker
Godot: godotengine.org
UE4: unrealengine.com
Unity: unity3d.com

> Models/art/textures/sprites
opengameart.org
blender-models.com

> Free audio
freesound.org/browse
freemusicarchive.org
incompetech.com/music
fantasymusica.org

> How to Webm
obsproject.com
gitgud.io/nixx/WebMConverter

Other urls found in this thread:

blogs.msdn.microsoft.com/jonathanh/2005/05/20/optimizing-managed-c-vs-native-c-code/
twitter.com/SFWRedditImages

Describe your game with a single picture

sage because I like the other thread's op image more :^)


Bokube is 100% sure it's bad graphics card.


Radeon is AMD, not intel, and it's able to run Assetto Corsa at 60fps at medium settings and native screen resolution. Put your name back on, Bokudev, then go fix your game. My GPU is not the problem. Your shitty code is the problem.

...

...

...

Reposting here in the PROPER THREAD!


might've gone a bit too far with the dynamic lighting, considering the rest of the graphical style, but it FUCKING FINALLY WORKS HOLY SHIT.


Inb4 this would've taken like a minute in Unity.

I'm really proud of it.

...

...

Looking great user, keep it up

I didn't forget to post a pic.

...

What ever happened to that black and white game that had small planets, it was 2d and a few of the webms showed entering a rocket and flying to other planets, going underground and talking to some tentacle or something.

A face portrait of a girl was a reaction pic around here for awhile.

Procedurally generated?

This would have taken a minute in Godot.

I was going to say "looks good, keep it up", but the other user already said it so now I don't know what to say.

what if I'm just brainstorming and concept arting to procrastinate on learning how to program?

>watching sailor moon
>tfw sailor v game is ded

...

top kek

you should make the map like this

I think he's still working on it. I don't remember his name though so I don't know if he has any social medias.

which one is better for performance? scaling texture in game engine or in shader?

>amateur gamedev general
>amateur

Either trolling or someone who is not him

Both, but there are also professionals here

...

Probably in game engine, since that is basically "pre-set" information and a shader needs to work in real time.

Depends on your game though, the different will probably be negligible at the end of the day for most people.

I hate it when the boxart of a game looks better than the actual in game graphics.

...

Flipping them around disables them now, the transition to upside-down probably has to be improved

...

I mean as in "game engine", I scale it in the game logic. I use my own game engine bt w, if you're wondering.

That's breddy cute

>Bullying simulator
This knight is a dick

Cheeky cunt.

demo when

You should always pre-bake whatever data you can. If you need the scaling to be dynamic, then you might have to do it in shader, but if it doesn't then you should absolutely do it offline.

Avoid making any unnecessary code in shaders. Every operation in a fragment shader, for example, needs to be performed per-pixel in the scene. It can get very, very expensive very quickly.

Ahh I see. I would still say the game logic in that case, since you still only have to set it "once" and then forget about it. I really can't pretend I know what I am talking about though if you're doing your own engine.

You'd better keep a hidden bully count and send in the anti-bully rangers if too much bullying is detected.

that attention whoring retard who can't into anything

Anubis

Underspace

Project Wingman

What is a good setting for a Dwarf fortress/Rimworld clone?

My ideas are:

>Vampire coven
>Necromancer cave
>Wild West frontier town

What's stopping you from making them all?

>Dwarf fortress clone
It has a proper UI and multi-threading
>Rimworld clone
It's good

Wild West frontier town. There's not enough wild west games.

How do you go about achieving this artstyle? Is the Sims 1 real 3d, or is it faked?

reposting in actual thread

added a charge attack, some screenshake, and a new crosshair

your ideas are generic and boring as fuck

yeah, maybe I will need to scale my texture down a bit
same thought, tb h. changing variables in shader program is a hassle.

thanks for advice. everyday I am making progress and one day I will have my own version of Knights of Honor

characters are 3d, the rest is pre-rendered

>enemies bounce around like crazy
top tier game right here.

A Wild West frontier town builder, with a vampire mechanic where your whole town could become a vampire town.

sims is real 3D with 2D items. most of the furniture and such are 2D while the walls, doors and people are 3D.

How does it look

Yo guys

I want to make a game kinda like the gameboy zeldas with a jrpg kind of feel but with no leveling up or combat, except that in some sections I want it to play like a simple sidescrolling shoot em up.

Do you think something like that can be easily achieved with rpgmaker or should I go with gamemaker2? (dont bother suggesting godot)
Keep in mind I'm a bit of a brainlet and have no programming experience. Basically I'm asking if the extra work and difficulty of gms2 is worth the extra freedom and flexibility or is rpgmaker good enough for this sort of project?

Reposting because you faggots forced my hand by creating more than one thread.

Why people say you don't need to learn C++ to create a game in UE? Won't that make your game an unoptimized mess? Also if you know almost zero about coding is it possible to start learning with C++ or do you HAVE to learn C first?

Left takes no talent, while right takes a lot of talent though.

Remove your name. You're not the only guy here who uses WebGL, bucko.

Oh that's not a map, just debugging the lightmap.

Thanks I guess

yeah

thanks

>left takes no talent

>Why people say you don't need to learn C++ to create a game in UE?
because blueprints are enough and C++ is a crashy mess, you can compile blueprints down to C++ for a performance boost once you're done with them

oops sorry forgot to remove it
we cool now?

hey wait, that's my game a couple months ago

Because Blueprints. You don't need to know C++ but you need to know proper coding practice. My pathfinding is in blueprints and works ok.

Blueprints aren't nearly as sluggish as people make out, unless you're doing advanced shit it makes no difference, and by that time you can learn C++ or look into nativisation.

>is it possible to start learning with C++
no
it's literally impossible to learn coding

I would say just bite the bullet and actually learn some programming - then do what you want to do in something like Unity.

I know that it's exciting and you want to jump right into making stuff, but I promise you that in the long run that learning how to program will help you tremendously.

Excuse me I said coding not japanese.

Most games will never need the level of optimization you'd gain from using C++. And in the rare cases where you will, you can just make part of it C++.

Has any 3d game ever done a properly animated large snake?

Resident Evil is the closet I can think of, but that snake is on rails. I want a big ass basilisk enemy that can slither around.

isnt it essentially the same?

Snake Pass

like a black screen

>installing UE
>130.6 out of 128 completed
I'm seeing this is going to be a wild ride.

Yeah, thanks.

C# is faster than cuck++. people only cling to the fiction that it's fast as a way to cope with the fact that they're wage slaves forced to use it by management.

Snakes in space counts?
There was one dev with some 3D space arcade thing

Won't that make your game an unoptimized mess?
No. Its been shown many times over in rigorous tests that for most routine tasks, blueprints perform only marginally worse than direct c++ code. There are obviously certain specific tasks that you might devise some genius algorithm to accomplish that can only be accomplished with the lower level c++, but if you are new to programming, that is unlikely to be the case.

go with gms2. any sidescrolling shooting you somehow manage to make in rpg maker is going to look pretty and play pretty awful. The engine is just too weak to do a lot of stuff with.

Leviathans in Subnautica were done well imo.

>Unityfags will fall for this meme

is it really that dark?
the only thing that is supposed to be visible are the seats and text

Coding makes sense for humans.

>C# is faster than cuck++.
not even close, it's like 30x slower

this kind of SPEED is only necessary if you are doing some serious large-scale simulations though

C# in unity is slow because the unity api is slow. if you don't use the api it speeds up immensely.

>People here are making games so high end that the difference between c++ and c# actually matters

That is one beautifully animated snake. It looks like the player just controls the head, the body follows with the spine moving horizontally in a sinusoidal fashion?

blogs.msdn.microsoft.com/jonathanh/2005/05/20/optimizing-managed-c-vs-native-c-code/

>A line-for-line translation of the original C++ code into C# ran 10 times faster than the C++ code.
>It took five different optimizations (one of which introduced a bug) for the C++ code to match the speed of the unaltered C# code.
>After Raymond's sixth optimization, his C++ code finally beat the C# code — because the runtime got down to where the 60ms startup overhead of the CLR made a difference!
>To accomplish this, Raymond had to:

>Write his own file/io stuff
>Write his own string class
>Write his own allocator
>Write his own international mapping

>C++ is slower than C#
>No it's not
>it dosnt matter fagget

>not even close, it's like 30x slower
>like 30x slower

How about you stop bullshitting.
It sure is slower in general but nowhere close to being even 5x.
In fact, you can properly compile c# in native code bypassing IL and it an be pretty fast. Still not faster than good C++ may be, but you have to consider it still better language in terms of usability and miles ahead of C++

I am not that guy, but yes I am saying it doesn't matter m8

So GDScript is basically C#?

I just DON'T WANT my game to end up in a situation like enter the gungeon, where despite the simple graphics it runs like shit, or assassins creed 3 where the GPU is barely used and for SOME reason the CPU bottlenecks everything all the while the game ignores all but one of your cores.

Where is the real gameplay ?
Seems like the only thing you add is silly meme shit for laughs

>is it really that dark?
yes, i couldn't even tell those were seats. Text looks like it's in the middle of fading in

IL isn't an interpreted language. it gets compiled.

That's exactly what I don't want to have. The leviathans were incredibly stiff, the moved awkwardly, and they had terrible path finding and collision. They get fucking stuck on everything.

Looks fine to me

It runs in VM retard
It is very slow.

if you need pathing for something really out of scale with everything else you should probably generate a separate navigation layer just for them

Your performance is going to be limited by your own shitty algorithms, not by blueprints vs. c++.

real thread over here guys

30x is an exaggeration but C++ is easily 2x faster than C# in every application, one stupid test from 2005 proves nothing since there are thousands of other benchmarks always putting C++ ahead.

And C#'s biggest issue isn't performance, it's the horribly high memory usage and garbage garbage collection.

If a game like enter the gungeon is running poorly despite being written in C# (Unity) or something like that, then it is more than likely just bad programming by the developer. When you have a full scale project, a lot of developers tend to forget about shit they left in, or once something works they think "fuck it, that'll do" and don't try to improve on it. This adds up to the final product not being as efficient as it can.