/agdg/ + /vm/ ~ Amateur Gamedev General

those are called installers + dynamic linking

The point is being moddable by changing the .txts

Its not hard to mod
it is hard to read :^)

Can you work within the free limitations with Unity or Xenko Framework or should I try to find a cracked open version? I have never gotten a straight answer to this question and because of that I keep floundering between engines.

For Unity yes, dunno about Xenko

In all seriousness, you should really reformat that file, it looks like such a pain to modify

Probably will before release, im just afraid the Json reader will glitch out if i add tabs or spaces

I am just curious how much of the game I actually own when I make it. I heard that you need to pay a subscription if you actually want to sell your game outside of a few exceptions. Thanks for the heads up, I might stick to Unity and see how far I can get.

Im not sure now, gotta check that again

if you are worried about licenses you should consider looking at godot and see if it'd work for you

Occasional OP here. I'll dump some logos for aspiring OPs so no one complains like that user in the fae thread. Don't let a lack of imagery stop you from becoming the biggest faggot.

...

These and the ones in are more meant as second images when you can't think of a fun edition. Except the christmas one of course, that one's for christmas.

Here's a few more.

Thanks user, I feel like I can sink my teeth into this engine!

...

Here's a vector of the AGDG logo

...

Saved them

Oh yeah, webms.

when was this made?

If you only care about 2d then you can check out love2d engine.

Unity, Unreal, and seemingly Xenko all work on a similar business model.

You're free to use them and sell games you made with them, but if you make enough money with your game in a certain period (for example 200k in a year with Xenko), then you have to either upgrade to a paid version or pay a certain percentage in royalties.

All useful features (to indie sized studios) are offered in the free version of unity.
The only useful thing about the pro version is full access to the source code (beyond what you can get via decompiling, which is a-ok btw, they encourage it).
The only reason that one would get a plus or higher license is either for removing the unity boot logo, or if you're required to get a license due to surpassing the revenue limit (pass a 100k for this year, you must at least get a plus license).

Anything made in engine is by law your intellectual property, and if memory serves they have a very similar wording in their EULA.

Xenko looks pretty quality as far as I can tell.
Once they get the same level of OS support for compute shaders, and a better compute shader API; I may try it out.

Thought I should share this, its Bisqwit's "Parallelism in C++" Tutorials. Its a pretty useful topic.

youtube.com/watch?v=Pc8DfEyAxzg
youtube.com/watch?v=MfEkOcMILDo

...

Ages ago, back on 4chan

Yeah, Bisqwit is a torah-following turbojew. Sucks, can't bring myself to watch any of his shit knowing that.

I thought he was just a "really devout" christian or something. I don't actually know.

Anyway if you can get past the first five seconds it stops being so Semitic and turns into a pretty good programming video.

well if he says shalom, it's hebrew, and is a language largely practiced/used by jews in the modern day (as otherwise the list of practical uses is very short).

youtube.com/watch?v=xVxSAi4sueQ

This is probably the most he talks about it. But he says in comments that he is a christian.
Really since the bible is a semetic work (written in Hebrew) its not far off. But, I suppose this line of posting is just going into the semantics of it needlessly, since I don't disagree with anyone.

I implied what you're saying there, alas, I was too lazy to type it all out… i'll clarify:
Meaning, theological purposes, translation, and other such things.

Seems he's christian, but he uses the original hebrew writing which is very uncommon for most christians nowadays; as it's considered highly impractical to learn an otherwise dead language.

Did a bunch more props for the church.

BSR stuff?

Looking nice

What's BSR.

Bavarian Scout Ranger, a steampunk project of my client, basically like EDF with gothic lolis I think.

that looks very nice, user

I was forcefeeding my faeries with health potions to keep them from taking starvation damage, but the apothecary started to fall appart due to lack of infrastructure maintenance and now everything is going to hell
I feel like im one step closer to becoming Toady One

Hey everyone.
I have a bit of a weird problem- I'm better at character design and modeling than figuring out 3D environments. OK, my character design and modeling might not be anything special either, but I at least have good mesh topology and I like how my model looks.

There's several problems to tackle when making 3D environments for games- level design, modeling, texturing, materials, lighting, etc…
What I could really use some advice for is the texturing (maybe it's worth tossing in materials as well since they're related to how textures will end up looking).

What advice and/or learning resources would you give to someone who is looking to learn how to create nice-looking environments?
It seems like it's a vast subject that requires a lot of different knowledge and skills- architecture for buildings, ivy and trees can be done with ivy/tree-generating software but what would be the best approach for texturing them?, grass can be done with a particle hair system (but a tileable grass texture ought to do just fine as an indie dev).

It seems to me like as an indie dev, a nice approach would be to have a set of tileable textures that work well with each other for most of the environment, with separate textures made for each prop (which should be kept to a minimum given small dev team size).

What do you all think?

How is Godot for 3d?

Really bad at first then it gets better but then worsens near the end and when you know all its fucking shit and you have wasted your time

look at stuff like GDC presentations, siggraph papers, and such things.
Plenty of material to study there… however, there's many different approaches; so it really depends on what suits your specific needs.

Also, lighting model, and in general post-effect are a stylistic choice.
Though PBR makes certain things extremely easy (mimicking metallic, plastic, rubber, and such appearances), and should be taken advantage of.
Utilize the Allegorithmic/quixel (w/e u prefer) software for creating your PBR materials (lighting settings applied to textures/normal map/etc), and procedurally creating the textures/normal maps/etc to apply the lighting settings to.

Are modders gamedevs?

What's so bad about it? Weren't they working on improving the 3d rendering? I haven't kept up with it in a while.

They require a similar skill set, but game devs use original assets and code excluding boilerplate code like in game engines.

Everything is fucking shit
The controverted way of making things it has plus the shitty comunity and the way of handeling images

3.0 (which has the rendering improvements) isn't out yet. A lot of the features were/are pretty buggy. I remember scenes kept re-importing, preventing me from making changes. Hopefully they've fixed that.

I think the communities pretty great. No worse than Unity, surly. What's wrong with the images?

Looks suspiciously similar to the churches in Resident Evil 4 and Dark Souls

There's different arbitrary tiers people will go with for being considered one or the other, but eventually each of these categories start to blend into one another due to the reality that each has similar skill sets required.
So, the real answer is that it depends on the specific situation.
Taking into consideration the skills used, tools built by themselves or provided, what exactly they're doing, in addition to how much is original assets/story/setting vs. in-game assets/story/setting; also some other stuff I'm not mentioning.

The graphic system is retarded and the comunity is more acidic than Lol's

...

I am by no means a pro, but i do it like this:

Most level design tutorial guys either charges a massive amount of money or is some faggot who will talk for 20 minutes with no substance and can't describe why they're doing what they're doing because muh art. Might wanna check out cgpeers for actual level design tutorial videos. Also check out some youtube speed level design to get direct insight of the process.

...

Man, I haven't touched Twatter in a long time. Not much ooh ahh artsy progress to post, and the only other use – virtue signalling – doesn't appeal to me.

Nope, that's not me. I think it's the epic tavern guy.

I learned the most while playing around in UE4 editor and figuring out what best to use standard architecture shapes, where to use BSPs, and where custom props would be needed (custom props of course, needing a specific UV mapped texture). However custom props can be made modular / tileable as well, and reuse some of them and the levels start to have a unique commonality that starts to scream they belong to the same game world.

No its my waifu sim. I dont post often.

OK, how's the gameplay on your waifu sim?

There's no gameplay yet it's just models since I'm currently nearing completion of my other game. I work on models at work since I have loads of free time. Once I finish my first game I'll download and start using UE4 for that 3D game. I also watch videos of it in my spare work time. Here's a video of some old ice map test.

Yes. In both cases you're developing a game.

If you make an original game in Doom's engine, are you a game dev or a modder?

If you make a mod so extensive that it transforms the game into a different game (see: DayZ), are you a game dev or a modder?

It doesn't matter, because you're doing essentially the same thing as a pure game dev is. A modder can be a game dev, but a game dev can't be a modder.
that's not the point.

"it's retarded" isn't an argument.

I guess editing an ini file is modding the game then too. It's the same thing, you're editing the files the game uses to modify its behaviour. Makes sense to me! :-)

Looks interesting, not sure about the control scheme, I mean, is it an M+K kind of game or made for a controller?

It's a matter of data vs. code.

"Modding" refers to the modification of both game data and code.
"Game Dev" refers to the development of both game data and code.

If you're only modifying data (and text data, at that), you're not much of a game dev or a modder, and whether you're a game dev or a modder, it's going to be the combination of good data and code- art and engineering- that will make you legitimate.

Are you per chance the autist that keeps convincing /agdg/ BO to delete everything because it "isn't game dev"?

I make mods for RimWorld, which supports injecting compiled code into the game via loading .dlls at runtime. But it's still not game dev. It's a matter of, "Did I design this from the ground up?" or "Am I using someone else's pre-designed game and changing/adding things?". Even total conversions are not game dev. They're still relying on the underlying mechanics that were designed by someone else.

If it buttfrustrates a moronic faggot like you then I'm the BO himself

I'll take that as a yes.

I'll be using this MH character as dummy for gameplay and animation tests until I can make a lower polygon model for easier sculpting, but I want her to look nice, any hairstyle recommendations? post pictures if possible.

Can Krita replace photoshop for texture work?

Sure. It's slow at large resolutions though, so it may be uncomfortable if you're doing 4k HD™ textures and want to use huge brushes or something.

It has normalmap painting brushes and wrap around mode which make it better than PS in some ways.

...

She looks kind of fucked up, reminds me of the main character from Fable with those fuckhuge eyes.

So if you make Pong or Flappy Bird, that's not game dev because the game was designed by someone else?

Yeah, MH is not quite good when for making stylized characters from the get go, but this will do for now, I can try to fix the eyes a little but I really can't think of a good hair style right now.

Her face needs some work. Mostly the jawline. Feminine faces aren't necessarily so thin, too. Some eyebrows would also help.

OK, that can be fixed in MH, I just need to re-export.

Her neck looks like a pyramid.

Cuz she was a kwaanzs and shieet nigga

Fine… I'll make a new one with less tampering and less stylization, sometimes I wonder if having nearly 35k polygons would work instead of having to make a base figure myself.

Even using using an OS you didn't write makes it not game dev cause your still relying on someone else's code :^)

Finally, another thing to add to the agdg humanoid 3d model wars.

Have you written a compiler?

What about shovel-chan?

I can't believe I forgot about that.

How do you deal with the fact that at any moment, you could come across a project that's almost exactly what you're making, except better?

Most of my game ideas are based on other games that suck in one or many ways. Games that I want to exist and be good but aren't. If someone made the game I wanted to make and made it well, I'd be happy, but I highly doubt that can happen.

There's only 1 game that I like and wouldn't know how to do better or twist into my own thing, and that's Factorio. Everything else just frustrates me because all I see is a million things I'd change or do differently.

Happened to my dad. He spent several years working on an invention just to figure out that a mutlimillion dollar company already did it. He's still pursuing it though.

Stopped making my game and played theirs for 1500 hours.

invest insane amounts of time and effort is your features, no one will make a cooking system so over the top like mine

By being that project that makes other wannabes quit.

in this modern era of ienumerators and LINQ, for loops seem absolutely antiquated

Truly a topic all game devs should concern themselves with.

Too blunt? The flying island name is already "Migas Town", should i remove the last part to make it more subtle?

If you are able to access a collection with indexing and need that functionality, arrays and fors are still the absolute best for performance.

Foreach doesnt expose indexing, nor does it allow the item or collection to be modified (more noticeable with value types). Arrays implement ICollection, so they are iterable in this manner as well (though their .Add() method exists, it is hidden)

Still, for has its uses, and iterators and LINQ is just a tool. Its great for collections and handling db-like functions (personally I like using anonymous functions inline for debugging)

...

They are magicians, not witches

I'd generally be careful with putting current day memes in games that will be done in a few years at earliest. They'll become old and stale when you finally release.

Good point, taking that last part out

this was almost you

Was that supposed to be a jab or just a reference?

Is she a muzzy?

A jab, please don't put memes in your game's script.

She is a corrupt fuck who flooded the empire with Goblins causing it to fall, the only human survivors were in the kingdom of Pollan who ignored the empress authority, closed the borders and erected tall walls

What about references to political events?

You're putting it on far too thick in general, and political events are also current events.

...

I wast to justify the existence of all homunculus races, i needed a war to justify the goblins and the chimeras and also why the nations/landscape changed so much since these nations will no longer exist when the game begins, this story will not be shoved into the player's face, it will be in books in the town's library which you don't need to even enter.

If I add cooking to a game, I will fucking destroy you

Then have the common decency and respect for your work to not draw a crude, one-dimensional caricature of current world events, change a letter here and there, and sell it as the lore of your game. There's nothing wrong with taking "evil ruler destroys country through immigration, except for this country that secluded itself" as the baseline of your lore, but there's absolutely everything wrong with the way you execute it. That's the writing of a leftist hack in right-wing flavour. Flesh it out, add interesting characters and shades of gray. Build a world instead of a political cartoon.

DO IT FAGGOT

That's fine, having politics internal to your game world is fine. Just don't reference anything including people, groups, political events/ideologies, or memes from real life.

Unless I guess if you're making something deliberately political/offensive/memey like "gas all jews"

to be fair, literally everyone told him to fuck off

Thats what iam doing
Its not as oversimplified as i posted here, in my devblog i have the more detailed version

Triggering lefties is free publicity, and i didn't even add anything with religion or ideology, im holding myself from going full anti-commie propaganda

Thanks for suggesting Godot for me though, I really prefer how it's just an exe instead of an installer that tries to get me to sign up in their super exclusive gamer dev community. I tried Love like the other user suggested but I must be a fucking dummy or something because that shit seemed way to obtuse to use.

I used it for like a hour.
It basically fully featured unity and maybe some more, for example there is build-in Sharp editor, so you don't need to switch to studio or whatever you got.

BUT IT'S PERFORMANCE IS SO FUCKING SHIT IT'S USELESS

I don't know wtf they were thinking and whenever someone tested it. And I mean editor itself not even games. Even scrolling text in said code editor goes with freezes. Crysis 3 or witcher works fine on my PC at max settings but soome weeb engine apparently needs more power to display simple scene without materials or a bunch of letters.

u wot m8

...

Love is very minimal. You use pure code, it's more a framework than an engine, not very comparable to Godot and the like.

Okay, that makes a lot more sense then. I thought I was somehow missing the executable that launches the "workspace". Going by that logic then couldn't I just use godot and export the code and then import it on love or are those using two completely different formats that would make it incompatible with each other?

65pok: i've just used ditto to switch in on a slurpuff
65pok: with belly drum activated
Anticop guy: MODS
Anticop guy: hes casual chatting
65pok: for some reason it didn't work
jumbowhales: Lionyx

back to fullderp you poopy

We're all dummies. If you want to easily run your scripts then install a löve-run extension to whatever editor you're using. They let you use a keyboard shortcut to run your active file/project/directory.

It sounds like you're after a visual editor so Godot would probably be more fitting anyway.

Add $2000 nightmare chan and it's golden

Since this is my first time doing this I figured a visual editor with a basic code editor built in would be a good start since I can see how the code changes while learning the basic structure of how the code needs to fit and interact with various scripts and assets. I'm off to learn more while shitposting. Thanks for the help anons!

Nothing wrong with having a political opinion in a game, you just have to be subtle.

So if I replace or edit most of the textures, swap out the music, edit the models, but don't change any programming, pouring weeks or months into the effort, is it still game devving if all the game mechanics stay the same?

Content and asset creation is still development

for loops get unrolled by the compiler (and, hopefully, optimized with SIMD instructions, if you're doing operations on arrays). "Iterators" and other meme concepts just obfuscate what the machine is doing so that you are thinking about the abstraction model that the languages creates and not how the machine works. This is why they are implemented in java and python.

Modern programming is not the newest name python has for what people can already do. Modern programming is the new instructions in the Intel Manual that compilers haven't implemented yet.

Yes, it is gamedev
But you still have to program something sooner or later or it was for nothing

If I make textures, music, and models for my own game, am I game devving?

You can do it like me and do everything at the same time. I can't focus on a single task for too long, so i do programming, some assets, write some lore, more programming, more assets, and it goes on.

You one-up them. Add more features/content or more gameplay depth.

This is the result of telling Vulkan to accept vertices that are formatted in this way:

typedef struct vertex_s{ union{ struct{ vec3_t pos, color; vec2_t tex_c; }; struct{ float v[8]; }; };}vertex_t;

And then passing it things formatted in this way:

typedef struct flat_vertex_s{ union{ struct{ vec3_t pos, color; }; struct{ float v[6]; }; }; }flat_vertex_t;

Pretty fun stuff. I'm glad it didn't crash.
Pretty soon I will have textured things in Vulkan, which will be very nice.

I admire enginefags like you sigmanon
Nice stuff

Thanks! I'm about to have textured polygons in Vulkan, this means that I can begin building the rendering engine based off of: cs.utah.edu/~jsnider/SeniorProj/BSP/default.htm , very soon.

That means, i'll have tons of things to show off soon, instead of the spinning cube. My program is ~5700 lines long, and that's still the most prominent feature…

For fuck sake.

Are you rewriting from scratch?

would not recommend

Neato, what model format are you going to support?

I have to agree on that one, if you want to add reference to the current memes or political events then do it in a subtle way and in a way that fits in your lore so that it doesn't stand out too much like make your own version of those stuff and change it enough so that it isn't too easily recognizable or something like that.

I dislike programming as it is, i do it because the game won't program itself
I do love creating content/assets though
Good luck on yours

yeah, if you download my old code you'll see why. Its on my page: ika.itch.io/redsky

The old code is more of a hindrance because of how it is structured. This new engine is so much better structured, so i'm able to be so much more productive with it, and I cant stand to look at how I made my old engine.

I don't know. I actually don't know anything about model formats. Maybe i'll write my own model format or maybe i'll use something really simple. I don't plan on loading in models anytime soon. At least 4 months.

Daily reminder

The EU/Merkel reference wasn't actually planned, i wanted a justification for the creation of goblins and the fall of the empire since i wanted all major powers out of the scene when the game actually started.
When the ideas started getting into place i noticed how familiar that was and added a bit of less subtle juice on the mix
Same with the Migas Town stuff, i was discussing the lore with my brother and we noticed how "Mages going their own way" was actually happening in the lore, i have nothing against MGTOW btw

I have already changed the Merkal name to Kelmer, hope that helps

This is a pretty bad example as in those developers made a 1:1 copy instead of doing at least of their own version, not to mention its pretty plebbit tier now.

Good.
Oh alright then, but yeah as I wrote it before and as other anons pointed it out do it at least in a subtle manner.

For what purpose? In case your game is going to be moddable then there needs to be people that is going to make a export/import plugin for Blender and unless your game becomes popular enough for a sustaining modding community that export/import plugin will get outdated pretty fast because nobody bothers maintaining it anymore. You could try using the .obj format which is the simplest 3D model format there is with reliable export/import plugin, don't bother with the .fbx one that shit is proprietary and the Blender devs have a hard time doing a proper support for it.

I hid a "David Duke for U.S. Senate 2016" graphic in two places in my last game. I think a few people found it, but the game took so long, that it didn't come out in time for the election.

Maybe i'll use .obj then. I would write it because I like writing all my stuff. You should see modding on Sigma I. A bunch of people messaged me about that, and I linked them to my map editor: 73.246.174.185/sigma_editor.html
(my DNS hasnt linked that IP to ikadev.net yet)

Nobody has made a mod yet. (I wonder why)

Really i'm not trying to make a modding community. I'll have OK tools for it, but, I will just be devoting my time to a more advanced engine after this is done.

Prolific or Perfect?

Isn't a perfect game destined to become prolific eventually?
Sorry im not sure i understand what you mean

Sorry if I was unclear, let me rephrase it:

Do you prefer to keeping make games ad nauseam with no necessary quality until quality comes by practice or do you prefer to autistically focus on one game and the skills will come as long as you keep working on it?

Why make a crappy game when you can work harder make your dream game?
Besides the market is flooded with crappy games, if you make one you won't satisfy yourself and neither will succeed, unless you stupidly lucky like the flappy bird guy.

Go for what I want to do, and try to make it good to the best of my ability. That's how I approach learning anything.

I think finishing at least 1 small-ish game is beneficial though, it'll give you a much better understanding of the process as a whole so you won't run into something you didn't expect, like how much work it is and how many assets it takes and how to program certain things.

Here are my latest troubles: Now Vulkan will accept a vertex that is 8 floats. I am supplying all of these now.

However, things are not displaying, which is way more boring that a bug where it shows off something funny. There are two squares on the spinning cube that display. That's one thing. So, I think this is something to do with the depth buffer, or back-face-culling, i forget what it is exactly, it something where the polygon is facing away (with its normal) and is thus invisible.

That sounds terrifying to fix, good luck

Is too wide, I don't mind having my mistake from MH there, I guess I can get over it, but at least make a collage or something.

I came across Haydee, is too soon to know if mine would be worse or better so I don't care, if there was a better version of any of my ideas I would follow them.

Yes that does sound like a backface culling issue the way you described it

8 floats per vertex doesnt seem that impressive? Even XNA let you define your own vertex formats and even the largest default one, VertexPositionColorTexture was 3+2+1=6 floats in size

Well, i'm slowly adding in the features. Previously it was 6 floats per vertex (3 for position, 3 for color). Now, I have an extra two floats for my texture coordinates, because i'm adding in textures to it. You do define your own vertex formats. I don't know what "impressive" is supposed to mean. It takes a really long time to add in support for more things, so that's why i'm still at "textures" and not something like "normal maps".

Its significantly less terrifying than the bug I had yesterday. I didn't know that there was a difference between "descriptors" and "descriptor sets" and this seriously messed me up for a while until a friend helped me. This was messing up the way that I initialized the "descriptor pool" and the "descriptor set layout" which spat out a strange error about how one of my things was null when it allocated the descriptor sets.

I expect to fix it in a few hours.

I'm not familiar with Vulkan so I didnt realize you were reinventing the wheel. You made it sound like you were implementing new features; I say it wasn't impressive because things from 6 years ago were using 6-8 floats per vertex

Actually XNA uses a RRGGBBAA bit format for colors, so it can squeeze it into 4 bytes, or 32 bits. The colors are represented with a packed int value that gets worked as a float value internally, so they can get away with 1 float per color vertex instead of your 2. But do whatever works, just sayin

Ok, I fixed this bug, i'll explain this because its sort of interesting:

so, in the old vkr_draw_vbo function, this line here:
vkCmdDraw(vkr->cmd_buf, vbo->size / 2, 1, 0, 0);

in this case, the size argument is the amount of verticies. Now, in my previous layout, a vbo's raw data was an array like this:
static const vec3_t tri_data[]

where every other vec3_t would be describing the color of the vertex. therefore there were twice as many vec3_t's in the vbo->size parameter, so to get the real number of vertices i would just divide it by two. Now, my raw data is declared like this:

static const vertex_t tri_data[]

and so, the vbo->size matches up correctly. so, there is no longer a need to divide it by two. So that was all it was. A pretty trivial bug. Its looking perfectly normal now.

Yeah, what you have to do is manually specify EVERYTHING. In XNA they do it for you behind the scenes, i imagine, but in vulkan there is a much more long-winded way of doing things:

void vkr_init_vertex_input(vkr_t *vkr){ //if the program crashes or doesnt do the cube correctly //check this out first.... vkr->vi_binding = (VkVertexInputBindingDescription){ .inputRate = VK_VERTEX_INPUT_RATE_VERTEX, .stride = sizeof(vertex_t), }; vkr->vi_attribs[0] = (VkVertexInputAttributeDescription){ .location = 0, .format = VK_FORMAT_R32G32B32_SFLOAT, .offset = 0, }; vkr->vi_attribs[1] = (VkVertexInputAttributeDescription){ .location = 1, .format = VK_FORMAT_R32G32B32_SFLOAT, .offset = sizeof(float) * 3, }; vkr->vi_attribs[2] = (VkVertexInputAttributeDescription){ .location = 2, .format = VK_FORMAT_R32G32_SFLOAT, .offset = sizeof(float) * 6, };}

This is how i'm translating the vertex struct described over here: into something that the GPU can understand. That vi_attribs array goes to a field of the "VkPipelineVertexInputStateCreateInfo" which goes into the "VkGraphicsPipelineCreateInfo" which you need to call the "vkCreateGraphicsPipelines" function. Thats a whole other thing though.

Are you using Windows 95?

I'm using the "classic" theme on windows 7.

Windows 95 would be better if Open Watcom supported compound literals so I could compile my code targeting it. Also does anyone make Vulkan drivers for windows 95? Probably not.

I guess they meant that the compiler runs faster because one tab is faster than 4 spaces??

Are you that guy who made that Doom clone?

These are my games: ika.itch.io

I'm pretty sure I am the guy you are talking about.

it's false either way. Once code is compiled, tabs and spaces at the front of a line are completely discarded

Right but I'm saying it's a completely useless fix, but technically it would be less for the compiler to parse.

And those churches look suspiciously similar to the churches in Europe.

What's my best option for developing a simple 3d game on Linux?

Suicide

Use Windows.

Unity, then pray you will be able to export to Linux without everything falling appart

Godot supports linux I think.

Start with the Xlib documentation

The rest of my workflow is already in GNU+Linux using Free and Open Source tools like Blender. Unity is for niggers and isn't anywhere near as good as proprietary UE4 chinese royaltyware anyway. Godot 3 will kill Unity and Monogame simultaneously if it's anywhere near as good as the hype suggests.

Godot is Linux native

There is no amount of "you're doing it wrong" to describe this

What advantage is Linux over Windows besides meme value?

Linux is a huge pain in the ass, most of the features are not even documented correctly (have you tried writing anything on X?) And the text editors are all shit. You're literally asking people to stop developing on the platform almost everyone uses and will play your game on just because there are one or two people on these threads who like to act like they are better because they use Linux but then program everything on a higher abstraction level and cant see how much of a mess Linux is compared to windows. POSIX clocks are broken (and sparsely documented) and x86 - x64 compatibility is also broken since nobody installs all the libraries. Its nothing to be smug about.

seconding godot, but be aware that it is single threaded and that you can't mix animations.

for starters you can't get distracted because it has no games, also workspaces

I have workspaces on windows 10. Is that all?

vim and emacs are shit? lol what
whats this nigger talking about?

well, you have games so you can get distracted.
Case closed

...

How many fucking servers are running Windows Server, nigger?

You can get put it on pretty much anything. An old laptop will run a new version of Linux better than it runs Windows 10.
It's free
More customization options
Updating software is easier
Less malware to worry about
A sense of smug superiority
The Unix philosophy
Doesn't actively spy on you
You can choose a distro based on what you value. Want stability? Debian Stable. Want to run a server? Fedora/Centos. Have too much free time? Arch Linux. Want something easier to use? Linux Mint or Ubuntu. Want to actually learn Linux, not just use it? Slackware. Penetration testing or acting like a total edgelord? Kali Linux.

I'm not saying there aren't any disadvantages, there are plenty mostly with software availability, but it's not like there aren't any advantages either. It's really just what you are comfortable and productive using.
OS wars are as stupid. It's a personal decision.

Not going to waste a week learning lisp so that I can configure emacs to use tabs and stop aligning my shit incorrectly. Vim at least has copy-paste on gvim but it just lacks all other features that I use and constantly makes "yourfile~" files that clutter up what i'm doing. You can waste time memorizing shortcuts but using an alternate configuration does not make your text editor better.

Nobody plays games on a fucking server, you retard. You know and I know that the vast majority of people play games on windows. Playing dumb doesn't change that.

meant for

How do you guys deal with not having time to implement ideas quickly enough?
I have a huge backlog of TODO lists of stuff i want to implement, can't make it fast enough.

you can literally configure them to your liking if you fucking desire to do so, I've seen plenty of people who use either emacs or vim on windows, stop being a baby
let me guess, booting into windows fucks up your clock on linux. As we all know, windows plays real nice with their competitors…
what?

Because it takes a while to learn or configure it means it's shit? You're a retard. Also you can stop vim making temporary files.

Are you using win32 or something else sdl or glfw on windows?

Linux doesn't come with x86 libraries on and x64 system, so, it causes problems if you're distributing your things. Unless you can prove anything about Microsoft doing that, there isn't anything to say about it. My main problem with it is how shitty the documentation on POSIX and Xlib is and how oblivious Linux shitposters act about these kinds of things.

Having a special snowflake control scheme means that if you are trying to do anything on Linux coming from other systems where text editors all act the same, means that Linux text editors are just a huge waste your time when trying to get normal things done. It would be fine if those editors weren't usually the only ones out of the box and the userbase for them could stop jerking themselves off over it, so you have to waste time looking around for something that's usable.

Literally the only things you two have contributed to this thread is shitposting about Linux. I doubt you know anything about how fucked it is if you program in python for godot, but if you try to do anything like work with X11, you'd see that its by no means a superior system. I had to port my game over, and my code for linux is like my windows code, except instead of reading documentation (there is barley any of it on linux) I had to stick it together without really knowing what kind of things would bite me in the ass.

linux is a kernel you dingus, what linux ships or doesn't depend entirely on the distro, in most they are easy to install(and most systems, and I'd expect even more so systems using linux are 64 bits, specially if they are being used to game).

vim can be used in windows as well.

did RMS bully you, or something?

Just move them to the planned DLC

I'm talking about Linux-based Systems. You know what I mean by that and playing dumb so that I have to waste time telling you that doesnt change anything. The only thing I don't like is that the program doesn't immediately work and people have to do further things to make it work. I guess on Linux people are used to it though.

Just because its ported to windows doesn't mean that vim/emacs are common control schemes.

Really I just don't like reading (1) shitposts like . There is using Linux, and then there is being a smug autist about it. If you program for both Linux and Windows you can see that the API works really similarly for most things, so the main difference is the shit documentation. I dont remember why i've been writing all these angry posts about linux, actually.

STOP

Wrong.

I actually have two free expansions planned already
Im already making the icons for the new items, can't stop now

Looks like I fucked up the shaders, and everything is black. Or maybe that's what a blank texture looks like in vulkan? I should find out.

it's not like it is odd for games to ship with their own libraries, when talking about steam they usually come with them, and when the game is installed from a repository it pulls the libraries as it needs them, obviously windows has better documentation with the amount of money and full time workers it has behind and the fact that x has been dragged for decades. At this point i dont care about x documentation because it isn't going to be fixed at this stage(and most people here use someone else's engine so they don't have to meddle with it), if anything I'm worried about wayland, it needs better documentation if it wants to start with the right foot.

I recall you can separate some things into a different thread, I don't remember what in particular but I don't think I'm making shit up.

time to pass an actual texture then, boyo

This is my first time seeing a texture on here, which is pretty cool. My problem was this, in the shader: "outColor = outColor * texture(texSampler, outTexCoord).rgb;" should have been: "outColor = color * texture(texSampler, outTexCoord).rgb;"

Now, I just need to make it behave like I would want it too. Right now its all over the place, but I can mould this into what I want.

damn sigmanon you work fast

my problem is that I am the engine-dev, so I do have to deal with these things. I'm always going to be targeting X11 because wayland supports X11, and therefore there isn't a reason for me to change. What I like the best about windows is the MSDN website. If X11 had this then I bet 90% of the issues would be gone.

On most systems my engine works fine except one guy who didn't have 32-bit libraries installed. Also fullscreen mode in X is horrible and doesn't hide your title bar (which, im not gonna fix that bug). Ironically, the engine is more tested on Linux than it is on windows, since all my friends use Linux.

I'm actually not working as fast as you think. For about a week I was maybe posting tiny bits of code on the thread without any new stuff to show, that's because I was writing the function that loads in textures. Then I worked on the command line for 3 weeks, and now i'm coming back to finish this feature.

After I get it working on Vulkan, i just need to get OpenGL up to speed, which is a walk in the park. Its legacy OpenGL using the fixed function pipline so I can just do anything really fast.

fuck, i meant to quote . I keep misquoting people today

I'm going to take a break and then un-fuck this thing.

One day I'll get good at XNA ;__;7

fucking user I remember you spending 20 posts bitching about how difficult it is to do animations in godot and how you regret every decision leading you up to this point and now you're turning around and encouraging people to do that shit on their own to spread the suffering

I really how I can just avoid all of the OS level problems by just using sfml
Seems like even a fairly basic graphics library allows you to avoid a lot of boilerplate stuff

Legacy OpenGL is pretty easy if you want to have a go at it.

Sfml isn't basic in the slightest. Just the interface that you use is basic. Its doing tons of non-trivial stuff that you never have to see.

New dummy I guess, do you think is much of a problem using a 40k polygon model for a game? if not then I will keep trying to make my own base but I still need some hairstyles for this one.

...

Only if you also have lower resultion LODs. Also, please post topology of the entire thing because i don't understand how your thing is having 40k.

It was made with MH, oh wait, is lower than 30K on the base mesh.

If you can't handle arguments against you then don't fucking speak at all you no-dev, faggot.

No, but you're not getting a particularly visually pleasing character model considering that it's 40k polys, the current poly counts you see nowadays are insane, 100k+ is not uncommon, gonna need to get those LODs up though.

These are my games: ika.itch.io
where are yours?

MH, MakeHuman? Anyways, what will the purpose of the character be? Or rather the genre of your game. Generally, the more shit and characters in your scene, the more lowpoly you should go. You should always try to maximize playability as much as possible. If only your main character has 30k, while your enemies 10k, and far distance LOD characters only 2k, then it shouldn't be a problem. There is a list somewhere on the internet that shows the tri limit for modern games. I think the most hardware expensive AAA PS4 games have about 10 million tris maximum per scene camera view.

Where are you games again? I'm not seeing any.

Are you going to be using per vertex animations for Sigma II or armature based ones?

Right now only my animation and gameplay dummy, but I want to make a relatively customizable character for a project, it would be pretty much an exploration platformer with some combat here and there, mostly shooting but nothing like an all out army vs you.

What do when you're the project lead and everyone involved works for free but we also need a guy to work with organics (and possibly animate because I'm shit) but you can't fuckin go around asking because there's no money in it for anyone yet.

I'm legit gonna have a fucking stroke you guys. Where does one find a 17-20 year old aspiring modeler who wants to make a name for himself? That's about the kind of crowd im looking for.

6 people are relying on me to magically shit out a modeler and either find an animator or drastically improve my own animation skills.

picrelated

SHIT-TIER

polycount.com/categories/c44

Thanks, I guess it's worth a shot.

What's your faggot project about?

I don't know. I don't consider it an important question yet because its so far off. I will think about it when I have a rendering engine for the game world, and can walk around in a 3-d level.

Making a non-shit voxel-based sandbox game with a focus on fear of the unknown. Infinite depth mining, deep sea exploration, etc. There's also PvP - we've got plenty of concepts regarding how to make melee feel skillful, how to properly address different tiers of weaponry for different levels of progress for various players. But you know, in these games it's kinda tough making PvP balanced since voxels make shit so much more ridiculously exploitable - PvP is taking a backseat just for now.

It started off as a one-up to the shitshow that Rust has become about a year after scrapping Legacy - but then we realized that there's loads of untapped potential in the genre without autistically focusing on PvP. And then, eventually we've gone from having just a single programmer to having three, one of them being a professional programmer, a proper 3D modeler, a sound designer, and the scope of the project blew up.

We're hastily finishing up our Rust rebalance mod (the thing that started it all) and moving on to this as soon as that is out of the gate. We can't just drop it, though, because we have an agreement with a large Rust server network to let them run it on their servers - it's good publicity to have before committing to something else. The only thing we're missing right now is someone who can model organics. Our modeler likes doing inanimate objects more, and since he's not getting paid - making him do it will just cause him to get burnt out and leave eventually. And it's a shame - picrelated is his work.

that's yet to be seen innit

I can make at that quality posted but I prefer to use Blender and even then

Ok, so opinion question… Today I incorporated a system for weapon/armor durability into my game. I need to decide how to notify the player when a piece of equipment has become damaged or broken.

Some attacks hit multiple times and a weapon could potentially break in the middle of a multi-hitting attack. Further, while the game is turn based, it does use a system of action commands so player timing is important and I don't want to throw it off by pausing the action in the middle of an attack to give a notification.

Your thoughts? Pic unrelated, I just hoped it would get attention.

Looks like I got an idea for the hairstyle, not very original but at least I can try making a decent wig.

Make cracking/splitting/shattering sounds that start randomly occurring at different thresholds of degradation, and make them different based on what kind of weapon/armor it is that's breaking. If cloth gear fully degrades your character's pants fall off :^)

Does an iron crowbar break in your game?

Well yeah, but if you go in with that attitude nothing that comes out is going to be any better than the attitude itself.

I explicitly mentioned that we deviated from that. Rust has zero focus on PvE. We have a LOT of focus on PvE. Moreover, Rust suffers from an insane amount of horrific game design. Hell, so does every single game that attempts to go into that genre. Look at Conan Exiles.

If I had to explain the current concept in layman's terms - it's very close to what would happen if Stalker, Terraria and Lovecraft's work had a baby in 3D. Except you know, without the horrible raw mechanics of Stalker, autistic grind of Terraria or pretentious self-fellating attitude of Lovecraft.

Not at the point where I've gotten into sound (or graphics beyond placeholders), but my concern with this is that it may be overlooked. eg How easily can the player distinguish the difference between the sound of a sword hitting and a sword losing durability?

If it's relevant, the specific system is that each piece of equipment has a number of durability points and attack with a weapon or a piece of armor gets hit, it does a roll to determine whether or not the item should lose durability. When durability reaches 0, the item us damaged but usable (though with reduced effect). If it takes further durability loss, it breaks entirely.

Fantasy setting, so while crowbars may exist (a quick search says they've been around atleast since the 1400's and I haven't decided a specific level of advancement for my setting), they aren't likely to play much if any role in the game itself.

Just don't have something breaking like in other games, specially something that is supposed to be sturdy, compensate by giving it a drawback like slower atkspd.

Two parts to this. One is make the attack sounds and the degradation sounds thematically different. If regular attacks sound make clangs or thumps or whatever stabbing sounds like, then a degradation loss should sound like a crunch, or something discordant and startling.
Second part is triggering the sounds. I don't know how "realistic" you want it, but instead of playing these sounds whenever the durability points decrease, make it so that they have a random chance to play when in a certain durability range, with a higher frequency the more damaged they are.

Just to clarify, by frequency I mean rate of occurrence, not pitch.

I'm not sure, my heart wouldn't be in it but it will probably make some good money so I could use any potential payday I'd earn if I worked on it to put more of the things I want into my own game, do you have a website or something so I could see what is expected from me content wise, or if you don't want to use Blender don't worry about it, I'll take a look at it.

Man you must have a massive list of assets you need made for this game, I'm surprised you pushed this as your first game.

Gonna work on it a little more later.

I finished the shaders for the lava level. Ignore my music in the background.

That looks better.

What kind of collision shape should I be using for the player character in a 3D environment?
Moving capsule (so lozenge-triangle intersection test)?
Moving cylinder (Not sure there's a proper name for the shape formed by a cylinder sweeping along a line segment)?
Something else???

Obviously will need to collide against other objects as well (so lozenge might need to also collide against other lozenges, or OBBs, etc.), but to simplify my question, let's just focus on the player character colliding against the environment (triangles) for the moment.

...

One of my ram sticks blew up and now i'm stuck with 2gb in this computer

stalker was made by a professional group of slavs with access to kraut magic and an actual budget and somehow you expect to do better

There you go, if you're interested I might as well send you the polycount thread I created.

polycount.com/discussion/188200/a-team-of-6-looking-for-a-3d-artist-animator-with-experience-in-character-design

I've got Mezdonian working on our concept art for the environments right now.

Not really. Open world games have insane levels of asset reusability, in a sense that every time we add a new asset, it adds to the entire game world proportionally. There's definitely going to have to be a fuckload of assets, as in any game, but it's nothing crazy.

You got anything to show off?

I specifically said that we'd do better in a very particular aspect. It's no secret that Stalker is pretty fucking amazing, but only an idiot would disagree that combat and gun handling is rather weak and lackluster.

Is your game going to have A-Life stuff? :^)
github.com/OpenXRay/xray-16/blob/dev/src/xrGame/ai_stalker_alife.cpp

wew

...

This is why you're a nodev faggot.

I've written lisp implementations better than that pile of shit.

Playing vidya is like avoiding the problem, but going to sleep is like giving up.

How do I put Denuvo in my game?

If you're asking that here, I doubt you have a game that could suffer from piracy.

truly the only person here who will ever escape this shithole

This is the first time I've ever heard of a ram stick dying. Did you fall for the overclocking meme?

Why contain it?

Liberty Prime was the best

Delete System32

Like I said before my heart isn't into it, I will contact you again in a few days in case you don't get anyone who may actually be interested, I legitimately think Rust is trash. I guess you do too, but I noticed that you played like 1600 hours.

is a very old computer and the ram is even older

Source uses a non-rotating box. Works pretty well but obviously you're going to have issues with diagonal doorways etc. if you don't make them all wide enough.

no man going to sleep is how you solve the problem

I'm posting Stalker because I have a bad habit of posting images with everything I say. I was talking about Stalker - there's a Stalker screenshot.

And no, we're gonna have to figure out our own approach to AI because procgen. Procgen is a huge fucking problem with this sort of thing - has always been, mainly because navmeshes.

Actually this was supposed to be an ironic character, which backfired because people love liberty prime.

I'm not sure the best place to put it, so I'll just say I would play the ever living hell out of a "Rise of the Triad" Style FPS where you go around shooting Antifa

Is there any limit to what they can fuck up?

It's ironic that they couldn't find irony.
It works on a meta level, so they succeeded

B T F O

I think Bethesda will always find a new way to fuck up such as paid mods and Fallout 4 which is a fucking heavily visualized garbage with a brilliant idea to cap FPS to physics. Though one may say that Fallout 3 is already a casualized garbage due to according to a few players it doesn't have the same depth as Fallout 1 and 2 did before. I only pirated FO 3 and the NV one, fuck giving them any money

Working on some programming right now. I've started that part over many times.

There needs to be some kind of object messaging system, and I don't know the best way to do that.

For example, all falling objects will try to move down, if not, roll left, if not, roll right, otherwise stay still. Light objects like keys won't harm anything, but things like rocks or gems need to send a "thud" or impact signal to the object below it so they can potentially explode (like creatures).

I don't want to have a bunch of different subclasses and stuff if I can help it (eg, I don't want Rock::FallingObject), rather define it with behavior; so I suppose I need some kind of scripting system as well?

There is nothing wrong with subclasses
but there might be a better way

No, but I just really like having the ability to arbitrarily assign graphics or behavior on the spot. It's handy

Hey it's me, a big retard.

For potential kickstarter rewards would a separate kind of game design document discussing game design decisions/logic in long form be a good idea for like a "game designer/design enthusiast" tier in combination with a separate art book tier. Or would it be better for it to be a single book that discusses art, lore, and long form game design.

Trying to create as many rewards as possible that have some sort of genuine value without having to put stupid shit in my game.

Your character looks like it has no neck.

Separating content up doesn't encourage donating more.
I'd just put it together with art and lore.

Did you end up fixing that random crashing bug?

I did think about kikestarter for my game and was trying to brainstorm digital rewards, but decided i will just open a patreon instead, i think the kickscam era is already over, and wouldn't recommend anyone who has a serious project to try it.

Try and add one, it's out of my capabilities to make this sprite any better.

The thing is I don't really think a book where you discuss lore is really the best place to have long form game design logic that isn't necessarily related to the game. It's more just game theory in a vacuum, and then probably some tangential correlation back to the game, or why I made the decisions.

Which one? The last "crashing" bug I had was just the game locking up due to collision calculations which somehow happened because I believe the player's velocity was set to NaN. I think it only happened while using the gamepad, so it could have something to do with it, but it hasn't happened since then.

People say that, but I don't think that's true. I think Patreon is only really viable if you want to make porn games and do no work prolonging your dev cycle so you can fail to deliver on empty promises. I don't think non-erotic games can really make a sustainable amount of money. There's also plenty of kickstarter games that have nothing to show that get well over funded. I think the landscape has changed, but Kickstarter still seems like a totally valid platform. I could be wrong though.

Two words
Dwarf
Fortress

people will always be retarded
kickstarter.com/projects/141658446/digitsole-smartshoe-the-worlds-first-intelligent-s

I don't think that's a particularly good example. Dwarf Fortress has been around for a very long time and has a relatively large fan base. Of there'll be exceptions, but the probability of being able to pull it off seems very low

Nah, it was just almost twice the asked ammount, back in the day this would have made a million at least.

I think it depends how much you are willing to be transparent on the development and build a community, but it requires more work than just making a single kickstarter video for sure.

Look if this cuck can do it, so can you
patreon.com/YandereDev

raise the head 1px

trips of aspergers

How do you go from this

To this

I want to understand how can they fuck up with the payment methods, it even stopped working with my credit card, while paypal still works fine.

Design people are crazy. It seems mandatory for tech companies to change their whole look every 6 months or so.

I am questioning about the long term too. Where would one go from here, for a more or less permanent game devving, if they did make a game with moderate success. They make more games just like before and become a full indie Dev, get a job at a studio, make their own studio or become patreon whore?

Reddit.

You need a sustainable cult of personality

I've thought about it a bit and I think the best solution is going to be to just track it all during the attack and give the notification after the attack has finished. For an example of how it would actually work in practice, let's say your character is wielding a sword and a dagger and you order him to use an attack that will hit 6 times total - 3 hits with each weapon, alternating which one hits with each attack. Without the weapon breaking, the attack would execute like this

with a brief (

Just because it doesn't have a lot of features doesn't mean its not finished. And that's not a demo , its a complete game, using the engine.

Look at company logos from early 1900 to now. They've steadily became more and more minimalisitc to the point where you can only tell the what it is because of the colors. This is what designers and illustrators learn in their education. Welcome to (((modern art))).

Some progress for my map pack for my microtank skullcrap mod, if I recall it correctly each Doom 1 episodes had 8 maps so I am trying to aim for that. My plan is to make those maps first before I bother with details such as additional objects since they require a DoomED number in order to place them, the first 2 maps should be relative easy to play but on the 3rd map the difficulty will be increased.

For now I am going to aim indoor techbase style only, the other theme will be outdoor and urban related and thats it I won't include anything hell related since it is not part of my lore. It kinda sucks I couldn't get GZDoom Builder to work correctly in Wine earlier else I would have started this project much earlier on, I have 2 other maps which one of them got some details and the other one hasn't got any work on it but I won't bother with those 2 for now since the first map was planned to be outdoor related with 3D floors and SLADE doesn't support it at all.

So that also means there will be some development changes to my mod which will theortically not be a mod anymore but a standalone game which requires freedoom2.wad at least but that is free anyway and doing all these textures from scratch would take alot of work. I don't have any real concrete plan on map design expect for this part that I wrote a while: pastebin.com/n6sAsP9M

If you want to guarantee long term use patreon to keep yourself afloat for dev period and sell the game cheap when ready, maybe hand a free copy to patreons, just announce that this is your intention from the very beginning

so, I'm making a game and I'm planning to release it on android, what aspect ratio should I be working with in mind? 16:9?

it should adjust automatically.

360x640 is the baseline resolution independant (1x) size you want to work around. From there you want to support different screen densities (mdpi: 1x, hdpi: 1.5x, xhdpi: 2x, xxhdpi: 3x, xxxhdpi: 4x). So if you have 2D assets make sure you have appropriate sizes for each density.

Most Android phones are 9:16.

Ask the phone what it wants. Its bad to make assumptions about that kind of thing.

which might as well mean "stretched to hell and back"

if anything ill only support xhdpi onwards, I hate asset making

If I want to make a game for toasters, would I use a 4:3 resolution like 640x480 or 800x600? Or can I assume something like 1024x768, or even a widesceeen game like 1280x720?

Don't assume anything or people will complain about their special snowflake monitors not having the right resolution. Getting a list of screen resolutions that the monitor will play nice with from the operating system is reliable.

Oh god i remember when i worked making crappy android games and porting ancient java games to android, it was a nightmare.
Sometimes your game won't work on certain devices for the most retarded reasons and the android emulator never simulates these kinds of problems, so you need to have several different devices to test it the hard way;

No faggot, I didn't mean it as that.

what is the correct way of modeling, or attaching clothes to a base model? my topology is fucked, so while I could extrude from it, it would probably end up looking very bad.
i realize the "correct" way is probably to not have shitty topology in the first place, but i really don't want to have to redo this entire model.

i've tried googling this and people recommend shrinkwrapping, but that seems like a disaster later on when you have to deal with weightpainting shit to not clip.
the people recommending shrinkwrapping also posted saying to "not worry too much about clipping, because AAA games have clipping xD" which is pretty infuriating.

sage because shitty beginner question vaguely gamedev related.

Sup, /agdg/, I'm new to this, how do I make game?

Seriously though, is there any reading material accessible enough that would let me make simple 2D shit to start with, considering I have zero knowledge?

Java is cancer.
Thus Android manages to be absolute cancer while still being kind-of open-source by forcing everything to go through the Java Virtual Machine, which is quite the achievement.

Literally open Unity and the documentation and try to go in small steps adding shit.

There are three ways to deal with clipping:
1. Delete the verticies under the clothes completely, if the character won't ever change clothes
2. If he does, split the mesh in different areas and deactive the meshparts below the clothes in the engine, and reactivate when clothes are gone
3. OR Use transparency maps to make the parts of the body invisible that are under the clothes.

It's advisable to have the same topology for clothes, but it is not 100% needed.
Many times you don't even need to have extra meshes, tshirts and pants often can be textures on the mainbody without their own topology. For example WoW armor is 90% textures changes.

I would recommend nehe.gamedev.net to get started with simple 2d stuff.

I got started by deciding on the type of game I wanted, thinking about the necessary mechanics to make that game, and looking up youtube tutorials.

It isn't that hard. Learn by doing.

well, I should be able to get a hold of different devices so it'll be easy to know if shit doesn't work the problem will be finding out WHY it doesn't work

It was always some bullshit reason, like having one filename starting on upper case
Sometimes image transparency didn't work properly
Sometimes a few android funcions simply didn't work on device A but works on device B, and both had the same version
I really don't recommed it

Do Roguelikes have to use dicerolls for damage? Eg a longsword does 1d8+X damage.

Some RPGs like FF usually do something like (STR×ATT)/(DEF/8) -/+ 12.5%

Why would it have to use anything? Just do what you want to.

Also see this.
redblobgames.com/articles/probability/damage-rolls.html

Unless you're planning to have Multiplayer be a core feature you should probably use GZDoom instead. It also recently switched to a fully DPL license or some shit and ZScript is fairly powerful.

Rogue purists are very particular about how the game plays

Honest question: Who cares. Do what works for the game. Unless you're deliberately trying to capture the roguelike purist audience it's totally irrelevant and most players won't consciously notice it.

If the game is fun people will buy, no need to pander to a super specific niche community if that is going to alter your game, Do you game you would like to play, there will be other people who will also enjoy it.

new progress on textured polys. I have to manually enter in the fucking texture coords on this cube but whatever, after this im gonna try and add an alpha channel to the colors so it can be semi-transparent.

Are you going to treat collision and physics stuff too or just graphics?
Nice work by the way

yeah, i'm gonna do my own physics and collisions. right now graphics are the only concern because i need a graphics engine before anything else can happen

what kind of game do you want to make on this engine? sorry if you already answered it, i cannot remember if you did.

first person shooter or something. I haven't given it a ton of thought.

But how is that possible, god

Since you are doing the engine from the ground up you should take the opportunity to make something more out there that are usually not supported by the physics of default engines, more interesting mechanics like realistic ropes, hookshots, swinging, ensnaring, mounting, climbing monsters, etc…

I have no idea, we had to simply not use some functions sometimes,
Keep in mind, it was arround 5 years ago, might be fine nowadays.

I know the quality is far from done, but does this look like it follows a good sequence for an explosion animation?

Yeah, I do like that kind of freedom a lot. I'll try to take advantage of it, but I am still working on the "base" engine, so i'll try and do those kinds of things once i get to that point.

So how do the shaders in Vulkan work? Your earlier snippet looks like GLSL, do you convert that to SPIR-V or something?

Looks fine, but if you give us a gif or webm it will be easier to judge
Maybe would be better if you added a second animation for smoke, or 6~9 more frames with smoke forming and then fading

yeah, you just write them in GLSL and then run them through a compiler to make them into the SPIR-V format. Its a program that comes with the LunarG Vulkan SDK. "glslc.exe"

I can add smoke and stuff later…
It's a little tricky because explosions count as a solid tile so you can eg roll rocks over them while they're happening, and explosions can be overridden by other explosions like in pic related (game I'm cloning).

im trying to make a bunker but there's something that feels off

you shouldn't see the inside walls of the bunker, it makes it have two separate perspectives.

well, normally you'd think that way, but i actually made a "model" out of paper(I can't 3d model for shit) and you could see a bit of both inner walls because of, well, having two eyes. though I still dont know if i should remove them completely or just tone it down

I made this shitty tutorial a while ago.

When you're looking at a rendering you don't have 2 eyes, your screen is a 2D image, so when you're going for that perspective both the y and z axis share the same lines.

He's right though, your image uses two separate perspectives. It's fine to show the inside you just need to do it in a way consistent with the exterior.

When you're looking at the mouth, it looks top-down. When you're looking at the tunnel, it looks like it has orthogonal projection (no depth change as distance changes). It's mixed as fuck.

Either make the mouth look more top-down instead of frontal, or taper the back inwards a bit

Sounds like you want a component based system, like what unity uses.

well, this looks visually pleasing

this any better?

Definitely. One other thing is that the lip seems to extend past the tunnel part, which is fine, but the way you do it seems a little weird, I can't really explain it

It does look better although the discrepancy is still there

Did 8 frames of door opening animation; original game had 4 frames. I tried to keep the style, but mine still came out simpler looking, but it seems fine, I guess. This animation plays when you unlock a door, consuming a key, and lasts a second or less, so it shouldn't be too noticeable.

Going to redo doors and gates to match this style.

Getting better at animation. Still, it's fucking hard.

Now do the goofy reload animations like that one guy did

Ranon Sarono. He actually got hired by Respawn and worked on Titanfall 2's weapon animations.

That's Hyper.
youtube.com/user/HyperMetal101/undefined

Finished my keys and doors now, too.

...

Which effect looks better? It will play when you walk over the dirt, removing it.

I'm thinking the left will look better, when moving rapidly across several grid/tiled dirt tiles

The one on the left would probably maintain the tiling better.

Yeah that was my thought. I'll just do that because it makes sense, and also it's a lot easier to just shit out if I need to change the dirt texture.

Also I recorded a few other animations. The big and small explosions are exactly the same, just different resolutions, but I need to remake the big one (small looks fine but needs a little tweaking)

Also HOLY FUCK that door is amazing, I didn't think it would look that smooth (I made it by hand)

I don't know how far your codebase is, or if it's even a viable option anymore for you, but a messaging bus system that described in "Game programming patterns" comes to mind.
A Scripting system might be neat if you want to encourage modding, if you do some research there's some really good examples on how to go about implementing scripting.

cornflowerblue.jpg

Fuck off mate.

haha, I know the feeling of that one. Give that book a read if you want to do some engine planning, it's (((free))), if you could believe that.

If I could program I would rather work on Godot instead of the pile of shit that GZDoom is.

This is a new "comparison" picture of my rendering engine in different settings now that I have ported the texture functionality to all sides. The first two pictures have no visual differences, however one uses Immediate mode and the other uses the GL_ARB_vertex_buffer_object extension. Vulkan doesn't have font rendering yet, that's my new frontier.

Good luck.

S-surely you mean they tied physics to FPS, right? It's a bad-practice rookie mistake that they just didn't think through and somehow nobody at the company caught, right? Right?

Making a scheme showing some jobs/items interations for the friday progress report, what do you guys think?

The Physic FPS cap got added in Skyrim and is still persistent on Fallout 4, on Skyrim when the FPS is over 100 it is going to act bonkers which includes all form of old physic bug that happened in older games with physic such as ragdoll. pcgamingwiki.com/wiki/Glossary:Frame_rate_(FPS)#Frame_rate_capping Fallout 3 and New Vegas did not have this issue at all if I remember it right. So either it is something that got introduced with recent Gamebyro engine changes or Bethesda added that shit themselves for some weird reason I don't know.

WHY? How they can break something as basic as that?

Well you tell me user, they managed to screw up the conversation style making a lame Ass Effects clone of it, probably messed up the gameplay depth even more and a heaps more messed up things I forget. After all its Bethesda mayne they are spending more money on marketing then anything else also add that to the fact that most modern developers hardly have any testers anymore. Also ID Software and Bethesda "Teamed" up to make Doom 4 which doesn't even know what it wants to be. How they exactly screwed up in the technical sense I don't know it all, I am not even aware if one of those interns leaked some details regarding Fallout 4/Skyrim.

However I remember there was a screencap posted on Holla Forums sometimes ago that describes how Bethesda and/or Zenimax screwed up other developers which goes so far that the game got cancelled or doesn't live up according to the OG developer vision.

holy fucking shit I want to kill myself

...

Psst

Jeez fam.

What's y'alls opinion using unity store assets like the easy code program or easy make set pieces?