/agdg/ + /vm/ ~ Amateur Gamedev General

my.mixtape.moe/tzqnxa.mp3
r8 my boss song

I was waiting for like half an hour with my thread, come on!

...

I am unwrapping my 3D girl before going for some sculpting, I need hairstyle suggestions for complementing her appearance.

It's really bad

I kept waiting for the intro to stop, but it never did. Hate to say it, but that is absolutely not boss material. It sounds like something you'd hear in a haunted house level, but it doesn't even work in that context because of the drum beat.

Speaking of which, the beat itself is decent, but everything else is totally out of time with it. It just sounds like a clusterfuck.

Got the barebones of a context sensitive menu. Would love to finish it but have to go wage slave.

Implemented a roll. I had to get back into my collision code for this, but that didn't take as long as I expected.

I don't play on making the roll invincible, just giving it a low hurtbox and maybe projectile invincibility. But I'll have to deal with introducing that to the average player going in with a soulslike experience mindset.

Sounds more like an "Escape before shit blows up" song.

Fucked some shit up.
This is actually progress.

It's dangerous to go alone. Here, take this:
umlautllama.com/Audio/theory/theory.html

nice.

ow

wtf I love hardcoding now

...

Alternate edition

Fucking jews

I get pissed off to an unreasonable degree when this shit happens

:^)

...

while I'm bitching, snipping tool is so shit that it draws random marks occasionally just for fun

Wolf clone is progressing smoothly, 5 levels, 4 monsters, 4 guns and I think my level design actually improved around level 4. Almost done with EP 1 of the g-d's chosen FPS.
a.uguu.se/JRkrIvVU2LY0_gs3d.7z
Shit's gonna get spooky

Rustfag here, rising from the ashes. I've been pretty busy, working on a plugin bundle for Rust - trying to unfuck the core game. I've got a pretty large server network interested in running it on their servers.

As the project is coming to an end, I'm continuing work on my "totally not Rust you guys", now working on some basic weapon placeholders and their animations.

Here's a bow shot animation, in 110 FOV, because anything lower sucks balls. Rate out of 10 and suggest improvements, please. The model is just a placeholder.

Also, yes I'm aware that my elbow is crumpling. I'm letting it be for now.

Just my opinion, but the bow raise and the arrow draw should be nice, slow and controlled. The animation looks way too fast, and it goes from raise, draw, and reel without pausing. There also doesn't look like there's much feedback from the bow when you're drawing and firing. The arms and bow doesn't recoil at all.

My lifetime favourite:

Googling how to solve things is like being in the desert and everytime you see water it turns out to be a mirage.

Welcome back, Rustfag.
I never know how to rate things like that. I'm just going to point out some things I notice:
>your elbow is crumpling :^)

I've only ever shot a bow once in my life, and I don't remember much at all. Unfortunately, it's kinda difficult to see the actual forces in videos, as well, because it's something that is usually felt, not seen. I'd like to convey it visually, though - in what way do you think it'd be best to show the power of the draw?

There's no pausing in the animation itself, because the actual aiming phase can be held indefinitely, it'll just freeze on that one frame until the string is released.

I've actually eased it on purpose so it slows down, but I guess that wasn't enough. I wanted to convey force, and it felt like a lot in the moment, but I can see that it wasn't now.

I may or may not have resorted to stretching both ends of the bow backwards instead of using IK to make them bend when the string is pulled. I thought I could get away with it, damn it.

Blender is gonna be the end of me, I swear.

Vibrate back and forth or chaotically?

Also, thank you. Good to be back.

Shouldn't you work on unfucking your model before animating?

Have these obligatory image, I remember there was another version of it with actual forum post but I dunno where it is now, fugg :—-DDD.

It's always nice to mix it up. It's not like I can't apply existing animations to a different model.

It is actually somewhat subtle in reality too, but for 3D animation you often want to exaggerate things because they don't look right when they're perfectly realistic. It's part of the whole gamefeel thing like how good gun animations can make a gun feel more powerful with the same stats.
Back and forth, but it's really fast. Visually it's somewhat like a guitar string when strummed. Not even sure you can show it using animation in a game due to FPS limit, it might be more noticeable in the arm of the person using the bow. It's kinda hard to describe, I just happen to have used a bow a few times so I mostly know how it 'feels'. Also, have a reference video I just found

your game doesn't wine

But it does user. 2.4 on 64bit Slack 14.1. Set the prefix to XP and make sure you have d3dx9 from winetricks.

why is godot doing this to me?the texture is the same size as the cell, why is there an outline?why am I poor?, the only shit i find online is about disabling filtering but that changed nothing

How did you make the tileset?

new scene, added the sprites(they are file separated), and exported as a .tres file

Preface: I don't claim to be an expert on archery, this is just my opinion of how I feel a bow should handle

That first person archery video is okay, but there is fuck all tension in that bow and the arrows are fired very slowly as a result. I did a bit of looking on youtube and I found this video. You get a view from the side of the bow raise and draw. The arrow also makes a nice satisfying thunk when hitting the target.

Sucks.

Stop feeling pity for yourself, open any IDE or Engine and start working now
Do something, ANYTHING.

i should probably add that if I add a sprite to the scene directly instead of using the tilemap, it actually is the size it is supposed to be, I don't get it

...

Why is every program a special snowflake when it comes to window compilers

Time to refactor my engine again

I should also add the problem i still present when using 64x64 cells with 64x64 textures. I modified the texture to add two red diagonal pixels in the top left and bottom right corner and it shows how the pixels that are being "sawed off" are from the bottom right

maybe a precision error if you made it from a scene, go to the tileset properties and see that they have the correct sizes there, specially region

I don't think we are working the same, from your picture it seems you are working with one image that holds your sprites(so you have to set the region) whereas my sprites are all separated so I don't have to do that. I hate it when everything should be working but doesn't and you can't find the solution, it really grinds my gears

Your engine's texture sampler is probably rounding incorrectly. See if theres a setting for nearest neighbor interpolation as well.

Ive had an issue like this before and I was getting texture bleed

the textures aren't bleeding, the outline you are seeing is the background itself. i dont understand why the tilemap is removing a pixel from x and y

I recreated your basic setup and it works. This was with the 32x32 dirt sprite you posted, slightly altered so it would tile on its own. I'm not sure what I did differently, I didn't alter any options from the default. Here is the project in a zip file:
my.mixtape.moe/fmowbc.zip
Btw if you want to disable texture filtering, double click on you tile image and untick the "filter" option

well, I found the problem, it is the tilemap itself, yours works properly

The tilemap node or the images of the tilemap?

the tilemap file, so the problem would be the tilemap images and not the node

I have two questions:
What's a good IDE and game API that I can use in Android to code on the go?

What's the equivalent of Game Maker in an Android app?

I don't understand why my tilemaps are coming out fucked transfer.sh/7KYjM/tile.zip

I re-converted the arid-tilemap.tscn to a .tres and it worked without any change. Do you have the latest version of Godot?

2.1.3, should be the latest stable release

Try re-converting it? Here's the one I converted: my.mixtape.moe/ppmhql.tres
Then I guess you should report a bug.
github.com/godotengine/godot/issues

neet mother? not to be that guy but do you owe her money or does she need your support, because old or sick? just saying you could leverage shit against her to get her off your back. or you know carry her weight. gotta take care of yourself, no reason to put yourself out there for people who are not there for you. cant let people get in the way of your dreams.

guess I will have top open a github acount, and try to fid a way to explain this

How's the progress everyone?

I have been procrastinating for 7 months, and the last few month or so busy playing with other games.

t. skullcrap modder

Keep in mind that the game I'm working on a is a fastpaced PvP game - as a result, shit's gotta be quick. Thanks for the reference, though - I'm gonna use it to polish the animation.

I'm doing what I hope to be the last major restructuring of my engine, then I can finally just like make game. I said that before my previous restructuring

Hoping to get it done by the weekend.

I've been finishing up the ranged combat of my game for a while, but things turned out to be harder than I thought, so progress is very slow.

After I put in a restriction system that prevents you from using the same attack twice in a chain combo, but that doesn't seem strict enough so I'm putting it on hold until I tackle the whole issue thoroughly.
What I want to do next is either animating and scripting the last few attack moves (rapidfire jabs, flying knee, not-genocide cutter, charged versions of existing kick attacks), or looking at movement, which would roughly be reanimating walk/dash & implementing ledge climb/wall jump mechanics.
Somewhat more distant work includes attack responses (more flexible velocity scripting, wallslams) in general, and enemy attacks/player hitstun are also an important goal for making sure the idea I have of combat in general works out.

If I make a hentai game, can I put it in my portfolio when I apply for jobs at a video game studio?

I guess that would depend on the studio. If the studio in question boasts family friendliness and/or is excessively prudish (looking at you there, Americans), I guess it could be a bad idea.
It most certainly would backfire, should you ever find yourself on the receiving end of a SJW mob. They would certainly use that as a stick to beat you with.

yes, you can put whatever you want in your portfolio even stuff that might seem unrelated to the field
Whether or not you do is all about how you want the company to perceive you

I'm taking a break from game design and learning C.

I'm trying to learn to draw and java at the same time, it's tough, but it'll be worth it.

I can make sprites if someone needs. Mainly 32x32 or 16x16 but I can also do bigger.

that'd be rather handy, but I want to get further into the logic of the game first

don't worry, I'm not on your side of the wall

They're chill with NSFW artists as they're generally liberal. I remember that Blizzard accepted an artist when they knew he did porn despite him not putting it in his portfolio.

Most companies don't freak out when it comes to vanilla sex or nudes, but they do freak out when you do weird stuff like nazism, loli/shota, or rape.

Why use spanish filenames?

default filenames from kazam and shutter, if you notice my other files have random numbers because I changed them manually, I just forgot to change it for those files

is there a way to make tilemap nodes destructible in godot?I tried to use a damage function just like with my tanks, but when the bullet hits a wall it doesn't get the wall body node but the tilemap, I can't work with this

Yes. First, get the location of the collision:
var col_pos = get_collision_pos()
godot.readthedocs.io/en/stable/classes/class_kinematicbody.html?highlight=get_collider#class-kinematicbody-get-collision-pos
Then, get the tilemap position from the world position of the collision.
var map_pos = get_node("Tilemap").world_to_map(col_pos)
godot.readthedocs.io/en/stable/classes/class_tilemap.html?highlight=tilemap#class-tilemap-world-to-map
Then change the tile:
get_node("Tilemap").set_cellv(map_pos, -1)# -1 can be used to empty a cell, set it to whatever you need
godot.readthedocs.io/en/stable/classes/class_tilemap.html?highlight=tilemap#class-tilemap-set-cell

If you want the walls to take more than one hit, you'll need to store a 2D array with the walls' health.

the problem is that I don't want the walls to be destroyed outright, I'd prefer if they slowly chipped away until they are gone which doesn't seem possible with this

In any case, that's the way to find which tile was hit.

actually, is there any way to change the texture in a cell through code(so it changes during the execution of the game), is it possible?I might just leave the wall as its own scene

Working on a salvaged crossbow type of deal. I want to make it look extreme enough to justify it being able to punch through armor. It's not finished, not at all - I want to add a lever on the bottom to draw the thing, and make it look real intense.

Thoughts?

Shortened the spring to make it stretch more visually when the crossbow is loaded.

You can't change an individual cell's texture, but you can change the texture of a tile.
tilemap.get_tileset().tile_set_texture(tile_index, Texture)
godot.readthedocs.io/en/stable/classes/class_tileset.html#class-tileset-tile-set-texture

...

...

But user! Mod support is coming in 1.9 1.10 1.11 1.12 1.13 for sure!

So I heard that someone was creating a Moonman game?

I remember the model thread from Holla Forums and said that the guy making the game regularly posts here.

The moonman doom mod or the one for GTA IV/5?

I don't know I was just told that someone who regularly posts on /adgd/ was making a Moonman game.

spent the last 3 months coding the behaviors I want, learning the tools, etc.
suddenly find myself without the creativity to actually design levels

The Moonman post about our game was the first post I'd made in /agdg/. We don't post too regulary.

Pls gib updates

You are aware of the fulcrum and lever principle right? With how little the spring moves it has to be extremely strong, and it has to be preloaded as well. In fact it will probably be strong enough to bend the cross bar.

If you wanted to make it look vicious, you should've just used stronger shoulders. Also it needs a winch, there's no way you could draw that kind of bow by hand.

Honestly just make everything thicker and reinforced then give it a winch or something to draw it so it looks like human strength is not enough alone to operate it. Then make it really fucking big. Arbalests were the anti-armour crossbows of old so you can use them as inspiration pretty easy.

It's not pretty, I can assure you.

For sure, for sure. Oh look a pig just flew by, and it's raining malted milk balls!

Is there any gud free e-book or video tutorial in how to git gud at diffuse texturing in either 3D coat or in Blender? I really have absolutely no idea how I am supposed to texture stuff at all, in 3D Coat I usually just use some brush with random settings applied to it and coloring the models that way, but to be honest when I look at all the models I textured that way it just doesn't really look good at all, I decided to pick up modding again and tried to re-texture one of my models again but I am just getting lost and starring at it constantly.

When I look at the texture of this weapon I am envious of it since none of my texture look even remotely close to it.

You know liberals are moral absolutists and are exactly the same as the conservative christian moms of the early 2000s, right?

I figured out how to import dev textures into Unreal 1

Its texture alignment settings are less than ideal and I'm still figuring out it's editor. I also had to download a fan patch to even run Unreal 1's editor properly. UT1's editor still has issues even on my Windows XP machine.

However I have seen a lot of impressively looking maps made in Unreal 1's editor so I might play around with it some more.

I might try and see if importing textures into Unreal Tournament 2004 is similar and see how they look in it.

only if its racist or not diverse enough sexual degeneracy of every kind is welcome with the liberals as long as its not a racially homologous couple having a life long commitment to start a family. finding good porn of my fetish is pure suffering quite ironically 4/d/ sometimes carries what i'm looking for

but comparing liberals to christian moms that bitch about violent vidja i mean fuck they do still btich and act in the same manner but their morality cant be farther from the same. (other than wanting to destroy racially homogeneous societies we are all brothers and sisters in christ right?)

Didn't some guy get fired by feminists because it got found out he was into BDSM?

But the result is. Horseshoe theory, friendo. Christians wanted to censor violence and sexuality, and so do leftists now.

This is a good reading about the details of that:
breitbart.com/tech/2017/06/03/tech-leftists-kept-political-dossier-larry-garfield/

tldr, he was fired for saying some vaguely unpozzed things on twitter

Make a properly modeled and textured gun using a pbr workflow, bake everything into a static diffuse map and then scale the textures down to 256x256.

What does HIV have to do with it?

So I need to make a highly detailed model, unwrapping and texturing with all possible PBR stuff with it and then make a lower poly version of it? Hmm, is there a way to increase the workflow speed in Blender? That sounds quite time consuming when the model has like 5-20k polygons.

They're tolerant provided that you have the correct politics.

user pls, I dont want to repeat that argument a few threads ago

I see you make this post in every thread, every time this word comes up.
This is the fourth time in, what, two weeks? And that's all the ones that I have seen. You're like that one guy on Holla Forums that would always post his little OC of Trump with anime eyes.
I know you regard it as the pinnacle of humor, but it's not that funny.
Also,

I'm heading to the beach next week. No laptop or anything. Anyone know any decent UML editors for Android? I found one that works decently enough, but it's kind of annoying to use.

UML editors are good for conversation trees and stuff, and I know I'll get bored as shit being at the beach for a week. fear of water doesn't help

Absolutely everything.

All that's left is rebalancing every armor/weapon/spell and figuring out how to batch a cool filter in GIMP over every texture so the game doesn't look like potato (see: Morrowind Watercolored).

A bit more progress.

Now to work on the tool for creating tile maps. Tiles are just a special case of small 2tri meshes. That tool will be quite shitty, and since the tile definition is .CFG based I don't even need an editor. However, making it will test out collision code and help me decide how the UI system should work.

I'll make a font out of the tile map and I'll have a new font system (so I can ditch the old potentially encumbered system).

If anyone wants it, here's a font atlas recreated to match the default PC BIOS / VGA font.

I'd want to tip my toes with canonical perspective, the game remains top down, but I'd to design my texture with a canonical perspective so it doesn't look as boring, wondering if there's any resources out there for this as I'm terrible already with simple texturing

Thank you, user. You've been very helpful.

I would play a moonman centered game
that looks nice right there

What other names are there for a "scene"?

As in something where things happen, a room or a world. The main menu is a scene, and the gameplay is another scene, for example.

Arena
Location
Place
Setting
Site
Spot
Stage

For a single-dev project, how many features are too many for a management game?

Could, say, Akaburs princess trainer have a few more features implemented without problem or is that pretty much the limit for a manageable project?

Can you leave out this part?

pretty nice font, pic related is mine as a PNG

right now I use WEBP but I am thinking that I will switch to FLIF or HEIF soon, when I want to rewrite that part of the code to use a better format. I still need to figure out what the compression ratio is, if FLIF is better than HEIF, im still not sure yet. I know FLIF is better than WEBP.

Each component is optional, I'm only doing the watercolored filter because after using all the fancy gwafix mods I decided they look like shit together because they lack consistency. The 5 modules are: main merged uncontroversial mods, gameplay overhaul, mesh replacers, watercolor textures, and grass/flora (optional for performance reasons).

...

Did you fix the tribunal expansion? I started up morrowing for the first time the other day and got murdered by the dark brotherhood every time I tried to sleep.

Yes, there's a delay until you do something notable like reach a high rank in a faction. For people who want to start tribunal early there's also now a way to start the quest without getting attacked.

Implying it's not NSAniggers just having a piss

I'm taking a break from game design and learning C++.

>oh wait, I did that ten years ago now…

Making a lot of progress, but I really am not looking forward to adding an inventory system. It seems so routine, boring, and impossible to design well.

why don't they just tell me that, then?

i just spent a week modeling probably the most appealing ass i have ever created, only to realize the topology i set up didn't allow for me to extrude clothes from it for the design i had in mind.

now my choices are to risk being labelled a nier clone and use a skirt, shrinkwrap clothes to it and go through weightpainting hell, or redo the entire fucking model

sage because barely video games blogpost

...

git doesnt tell you the size of the repository you are cloning, and still doesn't have resumable downloads, git needs a lot of polishing

why not just create a collision surface and do dynamic clothing? after all it's 2011+6 user.

This. How the fuck is this even a thing?

...

I know it's difficult to contain your autism, but at least give it a go

Did you know that Git is an SJW infested hub, and they can pull the plug on your project if its problematic?

they can also pull the plug if you use too much bandwidth, or if your repository is too large, or if they just don't like you. Pretty much any service can.

waifu2x the textures?

that's going to increase the size a lot

checked

Pay attention to the formats.

So I took everything you said into consideration and changed a few things. In particular, I added a crank to draw the string (meant to be several pumping motions to translate repetitive turns into a full, powerful draw), and a stirrup for no real reason other than to make it more visually appealing.

I haven't added the internals yet, in case if I don't need to. Depends on what the animation will look like from first person.

it looks neat i wouldn't spend to much time on it.

something, something, something minimal viable product

Why convert it to PNG? Waifu2x caffe lets you save as jpg as well.

anyone need a 3D artist to jockey all their shit on?

should elaborate i cant texture much complex shit

shit like pics related I'm mostly capable of but anything beyond pic 4 I'll shit the bed

don't abuse me

pic 3* rather

Can't even muster up the energy to open Unreal these days.

Does a Titloli-fucking Simulator count as a game?

Sure, why not?

how did you make the skin?

anatomically the model is a mess but I figure weebs can jerk off to it just the same

This is taking fucking forever but it looks neat as fuck. I also had forgotten how much fun webdev was

The anatomy seems off. I don't really want to nick pick every little thing so I'll just say look at a good front facing picture of a person to model it after.

anatomy's fine, I just think her tits are defying gravity and it makes it look weird

Her feet are pretty meh tbh. And the face could use some work.

HERETIC

i would buff up the shoulders and upper legs a bit and cut the calf's down looks like she does calf work all fucking day

well take what i'm saying with a grain of salt

Holy shit thanks for posting this, I didn't know what this was and after trying it out on some textures it looks great!

nah, tits aint fine

the shoulders are also borked

Better?

Also face is still weird, I know.

getting there

I'd like to stress this edge here because it's the most important part in selling the authenticity of the breast

this iswhere the shoulder and breast and armpit connect

I would also go ahead and study the muscle groups and base your model's anatomy around them

reminder that realistic anatomy is a lot sexier than simplified anime-esque anatomy in 3d space

i remember long ago there were a guy who posted game like terraria with isometric view, atleast these gifs and screens gave me feels like that, didnt seen him in a while, is gaem ded?

my progress is that restarting the video system works completely fine now. I also ironed out some other bugs in the code that parses console input.

do not use jpg for pictures. If you change formats to something like png or webp you wont have this size issue.

Well yeah but where I can read or watch more of that stuff? Is there are at least any gud jewtube channels for that? Does Blender even have the ability to bake normal maps on diffuse map? I really want to improve my texturing skill since it sucks right now.

Fixed up the face and a bit and added a swimsuit.

Animation is fucking hard, guys.

You gotta think about it a bit more physically user, where's all that leverage coming from? Have you ever tried to pull something like your character is doing? His right arm is going to have to be stable enough to allow his left to apply enough force to winch the springs back… which is also a great way to fuck up all the muscles in his arm.

It's coming from his immensely brutal, manly muskles. He eats nails for breakfast and brushes his teeth with barbed wire.

In all seriousness, the actual implementation would have to be the classic crossbow lever draw, but that'd include looking down and overall be too cluttered to actually do. This crank was an attempt to solve the problem of his left arm thing being "too strong" while still only using two arms.

I'm open to any ideas, man. I'm at a loss here.

...

In the meantime I went ahead and rendered it because why not. Please do give ideas as to make it less unbelievable.

Have him bring his other hand forward to a more central position, gripping it underneath the axis of the lever. It will look a bit janky but will solve some of the floatiness it currently has. Perhaps you may want to add a fixed handle there for reloading purposes. Just try to remember that nothing you animate has weight, and that the guy you're animating is basically a mime that you have to help create the illusion of weight, leverage and energy expenditure.

A side note on the winch animation, try out having it draw slower and slower the closer it gets to the end of the pull - the more you draw on the strings the tougher it will be to pull them further, so once it reaches the apex of the pull have it snap back more quickly. It might be easier to have him let go of it for that too… If you are animating in software that has animation curves, they will definitely help with this type of interpolating.

Also the most jarring thing for me right now is that the crossbow is 'pulled' in the opposite direction that most of the force is going. He's pulling on it and it rises up? It should waver slightly and droop if anything, and once it snaps back it should snap up as the sudden loss of forces from pulling it will be unbalanced by his other hand countering them, if only momentarily.

Thanks, I'm gonna see what I can do. About the whole pulling the crossbow in the opposite direction - I wanted the action to be performed with both arms - and that's the way I've gone about it.

I see. Perhaps this diagram will better explain what I mean… The crossbow will go in the direction of the arm producing the most force. In red I have indicated the forces acting on the crossbow from the winch hinge, with dark red indicating the leverage distance that arm has. In the position you have animated, his left arm is able to be fully engaged, from his deltoid/pectoral region right down to his forearms. Cyan indicates the forces his wrist(!) alone is applying from his right arm. Trying to resist that pull from the opposite direction are; the top muscle on his forearm, his tris and his lats… His leverage is also a lot smaller. I'm saying that even if he's the strongest man on the world, basic physics should still apply.
One of the lessons I remember from when I studied cg was that you can have all the in universe justification for your material that you want, but expected reality more often trumps reality. If people think fire hydrants are red you'll have more luck just making them red - even if where you live they are just holes in the ground with a water main poking up. If I look at your animation, the leverage and forces don't make any sense at all, his right arm doesn't have any way to apply that much force to the crossbow especially not from that position.

I tried to find a not-shit example to show you what I mean but this second image is really all I could find. The bit to pay attention to is when he pulls the cocking lever the gun is pulled towards him and then pops back a bit. Those are the prevailing overall forces on the gun. If your guy was managing to pull the crossbow upwards the winch wouldn't move at all because there wouldn't be enough backwards force applied to the winch. I'm being a bit autistic about this so I'm going to stop now but hopefully you see what I'm trying to say.

Thanks, man. I'm gonna screencap this and revisit the animation once we start implementing everything.

Cool, good luck with the project; I envy your motivation!

Thank you, user. If you want advice - think about whether or not you really want to do gamedev for a living, and if you feel like it's something you want to do for sure - erase even a prospect of another job from your mind. Keep pushing and eventually an opportunity will pop up.

There's nothing for me but gamedev - I've decided that much. Now the only thing left is getting there.

I already have a job, hence the motivation issue, but thanks

...

Why are the Unity forums so full of retards?
There's many cases like this but one that infuriated me the most was a guy asking how to fix his inputs always being delayed by 1 frame, and the rest of the thread was people arguing that its humanly impossibly to feel 1 frame of difference.

In UE4 C++, what's my best bet for making a list of checks, that if true, call a list of effects? The way I have it set up (in my head) now is:
TMap OnChangeMap;for(auto pair : SomeMap){ Spawn(pair.key); //and somehow send over the pair.value, probably when the pair.key is constructed}

The main idea is, I have a set of attributes (just UStructs containing floats and aforementioned TMap), whenever these attributes change, it would run a check (the UCheck), and if true, do something (the UEffect). E.g. if health goes down, it will look at the OnChangeMap for health, which tells it to spawn a UIsAttributeZero check (to which I'd pass a gameplay tag to tell it what attribute it needs to check), which if true, will in turn run a UKillCreature effect, and this should be applicable to any attribute (as long as I pass the correct gameplay tag). The reason I don't simply run a set function when a set attribute changes is because this way I can, per entity in BP defaults, decide exactly what happens when a specific attribute changes instead of having to program functions on a case-by-case basis (e.g. the goblin simply dies when health reaches 0, while the fire elemental dies just like the goblin but it also runs an effect that causes an explosion, an effect which I can feed what explosion to use, the damage of the explosion, etc).

The issue is I'd be spawning a lot of UObjects. Lets say I am running this whenever health changes (to catch if the player died, as above, but might also be more complex stuff), that means that if the player is poisoned (or regenerating health), health changes per tick, which means at the very least, the UIsAttributeZero will be spawned. That's already 1 UObject per tick, if there are more, more are added. This is assuming they fail, because if UIsAttributeZero finds that health is indeed 0, it will spawn UKillCreature (and this is the best case scenario where only one UEffect object is spawned). This is just for 1 attribute, there may be more doing the same at the same time. All in all in hectic situations it would be spawning and killing 3-10 UObjects per frame, which I feel will be quite the drain.

Question is, is there any other way to do it? I could define checks and effects as functions in a library, but I wouldn't be able to add them to a TMap in the defaults like that (I think). UStructs are much simpler than UObjects, but they are pure data assets and I'm not sure how I'd replicate this functionality with them. Should I just rethink my whole system?

Is it possible to batch static lightmapped objects? I made a cluster of big rocks and since they're the same material I'd like to have them rendered in one call.

here's what the frame debugger says

docs.unity3d.com/Manual/GPUInstancing.html

It was already on, but i discovered right after posting, that if you put static objects under a non static parent, it breaks instancing, the only 2 options are to have the parent be static before you attach the children to it, or change it to static and apply to children (which i want to avoid since it seems to break static prefabs that use light probes instead of light maps)

The body is passable, but I recommend studying up muscle structure in general. the eyes look a bit creepy.

You want breasts to not follow gravity in a t pose, so they behave properly when you do physics simulation.

Should be possible, yes. The lightmaps needs to be packed in an atlas (unreal can probably do this for you). Question is, do you want to? It's impossible to cull objects in a batch, so either all or no rocks will be drawn. That might result in quite a bit of fragment shader overhead. If you use forward lighting, lights can't be culled from individual rocks, so the lighting loop will be very long (with deferred this is not an issue). Culling is very valuable, but requires a granular scene. This is antithetical to batching. Also, batching is more about decreasing CPU overhead than increasing GPU efficiency Go ahead and try, but don't be surprised if you don't get any fps gains. Also, as your game grows it's possible any bottleneck you have now may change. So even if batching is a good idea now that you are CPU-limited, it might turn suboptimal when your scene gets more busy and you get GPU-bottlenecked. This is why they tell you to optimize last. There is actually no such thing as optimization. It's a myth. There are only tradeoffs, one of which will result in optimal hardware use.

Ok, I was wrong actually, didn't notice pic related in the editor.

You're right, but I'd still like to at least know how to set it up so it works if I end up needing it. A whole draw call for just 36 tris sounds like a waste, though.

Are you Bad Topology user or someone different?

Because it's if you are its like a switch was turned on once you applied textures and that model is now fuckable. Still needs some work, but good job.

What do you guys do for a living? How do you find time for devving?

I want know, where to start to learm 3d modeling since each time that I open blender or max my mind become blank and stuff , mostly for mods, mind you I'm a poorfag

I'm searching for a job and might have to homeless-dev if i don't find one in a month.

Everyone is different, but what helped me break through with 3d was watching tutorials on my phone when i couldn't sleep, it might help since you cant distract yourself with other stuff. Learning core 3d concepts first will make learning the confusing interface of blender easier.

i meant rpg playstyle like terraria and it was isometric

I have plenty of hours every day even after work.

If you live in a big city, move to a smaller area into a small apartment where living doesn't cost as much and you won't need to slave away for someone else 24/7.

The hard part is finding energy and motivation, not time.

glad it was nothing

...

A fullscreen quad is just 4 vertices, but on a 4K screen it's 8 294 400 fragments. Just because an object is geometrically simple doesn't mean it's cheap to draw, especially if it isn't culled. If it's hidden behind other geometry and it gets submitted before the occluding geometry, the redundant fragments will cause overdraw. If after, it might get culled by early z test, but that's not reliable. If not, the fragment executes, but no fillrate is used. The cost of an object is a function of both the geometry and it's onscreen size. But yes, it is still wise to learn how to do the batching so you have another tool at your disposal. I just want to make sure you have the information necessary to make efficient use of said tools.

At one point, I got the job to set up networking in Unreal for a project. This is an easy job initially, just using the built in system. Yet it wasn't working for some reason. My work laptop, which had Windows 7 wit virtually no updates at all, was starting to break down hardware wise so I bought a new one. Bug's suddenly gone. Turns out that Unreal's networking relies on some updated driver during the build stage and building on any other machine to test would have fixed it.
Our catchphrase whenever a bug came up after that was "buy a new laptop". I miss that project. I don't miss its terrible codebase and artists who didn't know what optimization is though, god damn.

that thought didn't even go through my mind for a second
my bad

The eyes are definitely a placeholder. I'll need to study up on how anime games do eyes, since from what I can tell most of them have the eyes as a convex plane with the iris and pupil as a separate plane that travels over them, rather than just as textured spheres like in most games.

Not me, I only made this very recently. Most of the times I've posted here before it's been to post military or 40k stuff I've made.

Graphics issues are the fucking worst. The program is giving me no indication about what's wrong and it's not like I can see how it's broken or draw debug graphics or anything to narrow it down.

alright, bear with my stupidity here
you've got a deck of 32 cards and draw 5
what are the odds of drawing 3 aces?
the way i understand it is (4/32)*(3/31)*)(2/30)
but what about the other 2 cards you drew, how do they affect the odds?

They don't, unless they matter.
Does it matter that you only get three aces and not four? If it doesn't, then the other cards are completely irrelevant and don't factor into the equation at all.

but don't the odds of drawing 3 aces increase considering that i've also drawn 2 other cards?

Drawing an ace would reduce the chance of drawing another. Drawing a non-ace would increase the chance of drawing an ace.

alright, what are the odds of drawing 3 aces and 2 kings?
(4/32)*(3/31)*)(2/30)*(4/29)*(3/28) ?
is this correct?

wake up dev go to work sleep wake up dev bonus points for cooking meals for all week on the weekends replacing work on Saturday with home maintenance & cooking replacing work Sunday is sacred and one should never work on a Sunday.

True enough. You don't have to draw all of them in order. And that influences the chances, yes…
No easy way of doing this, actually, I don't think.
You have to think of every single mathematical possibility and add them all up.

so, here are the possibilities. A for aces, x for non aces.
AAAxx
AAxxA
AxxAA
AxAAx
xAxAA
xAAxA
xAAAx
xxAAA

That seems to be all of the possibilities, but I may be wrong. So, we count all of them and add them up.

AAAxx - 4/32 * 3/31 * 2/30
AAxxA - 4/32 * 3/31 * 28/30 * 27/29 * 2/28
AxxAA - 4/32 * 28/31 * 27/30 * 3/29 * 2/28
AxAAx - 4/32 * 28/31 * 3/30 * 2/29
xAxAA - 28/32 * 4/31 * 27/30 * 3/29 * 2/28
xAAxA - 28/32 * 4/31 * 3/30 * 27/29 * 2/28
xAAAx - 28/32 * 4/31 * 3/30 * 2/29 * 27/28
xxAAA - 28/32 * 27/31 * 4/30 * 3/29 * 2/28

Add them all up, and I'm pretty sure you'll have the correct answer.
Maybe.
I think there may have been an easier way to do it, but I don't remember it, frankly.

I deserve a diploma, thank you very much.

isn't that only when the order matters?

the order influences the chances.
if you pull two aces first, then your chances of pulling a third one are different than if you pulled two non-aces first.

Graphics debuggers exist you shit. AMD CodeXL is what I use to debug my OpenGL code. Although you cannot introspect everything and its pretty buggy, it does allow you to look into textures/vertex/element buffers, framebuffers and shader uniforms.

It aint perfect but hey, its something.

I hope I don't mess this up.
you have 201376 possible 5 draws combinations. if you have 3 aces, then you have 29 remaining cards for 2 vacant slots and there are 406 possible combinations for that setup, so you have 406 "winning" combinations out of 201376, which is a chance of 0,201612903%
I was going to post this earlier but then I realized you weren't working with a 52 cards deck…

this is correct

VOGL exists too.

yes is called full render, and if you are using cycles you can choose what to bake

I listened!

Also, new demo soon.

Assuming you are taking out cards without putting them back in (i.e. keeping them in your hand) you have a space of probability of 32 over 5 = 32! / (5!(32-5)!) = 201376 combinations. Now, out of those, the probabilities that you get 3 aces and 2 kings are (5 over 3) * (5 over 2) = 100, which gives 100/201376 = 0.0005 = 0.05%.

That's the theory anyway, had my probability course a year ago, back then I'd know it by heart but once I pass an exam all my memories of the course kinda get purged, so I give it about a 50% probability of being correct.

I worked on gamedev for 10 hours straight today.
I can somewhat reliably spend a lot of time on visual stuff, but i have zero motivation for learning programming, what should I do?

This is for human heads but the baking process should be the same. Essentially you want a high-detail model with all the bells and whistles that you bake the details into the low-poly normalmap and AO map. For your tank you should add the details to the barrel of the gun, the treads, etc.
youtu.be/Ryi_pYqcavs?t=3m34s

You really only need a top-down shadow on Speebot. Just make it a checkbox to turn it on or off and leave the "realistic" shadows on always

This isn't really a solution, but you could use UE4's blueprints.

I'm using unity and c# and i want to get good. I may have worded my concern wrong, it's not that I don't wanna learn to code, it's that I can't find the motivation to do it and my knowledge is quite lacking.

Dude, that's lovely.

are you the roll a ball on steroids guy?

Will you translate it?

nope, most of them are incomplete tho.

Try programming something that lets you see the results instantly, i.e. make a character that you can move around. When you make something that you can play around with - you'll get your motivation quickly.

That would cause a lot of visual glitches that I'd like to avoid, so I think I'll keep the shadows consistent.

Thanks!

I might, what language would you like to see supported?

Spanish.
It would be nice if there was french and german too. Also japanese and chinese, but those are harder to create.

I can help to translate to spanish.

hmm alright do I even need to unwrap and texture the Hi-Poly model when its only purpose is for adding additional detail? Also since I am working on a game which doesn't has great texture support the AO stuff won't work out for me well since on the T-64 tank model I have there is some mesh parts which are duplicated in order to save some texture space.

You know, I was lurking the thread when I saw this and I remember working on a VERY similar problem. I was curious and wanted to make a program for drawing each possible hand in Secret Hitler, which is a /tg/ game.

Good Idea here but you can simplify it, assuming we have exactly 3 aces drawn.
First you need to multiply x-draw probabilities after you get the aces. You'll find the probabilities for each order of the cards to be the same. If you multiply the numerator and denominator of each hand separately you'll find you're multiplying the same numbers in different orders, which doesn't matter. was onto something. Yes the order changes the odds for the next draw but it's really moving a numerator number to a different location.
So after solving it once just multiply by the number of different orders it can come in. You can write it like this:
If you wanted the probability of AT LEAST 3 aces you need to either:
Correct me if I'm wrong.

this still seems fishy to me, i feel like the order in which i draw shouldn't matter

m80, I'm pretty sure this is the right answer

Kind of my point. The 8 in front is for the 8 possible orders the cards can come in. You can multiply the numbers within the parentheses in whatever order you like; it doesn't change the solution. There's a name for that but don't ask me what it is.

Guess who made progress

I started a hops yard and didn't have time to dev, so my friend did most of the unseen work for the last month

How is the rotation of your character handled? Just movement input or mouse direction?

I'd make the turn radius for quick readjustments faster and leave at the current speed for normal movement.
Looks kind of weird as it is.

How is the rotation of your character handled? Just movement input or mouse direction?

Diablo-like controls, the cursor just wasn't captured.
The movement itself is currently handeled solely by the navmesh agent so once the character reaches their intended destination, any rotation stops. This isn't the way it was intended or how it worked in the demo, but because everything is being re-written, this isn't implemented yet.
Rotation speed itself will be tweaked.

the potions of poison now creates a toxic cloud and the bottled flames now explode when thrown, applying a debuff to anyone caught within. The other potions simply apply their effect to whoever happens to be on the tile they land on.

when I was figuring out transforms, what I did to debug was apply my transforms on the CPU, then look at the output positions. Any triangle that is to be visible needs to be within ([-1,1],[-1,1],[0,-1]) after perspective division, so check if this is the case. If you have trouble visualizing the triangles from coordinates, dump it to an obj file and view the data in blender. If it fits within said cube, something else is wrong. Implement parts of your shader code in software and look at the values you get.

Protip: you can easily verify any solution with a computer. Learn python, then write a quick script that implements the problem (i.e. make list of 32 elements, set 4 positions to 1, perform random sort, pop 5 values, count number of ones, return true if condition passed, loop a billion times).

so how is your organ gore system handled? i wanted to make survival game or a damage system similar to rimworld where individual organs/ limbs took damage with overall health being nonexistent. also with so much of it being dynamic how does that hit performance?

that is pretty much a writer's lock, it means that you don't know what you are doing and should learn more about what you were trying to do has the right idea. going to kikepedia and clicking on the categories can get you started, going for a walk in google maps and porn can get you inspired.

I'd play it.

Finally learned how to normal map

teach me master

Created a jiggle animation. Pretty good considering I've never worked with jiggle deformers before.

now you can harvest circles from watermelons

Is the rendering within the 3D software? It will look different in the game engine, right?

I want to use toon shading in UE4, the problem is once I apply the black line edge detection, particle effects become nigh invisible.

The rendering is done in Maya, yes, though it's just using the default Viewport 2.0 to render. The texture/shader is a layered shader with one material being a toon shader and the other being a texture to darken/shade the nipples/nose/pussy/etc. The default render is faster, and for whatever reason Mental Ray is incompatible with rendering the toon outlines for some reason.

I guess the reason I asked upthread if this counts as a game is because I would be rendering out these animations in Maya, and then compiling them together in something like Flash, like what Seismic does.

Could have saved yourself a lot of work, and had a better final result, if you just made only the player's shadow topdown.

I am working on the command line for Sigma II. The last major question to answer is tab-autocompletion like in quake and everything else.

I am going to do a linear search, right now this is my code to catalog everything. I don't think that I will have over a thousand vaid keywords, maybe like 200? 400? so a tree isn't needed, although i think maybe that would be better.

Right now i just log all the keywords. This code is just putting them in memory, I dont use them (yet)

Did anyone actually enjoy these games?

i'm working on a command line for Unity with the main feature being network support and i want to fucking die

Don't die yet I still need to play your game once it's finished.

Since I was working on the inventory, I did not have time to further develop the gore system. It's in "experimental phase" currently so it only works by clicking on the character with the mouse cursor.

1. Ray is cast, it ignores all layers but the "enemy" layer. The ray is your typical ScreenPointToRay deal.

2. The hitinfo contains RaycastHit.textureCoord, this tells us which pixel on the texture we are pointing towards.

3. The texture pointed at is stored as temporary Texture2D and its pixels as an array of color values(Texture2D.GetPixels()). This is quite CPU-intensive unfortunately so the texture has to be low-res (512x512 seems to be optimal). This isn't a problem for me as most details are stored within high resolution normal maps.

4. The alpha value for every pixel in the texture is 1. I modify it to be 0 in a circular area around the previously acquired coordinates. I also add to the R values around that to create a blood/wound effect.

5. The modified texture is saved and applied to the material. Because it uses the normal/cutout shader, the 0 alpha area looks like a gaping hole.

6. The model looks a bit like a matryoshka, there is the skin layer, the muscle layer and the organ models hidden inside.

7. The organs themselves are fixed to the model with joints, which can be destroyed, but this part is not ready yet. It will be damage-based most likely.

8. The blood drops are something that was already present in the downloadable prototype

Hope that makes sense. As you see, this is a bit convoluted and would be hella easier if I knew anything about shaders. The effect is satisfying/efficient enough that I'll most likely stick to it though.

I'm shit at shaders but I'm 99% sure there's a better, easier, faster way

Can I give some constructive criticism?

No. /agdg/ is a safe space, criticism is forbidden

Just make the textures themselves a render target or drawable so you can leverage the GPU to do the work for you, or else avoid copying the data every time

You could also maintain a pair of textures per layer, one 512 one with the gore etc and an any sized drawable one that just contains alpha information for when the other one is drawn

As am I. There's always a better method and I'm always open to suggestions.

like this one:

Rendertextures then? All available documentation and forum discussions always has some sort of getpixels() setpixels() Texture2D.Apply deal going on but this is an interesting concept.
And I just found this:
gatemedia.ch/software/2016/01/29/high-performance-real-time-painting-in-unity.html

The only possible shading techniques i know for that are blending with textures or with vertexcolors. Search for terrain shading tutorials with textureblending and you'll find what you need.

RenderTextures is what you're gonna have to be looking at.

Basically what you do is instead of rendering to the backbuffer (that is, your window), you render to a texture instead.
I don't have experience as to how you set this up in Unity, but this basically means you'll have a setup in the lines of:

You'll be swapping between textures A and B every time you do this, but its still significantly faster than copying data between the CPU and GPU.

I don't know to what extent Unity supports multiple texture targets, but you might even manage editing multiple layers in a single draw call. This would be even faster, especially if you implement a lot of layers.

That looks comfy af.

Nice

Start with small stuff you can get done in minutes
progress => motivation

do you mean passing commands over the network?

I still have to make the server-side of my program, right now i'm building the client. Once I start programming things like character movement i'll have to set that up.

Richard Williams was right, animating a sidescrolling walk and doing it well has to be the toughest thing in 2D animation. I've worked on this animation for several hours already today, so advice on where to improve would be greatly appreciated since I think I'm getting stuck on my own.

Add one frame between the legs crossing and the legs fully extended.

didn't mean for :^) to be on

...

this
12:16-12:18

And I just discovered I forgot to save the animation file itself, so all I have is this .gif and an outdated spritesheet. If that isn't a motivation boost I don't know what is.

The whole animation kind of needs inbetween frames, but I don't really want to add them until I'm confident in the main ones.

I tried something in that direction the first time, but it turned out absolutely laughable, so I'm posting it for you to laugh at. I'll have another go at proper footwork for the third try though, since I already lost my data.

Thanks for the reference.

This started off as a doublepost, so I saged, but it's become a bit more than that.

yea, the back leg is limping too much in that

Stance's too wide in my idle in the first place so if I want to use a proper slip&slide I'd have to start with the back foot, which is kind of nonsense as far as I understand that. Problem is, it's the idle so I probably have like 20 frames that build on the too-wide stance too and I'm not quite motivated enough for that shit today.

A side scrolling beatem up with a nun? I'm game.

The movement feels very unrealistic and there is no balance.
Front leg does the huge stride compared to back leg's tiny tip toe movement.
The hair jumps up but I don't see this "jump" in movement of limbs. The head also is dragging along as if the neck is limp.
I would add one more frame when character is in the air, when neither foot touches the ground (if I see this correctly)..

I know that thing is a piece of shit, is what I use currently.

Pretty fucking slow, as we can't get much in the way of gameplay or even concept art. What art we do have is good, and the artist is soon to have more free time, so things might take a turn for the better.
That being said, we've decided on the engine, and just need to get to work.

testing fix

Are you still using Vulkan for your engine ?

yeah. I was away from my desktop computer for the past 3 weeks, which is why its been saying "Intel 965" as my GPU when I post screenshots of the startup text. This means of course that I haven't been able to work on vulkan, so everything you've seen lately is using the openGL renderer. Since I haven't been able to run any vulkan code, i've just been working on the command line. Once I have text rendering in vulkan, all the work I have done for the past few weeks will be available in the vulkan renderer, since I havent been writing renderer-specific stuff (although the text rendering only works in openGL now)

I'll give you guys a download of what i've got soon. Although, what you see on the screen is the tip of the iceberg in terms of the retarded amount of code its taking

let me guess, next youre going to mod coin battle back in and for his final smash he rubs the kike wall

guys, I know nothing of sprite work, I was working on this yesterday and want to know what could be improved/added, and I'd like to have some pointers regarding doing animation

How are you doing kids

the tires could be rounder and the separations equal. You can also add more detail.

finding a good separation difficult for me, however, I'll probably add more holes to the wheels so they don't look as plain and help make look like the animation is…animated.

I might use the power of math to find a better tread width later, though it's not priority

So somebody made a tutorial on Jewtube how to hide parts in Substance Painter 2, welp I guess I am going to use this shite then, I was kind of a "crisis" before as.. I mean frustrated that 3D Coat doesn't seem to do diffuse painting very well and thus leaving me stuck on improving my textures. youtube.com/watch?v=zq2waSmYFbs

what do you want

nothing actually

just passing by

Hey guys.
Does anyone know a good comprehensive guide to rigging in blender? All the stuff on YouTube just shows simple joints.

I can't figure out how to correctly set up constraints and pole targets. It's really annoying.

Is there an add-on or other programs I need or is there just somthing I'm missing

If you're looking for quick auto generated rigs, you can use Rigify. Just be sure to enable python scripts in your options or it won't work.

If you're looking to create your own rig, then here would probably be more useful:
youtube.com/watch?v=cGvalWG8HBU

PUBLIC SERVICE ANNOUNCMENT

I had to recently sanitize my online identity because of dumb drama (completely unrelated to Holla Forums or /agdg/).

HOWEVER, I went through my Google Drive, and it turns out it keeps a full history of every file you have accessed with it. I was seeing Holla Forums projects from like 2014, with it and accidentally found the first and last name of the person I was arguing with because of it

Just a friendly reminder to keep track of what you do online and where, just in case someone doesn't get along with you and can use it against you.

People reap what they sow.

...

Speebot user did it, too

What keeps you going?

The Krebs cycle.

if speebot user kills himself will you follow suit as well?

there are no excuses. just because speebot user did it doesn't mean it wasn't a bad idea.

Game development.

To elaborate, gamedev is the closest thing I have to a hobby, and the only thing that distracts me from the fall of western civilization, the game industry going to shit and society in general being disgusting fucking degenerates. It's the only thing that takes my mind off of wanting to off myself.

Another part of me really wants to prove everyone who ever told me that I can't make games - which is everyone, really - to prove them wrong. Because fuck all of those bastards.

why do people tell you you can't make game?

rate my tiles

fuck

...

I bet he is another yanderedev

the first one works better as a bump map than at whatever you were trying to accomplish, second one is just noise

Hey, I dox myself on an almost daily basis yet nobody cares to notice.

Nobody cares, Anthony.

Yeah, nobody cares!

If you use anything related to google then you deserve to have all your personal information and activity displayed publicly.

that reminds me, earlier today i had to confirm that i was human before doing a google search

That's fucking great, even with a hidden opacity it won't let me paint the other faces. What a bunch of fucking MIT niggers that claim they are pro but can't add a geometry hiding function.

For some, it's because they simply don't like me as a person for [insert view that I hold here]. Others, like family and stuff, who have known about me wanting to do it for years now, tease me about how I haven't put anything out despite working for so long, and imagine that gamedev has to be AAA and one man teams are impossible.

The day you faggots stop giving me shit is the day you're useless and dead to me.

If you want to paint on seperate parts without worrying about other parts you can do one of these:
1. Select UV faces in the 2d view to paint only these
2. Use a Color ID mask
3. Seperate the meshparts before export

Substance painter has majestic materials but they're total fucking niggers when it comes to standard features. Also jews for the shit they pulled with SP2.

Well I already used the Color ID mask in order to use the opacity/Color ID masking but when I give each mesh its own materials does it even export all the textures layers stuff into one diffuse texture?
It has that UV locking feature does it even work in the 3D view?

Indeed painting on that program is pretty neat actually and provides a nice set of materials, brushes and what else and on top of that they have a site to download more of that.

Yeah look at this shit it got almost 200 upboats, the request is from 2014 and about 3 years later they still don't have this most wanted and basic feature, how the hell can they screw up that one?
allegorithmic.uservoice.com/forums/261284-substance-painter/suggestions/6625997-hide-selected-geometry
What did they do this time? I pirated the Substance Painter 2 program I have anyway not enough moni for it.

Well don't stop user. You have enough common sense to at least be aware of the truth. Also, make the bastards eat their words yea?

It does, every masking works both on 2d/3d, but hiding faces with texture opacity in a roundabout way to paint faces that are behind other faces is a massive pain in the ass. I am always breaking apart meshes into several parts before exporting to painter. This way you also don't need to make cages when baking shit and nothing is overlapping

When i found this program first, i was amazed, even with its fault it was easy as hell to make shit look real with PBR materials, so i bought it because i needed it for commercial reasons. They've promised all kinds of features that will come in the future and many updates. Not much later SP2 came out. Assholes deliberately didn't really announce it so people would buy SP1 instead of waiting for SP2. Worst of all, SP2 is pretty much containing all the shit they "promised" for SP1 and was nothing more than a slightly bigger udpate, but want people to pay for it. I don't know if they've deleted all the comments but everyone was mad when they announced people should pay full price again for something they promised earlier. Anyways, i'll never buy shit from them again and SP1 is enough for a while.

Hmm do I need to give each mesh its own material too? I separated it now and they all have the same materials and I imported it in Substance Painter 2 now as .dae and it loads as 1 mesh instead of several meshes.

No, you treat the materials, UVs and everything as you normally would. I only seperate the mesh into parts to make baking and painting easier.
For example in Blender you make a trouser mesh and a robe mesh, but the robe is partly hanging over the trouser which makes painting a pain in the ass, so i move the robe aside for a meter on the x axis and have them in seperate positions to make it easier. It can even be a single object and single material, you only do mesh repositioning for easier painting purposes.
In case you could seperate the inner barrel from the outer barrel to paint it easier.
This is generally helpful when painting mechanical parts too.

Which flag for the human empire?

neither.

i tried the lotus emblem without any detail, it was too boring

Did you try making some kind of design with just the lines, or the shapes between the lines?

Tried again, its less bad than last time, what do you think?

As a flag it's better, but now it just looks like a flower, and the yellow colors are still too similar. The contrast between white and red is so strong that the different yellow shades are hard to tell apart from each other.

Also, it's usually bad to make any color touch both a brighter color and a darker color like the yellow does in your design, it tends not to look very good.

I don't know what you're going for, but pic related is what the designed in me immediately wants to try.

Definitely an improvement, but isn't a flower of any kind as a symbol of humanity kind of faggy tbh?

Most flags don't really have overtly elaborate designs on them, and all the details are made up of solid colors. One exception comes to mind, and I don't even remember what country it was - it was an island Britbongistan colonized at some point, I think, but my geography is cancerous

I know what you're trying to do, mate, and I tried to play around with the colors while retaining your design. This is probably the closest you can get to a "realistic" flag. Red and yellow just doesn't work for me - the yellow does not pop out.
Though I think the design is inherently the issue and you might have to lose a lot of detail.
Are the elements really that inherent to society's daily inner workings to where you must reference them in the flag itself?

It's a symbol of purity and shit. Wouldn't surprise me if "humanity" jerked themselves off like that.
Though you did guys did just remind me that I need to design a flag for my own country as well.

Try something like this with black lines. Pure red on white is eyecancer.

...

The Elemental Lotus is a secred symbol for humanity, as humans are the best elemental mages, and magic is used on every aspect of human life
Thats very British

I see, gonna try it, but i have to leave now, will post results later

Probably.

Even worse. At least flowers were invented by God.

5s loop/10
Tracked music is very rewarding (especially the file sizes), keep at it and you'll get there.

R8 my title screen thing.

If Im in the process of optimizing a mesh. is it okay if I do pic related? As in, instead of connecting EVERY point of an arch for a wall, I skip every other one since you wont ever see the overlap. Red lines show where I should make more edges if I was to connect the arch to the wall at every point.

unless those small faces run along the wall the same triangles still exist

Unless it's a heavily instantiated mesh it's probably a waste of your time. What I'm talking about is stuff like trees that are placed literally hundreds of times in your level.
When talking about polycount in regards to rendering performance, what you care about the most are LODs. These are what it comes down to (assuming you're not working with a terribly unoptimized engine that doesn't cull what isn't seen).

Do any of you have any experience with id Tech 4 and would you recommend it? I was thinking of using one of its libre forks or an updated version of Doom 3 BFG Edition's engine for an FPS with pretty dark environments preferably not prebaked for mechanical reasons.

Return of the Obra Dinn/10

It has a potential to look nice but as of now it's too raw. The composition is bad, there is a lot of weight in the middle but then instantly pure void at the sides.

There should be more reflections in the water on the left and right.
Sky looks too clean, too empty.
You may want to redo that smudge next to "D", maybe the way it is now it references something in the game but to me it looks like you just drew it randomly without thinking if it looks interesting.
The version without the title has all of the problems mentioned but the lack of text makes it's extremely unappealing.

Also, the moon it too close to the border of the image at the top.

And the reflection is in the water is too narrow compared to the size of the moon.

God damn it I hate working on graphics.
There go my plans of finishing shit within a year.
It was based on this picture, actually.
The D is a sideways sunrise as well, actually, guess it doesn't look too good either.

Thanks though, I'll try messing around with it some more. Maybe after a month it will look tolerable.

Tried shrinking the sun to make it fit the reflection more, as the reflection's harder to rework.
The logo itself, though, now looks way too small.
So out the fucking window with that, then.

I hope that image is royalty free and/or you're not going to sell the game.

Both, actually.

You don't have to shrink it. You can retain the sun's size and add more space between sun and top border by moving everything (beside title) down.

Shrinking was to make it for the reflection better.
How much space DOES there need to be from the top, really? Is this fine, you think? I'll redo the reflection later. That, and more detail in general.

Any good places to, like, read about this sort of shit? I'm going to the beach for a while, could use a good productive read.

There are no gods in my fantasy setting

You moved the title down making it unbalanced again. If you moved the text back to the center and the background a little bit up then I think the general composition should be ok.

Unfortunately I don't know any books about composition and my theoretical knowledge is almost non-existent, I base my advice on gut feeling I got from some practice. Only thing I have read is "Framed Ink", which I recommend but it's more for movies and comics.

Also pic rel, they are hidden now but if you move it they may be visible again

Unpolished+low effort title screen/10

Why do nations have knowledge of what's the equivalent of an omniscient being (you the dev) in your universe?
Doesn't make sense.
Better to have some type of approximation, or interpretation of your symbols based on said nations culture (or favored towards an elemental affinity would be easiest).

The Elemental Lotus Symbol was made by human magi
Its like our representation of atoms in real life, its not a perfect description of the universe, its an abstraction.

Also Scarlian Empire flag done, next the Fae's Alldrasian Republic
The rebel flag will have a lion head instead of a flower, using the petals as a mane

How would you handle walls in a top-down building game?

I'm trying to test some graphics but I realized the walls are a bit more complicated than I thought. If I make the walls tall enough to put windows and paintings on, then they take ~4-6 tiles of space which is a lot. If I cut out the top of them when you go inside, you won't be able to see the windows or paintings on the walls. If I cut out just the front wall, then it becomes complicated to determine which wall is supposed to be cut and when.

Gotcha makes sense.
Then what's the true form of said elemental lotus? What does it actually abstract, and what's the perfect interpretation?
Do u not have this planned yet?
Could be neat, as different ideologies/factions could war/have conflicts based on their interpretation, and the different means to attaining this knowledge could influence the actual interpretation + religious practices.

**quick terminology:
common elent: the 6 in the middle (fire,water,air,earth,dark,light)
composite element: the 12 on the second layer, each made of 2 commons
celestial element: the final 8 elements, made out of 3 composites each
total 26 elements representing all possible elemental combinations, since you cannot combine opposite elements**

I was thinking of a second scheme where Austra and Borea (the elements in the center) are outside with the 6 other celestial elements, i could drag two extra lines directly from each common element into the the far away composites they form, so they don't go through other composites, there are several ways to arrange the lotus, this one is the more readable i could make.The Lotus represents how elements are combined, it could alsol be represented with tables or a list of the combinations.

i fucked that spoiler sorry

You could do it the way a Link to a Past does dungeon walls. On the outside, have the walls slope inward, and when the character walks inside, the walls slope outward.

alright, you're trying to find a plane
it's perpendicular to the plane 𝛽: 𝑥 + 𝑦 − 2𝑧 + 2 = 0
and the line 𝑔 {
𝑥 = 4 + 𝑠
𝑦 = 3 + 𝑠
𝑧 = 2 + 𝑠
, 𝑠 ∈ ℝ}
goes through it
wat do

Need some help to fill the blank space here, i have no idea what to add
This is the Fae flag, the circle thing is a representation of the alldrasian flower, the flower they drink nectar from
But i have no idea what other colors and shapes would look good here (other than stripes)

...

krikienoid.github.io/flagwaver/#?src=https://i.imgur.com/W7lh3DB.png

It looks fine without anything extra.
sage for double post.

That's a bit too much top-down than I want.

I think I'll just try to make it so that enclosed spaces are treated as rooms, and if you go into it all the walls on the front/inside get cut out like pic related.

Here's a few tricolors I just shat out

Oh boy time for another round of "What does Imgur think about gamedev" imgur.com/gallery/kbKY5

thanks, but i went with the

suggestion

And here the last one, the Dracari Horde flag
Now time to draw the map

...

The featured games for those are literally all mobile shovelware.

Finished a few more building props for my game. Still need a shitload more for a basic map. I'm going for a pueblo style setting.

noice

Recently pirated Rpg Maker MV and I need thoughts on this fighter class i'm brainstorming. The game will be turn based with an ATB system.(think ffvii or something)

The core mechanic of this character will be a combo system. The player will (eventually) make use of certain attacks that string up to a four hit combo that has to be executed in a quick fashion. To encourage the player to make use of the system, simply using the skills as is will do pathetic damage. In fact, depending on what skill you fucked up, the character might either hurt himself(falling on his ass, getting a comical cramp) or suffer a status effect(accidentally immolating himself when trying to do a fire punch)

Each step of the four hit combo will have its own unique property. Step 1 will be an attack that sets up the character. Step 2 will be an attack that temporarily cripples a single enemy such as a disarm or silence for a split second. Step 3 will be an attack that inflicts a debuff that can benefit a party member(decreased magic defense, decreased physical defense, decreased projectile defense, etc), and Step 4 will be a finisher that deals massive damage along with certain elemental effects.

Another thing to mention is that since this is an ATB system, I was pretty much able to fuck with turn speeds. So to simulate using an actual combo, I had each step to be modified to increase my turn speed. To keep this character from being too powerful, there is a downside. For each step into a combo you progress, the damage you can potentially take increases dramatically with the values of +25%, +50%, +75%, and +100% respectively. There's probably more nuances to this character, but i'm just explaining the basic part of it.

forgot to mention something about the last part

The damage you can potentially take resets after you complete your turn. However, if an enemy attacks you while you're doing that combo, then you will take damage according to your current value.

I still don't get exactly how it works, but i would like to see something like a flow of possible combos

StarterA (low damage) combos into Step2A or Step2B
StarterB (debuff enemy) combos into Step2A or Step2C
StarterC does nothing, but if you take damage combos into FinisherB
Step2A (high damage, but deal less damage each consective hit) combo into itself or Step3A
Step2B (deal no damage, counters next attack) combo into Step3B
Step2C (no damage, debuff enemy defense) combos into Step3A or Step3B
Step3A (high damage) combos into FinisherA
Step3B (moderate damage) combos into any Finisher
FinisherA (very high damage)
FinisherB (moderate damage, debuff enemy defense)
FinisherC deal no damage, greatly buffs attack instead until the end of the battle

The fun about this would be collecting all moves, you could have several secret moves that you need to travel the world and do sidequests to learn, maybe mix this with a stance system that powers up/unlock certain type of moves, and some moves change your stance.

But i don't think RPG Maker supports something like this

That's more or less what i'm aiming for with this character. In the game i'm sort of working on, teamwork with your party is essential to winning battles. His utility is to be a high damaging opportunity creator for your group at the cost of his defense. Also, i'm pretty sure that's all very much so possible in RPG Maker. It'll just take a bit of work.

New bread