Amateur Gamedev General ~ /agdg/ + /vm/

"Demo day is less than 3 weeks away" edition

Resources:
>>>/agdg/
>>>/vm/

Links:
>Wiki: 8agdg.wikidot.com/
>Beginner's guide: >>>/agdg/29080

QUARTERLY DEMO DAY SCHEDULED FOR FEBRUARY 2ND
Please contribute to the wiki if you can!

Other urls found in this thread:

jax184.com/projects/ultra64/Nintendo Ultra64 Programming Manual Addendums.pdf
psxdev.net/
glampert.com/2015/03-23/ps2-homebrew-hardware-and-ps2dev-sdk/
n64squid.com/homebrew/n64-sdk/
wiibrew.org/wiki/DevkitPPC
8agdg.wikidot.com/beelzebox
reddit.com/r/IndieGaming/comments/75yjlt/speebot_its_the_rolling_and_hopping_robot/
reddit.com/r/gamedev/comments/76i525/how_i_wrote_my_own_3d_game_engine_and_shipped_a/
drdobbs.com/parallel/graphics-programming-black-book/184404919
archive.is/jvaij
cs.utah.edu/~jsnider/SeniorProj/BSP/default.htm
altimit.systems/articles/Doom Engine Review/page3.php
bisqwit.iki.fi/jutut/kuvat/programming_examples/portalrendering.html
fabiensanglard.net/doomIphone/doomClassicRenderer.php
fabiensanglard.net/duke3d/index.php
mobygames.com/browse/games/action/1st-person/list-games/
en.wikibooks.org/wiki/C++_Programming
en.wikibooks.org/wiki/C++_Programming/Operators/Operator_Overloading
en.wikibooks.org/wiki/C++_Programming/Programming_Languages/C++/Code/Keywords
8agdg.wikidot.com/recording
cs.utah.edu/~jsnider/SeniorProj/BSP/default.htm
fabiensanglard.net/doom3/dmap.php
regurge.at/ql/
a.uguu.se/LH65218EvxoJ_sigma2_rel11518.zip
xseedgames.com/about/
lua.org/pil/contents.html
en.wikipedia.org/wiki/Lexical_analysis
disagreement.gg/7Fvzub
godotengine.org/showcase/
assetstore.unity3d.com/en/#!/content/10374
twitter.com/NSFWRedditImage

CAN'T MAKE BULLET TRACERS FUCKKKKKKKKK

Also, Notch's current project looks pretty nice, even though it's basically just throwing shit at a renderer. I want to say its HTML5, but I have no idea what he's doing

It's just nice to see him making stuff again, I think that's when he was at his happiest.

If I'm remembering correctly, from his earlier Tweets, I'm pretty sure this was done in Java/Javascript. I remember because he was trying to make a point about how it's a perfectly valid language to use by making a really powerful voxel world generator that's way better than Minecraft's.
Glad to see he's moving forward with it though. I fully believe the future of open-world (Particularly survival) games is in voxels. Would love it if he shat all over Minecraft and the shit direction Mojang has taken with it with this project.

A successor to minecraft with better combat and a return of some of the more harsher survival mechanics Notch wanted to implement would be great.
and proper mod support, for fuck's sake.

To be fair, Minecraft is 9 years old this year, and you can't be expected to get a game right the first time, you have to come back to a project with refined skills and direction for it to succeed.

Not to say MC wasn't a success, but that's a long time, I should hope he knows what he's doing this time around and I'm sure he can do well

Well fortunately he can't be corrupted by money again. right?

Only time will tell.

Notch comes off like one of those devs who gets way more credit than he deserves.

what game did he borrow from this time?

Why not just have a glowy projectile?

Minecraft :^)

Has anybody tried to dev for old consoles ? I'd like to make an engine that you can use to make games for the Ps1/2, N64, Gamecube, 3DS and other consoles. The 3DS has a port of OpenGL but what about the rest ?

jax184.com/projects/ultra64/Nintendo Ultra64 Programming Manual Addendums.pdf

Writing something that targets all of those is a pretty challenging task and any games made on such an engine would not use the full potential of the platform. OpenGL is too bloated and slow (every version), you will have to develop a cross platform graphics library that only supports features that every single platform can do as first class features. Similar to how 3dfx Glide was preferred on voodoo cards because OpenGL's featureset is was too bloated for 3dfx to be bothered to implement fully in hardware. Basically imagine textured polygons, and that's almost all you can do, PS1 doesn't even have a Z-buffer in hardware.

Contact the people here for Play Station 1 development.
psxdev.net/
This guy looks like your best bet for homebrew PS2 Development. Good God the PS2 looks like a nightmare to develop for. 10 processors?
glampert.com/2015/03-23/ps2-homebrew-hardware-and-ps2dev-sdk/
N64 development
n64squid.com/homebrew/n64-sdk/
Gamecube development
wiibrew.org/wiki/DevkitPPC

I have. Doing NES dev now. I've done dev for the DS, 3DS, and N64 before. Good luck. Maybe PS2, Gamecube, and 3DS, but N64 and Playstation are still really pretty low-level and you have to use very specific programming practices to squeeze performance out of them. Even GameCube and PS2 have some weird hacks you have to use to push the full performance.
It's really not worth it, and you aren't going to get anywhere with that kind of engine unless you limit it entirely to gen 7 and above.
Seriously, learn development for all those consoles and implement some simple games in them, and it becomes quickly obvious how futile such a goal is.

I would love for you to prove me wrong, though. More of that kind of shit is always welcome, just know ahead of time that it's probably an impossibility.

Holy shit, what a mess. Would It be hard to just do a raycasting engine like Doom instead ?

AFAIK the GameCube "Flipper" GPUs native API was a proprietary Nintendo/ArtX API called GX. Not sure what version of GL that would work well with it on the devtools side

pdf related

Not really, we already have an user who is doing that and he seems to be making progress at a good pace. I can't find the wiki page though, it's not on the active projects or else I cant recognize it. So, go for it! It's easier because you don't need to support true 3D.

Anyone know anything that might make good background noise? I usually afk spectate random games, but the list is shrinking as old games become smaller and e-sport shit takes over.

...

I need a little more than music. Just something going on in the background.

I've been listening to Red letter media's half in the bag lately.

I'm messing around a little, but why not older e-sports? You can have things like the WCG 2006 Quake 4 final between Fatality and Toxic, or other games from this channel. It's done in a much different style than modern e-sports, so maybe you'll like it.

I have a question for you programmersfags, specially those with experience in unity.

How long and how much money would you ask for making a visual novel basic engine (like renpy) + a really basic strategic game. Something like fire emblem but a lot more simplified, very little units, little actions, a handful of maps.

I'm only asking about programming part. No the art or anything else, just leave it ready so it could be used immediately.

I'm working with a friend for something along these lines and he's taking care of programming and, well, he's taking long and I'm not sure if he's actually doing a good job or if he will need some help. However I'm a writterfag so I don't know how the market for programming is going and I need references.

Sounds like you could make it in a little over a month or so, I mean if you were working on it full time you could have a working visual novel engine in like two weeks, and a working battle system in say three weeks, and then of course extra time polishing it and taking it from a bare minimum state into a functional, working engine. I don't know how much money I would ask for, I guess enough for rent and food and a little more than that. So like $1000? I don't know how negotiating works but that's like over a month of living expenses. That would probably get you a basic engine that does the bare minimum of what you asked at least.

Granted I wouldn't ask for actual money, I would just ask for some % of profits since, I don't think that any hobbyist dev can afford to hire people in the same way an actual company hires people, unless you're getting paid by some richfag to program his furry vore game

my codebro quit work and is working on procedurally generated rooms.

the white rooms are his attempt, the super random dark ones are mine. In the end we'll likely join forces.

Things look optimistic, although we're far off from achieving something symmetrical and gothic-like architecturally, like the level I put together by hand.

Is there anything in the layout you'd like to see?

procedural generation that isn't just the same 10 or so rooms in a different order

duh.

Secret passages/enigmas/puzzles.

The simplest part to understand, yet simultaneously the hardest part to get right

that sounds cool. How would you like to discover them? Hidden switches? Levers? Breakable walls?

I completely agree. And I'll elaborate: we're still both learning and gathering ideas of what we'd like to see/what the rules of the generation should be.
The attached Diablo 1 map is kind of the way we'd like to see it generate, with maybe a bit more symmetry and rooms-within-rooms. Players also told us that really claustrophobic rooms make the gameplay annoying so we'll be trying to minimize those. My codebro is pretty talented so I expect the results to be more like the first two Diablo games than what nuBlizz graced us with in III, which is exactly what you called it

I have to ask, does your game have elevation differences in it? You are the guy who is making 8agdg.wikidot.com/beelzebox right? I think that it would be cool if you made areas with elevation differences, like rooms with bridges and bottomless pits and stuff.

I guess what I want for the actual generation is, twisting hallways that go through the map, also rooms that have lots of cool pillar layouts. Guess I can't say what I do want, I just want something where I don't have the "Dega Vu! I've just been to this place before….!" reaction. Like some way to change the fundamental way of setting up the level many times.

A mix of all of these. Breakable walls, or overgrown vegetation to cut with your sword. The occasional huge empty barrel of wine you can go through, maybe some secret passage behind a bookshelf or under the bed. Something with statues where you have to make them hold certain items or whatever.

Yeah that was my goto for a while.

Good idea.

>You are the guy who is making 8agdg.wikidot.com/beelzebox right?
yea, that's me. Forgot to flagfag.

Not currently. With the ortographic/isometric camera we're using it wouldn't be too visible, at least if players and monster were to be placed on different elevations.
Bottomless pits are a whole different thing and should be doable. We're planning to have different dungeon types the deeper you go so they might appear in a cave. We'll have to figure out how to implement them, though. Likely a terrain alpha mask. We'll see.

Yeah, pillars in gothic architecture make me hard.

I see what you did there. We'll see to what extent that's doable. I never had that feeling in Diablo, even if the number of assets was limited. There's plenty of resources on dungeon generation algorithms, I'll try to make use of 'em. although the mose useful one would be the source code of the legendary abandoned D1 remake that had 'em use a decompiler to get all of its secrets

that will most likely be a feature of the cave maps, don't know about the dungeon/church.

cool. I'll try those. This idea is really something. I wonder to what extent we'll be able to make it randomized/procedural but should be possible

How do you guys navigate your code? I just broke 1500 lines in my character movement code and I'm starting to get really lost when searching for a particular function.

Yes but it's rude to me sometimes

I got scolded when my inventory code reached 900. Better refactor it or place in separate, smaller scripts.

Also, if you can't do that, logical names for functions help a lot

Different source files in subdirectries. If your movement system is that complex, divide it into subsystems in your head, and then move the relevant functions into files named after what they generally do. Refactor if you can't keep track of it anymore, basically. Multiple windows in your text editor helps, of course- so you can see different parts of your code at once on the same screen

I got scolded when my inventory code reached 900. Better refactor it or place in separate, smaller scripts.

Also, if you can't do that, logical names for functions help a lot

Does having a child affect your short term memory? Asking for a friend.

Lack of sleep does. I deleted that previous post when I realized I replied to the wrong post, you just see what's left of it.

ok i'm about done with the arm i fixed the hand looks good to me. but i think im wrong about how the arm connects with the torso, and i would like a second opinion. next the the legs and how they look before a boolean, however if i plan to do clothes do i really have to worry about that, and this applies to the feet?

Are you doing this from memory/imagination only?
There's plenty of things to fix if you do this with an anatomy book side by side.

The neck is rather long, the shoulders are almost non existent, the knees are too low.
You could greatly benefit from doing a skeleton first.
It's definitely not bad, but you gotta do some rearranging.

I feel like my blueprinting is slipping into Diploma Dev territory.

meanwhile I can't even remember how long it's been since I did anything substantial with minecraft's source, the last improvement I made was simplifying block harvest with a list that associates the item dropped and that must have been over a month ago

Blueprinting is diploma dev by nature.

Voxels and such still vex me. I understand the principal behind quad/octrees but I haven't implemented them because I always get confused about how to actually do it.

In my own case, I would either do it as a way to arrange terrain/tiles, or else collision detection, but apparently they would use two different implementations of said tree.

Also, another difficulty I have is converting data into 3D points. My grasp of 3d is tenuous at best, but to generate a mesh on the fly and connect the points? No way

Also, promoting your game. I still maintain that Twitter is the best platform to do it, but there is also Facebook, Reddit and Tumblr, depending on how you feel. Personally I have a main blog on Tumblr for big, meaty posts, and a Twitter that I use to shit out random screenshots.

I feel like Twitter is good because it has a larger audience (including bots), and great built-in analytic tools. Bots are actually a useful tool, too, because most people follow two or three pertinent gamedev bots, which reblog certain posts if they're tagged appropriately; basically if you do it right, you're forcing a whole bunch of people to look at a popular channel and showing off your game that way.

I use #agdg, #gamedev, #indiedev, #indiegamedev, #programming, #screenshotsaturday, and #pixelart, with the last two being very popular tags.

Just reminded me, since it's saturday.

I'd consider Twitter and maybe Tumblr because no one uses RSS feeds these days. Reddit isn't worth it because their feedback is actively harmful and all their content is just reposts of other sites' stuff, so the best way to get reddit's shekels is to post your stuff on Twitter or Tumblr, wait for some redditor to repost your shit on his favourite subreddits, and never interact with the cancerous faggots at all.

I'm pathetically inexperienced when it comes to programming in general, and java introduces concepts I'm completely unfamiliar with, but minecraft's code is a bag of tentacle rape, everything has been so meticulously defined that you really can't reuse anything without rewriting the class or implementing a new one. From one class to the next I can't remember what method did what, what was public/protected/private, and over the past few weeks I can only guess that minecraft performs a fuck ton of unnecessary read/writes when it comes to player statistics, writing to the stats file every time something is updated (like JUMPING) instead of just storing the data in a public hashmap and writing to file on world unload.

I see, thanks user. Well he's not working full time but it's been over 3 months already and we have almost nothing (actually it has been longer but we kinda were just messing up trying engines and shit).

I'm in a bit of a bind because since he's a close friend (like my only irl friend) just coming and saying "hey I want more people to join to help" could probably bring some disputes. Or not, I don't know to be honest.

Working with friends is kinda shit to be honest, I think in the future I would prefer to work with people that can proof some experience in the job.

The thing with reddit is they blindly follow the herd. If you just ignore everything they say they can't poison it. At least I hope.

Getting started with making a 2d shooter, not going to be a roguelike like other games in the genre but instead follow a linear story. Got basic camera working, adding some guns and enemies next.

Is this a tranny:?

This is AGDG, not AGDQ, mate.
don't worry, it's hard to tell for me too.

I don't have a problem with what Minecraft is. Just that after all this money and time Mojang never thought to fix the spaghetti code the game is built on. I'm pretty sure Notch himself said Minecraft is far more computationally intensive than it needs to be because of his inexeperience when he first made it.
They got bought for $1 billion and they never thought to invest any of that into refactoring the code and that's what I take issue with. Instead they left the code as a big mess and continued to add to it, further crippling the performance.
Apparently there's a new version that runs much better because it's built on C# but it's only available through the Windows 10 store, and fuck that.

That UI looks very slick and very MGS3/Peace Walker

I miss the good ol' days where you just download a .zip and run the .exe. Too much corporate bloat, even on simple games

(checked)
It's JS and WebGL. It's impressive, but I personally don't like the idea of it. Browsers are quickly gaining ridiculous amounts of bloat. Any major browser is like the Emacs of text editors; they're so bloated you could just about use them as a replacement for an OS. Hell, with the exception of a browser most normalfags today only use webapps on their devices anyway. I don't really know if this is a bad thing or not; it just feels like we're headed towards a future where everything is a webapp and the browser makes up 90% of people's computing experience.

Also he's apparently still a shitposter at heart.

he never was and just used 4chan to get his game popular

The bigger issue about browsers being gigantic monoliths is that they are gigantic monoliths made up of the world's worst possible technologies.

I want no bullshit. If the player's crosshair is on target and he clicks, it's an instant hit. No questions asked. This is crutial. I've made "bastard" tracers out of a stretched mesh and make it visible for 0.01 seconds but i can't snap the vertex of the other end to my raycast collision point. Unless i make an aiming system similar to Metroid Prime hunters and/or Source mod "Underhell" the tracers are never going to be on point with the aim. Yet to do such aiming system i need to clamp the mouse inside a circle or a square inside in the middle of the screen. I need to somehow place the mesh instance between the point where it shoot from and the raycasts collision point then scale it accordingly in between the two points. I'm not a programmer so this is proving rather difficult.

Do publishers buy your html unity games?

like it's possible to sell them your shitty game?
like I can make some hundred bucks with it?

Redpill me on publishers.

If my file gets over 1000 lines, I break it off into two files. Of course I can only do that if it makes a lot of sense as well, so I do have some files that are maybe 1350 lines just because splitting it wouldn't help me navigate.

What you have to do is write a self-promoting post disguised as a tutorial or some kind of story that only incidentally happens to be promoting your game. Instead of "My video game" it's "How I did X and Y and made my video game". This will get you to be popular.

Compare:
reddit.com/r/IndieGaming/comments/75yjlt/speebot_its_the_rolling_and_hopping_robot/
to
reddit.com/r/gamedev/comments/76i525/how_i_wrote_my_own_3d_game_engine_and_shipped_a/

If it's his first time making something like that, then it would take longer than how I described.

At least it's about Speebot, i'll let you pass.

Sorry user, I am just trying to explain how shilling on the site works

...

Quite literally every post on his profile is self-promotion. It's not like he's participated in the community beyond trying to get his game noticed by more people.

that's not what fucks me off, it's the whole people here think it's a /adgd/ game one of us anons made but actually he was working on it on some indie forum well before he posted here but to see him doing that after the fact is really telling

Believe it or not, user, a lot of the successful devs from here have long since branched off of Holla Forums to participate in other communities. You can't build a successful career off of one site alone (a niece one at that).
It's just how the indie scene is.

that isn't in question though.

The problem is with you, not him.

you misinterpret my point

I've seen his progress since the very beginning on here, so I don't know what you are talking about. Granted I don't have archives of this but his first posts that I remember were about drawing a bunch of trees in a grid in one draw call for his engine.

I can't seem to find the indie forum I found abit back so I can't really back my self up here

I have actually seen it get linked at one point, so I do know what you're talking about, but I don't think that he was posting on it first, since I did see his earliest progress on speebot here.

ah I see

FUCK THESE GOD DAMN TRACERS I'LL SETTLE FOR THIS FOR THE MOMENT AND IT FUCKING SUCKSI'LL JUST REVAMP THE (((GRAPHICS))) FOR A WEEK

Can't you just
tracer.pos = lerp(ray.start, ray.end, 0.5);tracer.pointTowards(ray.end);tracer.scale.y = ray.length;

Still making random lewd animations. Won't be for another month till I start putting it together in Unity for VR. I need to retopo the hair and make it look better.

At least make another girls for others to fuck

Who said he's including it in the game?

Easy guess.

I don't think he has any plans to release the game

…Or i could just

That's still not up for brainstorming at the moment but is possible since the head is its own mesh. But I am going to be releasing it. This is a hobby project so I dont need it for moneys.

how do the lips werk?

or well, the mouth, i don't see a rig for that so

There is a bone for just the lips that only moves the lips and not move the jaw or teeth. So its realistic. There is a shape key that is made specially for fellatio "fera" that is binded by a driver to a bone Z position for influence so i can animate and easily adjust the mouth O. The face does have a bunch of controls though.

...

I'm effectively done with animation for the time being, at least until enemy attacks/AI and player hitstun stuff is coming up but that's a bit unlikely for the 2/2 deadline. Having nothing on the to-draw list is a pretty surreal feeling. What remains is to import everything and script everything, which hopefully won't take too long since my in-engine animation edit tools should work fairly well now.

Wait, is this a potential Minecraft 2?
Will this one not get shit up, and really be a 3D dwarf fortress?

The patience of a sloth… Good work!

this is literally like one hour of work fagget.
get gud.

more 50/50, i just want to see if i can make a decent model that i can use i just want to make 2 minimum, 3 max. as of things i am using for this it is mostly from tutorials and speed modeling that i see off youtube. when i have something wrong and i dont know what it is i post here, i know the shoulders are not there let me explain. pic is what i am using for proportions, but as you can see i had to do it in T form so to fix that i did it has pic then rotated the arms so its in T form, and that what i have here. but im looking for realistic, im trying to go for PS2 era model.

I think I fixed his leg problem, if the user who pointed it out earlier could give me a heads up that would be good. Always looking for critique.
I'm also wondering if anyone has a good hair sculpting guide or technique, since the way I was doing it seemed way too tedious for a normal workflow.
Reference is Magnus the Red for anyone not following from the previous thread.

If that's the case, it's already looking good.
I'd say lower the head and neck a little and work on the legs.

Well, you're not wrong. It really isn't a lot of work to draw one animation in roughs, 30min to 2h at most depending on how well I understand the motion and how closely I work with reference animations. The problems come when you want to have like 30 of them, have limited motivation and constantly redo some of them because of changing the moveset and wanting a different move, so that's why I'm really happy I managed to finalize my moveset and draw all of that in roughs. That's, for now, the end of a chapter.

the bulge needs to be alot higher up as well as that leg bit protruding out alot more, like someone took a sledge hammer to his ankles

Bit of progress from the last week. Tossed some Hexen textures back in for fun, as a break from the "128" placeholders.

Hey, do you have a tutorial to share for that kind of engine ?

Here's a bunch of things I used for reference along the way, which are a mix of BSP trees and portals since I moved from one to the other.
drdobbs.com/parallel/graphics-programming-black-book/184404919 (chapters 59 to 62)
archive.is/jvaij
cs.utah.edu/~jsnider/SeniorProj/BSP/default.htm
altimit.systems/articles/Doom Engine Review/page3.php
bisqwit.iki.fi/jutut/kuvat/programming_examples/portalrendering.html
fabiensanglard.net/doomIphone/doomClassicRenderer.php
fabiensanglard.net/duke3d/index.php

Do you guys know if there's any gallery out there with lots of various UI from different FPS games? I've ended up realizing that I need to rework mine because it's not easy enough to read with peripheral vision. As I keep implementing things like parkour-like free movement, and stamina-based melee combat, this old-timey RPG-like interface isn't quite cutting it, at least the HUD doesn't.

I would really like to retain this style, but also increase the size of the health, stamina and ammo display. I haven't come up with a redesign yet, though. So, do you know if there are any resources that could help me with that?

Howdy, newfag here. I just had an idea for a 2D puzzle videogame and I want to try making it a reality. I'm an Arch Linux autist that cares about muh freedumbs, so I decided to use Godot. I watched some tutorials and so far I'm loving it. But then I realized that my coding skills are pajeet-tier at best and gamedev might be out of my league. The most complex piece of software I ever made was a 200 line python script. Can I just wing it with what I already know, or do I need to go read SICP?

It's almost an actual game at this point

The best way to code, from my experience, is to look up things you don't know how to pull off. Sometimes you might need entire features, like movement or collision, and at other times, just a command or two. Optimization will just come from logical thought.

This came out way better than I expected it to. Subweapons are fucking awesome.

This is really cute.

Exploding death by caltrop made me laugh. Looks good.

Do you guys have any info regarding the way classes in C++ can be used? What the fuck is a class, anyway? I learn through tinkering and I hardly can ever just sit down and read things because I just don't memorize anything at all, and can't focus for more than 5 minutes - I blame it on my head trauma. I was under the impression that every "thing" was supposed to be its' own class. You create an instance of that thing, and then it's stored in memory, right? Well that made sense but then I was looking through the way this one guy had his voxel engine set up, and the individual cubes were classes. I was very surprised by that - wouldn't this kind of thing rack up serious performance cost? Am I misunderstanding the way classes work?

I thought I'd be making structs representing cubes and then storing them in three dimensional arrays. Am I fucking retarded? How do I store voxel data? I have a lot of questions and I'm just not able to derive the answers from the information I find online. Please help me.

first thing that comes to mind, it's not a gallery of UIs per se but …
mobygames.com/browse/games/action/1st-person/list-games/

obligatory LEGO comment

hell yea. And I don't think the captured gameplay is bad for the eyes. Looks good to me.

I really am behind with the wiki. Good stuff, my dudes

With any language, I've always found it helpful to learn the operators and keywords for quick reference, since they're things you can "do":
en.wikibooks.org/wiki/C++_Programming
en.wikibooks.org/wiki/C++_Programming/Operators/Operator_Overloading
en.wikibooks.org/wiki/C++_Programming/Programming_Languages/C++/Code/Keywords

When learning OOP, they like to teach you with tangible "things", like a Book object has Pages, a Title, etc, or that an Animal class has a Walk() method, and a Snake is a subclass of Animal that overrides Walk() to perform it a different way. Personally, I've found that to be a completely shit way to approach it. Instead, I just think of a class as a group of data, with methods to operate on said data.

Now most of my experience is with C# so take what I say with a grain of salt, but in C++, you have classes and structs. These can be thought of blueprints. When you malloc or call new, you are creating an instance of that class (building a house from the blueprint). Actually this is a good analogy; if you wanted to create an array, you set aside a chunk of memory. This is analogous to bulldozing a street and building sidewalks. You're set aside space for the houses in memory, but you haven't built them yet, but the memory is still there.

So I can just wing it, learning as I go, and I'll be fine as long as I'm not a complete dumbass? Alrighty then, sounds like a plan.

mMedia Recode is a black magic

But I guess even with the relatively low quality the results get across.

This is rapidly becoming an HD remake of Diablo 1 and I am okay with that

Thanks for this explanation! I'm actually feeling much more free in thinking about classes now that you put it that way. A unit of specific data with functions to handle the data as a thing of its own.

So then - I'm thinking of a game object system (object as in anything, really, a chair, a rock, a huge chunk of terrain) where I'd like to represent the shape of it using a mesh approximated to match voxel data. That's all well and good - I understand how to do that in general terms, and I'm sure the particularities will be sorted out as I'm working on it. Funnily enough, the foundation is what gives me trouble. In this particular system I have in mind, I'm planning on subdividing the voxels into smaller ones in the space of the larger one. Now, that's where I'm stepping into a bit more of a foggy area in my understanding.

I get that I should be creating a class representing a voxel, which itself is set to either be an actual voxel, or a space that will contain 8 more nested, smaller cubes, and so on until the defined LOD limit. How do I store a multitude of voxels? I'll most often be having to store the voxels in sets of 8, so do I use arrays? Is this a good application of it, or is there a more applicable, sophisticated way to store objects that I'm just not aware of due to my overall new-ness to programming? I feel like I've almost fully grasped this whole programming thing, at least on the surface - but there are still these white spots I have, like the way I'm supposed to handle large amounts of data and such.

do you really have a 4k display? are they worth it or just a meme?

There's a few fundamental data structures that are commonly used in programming, especially to implement other more complex data structures. You have stuff like a bag (unordered set of objects), a stack (first in, last out), and a queue (first in, first out). There's also trees (very important) and lists and linked lists, but also stuff like hash sets.

In terms of performance, the best speed for read/write will come from using arrays - data is packed tightly all in a row, and the only thing you need to know is the offset from the start. A tree might be sorted in some particular manner, but it tends to use pointers, but even if it's backed internally by an array you don't see, the objects won't be together in memory, you'll still have to traverse the tree (usually in O(log) time). This is where data theory and stuff comes into play. Anyways, if you need to sort something, you should think "stack". If you need to use performance, think "array".

Of course, the effort you put into optimizing it into an array might be more trouble than it's worth and you should just use whatever is easier / makes the most sense / is easiest to maintain, because it probably won't actually be a bottleneck in the first place.

Sorry, I actually fell for a completely different meme and bought the famous Sony GDM FW900, a 2K widescreen CRT. The 4k resolution is Nvidia's DSR upscaling.

Thanks!

I am jealous, that must be very nice to own. I used to have a Diamondtron 2060u but it broke in the mail.

I've been trying to redo my gems, but I can't quite get the shape to look right.

I basically want them to fill up as much vertical space as possible (30 px tall), not be "door shaped", and sort of classic diamond shaped, but I can't quite get the tops looking right

It's hard to tell what perspective you're trying to go for but this might help you.

Here's a quick something I did.

Alright, this is more or less what I had in mind, thanks.
I need to adjust the brightness and shade it properly now, but does the top or bottom set look better?

The bottom. I took the bottom row and shrunk the middle section a bit. since whenever you see gems, the bottom section is about 2/3 of the gem height, and it's about 1/2 on yours.

woo, an original diamontron, nice. I have a Lacie Blue IV, which is a licenced diamondtron. In all honesty, the image seems better (both colors and sharpness) on that one, while the Sony monitor can handle higher refresh rates and resolutions.

Adding some shine to the edges of the cuts would help sell it a lot I think.

This is the AGDG thread, not the AGDQ thread.

it'll never happen I just know it

but I don't want to do pixel shit

I think they fit your artstyle nicely

and I think the top row is better

The center of the diamond is too tall and square. Making the cut a little deeper would probably do you better, similar to what suggested.

I know you don't want a "door-shaped" gem, but an emerald cut gem would suit your dimensional needs better than a diamond cut gem.

I generally like the bottom row. call them something other then diamonds and have them tbh

Pixel shit is only pixel shit if you do it wrong(undertale,fez)
You can do good stuff with sprites,but it's a lot of work,you can always just draw them.

Here's some examples of gems from several games that have face on diamonds.

and now we've got snow as well!
all at my fingertips, with a single checkbox to turn on or off 'snow' for any level.

i'll see if i can put together a webm or something.

I fucked up.

goddamnit, i need a decent fucking gif to webm tool or something, anyone have any recommendations?
15 mb of GIF is not quite what i would call acceptable for web-usage; i apologise to Holla Forums's bandwidth >_>

8agdg.wikidot.com/recording

Usually you want to break up your classes by then. Don't think of it as a single monolithic class, but as several smaller classes working together.

ffmpeg -i your_gif.gif -c:v libvpx -crf 12 -b:v 500K output.webm

Why the hell would you record to GIF then convert to webm, instead of just recording a webm? GIFs have horrible limitations, it'll end up looking like shit.

looks comfy

Welp, wrangling together assets for this took forever. Once I started writing code, I was able to easily generate multi layer icons making use of TextMeshPro, with a little hack to force them to draw in a given order.

Next thing is to make a simple format string for easily inlining these icons into my data and attach them to items/skills.
I still need to make systems for items/skills

For some reason I'm really, really happy with how these turn out.

this

Bro just download OBS and directly record MP4s, and convert to WEBBUM only if you need to crunch the filesize.
Seriously I don't know why people have trouble with video.

Nice progress man

Kinda looks like you are also getting hit by caltrops, or is that whatever the baddies are throwing at you?

The caltrops are completely harmless to players, I was getting maced because I was standing still like a nimrod.

You mates might find this interesting.
8-bit Guy made a game called Planet X2 for the C64, and is now making a sequel on DOS, except he's detailing the process as he goes along this time.

Reminds me a lot of the Canadian level in Sly Cooper.

That's how I feel about posting progress in this thread.

Pretty much. Nothing too impressive in my own mind about showing off some functioning collision, even if it took a lot of work, no one will ever really know that.

I feel shitty because I've only been working on art assets for like two months, when I wanted to get my behavior set up for the last demo day

We're posting to suck off eachothers dicks in order to keep ourselves motivated.
nohomo

And unfiltered criticism. If I just wanted people to suck my dick I'd post to reddit.

This fucking drove me mad.
I was using a plain C# project to quickly test some regex code to eat formatted strings for 'glyphs' to turn them into the TMP RichTextTags for displaying sprites

There's some inconsistancy between the way regexes are handled.
MSBuild or whatever handles `\[.+\]` properly, but Unity's MonoBuild needs to be clued in that the last ] is greedy by using `\[.+?\]` instead.
This caused me an hour of headache, and it's the simplest regex I'm using to scan for info with.

Besides that, this system came together quickly, and I'm very happy with the results I can get by quickly typing things, this will make cramming these icon glyph code thingies into data very easy.

It helps me keep working when I share things semi-regularly, even though I get almost 0 feedback sometimes.

Currently still following tutorials :)
About half way done on the coding tutorials, then im going to add pretty much everything I have planned feature wise before I even think about working on art and sounds

I am back to working on heavy programming tasks for Sigma II, and as such there is no progress that you can see visually. I'm writing the BSP compiler, this one should be much better than the last engine's BSP methods. I'm going to implement:
cs.utah.edu/~jsnider/SeniorProj/BSP/default.htm

But, I also have to do a reduction on the BSP tree, using the kind of method that is described here:
fabiensanglard.net/doom3/dmap.php

So that I am not just wasting a bunch of space, this should help keep the polygon count down a lot. Luckily I am not absolutely retarded at programming like I was when I worked through part 1 of this tutorial in 2015 but it is a shame that most of my progress is rather hard to see, since I have so little to post on a project that, I think I started on Jan 4 2017, so it's a year and 10 days old now. I should add texturing into my map format so that my levels look more interesting, right now it's rather dull.

I also need to learn how to make 3D models, I think that I will use milkshape3D since I don't want to use blender, it requires too high of an OpenGL version for me to use it on any of my laptops. That is not something I will have to think about for a while, though.

No homo, all bromo.

That's some pretty cool shit

Christ that brings me back

Milkshape 3d is complete ass, try and get another pc that supports the open gl version that blender uses or try and get someone to help you.

I have a PC that can run blender, I just don't know if I want to use it. I don't know how to use any sort of 3d modeling program aside from anim8or which I used to make all the prerendered graphics for Sigma II.

I only plan on making low poly models, so I don't really know what I will use, MS3D seems like a good choice since I can run it on my laptop though. I don't plan on buying a better laptop. I don't know the controls to any of these programs so they all seem similar.

Thanks m80.

The only value that remains in ms3d is for exporting to archaic model formats, you can use it but blender is easily the superior toolset, especially for texturing and animating.

...

lol i didnt read the thread :D

This isn't as cool or dramatic as the glass knife I made, but the spread knife is just one of those utilitarian upgrades that's just nice to have.

Finger is still broken, but I made some good progress on my map editor.
I can now place entities, which are dynamically loaded from the file system, and can be either dynamic or static. Static entities are batched (when possible), dynamic entities require separate draw calls. The engine batches static objects with common texture, which means that if the tileset and all static entities use the same texture - the entire map is compiled into a single model. Entities can be scaled, rotated, duplicated, snapped to grid or fine-tuned. The UI system also supports proper input fields now.
Additionally, it's possible to add sound emitters, arbitrary points and trigger blocks, which will later be accessible from code by IDs.

Also, both of my games on Steam are on sale right now - Hypnorain is 90% off and Speebot is 50% off!

I'm pretty sure I've seen this episode before.

Congratulations on your recovery. Hopefully the sale goes well, Speebot deserves the love.

How's typing with a broken finger? Which finger is broken, anyway? I fucked up my hand a few years back, made coding interesting for a while

Holy shit, that's fortunate.
It'll take me a few days to implement, still, but it's neato nevertheless.

Anything good happen to you anons lately?

Why did you chose love2d over GMS? I'm asking because I'm trying to decide which of these I'm going to use.

Make sure the level design supports it. There's few things more disappointing than getting nice AOE skills in games where enemies tend to be spread out

I started on the tedious task of menus and settings.

I began development in GMS (1.4), mostly as I had had six or seven years of Game Maker experience under my belt. Casual experience from joke games but experience nevertheless.
It was going fine enough, though at times I did need to cheese things to get them to work. A benefit of GMS was that every feature had functioning examples already uploaded on jewgle somewhere if you looked hard enough.

I ran into an issue around the holidays, though. I unplugged my controllers from my PC to go play some Snowboard Kids 2 with my little sister, and apparently, the way I had implemented controller support made the performance tank to sub-10 fps (from an average of 1200-1600) when there were no controllers present. I tried to band-aid it by reimplementing all the controller support, but no dice, it improved but not anywhere near to be playable. I believe it's an issue with the way GMS implements Windows' native controller support, to where it asks the system itself for a list of all USB devices and it takes a while. There were suggestions on how to unfuck it, but none of them worked. It was possible, you see, to implement xInput controllers just fine (those being Xbox controllers and some newer controllers), but DirectInput ones (those being every controller that doesn't cost over 50 bucks, and also happen to be the ones I use) just would not work.

I tried, then, to port the project over to GMS2, which had supposedly fixed the issue. When I switched it on, it was the most hideous thing I had ever seen, it's workspace is broken, it's impossible to put certain windows behind other windows, you have to scroll around their new interface which was a fucking pain as there were no bloody scroll bars, so if you wanted to move sideways, you had to zoom out and zoom back in on the spot you wanted.
It was a fucking mess, and from what I read on their wiki, YoYo wants to force this shit on people and aren't planning to fix it, or even give people the option of using the old layout.

So I bit the bullet and moved to Love2d. It was overwhelming at first, but after days of jewgling really hard, I got basic functionality, and started porting over features. I got sprite animations, movement and collisions working, as well as ways to store in-game rooms in separate Lua files for better organization, and so on. The absence of a Room editor is strange after coming from GMS, certainly, but I'm getting used to it. There are libraries that offer room editors but I don't really need one. I use GMS' room editor to complete a basic layout for things, and then just punch the coordinates into Love2d.

The two are very different. GMS is simple, easy to get into, bloody well documented (though inconsistent if you're coming to it from an actual programming background, according to some anons), and it's OOP is easy to understand.

Coming from GMS, Love2d was overwhelming, and you have to do a lot of things on your own. Functions like rounding up numbers, checking whether two objects are colliding, or comparing the distance between two objects are functions you have to write yourself, whereas in GMS they are already there from the getgo. But, as you might have guessed, the controller issue with GMS1.4 and the disgustingly unusable workspace of GMS2 makes Love2d a better choice in my eyes.

And really, Lua is bloody well documented too, it just takes more work to get things done. There's just enough of engine-development in Love2d to be fun, but enough is done for you to where you won't waste ten years and have no game at the end of it.

I wanted to make that my next task, but I want to implement a few more features, mainly in-game time, and then develop a save system before I make the menu.

CS student here 3rd year
Should I make a game for my senior project? any of you tried this? no idea how to game develop. maybe ill start early

yeah i kind of stuck on the legs and feet, mostly because i'm going to put clothes on it and that means shoes, so what am i sculpting, the shoes the feet or a little of both? also going to lower the head once i boolean the ear to it. so how does it look so far, and what needs work?

ur a bro, my dude. Although look, I'm not using pentagrams as a selection symbol!

I'm still working on the menus. The one I made for Mr. Gigafist actually got really useful. There's some tweaking to be done like a confirmation prompt but the results will be good enough for the demo day.

turned off texture size graphics option because it fucks with the generated item icons

hope you'll get some sales. I already own both. Did you decide on what the new YUME game will be or is it just tinkering with the engine?
tell us about your sales

Oh I could watch that one all day. although I remember the static being more literal white noise when I was young

so how about that .webm?

Damn this nigga's progress is fast

he's had YEARS from minecraft to learn how to into this kinda shit

Yes, the new GM studio versions are absolute cancer trash. It feels like subscriptionware built around a game engine, not the other way around.

I always thought the consensus was that everyone hated him and yet everyone, given the chance, would make the exact same choices he did.

I don't really hate him.
I sympathize the downward spiral he went through after his rise to success and suddenly people started turning on him or trying to take advantage of him, and sometimes he can be a bit of a cock, but I thought he was mostly an okay guy.

the trouble is people wanted alot more out of his game then notch really cared about doing

Is there any framework for SFML that helps you with animations or how do you animations in SFML?

Modding Dark souls and probably going to convert the entire game at some point into a berserk game. Or maybe I'll lose interest like I do with everything
I'm figuring out this editor a lot in just a little bit.

2D or 3D?

If it's 2D all you really need to do is store a series of texture coordinates, and interpolate between them. If it's 3D you can just have a transform matrix per bone, structure it as a tree, and then attach it to a mesh group and cal it a keyframe.

That's the most naive, simplest way, anyways

2d, what about the time component? how to make it independent from the frame rate?

None that I know of but I had to do this last week.

Basically what I do is I have a two dimensional array that I populate with the sf::IntRect coordinates that I loop through.

I change the objects' sprites' rect to the returned rect for that particular frame.

I use the sf::Clock to hold onto a timer that only resets when the next animation is called (so that the framerate doesn't affect it)

Yep that's what I do lmao

What about it? This is how I do it.

My Image class is literally just a reference to a texture object and a Rectangle[], which holds all the frame data. Images are freely shared between objects since they're supposed to be immutable. The actual objects that use the Images have their own Timer, Transform, Color etc data that tell the renderer what to do, but as far as the "animation" is concerned, it just fetches whatever subimage it gets told to.

So for example, my Timer has as GetPercent() method, that is guaranteed to return a value between 0.00 and 1.00, I plug this into Image.GetFrame() and out pops the texture coordinate I use

White theme? I dont think we can be friends.

FWIW This is how I do it.

I don't know how to easily change it in visual studio, and Notepad++ doesn't have the auto indent shit I like

you need better references, i saw the picture that you keep posting as reference, but it doesn't cut it, you need to google references for each individual part and do them separately, once you have that, you can create a whole humans, even from imagination.

Does anyone else get really invested in creating expansive option menus? A part of me just keeps coming up with obscure options that someone may need and implementing them.

I know that a lot of controllers are very varied, and certain features are implemented strangely in them (Eg. my chink NES/SNES-style controllers have no analogue sticks, but for some reason, Windows thinks that they do and they're always tilted up).
So I'm adding really stupid features that someone might need, like the option to switch the X and Y axes on the joystick, mirror the axes and disable the joystick entirely in case there's a cheap piece of shit out there that doesn't work properly and needs one of these settings to work fine.

What utter nigger invented Lua anyways. I swear to god I feel like I'm just one rung above the idiotic script language Morrowind used

I'll take that over imperfect spacing breaking all functionality in Python, especially with my shit eyesight.

This is annoying but at least forgiveable: It makes embedding functions/tokens a lot easier, apparently.

But also

I actually like python's indentation based scope. It keeps the number of brackets down which makes for more quickly readable code where you have plenty of brackets already, at least for me.

1-based indexing threw me off I can't honestly criticize it, as the distinction between starting at 1 or 0 is entirely arbitrary.

He lives in a Beverly Hills mansion with enough money in his accounts for the next 8 generations of his family to not have to work a day of their lives. He quite literally has all day to dedicate to this and I'm glad that unlike many people in that situation he's actually doing so.

Modding Dark Souls is an uphill battle, I salute you

I personally have not used any model formats, and I will probably use a custom model format similar to obj but compressed with LZMA to save space, also I don't like how some of the tags in obj work either, and I want my own tags.

Now, I don't know much about making models but I could also instead of using other people's software make my own model creation software, which would be incompatible with everything else (obviously) but my own formats which I will stop supporting in two years anyway. That sounds like the coolest thing to do although it would have the downside of taking a while. So I will probably support some subset of .obj and then have a seperate program that lets me edit them further. I don't have a plan though so I will start with .obj.

Just implement a command line. Basic options that everyone uses go into the GUI menu. Obscure and esoteric options are only accessible from the command line. This is how it's done in id software games and it works great, just because there is no way to fit hundreds of options into a GUI, so a command line is the best way to expose them. Check: regurge.at/ql/ . It's obviously totally impossible to use a GUI for that, and QL already has a 10-screen GUI on top of that giant manual. Plus this manual is probably not even a full, complete guide, it's pretty reasonable to think that a few commands aren't documented in there just because of how many there are.

...

Whether or not that's a good idea depends on the game. Your typical game has far more things attached to it than just the code/software architecture. Sounds, art, etc. are going to be overhead as far as CS is concerned. You have to create them, but they probably won't add anything to your score, because they're thematically unrelated.
One thing you could do to solve this conundrum is to rely heavily on procedural generation. A simple 2D game that is mostly generated might work. Think about generating textures and synthesizing sounds on the go. The problem with that is that it could easily become too much. You can write a doctor's thesis on synthesizing sounds alone.
If you actually want to do this, you really should ask a professor.

2 player hotseat vs mode thing for demo day. It's functional, right now I'm just working on bugfixing and cleaning up things for player benefit as much as I can while I don't actually have my own art assets.

I am no longer sure if you are going too far or not far enough.
Hows your map editor will users be able to make intricate architecture as in quake or will it still be quite limited similar to that of your first engine?

God bless

Is there any better way to handle config files for rebindable keys n shit?

This works but it feels dumb.

Switch statements are made for this kind of thing.

Yeah, basically what said, just use a switch statement. In most cases they should perform identically, but some compilers can optimize a switch statement a little bit faster (not that it matters at that scale)

In either case, an if/else block is a very good sign you should use switch

Especially if each statement is just checking equality for a single item.

...

the drain flies just WON'T FUCKING GO AWAY and I can't concentrate for more than 10 seconds before another appears and I have to kill it

If you want it to look neater, use a std::map

This is akin to my interests

It would be cool to make my own 3d modeling software. But, I don't really know what I will do until I get there. My map editor will definitely be more capable this time, but it will also be more cumbersome than the kinds of tools quake gives you. I plan to make it so that you can make 90% of the stuff Quake can do , if not as quickly. My current map editor is so much better than my last one, so I am going to be a lot more efficient with it.

You can try it now, its missing a lot of features I know but the general idea is there: a.uguu.se/LH65218EvxoJ_sigma2_rel11518.zip

Okay, engine is finally organized-ish, working on behaviors again.

Thought i'd show off what i have for my game so far i've taken inspiration from games like m&b, freecities and COC

I'm not talkin' softcore Nintendo lounge…

Tried to do up some directional crates, but they definitely need more work

Ok dude: I don't know what engine you're using but there has to be a way to get a bullet's position/direction. With that info you can just render a line (make it a gradient for bonus points) from the position of the bullet and towards the opposite of it's direction. Doesn't sound like it should be difficult.

Ok dude: I don't know what engine you're using but there has to be a way to get a bullet's position/direction. With that info you can just render a line (make it a gradient for bonus points) from the position of the bullet and towards the opposite of it's direction. Doesn't sound like it should be difficult.

Diablodev, I'd like to become part of the project, if you'll have me. I can code (java, javascript and a bit of C), have experience in mapmaking (hl1's hammer editor way back when), I also do 3d modelling/animation (I work in VFX and motion graphics). I just want to work on vidya.

how do you fags do it?

Just like make game

Go the Nintendo route with it. Don't think about what you'd like to see from a video game, think about what you enjoy outside of that and make it into video game from there. I'm really big into Tokusatsu but the lack of a modern and comprehensive experience for the genre has forced my hand into game development. The latest game for the genre was the terribly optimized and extremely underwhelming Godzilla game for the PS4. Fucking disgraceful, I'm sure I can do better than that game that looks and plays like a university exit project.

Wanting to make a game you'd enjoy playing is a deceptive goal. Before you'd be able to play it you have to spend shitton of time working on it. You should rather think about what type of game you'd not only like to play but also what game would be fun to make.

What game would challenge you in fields you like but also use little of shit you find boring? The answer to this question is just like make a game, if you don't like it than just like start making a different game, repeat until you find what you like.

...

What's the idea?

You just gave me an idea for a neat game. Giant, feral, partially biological mechs fighting in abandoned cities against piloted mechs. The main combat mechanic being throwing and kicking like Dark Messiah and knocking buildings onto them. Changing the landscape by knocking over skyscrapers to form barricades and block paths of travel would also be a must.
Thanks, user. I'll start small and make it a 2d turn based game then look into how to make it more detailed.

Anime girl urbex/general exploration game.

Sounds good, I was considering GMS because it looked simpler enough for someone whose only previous real experience was with RPG Maker, but I'll give a chance to love2d, thanks mate.

It's weird, the little experience I've had with GMS and GMS2, I actually prefer the interface of GMS2, but I guess my opinion should be taken with a lot of salt seeing as I've yet to sit down and really try to make anything with it.
I want to really try to do something with GMS2 in the future, my friend helped get it when it was on sale.

Shoujo Shuumatsu Ryokou game?

the gauntlets been thrown niggas are you gonna up the ante?

and what has this guy made?

fairy bloom freesia, astebreed, sakuna of rice and ruin

I don't think he's wrong but I think Cuphead at least deserves a pass somewhat.
Who?

Some nobody jap who pumps shit out constantly is “throwing down the gauntlets”? Amazing.

talk shit when you get xseed asking to distribute your games for other countries

So does this guy have a job or does he do this shit full time? Is he a 1MA?

I would be happy if I finished my shit in 5 years and have no desire to compare myself, my game or my development speed to any other indie game developer, so fuck off.

ftfy

he does rice farming

...

That's really intesive work a short period per year, and a whole ton of waiting for the rest. So no, most of the time he does not have a job.

I think he's fair to criticize the games themselves but it's up to a dev how long they spend on making it. Examples like Owlboy and Iconoclasts aren't really indicative of the average dev cycle.
Plus, a lot of times it's not focused work. I've been making concepts and writing shit for my game for about 3 years, but haven't gone into full production since I haven't had the time. I wouldn't really count those 3 years as actual dev time.

good, you should feel motivated to work harder and get better at your work instead of getting hurt feelings

I just don't feel like what he says applies to me at all, much like how people criticize pixel art. People usually look at the worst cases of developers who don't try and make generalizations like

What really motivates me is seeing other devs succeed or fail because it reminds me I can succeed as well or do even better.

Running out of names for weapons sucks.

to be honest though, most 'pixel art' games are not just visual fucking garbage, but are also garbage in every other way.

i recently read some article about some dumbfuck hippy who quit his job to make a videogame, even though he had no experience except some coding abilities, he made some fucking windows 95 screensaver-tier mazegame, and was shocked that nobody bought the game.

absolutely pathetic.
sure, there's exceptions, some pixel art is done fantastic, or has a great style, or it's got some gimmick of haivng the visuals interwoven with audio, or just REALLY solid gameplay, but EASILY 90% of all 'pixel art' games is literally fucking cancer wasting space on the internet, clogging the intertubes.

If he's NEETing it up, chances are the NEETdom itself was induced by mental illness

I have a job and plenty of free time. I can get into a routine but I'm incapable of keep it up for very long. There's literally nothing in my way to making a game other than mental issues.

The only advice I can give you is to live life in a way that makes you happy, and anyone who stands in the way of that should be removed from your life.
Stop listening to the typical advice, whether it be from normies, (((psychologists))) or whoever else and figure out what helps you. Generally, waking up at the same time, getting at least 8 hours of sleep and eating better helped me, as well as removing all people (except for one) from my life. I used to desperately seek friendship because that's what everyone said would make me happy, but it took time away from my hobbies (mainly gamedev and cooking), and at the end of the day, I was tired and miserable. Limiting human contact to one person and anonymous interactions on this site helped me immensely.
I don't know what will work for you, user, you have to find that yourself, but you can try this out, at least.

Sorry, I was looking at the wrong link. The games look cool but there is no way that this guy doesn't have a lot of money and a small team. I can't find any info though, since the site has so little information.

Not true. At one point I was NEETing it up for a year because I didn't know where to go or what to do. I was depressed as shit but got out of that pit once I found something and started going to school again. Routine goes a long way.

I suppose find the reason or cause for what makes you feel like shit. It varies from person to person but it's almost always fixable. Really all you have to do is find the cause and avoid it or at least mitigate it as best as you can.
It can be something that's not quite obvious, like a fucked up sleep schedule or lack of social interaction. For instance, I've noticed on some days I feel awful simply because I don't get to talk to my friends, but even that is nowhere near as bad as when I was actually isolating myself entirely.

If you don't know the cause a therapist can help (the one that talks to you to figure out where shit went wrong, not the one that tries to put you on meds), but you can also just do some self reflection or talk to friends about it to figure it out.

Also has some good advice. I wouldn't cut contact with everyone I know, but definitely try to distance yourself from people who stress you the fuck out on a regular basis.

What IDE/text editor do you guys use?

What if he's a diaperfag?

So that was my original idea, but C++ apparently doesn't like switch statements on a string it complains because it's not a primitive type like int.

When I was googling apparently most people make a function to take the string and turn it into an enum but that looks just as dumb to me.

Why hello there, Yandev.

Hematite Thirst, is what I'd call that dagger

Pls no bully I am smileydev k thnx

I use Microsoft Visual Studio 2010 C# Express Edition

oh oh oh oh?

i appreciate the references, but i mostly use pic>>14156478 for proportions as said in post. as for references i sometime use my own body like hands and arms, the only time i look for references is on a body part i cant look at like face or ear see pic. however i'm saving those pics they're too good, you wouldn't happen to have a female version?

Switch on string fails because it needs to be an integral type, not just a primitive. It needs this because the compiler generates a jump table from it to make it far more efficient.

You can do successive if statements in that case, but that's stupid. You're far better off doing an unordered map to map strings to enums if you're going to do it like that. Something like:

const static std::unordered_map keyCommandActionsMap = { {"moveUp", keyCommandActions::moveUp}, {"moveLeft", keyCommandActions::moveLeft}, {"moveDown", keyCommandActions::moveDown}, {"moveRight", keyCommandActions::moveRight}, {"inventory", keyCommandActions::inventory}, {"pause", keyCommandActions::pause}, {"actionOne", keyCommandActions::actionOne}, {"actionTwo", keyCommandActions::actionTwo}, {"actionThree", keyCommandActions::actionThree}, {"actionFour", keyCommandActions::actionFour}, {"actionFive", keyCommandActions::actionFive}, {"modOne", keyCommandActions::modOne}, {"modTwo", keyCommandActions::modTwo}, {"modThree", keyCommandActions::modThree}};

And then you can just set it with:

const auto it = keyCommandActionsMap.find(config); if (it != keyCommandActionsMap.end()) { rebind(key, it.second()); }

Maybe putting an "else" on there for invalid config values or something. Personally, I'd embed lua or something. Using Lua as a config language is pretty easy.

guess i'll just keep working on art and music and hope I can push through.

It's just that Yandev also made this exact mistake, not knowing what switch statements are and using strongs to compare a lot of stuff. The difference being that you're actually processing text from a text file, while he just used it as logic variables, which is slow as fuck.

Okay cool, this helped me alot. Thanks man.

I'll look into lua as well. I haven't done any scripting involving another language before but a lot of the stuff I've been reading lately keep mentioning it. I'll hit the googles to learn more.

I think I saw a diploma-meme with his code before like pic related or that might have been cooking fairy dev?

Does fairy dev post anymore?

He still posts pretty regularly. Lately he's been posting character models.

Guy has a team of more than 10+ people though. Astebreed has 13 not including individual instrument players and publishers.
Undermeme was mostly developed by one person with kikestarter funded assistance coming later to get the game out.
Stardew valley was also one guy who got help from Chucklefish Games who were also working on two other games at the time(so he either had less than 10 people working on it at any given time or he was getting little bits of help from all them)
Salt and Sanctuary is from a two man team
Inside is complete art game trash anyways
Cuphead took 7 years for hand-painted animation cels, it gets a free pass
Iconoclasts isn't out yet but from past game credits Joakim Sandberg does his games entirely on his own with very limited outside assitance(For Noitu 2 the only outside help he got was "Guest Boss Design" and testing)
Owlboy was made by a studio of 6 people
Really easy to say games take a long time to produce and aren't very good when you have more people working on your games than they did. I bet if any of these games had 13 people on staff they'd be just as good and probably better than the games he has made.

"Sponsored by Microsoft" doesn't mean anything when Chad had to put two more mortgages on his house to fund the game's development. I know it's a semantics argument but Cuphead is for all intents and purposes an indie title

If you go the Lua route, I'd go through the blue PIL book first (it's really not incredibly long, and it's free online) to get a feel for the lua language and a taste for embedding, sticking with version 5.1 of the language at first, because it's largely compatible with 5.0, which the book covers, and LuaJIT is 5.1 if you want to use it in release builds for the best performance: lua.org/pil/contents.html

I would learn it even if you don't use it for your config, because it comes in handy in tons of little places as well. It isn't the perfect scripting language, but it is a scripting language with a good-enough syntax and with the best available embedding API and damn good documentation, not to mention a massive library of community libraries. mruby's API is close, but the documentation is very shitty, and there isn't a real good tutorial to get started with embedding it; it also doesn't have an incredibly fast JIT implementation available.

im here, still working every day, im not posting much because there isn't anything exciting to post. Until my bro finishes the last touches of Adam's cloth configurations so i can show him walking around in the engine, or until this model is complete, which will take a while since we are planning on making this one the base for more than half of the females (body proportions will change by shape-keys, stuff like breast size will be easy to configure)

if you look at the credits for astebreed you get 6 people who contributed directly to the game play and technical aspects of the game

I didn't know that he did that, now that actually makes a lot of sense.

I did look. Artists need to be included too, maybe not the music composer but to say a game is made independently of artists would be dumb. Considering these games are 3D and gameplay features dependent on animations for a good chunk of them. You have to include artists because for those one and two man teams they did most if not all of the art themselves. For this kind of discussion you can't shift the goalposts by talking only about gameplay designers and programmers.

I've heard tetradev was coming back. Is this true?

no

I hope he does

I wish he would, and that he'd learn something from his spergout, and actually make his fucking game instead of getting pissed over another tiddy-action game coming out with a similar character design.
His game looked decent, and I was looking forward to it. Getting pissed over another white-haired animu chick is stupid. If they had stolen code or a unique gameplay style I'd understand. It almost seems like he was looking for an excuse to drop it, as either just trucking along anyway or changing the character's design would have been the rational choices.

I'd like to reiterate that cascading ifs for what you're doing isn't really that bad of a solution. It's something that's only parsed on startup, so performance impact is incredibly minimal. If you want a stupidly-simple configuration file format the way you had, and you absolutely don't need more power, it works. Either way, you need to take a string in the file and match it up to the functionality, so you need a map of some sort (by "map", I mean the concept of a string-to-something lookup, not necessarily std::map). You can do a cascading conditional, or switch to a language-level map that makes it more clear and faster, or you can switch the config format to another config format that does that lookup for you (if you do lua, you'd either load a function in and use that in the specific slot, or you'd set the global variable and have to do the same lookup anyway; Lua is mostly just future-proofing for when you need more functionality and making it so you can sanely handle strings without having to fuck with it later).

The main reason you'd want to replace strings with enums is for simple logic and clarity. Strings don't have inherent meaning, where enums do. It's the same reason for any parsing why you'd tokenize the data rather than handling it through as strings (See en.wikipedia.org/wiki/Lexical_analysis if you want more details). Don't be mistaken with what you're doing. Even if it's a dumb config file format, what you are doing is parsing it, so the code that does this is, in fact, a parser. Your old setup was skipping tokenization and using the strings as tokens. The solution I dropped with the map is a simple form of tokenization. When you have proper tokenization, it gives you the ability to reason about your code more easily and faster. There's nothing that you really need to do that differently, but knowing that what you're doing is a wide field with tons of existing work can make your life a little easier, because you know what to look for when you need help, and you know practical applications for the things that you are doing.

Thank you, it's really flattering that you want to help out with the project.

My codebro and I aren't currently looking for help but please feel free to hit me up either on [email protected] or @shodanon on twitter. Hopefully you'll still be willing to help should we get back to you.
We'd most likely need help with modelling, as code (C#) and maps (generated) are taken care of.

Most anons want to one-man-army their games but I'm sure even in this thread there are anons who'd appreciate your help. I think GDIdoujins even hires people, haven't seen him around for a while though.
Also, the /agdg/ disagreement is a quiet and slow place but anons that reside there are up for collaborations, should you be interested. disagreement.gg/7Fvzub The best time to get some work done on vidya together is around Ludum Dare jams.

If he sperged out once he will sperg out again. I think if we keep talking about him eventually he will snap and and come here to tell us off and that will be our best opportunity to make sure he stays this time. He might still be working on the game in silence for all we know.

You're right. It's not like the famous image of Cuckdev's code where it's cascading else/if statements that are checked every frame and every NPC has some variant of it running in their script(of which there are like 50 in his meme game)

hey /agdg/
I'm thinking of reviving a very old project of mine
thinking of pic related, in a ecchi slave girl training scenario
one of the reasons I lost interest in it, was my failure to find a an interesting core gameplay mechanic
like in both pic related, that is to train slave girls in a sensible way, while still being fun and able to retain arouseness

for example, in case you don't know how these two games work:
CRV: you recruit a girl with custom attributes in school, train her ecchi skills until increasing her stats, through a series of repetitive skill training you can do daily,
the training itself is a VN style scene, but since its so repetitive you will mostly skip all of it, still there are a lot of redundant "skills" you need to level up while somewhat managing her stats
you make money by having her do prostitution tasks (no scenes), selling the girl (nearly pointless), you expend money by upgrading the school and recruiting new girls
the juice of the game however, is unlocking special scenes in which she needs special attributes/skills to complete, which have a saucy scene and big money prizes

JoNT: you buy a slave and have to train her to the best of your abilities while managing her welfare and mental status
you need to train her the most you can and sell to the highest bidder while managing to not go bankrupt,
using money to buy new stuff to help out or augment your slave, looking for the best buyers is also essential,
the base objective of the game is selling to everyone from neighborhood to neighborhood until you gain approval of the whole city
however the ecchi content is pretty much just random scenes while training and a somewhat broken sex mini-game,
all of that using random stolen lewd pics which really don't suffice

so overall, neither games provide a suitable combination of gameplay depth and fap worthness, which is the main bane of hentai games, either being to grind on gameplay or plain old jerkfest scene hunt
you guys have any suggestions?

>Making game in Game Maker 1.4
What.

I had a similar issue when the game reset it's display mode when going between rooms (IIRC interpolation didn't work otherwise) while the window was inactive. I would have to tab into the window (but was unable to see what I was tabbing into since the game was still in fullscreen), and couldn't quit the game.
Had to set room transtions to only happen when the game was in focus.

there's afew ways you can go about it, the first building up a brothel and engaging in a sort of animal crossing like game, or having a business that is of the mafia where you have to extort, threaten or blackmail your ahead of the other families, finally is some kinda battle arena game where you also train girls to fight in giant mechs

This looks pretty sweet, I love the motion blur with the camera zoom. I know it's a design choice that people seem to always make a fuss about, but I guess I'm a sucker for it.
CoC? Corruption of Champions?

yeah corruption of champions, also the motion blur can be turned on or off i just forgot to record the main menu which has a full set of options

...

I think I figured it out. It seems it was being caused by ShadowPlay malfunctioning, which was a lingering effect from trying to get it to record Yuri's Revenge earlier, which causes it to freak out for whatever reason. I don't know how drawing technology has changed since 2001, but ShadowPlay doesn't like it.

I'm not sure you got what I was trying to say, but neither of these really sound anything interesting, much less arousing

for example, what I thought was, keeping the slave training aspect of JoNT, simplify some redundant attributes
have the sexual content be in form of scenes, on jobs she can take around town, depending on her attributes,
while having content progression, like for example she could simply work at a pub if good at service, but if high lewd skill the scene would unlock more content
with 2d in mind, most of this of course would depend on good writing for more arousal

you're basically just making a akabur game then, just have decent art work and you'll get enough people who'll enjoy it

Is this what you want the character sheet to look like?

Original gate+door sprites are on top. Mine are below. Guy thought they were too similar, so I made a squared red door. That looks pretty distinctly different, yeah?

I'll still have to redo the gates, though.

sorry to say but you're just being dumb there,
akabur's games have the mechanics for a brain dead user,
its literally just grind clickfest to pass through whatever it needs to do
and even then it usually just has handful of scenes and half baked dialogue

that is pretty interesting indeed, a good concept, and having a game of that genre but in a 3d world would make a ton of difference

nice pic, is that your job user?
scored autism/10

It's from /tg/

I made parts of it, but it was a group effort as said.

/trainer general/?
I heard that place is kind of a shithole
anyway, if you're in the genre might give your input on

well what else are you going to be adding?

/tg/ = traditional games

How the fuck am i supposed to release anything without my game getting drowned in that shit? Is this how pixeldevs feel like? Do i have no choice other than going third person now? Fuck.

Make a good one. That's the only course of action unless you want to drop your project. I'm in exactly the same position.
To a certain degree the problem is a matter of perspective.

there's been some great first person horror, Monstrum being one

Now if I made my own extension method that returned a random point inside a circle, it should logically return a value between -1..+1 on both the X and Y axis and never on the circle edge itself, right?
Basically I should follow their convention?

you're worrying about ~0.000000000000000000000000000000000000000000001401298
don't worry about that

Also I made some particles again

Y-you too

Most HG devs these days are faggots and fuck their games up with shit nobody wants, like Fenoxo made TiTs but its got no fucken ayy lmao's in it just fucken Japanese memes, I"m so goddamn sick of these fucked cunts putting to no end these communist fetishes in my H vidya

Fucken
Vore, Scat, Netorare and Netorase, rape of mc, unavoidable rape, black people in general

all of these "Fetishes" are for bad taste goys who enjoy black man endlessly raping their waifus until they're mind fucked, i subscribe to the fetishes for true men Pregnancy, hand holding and Hitler i say REEEE to these cucks

also i got the date and time working like mount and blade so it displays stuff like morning > noon > afternoon instead of the actual times so you dont see the minute character spazzing out because of the time scale

Anyone want keys for Metasequoia 4.6.5? Don't worry about ethics, the creator's a shitheel who stole code from plugin writers and slapped a price tag on a stability patch.

Well, the biggest problem will be to make players aware of that on first glance. I just think they'll assume it's another garbage fp horror game or skip it prematurely due to the oversaturation of the market. I don't fear people hating it, i fear that people don't even get to play it. You're right though, normally there should be no problem if my game is better and starting over again halfway would be a waste now too. All i can do is finish it and pray.

Wew. That game is actually pretty similar to what i wanted to do at first, except in my case you would have to escape a nightmare by finding a way to wake up. However i've decided to make a story based game later on.

At least, that's what the GameMaker forums say. Do other engines experience things like this, or is this the kind of thing people are talking about when they shit on GameMaker?

As far as I know, even OpenGL does this, because going from fullscreen to windowed is a graphics context change.

Among other things, this is one of the reasons why windowed fullscreen is basically the best

If nothing else, I'm sure someone here will play it on a Demo Day. Just do what you want to do, and make it the best you can.

Doesn't windowed fullscreen require significantly more power? Not that it would matter for a game as graphically simple as mine, but I thought it was supposed to be avoided as a matter of principle or something like that.

Fullscreen requires less juice because it dedicates the graphics card to the application via an exclusive mode. Switching to/from this context (eg alt tabbing to look at a guide) kills the graphics surfaces, and need to be re-created, and can be a bit of a lengthy operation (you might have to re-copy the texture memory back onto the graphics device itself?), and prone to crashing if you're not careful

Windowed fullscreen is much faster and flexible for switching around, and most people prefer it, from what I've gathered

discuss

The joke about lisp could have been better than just "lol, semicolons".

...

...

I'm a fucking retard.
I meant parenthesis.

Worked on the hair today. Found a guide that suggest making individual chunks from a large base via extract, turned out pretty well I think.

Don't you feel it makes it look like his skull has a weird shape?

I guess, but I didn't design it. I suppose I could change it just a tad near the skull.

What would the Love2D equivalent be, I wonder?
I'm not exactly a comedian but i'd make it something like

That's the last subweapon, all that's left is to do some busywork nonsense.

I've heard of bread knives but this is ridiculous!

shit fetish, next thing you gonna be catering to head patters, filthy degenerate
but you seem to be building a solid base for a game, so I'm interested

for one, there is detailed slave management, if you ever played JoNT, that would be directly based on that system
so you have slave obedience, trust, fear, despair, so you have to kind of stockholm syndrome her before she loves you

second, actual ecchi content, so when she goes somewhere to be gangraped, you get to actually see that scene
akabur had the biggest cockblock, where the princess went to something like that but you only get to see the aftermath (aka lazy content)

Is it possible at all on Blender to bake high-poly models to low-poly ones without having to separate all those pieces and moving them around, in order to prevent "bleeding"?

that neck though

dont make it html and put on steam

Here is the short version of my plan.
1. What struck me when I visited ModDB/IndieDB regularly was that most people create their project profiles there far too early. Nothing makes me go "Nope" faster than seeing a new game's profile, when all it has on it are ideas, hopes, dreams and maybe some concept art. So the first thing to do is shut the fuck up and do some actual work before you announce your project. In the case of IndieDB (which I recommend all of you to use as development blog) adding your project makes your game appear not only on the front page of the site, but also in the inbox of everyone who subscribed to the engine you use in the form of a notification. If someone who is interested in horror games clicks on it and is greeted by a profile page that had some effort put into it, they're likely to subscribe and get further notifications of your game's development specifically. IndieDB has a really good format for doing a development blog, where you go into the nitty-gritty of what systems you implement and how you do it, as opposed to your average announcement on social media.
2. Speaking of social media, you should try to reach people in as many places as possible. There is nothing wrong with posting about your game on tumblr and reddit. That doesn't mean you should pander to them, but you shouldn't pander to /agdg/ and Holla Forums either. Don't just post in general gamedev places, but also in more specific ones. I create my art using Blender, so I'm going to post it on blenderartists and maybe some other places. I'm using Unreal, so I'll make myself a presence on the official Unreal forums and maybe some bigger, unofficial ones. It's a horror game, so I'm going to find some horror game communities. Say your game is similar to another horror game X that's already been released. You can drop by in some fan forums and make a post along the lines of
If this rubs you the wrong way, you can always ask a moderator if it's fine first.
Releasing your game on Linux, too? Post about it in Linux gaming groups. Don't restrict yourself to the biggest communities. If you're going to post an announcement on Facebook anyway, copy and pasting it to some communities on Google Plus isn't going to take much time.
3. Find excuses to make high quality posts on gamedev forums. I'm going to open source some reusable parts of my game and post them on the Unreal forums.
Write tutorials in which you reference your game. Some communities, like the Unity and Unreal ones, have Q&A sections. Answer a question a day. They have a community wiki? Write tutorials about the stuff you're doing and use your game as an example. You can go the extra mile and have a YouTube channel with gamedev tutorials. However, that's more work, because, just like your profile page on IndieDB, the threads on forums, etc., you should hold yourself to a reasonable standard. Nobody wants to see a tutorial that's so low resolution they can barely see the content, has a voice over by a guy who barely speaks English, lots of background noise and other problems. If you have those problems, find someone to help you with that.

If I may ask, what are you working on anyway, user?

homie, those are parenthesis, not semicolons;

On another note, I'm using Godot on my project
I fell for that abstraction layer trap, where instead of manually making that each part of the game, for example in godot being a scene
I had every single thing relating to the world as plain json files, for example navigation, you have a background image, row of menu files etc
and the scene is generated at runtime based on what is in the text, simple enough right?
the problem begins with the slight variations you may have with each scene, for example trigger checks, or having a different menu or popup
then things get way more complicated specially considering if everything needs to go through a same pipeline an whatnot
for that reason I'm sort of re-considering building the whole game up again manually, or at least semi-manual with a more sane abstraction
what do you guys think about it?

one more thing, I've seen people recommend gitgud.io
are they user safe in some way? is sapphire trusted?

He's been posting engine progress since day 1, we gave him the name speebot, and he's been an active member throughout all of it. Speebot is an /agdg/ game, and shilling your game on social media sites doesn't make you a member of it.

They aren't politically curating the content unlike GitHub for example. The #GG current happenings are also hosted there.

Yes, do it. One of Godots strengths is it's UI system. Generating scenes from a JSON file seems mad.

that sounds interesting, I'd play that so long as it's not "it's a dream world anything can happen xee dee"

It's made by anons who were pissed off when the Gamergate github was banned.

I'm not a fan of gamergate, but that sure makes me a lot more trusting of their services

it wasn't so bad, you had a few things like background, dialog and a menu,
for example the menu was just an multi-array, and their identifier is the same name to calling other scenes
it was sensible somewhat, but things would get pretty the more eventualities you need, like a random encounter

Added a simple clipping region as a camera so I can filter out which tiles to draw, and did some more stuff with "particles". These are actually animated objects that are moving towards the mouse, not true particles (and incidentally kill the processor when I have 2500 of them interpolating shit every frame)

Anyone familiar with Godot wanna help here?

My character is supposed to be aligned with the grid in a Tilemap. Right now it seems he is offset which is affecting movements into walls (collision right now is just a check if the next tile is a wall or has anything blocking on it)

Is this because of position offests on my player object or what

It's probably drawing from the center of the sprite, aka the origin. Try offsetting it by half its width/height

that seemed to fix it, thanks!

however, this seems like a hacky thing to do, to just offset the position of the sprite on the object in order to get the position to line up on the grid. Is there a better way to do it?

I'm trying to set up something in the Unity Scene editor that lets people click and drag to draw, is that possible? I really don't want to have to use the game view to do this. It's important.

Fullscreen can be done without recreating the context from what I have read, however OpenGL is not aware of any of this, since OpenGL context creation and fullscreen mode are not actually a part of OpenGL.

What you can do is change the display mode with ChangeDisplaySettings and then set the window to a borderless style using SetWindowLongPtr, this should work fine but I have not written any programs that test it. So, in theory GameMaker could be programmed to seamlessly switch to fullscreen mode.

On X windows how this shit works is totally missing from the man pages, you can find some really terrible code samples if you look hard enough but I think that it is almost lost knowledge at this point, programming for X is like a lost art from the 80s-90s at this point.

Windowed fullscreen has the advantage that the window is actually at the same resolution as the rest of the desktop, and thus there is no change in display settings. This makes going in and out of the window seamless. However a fullscreen program will actually change the resolution of your monitor and this takes a while.

This is not how fullscreen mode works. The bottleneck is that the monitor's display mode has to be changed, the GPU's context or texture memory is not actually affected.

You could account for the offset in some kind of GameObject class or something? I ended up doing something similar with SFML Sprites.

pretty much what said
check if the sprite is based centered or if you have the option for the sprite to be based at corner
basically the trick here is accounting for the origin + size
it just seems weird for you because you're thinking of the object as just a dot

What I ended up doing was alter my function that sets position based on the grid to include this offset, rather than offset the sprite on the object.

The reason I did this is offsetting the sprite node gave me some graphic glitches when it came to 2d lighting occlusion.

every thread now

what it looks like now for reference

How viable is a tcg that uses a blockchain to implement a system where you can ACTUALLY TRADE CARDS?
I wish I was kidding and wasnt actually considering this.

Watch vid related. Here are some things you're going to have to figure out, in particular how you are going to incentivise miners, who's going to store the block-chain, how you are going keep certain cards at certain rarities, and how you are going to introduce new cards into the deck both "new" as in a copy of a card that already exists and "new" as in a different card.

So what, are you using an alpha mask on the light or something?

yeah basically. in Godot it's just a Light2D node with a radial texture. The wall tiles have a light occlusion node attached in order to cast shadows.

Also I just darkened everything else to make the light more realistic.

oh fug :DDD

Penguin = Linux
Seal = Proprietary software

top fucking kek

Here's some tiles I whipped up in 10m because someone else wanted them I guess

Why wasn't this the first thing that was worked on when making the rendering engine? It's basically the most important part, what separates an actually useful rendering engine from a simple model loader. I actually thought that this engine was doing some pretty cool stuff but now I see that it's really unusable for anything beyond a fighting game, or rocket league clone, right now…

Is it me, or do tool devs often deliberately use terms that are uncommon to describe very simple concepts? For instance, why call it an armature instead of a skeleton? It doesn't matter if there's a more direct comparison to be made to the sculpting tool, because almost nobody knows what that is, while everybody knows what a skeleton is and what purpose it serves.

Top-down games don't need much occlusion culling either unless your level has a building with multiple floors. If the other team members are dead-set on a camera with more freedom I'll have to look into Urho3D or something similar.

A lot of the terms that engine devs / tool devs use are passed down from the academic papers that game engines and tools are based off of. So, you can blame computer graphics academia for some of that. But who calls it armature? That is new to me. Other than that, "occlusion culling" is a simple enough concept, but there aren't any terms that we can use to describe it that are shorter and easier to say. If you don't know what it actually is, when a thing blocks another thing from our point of view, this is called occlusion. So, we want to cull all occluded polygons as they don't contribute to our scene. That is what occlusion culling does.

I know, I am just saying

Sorry, I forgot to finish my post, I am just trying to talk about the kinds of games that need occlusion culling to get any kind of good performance, like an FPS game. Although on the site, you can see that the vast majority of the games made with the engine don't need it to achieve acceptable performance on modern hardware.

godotengine.org/showcase/

The fuck are you on about? That's a brilliant fucking idea dude. TM it.

Ever read an academic paper? The audience they write for are the kinds of people who regularly read academic papers, not your average Jo. Academics want short phrases and words to describe a lot of stuff very accurately. Their lingo isn't complex as much as it's very dense with information and precise. Rather than saying "not rendering polygons and elements that are not visible to the camera" an academic just says "occlusion culling", "hidden surface removal", or "visible surface determination" and all the other academics reading their paper know exactly what they mean.
The industry is informed by these academic papers so they use their lingo.

People do call the armature a skeleton. In UE4, for example, the file that holds the bones is labeled (Skeleton).

Blender.

To be honest i havent thought much about marketing so far other than reaching out to social networks, lets players and horror forums/blogs. Your post helped me out a lot in that regard.

The game i am working on is pretty basic. It's a first person survival horror game, similar to current walking simulators but there will also be weaponry and very limited ammo so that the player can fight back if he can't run, making him feel pressured with each enemy encounter. I guess a mixture of SH1 and RE7 comes the closest to what i'm doing. I don't want to spoiler the entire story but basically the main character has a incurable illness and is the first person in history to be put into cryostasis as a voluntary subject in a laboratory for experimental technology, keeping him alive until they found a cure. However when he wakes up the entire laboratory is empty and shit gets more nightmarish weird and surreal as you progress trying to get out of the place. There will be actual pseudoscientific reason behind shit happening instead of the lazy "it's just a dream" or "it's just all in his head" writers normally opt out for.

I like rings that power up the player.

assetstore.unity3d.com/en/#!/content/10374 (free)
The only thing I don't know is if that support angular2+, although it will most likely support after webpack builds it

After having to work with html5, I can't make UIs in anything else anymore

I will test it out this weekend and report or hopefully, someone here already uses/used this and can give some feedback instead

...

...

Alright then, new bread.

sounds like overblood, skip to 1:13