Amateur Gamedev General ~ /agdg/ + /vm/

2018 edition

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

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

QUARTERLY DEMO DAY SCHEDULED FOR FEBRUARY 2ND
Polite reminder that the wiki exists, you are encouraged to contribute to it if you can (even if it's just your game page)

Other urls found in this thread:

store.steampowered.com/app/731320/Speebot/
wikidpad.sourceforge.net/
docs.microsoft.com/en-us/dotnet/standard/design-guidelines/choosing-between-class-and-struct
youtube.com/watch?v=AXUvnk7Jws4
docs.unity3d.com/2017.1/Documentation/ScriptReference/Playables.PlayableGraph.html
benchmarksgame.alioth.debian.org/u64q/which-programs-are-fastest.html
github.com/godotengine/godot/issues/15321
nier2.com/blog/2014/01/29/possible-connections-between-yorha-dodnier/
gamedev.stackexchange.com/questions/72978/shader-that-cuts-hole-through-all-geometry
en.wikipedia.org/wiki/Event_(computing)
love2d.org/wiki/love.keyboard
love2d.org/wiki/love.mouse
stackoverflow.com/questions/29085942/stdbad-alloc-when-inserting-into-stdunordered-map
8agdg.wikidot.com/love2d
8agdg.wikidot.com/godot
twitter.com/NSFWRedditImage

benis :DDDDDDD

It's a weird feel. Got so used to knowing exactly what needed to be done every day. It's fun, though.

Is it better to build a game on a high end pc or a pc with the specs that you expect most of your players to have?

Developing a game takes more processing than merely running it, so developing it on a rig meant for playing vidya might be slow at times. What I do is develop it on a mid-high-range PC and then test it on a potato every couple of months.

I hope something else means full time game dev

Don't build the game on the targeted specs, because you will run across not having enough resources to run test builds. Unoptimized test builds WILL eat up more resources. So you'll either end up with test builds running like absolute dogshit on your targeted specs. Or you'll accidentally make your game's graphics and features not utilize the full potential of your target hardware because you ended up optimizing for a lower spec range to get your test builds to run smoothly.
There's a reason why many console devkits have more RAM and sometimes are even slightly overclocked and this is it. Build on a high end PC and do occasional tests on the target hardware to see how much optimizing you'll have to do in the future.

Not quite. I just finished designing an important location for my game, and now gonna focus on some side-stuff that you can do there.
Full time starts at February and, if I can help it, never ends.

Repostan because end of other thread nobody is going to see.
Optimized the fuck out of my server code, and started working on more visible stuff again
1/2

2/2

How can you ever expect to make it, user?

I have "target hardware" which is a computer with a pentium M and a radeon 9000. I run my program on it every few months, it's hard to use because it has no networking unfortunately

Is there any particular reason it doesn't have networking? Mobile device with a broken adapter or something, or you just haven't gotten around to digging up a NIC for it?

No, as the others already pointed out.
>you need to be able to test max settings before optimizing
I sourced a rig with used components off of ebay. It's difficult to determine the lowest system requirements before you have finished your game, but I used another approach rather than trying to determine the exact requirements.
Since I use UE4, I basically checked the min. system requirements for various games (not restricted to UE4 indie games) that are modern. My thought process was that if some game (that is more graphically, computationally intense than mine) runs on X hardware, then it would be embarrassing if mine didn't.
In the end I settled for:
I will optimize my game to perform acceptable on this hardware. To me this means low settings, 720p, >=30 fps at all times or better. If it turns out my game can run on this hardware on medium settings or higher resolutions, then great. All I'm after is ensuring that this toaster is the upper limit for minimum system requirements.

Makes me wish I kept my old GTX 480 instead of selling it to a friend. It's the perfect example of an old but capable card that'd be a great minimum spec card.

Damn miners are ruining prices

Is it realistic to expect success with Dota 2's original business model? (All gameplay stuff free and direct sale of cosmetics only) or do Indie games not have the player base to make that work?

Hey, I want to do a raycaster that can run on super old toasters what do I need to know ? Do I need anything other than the oldest version of OpenGL and know how to use it to draw a line ?

You probably don't have the playerbase to make that work. It works for big multiplayer games because they keep releasing content which keeps people playing. There's extremely few multiplayer indie games that still have a large playerbase. Especially competitive ones die out really quick because they simply can't sustain themselves.
And if you have a singleplayer game it doesn't really make sense to sell cosmetics. It's not gonna reflect very well on your from a PR standpoint because people would rather have it be part of the game itself.

You shouldn't worry about monetization in an indie game anyway. Just like make game and sell it at an honest price.

I didn't know Bjarne posted on full.

Does anyone know how to make Godot's GUI controls use the input map? I have an ItemList that I want to control using the movement keys and my accept key, but it defaults to the arrow keys and enter.

Probably should know x86, and how to use MMX/SSE extensions. Also OpenGL is probably not going to help you very much unless you want to program your shaders in ARB assembly, even then it's still only going to work on graphics cards post-2003. You probably should just do software rendering with an optional shader implementation. Drawing a line in OpenGL 1.0 is a lot different than doing raycasting.

I installed some networking driver but it didn't work, so, I might try another driver if I can be bothered. It has wi-fi, but it cant connect to a single network, so maybe its adapter is too weak. And ethernet doesn't seem to work either. So, I guess its just the drivers. I was able to get the graphics drivers to work so at least I have OpenGL 1.3 on it.

You don't need a large playerbase to be successful as long as you know what you're doing though. Look at games like Vermintide or Depth. Those games usually only have a few thousand players playing at any given time. Yet Vermintide is getting a sequel and Depth just celebrated it's 3 year anniversary not too long ago.
Any model can work as long as you get two things right:
1. It's fun enough on it's own to be played for more than a month. Easier than most people think, just make sure the core gameplay loop is enjoyable and there's nothing that's absolutely bullshit. Example: The Mean Greens died because everyone (regardless of gamemode) spawned with a rocket launcher that'd one-hit kill (The explosion didn't even have any damage dropoff based on distance from epicenter). By the time they nerfed it (Which is the wrong decision, they should have made it a pickup power weapon instead) people were already so frustrated by it that they stopped playing and now it's a dead game.
2. Whatever your monetization model is, people should be able to get SOMETHING for free/by playing the game every single day. Going back to Depth, it has really simple daily quests (Get 25 kills with X weapon, Win 3 games as X shark, etc) that can be done in just an hour or two that gives a free lootbox. It encourages daily play and in a way that isn't asking a whole lot of you. I'm vehemently against monetization outside of the base pricetag and expansion packs (We're talking Age of Mythology: Titans Expansion sized content packs), but if there was a gun to my head, that's probably the model I'd go with.

You'd be surprised how much toaster shit is 64-bit and OpenGL 4.1 compliant. Pretty sure my old laptop from 2010 that has a GT310m, 4GB ram and a crappy dual-core i5m is both those things. And it's a toaster that struggles to maintain 60fps in Source games.

I don't disagree. I simply think it makes more sense to foster a relationship with those few players than trying to squeeze them dry with monetization, especially as an indie dev. Doesn't matter how unintrusive those microtransactions might be, it makes more sense long term to not have them.
That means the actual sales are much higher. Again, not something you will feasibly achieve as a run of the mill indie developer, or at least not something you should rely on. Vermintide was a relatively big production and had the Warhammer name behind it. Meanwhile look at something like Secret Ponchos or Lethal League, really well made competitive games that are now essentially unplayable because there's simply no playerbase left after the initial release period, save for a few diehard fans. Adding monetization when you have those numbers isn't going to make you a lot more money and is simply going to hurt your relationship with your customers.
That especially goes for singleplayer games. You can get away with it if you make mobile shit I suppose, but I seriously hope none of you guys do this.

For Wi-Fi, it might be using some older protocol. Might want to see if you can force your router to use a, b or that legacy set of channels with just WEP or no encryption, to see if can get any sort of connection. Many older adapters don't support anything beyond WEP. Incompatible drivers could always be the issue as well.
Interfacing with older hardware can become quite a fiddly thing to get running. Probably not worth sweating over too much. Given enough time, you'll most likely run in to the day where you spend ten minutes on it, get frustrated enough to hammer away at the problem for however long it takes until you get it working. Or cut your losses and just streamline whatever method you're using to get data on there as much as possible.

You lucky motherfucker.

Hopefully I can find a job that won't suck my motivation so I will be able to focus on devving more. Part of that is also other things being on my mind, but it's a start

Also, is the Speebot dev really from here? Is it any good?

Yes, Speebot Dev regularly posted to the 8/agdg/ thread during development.
The game is fun.

Speebot is simple but fun.

I guess I'll have to check it out. I just jump in these threads now and then, to be honest, but I will help some /agdg/bros out. I have fond memories looking forward to tetradev's game.

user, you have no idea how shit that graphics card is. I'm not happy with it at all because it still uses the old AMD driver under Linux which sucks ass. The problem with the new driver is that it isn't supported on old hardware and even AMDs lowest-of-low-end graphics cards from the past two gens (that use the new driver) are a lot faster than this piece of shit. You can check userbenchmark.com (which is quite useful for this sort of thing). The RX 550 has almost three and the RX 460 almost four times the performance of the HD 6670; and that's under Windows where the driver situation isn't a travesty.
As for 64 bit support: I checked the Steam HW survey. The 32 bit installations of all operating systems make up for just over 2%; and that's installations, not hardware. I just checked google: There hasn't been a 32-bit x86 CPU release in over a decade now. AMD introduced the x86_64 architecture back in 2003. If you actually have a pre-2003 CPU then I don't think you get to complain about not being able to run 3d games from Current Year+3.

Wait. Do you have fond memories of looking forward to his game or do you have fond memories and are lucking forward to his game? I wouldn't hold my breath.

I thought he stopped working on it now?

Yeah, he isn't working on it anymore. At least he wasn't when we last talked months ago.

Man, if only he promised that fucking linux port.

Why live?

You boys better get optimizing, stat.

A Windows update killed my linux dual-boot set up, and virtual box does not support OpenGL 3 on linux targets so I can't reliably use it either. If I get a separate laptop in the future I'll use it as a dedicated Linux machine to port my games, including Speebot.
Until then I'm afraid you'll need to use Wine to play Speebot on Linux.

How has it "killed" it? Windows updates have the tendency to overwrite the MBR, but that doesn't kill the installation and it's just a minor annoyance.

We don't have enough computers for that

It fucked Grub and something else because my usual restoration tricks didn't work. Maybe I will give it another shot.

There was a Windows 10 update a while back that outright deleted Linux partitions. (((Completely accidental))), nothing to see here.

Last time we heard of him, he got sick or something. He's been away for a while, but I am fairly certain he'll eventually be back and deliver that linux build.

I sincerely hope none of you niggers are using windows 10. I want to believe none of you guys are that retarded.

Lucky me, I've got AMD.

I-I was forced.

I am still on windows 7

How were you forced?

I use comfy Windows 7 with the classic theme.
but i'm sweating bullets right now about how my dinky little core 2 duo is gonna take the (((performance hit))). I fucking swear I'm gonna get a Talos workstation.

Iv'e been gone for a couple years.

Have you guys actually made any games?

I've made two barely-mediocre games and a bunch of mods. Working on a third game right now.
More importantly, though, Speebot is out and cool.
store.steampowered.com/app/731320/Speebot/

I didn't like to upgrade to pic related, but I need my PC to work first and foremost. I dual boot with Linux, so I reduced my Windows usage to gamedev and Windows only games.

Belated happy new year aggydaggy. Back from the dead with some visible progress finally. I ditched my dumb wall texture mapping shader for meshes and finished moving to portal based drawing. I decided to focus making a small scope fast instagib multiplayer game with it first and bust my balls to get it done this spring, then I can move back onto the FPSARPG idea with better foundation in place.

I'm just using a flash drive to copy my game over to it, which works fine for now, since I haven't got a need to test online multiplayer yet.


I don't consider it to be a toaster if it has OpenGL 4. Just because the Source engine isn't fast doesn't make it a slow computer- it means that the Source engine isn't optimized enough.


x86 was only phased out 10 years ago- so that's toaster hardware. x64 is too new usually. Anyway I really can't agree with you about such a card being shit, I mean you might not run that many new games because of bad optimization but I bet it can run all older games fine. Anyway that is your idea of what a "slow computer" is, which is a lot faster than mine. In the end it's just semantics of what a slow computer really is, I just think that games should target computers much older than OpenGL 4.1.


My nigger. I have one on my laptop, so I can develop stuff on the go.


Very nice looking, I know you can do it!

Time to bring back booter games. Can't slow down the OS if there's no OS.


Chunky pixels. There's something very appealing about those low-res fonts, and many older games have done wonderful things with them.


Classic theme best theme.
Haven't touched Windows at home for years, got forced into Win10 at work, though. I don't understand the appeal of the Metro look. It doesn't read well, there's little delineation between elements is usually almost non-existent, the amount of options you have to make things suck less are limited, and just about everything to do with the menus and interface has somehow turned been boiled down to close to nothing and still made more convoluted to use than ever before. taking away as much granularity they could get away with and then some, replacing vast arrays of tweakables with a toggle switch. Suck more/suck less, suck/suck in a different way. No off. And somehow, even after whittling away these options to a small sliver of their former scope, Microsoft has still managed to make finding what you want as irritating as possible.

Aesthetics over functionality taken to a stupid extreme, and without outside tools you can't even mildly deviate from Redmond's prescribed solution. While this permeates through the more technical customization, like the dumbed down networking, it doesn't stop there. Even something as simple as the damned background. Prescribed colors. You want a solid black background? Too bad, you have to make a fucking black image and set that as your background, we only have these colors, and they carry through to the rest of your interface. Fancy distracting effects everywhere. I had to go into the disability options to turn some of the gaudy shit off, clearly as only the blind have no use for Microsoft's superior taste in design.

I got mildly pissy at what Windows 7 did to some of the settings, but those were still understandable decisions, as the backend was just as customizable. Windows 10 just crippled everything.

And I don't even have to do all that much outside of certain applications, but every time I have to touch Windows itself to make my workflow a little bit more convenient, it's a goddamned shitshow of trying to override whatever their idea of "sensible defaults" is.
Fuck.

I'm sorry for your loss, I thought there was a work around to get updates on windows 7 with new cpus but apparently that doesn't extend to drivers.

I'm talking about the line going from the top to the bottom of the screen for each ray being casted in order to show an image.

I guess we have to agree to disagree then. To me, the question of whether a piece of hardware deserves to be called toaster-tier is primarily a question of performance, not age. Sure, newer hardware tends to be faster, but you still have the situation where newer low-end hardware is slower than older high-end hardware.
I might end up supporting 32 bit builds, if I find out that I don't use more than 2GB of RAM at any point.

Well, whenever I feel bad about it, I remind myself of why I switched:
And I didn't have a bad CPU before. I had a Xeon E3-1231v3. That's not a bad CPU at all.

I see. The issue with OpenGL 1.x is that it's not programmable, so you have to use extensions to access any kind of progammability, you'd be writing pre-GLSL shaders and would require non-standard extensions. If you want to do polygon rendering however, it's very capable. So software rendering is probably the best way to go.


I'm really just rating it by features. If it has OpenGL 3 or up, I don't consider it a toaster. Just because the featureset is what I am most aware of when programming.

I see. I'm looking at it from the end-user perspective. You're not going to have a great time when using that hardware.

fucking game maker

Things I have done since last posting:
>attempted porting to monogame, decided to stick with gml because progress was too slow
Vid is a supercut of some of the stuff I've messed around with. Pic is the characters as well as a demo of the modular sprites.
I hope to get the color pallet swapping working to get the clothing of the modular characters to be less patchworky.

Damn, looks R O B U S T

I think my idea of slow is well inline with the general public. If your computer can't maintain a constant 60fps on TF2 or CS:S then it's toaster tier.

not when trying to run new games on their higher settings, but old games and well optimized games should be fine.


I guess so. The game I would rate something by is how well it can play Quake 3 Arena.

Model is getting there. Needs more fix on the back near the armpits.

I was curious about how excel worked, since I pirated excel 95 a few months ago. So, I decided to make a graph of my progress since I started the project, based on a few of my backups.

Autism as fuck, good job on your engine

I have a great idea for a game called Uncucked. Its like Contra 3 but with two fit white girls. The game involves them shooting the fuck out of nu-males and fat lesbians while on a quest to get some McRibs.

>still have to rework all controller support

butt is way too low imo

I want to engage in coital activity with that physically attractive older woman

For the new model right? I raised it quiet a bit already. Could it be just the crack that is too low but the glutes are in the right place?

finally got an internship as an assistant for a introductory C# course. pays around a quarter of minimum wage, but it's still a job
meanwhile i'm about to fail my algebra exam for the 5th time and drop out of uni because of it
university was a mistake, but at least now i won't be a complete disappointment to my family

This isn't your blog user.

After 2 days of autistic research i found a way to decently sway the weapon left and right and so on. To the user who recommanded it: it is done.

Looks great
It's nice to see someone doing progress so fast

Thanks man, may your trips empowers our motivation to carry on!

have I found a soulslike experience? :^)

make it rotate at the base of the weapon instead of the middle

I plan to made them blur their backgrounds and make the player shoot them, like in Metroid to open them. I still have no clue how the fuck am i going to do this.


I'll try that in an hour or so.

I hope vidya doesn't die anytime soon, I wanna work in the industry and have fun making great games. Hopefully ones that dont require SJW bs.

Are you kidding? The best thing that could happen would be for it to die. Then innovation can rise from the ashes and vidya will be made great again. Why yearn to work in a shitty industry where you have to constantly be looking over your shoulder for SJW shit when instead you could rebuild the industry from scratch?

rebuilding an industry is not easy faglord, it is literally Refactoring: business edition

This guy is still working on his graphics engine, it's looking really good now

Use these material parameters:
Flags:
Albedo:
Roughness:
Refraction:

I went to school for programming in 2008. Graduated in 2012, paid off my loans by 2017. All I have to look forward to is pajeets in (((Toronto))).

Though if you go to school now, maybe the industry will mildly uncuck itself by the time you finish

Right so I don't know if anyone remembers me, I was working on this a few months back, then stopped working on it because I had fucked up a bunch of things and because of a few problems with webgl.

The game is basically an ero-RPG inspired by stuff like Violated Heroine but with more focus on player agency and branching dialogue.

Before the game view used to be on one side with all of the text being shown on the other side, which I have to admit looked pretty jarring and not fun at all. A bunch of people told me they didn't like it that way and honestly neither did I.

So I tried to rework the whole dialogue system to a more RPG-maker / Fallout kind of thing and now it works much better.

I also reworked everything from dialogue, the way quests are handled, the way NPCs and maps are made so that anyone, even with minimal knowledge of programming would be able to add branching dialogue/maps/events to the game by just adding objects to it.

I'm not going to rework some of the sex events and pregnancy stuff

Also you can play as a "loli" if you want

now*

Tiles and menus are still all placeholders, I can't really into pixel art so this all I could get for now.

how do i make good textures???

Go on wiki > graphics > texture maker 3

fuck it's meant for

...

I told the guy to find a permanent link… Or it's not allowed to outjew the jew on mega.

Blocking out the fishing mechanics. Casting and reeling now function. No fish yet to catch, though.

how to good butt topology?

Done it. May have abused too much but can surely look better than first try.

Now that's gonna be fun to code, except that i'll have to make my weapon shoot first, and i have a feeling that i'm going to struggle doing that.

here is mine, but there has to be a better way to do it.
onready var bullet_inst_scene = preload("res://media/sprites/particles/bullet_instance.xml")func shoot(): if cooldown_shoot

I'll try a raycast-based approach and see if i can do immediate geometry to create lasers rather than instancing but if it doesn't work the way i intend it i might just do it that way. Honestly wish me luck on that one, i see you may be much more experienced than me in programming i have yet to know how groups work let alone reparenting nodes by script…

Hey there guys,

I've gotten really tired and frustrated with my GDD - it's a long PDF and it is no longer a practical thing to use as a reference. I thought about what could be a good alternative, and I've realized that a Wiki-style resource would work well.

Any ideas what I could use? I kinda want to avoid using Wikia because they're known for deleting wikis randomly.

In this case, I highly recommend a program called Wikipad. It's basically Notepad but styled as a Wiki. You can create and delete pages, and export to html and have automatic page linking and shit, too.

It's pretty lightweight and uses a local SQL database to hold its shit, maybe only a few MB big at most

...

It's called wikidpad, with a D between wiki and pad.
wikidpad.sourceforge.net/
Takes a tiny bit of getting used to, but so do normal wikis to be fair.

raycasting is pretty much the same shit, just get translations instead of transforms and feed it to your raycast node.

A concept is only bloated and unfocused when you come up with shit on the fly instead of planning using a GDD. Anything more complex than Pong needs a GDD.


Thank you. This seems to be what I had in mind.

can you make flowcharts with it or do i have to keep using blender for that?

That's pretty nice, you made all those sprites?
Im trying to make something like that on art.

Only for the process of procreation right?

Tetris didn't need a GDD and I doubt Toady has a GDD for Dwarf Fortress.

Pretty sure that's a no, but I don't see why you would use Blender instead of some Mindmap program for that purpose.

Game would probably be 1000x better if he did

Blank textures mean I don't have to put effort into what a room looks like. Yay!

Im trying to make an stealth artificial intelligence for my game.
Here how its going:

If you think Tetris didn't have a document detailing all the ins and outs of the game, you're nuts.


That's why his game, while fun, is the definition of feature creep. It's a rare situation where an amorphous blob of a game benefits from constantly expanding in all directions.

Minecraft also was like that for a bit. And then it wrecked itself with that approach in Beta 1.8 when they added garbage fluff.

Did you make any video games today?

Sorry for the grammatical mistakes.

I made levels load from a file again, and greatly cleaned stuff up. Now I only have 2 or 3 errant classes that need redoing and I can finally fucking work on behaviors

I gave up.
and started porting everything to Love2D. The only reason I stuck with Game Maker for so long is because it just worked™ but it stopped just working, and GMS2 is the most dysfunctional pile of shite I have ever witnessed. So progress is fucked as I now have to learn a new language, but it should be beneficial in the long run.

can you recommend some?

i see the pixelshit of the future

cool

Any musicfags around here? I want to know what program you use to make your music, I'm using lmms because free. Do you have nay recommendation of classical instruments? I'm just randomly searching around and downloading anything until I find what I want. I also realized that using vst's with lmms is the absolute shit, nearly all the plugins can't load properly, so anytime I want to open the music file I have to remove them from the folder and re-import once the music is open, sf2 player seems to be the way.

Anyway he's my first music, tell me what you guys think of this. It will be the main menu theme, there will be more ambient sounds like birds and such and the music will be looping forever. My inspiration was Garden of God from Chrono Cross, I was thinking that using an existing song as inspiration was bad but I learned that you have to listen to the song so much that you stop enjoying listening to it, then you stop trying to copy it and your brain actually allows you to be creative and derive from it.

I've spent the whole day applying undo and redo to the editor, it's not done yet, but now is a matter of merging every state-system in a whole, which is quite easy, but I'm tired as fuck. I'm thinking about starting another project since this one is almost over, but dunno maybe it's not wise.

Can anyone explain me why would I ever want to use polymorphism (C++)? I know what it does for the most part, but I can't find a good reason to ever use it and the examples I find only complicate the code by splitting it into many classes, making the whole code look like over complicated shit.

Last attention whoring for today, i'll come back when i have something worthy to show.
The blur looks amazing, thanks alot to the user to showed how.


Although i have the same VST plugin as embeded artist, i just can't get around producing music. Also on LMMS. It's a pretty piece though.

what is polymorphism?

My screenshot got blackholed.

its okay until you get a diamond, that's when you need to stop


it ranges from

Good luck with Lua, a lot of people don't like it. Not that fun of a language when you come from a stricter OOP mindset.


Mostly just importing sprites and bashing my head against even more, fresh animation editor bugs. So not much.

Most code I write tends to be flat, and only 1-2 object types deep. You don't want a retarded nested inheritance tree 10 objects deep just because you added one variable.

A better use is with interfaces or such.

For example, you could make a basic GameObject that can be added to a list for updating, etc., and make a more defined type that explains HOW it's updated, eg

OOP really has no place in the core engine. The fact it's used outside of obvious cases like entities (light entity, prop entity, player npc, etc) for things like the fucking engine itself should show most people who force OOP have no clue what the fuck they are doing. Hell. When I took a stab at writing my first engine I fucked myself over pretty hard by doing it OOP style, because of things like cross-cutting concerns which OOP has no way to deal with that wont over complicate the fuck out of everything. OOP also rapes cache, and thus is slow as shit.

I came from C# and it was pretty jarring having to do all the OOP shit manually, but it's not that bad.


Fucking this. People meming about OOP being cancer or OOP being fucking ambrosia both need to shut the fuck up. It's a tool that has its uses in areas where it actually belongs. Trying to applying OOP to anything besides the obvious is fucking cancer.

Yeah, most characters were drawn by hand but some are pre-rendered, the background I generated semi-randomly and the trees were made by a different technique all together, hence the four distinctly different art styles.

I don't know whether to be impressed by the ingenuity, or disgusted at the potential for optimization.

neat

So we agree?

Bored on Wednesday evening? Have some hardware exploits, one specific to Intelshit and the other affecting basically all modern processors under the sun.

It's was a two days work, give me a chance.

Is C# really that slow? RimWorld is coded completely in C# and it works pretty well. It's also my favourite language to program in so far. I've only used Python and Lua besides

It's not slow but it is bloated since it's running in a VM.

It took me a whole day, but I managed to get this working the way I want it by pure chance because something happened with the engine's tools that made absolutely zero logical sense.

Fuck gamedev.

It's not that slow in and of itself, but if you treat any language like a kingdom of nouns, you program will be bogged down simply from the verbosity of trying to express anything.
Latest versions of the .net runtimes will JIT most code down to assembly segments and execute them natively.
If you use structs and don't have a lot of alloc you can minimize the GC hits to the point where they are almost nonexistant.


Oh yeah, there's that fuckhuge standard library that comes with it that. That's both a good and a bad thing.

docs.microsoft.com/en-us/dotnet/standard/design-guidelines/choosing-between-class-and-struct

I didn't say use structs for everything, but you use them where applicable, you get a good speed boost from not heapallocing constantly.

Also forgot about this: structs implementing interfaces in Unity's upcoming job system:
youtube.com/watch?v=AXUvnk7Jws4

And Unity's playables API is an interesting use of structs for EVERYTHING without performance getting fucked.
docs.unity3d.com/2017.1/Documentation/ScriptReference/Playables.PlayableGraph.html

benchmarksgame.alioth.debian.org/u64q/which-programs-are-fastest.html

C# is fundamentally slower than native languages without GC, you have the VM that it runs on which adds some overhead, and the GC which prevents you from using memory as efficiently as you could be. So, you really can't expect to get the same speeds as other, faster languages because of that. Obviously like said you can try and minimize the drawbacks of the language, and maybe that will give you an acceptable amount of speed, if not then it's time to use a faster language.

be sure to document it then

Hello my favorite peeps,

I want to get into OpenGL (with c++ or c) but I don't know where to start. There are so many tutorials I don't know where to start.

Anyone got a link to an actual decent tutorial or maybe an actual updated documentation base?

I tried making a subtile background layer, so you could decorate the black areas between tiles. I don't know if it's too distracting or not.

nehe.gamedev.net is a good place to start on older fixed-function OpenGL although I think that the programming style isn't the best.

I actually am unsure of what a good tutorial is, I wrote my first OpenGL programs with nehe's site, as well as winprog.org/tutorial for win32 api stuff. But, none of those sites tell you how to implement a good timer, which I had to do myself.

Which works absolutely nothing like modern OpenGL so avoid.

This and drawfag threads are pretty much the only stuff I still lurk on imageboards, so I wish you fellas a happy new year.

it depends on what you are trying to do, really. If you don't need the functionality of modern OpenGL, it's a very productive library.

someone take black and white and breed it with populous the beginning where the villager the god makes into a sharman can be controlled by another player

Thank you anons, I'm going to be using wikidpad for years now, this is pretty awesome stuff for planning and taking notes.

Cutscene system is working great.

I'm just trying to get into 3d rendering. Start with something simple maybe work towards a minecraft clone as one of my first bigger projects

If you just want to make a minecraft clone, then fixed-function OpenGL is fine for that and a lot simpler than modern OpenGL. More complicated things would require modern OpenGL, though, and if you just do fixed-function OpenGL, you'll definitely know enough to easily transition to modern OpenGL, but you wont learn as fast as you would if you just go right into modern OpenGL.

make it with the best hardware you can afford, when the game it's ready your hardware will be years old

Oh man oh man what an intriguing idea. Potential for better performance, though no alt-tabbing. I'm reading on it and the biggest drawback is you have to have your own mini OS, complete with drivers and compatible different hardware architectures. A standard one could be open source and free to use, but if you're going that far you might as well modify some linux distro.

I want to fucking die, but at least it's over now.

Are Sourceforge links no longer botnet? I remember them including malvertising software in their installers but it seems like people have forgotten about it. Did it stop?


Considering 99% of the PC gaming market owns a phone or some other screen, this isn't even entirely a drawback. Drivers are definitely an issue though.

The flying eyes look kinda cool, actually. For a stealth game, I think it would be elegant if the AI were implemented upside down, I mean enemies acting like they are intelligent, but they have almost nothing going on inside of them (basically just a number that is the individual alertness state), instead there is something like an intelligent tentacle-carpet thing that is oozing out of the player's character all the time and telling everyone it touches what to do. I would use two different patterns for sound detection: One for ground vibrations (would be a circle like yours, but usually smaller, so it doesn't alert people through thick walls) and another pattern that tries to fill a certain area, so the sound travels further in a narrow hallway. The player's movement could also leave a slowly disappearing trail (three different types: blood, which raises the alertness level tremendously; foot steps which have different visibility on different ground and enemies below some threshold alertness state don't even register them; and smell, which is only registered by dogs/humans with dog on a leash).

how do make geam

just like

What the hell happened while I was gone?
I thought /agdg/ threads were better then this.

Nowhere is save from shitposting anymore. It was nice while it lasted.

This is pretty tame compared to the literally insane shitposter I've seen pop up on the other /agdg/

Oh hey, Wolfspider

At least our shitposters still have their dicks, they have Rotatedev.

kinky.
mods if you ban this man or delete his post itll show youre jealous of his peener

Good point.

There's still devs there that circlejerk around Rotate, though. That doesn't make it much better. The last thing I want is these threads to turn out like theirs.

I'm almost afraid of asking…

There's also no googum or Boku and thank fuck for that.
But no Risk of Rain or fight knight. Something great needs to come from 8ch's /agdg/

Human anatomy is hard

At least it's saved for eternity.

Does Hopoo still post there?
I'm friends with the FK dev, he barely goes to the threads anymore.

Rotatedev was/is(?) a shitposter who back before the exodus kept posting webms of an extremely shitty patformer with a rotating protagonist, with just about zero actual progress at any point. At some point after the exodus I heard from someone here that it turned out to be a "girl" and then later that it's a tranny. That's about all I know though since I never go there myself and have no interest in doing so.


Speebot is plenty great, it just never got nearly as much attention.

This, speebot ruskie made 2 full release games while most people here are nodevs or went insane because of yoko taro.

Needs might not be the right word for this. It'd be wonderful if someone from this community made something great, but it doesn't need to happen, and it should absolutely not be forced. Just make the game you want to make, fill it with love, and finish it. That's already great enough.

I haven't seen Hopoo in ages, but I've seen FK dev post every so often, Googum or someone pretending to be him had the audacity to "who?" him and then the thread blew up.
I'm really excited for fight knight, and you tell him I'm rooting for him, he already knows, but it's nice to be reminded.


I really need to get around to checking out Speebot, but you also forgot about Sourcedev and his EXTREME SAMEFAGGOTRY and the monkmonk guy, which I think is the same person.

Doesn't the legend of doom modding post here in like every thread?

Which legend?

I haven't seen any space pirate posts in a while.

lurk moar

I've been lurking here for awhile now, which guy you talking about? Memoirs of magic? GMOTA? Booty project?

Classic. It's pretty sad that shit hasn't let up after all these years.
Sure thing. He's at a convention right now, but I'll let him know when he gets back. I'm glad he's finally gotten his break, I've known him for nearly 10 years and he's had his share of road bumps. It's a nice reminder that anyone can make it.

8 agdg is purgatory.
Our games are so good that we have be contained here so that we never finish them. Its a special kind of hell

I still think the first floor battle theme needs work, but god damn the desert battle theme is dope. There's a lot to love about Fight Knight with the cheeky writing and satisfying punchy combat.
and cute bug grills

manlet simulator

Ah right, that guy. He's getting back on track last I heard. I'm rooting for him, too.

Holla Forums, and especially not agdg, is not a fucking 4am thread.
You've already had your posts deleted once and I haven't seen you contribute tp these threads in any meaningful way beyond attention whoring.

I politely and unpolitely will tell you to fuck off, and report all of your posts on principle.

Manlet Simulator and Space Pirate/Booty are made by the same guy, Terminus. Also known from a few other DooM mods such as Demonsteele and Samsara.

Do NOT reply.

There is literally no reason to believe that will happen as long as the West continues on it's current path user.

pardon my newfaggotry on the topic but why would he cause devs to go insane?

I read the correct topology for boobs needs this loop, is this right?

I think that's the first time i've seen polygon on a tits,it's kinda magical.

Tetradev thought Yoko Taro stole his idea or character design or something. Going by some of the stuff he's said and done, he's not very mentally stable.

How the fuck do you fail algebra?

The weapon sway and it's position on the screen remind of Metroid Prime, which can only be a good thing.

It's already uncucking itself.

If you want to see the pixelshit of the future, just play Diary of A Spaceport Janitor

fair enough i suppose, but how exactly would that cause devs here to go insane? emulate him at all costs?

>>>/loomis/

people bullied him like they did yandere dev because they just wanted more drama like that

H-how?

I think it was just Tetradevs situation in general that made him miserable. If I remember right his family had financial trouble of some sort and he had a lot riding on the game he was developing. Problem is he was taking pretty long to make it. I guess when Nier Automata came out and it was a vaguely similar concept as his game, coupled with Yoko Taro posting on Holla Forums exactly once, he felt like he was being "stolen" from or something. It just seemed like straight up paranoia.


That had nothing to do with it. People made fun of him after he flipped his lid.

Well, to be honest i wouldn't be that surprised if it's true because taro browses 4chan, would even accept it if someone claimed Mario Odyssey stole it's hat idea from A Hat In Time, what i don't understand however is why tetra lost his shit to this degree because of it. It's regrettable when someone else takes your prize but if you post your game progress and gameplay out in the open then you have no right to be mad if someone steals it. It's like running around telling everyone about your great invention in detail and then wonder why someone patented it before you did.

Gravity is applied to tits post-modeling through jiggle physics you tard. If you add gravity to them while modeling you get double-saggy ones in-engine.

...

he didn't flip his lid until more and more people took interest and take a guess how people took interest

The only thing he had to go off was because his character design was a white haired android and was in an action game. Even if someone "stole" that idea, it's still just that, an idea. Not a particularly new or ground breaking one at that.


Nah, most anons were supportive of him when he still posted. I didn't care for his game but I don't remember any hostility until he went full retard.

well you don't remember how someone impersonated him then

And here I thought Godot2 was shit to make 3D stuff.

No one here was hostile to him, when he went into a Nier thread with his complaints the people there started following him.


He never denied being the person that posted there, and acted the same way as in that thread on his tumblr account so I have no problem believing it was him. No impersonation.

not denying or confirming doesn't really mean anything tbh

I don't. But if you think it's normal to drop a project you've worked on for years because you got impersonated or someone said something mean to you, you're wrong. That's not a course of action anyone sane makes.

considering the stuff he's gone through the answer to that question isn't likely that's he's gone insane but was forced to give up, the most likely answer is because kicked out of home with no where to go

That doesn't excuse insulting people or attacking random artists on tumblr because you think they drew a comic that makes fun of you. There's been other anons who said they've had to quit dev but didn't have a whole ordeal like this.

clearly you've never been under heavy stress then

I have. Again, his reaction was not a healthy one.

ok here's my two cent, have you ever heard about the death of a author? it's tells that ideas can be implemented into anything no matter how original it may be. in the end yoko will die and someone shall take his place.

I had an email exchange with him a few months ago.
The shit goes way back. And I mean way back to before Nier. I don't know what to make of him. The entire situation is a clusterfuck and I don't think it's going to be resolved, if I look back at the emails. And the worst part about this is that it could; at least I don't have reason to believe that the situation couldn't be worked out after he explained himself to me.
The thing is that he is a lot more justified in acting the way he did than Anons give him credit for, because there is a lot more to it than Nier, and yet it's the only thing people ever seem to talk about. The problem is that I don't think he's ever going to explain himself, which upsets me, because I don't think that anyone here would hold a grudge against him, if he did (the occasional shitposter notwithstanding).
The current situation is a deadlock where /agdg/ expect him to apologize while he expects /agdg/ to apologize. So obviously it's not going anywhere. What would have to happen is that somebody gets him to somehow come here to tell everyone what he told me, albeit with one caveat. He would have to calm down and be a lot less unfavorable than he was toward me.
I suggested something in our email exchange that upset him and I didn't understand why on earth it would. He eventually explained his point of view and made a solid argument. However, I pretty much had to pester him, I had to insist on him explaining himself. Just assuming that I had the worst intentions and brushing it off with (quote):
without any explanation is not helpful at all.

Anyway, I don't see that happening. I'd love to see him coming here and working everything out (or at least give it an honest to god try), but I'm not optimistic about that.

subbed for next gen revolutionaries

how did that happened? meant for >>14096442
totally not my fault

i give up posting

The sum of experiences is a crucial part in creation, but the synthesis of those experiences is what makes an author's creation unique, and therefore theirs.
pic unrelated

That's Godot 3.0

Polyobjects are cool shit. They're roundabout shit, but cool shit.
The short of the matter is that I make objects actually outside of the map, and then have linked pointers in the map referencing them as interactive entities. When the player interacts with the pointers, they look at the original linedefs and move them instead, which causes the lines in the map to go as well.
So I can make sliding doors, traditional doors, or more.

I usually just see anons lamenting he left. He made a fool of himself sure, but I'm not sure if that warrants a direct apology.
Asking for the impossible. Not in the history of anything has an imageboard ever collectively apologized for something. If he can't handle anonymity he should post his progress elsewhere, and I don't mean that in a degrading way. You do put yourself out in the open to strangers by posting your progress and it does make you a potential target.

Your post doesn't really shed light on what's wrong with him, though. Just that he's stubborn and dismissive, which I recall from the few exchanges I had with him.

Hope you are using subdivide. Make a 2x4 square grid of points on the center of the cheek (laying down rectangle), then cover it in a circle of points and form the butt shape radically outward.


I think you have too many vertices to shape with. The less vertices you work with the easier it is to shape. That's why we use subdiv. Never make edges with more than 5 vertices (called poles).

Jesus Christ, how lazy can you get.
Anyway, the death of the author is one of the fundamental concepts of post-modernism. If you boil it down, the idea is that the author's opinion on what a text is about bears no meaning whatsoever, because he doesn't deserve a special status. He is just a person, so why would his interpretation be any more valid than anyone else's? And since your interpretation of a text is influenced by your opinions, knowledge and time and place you live in, the interpretations will change from person to person, from time to time and from culture to culture. So in other words: there is no canonical interpretation of a text.
And while I don't know whether the guy who came up with the death of the author was a post-modernist faggot himself, we sure know what the post-modernists made of the that idea.
And if you apply that reasoning to things outside of literature, you have the justification for the complete dismissal of reason, truth and logic that you see being played out right now.

I don't want one either, but that's how he sees it anyway; even if it's just implicit by admitting that he did anything wrong.
I agree, but I don't think that's what he's asking for to begin with. It's difficult to explain, but I don't want to upload the emails here. It was a one on one conversation and making it public behind his back is a shitty thing to do, so I'm not going to.
I never got to the bottom of it myself and instead just scratched the surface. Our conversation didn't go anywhere, partially because I'm a fucking idiot who kept putting words into his mouth rather than just asking him to explain himself. It's not like he didn't do that himself, but I should have known better, just shut my face and let him talk.
In the end it was two strangers talking past each other until one of them got fed up with it.

That's understandable. I wouldn't beat myself up over your exchange either, I don't think there's anything anyone can do at this point.

Thanks for the ideas, i will definitely implement some of these.
Making the noise tolerance bigger when the enemy is on the other side of a wall is pretty easy, i don't know why i didn't implemented this before.


Here, I made some progress today and optimized the field of view and also added a console.

Holy shit, the wait is almost over.
github.com/godotengine/godot/issues/15321

You show who you really are under stress. Trying to pretend it wasn't him that made those posts when they were in line with how he's acted since doesn't help him and just adds fuel to his paranoia.

Am I the cancer if I actually enjoy this aesthetic? The way the world looks kind of reminds me of Megaman Legends.

If you want context for Tetradev's actions, it's probably because of his situation - the game was supposed his way of escaping it. Tetradev lives as a NEET on neetbux and he can't work because he has a huge amount of debt that he inherited. So, if he works, the money will just go into the debt. He lives in some slum in Poland and he could be kicked out of where he lives by the police at any time.

Basically, making a game like Tetrachrome and being put on a pedestal like yandev was before he had a falling out with 8ch, was (is) his only real shot at getting out of that.


The thread was a "what is your favorite game" thread. And it was definitely him who made the post.

While Earth is a huge recycling plant of ideas, it's the author who must get off his lazy ass to and put some effort into a given text or whatever he's making, and in doing so, there's a chance that at least something completely new will be born. Art is more complex than two lego pieces put together. As a bonus, the idea of cultural appropriation is thus shot down - I might be making a leap in logic, but in all honesty, fuck it.
I object to the idea that a given text or a piece of art can have an infinite amount of interpretations unless that's exactly what the author wants. Abstract paintings often display a lack of ideas and skill and are a huge amount of steaming nothing, so I'll accept that the death of the author applies to them. You can see literally anything you dream up in an empty canvas.

Pardon my interjection, I haven't engaged at all in this discussion but from what i understand, Tetradev was in deep shit, turned to gamedev as his last resort to get out of being NEET?This sound too much like my situation right now except that i'm a stereotypical mom's basement dweller.
I wouldn't be surprised if most of the people trying to make a game here have similar reasons for trying hard to gamedev. Autists with dreams: Dreams of getting out of misery.

Tetradev clearly has paranoia issues and while I can't blame him for the shit situation he's in he was just acting like an utter dumbass. He legitimately believed Taro stole his ideas when he came to visit 8/v/ that one time– despite the fact that Taro had the concept for YoRha battle androids in 2015 at least. I feel bad for the guy but what he needs is psychotherapy, not a apology for getting dunked on for acting like a crazy dumbass.

t.etradev


If he would bother explaining himself I wouldn't have a problem discussing something other than Neir. Frankly the only other rationale for his behavior is the rumors about family debt causing him undue stress, which is more of an irrationale.
Why don't you? Post pics of the emails and tell us what happened from your POV. Tetradev may be unwilling to talk here, but you obviously aren't.


Then ask him for permission to post the e-mails (or a paraphrase of the e-mails). If he says no, don't post it.

Sage because offtopic drama

Didn't know the current build was usable, I thought everybody here was waiting for the stable release like .


Amazing.

The game was in development since at least 2014. So, he did have the concept before Taro had the concept in 2015.

This is definitely out of the etiquette, but anyone have ideas for what sort of game would go with art assets like this?
Like genre or gameplay style, a couple friends of mine are making a game around it but we're all pretty stuck for what could work.

Adventure/exploration game with puzzles and little to no combat.

Hyper light drifter? some sort of maze game? or some dwarf fort circuit boards game

but i think you might be working backwards

Try a shmup maybe?

How would that work?


Definitely, but if anything we know who makes this stuff and can get more made.

The point is Taro did not "steal Tetradev's ideas" when he visited 8ch.

Beta 2 has been working a lot better than the Beta 1 build, for me, personally, so far. Only problem I've ran into is your exported variables don't seem to show up in the editor in a C# project.

Sure, I'm just pointing out that the evidence you used isn't valid because Tetrachrome had the same battle android girl as the protagonist since 2014 at least.

Just like a standard shmup, things like first pic being used for the enemies. 2-4 I could see as boss battles or just part of the background.

These pictures remind me a lot of Hyper Light Drifter, so you might not want to go the hack-and-slash route. Maybe a top-down twinstick shooter would work?
Choosing an aesthetic before even a genre is definitely unorthodox.

strawberry cubes if you hate yourself

I just don't know why tetradev didn't try to repurpose his game if he felt that Yoko Taro beat him to the punch with female battle androids. The hack and slash combat of Nier Automata isn't even that impressive for the genre, he could have probably made a better game by virtue of it's mechanics alone. It's probably easy for me to say how he could have continued his project, but still.
>But wait, the MC isn't an android, it's actually an exosuit piloted by a loli
Was changing his game from his original idea too much for him? Not getting married to an idea and keeping flexible with a project are good skills to have, but maybe he was too fa along with the game, idk.

Polite sage for offtopic, probably beating a dead horse here.

nier2.com/blog/2014/01/29/possible-connections-between-yorha-dodnier/

Multiple people including me repeatedly told him that if he was that bothered by Nier, he should change his shit up, but also that he shouldn't be bothered by Nier because they're only superficially similar. He'd then pretty much always got pissed off though because he only seemed to interpret that as "you have to change shit" and he didn't want to do so.
But yeah, we might wanna limit this Tetradev talk to after we hit 300 posts, so we can use as many bump posts for actual progress as we can.

That's better

I got some me-time finally and used it to make a better script for the wall fade. What do you dudes think? I'll be testing a community shadow-receiving transparent shader soon as well.

That game doesn't deserved acknowledgement.

The shading looks pretty good. I hope the character is a placeholder.

The transparency seems to be lagging behind the player, also, the amount of transparency seems to be done "per arch" instead of as a smooth gradient, which means that each arch model is really noticeable instead of a seamless gradient of transparency around the player. Other than that I thought it was pretty cool how it doesn't take effect when you are in front of the arch, and only when you walk through it.

Do you guys mind pushing me in the right direction? This should be a simple problem, but I'm honestly pulling a blank. I don't know why, but this is just fucking me up. I'm trying to align my horizontal acceleration vector with the surface normal, to make sure the speed doesn't vary when going up slopes.

Assuming the original horizontal vector is "aligned" to world up vector, I'm guessing I need to get the rotation difference between that, and the surface vector and then rotate the acceleration direction?

You guys mind tossing some pseudocode at me?

Why not have something like a "window" of transparency? As it is now it looks kind of cheap.

Vec3 Right = Cross(Horizontal, Normal) //make a right-vector, if you don't have one yet
Vec3 Dir = Cross(Normal, Right) //creates a vector in the direction you want to move
Vec3 Mov = Normalize(Dir) * Length(Horizontal) //makes the final movement vector the same length as the original

I found this
gamedev.stackexchange.com/questions/72978/shader-that-cuts-hole-through-all-geometry
That may help you have more Fallout 1-2 style holes in the geometry.

I try to avoid stupid questions but google hasn't been helpful in this regard.
In Love2D, what's the best way to run certain functions not on every tick, but every few ticks (say, 1/120th of a second or so). Should I just use a counter that adds up (dt)s until it reaches the interval I need and then reset it?

I ask because checking controller inputs about 8000 times a second seems like a bit of a waste to me, and running that many commands that many times a second is bound to tank performance.

I'm trying to learn unity atm and I'm reading through the manual first before I do anything substantial so I can at least get the info in my head first. I read across this part


in the "controlling game objects using components" section and I'm a little confused. This implies that the entire game must be static and nothing can be dynamically created or destroyed if it must be connected in the editor first… what the fuck? am I fundamentally misunderstanding this? Can someone perhaps make it more clear?

Usually for input you use an event system, rather than updates. en.wikipedia.org/wiki/Event_(computing)

Look up event handling for your engine and see if it has native support, otherwise you'll have to write your own setup

Thanks!

I'm not sure how well you understand components so I'm just gonna explain from the beginning. All objects use components, they all start with a transform component, which means they all have a position on the world, other objects like the camera have a camera component, any visible object has a render component, light has a light component and so on, all scripts are components too.

To modify these components, like modify the transform to make the character move, you need a reference to it. References can be acquired in two ways: through the editor or through code.

When you make a public variable, it shows up on the editor on the inspector tab of the object (like in the picture, those are the public variables of my character controller script), meaning you can modify it any way you like, even during runtime (runtime meaning when you are still playing the game through the editor, not when you build the .exe and play though it). This works for references to objects too, you can drag and drop an object from the hierarchy view into the inspector tab and into the variable and it will work as if you gave it a starting value, it's the same as if you did this:

Assigning values through the editor should only be used for debugging, in the build the best practice is to only assign values through code. You can get references to objects by directly assigning it, like:
Or by using GameObject.Find("Object name here") which will return a GameObject variable, references to the components of the object can be used by calling GetComponent on the reference to the GameObject you have.

I've read the page you are quoting and they used a confusing way to explain things.

Love2D has callbacks for keyboard, mouse, and joystick (gamepad) input:
love2d.org/wiki/love.keyboard
love2d.org/wiki/love.mouse
(look at the "See Also" section for the callback functions)

Ok, this is all understandable. I think I might have taken that paragraph out of context then it looks like. Thank you

I spent all day modifying a soundboard and have ended up almost completely revising the whole thing because I can't figure out why it's not updating correctly after playing sounds and for some reason causes horrendous spikes in CPU usage in between because I think the thread for playing the sound gets stuck.

I wanna quit gamemaker.

Good news, it's pretty much gone to shit these days, from what I hear.
Godot might be a suitable replacement, though.

A J E W - F R E E E X P E R I E N C E

That will be the best gamedev decision you ever make. GameMaker is steaming dogshit.

So, I'm not sure if I want multiple players or not.

Video related is from the original game. You have a "player one" and a "player two" object, and each has its own set of input it responds to. The brown player uses the arrow keys, and the blue player uses WASD. If you have multiple copies of that object, they all move at the same time.

In all 400 levels of that game, I've seen less than 10 puzzles use both players at the same time; more likely, you'll have multiple copies of your own player. Is it worth the effort to split inventories, etc?

Finished my input.

Took another anons advice and have a .cfg file that gets loaded.

This binds keys to various integers that get pushed to a command (basically a entity id paired with an integer) stack that gets processed by the world class and is applied to the associated entity id.

This doubles as a pipeline I can use to script behavior.

I can also now tie objects together as groups if i associate the ids together.

Shits gucci bois and I'm still feeling productive.

Now I just gotta iron out my GameObject's Graphic / Physic Components and I might be able to put together a demo for Feb.

...

...

Writing my own collision code from scratch is the most painfully unfun thing I have done in programming so far

What kind of collisions are you doing? Rectangles are pretty easy and serviceable for most applications. Polygonal collision took me some time to wrap my head around, but that was pretty satisfying.

Spent the last 2 days coding a undo and redo system in my editor. Finally finished it, I think. I would post visual progress but I'm beat, tomorrow, I'll record a gif or something.

I am writing code like this and I kinda want to kill myself, but it works I guess

Just rectangular collision. I'm not a terrible programmer but I'm a complete and utter retard when it comes to straight math.

Pls no bully

I just got it working though, so I'm not hopeless.

These are some bugs i'm having in Vulkan, I fixed a lot of issues but now i'm trying to re-use the command buffer when I can instead of re-recording it each frame, but i'm not entirely sure what exactly to disable and enable so i'm slowly making progress on it. Luckily the rendering code is rather simple right now.

Good. Make sure though that you test for each collision only once per cycle.

That webm gave my computer cancer. Seriously, everything plugged into USB kept reinitializing until I stopped playback.

And here it is while re-recoding the command buffer each frame. It's slower but whatever, I can fix it later since I have other more interesting problems to solve.


I recorded it OBS and then re-encoded it from flv to webm in webm for retards. I have a version of OBS that is a couple years old from when I installed it, and I probably have the first version of webm for retards that was released on 4chan when webm support was added in 2014 or something. This mp4 is encoded in a newer version of ffmpeg (maybe a year old) from flv. Hopefully it doesn't kill your computer.

No worries, everything's fine now. It was probably something to do with how my browser plays VP8 video.

should i actually think or use path2d in godot?

Not sure I'm thrilled with my solution of a grid of collision squares. It works but it's pretty wasteful in terms of how it is stored as a bitmap grid that's full of empty space. I think I'm going to try polygonal collision instead, as checking if a point is inside a triangle is mathematically trivial and it would be a lot more efficient in terms of memory and would obviously allow angled walls by design.

Yeah, collision is rough. I threw my own together some months ago and while it still needs work, I hated it so much I haven't been able to motivate myself to improve it. Squares alone aren't so bad, but when you get into slopes, that's where it becomes a pain in the ass.

I have a decent amount of experience working with existing 2D and 3D collision systems so I'm pretty familiar with dealing with complex angles, but writing them myself is a kick to balls.

Just use some basic compression, for example never store things like: [ 1 1 1 1 2 2 2 2 3 3 4 4 4 4 1 1 1 1 ], instead you could just store something like: [ 1 4 2 4 3 2 4 4 1 4 ], this works very well since most 2d games like to repeat tiles for long stretches, so you could see some really fast gains this way. Granted polygonal collision is far superior than this if you want angled walls, although you could use a grid of polygonal shapes to get the best of both worlds.

as many people do. I'm still looking for something to replace it. Not that easy, I grew fond of my blobby bois


it's lerped so I can tweak it to lag less. Check out the demo day build, it was instantaneous there.
The "gradient of transparency" we both would like to see is unfortunately something that will require more shader knowledge than I currently possess. But maybe the solution from the other user will bring me closer to an effect like that. If not, it will remain 'per arch'.


Originally it was something of a quickie last-minute solution before demo day but I grew fond of it. People asked for the transparency effect to be less jarring, so here are the results.

I did find that stackexchange thread when looking for a solution back then in November. Perhaps I should return to it. I wonder if I can make the borders of the "transparency window" blurred, though. If not, I guess I'll change it if/when I learn more on shaders.

Transparency just needs to be per-vertex and not per model, the way you do this is with color blending. Each vertex has an RGBA color associated with it, usually this is just set to white with no transparency which does nothing, but you can adjust the alpha component of each vertex to adjust its transparency, this achieves the gradient.

I'm not that great at shaders so I forget how it's actually written in GLSL or whatever, but that's the concept of how it works.

Advanced tile map editor in YUME 2.

why make a map editor for something like that

Aw sheit it's Keyreal. Welcome back man, been a while since we've seen you right?


Because you can use it in the final game, like how Speebot has a map editor? Plus it's just a ton easier than dong it with text files and you can compress the map format more.

more like shitbot tbh

I lurk every thread and post occasionally, but I haven't posted any content lately on account of my broken finger. It is slowly recovering so I'm getting back on track.

Are you doing anything fancy with how the data is stored and operated on, or is it just an editor loaded with JUICE

There's some optimization going on but nothing too complicated yet. All static tiles and entities (which can also by dynamic) that use the same texture are batched together during map compilation into a single model. This significantly reduces the number of draw calls.
Entities and tiles also have their own bounding boxes and spheres, which will also be grouped and optimized as much as possible during compilation in the future.
All data is stored in files that are hotloaded by the engine, so I can create all the components and maps with those components in one session, without recompiling or restarting the game.
This also works well with git as I can see which entity, map, tile or other component was changed in a commit.

Looks good user

I fucking hate collision code, from SFML to unity rigidbody2Ds to writing my own AABB

Also, is there a particular reason why it's an 8x8 chunk?

Seeing this thread all the time has inspired me to get back into coding and game dev. When I was senior in HS I actually got pretty far with my knowledge in C++ and using the SDL library, but I haven't touched anything since now. I think as I've gotten older I can grasp coding concepts a bit better, but its still hard to understand at some points especially when I start getting into classes, arrays and pointers. Sage for blog posting but I see all the cool shit you anons are doing and its damn motivating.

Damn Satan

Turn them into some easter egg or enemy.

No, there's no limit on the dimensions.

I post my progress on jewtube sometimes but for you guys it's gonna be raw mp4 uploads. Post your shits too sometimes.

Honestly I look at these designs and I see floor plans for levels. They look like they could be maps for an ancient civilization with a taste for ornamental architecture and very deliberate city layouts.

i finished the head, now I'm just finishing up the body

are you the shota drawfag?

Nice filename    fag.

I demand a testimony.

if you mean the logan paul meme drawing I did and spammed everywhere for shits and giggles then yeah, I'm that guy.

it's true though, he's suppose to be a monster boy.

no no I mean the guy who did this

Kanker homo

yeah I did that
jesus that's old and shitty art

I'm pleased you're still around. I miss alot of people who seemed like they used to post here but I'm glad at least you're not gone yet

thanks again for this

aw, thanks.
I post my work recently more on loomis.
my art hasn't really changed that much

anyway on to the /agdg/ posting

is your gayme about cute boys doing cute things and also dicks?

...

how much money has speebot made you? was it worth it?

...

oh you're the nigger on /loomis/ too huh

HET HOMO GEMEENSCHAP


You may be a fag but at least your 3D model looks pretty good.


My guess based on Steamspy's data would be $6600, so I'd be interested in an accurate number as well to compare.

...

Looking at his steam spy page now, it looks like people spent $10989 on the game, since Steam takes 30% of that, so it would be $7592 that keyreal gets before Taxes are factored in. If you guess that his tax rate is the average US 2017 effective income tax, of 13.5%, then it drops down to $6567, then subtract the $100 he had to pay to get his game on the steam store, it goes down further to $6467.

C++'s STL is so shit. This isn't even considering the other problems map has.


No wonder why AAA and soyim with rich parents are the only people who can survive in this industry.

you're disgusting mark


isn't actually alot

stackoverflow.com/questions/29085942/stdbad-alloc-when-inserting-into-stdunordered-map

Jesus christ, I'll stay in my C# managed safespace, thank you

TBH I don't know what the fuck that nigga is doin

Why do you keep saying "flash game"? It's not very much like any kind of flash game. It's more comparable to a phone game actually, I bet it would do well in that market.

it's that kinda game that doesn't have much going for it other then the fact that it's very solid for what it is

it's that kinda game that doesn't have much going for it other then the fact that it's very solid for what it is

If you look at his other posts he seems to be butthurt about speebot for some reason rust-fag-dev 2.0?

They're well drawn but fuck they look like traps.

and?

I don't dare imagining rule 34 and that's a problem.

it's just flesh at the end of the day, flesh that isn't even real even

As promised.

here be waifus, opinions??

I'd fuck him

I wouldn't mind that hunchback golem (?) as a character, but it needs some animation and juice.


Looks pretty gud so far. Any idea what your next game's going to be?

Glad to hear. Hope you heal up soon!

Friend loved the cutscene art, so she made an HD version of it.
Terrifying.

Why is there solid diarrhea coming out of her butt?

4c725b should be able to appreciate this.

wolfgirl waifu has wolfgirl tail, a furfag such as yourself should be able to appreciate this!

Shouldn't the base of the tail be a little higher? I'm no expert; it just looks low to me.

gotta be honest, i think you might be on to something here -.-

Cute but tail shouldn't be coming out of her butt.


Not cute.

A post processed 360 degree turn is not good enough for me to give you good advice. Post topology pics of upper body and lower body. Did you use corrective shape keys for biceps? What engine are you using?

From that gif alone:
Good shader, bad deformations, you didn't use subdivide or your weights are weird, swimsuit is nice but doesn't seem to fit with body topology, face is too flat, proportions make it very manly (unless you are going for a gay furry anime thing). Need more context.

It looks like it's coming straight out of the butthole, you have to give it space, make it look like it's an extension of the spine. Look, even this stupid cartoony character gets it right.

Anyone know how to work out an optimal acceleration to achieve defined maximum velocity in defined length of time (from zero) given current velocity and current tick delta seconds? Trying to make my movement feel good.

Use the difference between the current and desired velocity as acceleration. So you basically want to accelerate the object by the remainder times delta time times some constant factor, which will determine how quickly you reach the top speed.

I mean I get that I gotta multiply by delta somewhere there, but it's not clear what the constant you're referring to should be.

This is called once on the beginning of the acceleration.
VelocityToChange=CurrentVelocity-DesiredVelocity;
Acceleration=VelocityToChange/DesiredTime

This is called every frame:
CurrentVelocity+= Acceleration*DeltaTickTime

I also advise you to look into position, velocity and acceleration curves, they will give you a better idea how your character will behave.

Alright, thanks!

Love2D or Godot2D? I've never tried 2D gamedev before.

3d is just 2d except you don't have a w/z axis there's still an actual z for what gets drawn on 'top'

Love2D is a framework, so you'll be managing updates and content yourself.
Godot is a full engine and does what Love2D does, and additional management.

Both are free and open source, it just depends on the kind of work you want to do, and the kind of game you had in mind.

Is it possible to apply Rigify rig to Mecanim in Unity? I tried to find some tutorial in thread archive or google, that's retard proof enough for me but I got nothing. It's either too vague or doesn't work when I try it. I tried to find the proper hierarchy of bones through trail and error but there's always some shit with the way either shoulders or legs connect to the spine. What do?

We've got pages on both on the wiki 8agdg.wikidot.com/love2d and 8agdg.wikidot.com/godot in case you didn't know.

it seems that the problem is them being dislocated, try joining them and re importing

Went with a dual rect/triangle collision system. Rectangles are obviously quicker to compute and it should make it a bit easier to fill large square areas while also having the ability to use angled edges. No gifs/weebms since I'm working on this on my phone.

The number looks small on paper but when you consider what that number is worth it's actually not bad.
I could go full-time game dev for 3-4 months with that much money

The game took a year to develop, if you take a conservative estimate and say that he spent 1000 hours on it (he probably spent much more) then he made like $6.40 an hour for his work. Granted I do a agree that it would be worth it considering that nobody else here is making any money.

Actually I just thought of that again, it would cover like 6-9 months of my living expenses. If you live in Poland like half of /agdg/ seems to do I bet you could do a lot better.

but then you have to account in your local taxes on top of us's

>Ontario just raised the (((minimum wage))) to $14.00
Seriously, the cock of living just went up, because for many people that shoves them into a higher tax bracket, while at the same time scaling the cost of goods and income, while also getting reduced hours. It benefits nobody

Can confirm. My annual living expenses are ~3500 dollars.

t. Eastern European nodev

That's a nice freudian slip right there

Kek. I heard some people pay such amount monthly in commiefornia. Sometimes i think of moving to my relatives in poland, but then again i don't want to be the same as rapefugees who run away from their country instead of trying to fix the problems within the country i live in right now.

"The cock of living just went up" sounds like someone trying to directly translate a saying from some kind of hindu or dunecoon expression into English. Like the weird kind of profanities you might hear spouted by someone who's clearly not a native speaker of the language.

I'll stick to vid related for now even if the lighting is a bit fucked but I'll keep your suggestions in a todo list. i have other tweaks & things to fix that pointed out way back during demo day.

Added extra bumpmap strength to make sure it's clear this is a wall and not a walkable arch, I was told it wasn't clear even with the shadows…

big meanie butt ghost
very cute


I want to see the qt animated. She looks good at first glance.
obligatory tail-out-of-butt comment


glad you're back on track. I'm not sure if you want to answer this question but I hope it went at least as well as in case of Hypnorain


good stuff. Is it a part of a project or just training?

I hope the items, head bob and an animated face will provide just that. I really need to start working on the armours and shit


I'll be looking into mesh deformation to create 100% randomized horns for 'em at least.


thx bro

If I moved out of where I live and into a trailer park in a cheaper state, I could probably live off of that for much longer.


Arab insults usually have to do with shoes. If you ever see anything like that, it's definitely an Arab.

Stones, shoes and spitting as far as I know

I did manage to catch myself before posting, but I thought it was funny enough to leave in.

Anyways, suppose I made a d20 styled game like Baldur's Gate or whatever. Would a dice rolling system work if I used Random.Next(0,20), or would it be preferrable to do Rand.NextFloat() * 20 instead?

Also does anyone here know how the NPC movement in RPG maker and Pokemon is handled? Specifically for events.

I tried creating an object containing objects with X and Y values and then go through them with a function that animates the NPC from where he's standing to either X or Y, however it's just fucking up a lot and the movements are happening together (making the character move diagonally) which I don't want.

Hey, that's fun. I always put root namespaces in whatever I was using, by habit (eg if I was using System.Collections.Generic, I'd tend to include System, as well). But it turns out it has a lot of extra junk, like C#-specific type names (eg Int32) versus the .NET-defined ones (eg int). Shit like exceptions are also there.

One of the interesting things was the Void struct. Methods that return void are a distinctly separate thing than the Void struct. This is a struct with no accessible fields and is mostly used as a placeholder in reflection.

Not really relevant to anything, but I thought it was neat.

Dice are integer based, in a range from 1-20 inclusive. You should check to make sure your implementation of Random has a good distribution, though. Some implementations are not distributed nicely for the sake of speed or because they were implemented by retards.

Well I mean I'd just be using the default system implementation for Random, which is fine. But the question was whether to do true dicerolls, or generate a float, scale it, and round it off (which I think is what the integer random functions do internally, anyways)

For gamedev purposes, it might be a good idea to get or build something that allows you to specify a seed value for debugging purposes, and log the seed on every play. Or maybe to save the generator state so loading the game will always get the same result. And to actually make sure the seed actually changes so you don't get the same result every time unless you actually want that to happen.
Beyond that, just use whichever is easier to implement and check the distribution with a million simulated rolls or something and see that you get roughly the same amount of each number. Probably integers. If it's all kinds of fucked up, maybe get an external library or use floats and check the results again.
If you do decide on floats, take care that you don't screw up with the edge cases. Plenty of implementations round numbers off wrong and end up getting only half the probability at the very extremes. I also recall reading some rant about a random float generator implementation that had a minute chance of generating a 0 or 1 and causing problems because the author's program was expecting something strictly between those numbers.

So, uh, if you don't care about the seed stuff, just use integers, check the results and work from there.

long random_at_most(long max) { unsigned long // max

...

At least the job is pretty fun though plus i make decent money

sad, good luck user

Can anyone point me in the right direction on how to get vectoring thrusters to evenly distribute thrust around an object like how this game does it? I don't know the first term to look up for more info on something like this

There are ways to work around being in a higher tax bracket, like donating to registered charities, or putting them into certain kinds of investments.
I'd do some research on how if that does affect you, since the cost of living really only went up for stupid people.

Seems just a little bit hard to see the player from behind the fade
Other than that good work user

She's really fucking cute

Question for all the codebros:
Are you guys good at math?
Do you enjoy it?

I don't think so.

Yes. Learning it is fun.

I hate formal math, I like math in programming because it is more about understanding relationships than memorizing all of the rules yourself.

I'm literally asking which syntax makes more sense.
I'm not implementing a random function.

I graduated high school with a 9/10 for maths, with the more advanced course option. Yet I wouldn't call myself good at maths. Can't remember shit unless I crammed it the night before, but if I did that or am allowed to look shit up I'm "good" at it.
I would say I am good at and enjoy problem solving. That kinda counts.

Why don't you get a fulltime job gamedeving then user?

also
whatdoyoudo?

I think I'm above average in math at uni and I love math. Once you start to really grasp it it's like your vision of the world changes, everything "makes sense" as long as I can use math or physics to explain it, it makes the world seem less random or chaotic. I always keep wondering if there's a better way of doing things using math, the more advanced forms of math could further improve my vision of the world, and with that I could find other ways to improve the way I code.

Nope and nope.

If I have to use the math for what I am working on , then I become good at math. However I am really bad at math class and really hate it a lot because of how bad at it I am. I'm really shit at calculus and I can't do a linear algebra worksheet because I have learned linear algebra exclusively through programming.


Memri TV is the greatest thing ever, if you don't watch it already

NEW THREAD WHEN

Shove that forced meme up your anus, user.

page 13

Y-you too

NEW THREAD

software dev
also i hear gamedev jobs are pretty shit