/agdg/ + /vm/ ~ Amateur Gamedev General

Looks pretty good. How many tris? Seems to be lower poly than most stuff you'd expect in UE4. Is that animated too?

See you in two years, user. I should be releasing my game by then.

I get the feeling that I'll try to treat them well and inevitably fuck something up by accident.

Well, I'm hoping to be one of those outliers.
Vidme's an option, perhaps. But then again, if I knew PR, I wouldn't be in this predicament in life.

You don't have to draw user.
Just use blender or suming

It has 8156 tris. I have yet to animate it.

When the faerie village is going to shit you can cook them sweets like pudding to fill the food and amenities resource bars in case you fail to automate everything or an event causes a disruption

The cooking system is done, the village management is about 50~60% finished

Since im going to leave my job and focus on gamedev 100% next month i will probably have this minigame/feature demo finished by late June

I think what you have to do to attract potential collaborators is have a presence on the asset stores, youtube, and/or the various other social media showcasing aggregators.

I just imagined /agdg/tan as an ara that gives birth to babies and raise them into lolis, so they can play with Holla Forums and vivian

CAN'T

this makes the most sense

Don't start a studio.
Don't even quit your dayjob (unless you're already a NEET)

Google "writing indie games is like being a musician" and you see someone who's DECADES in the industry deciding his next game will be Assetflip: the Game. The industry is that bad, so unless you already have a Journo Cock you're sucking, the only way to move forward is to go cheap and just focus on making an enjoyable game and get a return of $500 for your $100 game, in which case i say, congrats, your game is profitable.

Believe me, I'm one of the anons who commissions shit all the time and I'm only doing this because I want a little quality and have a day job, but otherwise its a net debt and if I were really seeking profitability I'll be doing what all the asset-flipping Russian devs are doing.

Is there some template to follow and or a general lists of "dos and donts" floating around out there for Swappers/Matching games?
The stuff I've been able to find on my own is generally useless and no one seems to want to give up the goods on how to make them in a presentable way.
Hell a pal of mine suggested just buying a nearly complete generic one and replacing it with all my own assets - - That's a possibility, but I do not want to go down that route.

I want to make something somewhat presentable by myself within the next two years.

I want to add a skeleton tracking system in the game that the Player has to watch over and analyze for injuries, but I'm too worried that it might be too hard for Players.

...

It would make the most sense if /agdg/tan is the kid of vivian.
Though, vivian is too young, so yeah… that's off the table. Also, vivian as the kid just doesn't work imo, but that doesn't mean /agdg/tan can't think of vivian as the equivalent of their own child; despite having no "blood relations".
Overall, I think it fits to have an ara as /agdg/tan.

Also, her kids are the games created via anons of /agdg/; that "can play with Holla Forums and vivian".

If you go full medschool, then maybe, but don't dumb your game down just to appease the lowest common denominator.
Don't work to appease the chick from embed related.

...

How can girls have prpblem eyebrows if they shave them and paint them back on

Please tell me it's fake.
I don't like BOTW but this level of stupidity has to just be satire.
right?

We are totally gonna need to borrow some rope from Holla Forums or /suicide/.

You cant give it back if youre dead retard

God, I HATE this bitch.

jesus fucking christ fucking jesus fucking christ fucking jesus fucking christ fucking jesus fucking christ fucking jesus fucking christ fucking jesus fucking christ fucking jesus fucking christ fucking jesus fucking christ fucking jesus fucking christ fucking jesus fucking christ fucking jesus fucking christ fucking jesus fucking christ fucking jesus fucking
CHRIST

Add it. Then play around with it to see if it's fun. If it is, keep it, if it's not, remove it.
If you appeal to the lowest common denominator you won't create a game that you love.

Wait, you wanna make money or just make a good game? Have even made a game at all, even a simple one?

If you want to make money, just create a couple of simple but fun games and get them out there (have youtube channel and other social media, promote them anywhere you can) so you can at least be on people's radar, if you see a good reaction then start a more ambitious project, at that point you might be able to have a patreon or something like that.

The only thing good about stupid as fuck people is you can say "At least i'm not this fucking stupid"

...

sorry user
this is the bad WW2 timeline

Do it you jackass. The only concern when you design a feature isn't how dumb the player is, it's how about impractical it is or is not in regards to time to implement.

The only thing that matters is how you want it to be. Pandering to others does not end well, and the simple fact is that it's because you cannot read minds to understand the true wants instead of the projected wants.

There's an old saying: If you build it, they will come.

And that's all you have to keep in mind.

Release your game, unfiltered and full of your brilliance, missteps and character. Create the thing that you want, and other will want it too.

No one wants a bland thing, maimed and castrated and put on display.

fucking
DO IT

I believe those things are synonymous after a certain point. Sure you can make something really cheap and low budget that can be good, but you will always be limited to the scope of one person and no budget.

Like right after I posted that I found a vid of a dude interviewing a guy who made an obscure game in the 90s with a team. He basically described it as a bunch of people hanging out in one dudes house all day coding on a game. It seems every indie-game back in the day use to work that way, but nowadays it's just people trying to 1 man team which I don't buy into.

Personally I don't really see patreon as super viable for games. It's like a salary for one person, instead of a budget for an entire studio. I'd rather go with the loan option like frictional games did, but the state of the industry has made AA fucking impossible.

I modded a shitton and wrote a few things in OpenGL. The games I made where back when I was fucking with shit in gamemaker ages ago and aren't even worth mentioning.

Is that the girl from King of the Nerds Zimbabwe?
virgil a best

Yeah that has to be her.

goddamn nigga he fucked up

Okay, let's go crazy with this feature then.

Good luck.

So that idiot wants all of the content to be done in exactly 1 run
he doesn't want any games to have any shit new on replays, so essentially he wants games to be stale as fuck on replays, making that $60+ game only good for 1 run
What a retard, wanting less like that
I bet he's a hit with the ladies, talks with them once for hours then has nothing else to talk about

Now I'm just picturing him having a conversation at a bar.

I'm thinking of doing everything myself, but pulling a Tom Clancy. Like "Tom Clansy's sneaky man simulator" that only I developed and let Unreal Engine and Epic employees and youtube tutorials soften the blow for me when it comes to coding.
You know like Yang Bing, the ONE MAN studio machine.

...

you know he buys all most of his stuff right?

...

Reminder

ONE GUY
He made the game he wanted to make, who cares? It also doesn't look like shit because apparently the guy has standards.

Major outlets are for major players and cock suckers but they appear to show unrelated people from time to time.

How about trying to advertise through many smaller people and gamesites instead of focusing just on big players.
For example:

At the same time you would try to contact bigger media outlets as normally a dev would do but since your game is unknown anyway it wouldn't hurt to contact smaller guys and give away your games

So what you're saying is I should make Gone Home?

Gone Home isn't a game.

You have to be one of those people who love the game.

If that's unironically the message you got from it then I don't think there's any hope for you.

Seeing that you think buying public assets just to get higher quality rather than make something original is good, I really don't think there is.

c'mon user, you're better than not knowing that

I'm trying to write a basic air/water simulator that has pressure physics. Anyone know on what I should read up on for this sort of thing?

Large scale or small scale? Are you trying to simulate planetary weather or an airlock, basically.

Small scale, I guess. All voxel-block-whatevertehfuck based.

That sounds like a good strategy. I kinda forgot about the smaller guys.

Gamer girls were a mistake

That's precisely the problem

How do games like stronghold, blitzkrieg or diablo organize the millions of tiles needed for their maps?

I use a local SQL server for that sort of thing

Not what i meant.
Assume you have a 1000x1000 map, how do you efficiently store the 1million sprite tiles in memory? What if you want even bigger maps?
Considering those games are old there must be an efficient way to do this.

that's what I said. Well, more specifically, I store each part of the world in chunks, and their data as binary.

Chunk
X - 0
Y - 0
Binary Data - [Blob]

It'd be silly to have them all in memory, of course, you just pull in the ones you need (e.g.)

var mychunk = server.Where(a => a.x == myAreaX && a.y == myAreaY);

software.intel.com/en-us/articles/fluid-simulation-for-video-games-part-1

Also brush up on your calculus. The math behind fluid dynamics is no joke.

Sage for double post. I forgot to mention that's part 1 of a 19 part series of articles. You can find all the links at the bottom of the article.

3d fluid physics in games always looks like shit though.

Test posting, Holla Forums won't let me fucking post.

Yes, they look like shit, but that's with all the leading experts in the field working on it and it's still shit.
Would you rather not have them? Like, 2005-era water effects based on basic trigonometry?

It's not 1998 anymore. Any cutting edge technology wont have a clever use like it did in the past. Devs will just fucking cram that shit into whatever at the expense of the game, even if it's botched so hard it's pointless, and a few people here and there will make smart use of it.

I guess it's better to have it than not to have it, but I remain skeptical.

...

Why are you telling me? I'm just giving the guy what he asked for.

you don't store them all in memory, you store the ones you have (maybe 100) and assign them to certain places on the map. most of the memory management behind that is built into the graphic cards via OpenGL/DirectX/Vulcan, you as a dev are not supposed to write that from scratch unless your willing to compete with those 3.

For the hundredth time, this ONE MAN UNREAL ENGINE game meme has gone on for too long.

Looka here, he has a Chink army:

ultizerogames.com/enindex.html

It only takes two people to make an Unreal Engine 4 game /dev

So, is there any reason I shouldn't use Ubuntu?

I personally think it'd make most sense if she were a 20-something, I really get the impression most people in here are about that age.
A lot of tans are of that or some indeterminate inbetween age, like Holla Forums, /vr/, /x/ and to some degree Holla Forums he looks a bit more mature but that's probably just him being a well groomed and dressed man as opposed to /vr/
Usually tans are milfs or lolis because that relates to their subject matter somehow, like how /a/ is filled with loli weeaboo faggots or the Wii U was released a year earlier than the other consoles of that gen.

That was exactly the point was making, but it seems you're too autistic to read between the lines.

oh you rigged the model, was it hard?

Why aren't you using Gentoo/GNU/Linux/KDE?

It was certainly time consuming, but not too hard.
It was a sculpt so I had to retopologize it. There was almost no edge loops to it. I reconstructed his posed right arm, so I had to fix the UVs for that too.

Basically, I don't want to be a windows cuck, but I'm not some fucking wizard who can use Gentoo/Arch.

Also, if CUDA is really important to fast rendering, I really should switch to something that is supported by Nvidia.

That's what my thinking was with the analogy of the "games being her kids".
It relates, and matches perfectly imo.

Eh, yeah I suppose most here are likely that age, but does it match the persona of the character; as it's more an embodiment of the board's essence.
e.g. vivian james = vidya games
appearance w/proper contextual objects implies she's more obsessed with games than life
also she's young as video games are a hobby most picked up when they were kids
green/purple theme -> daily dose
etc etc.

/agdg/ is definitely something done when one matures, and gets the dedication/willpower to focus on something as insurmountable as making a game; in addition to /agdg/ fostering that love/dedication into a form that's self apparent (analogy, game = kid, nurture = /agdg/ threads, dedication = good kid, game influences = kid's friends, etc).

If there's another suitable analogy that you can come up with, that embodies the essence of /agdg/ then lets hear it.

use compute shaders (windows, linux, mac supported by ATI/nvidia), or OpenCL (windows, linux, mac, supported by ATI/nvidia).

nice retopo, I can never do the hands properly in topogun.

How much worse is OpenCL?

That reminds me of something. I've gone through a bit of a tough choice some months ago. I was developing a game for more than a year until September. Then, I had to start going to college, which made me stop developing completely. I had to choose, and I decided to do what was best for me: abandoning gamedev entirely. I decided I would focus on college, then work in my free time in summer to pay costs, and do small projects here and there, probably commissions for 3D models.

But every time I remember my game, I just want to go back to working on it. But finishing it would take a long time. Anyone in the same predicament?

OpenCL is great imo, and I would be using it for my current project if there was a proper port/wrapper for C# without massive overhead (as source is C++); it can use dispatched threads from the GPU or CPU (though, GPU is best).
OpenCL also has superior tools, and documentation.

Compute shaders is more graphics specific (written in HLSL), but can do mostly everything OpenCL can do; while being specific to only dispatching threads from the GPU.

eriksmistad.no/getting-started-with-opencl-and-gpu-computing/

FUCK YOU YOU NIGGERFAGGOT I JUST SPENT THREE FUCKING HOURS TRYING TO INSTALL CUDA AND IT WOULDN'T LET ME BECAUSE IT WANTS ME TO CLOSE ALL OF THE X SERVERS

I TRIED CLOSING THEM ALL BUT IT DIDN'T DO SHIT, I TRIED USING sudo init 3 BUT IT JUST BROUGHT ME TO A BLACK SCREEN WITH NO TERMINAL

I FOLLOWED ALL OF THE DIRECTIONS AND IT STILL DOESN'T FUCKING WORK

FUCK THIS SHIT I'M INSTALLING UBUNTU THIS IS ABSOLUTE HORSESHIT

also, CUDA is too limiting in terms of GPU manufacture support; as it's only supported by nvidia cards (unless u use an emulator on ATI cards).
thus, it's nonviable for usage in a game, but for tech demos/non-distributed projects it's fine.

I didn't touch the hands. I didn't need much movement in them.

WHAT THE FUCK IS AN XSERVER EVEN

WHAT THE FUCK IS RUNLEVEL 3 EVEN

WHAT THE FUCK IS CUDA EVEN

WHY DO I HAVE TO DO THIS SHIT TO HAVE DECENT RENDER TIMES

WHY DID MICROSOFT HAVE TO SHIT THE BED AND MAKE ME SWITCH TO THIS NERD SHIT OS

Calm down and rm -rf /

One more for ara.

Let's be autistic about this. Isn't /agdg/ older than #GamerGate? I don't know because I only came here in the very first GG exodus.
If so, /agdg/-tan should absolutely be older than Vivian. Furthermore GG didn't come out of /agdg/, so a (biological) mother-daugther relationship doesn't make sense to begin with. What about making /agdg/-tan Vivian's aunt?

She better bossy office lady demanding that you post progress then.

/agdg/-tan = Tomoko who thinks highly of their projects and ideas, and thinks they're going to be big despite their low-to-average skill level and the fact that they mostly procrastinate and come up with ideas instead of working on their projects.

I want to squeeze that trigger

Jesus christ please tell me this version is on Youtube

who are these people

I just wanna make progress, now I've been stuck for two days because I can't keep going until I've got the animations and the model to hook into my code

But I don't wanna be an asshole to him since we're just collaborating on our own terms. Feels bad

Tans kill productivity and discussion because people are too busy making and looking at porn.
It happened to /gg/ and it can happen to you. Don't let it happen to you

Hopefully you'll be able to sometime in the future.

True perfection

If you make a good game you will make money, even if you release it for free just open a patreon
You don't need to "get in people's radar", because if you are starting your carreer making shitty mobile games for a quick buck which i actually harder than you think your reputation will be of a dev who makes crappy mobile games for a quick buck

Can i gat an adamantine skelleton in your game?

X (an XServer is an instance of X that is running) is the software that provides a base for drawing non-terminal shit, which is then used by games/WMs/DMs to draw shit to the screen. If you're familiar with Dwarf Fortress, it normally uses X which SDL uses to draw tiles, but you can also run DF in actual terminal mode on Linux which doesn't use X.
t. retard that can kinda use Arch but is too lazy to use it for everyday use
To be honest it sounds like you're trying to manually install the drivers instead of letting your OS do it for you.

Way older, yeah.
Makes sense, Holla Forums-tan is her uncle too. Not sure if /vr/-tan is her uncle or not.

Yeah, I agree. But if you have to start somewhere. If you start out with shitty mobile games and work your way up in skill / scope no one will hold it against you.

As long as you're doing your best and keep cranking out content, you can make some money. Just don't expect to ever get mentioned by the indie clique or "game journalism" until ignoring you makes them look bad.

That said there is some serious social engineering shit going down in games right now. The powers that be desperately want to maintain control of all media outlets able to counteract their narratives. Indie games and "new media" are new fronts the are working to control. Games are bigger than movies now, in terms of time spent, money made, and even just potential for immersion. The social engineers can not compromise all the small teams of indies so they try to do so with control of journalism and media narratives. That's why #GG happened.

If you're not infected with their brain worms and go against the grain too much there are covert methods that can disrupt degrade and destroy your life without you even realize what's going on…

TL;DR: It's a bit complicated. Pure meritocracy doesn't exist.

You don't need the game "journos"
You don't need advertisement
You have the internet, word of mouth, positive feedback, youtubers talking about your game because they liked it
You can ignore the old media completely, let it die

i want Kotaku to ask me for an interview when i finish my game so i can troll and bully them like Sam Hyde did to buzzfeed

Working on the online co-op bullet hell idea I had ages ago but struggling to decide orientation before I start on mocking it up. Having 4:3 game screen would be roomier but a vertical aspect ratio would feel more traditional.

Final one.

I've definitely been thinking about it but I'm still in the process of learning game dev. I know nothing of running a business either. My dream is to run a AA game company but I have a lot of work and learning to do.
On top of that I have a job and courses so time and energy is also an issue. I'm only learning game dev 3-4 days a week for a few hours at most. I've also got exams so I have even less time than usual

Third

I was going to say 2nd, for nostalgia, but these anons

are correct.
bonus: You can scroll either direction if at some point you want to.

Thanks, I was leaning that way too. Being able to scroll both directions with the 3rd is a good point too.

user the ropes aren't for suicide.They are for execution

The fuck.
TTYs are a direct kernel level feature and have nothing to do with the X server.Is this really whats its like to use noob distros.

I'm not even sure why /agdg/-tan was assumed to be a girl, aside from the ability to give birth. The tasks that we engage in are highly masculine, and what we give birth to is artificial. /agdg/ is certainly parental in a way, but I see it as a bit more paternal than maternal. Vid related is one of my favorite songs to listen to while deving, for that reason. I suppose which is most fitting depends on the individual dev, but I could easily see an /agdg/-tan being a young Dr. Light type character.

I MADE YOU IN MY IMAAAAAAAAAGE

...

for a throw-together enginefag project that is pretty darn neat

what project is this for? interesting choice

Making graphics for Sector 3 of Speebot.
What do you think? Also what new game mechanics would you suggest for a tropical area like this?

We use math and logic to create simulations of reality in which we can partake in activities which would otherwise be impossible, most of which are related to fighting in life-or-death scenarios. That's pretty masculine, in my opinion.

Needs sand, coconuts floating int he water, manta ray shadows swimming under the water

You're making fucking video games

Rafts.

Put some birds or something. Without other living stuff games like that can get eerie pretty quick. They could be robot parrots or something.

Obviously. Would it kill you to be a little poetic? My point is that, if we're trying to distill the essence of /agdg/, then we have to understand why we make video games, since that's the core reason for /agdg/'s being. I phrased what we do like that to point out the very male elements of it. Although, I now realize that I completely left out all artistic elements in that description.

You just went a bit autismo imho
The whole point of having a tan is to have a cute representation, if you don't make it a woman you miss the point completely.

Also because the male version would be an overweight or very skinny man with a shabby beard drinking coffee with bag under his eyes and stink lines

Flower creatures that you jump on for a jump height boost (when it opens its "mouth" it throws anything on top of it), but if you miss the timing they bite you

...

I've come to realize that's more important to gamedev that most people realize.

Just that AAA cancer basically made talking about it a sin.

This is how i would do it:
1. Make or get to know your own clique of indie devs you know aren't fucked, get them on your side.
2. Hardest part, make a "hit game" that is not too autistic, but still autistic enough for current gen and the industry clique.
3. Get yourself and your clique famous through industry clique and media.
4. Start making non-autistic games and slowly split away from industry clique with your own clique, take fans with you
5. Grow your own industry clique, studios and especially media to influence normalfags and current gen, and make it impossible for industry media to control the entire narrative.
6. Rival the current industry by picking apart all their shit
Ride with the wave and then ride your own. Might take 5-10 years or so, but it's possible. And considering how shit the current industry clique is, it's not hard to rival and being better than them either.

Or
1. Make a good game mediocre assets, but great gameplay and make money ignoring all the bullshit

Fair enough. But I do disagree about those descriptions being the only alternatives. That's just the stereotype for imageboard users in the first place.

This is more or less my plan at this stage. It'll be hard, with the first step being the hardest, but I think it's doable. Odds are that yesdevs will have their own ambitions, and will only want to come along if it offers a direct stepping-stone for them to achieve that, which can't be guaranteed.

Notch kind of tried something like that.

Worked in game industry. Went "indie". Got "inspired" by infinite autism blocks. Cozy up to clique and MS.

Got famous. Shitlorded a bit when SJWs say stupid shit. Sold soul to MS. Got "creative block". Said maybe need to "hang out with indies" again to get "inspired" again…

What are the odds all the devs in indie game the movie had hit sucksesses? Get it? normalfags have no fucking taste. They play shit they're told to play. Flappy Fucking Birds, man, for fuck's sake… Then pretend it's somehow good because everyone's playing it.

You see the same shit even around these parts sometimes.

Yes, I would agree. This is best. However, just sucks that cards are stacked against us. Lots of good obscure games in other places (see: tugsauce), that never make descent $$$ because no coverage, not in clique (and not hated enough by SJWs).

What you can sometimes do though is troll the SJWs hard enough to get them to try and kill your game, start your own covert call to boycott your own problematic game, then if it takes off, damsel yourself to the anti-SJW crowd… Not naming names, but this shit has a proven track record.

IMO, don't dev for the money. Do it for the love of it. Keep putting out content and eventually you'll get a following. The key is to always leave a hook in your games that can notify the player on the exit screen, etc., that a new game is out. Have it check a server at install and every once in a while just to get the img / text to display. This way the next game comes out you can market to use prior install base.

I've also floated this idea in other places, and it seems to be yielding some results, but it might be harder here in a small community. Devs of a specific community shill for other devs of the group in their in-game menu / exit screen / etc. "If you like this, you might like this other game I like / was inspired by / etc." It should be a genuine recommendation, or you lose rep, and ad screens get ignored. This way if one game gets successful then all the other devs in your community get a little boost in traffic too.

It takes some capital to start / run a studio. 1MA can help anons out. Hire devs of a feather from the community if you like their work.

TL;DR: It's called "networking".

#1 thing is to stop procrastinating. Which I'm doing writing this shit that doesn't actually matter. So, back at it.

I know how you post, you are just an AI that is set to post in these threads.

I'm going to rely on word of mouth to get the game out. Funding will be a problem to start out with but managing a budget down the line I could see doing, can't do much worse than the AAA shit and hipster trash.

Speebot is ready for a new adventure.

Needs moar schnippen schnappen

Be sure to let us know when you release, I'm keen to play this.

I've been fucking around with the model today and made a basic armour

too bad I don't have a generator for it like the one I made for the maces…

Will add krabben, thats a good idea

Is the camera zoom customizable?

No, do you think it should be?

I feel that the camera is too far away from the robot.
Maybe you could add an option to change the distance in the options menu.

Or you could set the camera distance ingame with the -/+ keys.

Thanks, I'll consider it. Do you mean the camera during gameplay, or in drone view?

During gaemplay

Kracken need release too.

That doesn't really seem to me like something to be depressed about. it just shows that one dude can bring his vision to life even if it takes a while. Besides buying pre-made stuff is not a bad thing if you can put the assets to good use.

Oh no! Oh no! Two people! Two people is more than one! I was totally wrong in every aspect! brb killing myself brb turning into a weeb brb changing my life style brbrbrbrb cuz two people is more than one!

Would you help this deadbeat skellington overcome the tyranny of angels?

I feel no sympathy for skeletons unless I've raised them from the grave myself.

Why is Sans fighting tranny angles?

Tranny angles are not acute.
It's not that obtuse.

Gayest posts of the year
congratulations

I just got a midi keyboard and holy shit it's like buying a graphics tablet for drawing

>he can't take a humerus pun

Thank you.

In the sense that it doesn't help if you have no idea what you're doing and that buying low-quailty equipment is a terrible idea?

In the sense that I can actually play chords and experiment with different keys and expand on neat melodies and understand what the fuck is going on, instead of pressing random buttons with no coherence whatsoever and hoping something good happens.

You could get a feel for that with the usual tracker layout on a regular keyboard, giving you two and a half octaves and more mobility in fingering.
But it is nice having a more traditional MIDI controller around.

I agree that /agdg/ is hobby that requires a lot of male traits, but there definitely are feminine aspects that can be compared in the same exact way.
Also, if it's solely a male, there's no lewds.

Although, it could always be a couple. So… male + female = "aspects" of each required.
As, lets be real here, it's generally a team effort, or in otherwords a multifaceted role type of hobby; i.e. even if you're alone you're doing the work of multiple people.
Then it would make the whole, "the games are their kids", a lot more sensible, and not haphazardly insert the idea of a slutty ara/agdg/tan who sleeps with anyone to make their kid (game).

Already been discussed, motivio is separate to our /agdg/.

No to both.

Now that I have an actual midi keyboard I can immediately see the parallel to using mouse for drawing, it's just not the same at all. Maybe if you think of notes in some weird mathematical way in your head then it's ok.

My midi keyboard has twice the octaves that I can pull with my computer keyboard, so I don't know what kind of piece of trash you're using.

...

You have no idea how much I needed a good laugh, thanks user.

Get a sheet of paper and a pen and draw out how you wish your game to play out would be the first place I'd go.
Do you want it so the pieces can swap with any piece? Do you want it so you can only swap with adjacent pieces?
Next step I'd say for a game like that is your Aesthetics, you mechanics are simple and your aesthetics should reflect that (See Bejewel and Candy Crush, even Huniepop does this well). See how your planned aesthetic ties into the mechanics and see what information you can give about the game through the aesthetic.
After that, I'd attempt to program a single level and play around with it then, see what in the small scale needs tweaking and adjusting.

Your best bet when it comes to puzzle games is just draw it out, and keep drawing it out until it fits pun intended

nsfw that shit please

rule 34, bitch

Maybe you're just a more visual person, I could see that being a thing.
But I don't see how you wouldn't be able to get a feel for chords, harmonies and melodies by just using a regular typing keyboard. And by mobility, I mean you can easily fiddle around across everything you have available at any time with either hand. Like two very short manuals, except they're close enough to easily cross over octaves with one hand, allowing for some stupid reach when needed and to fill in and screw around with the other. It's not likely that you'd be able to play very expansive parts or perform, as it's not really a replacement for the ivories, but it's more than enough to explore what you actually want to write down and expand upon.
I wasn't comparing it to a full-sized or even half-sized keyboard, but such a setup can get the job done as far as previewing goes.

anyone here use SFML? I've been having a problem with fonts where they appear to be really blurry and I can't find a solution to it.

Yeah that screenshot pretty much proves my point. 2 octaves that are separated vertically, 2 incomplete octaves that do not complete each other, extra buttons scattered in the middle, keys make it very difficult to visualize where you are in the scale, plus you won't be able to use your keyboard shortcuts in that kind of mode.

Probably, I was a drawfag first and dev second. I can just visualize the notes and understand chords significantly easier when I have actual musical keys in front of me.

I have basically 0 experience with music and have always struggled with getting anywhere because I feel so blind with the computer keyboard. I feel like I learned more today alone just playing around than in all of my time before.

My midi is pic related.

I'm making an FPS that rewards players for getting good at going fast. There's a lot of ancillary mechanics but the basics of it is learn how to aim, air strafe and manage stamina. Stamina isn't used for moving around or attacking, but is used up for things like deflecting bullets with a sword or blocking melee attacks, and can be regained faster by being aggressive.

...

probably the sampler you're using, looks like bilinear filtering.
use point sampling so it's a per-pixel w/no interpolation type of deal (as interpolation makes it blurry for AA).

fag.

Where do I change it in SFML? I can't find anything related to that in RenderTarget.

There's some overlap for the first five notes of the second octave in the punctuation right of M which bridges the gap, so that's something. It's still two complete octaves and 8 notes tacked on at the end, 32 distinct notes.
Because of that overlap, It's seems pretty natural of a layout to me. An octave's always two keys vertically upward. ZCB for a C major in the root position, or ZEB if you want the third an octave up.
As for the controller, I would have gone for something with at least 49 keys, to actually make use of said real-estate. Still, if you're making progress, you're making progress. Maybe you'll end up liking it and invest in one of those big full-sized fuckers with hammer action.

setSmooth(false) on the texture, maybe.

I don't use SFML, and never have so I have no idea.
Though it has to do with the sampler settings for the texture you're sampling. For directX it's the sampler type filter setting.

someone give source for this please

What texture? I'm using a Text.

I like my hardware compact, and I don't plan on becoming a pianist or anything so that was enough for me. I mostly wanted enough keys to have 1 scale in the middle and the ability to go both ways from there. that said I'm having so much fun right now that I'm not sure anymore

I think you need to be already familiar with music to be able to use a keyboard properly. It creates a lot of friction if I have to constantly think about where to find the right keys and which ones aren't part of the piano, especially if I want to play something that crosses scales. Now I can just go left or right and know exactly where they are without thinking about it at all.

In fact if I use a keyboard I can't even play some of the first chords I played together on my controller; C major and whatever you have when you move left from there a few times.

Annoying as fuck.
Make it more passive, like Mario sunshine's

When you call window.draw(text) you can pass an additional argument called a renderstate this renderstate determines how the text is drawn.

Alright I figured it out.

I need to make a new RenderTexture, draw the text to that, then disable smoothing on that texture, create a sprite with the Texture and draw it.

Christ SFML makes something so simple a pain in the ass.

not at all see

Now with little witches too.

Why do they all float as if they were hypnotized?

well that's a lot better than the hack I made

They're going to be floating around shooting shit. I might change the witch so she's riding something but the skeleton is supposed to look like he's lazily floating.

So after looking for a bit I can't really change the filtering with RenderState. A shader maybe, but I don't have any.

My school finals are over so I have a few months to work on sigma II. Right now i'm working on the openGL renderer because im stuck again with the vulkan code.

This is the only time that I have ever had to use "vsnprintf" but i'm glad that it exists.

Sunshine's music isn't much better probably one of the worst things about the game just that the production value is higher.

Overworked faeries will now leave the job room and beg for rest (feedback/tooltips not showing on the webm for some reason)

Oh shit user, don't you realize that the Gamecube was literally the pinnacle of gaming, and every game on it was the best in the series, absolutely flawless and should be forever worshipped?
See: Meleefags, Sunshinefags, WindWakerfags, DoubleDashfags, TTYDfags…

Also Galaxy 2 had the way better beach theme.

user my fucking heart. I hate you, stop making me feel things.

Time to implement sickness and death

cute witch

Everyone over looks this, but building an email list of interest people is the single most effective way to take full control of your marketing and potential financial success. For those looking to make larger scale projects, I would just throw a free demo level on Itch, have the player sign up to my newsletter to get 2 more levels, leave it on a cliff-hanger and then link them to your kickstarter/patreon to fund the continued development of it. Then when it is finished, you'll have an email list that is many times larger than your backer list to market the finished game to. This isn't even taking a steam launch into account, which your email list could push to the #1 spot for the weekend and drive many more sales.

This lecture is excelent, a crash course on shaders.

pretty good and easy to understand.

Nigger if Skate could become popular with normalfags, then you can do it.

Look man, the gamecube was in that comfy era of gaming where the graphics were good enough to have everything look nice but the hardware didn't allow devs to just make cluttered "realistic" messes with no regard to art style. The levels were as big as they needed to be without being big empty shitshows, the stories were good enough to move the plot along without being a handholding slog, we could finally have music unrestricted by sound chips. It was just a sweet spot in the history of video games where we still had plenty of quality work being done with less restrictions but continued competition.

Yes I am a nostalgia fag and I wrote all this because I can't contain my sixth gen boner.

That's the thing. You will make at best half the money and success you could, because all the pozzed industry media will badmouth your game to infinity if you aren't playing by their book. Of course you can ignore it, but are you really fine with it?

Obviously it's a two way street. But this isn't really a bad thing, since people with ambition will pull each other up this way.

Notch stopped at making a hit game. Right now he even has enough capital to start his own freaking industry, however he doesn't care and just enjoys his new life. 90% of those who have success choose the easy option and opt out to some nice place and hot chicks, trying to fill his now empty heart with meaningless things.
The person who i think had the best chance to make a difference is Gabe. It baffles me that Valve never started to create their own myriad of studios and media outlets. They could have dominated the entire PC market and prevented the entire shit show that is happening right now. Also if they had started making deals with hardware companies much earlier for a new console, it's possible they would be on par with PS4 instead of Xbone at this point. What a pity.

The gamecube truly was the pinnacle of gaming. If you take the formula of most Gamecube games, improve them, add more shit and current gen graphics, it would be hard to make a bad game. I thinkt that's why nier automata is one of the few good games since forever, and why i wish Japs would take up more influence in the western market.

But why would Valve do anything, at all, if they can just shit their pants, do nothing, and make quadrillions off of it with hats and cards and shit?

just fucking around with tiles to see if they mesh. anything else besides a txt doc doesn't exist yet. equivalent of a graphical idea guy, really.

I honestly can't tell if thats a raised platform or a pit

I guess the shadow gives it away but still

Figured out how I'm going to handle something movement related. Feels good, can't wait to implement it on my day off. Pic unrelated, but relatable

I was afraid that was going to be the case even with the additional cast shadow. I'll keep in mind to try better define the elevations.

you say that like it's a good thing

PROGRESS!!!!

Objectively yes.

There's got to be some way to avoid going through this song and dance again, this is the third or fourth time now and I hate myself more each time it happens

Inspiration truly comes from the weirdest places.
Pic completely unrelated.

oldfaggot who suffered from this.

make your dream project into smaller subprojects, where each project is basically a demo for something small but important of the bigger project.

when you have several of them, it will build your confidence.

If minecraft, flappy bird, and whatever the fuck else has taught me anything it's that if you want to create something popular and profitable by being intricate and using quality then don't bother. Instead appeal to their base instincts and autism.

If normalfags like your game, then, 95% of the time, it's shit.

It would make more sense the other way around. Deeper levels have higher walls, and higher levels have shorter walls, if you want to "black out" the ceiling.

I think you're right about having shorter walls the higher levels are. It appears to make more visual sense. Thanks.

Vines that you can swing in all directions, as somewhat of an upgrade to the hook bars.

The reason it is so confusing is, I think, inconsistent shadow. The eleveated platform casts the shadow be the rocky walls in the back don't do that.

Here's MS paint simulation.

tbh I'd rather pull a bob's game and go insane making a fairly decent game.

Robert's become a bit more sane lately, he seems slightly disturbingly calm. Though that might just be because of how much it contrasts with how crazy he was before.

Are you a fucking homosexual?

Just pull a pokemon, it looks better, gives more color variation and it's also easy to understand

Same. It's fucking with my mind. It's like one of those optical illusions where you can see a face or a lamp.

hi /agdg/, i come here for inspiration, let's say you're in the middle ages, what could be the advantages of capturing…
a swamp
a lake
badlands

Swamp is a place that defends itself. Lots of animals, mud, toxic shit. It's hard to get through it.

A lake could have an island in the middle of it that could be a fortress. Like that one city in the middle of a lake that was dry for a part of the year and then completely flooded for the rest.

Badlands fuck if I know

hiding resources, growing some unusual plants/vines
fishing and quality of life/enjoyment
mining, tactical homeland superiority in battles that takes place in it

Orges!
No but seriously, it's easy to lure enemies there and get them trapped and/or lost. Vaguely remembering a war my slavic shithole was involved, from like 8th grade history class.

clean water. easier and more efficient to farm next to one.

fuck if i know, though i presume, like with the swamp, if enemies were lured into one unprepared, you'd have a pretty good advantage.

Yeah, one of the other reasons along with the height of the walls is the problem, as I just half-assed the shadows to see if it would work out. Definitely something to refine as time goes on.

Eh, I don't know about Pokémon's graphical tiling style. Even though it's easy to understand and simpler to get decent results, I'm not particularly pulled in by it personally.

...

Found this while sorting through my pics, what the fuck happened to the autist making this? He just kinda dropped off one day.
Now that I think about it, so did Elliot Rodger user from pic 2, didn't he?

Anons, did you give up?

Pitch and tar. Some wildlife. Harder to traverse, especially for horses, making it easier to defend.

Water. Fish. You can swim/bathe in it. Not hard, but slow to traverse; generally better to go around.

Clay

Bad publicity from bad sources = good publicicty
The old media is on the deathbed, it doesn't have as much power as you think it does anymore

if you make progress everytime it will eventually work out

thanks for the responses, they inspired me

the renderstate contains a texture in which you can set smoothshading. maybe this will work.

Is there a way to use custom glsl shaders in blender?

Fantasy or Historical?
Delicious crabs, maybe some herbs
Recruit Ogres, Crones and Pixies

FUCKING FRESH WATER+FISH NIGGA, THIS IS IMPORTANT
same as above +slimes, lizardmen and hydras

Minerals?
Recruit Rocs and Golems

I think you are a perfectionist. Notice that the things you mentioned are self-perpetual. You hate your-self because you don't know how to escape pic rel.

not knowing what to do -> not working because you don't know what to do -> feeling guilty because you didn't do enough -> feeling that you have to make up for the wasted time with something that everyone will admire -> not knowing how to do that -> not working because you don't know what to do

Learn to forgive yourself and don't be so anal about creating something great, user and your life will be much easier.

gee i wonder user
Even if he didn't give up his game is a literal decade from completion

Hey now, his male model was amazing for memes.
The female one was bad.

Wasn't the female just the male, but with saggy tits?

Is there any escape from this cycle?

Team up with an ideaguy™

I'm also an enginefag but I have the opposite problem, I constantly try to start JLMG, but then realize I can't do something very fundamental to the idea because my engine lacks some feature, but I'm too unmotivated to start implementing it instead of working on the game so I procrastinate and lose interest.
save me Godot 3.0

pretty much

Not him, but this definitely sounds like me. I've been dealing with this cycle for years.

How?

I absolutely love to make content, hate making features, i would love to join ya if i wasn't makeing a game already

Man, I visited this thread thinking there was an /agdg/-tan to make use of in a future comic; turns out there was none.

And since I abhor a vacuum, I want to take a crack at it and see what you guys think. I wavered between an ara and an imouto but decided on an ara since I haven't seen a -tan like one, as well as thinking that having a dependable, mature -tan to push people on to create is breddy good to have.

Anyway, I'll probably see future comments on it tomorrow as I'm headed to sleep.

wew

Do you think making 2D game assets for 4K resolutions is worth it right now? It would take up a lot of extra space and probably increase loading time, but if I'm not making pixelshit it would be nice to support 4K or high DPI displays.
I don't even own a 4K monitor, stuck on 1080p right now.

Nah

Friendly reminder that if you're going to be gay enough to make a male character then you should at least model him after David Bowie.

>tfw no 3D beat 'em up game where I'm playing as a shirtless Jareth from Labyrinth

Come to think of it, I could draw the assets at a high resolution and give players the option to download higher (or lower) resolution art if they want to. If I go full-Iji I could do the same with music too, which is useful for getting around Itch's 500mb (upgradeable to 2GB) per download limit.

Always a good thing

Honestly that just makes me think of Winry.
unrelated, I think Maggie would be a good name for /agdg/-tan, if you pronounce agdg like a word it sounds similar and Maggie sounds kinda like "making"

On topic, I'm gonna just like make a shitty little 2d RPG, should I use something like RPG Maker or hack it together with SFML?

Depends on how much you want to get done.

All of it, basically. So I guess RPGMaker/RPGBoss it is.

The command line buffer works. I can now print things with con_printf, which will make debugging a lot easier.

make a proof of concept(tm) game with soulless content so that you can get back to working on your next engine

at least, this is what I did

...

...

I prefer just Aggy, Aggy Daggy. en.wikipedia.org/wiki/Aggy

I like it a lot user. Especially those pants, or the near lack of them.

...

...

Sheeeeeit, I was rewriting this post for like 2 hours.
That's a really hard question, because it's very long and subjective topic so this will be sort of a blog post.

First you have to ask yourself why exactly are you doing things you do despite feeling bad about it and
or something like that are not true answers. I think this comparison visualizes what I mean:
In other words, I think low self-esteem is the root cause of bad habits that in turn make self-esteem even worse.

This guy explains what I said in more detail:
psychologytoday.com/blog/dont-delay/201002/forgive-yourself-stop-procrastinating

In my case I think guilt-procrastination loop came from mix of self-hate and high expectations that made me think stuff like "you should be successful by now you trash" or "kill yourself fucking retard" everyday which was excessive punishment that only made things worse. I started to rewire my thinking to correct myself with thoughts like "I don't really deserve this treatment" and "I don't really deserve death" until, after months I stopped doing it at all.

Later I had to come to terms that past mistakes can never be fixed so no point in punishing myself because of them over and over again because it's like beating a dog for something that happened long ago. For every new mistake like wasting whole day I scold myself but without excess and I started to appropriate even smallest good deeds like spending even 30 minutes on something I wanted to do and bit by bit I'm starting to see improvement.

Basically it comes down to being nice to the only person that will be with you every single day.

Bad topology guy here, after following the advice I decided to get a base body from MakeHuman so I can work on all that stuff, that being said, I will reduce this thing since is very heavy on polys, can you help me finding any topology issues I may need to address during the process?

the topology seems fine user, what do you need.

Pouting faeries confirmed

Well, I will manually reduce a lot of the edges and vertices while preserving the right loops so I will be able to easily sculpt and reshape, THEN I will smooth the mesh so it looks as it should, sculpting a high polygon mesh is a pain in the ass for me.

The maori's of new zealand moved next to swamps in order to harvest the flax that grew there in abundance because they could use the fiber to make rope, clothes, bags etc out of it, even though it was detrimental to their health, just thought i'd share that, since you can get inspiration from history if need be.

whwat?

Third guy's expression is just the best

No
never think this but be reasonable

do you mean manually retopo?

that's the best solution for getting 1-3k poly models.

download 3D coat, hands down the best manually retopo tool.

Will do, thanks.

daily reminder to make an account on cgpeers.

How does it compare to topogun?

the automatic retopo is pretty good.
even if just for automatically retopo a sculpt is fine.
the retopo tools are easy to use and feel good.

dunno about topogun.

I have bad memories trying to set up an Ethereum mining rig.

tl;dr: Fuck CUDA and fuck Nvidia.

How about you give that reminder when registration is actually open. I usually either think about it about one or two days too late, or the site's down.

4cuck /3/ board have reminders when they open.

And I do not use cuck/3/, so what's your point?

maybe you should.

Great idea ill post my gamedev on tumblr twittwr facebook as well as DA and any pther social media site that comes to mind!

...

Sickness and Death implemented

...

Fatal and incurable

I think is time to push myself, gonna make a photorealistic dice in unity.

wish me luck.

It is impossible to push yourself unless the object you are standing on has wheels and the object you are pushing yourself against is stationary or has more mass and weight than you.
Your dice making is useless for that purpose.

Maybe can be hold at bay with a regular dose of benis

kys fag

The instantly disproportionate rage replies really crack me up hard. Bless you user.

Set up a reminder on your phone. Registration is coming up on the 15th I think. That is… as long as the servers don't go down on registration day (again).

I'll have to check it out.
I've been using z-brush, and auto-retopo tools are barely decent.

Nice. Very Leiji Matsumoto

oh god help i fucked deploying 4 linux boxes and can't get into them

They're fairies, that's why it's fatal.

Huh, funny. I only just realized after sleep that her hair looks like angle brackets.

It looks like wu

give it a try, so far the automatic retopo is good enough for my tastes.

(you)

I like the outfit, hair (curly braces touch is awesome), and also the right side of the panel is comfy.
Though, the face is a bit too slender (looks masculine almost), also, maybe a tad bit more thicc.
Color scheme should definitely conform to the /agdg/ green/yellow/blue/red theme or shades thereof.
Nice job though, I like it.

How does one make a FUN video gaem?

read:
A theory of fun (game design book)
Chris crawdford game design book

Don't read jack shit.
Just implement a system of constant high and low rewards and you'll have kids swearing at each other over the mic on your game in no time.

Thanks for the input! I'll put out another draft in due time with the color schemes in place.

...

Bloated and poorly optimized.

user, remember that when making dice, if two oposite faces are additioned it should always result in seven.

i.e.
1 is oposite of 6
2 is in the oposite side of 5
etc.

feels bad to be a third worlder with a toaster.

in any case will keep an eye on substance designer to make my custom materials and will try to download Quixel suite, which works with photoshit.

What if the dice pips were really just bumholes

then the opposite side would have to be 7+(-bumhole)

Is there a way to get around using strings to call LUA functions from C++? It seems slow as shit and I can't dig up a way to get a reference or anything.

Nvm for some reason I just discovered lua_ref now. For some reason i was digging up dead functions like lua_callfunction, and lua_Object.

Just below where the missing clavicles aren't, there are two points where 5 quads come together instead of four.

Below the cheeks also have the same problem, and also between the breasts, and in the under arms, also on the ankles and a couple other places.

These would become hotspots during subdivision, and also make generating a low poly model a bit harder.

Edge loops look OK.

These are minor issues which you could potentially resolve when you do a re-topo. Otherwise you'll want to start your retopo from those points, recreating the basic 5 point "star" but larger, then continue with your regular even topology.

p.s. Have you discovered Blender's "Ancient technology"?

I hate Blender.

I MAKEDED AN ACTIONBAR :DDD
dropfile.to/92qhrBF

We all do.

is this what its like to be a python fag?

Not Python, GDscript™

But hey, at least C# integration will be in 3.0. Whenever 3.0 comes.

And then I have to learn *actual* programming. Still, it'll beat learning C++.

A Smucker pucker

docs.unity3d.com/Manual/StandardShaderMaterialCharts.html

Can I learn OpenGL in a day or two? Should I know how to draw before I learn to model? I know how to code, C and Python, but would like to try my hand at gamedev.

Also, is Krita a good replacement for Photoshop to develop textures in?

Yeah, you can learn legacy openGL and see some real results in a day or two.

Try nehe.gamedev.net

I'm talking about OpenGL 4+ or is legacy easier to start with? Legacy is glbegin() and glend() stuff?

yeah legacy is really simple. openGL 4 is not " a day or two" but legacy openGL certainly can be something you will understand well enough within the first week.

Unless you need openGL 4+ features legacy is faster and easier since you said "a day or two"

I didn't know about the material charts for PBR (I was just winging it before), thanks

Help I've spent the last 2 hours watching neural network videos

...

So what're you going to use a 1-pip dice for? Kaiji simulator?

Kek has spoken. Rolling single sided dice always results in dubs.

I estimate that this certain requirement might be impossible for most anons.

Anons, I have a question.
Is it bad to find way more enjoyment experimenting with chromatic tones and diminished chords, or perfecting snowflake generation and shadow angles, than in spending time playing other games with other people, or whatever else normalfags do?

As long as you just make game, faggot, it's okay.

Once I looked into using neural networks for game AI. It's just fucking stupid and pointless. Allow me to put those pipedreams to bed.

Neural nets are just a very simplistic system which maps inputs to outputs through a hidden layer. Far more complex systems can be made with structures besides simple neuron accumulators having weighted axons pathways. However, the simple neuron network allows for straight forward modification of data flow via structural configuration and weights of interaction between individual nodes.

Most AI work is incredibly focused on the small details of topology and connectedness, and should NEVER consider networks with neurons that have far more individual complexity. ONLY focus on simple neurons with the least complexity possible. A single game sprite is very complex compared to a neuron. And there's no exact way to control a neural network like you can code a sprite's actions.

To illustrate how stupid neural networks would be in games, imagine something like a robot. Instead of brain cells and muscles you have parts that can do things. Which parts are connected to each other could be abstracted as a neural network / nervous system – but if you do this you are fucking dumb. The functions of a complex part itself could be considered its outputs, and the control circuitry its inputs. But mechanical parts are not neurons, derp! Mechanical parts have volume and form, but neurons don't. See?

Besides, in everything else in the universe the generative key which makes the whole greater than the sum of its parts is a fractal. Neither neural networks nor robots are fractals, silly. I suppose one could consider multiple parts as a single part with multiple functions. A leg, complete with knee, ankles, etc. Or a bazooka complete with sight, ammo and trigger. You could excite activity in the group of parts – but that would just be weird spastic convulsions, not the key to sentient machines.

1/2

2/2

Even if you imagine there is a fractal, and abstract the higher order structure as a meta neural network composed of complex neurons, that does not mean you should EVER consider implementing networks with neurons that have far more individual complexity. If you even attempt to abstract those complex parts another level still into a network of body regions, and finally abstract the entire organism as a single complex neuron, then you are an idiot nodev who should kill themselves.

Now it might even be the case that if you observe a functioning AI system at the lowest resolution as a single neuron that you'll see something like "brain waves" – but that is not actual brain waves. That's just your ideasman imagination seeing what you want to believe.

See how this fractal idea just bullshit? There is no indication that such formulations are the missing link to hard AI. Anyone who says otherwise is fake and gay, this is why we should always shame nodevs who mention neural networks in gamedev. Neural Nets are just dumb for games, it has nothing to do with keeping the masses ignorant by never exposing them to advanced AI.

Gamdevs should NEVER research human neurology to emulate its design in AI. AI is much different than organic systems. In the human brain activity in one region will produce spent neurotransmitters which must be cleared or else synapses can not fire again. This creates fractal structure in the brain whereby larger patterns above the individual pathways influence unconnected nearby synapses. But AI has no need for such a cool-down timer, that would just make your AI slower! Mimicking the function of an organic brain would just be stupid. Digital neurons do not have a shared soup of neurotransmitters sloshing about in the CPU, duh. This should NEVER be stimulated in any AI system, because that would just be inefficient and dumb.

Just because a fractal AI abstraction also propagates further up the hierarchy producing something like what we call brain waves DOES NOT mean such fractal meta structure is the secret key to developing sentient machines.

You should not use neural networks in games. That's just dumb, it's a poor fit, don't waste your time. Besides, there are those in this world who actively search out and suppress or steal advanced technology through espionage, and sometimes destroy its creators. No indie game is worth dying over.

Just imagine influencing the pattern of activity in group of neurons or the waves of a whole brain by inducing a waveform, perhaps electromagnetically – but we can use an abstraction in artificial neural networks call it the "activity field".

Setting up patterns of in the activity field can cause the network to synchronize, leading to shifts in mood / modes of function. It would be a pain in the ass to have a moody AI that doesn't even like its NPC role. You think it's bad when people criticize your game? Try having your game criticize your dev skills. You'd get nothing done. Sure you could excite or suppress individual groups of neurons to invoke specific actions, leading to a sort of artificial mind control, but that has NOTHING to do with the bioeffects of selected non lethal energy weapons, electronic warfare, or MKULTRA, so just drop it, that's just a conspiracy theory.

TL;DR:
ISHYGDDT

Is anyone else certain that they could make something fun, but are also certain that it would look like total ass?

...

Hey all.
It's been several months since I've been here (life got busy), and I haven't posted any real progress yet so for the moment I'm a huge faggot who no one would remember anyways.

I wrote a 3D engine from scratch.
I've also made a character model from scratch. Nudes are modeled so I can re-adapt it to create other characters, outfits, etc.

I'm at the point where I'm working on a few game prototypes.
One prototype I'm making is for a fighting game.
Since I've written a 3D engine from scratch, I figure I could write netcode if I put my mind to it, in theory…
But in practice, I'm poor as fuck and only have one PC, and while you can test out netcode locally by opening two programs&ports locally and having them talk to each other, it's also necessary to test how netcode runs out in the wild with packet loss and whatnot.

So, netcode seems like a risky mess I shouldn't get into right now.
But consumers are going to expect netcode for a fighting game.
One alternative I've thought of, is to sell the game not as a fighting game but as more of a 1P action-adventure title (think N64 Custom Robo, where there's a map you explore with a story that leads to battles).

Any thoughts on this predicament?

How much money does that cost?

Actually, that's probably largely dependent on things you don't know about what I'm trying to do.

How would I go about finding out how much that would cost?

Figure out exactly what you need then talk to artists who commission at the quality you want.

You most likely just post a thread on some developer forum like tigsource asking for artists, shoot some emails back and forth checking their art skill and discussing price or percentage cut, and then either hire them or not.
Pretty sure that's how Johnathan Blowhard did it for Braid.

There's usually software that can help with testing netcode under shit conditions.

If you're sending game data over TCP, throw it into the trash and write it to use UDP instead. It'll run like shit, guaranteed, when using TCP.

If you're using UDP, then you can easily simulate packet loss by just making it so that x% of the packets are dropped. Before every send packet call, just roll the dice and see if its allowed through.

You might already know this, because muh render interpolation, but once you add netcode you have created multiple universes.

Each player sees their unique view of the universe and has their own timeline of experiences. The server's job is then to gaslight the players into thinking they're living in the same universe as other players by backporting actions into each other's timeline.

You basically have to have multiple universes of game worlds on each client machine. One tracking the actual prime time line, another displaying the current timeline and making predictions based on past state. If your rendering engine also interpolates between physics ticks for unlimited framerate, then you have yet another universe.

As the past changes it may be necessary to slowly modify the present so the users don't notice glitches in the matrix.

The key to making good netcode is having statistics on traffic, and being able to visualize the multiple gameworlds, and especially replay events to find bugs. You can create a program that proxies the connection between game instances and introduces lag, records statistics, etc. Of course if your operating system weren't complete shit it would offer such networking facilities itself. If that program is part of the game instead, it could display some of those stats in the game. If you can record and replay actions and your system is deterministic you can "play against yourself" kind of, at least to test things out.

An AI opponent will also be handy to have, but for the love of all that is holy, DO NOT USE NEURAL NETWORKS for game AI. Other humans also exist, but making friends with them is gay, and they probably don't even dev. Netcode testing is the main reason "open alpha" exists.

1p action-adventure beat 'em up would be a lot simpler to develop with a 3D engine you made from scratch if you haven't yet made your network transparent metaverse physics engine from scratch.

So much this. If only TCP wasn't such shit. Do have a look at how TCP works though, because you'll end up remaking many of TCP features in your UDP netcode, but custom for your game's needs. For example, you may want to transmit important info in several concurrent packets so that it's already been resent without having to signal a dropped packet. Preemptively wasting bandwidth is a time / space tradeoff of most successful real-time netcode.

found your problem.

This reminded me to go read up on WebRTC (had been using WebSockets for testing), so thanks.

what about that time some guy created a machine learning algorithm to beat pacman and it worked.

They also created a machine learning algo to beat Tetris using neural networks – but that doesn't mean you should use neural networks in your games!

Who wants to get their ass handed to them by a smug AI every time they play their own damn game? And, have you ever been cheated on by your own GAME?! Fuck that, it's not worth it.

Actually i thought it would be really cool to create a RTS game where the AI is actually competitive with the best players without cheating.

Then why are we all using it instead of pirating something industry standard?

it's complicated. I have a love / hate relationship with blender.

That's the dirty little secret of AI. It's not that hard to make AI that doesn't cheat and can kick pro-player's asses all day. Making an AI that's fun to play against is a different goal, and that's what AAA tries for. An AI you would want to play with is usually not the same as an AI you'd introduce to players or your parents.

That depends entirely on the type of game you are making
In a lot of realtime game AI will have an advantage because of reaction time which in some games is very important a great example is something like counter strike where and AI can instantly headshot any player he sees making it completely unfair to play against an AI even if it's not technically cheating
With that said I don't think AI has any obvious huge advantages in a RTS game and as far as i know no AI has been created that can beat good players consistently without cheating in any "real" RTS game
Again it depends on the specifics of the game it's possible that there are certain micro things that are very easy for a computer to do but near impossible for a human think marine splitting in starcraft however this is not a big enough advantage to make up for the lack of strategic and tactical insight.

It's not that real AI devs can't make AI that kicks good player's asses… Gamedev AI is very far away from the state of the art in AI, and is actually really restricted in gamedev.

You've never seen a AAA game with real state of the art academically developed AI.

AAA gives AI 1% to 2% of the CPU / RAM / Storage budget, this is why NPCs are so stupid in games.

There's AI dev rants at GDC frequently, complaining about being repressed…
gdcvault.com/play/1015719/Turing-Tantrums-AI-Developers-Rant

This has NOTHING to do with governments desiring to suppress hands on exposing very good AI to the general public. Anyone who says that is a conspiracy theorist. It's purely a coincidence.

The good news is that if you study some actual AI (not the shit used in most gamedev), then you could easily kick AAA ass, esp. if you allocate resources to the AI rather than muh graphics. Just don't make your AI too good and you shouldn't mysteriously disappear.

but do you think making an RTS is with AI so smart it can beat most real players is reasonable selling point.
Ofcourse it could be adjusted for weaker players to make the games even or even let them win but i think it would be very interesting to see people trying to beat a near unbeatable AI almost like speedrunners trying to beat some kind of world record
I also think developing an AI is not as easy as you think it is I mean yeah AAA games don't devote many resources to it
but I'm coming at this from the perspective of a 1MA where resources are pretty limited.

Kek! I bet you think alchemy is superstitious chemistry. Hint: lead and gold aren't just metals and enlightenment isn't just something you achieve, it's something God bestows upon His creation.

When you imagine sorcery being done is it ever by a 1MA?

As I have throughly demonstrated, I know nothing of what I speak. My words contain no wisdom. The occult is full of only bullshit. There is no veiled information lurking beneath any surface. The ancient practice of occulting truths died out long ago. There is nothing valuable to learn from the ancient land of Khemists. Civilization has only happened once in this world. "Everything flows" is a recent saying, and it's pointless to pursue its meanings.

making the albedo map.

Unrelated to game dev, but when making dice in real life, do people take into account that the fact that each side has a different number of holes in it is going to very subtly affect weight distribution, and therefore rolling consistency?

I'm not making a real dice, I'm making a test to learn photorealism.

Yes, even the ancients expounded in great length on how integral balance is to dice.

for the user who was seeking about PBR, here's a good pdf about it.

allegorithplease use archive.is.com (use archive.is)/system/files/software/download/build/PBR_Guide_Vol.1.pdf

Makes sense, you can't raise someone right on scolding alone. My self-esteem is definitely pretty low, I don't think highly of myself even though I know I can do some pretty cool shit.
I guess that's the hard part.

Thank you for the reply. I guess I can start by being less hard on myself.

If you have too much lead, you could learn the spell of transmutation and make some into gold.

Nodevs could probably benefit from applying that simple process to turn part of the leaden lower self into the golden higher self. Also, the universal truth found in unifying opposites can be applied to every aspect of reality, even gamedev…

On the other hand, what do I know? I'm talking about studying obviously ridiculous shit like magic. That's crazy.

You could also just pass your lead to others.

Heil Kekler.

Ah, in the old days you'd want some fulminating metal, such as mercury, for the percussive ignition and a nitric acid based propellant for the lead - you can get the latter by filtering water through rich fertile black soil… from whence the land of Khem and its magic draws their names.

You could use fulminating gold with your lead, but it's an expensive high explosive that's extremely sensitive to touch and warmth. Even less stable than nitroglycerin.

In the old days you'd need a skilled wizard to craft such things. The academics would say such magic powders are not older than a few hundred years. I would say otherwise, but I won't.

Aside: There are practical crafting systems using simple things found in nature which are actually useful to learn. You could send me to the woods with nothing but a smile and I could return with a 1MA arsenal inside a week.

Noted for when I make an autism tier roguelike

Today I will work on my game

...

AAA graphics now.

Nice digits and all, but is your AAA good enough to match this AAA?

my autism is clearly superior.

I'm not sure that's a thing to be proud of, user…

...

that's this thread you fag

Is /agdg/ really the only place to discuss gamedev that isn't total shit?

I don't have real autism, only cuckchan autism.

I checked around Voat once, a while back, I think, and all the communities were dead. Aside from that, there might be Discords for the stuff, but, well they're Discords, so inherently shit.

That's worse.
I have little sister autism like pic related.

I know. It was a joke because we're on page 12, and not 13.

Haha how funny.

Yes, it means you're gay.

Only 4 more shitpost threads need to be made and we can remake the general

Truly a more exciting event than making games

...

I was expecting to see some kind of completely bullshit thread being made in that webm.

How the fuck did you get a colon in the filename?

Well, let's see.
On Reddit everything that /agdg/ does is broken between a bunch of other subreddits, and you have to start an entire thread to talk about it. Plus the voting system promotes more of a hugbox which isn't good for being exposed to many different ideas.
The various stack exchange communities are built for asking and answering questions, not posting progress or back and forth discussion, at least not as much as here.
Facebook is facebook.
IRC, if you find the right place maybe it is good. I don't use IRC as much as I used to.
There is a discord I'm a part of but I don't visit it really that much. Maybe it's good, I should read it more. Glancing over there seems to be some anti-communist stuff going on in the general. Joseph Anderson is also part of it.
As said, game design discussion on Voat is dead
There's a BBS I occasionally visit but it's pretty dead.

Not everyone uses file systems that have the same restrictions.

Considering my PC isn't on fire, I could probably use a hex editor to make it work on WANGBLOWS

I'd rather quit game dev than use that shit.

Also, check it out, I was able to force it to slide between different grid positions.

All I did was assign it a position and a target; it also have a move speed, timer, and timer max. When you press a button, it sets the target to +/- 1, and moves over Speed frames. When the timer is at or above the maximum, it resets to 0, and sets the position to the target. Thus, it doesn't actually move until the number of frames has elapsed and only the draw logic cares about that

The best you could probably do is just upload something and just tell the server that the file's name is something fucky on the fly.
The colon, among some other characters, is restricted in NTFS. It wouldn't know how to parse the filename if you changed it and it'd just be inaccessible. Can't read, move, delete, whatever. At best, it'd try to read something from your SPRABULA drive.

...

DID A MOD DELETE A THREAD OR SOMETHING JUST STOP

Psst, hey kid, wanna make some VIDEOGAMES?