/agdg/ + /vm/ Thread

Factorio is made by "game engineers".

It's very possible to do that, but you do need to dedicate more time into learning the craft.

What's "possible" supposed to mean?

I do html5 games primarily because it's fun, and the biggest concern IMO is the fact that if you host it on a website, it can't be too big. Your graphics and sounds have to be relatively small in filesize, because the player has to download them every time they play. You'd also need to be able to host it in the first place.

You could provide it as a downloadable game, but if you run a html game from the desktop by opening the .html file, your browser may throw security errors at your face, especially Chrome and IE.

If you're talking about profits, then you should probably look into making a multiplayer game like diep.io.

Any suggestions on how to make a sweet ass handguard? My current attempts are weak.

Anyone have a good source for studying spears and body movement/stances while using them?

upper doesnt seem like it's remotely to scale with the barrel, are you modeling it based on a perspective reference image or are you modeling it to real world measurements like you should be doing?

also why use a short handguard in the first place with a >16" barrel on it? the literal only people who do that are americans trying to comply with silly atf rules, if you're a soldier or a cop or whatever you're not going to have a fucking atf compliant barrel on a shorty

Oh fuck there was a new thread?

Here, I'll just post this here too

I'm gonna allow it if it becomes a legitimate strat, if only because that's fucking rad.

There will never be enough time to do what needs to be done. This causes eternal stress and I love being stressed, pushes me to do shit. Gotta learn to just do what you've gotta do to the best of your abilities.

...

Maybe make the sword look like it's stuck in the ground when it's landed?

Yeah, when I get some proper sprites, the sword's gonna spin through the air and stick in the ground because that's fucking RAD.

why even bother with online, it's a pain

It should be, this is the whole thing. Stock, mag, sights and muzzle brake are in different files.

Would going with Gameboy Advance resolution/constraints be too much of a lolretro thing?

Suffer for the players user.

Is texturing user here? I need a few appliances that need work.

depends on the game id say. what are you making?

Speed-running platformer type of thing.

without seeing exactly what the gameplay you have in mind, I'd say it's not bad idea. When everything is going fast, the eye doesn't really focus on the resolution as much as the action. If you can keep the framerates high (low res helps), and make some cool art, it should be fine.
I guess it also depends on what kind of screen you're intending it to be viewed on, but I think low res can be nice if done well.
you already have an art style picked?

Megaman Zero felt very limited by its screen size to me.

Not really sure yet, I'll probably start mocking stuff up soon. Intending on using 3D->sprite workflow with indexed colour painted stuff for environments, since I've had okay results in the past with those.

Yeah if resolution ends up being a big concern I guess I can fudge it a bit for the sake of usability rather than keeping myself constrained to GBA size.

Right now I'm at the point when I need some sort of animation (mostly for skills) to be able to see more clearly what is going on in the game. So I ripped sprites from Kamidori Alchemy Meister, but there are very few particle sheets there. They were using .mpg files for special effects, which are unusable (no transparency when converted to images).

So, I would like to rip some other game, preferably with tools provided on this website asmodean.reverse.net/pages/tools_index.html but i have no idea which one of those games on the list (most of them seem to be visual novels)… When I search for images or videos from those games, all I get is porn scenes. So, can someone check the list and tell me which game should have what I need?

Or link me to some place where I can download already ripped stuff, that would also be great.

BOXES

Lolretro being shit doesn't really have so much to do with the particular constraints as whether you actually try to get the most out of them.
That is to say, your game doesn't have to look like Aria of Sorrow or whatever, but it shouldn't be full of shit sprites or lack detail.

Stop using shitty proprietary software

install blendoo

*gnu+blendoo

I'd just like to interject for a moment. What you’re referring to as Blendoo, is in fact, GNU/Blendoo, or as I’ve recently taken to calling it, GNU plus Blendoo, Blendoo is not an program unto itself, but rather another free component of a fully functioning GNU system made useful by the GNU mesa, nouveau and vital system components comprising a full program as defined by OpenGL.
Many computer users run a modified version of the GNU system every day, without realizing it. Through a peculiar turn of events, the version of GNU which is widely used today is often called “Blendoo”, and many of its users are not aware that it is basically the GNU system, developed by the GNU Project. There really is a Blendoo, and these people are using it, but it is just a part of the system they use.
Blendoo is the program: the OpenGL in the system that allocates the machine’s graphics resources to the other programs that you run. The OpenGL is an essential part of an operating system, but useless by itself; it can only function in the context of a complete operating system. Blendoo is normally used in combination with the GNU operating system: the whole system is basically GNU with Blendoo added, or GNU/Blendoo. All the so-called “Blendoo” distributions are really distributions of GNU/Blendoo.

Anyone know what these nodes are in the material setup to this tutorial?

wiki.unrealengine.com/Skybox_from_DDS_cubemap

I wish the nodes told you at a glance exactly what they were, but left to right: Constant3Vector, ReflectionVectorWS, TextureSampleParameterCube.

From left to right
Constant 3Vector, ReflectionVectorWS, TextureSampleParamaterCube

Here is the documentation.
docs.unrealengine.com/latest/INT/Engine/Rendering/Materials/ExpressionReference/

Damn it I was late.

Thanks

don't be afraid of being late but take pride on being there

I will never be enthusiastic fr something small in scope ;_;

...

Sure you want to make a game and not direct a movie or write a book?

Rip, there's only so much I can do on a toaster.

That's just a great way to make sure your game runs well on older machines. I'm not sure what kind of game you expect to be able to make anyway.

i'm stuck on story as well

okay faggots, not a burger, but I dont wanna spend years in college.
my life dream is to be a game programmer, so I want to put serious effort and do my own MIT challenge.
but I dunno what topics to study.
sure, there's the MIT challenge list of topics, but I want to know what are the mandatory list for a good game programmer to know.
So far I know I need to study:
Calculus.
Linear algebra.
Algorithms.
Clasical physics.
Optimization.
networking.
C.
C++
computer graphics.
Data structures.
So far these are basic shit, but what else do I need to reach carmack level?

Great job copypasting your shit OP

make Commander Keen
then Wolfenstein 3D
…Doom
And then Quake

rinse and repeat till you go to Quake 4 stage

i mean, if you're gonna study calculus and algebra, you mind as well just go to college, that shit's mandatory for anything IT related there

I don't really expect much, but I would like to have a nice 3D environment for my game, and have it run at least 30fps.

install gentoo and use emacs

You're forgetting Rage and sparse voxel octrees.

:::^^^⁾)))))

Anyone interested in testing out my Minetest mod?

I would also recommend studying discrete mathematics, calc and linear algebra really only come into play for 3D graphics or physics simulations. Discrete math is used everywhere in computer science and would help more with learning, optimizing, and implementing algorithms.

In fact, you should probably hit discrete math, algorithms, and algorithms the hardest.

algorithms, and data structures.

Why not get some Anons to do it? I used to see people put test builds up all the time when we were on halfchan, but not so much here.

That's the plan when it's more fleshed out and less of a prototype.

Bump.

Get anything done today?

What are some good modeling software's then.
Blender has the weirdest hotkeys.

Listened to dank retrowave and fixed a bug that caused my game to crash due to a stupid oversight I did when setting up my new AI system. That's what happens when you code while sleep deprived. I now have HTN planning working in UE4 seemingly perfectly.

Now I'm gonna see if I can get EQS to work efficiently with it (using c++) for movement code since EQS from blueprint is too slow to allow for lots of AI.

Next up from that is probably implementing melee weapons or grenades or working on replicating destruction and physics which I've been putting off.

I think you can change Blender's hotkeys to Maya's in user preferences. I've never tired it but it looks like it changes everything.

no coding, but I did do some planning.

I rewrote how my voxel system works a little bit ago, but I've already thought of several ways to make it better. While I don't want to optimize too much too early, I realize that if I don't set this up right from the beginning, it will be just that much harder to get it fixed down the line.

So I spent today planning on what features I want to add for this next version, taking the day to really plan it out so I don't have to rewrite it again so soon.

I put a revolver into the game, but I need to fix it so that more of it is on screen. Also I've finished the reloading animations.

I love it when the first thing I try works.

trips get

...

I completed my first Unity tutorial. Did the one with the ball rolling around collecting squares. Took a picture when I started.

Hooray for following dreams!

Just an FYI, many of Unity's officials are very old and out of date. So they won't work if you're coding with a newer version of the engine

...

Nope! I was super psyched to make progress today, but every time it looked like I was going to be able to, I got dragged off to do literally everything but make progress. Now it's the end of the day and I have nothing to show for it. I know I'm obligated as a human being to participate in things with other human beings, but I wish real life would fuck off. As if a cosmic entity was making a punchline, I was also interrupted while writing this post.

Yeah, their set up was different and was using MonoDevelop while I was using Visual Studio. Had to hunt down the hotkey for Unity API help and make a macro for it. Everything else pretty much worked fine though.

what do

Like base C++ templates? Google the msdn and hit up stack overflow if something stumps you. A technical article or two helps a lot as well

...

More Tesseract mapping tests and fiddling with the lighting settings to get the right look. Also looked for other engines that supported realtime GI out of the box, couldn't find anything but probably bloated RenderPipeline and nodocs Cryengine. I could always try implementing it in something like Panda3D or Urho3D but I'd rather focus on making vidya and adding neat shit than implementing radiance hints myself in another engine.

...

C++ has templates regardless of platform it's a language feature the fuck are you talking about

I've spent a couple hours working on a design document and I gotta say it makes me feel like I could actually pull this off. Once I get done hammering out the scope of the content I'll have to start thinking about how to implement all this.

I think he's referring to the msdn user.

Sorry some of us can't afford insane rigs that let us run UDK4 And Lumberyard.

Well, I finally got my camera view and controls up and running, at least to a degree that allows me to test shit out better, for my flying game. Physics are being a pain, though.

It was amusing seeing Unity start to flip out at high velocities. Next step is to make some sort of environment, so I'm not just flying around empty space without a clue of my bearings.

guy who asked about the axe thing from the last thread or so here,
I couldn't give out my thanks because I was on my phone and my computer was acting like a bitch, so thanks to those who clarified and help me.

I decided to go a bit more back and just making objects and such by looking at tutorials.
so far cups

Good choice going with the tutor4u tutorials. He always keeps things at a good pace for beginners and explains all his steps.

the very first video I started out with was Udemy's Blender tutorial. It unfortunately only had one volume and some parts were missing

other than that, I loved it- the lessons taught me hotkeys and how things worked etc.

I'm digging around data driven programming

I mean why do I have to keep like 40 different booleans and integers defining projectile behaviour inside every single instance when I could just keep x, y, velocity and direct it to a "driver" object that makes it work like normal bazooka shot or grenade or sheep or whatever.
And when I want to modify projectile behaviour on the fly I can make a new instance of a driver object.

Yesterday I did nothing but what I can only describe as game design, before I got distracted by draw thread. There's some parts of my game where I'm not sure how they would function, but now I'm much closer so something sensible. You could call it ideaguying, but I tend to actually do it in the game eventually.

I want to work on the wiring and/or space ship system in my Terraria-like next. What I was unsure about how the construction and control of the ship would work in practice, because I want everything to be player-built. I came up with a block that would take away control of your keyboard, and another that would control your mouse, and you could assign keyboard/mouse controls to a colored wire signal from it. You'd probably need to be able to acquire a pre-built ship somehow because I doubt majority of players would want to or be able to build one from scratch.

I also now have a better idea of how the actual electricity network could work, but I'll have to test it before I can know.

Next step is to learn smooth edges. This will highly improve your productivity. When making models for anything, you gotta imagine a poly count like a max budget you

Do people really do that? All my projectiles work on the same code and have the variables, just different values. It's really nice and easy.

well this looks better

multiple instances of the same projectile type will have the same values for each copy, so why not just keep them somewhere else and just refer to the templated data whenever you need to know how much the projectile weights or if it drops any clusters or whatever

The majority of their data isn't going to be the same. It would be more efficient to do what you're proposing but only barely. Not really worth the time to rework it when I have other stuff to do.

The sad thing about this is that the actual builds of UE4 run a lot better than Unity ones. It's just that the editor is a hog when compared to the Unity one.

What are your specs, user?

I sure as hell hope you didn't model the entire thing and used the mirror modifier.
When I started working with Blendoo, taking the time to look into how the modifiers work was well worth it.

You interested in custom enemy graphics for your mod?

I do, but that'd be fairly tough what with rotations and shit.

What's so hard about rendering a model from multiple angles for each animation frame?

For the inexperienced like myself, quite tough.

But I could use some sick as fuck sprites for at least my mastermind replacement, but I'd love to get some for an archvile, cyberdemon, and mancubus tier replacements too.

I'm still not sure about what the spider mastermind would be visually but he's definitely gonna be some kinda magic bastard that summons other monsters and shit.

The arcvhile replacement would be the BONE LORD who can shoot bones n shit.

The cyberdemon already has his mechanics and behavior in the mod, he shoots lasers and shit and stomps around, so I like the idea of him still being cybernetic/robotic.

and finally I plan on making the mancubus replacement into a really fat armored guy with a cannon. His death is gonna be the best because I'm gonna make him cause an earthquake when his fat ass keels over.

For my vehicles I could get away by just parenting the Mesh to a empty and the empty will be used for rotation. Then I did a animation render and packed them into a sprite sheet with texture packer as it was done the most quickly that way.

I remember there was a plugin for Blender that would handle the additional animation stuff as well but I dont know anymore where it is.

What is that? A skyrim mod for doom? Get out Todd.

but Fallout 4 and Skyrim is a pretty good quality game. Unmatched by other AAA game titles.

Do you see the Mountain? You can climb it.

Just give me like some concept art or something to go by, I'm going to be busy for the next few weeks though but I'd like to make a start in the next few days.

That's a pretty good technique, but I don't think I need to lighting to be consistent in the same way, I guess I will figure it out when I get there.

forum.zdoom.org/viewtopic.php?t=45155&p=743601

Is this what you were looking for? I guess that's specific to Doom though.

Thanks, that looks useful. I guess that should motivate me trying to make a atleast okayish looking infantry render.

I guess soon I have also a bleeding, burned, acidification animation and what else once I git gud.

Can anyone please tell me what I can do for a mere Android 4.4 tablet?

I'd really like to try making a platformer from scratch or an RPG with some unique or at least more uncommon elements programmed in. Plus crammed with extra content. I'm trying hard to understand and learn programming, but I don't understand why code has to have so much shit typed in just to flip around a string of numbers.

stop

No, it's Ghosts n Goblins, Magic sword, and those Dungeons and Dragons beat 'em up games, combined with Doom.

So you know, actually good games.

Oh dear. Let's just toss some spritesheets up here and I'll just describe a little more.

For the fat armored guy, I picture something like this, just with more armor, and packing a cannon of course.

the wizard that replaces the mastermind would probably be a typical evil wizard bastard, but if you have any ideas on how to make him look more intimidating or impressive, feel free to do so.

The cyberdemon replacement has to be some giant robot knight guy with a big horned helmet and a fucking arm cannon, players should get one look at this enemy and think "Oh fuck, he's the guy in charge"

and finally the bone lord, I picture some kinda commander skeleton, definitely with a cloak or something to signify he's higher ranking, he doesn't -have- to wield a staff but a weapon of some kind in his hand wouldn't be a bad idea, his attacks consist of shooting bones, bone magic, or firing a hookshot at you.

he's also gonna have a bone-nado spell. Because that's awesome.

If you somehow make -any- of these monsters, even just one, it'll be massively helpful.

Okay, I probably won't start tonight though I'm just about to get in bed, will probably start wednesday, I'm not doing anything on wednesday

The fact you're offering help is greatly appreciated, user. Thank you.

Don't worry, I'm probably going to dangle some of the assets in front of you to make you change your game like making your own god damn levels.

Unless you have no intention of making that change in which case, eh, It still looks good in my porftolio.

It's just a gameplay mod for now, to make it a standalone thing would be a massive undertaking, and I still have another three characters to make on top of finishing Blaz. Custom maps are pretty low on the todo list.

and at that point I'd sooner just make a proper standalone game in another engine.

Is it possible to make a level pack separate, that's kind of prepared for use with your mod?

does your mod work with strife?

It'd be fairly easy to make a GMOTA-themed mapset honestly. Pirate Doom does that, you can either play the campaign it comes with or use the weapons and monster replacements on any custom levelset.

As for strife compatibility, sorry, I only made it Doom compatible, and from what Term told me and the fresh hell he had to go through for Strife compatibility in Samsara, I'd rather not put myself through that.

I was just interested in seeing if it was just compatible out the box, I didn't think it would be.

not him but, you mean map pack that can be played without mods but when it is used with the mod it has additional shit and the like?

Well for me personally I would do this, a map pack that is actually optimized for a mod but with the usage of a "map loader" the map loader would contain additional actors with DoomID and the like so that it can be placed for maps.

I haven't tried it that method yet because I would need to make 1-2 maps to actually test it out this function.

you could try making a city builder that isn't a copy of clash of clans

I don't fucking like only having this to use either nor do I want to look like every other egotistic faggot out there who just wants to pander to retrofags for easy bucks when I don't even want money. I've long since decided the first ever game I'd want to make is a platformer just to see how I can make it my own with combat and puzzles being a major focus.

I want to start making a game now and I can't be picky. Something like this is the most I'd be capable of. I just want an actual human being to translate the most essential pieces you need to make objects. I figure I could at least draw lines then find a way to give them depth and properties, but I can't find anything like that.

I was hoping maybe if it was possible for a mobile device to do something like create objects, character models and environmental shapes in other programs then copy paste them over into whatever I have to use and flesh them out. Somehow. From there figure out just what I exactly need to code to make things at least work the way they're intended.

I don't want to make a little app. I want a full fledged game.

Making a platformer for a touch device is like making a hockey stick to a tennis player.

Kirby seems to have proven it can work. It's just not used to its full potential in either Curse. They're fun, but way too easy at parts and only focus on getting around with the occasional puzzle.

I want to expand on that type of gameplay.

You can do it, you'd just need to think outside the box

Yeah do that, those games were fun.

What if I wanted to remove the whole line mechanic and focus more on evolving abilities?

I spend the weekend trying to get Arcadia working ( REPL-ing live into the Unity editor with ClojureCLR )

Somewhere on the github one of the project leaders said the current master is completely broken.

And holy fuck - he is right about that.

I had the Editor dying on me, like every 10 minutes.

Even if i have to rewrite the whole damn thing and reverse engenier half of Untiy to get there - i won't give up.

The normal Unity workflow is just way too painful and i'm not free to use something else.

I'm starting to get the idea I should fuck around less and get more useful things done. Like spending an hour or two fucking around trying to optimize things that don't need to be optimized yet.

Or implement something that isn't loading .pngs into map data.

Refactoring, fucking around with collision and, on a more general note, getting a feel for how to best avoid unnecessary operations, I guess.

I wonder if avoiding things ever getting into or spawning inside of walls is a decent enough strategy, or if I should worry about how to get things out of walls if they do somehow get in there.

Have you finished any little apps?

No. I resigned myself to trying game creator apps to get my feet wet, but they were terrible and painfully limited to preset objects or in general scale.

I've at least already learned about APK and swf files. That has to go a long way towards making something. I found a highly rated pixel art program called Pixly I could potentially use to simply save files of everything, but I still don't know where I'd move everything before trying to code it properly.

I'm hoping I can do something simple enough like make a block that would be part of a stage then type a specific line of code to make it inflict damage to, shatter when hit hard enough by or just knock the player model away altogether.

how many of you guys remember voar.io?

do any of you think you could make a clone of it, though with a mechanic to prevent bots from spamming new ships?

maybe there could be a 1 second delay for every new ship, though not 1 second per player

well actually no it would be 1 second per player i guess, or maybe theres some other way to prevent botters from throwing out 3 ships a second

Making multiplayer games is a lot of work and requires money to run, and preventing bots is not just a matter of pressing a button to block them.

How do they monetize these games? I heard that slither.io was making obscene amounts of cash a day. is right but if you got lucky couldn't you theoretically start it off with a small amount of money and then keep feeding some of the profits back into more ads?

Garbage.

I'm back from not posting for about 2 weeks, I have some new content but I was not able to work on my game for about a week.

This is some new gameplay, what do you think?

Are you sure you want enemies to soak up that much damage?

no, i'm not, i'm still balancing the game. I'll make it easier to kill the enemies.

So, I'm working on making an open-world type of game.

Does this mean I just make one fuck-huge map? Is there anything in particular I should take into consideration (optimization and such)?

Depends on how your open world is structured. Which will be determined by the setting, and how you're going to be building the world (procedural, by-hand, mix of those, etc).
So… the most common methods are doing individual scenes connecting areas (easiest), or one huge map that's generated and/or loaded at runtime (more difficult), or maybe some hybrid of the above.

Again, depends on the above considerations.
You should probably think that stuff out… and then search up on what's the most suiting method based on that information.

Do you genuinely like it when the floor is ice? I assume you made the movement like that by choice and I find it hard to believe that anyone would enjoy that sort of movement. Minecraft also does it and I hate it there too, but maybe I'm the odd one.

Looks fucking rad. I know you've been doing mostly engine work so I'm pretty sure most of that is placeholder, otherwise I'd bitch about it. You do need to tighten up movement like the other user said. Quicker deceleration and higher top speed would improve that a lot.

The first of the in-dev HUD for Space Pirate. This'll be the "stylized" HUD, compared to the other HUD, which will be significantly more simplistic.

Not finished (the fonts need changing, the actual helmet graphics need changing, there needs to be stuff for a couple more features) but I'm trying to keep everything fairly streamlined without getting into information overload and "what the fuck does this mean".

Still, you got your health (bottom-left, with the colored lights giving an instant indicator of your general health level), jet fuel (red bar), ammo (bottom-right), keys (top-right), total yen collected (top).
The vital need-to-know information is all on the bottom, while the extraneous information is on the top.

I've been slacking lately. No new visuals to show, just been working on stuff under the hood.

I've successfully split up my code into a few pieces and am now working on a save/load system for all my voxel shit. I still got a lot of work to do, but at least I'm getting closer to actually programming the game rather than just the framework for it. Makes me feel optimistic, even though I'm just scratching the surface of what I need to do.

While I'm not the greatest programmer ever, I'm quite pleased with the progress I'm making, even if I feel like an idiot when making it.

Added a warp-dash that lets you dash through certain things and parsing map data. So threw together a test level in Tiled to test online with a pal he didn't dash so good by mashing the button, makes it look like he's jittering all about. Probably need to do some placeholder sprites so it's easier to see direction and feedback for dashing and shit.

Next up camera probably.

i hope i don't accidentally shit myself
at least i don't have any dragon dildos around me

Don't worry, skype doesn't have smell-sharing yet. Good luck.

I feel for you.
At least the GPU isn't as shitty as the one in the Laptop a friend of mine uses. He has some couple generations old Intel IGP on a mobile Pentium CPU. According to benchmarks it's even slower than yours. Here's the unfortunate thing (for you): It runs UE4 games, but struggles with even the simplest Unity ones. Satellite Reign for example. He can't set the settings low enough to even make the menus to stop lagging. He has tried various Unity games and reported similar issues. Meanwhile he tested Ben & Ed (he knows the developers personally) and it ran perfectly fine.
You should get a desktop, user. I know that it's easier said than done, but even a cheap, sub 300$ desktop with used parts would be orders of magnitude better.

Updated the rig a little. The original skeleton was made with a very far camera in mind, and didn't include twist bones or arm IK.

Thanks. I've made it less icy in earlier builds but I guess its still there…

The graphics are not placeholders, but some of the enemies are not completely implemented or balanced. I'm still figuring out what works.

I'll post some progress soon on new things that i'll be working on.

welp, finished my short interview
i think the dude liked me, but since i have very little experience with mobile development, he told me to try and convert one of my games to android, and make it relatively playable
guess i'll just take my rpg system and make some shitty zelda clone with somewhat playable controls

fuck

can someone give me a guide on the basics or fundamentals of 3d modeling?
I see alot of tutorials on how to make shit or use a software but I never see anything on what not to do etc.

also, anyone got any tutorial on how to send push notifications for android or other devices?
i can't find any tutorial on this that doesn't just tell me to use someone else's assets

Porn when?

Can anyone tell me how to make proper fucking forearm twist bones in Blender? I swear, it's fucking hopeless for me.

Looks great by the way.

No idea, but I can tell you how to make a shitty one that kind of works.

That already works better than what I have. Tell me more, please.

When I started to learn blender years ago, I read Blender: Noob to Pro. It's fucking massive (about 2000 pages) and contains anything you would need or want to know about blender. Here's a link to the free pdf:
upload.wikimedia.org/wikipedia/commons/2/20/BlenderDocumentation2.pdf

no clue if anyone here has bothered with this
i'm trying to get unity remote to work
i've installed all the necessary bullshit to get it running, built the apk, put it on the phone and installed
i run the app first, then turn on unity (supposedly restarting unity helps) and then i press play
if i do any kind of input on the phone nothing happens, but if i change something in the editor, it also changes for the phone
and when i stop playing in the editor, the phone's app controls fine
what am i doing wrong

youtube.com/user/cgboorman/videos

This guy covers everything. EVERYTHING.

ah fuck me nevermind
unity remote "works" except that in a normal mobile application the "back" button is the same as KeyCode.Escape, but while running unity remote i have no idea what the back button is mapped to, if anything

Does it cover particles?

DID YOU MAKE PROGRESS ON YOUR GAME THIS WEEKEND user?

I was out of town and couldn't get anything more done than just writing things down in a .txt file.
I'm working on it now, though.

I read over 200 pages of programming books over the last 2 days, hope that counts

i made a shitty rigidbody controller for a shitty mobile game
now i need to make it not shit

This shit takes way too long to do.

I was working, so not much, but I did get my combat menus working mostly. Still more to do, but the basics are all there.

I work on a game 5 days per week as a job, I need time off of gamedev in the weekend.
Just a few more weeks working with this piece of shit codebase thrown together by retards with zero knowledge of UE4, then I can finally stop and rewrite it from the ground up with a small group of programmers underneath me.
Today I figured out which code written by the most retarded of the previous idiots caused a crash whenever a client killed anyone, now it doesn't crash but does bug out. Still, that's progress when I'd been breaking my head on this bug for 2 weeks already.

no

I was thinking of having it sort of like WindWaker with airships/craft instead of boats, and islands in the sky, that sort of thing. Air combat would be a main gameplay feature (I've already got a basic flight model in place and working). For the sort of story setting I'm going for, procedural wouldn't work in my favor.

So the only two options I can think of are either one fuck huge map, or individual sectors connected like a grid that load in when you get close enough (I think that's how WindWaker did it).

Thanks for the reply, though. I know I'll have to do plenty of research into whatever method I decide upon, I just wanted to check here as well for good measure.

Unless writing a document I will use to pitch my project to a potential partner counts, no. I learned for an exam I have tomorrow, early in the morning.
Truth be told, I don't remember the last time I felt so well prepared for an exam. I literally spend the last days learning for the next one, because I feel that gud.

Ideaguying all day every day. I'm trying to figure out how my game would work in practice with the plans I have in mind, and what kind of things I should consider.

It's either a good sign when I'm having fun trying to design a ship in a certain way before it's even in the game, or a bad sign when I'm having more fun drawing concepts than programming.

I got my aircraft to fly, albeit arcade-y. At least now it flies in the direction I want, up to a max speed I want, and without drifting all over the fucking place.

play skies of arcadia for ideas

A glitch in the gamecube version caused the random encounter rates to plummet, and people were really upset by this, so learn from their mistakes and make enemies appear every 2-3 steps

planning is progress in its earliest stages, acceptable.

if you're going to be using it to make better games, then sure, it counts.

making something shitty is the first step to making something not-shitty, good job.

You don't want to be a nodev now do you user?

A step in the right direction, that's good. Understandable if you are busy.

Nice job bug crushing. Anything interesting you have planned for the rewrite?

YOU HAD ONE JOB user.

That's progress, lots of games need multiple people working on them, so its an essential step in my opinion. Good luck with the potential partner, I hope it works out. Good luck on your exam too.

If you are having that much fun designing before its even programmed, that's a really great sign. When do you think you will actually start coding it?

Nice dude, that's a good step. How realistic do you want it to fly? Will you write the physics for it?

Good work guys, keep it up. These games ain't gunna make themselves.

Y-yeah, I did a bunch of stuff that had to be done but had no actual effect on the game yet

Well, I'm well aware of how empty most open world games tend to be, and am gunna try to actively avoid that.

I'm trying to make the air combat feel like War Thunder, with a setting similar to Last Exile or Storm Hawks, and with technology reminiscent of late-WW2 (early jet tech). It's turning out to be a lot easier said than done, but it's making for a damn fine way of learning how to do all the dev shit (scripting, modeling, textures, mapping, etc.).

It's physics-based already, but I've eliminated momentum for the most part (I'm hoping to make it semi-realistic: with stalls, high velocities ripping your plane apart, and engine overheating) but arcadey enough for it to be fun to fly and cause all that shit to happen on purpose.

Basically, I just wanted to make a game around flying aircraft really fucking fast. Now that I've gotten things to a working level, it's been a blast making improvements.

That's important, i'll count it.

That's cool man, how long have you been working on it?

I'm at a similar place in that I'm just working on the underlying pieces of the game before I can really start making the actual game. Slowly but surely though, it all comes together.

Not necessarily making a game but I'm almost finished my current project.

that's pretty cool user, whatcha gunna use it for?

What if all of my progress was from this weekend?

Then why are you posting it here

At the moment my game is at a state of total disrepair because I was refactoring core functions but never finished them. I tried to build a mockup of a ship in-game to get a feel for it, but it bugged out. I've been contemplating on just starting over.

I'm making Warhammer models for people to use in Table Top Simulator

I figured this would be a good a place as any for people developing stuff for games, given the title.

Do you have a better design in mind for how to reprogram your work? No shame in starting over, especially if you have a better design in mind. If its any motivation, your sketches seem pretty interesting and from what I intuit the gameplay to be like, it does look like it could be a lot of fun.

Bloody hell that looks gorgeous.

A couple days each week, over the course of a couple weeks. I'm using Unity, so not exactly working "from the ground up" per se; it does a good job of testing concepts quickly though.

I still have an old version of 3ds Max (it expires in 19 days), I've tried blender but hate it.

Right now I need to make some sort of basic level so I'm not flying around empty space.
Pic related is a general idea for what I'm gunna make, albeit not nearly as huge, just a small port with landing strip with that kind of layout floating in the air (haven't decided upon how it's floating).

Definitely. I've learned some very useful programming related things while working on other projects, and some of the current code looks completely retarded in comparison with how I would do it now. I've been working on this on/off since 2014 and have redone everything several times, I think this was actually the first game I ever started.

Thanks, I sure hope it'll be. I'm making this game because nobody else will.

thanks a ton for these
I just read a few pages of noob to pro and its fantastic

Same guy as here
I cant get the last two steps to the tutorial. Am i selecting the wrong type of actor? Wrong type of blueprints? I'm new to this and very lost.

No, not everything never mind. He gives an extremely good base coverage of MOSTLY everything. To the point where you know how to use blender enough to screw around and get what you want,

Never mind figured it out

I'll get to work after I finish eating this apple.

archive.is/xAusG

zdnet.com/article/epic-games-unreal-engine-forums-hacked-in-latest-data-breach/

WHODUNIT?

It was past time for me to change my password anyway.

Things are coming along slowly. I'm working on a small airstrip and airship port pit stop market place thing.

The pink object is a placeholder I'm using for size reference (dimensions for a large carrier). The small barely visible jet inside of one of the hangars is the current model for the player's aircraft, again for size reference.

just changed my pass, just in case somthing happens.

Oh user, here's something fun to do. Massive security breach on a social level. So, Pokemon GO, right? People sometimes pay others to farm for eggs and catches, right?

You can basically get a full list of all their passwords in plain text. Chances are they're normalfags and have YouTube, Facebook, Reddit, a number of email sites, even banking logins saved.

Why do people use Chrome again?

It's idiotproof and mostly secure, unless you ask it to be convenient. Great for normalfags.

But user
When I signed into my phone to play Pokemon GO, it didn't ask or send any verification emails

first visual update in a while.

I changed how the code works from one monolithic monstrosity, now into a few convenient functions.

Once I did that, I changed it so that it would generate a new chunk every time, rather than copy the test chunk over and over.

Next up is the save/load system!

I spent a while writing an optimizer for loading. It worked really well. But I changed my mind about how it's going to save data.So I scrapped the whole thing. What have we told you about optimizing before finalizing your design, user? I don't know why I expected anything different.

Man, there are blank pages
Like page 394 "47. Penguins from Spheres"
Completely fucking blank beside that
And there are more
I really wanted to do a penguin from spehers, but now I don't know how to :(

en.wikibooks.org/wiki/Blender_3D:_Noob_to_Pro/Penguins_from_spheres

Many thanks, man. Sadly the pdf is fucked.
I guess I'll need internet at all times to learn it.

Anyone here ever try using UE4s EQS system with C++?

HUD while in civilian mode of Space Pirate. Since there's less overall fancy features, the HUD is much more simplified.
Health, keys, yen, bam.

unless you constantly earn or lose money in civilian mode you should keep it on the center

finally add a UI with unit cards and such
though it's still not interactive that will be a real pain in the ass.

fucking why

...

what are you using a gif for?

i wanted an animated sprite from an existing gif
but i didn't want to have to take a gif, put it into photoshop and manually make my own spritesheet

You do know there's tools to do exactly that, but automatic, right?

if i knew, i wouldn't be complaining about lack of gif support
spoonfeed pls

Well, I have come to the conclusion that engines like Unreal, Godot, and Unity are just too bloated to get the performance I want for 3D games in general. So, I think I'm going to write an engine. Maybe not a complete engine, but just enough libraries that I can cobble other things together and get something I have enough control over.

Yeah, sorry about that. It seems that they cut a lot of the actual examples out of the pdf, but I guess it would make decent reference material.

sites.google.com/site/stevedunns/convertanimatedgifstospritemaps
One of the first results from a searchengine looking for "gif to spritesheet". This one even has available source code. Just type things into google, that's gamedev 101.

don't lecture me on google, i tried one of the things that way and it didn't work so i gave up

It's no one's fault but your own that your google-fu is weak

Try more. Modify an existing one. Write your own.
Or, if you only intend to rip a few, maybe consider doing them manually.

I fixed a bug that shoved you across the entire stage if you jumped into a pushable object standing next to a wall. Now I'm trying to implement hit detection, which needs me to first change my animation scripts to hopefully fix a weird inheritance bug which makes the hitboxes malfunction.

Does anyone here have anything to comment on for Godot's scripting language? If it's any good I might consider using it over C++

It's alright, I'm not a big fan of it because I hate dynamically-typed languages. Honestly, I'd just stick to sing C++ instead of gdscript, why bother with some high-level scripting language that needs it's own interpreter when you could just write it in C++ anyway?

So do I. I just accepted that I can create the game, but not all on my own. At least not with the deadline I have.

Thanks. I think I might have scored a perfect score in (x86 assembly).

can someone help me imitate the "best fit" option in unity's UI text?
my issue is that i'm adding chars to the string over time, like in most rpgmaker tier games, but with best fit, it ends up changing the font while the text is still being typed out
i need to do it when it starts, but i'm not sure how they get the exact value for the font

Not sure about your actual question, but appending chars to strings is a bad practice. Strings in javascript and c# are immutable; meaning everytime you change a string you are creating a copy of that string and appending new chars to it. If you do this often, you could end up with a ton of useless strings, and your memory usage is going to be shitty because of it. If you want to add chars to a string, you need to use a buffer (I think it's called a stringbuilder in C#).

so like this?
public static string GarbageFreeString(this StringBuilder sb){ string str = (string)sb.GetType().GetField( "_str", System.Reflection.BindingFlags.NonPublic | System.Reflection.BindingFlags.Instance).GetValue(sb); return str;}
in my case i was using an ienumerator to add a delay

Haven't posted in a while. Here's some recent garbage.

Looking it up, when you normally return a stringbuilder object using the toString method, the stringbuilder is no longer needed and can be wiped by garbage collection. The code you posted is part of a small hack that lets you use a stringbuilder without freeing the stringbulder object. It allows you to extract the string straight from the stringbuilder, and when a change is made it can overwrite itself back to the stringbuilder, all without using toString and generating garbage.

Now, I don't know if you want that extra kind of optimization or not, but I'm sure using the toString method without the hack will be sufficient for your game. It's still 10x more efficient than copying over new strings anyway.

implemented maps last night, today im gonna take it easy and start working on gamepad support

question for the 2D devs;
how do you go about z-ordering?
my idea was to do have a player-entity bounding box check where the players height is subtracted/divided by, for example a quarter to half the sprite.
i cant find anything on dynamic z-ordering that explains any algorithmic way of going about it, so i've been pretty stumped

looking good as ususal
are you hand making dungeons?

Ported the new skeleton to the engine.

send help

dude this looks sick
got any more?

anyways, here's my garbage for the day, i made sure to use tons of free sprites, i think i only stole the music
issues so far:
can't get text to fit rectangles before it gets spelled out
can't figure out how stringbuilder works

so i just make a new StringBuilder object, append a few chars to it and then do .ToString() for the string builder object?
i tried something like that and for some reason i the returned string was all numbers

i require more lewd IKs

Please, put transition when stopping or starting to move in the animation
It looks too good not to have that

You got any tips on forearm twists?

What engine does dwarf fortress run on?

pure concentrated autism

...

More bugs? Sure.
I need to reimport animations, the old transforms were fucked up.

so what just opengl or something?

There isn't anything you can do about it, because Unity expects a string property for the text. One way or another, you're going to produce strings that you'll throw away once you need to display another.
Strings being immutable isn't a C# thing though. Many implementations are, because the "natural" way of creating strings (backing them by character arrays) makes them fixed in length by default. Many of the commonly performed operations on strings (concatenation, inserting, substituting, formatting, etc.) change their length though.
How to minimize string-related garbage in Unity:

Opengl? user, you sure about that statement?

I posted everything up there already.

nope, i just want to know what df runs on.

SDL for normal rendering, but it can output pure text on linux systems too. Rendering is independent from the game itself, as it should be.

That's it?

AN ASS THAT TRANSCENDS GEOMETRY.

Google, mane
Or any other search engine
facepunch.com/threads/918200-Dwarf-Fortress-Engine

Yes, that's how her wrist works.
This will work for ankles too, but shoulder/thigh twist bones will likely require entirely separate IK/FK rigs.

thanks

Weird, I guess I'd have to see your code. Either you're appending the wrong things to the string,, or you're getting a memory address somehow.

public IEnumerator SetText(Text txt, string target, AudioClip clip, AudioSource tar) {StringBuilder sb=new StringBuilder(target.length); float speed = 0; Transform cam = Camera.main.transform; if (clip != null) { PlaySound.Play(tar, clip, 1); speed = clip.length / (2 * target.Length); } else { speed = 0.025f; } txt.text = ""; for (int i = 0; i < target.Length; i++) {sb.Append(target[i]);txt.text=sb.toString(); // txt.text += target[i]; yield return new WaitForSeconds(speed); } yield return null; }

uh yeah, ignore the camera/sound stuff there, that's just stuff that currently does nothing of importance
basically what i'm doing is: sb.Append(target[i]);
txt.text=sb.ToString();

Monster Girl Island-san responded to my question

I don't have the language docs handy but is that actually setting the length or giving the sb an initial string of numbers

Move txt.text=sb.ToString() outside of the for loop. I'm assuming that txt.text is going to be used outside this function to change the UI text?

as far as i understand it's just the max number of elements
msdn.microsoft.com/en-us/library/h1h0a5sy(v=vs.110).aspx

You're also invoking .tostring every iteration
What

Stringbuilder exists so you don't incur the overhead of constantly mutating a string. If you're doing something complex you make a stringbuilder then append it and only after everything is done do you invoke tostring.

Not sure why you're doing it in the middle of an iterator block.

i already mentioned earlier, i want the text to be typed out like in classic jrpgs, which is why i was doing the tostring mid loop
i guess i misunderstood what the stringbuilder does then

some more progress on my level, and I changed some things about it.

Oh, ok. I get why that was there then. Calling stringbuilder like that is marginally better than just concatenating a string dozens of times, still there should be a better way to do it. Maybe you can call To.String after the for loop, and just write that string to the screen char-by-char with some delay in between?

Stringbuilder exists as a good performance alternate to doing "myString+=char" in a loop.

It just seems weird to do it in an iterator block. If something has an IIterator I expect to be able to use a foreach loop on it which would return each set of letters like you want but it would do it all at once. You could manually advance the iterator but then you're basically making a one off iterator that isn't generic in which case why use it?

Stringbuilder is fine but it's the iterator part that throws me

well what the fuck am i supposed to do then

StringBuilder supports replacement of single characters by index.
Make an empty StringBuilder the size of the original. Replace characters one by one. Push the builder's contents to the UI.

StringBuilder is a tool.

This is the correct answer.

how is that different from what i'm already doing?

Your UI has a Text variable of type String. You want a new letter to appear every 40ms or whatever as the text is revealed. To do this, you just need to update the string that it holds. That's what your goal is.

It doesn't matter where the text comes from. You can append it manually, you can push a StringBuilder to the UI, you can take a char[] and convert it ToString(), whatever you want.

As I've said above, the problem with your approach is that it makes use of an IEnumerator interface. This means that your code needs to guarantee that it works with the foreach construct. It doesn't. If I gave it the string "The quick brown fox", and attempted to iterate over it, I'll get:

But the problem is that it'll all happen sequentially, and instantaneously. If you were to draw the results of the operation by doing something like:
foreach(var item in SetText("The quick brown fox")) myUI.Text = item;
The screen would only have the last, final result drawn by the nature of how iterators work. It really would step through each element, but you'd just end up with the final result. I should also point out that "string target" feels weird to use; if it's a UI operation, you should just pass the UI element directly, or expose SetText as a method on the element itself

What user means by doing it differently is to just make a method with a state-tracking object that advances the text each time it's called. This is similar to an iterator, but simpler and doesn't need to meet the guarantees that it requires. It can be as simple as a String, an int, and a GetText method that basically returns the string from position 0 through N. By completely abstracting it like this, you avoid constraining it to any one thing, and it gives you a ton more flexibility.

That's already working, via the magic of coroutines.
The yield statement will temporarily halt the loop and resume execution later.

well how else do i create a timed delay?
i thought something like WaitForSeconds only worked in IEnumerators

That's not how iterators work in C#, at all.
An iterator is (usually) a compiler-defined state object. All it does is remember where it was between each step of the foreach or manual iteration. There is only a single thread and no coroutines.

Okay, so I didn't have time to test it to see if this even compiles or works right, but this is what I would end up doing in your situation:

public sealed class TextRevealer{ // Constructors public TextRevealer(string text) { if (text == null) throw new ArgumentNullException("The specified argument must not be null.", "text"); Text = text; Length = text.Length; Current = 0; } // Properties public string Text { get; private set; } public int Length { get; private set; } public int Current { get; private set; } // Variables private readonly StringBuilder _builder = new StringBuilder(); // Methods public string Next() { if(Current < Length) _builder.Append(Text[Current++]); return _builder.ToString(); }}

For brevity, it's immutable. If you want to change the revealed letters, then you make a new TextRevealer object; it's much simpler to do that than check states and bounds and bog it down with extra logic.

So to use it, you'd do

You'd have a timer in your update logic that determines how fast letters appear, and when conditions are adequate, you'd call tr.Next() and update your UI with the new letter. You draw the UI element every draw step.

Unity is abusing them to make functions that can work over several frames.

Christ, I should just stop trying to help anons using Unity, since I don't use it and it just promotes all sorts of weird programming habits

tell me if i'm crazy or just stupid

wait a minute, i still don't get how that works differently from mine
in both cases in each loop cycle, the string builder appends one character and the visual text gets updated
as far as i understand your real complaint is that i'm using an IEnumerator?

Yeah

But I guess if you're doing it The Unity Way™ then it's fine

well the alternative would be to make a float timer, and increase it by deltaTime every frame
which i choose not to do because it just gets more confusing that way

Don't do that, always calculate a "next time" in advance and compare against Time.time, it's one operation less per frame.
Anyway, the docs say that coroutines have next to no overhead so they're probably faster anyway.
As for your original problem with text scaling, try fiddling with UI settings to simply make it not scale at all.

you want me to predict how much time it's gonna take to reach the next frame?
anyways, i do want it to scale, having some small text on a nearly empty font looks silly

Next time = current time + 1/60th of a second?

...

what are tiny games good for practice to make in 3D?

Next time meaning Time.time + however long you want to wait. Then just check if Time.time has passed that mark.

Make an Intelligent Qube clone.

Practice what, 3D programming or 3D art?

eh, i still feel comfier using IEnumerators
having extra code just for the sake of it being in an Update function doesn't seem worth the effort

both
using blender and unity or other free softwares

gamefromscratch.com/

Make a physics-based puzzle game.

That way models are fairly straightforward to make and texture, level designs can be as easy or complex as you feel comfortable making them, and you learn how to script and apply properties and learn how that shit works at the same time.

This way you learn a bit of everything, can test it all out pretty quickly from the get go, and have some fun with game dev tools.

That's pretty much what I'm doing, except I kind of picked a harder genre to start off with (flying game).

Sure thing, my advice was regarding any sort of time counters.
Though now I'm worried about Time.time losing too much precision after some hours, since it isn't double.

Oh yeah, by the way, if anyone wants a free copy of Pico-8, I bought it a while ago.
lexaloffle.com/pico-8.php

drive.google.com/open?id=0B_5ZkqdDFIVdZklYbVpYYTcwNzg

Of all the things to get stuck on

For some reason my deltatime does this weird thing where it goes up and down randomly, it should never go below 16 but sometimes it goes up a bit and then all the way down to ~3. The only way it could go down to 3 is if my display was refreshing at 333 frames per second.

sounds good
good luck to the rest and may you all be blessed with good devving and less shitposting procrastination

Agdg picojam fukken when?

I am not sure how those numbers relate to each other

The gif shows FPS over the last 1 second. "average (min/max)"

I'd be down for that. Using pico is pretty fun, but I haven't done much with it yet.

Just keep in mind that it's basically a gamedev toy. You can do fun things with it, but I wouldn't count on releasing a serious project. It is neat to mess around with.

Got a helmet and some boots on that bitch ass character. I didn't know that UE4 supported vertex colours. They're used in the helmet.

Does anyone have a torrent or link for a pirated version of Hammuer (The UE4 plug-in)?
Help a poorfag out

Yeah, I am, I might make an end game proc gen dungeon, but I'm not sure yet.

it begins

I remade my physics stuff, instead of hard coding everything, I can now attach physics engines and controllers and AIs into anything, and it'll move accordingly. The only difference between the player (yellow) and the NPCs (red) is that player is moved with a keyboard instead of random directions. All the physics and movement stuffs and controls are identical.

Pretty excited to use this properly.

Hey friend, my water is doing that refract around objects too, do you know how to fix that, if not would you like to team up and figure it out?

How are you "attaching" things?
I've always had an issue where I don't know how to tell an object what to expect

I have no idea, and I haven't had any interest in figuring out until now. I haven't even looked up any hints on how to avoid that.

after running in circles figuring out love's current controller input, i
figured out that it uses some random txt binding method now and there was already a well known txt with hundreds of bindings

still have no fuggin clue how to do actual axis input

you could always do dungeons binding of isaac style
create rooms/areas by hand and randomly string them together based on a set of rules
of course handmade dungeons are the better decision, and would allow for a more seamless experience

For your own sake, don't ask me for programming tips.

I don't know how you would do it with "proper" languages, with javascript you can change any variable into anything. So if I want to change my AI, I would just do something like: object.ai = physics.ai.flyingBat

Then when I call object.ai, it runs physics.ai.flyingBat.

love.joystick.getAxes()
love2d.org/wiki/love.joystick.getAxes

Its a feature!

The revolver is almost done. I'm encountering a bug where the shell particles are instantiated quite a bit above the cylinder rather than coming out of it when unloading. Shouldn't be too hard to fix.

I'm still happy with the project.

hello sweets

Fuck I don't know why I replied.

Don't think I didn't notice that wall pose.
nice detail though, seriously

porn when?

This thread's been good for a fap or two

Don't die on me

pls no bully the music, youtube does it's autoplay thing and I stop noticing when I'm devving. I left it in because it beats silence

I am making a tablet game wherein you control a carrier. It's kind of a hybrid between a TD and FTL. Still buggy and I just started basically today more or less, but I'm sort of happy with how it's starting to look

It's past bumplimit user, someone's making a new thread in a few pages.

Wait, you mean to tell me this UE4 space game I've been seeing all these months is just going to be a tablet game?

This you?
8ch.net/agdg/res/24615.html#q25651

I actually really like the music.