/v/ - Video Games

>>15839122

>ded thread

well, guess I'll post something then. Been working on a text based adventure game, with turn based combat, something along the lines of Final Fantasy but with more unique mechanics. I'm not very far along, certainly not far enough along to post any progress that doesn't look like crap, but hey - I was always a nodev, so actually working on a project is quite something. I'm using Twine because I'm a massive retard and I don't plan on putting any real graphics in it, so I figure why not. Plus I've worked in html, css and javascript before, so at least I'm doing something I have experience in.

anyone have any good tutorials/examples for monogame? I've been searching for like 2 days for good tutorials but most ones I've seen either are outdated, use deprecated libraries, unqualified speakers, or something is always up. Anyone know any good resources for it?

>>15839840

read a book nigger

hi im a girl i want to make games where do i start?

>>15839487

Good luck anon

>>15840140

First you need to download more ram

>>15840140

Whether you're serious or ironic, post tits or GTFO.

>>15840229

didnt you hear me? I said i was a girl

>>15840140

Go cheat on your boyfriend with 5 different reviewers so they make you money with good reviews.

anyone done instancing in unity using Graphics.DrawMeshInstancedIndirect?

i'm using a vertex shader that takes a float4 compute buffer (xyz = position, w = rotation.eulerAngles.y) to displace them

problem is that if i enable shadows i just get one stacked up shadow at (0,0,0)

unity docs say you just need to add addshadow to the shader pragmas to generate an appropriate shadowcaster pass, but that doesn't seem to do anything for me

i also tried manually writing my own shadowcaster pass, but there's a ton of problems there, mainly the fact that the original pass is a surface shader, and all working examples of shadow casters are fragment shaders where some of the base code has to differ

right now what i'm doing is just have 2 shaders, one for the regular mesh drawing and one that acts solely as a shadow caster (but still as a surface shader)

the shadow caster shader does correctly cast shadows, but the problem is that it still leaves a black mesh at (0,0,0) that i can't get rid of

>>15839840

Did you check the monogame community for resources? Found this right at the top http://community.monogame.net/t/making-some-game-dev-tutorials/10818

I don't know if they're any good, but generally speaking, the best place to find current and up to date resources is to consult directly with those currently using the thing you need resources for.

>>15841151

How much exposure does one of those white squares cost?

>>15840229

There are occasionally good ones. I forget if it was Spyro or Prince of Persia, but one of the devs mentioned in an interview he went to GDC and the presentation he went to laid out the core pieces of the university he went to.

>>15843186

Doing web games? I tried it once for 3d. It was promising at first, but I was only doing extremely basic things. Eventually as a stress test I tried drawing a couple thousand squares (not cubes and not even more than you'd see in minecraft) and it lagged hard. That tells me you still need to do the usual OpenGL fuckery with buffers to get anything powerful out of it. I never understood the point after that and have been just doing WebGL.

>>15843297

It might work. I don't really understand what it gives over straight WebGL though.

>>15843327

>Also I tried like three game dev books and they all heavily relied on the provided CD that has all the code samples

This truly is cancer. I gave up on books a long time ago.

>What I can say is that I want my game to be played without a browser. The bloat of something like electron seems dumb and I want something like GTK or QT so I can open a window directly without having to rely on a 60 MB chromium binary.

You're going to get bloat out the ass if you want javascript and no browser. You should really just accept it.

>I want something like GTK or QT

Those are just window libraries. They don't give you 3d just a window to draw on. Three.js actually gives you more since it has a few drawing functions implemented, but at the end of the day you have to do the painful WebGL sooner or later. That's why I said I don't understand the point.

Basically, you're not going to be able to get far on Three.js alone from what I can tell. You're going to need libraries out the ass.

>>15843373

Yeah. SDL does it to, but you still have to do all the painful opengl stuff on them. Three.js gives you a bit more room, but like I said you still need to do the painful opengl stuff eventually. Maybe another anon knows more, but from what I seen it doesn't get you that far.

>>15841129

Please note that I made my grass more than 2 years ago.

The relevant shadow parts of the shader:

#pragma surface surf WrapLambert vertex:vert alphatest:_Cutout addshadow interpolateview halfasview fullforwardshadows nolightmap
half4 LightingWrapLambert(SurfaceOutput s, half3 lightDir, half atten) {
	return float4 (s.Albedo * _LightColor0.rgb * (1 * atten), s.Alpha);
}

I don't remember what most of these things mean, I had problems with shadows, but besides avoiding using copies of the same material, there is not much I could say to help fixing it

Thanks to the anon for suggesting SDL_FontCache, but I don't think I'll use it.

I glanced at the code and it's pretty much loading fonts the same way that I do (font name + font size)

Then it keep the resulting render on a surface and blit letters as needed. As for me I keep the full string result in a texture array on GPU, the needed quad is already sent to GPU too so I only need to issue a draw command for as much quad as configured.

I'm not making an RPG so it's fine as it is right now, but if I do I think I'll go more with a triangle strip with correct UV mapping route.

Anyway, sprite manager is the cleanest rendering implementation I did so far, I'll revisit model rendering later since it's working. Currently coding bars for hp/shield

>>15845008

It has a lot of shortcomings in the 3D department (no occlusion culling for instance), but development is steady thanks to its patreon goys. If you're planning on releasing a big (performance-wise) 3D game in the next 6 months, just cave for Unity or UE4, otherwise Godot's a decent long-term bet.

>>15845054

Are there any other decent 3d framework or engines that aren't dead as fuck? I don't care what language. Just need it to work and actually have documentation as well as people using it.

>>15845105

Panda3d with PBR and Armory3d are my top 2 choices for 3d engines outside of Unity and Unreal.

>>15845008

theres also Urho3D, which looks pretty promising and also Lumix Engine, which looks interesting as well and has its own editor. both under the MIT licence, iirc Lumix Engine had some unported stuff/shortcomings on GNU/Linux though.

Haven't used either of them so can't give more input than 'they exist', but might be worth looking into. they've been on my ideas log for 3d game for a while.

>>15845105

Just found another that looks pretty neat, they even have interactive code snippets running in the browser here; https://heaps.io/samples/world.html

How come most of you guys build an engine before attempting to make a video game? Seems very time consuming and daunting.

>>15845817

The best reason to do it from a practicality stand point that I've ever heard is that you own the code. Granted there are open source options, there is still nothing better than you having soul owner ship of your entire codebase.

>>15845817

Most of us do not. Sigmadev and the robot guy are anomalies.

>>15845817

because it's fun and i don't know of any other tiny FOSS RPG engines written in C and with minimal dependencies

but mostly because it's fun

>>15845817

There's a standing theory that you "learn so much" when you write an engine yourself. That's fine and all once you've got programming down pat, but it's basically 2+ months NOT making a game, which is objectively a problem for a hobbyist.

There's also the occasional elitist fuckwit who looks down on others for the high crime of not building their own engine as anyone less than programmer himself has no business even attempting to pretend they understand the API. These people completely fail to understand that their prescribed methodology would force all coding and scripting workload for all games using a given engine onto just one person.

The real reason to build your own engine is so that it does everything you need it to and nothing you don't. This requires advanced planning skills and a strong commitment to the project as written in a design document without modification, or it requires frequent modification of the engine code itself.

It's not bad to make your own engine, but there's nothing Unity, Unreal, or Godot can't do with a little effort applied

>>15845869

the fact that you put "learn so much" in quotes makes me think you condemn learning or improving for the sake of it, which means you're probably a shitskin

>>15845900

>I haven't posted in a while though as my GPU is nearing death and I can't open google maps without crashing, let alone work on a video game.

that's what you get for buying a 1 kw graphics card

>>15845900

the necessary features part can also be achieved with modifying a stable FOSS game engine, might be easier and if i was going for a 3D game then i'd most likely do that, unless i wanted to put time into learning graphics programming.

>>15845900

> but I remember talking to an anon that told me he worked on his for 2+ years

You can put a workable engine together in a few months, provided you know what you're doing. The problem is that most people who try here don't know what they're doing yet, learn as they go along, and then learn (part of) what they're doing that way and realize they have to redo it all. And that repeats for a few years. Either that, or they get hung up on quality of life improvements or overoptimization.

>>15845903

The R9 390 worked fine for the three or so years I had it. Tried to fix it but no can do. Buying a new one soon.

>>15845916

That's where the brainlet part comes in. I'd probably have to learn everything from the beginning to adequately implement anything.

>>15845919

>most people who try here don't know what they're doing yet, learn as they go along, and then learn (part of) what they're doing that way and realize they have to redo it all

Sounds like my kind of people.

This is probably a stupid student tier question but are there resources, guides or starting points to recommend when it comes to doing it? As it is now it seems undoable for me.

Have your ideas ever been stolen while posting on these threads?

>>15845919

>Either that, or they get hung up on quality of life improvements or overoptimization.

that's what my previous attempts got hung up on, trying to do everything as perfectly as possible, but it ended up causing me not to make any progress and get uninterested in working on it

>>15845929

>That's where the brainlet part comes in. I'd probably have to learn everything from the beginning to adequately implement anything.

i don't know if its any different for anyone, if im in a new codebase then it takes some time to understand what the heck does what, i think its normal lmao

read up on finite state machines/gamestate implementations, that's what drives the "modularity" of the game (in menu, in game, in editor etc), then its just a matter of adding enough functionality to paint pretty pictures on screen according to what gamestate the player is in and handling input according to gamestate to change the pretty pictures.

>>15845817

I think using already game engines are appealing because most people are making games that could have easily come out 20 years ago with worse graphics. What if you want to absolutely push the hardware to its maximum limits? That's when you make an engine.

>>15845950

I'm going to have to read up a bit about it definitely, though I'm afraid I'll fall into the trap I mentioned. Where did you learn about it? Youtube, books, articles? Sorry if I'm coming of as a faggot begging for a spoonfeeding.

>>15846004

Wouldn't that kind of an engine take longer to create than a couple of months? Wanted to create something with relatively modern graphics, hopefully existing engines will suffice.

>>15845817

Enginefags don't really like making games, they like solving problems. I like writing systems in existing game engines for character creation, fighting, inventory, some bullshit animation system, movement and so on, but when I get to the point where I'm supposed to make some actual gameplay I immediately lose interest. I guess enginedevs are similar, but prefer higher difficulty level.

>>15846044

i just read an article[1], i'm the

> learn as they go along, and then learn (part of) what they're doing that way and realize they have to redo it all

type >>15845919 mentioned, i prefer to jump head on in and just wing it, so i haven't got many reading tips.

[1] http://gameprogrammingpatterns.com/state.html

found it in my history after all

>>15846044

>Wouldn't that kind of an engine take longer to create than a couple of months? Wanted to create something with relatively modern graphics, hopefully existing engines will suffice.

You can use an engine to make any modern game. Any modern game could have easily been made +20 years ago with downgraded graphics.

>>15846095

By turning my brain off and mindlessly fucking around with shit. Even if it's something that is trivial.

>>15846095

Gamejams tend to help, as long as you can bring yourself to actually drop the project when the jam's over instead of using it as an excuse to drop your old project.

Also forcing yourself to do some other hobby for at least half an hour to an hour every day. I'm a programmer but lately I've been scribbling some shitty art onto paper every evening.

>>15846144

>forcing yourself to do some other hobby for at least half an hour to an hour every day. I'm a programmer but lately I've been scribbling some shitty art onto paper every evening.

I've noticed this to. I think it is related to cross neuron formation. LIke I found my ability to compose music seemed to improve after a few months of regular drawing practice. After learning new things about music for awhile, my art seemed to improve. In both cases it felt more like my ability to understand them was what expanded. There is also research that shows adult learning can "overwrite" past trauma, so I suspect that some of the improvements can also be from suffering less from the stress produced from past trauma mental scars.

>>15846061

I actually bought that book around the time my gpu started screwing with me. Good to know it's a good resource. Really want to brush up on my C++ too as this seemed way out of my depth.

>>15846051

Good to know cause I have some ideas.

>>15846342

Ask your lord and savior wikipedia https://en.wikipedia.org/wiki/Hypercube

A tesseract has 16 vertices, 32 edges, 24 faces, 8 cells and 1 "4-face".

If you want to know exactly what the 24 faces are, you can construct them as follows:

>create 4 vertices, forming a square ({0,0}, {0,1}, {1,0}, {1,1}), these are your "varying" coordinates

>of your 4 dimensions, pick 2, these will vary according to the scheme above to make a plane

>create 4 4D vertices, replaces your chosen dimensions with the varying coordinates, and fill the remaining 2 dimensions with either 0 or 1

>do all combinations of the above to get all faces (there's 6 choices when choosing 2 dimensions, xy, xz, xw, yz, yw and zw. There's 4 choices for filling the remaining dimensions, 00, 01, 10, 11. 6*4=24 faces.)

I don't know what you constitute as "unique 3D faces".

Is a face any set of varying coordinates? Then there are only 6 in a tesseract.

Is it a particular direction a face is facing towards? That's fallacious, as {0~1,0~1,0,0} forms a square over xy, which z and w are both perpendicular to.

When you're adding dimensions things get very fucky and you have to rework your line of thinking entirely. Don't rely on 3D projections of 4D images

>>15846582

I took it more as something like from Resident Evil, Silent Hill or old school adventure games, where the protagonist can give their thoughts on the environment.

>>15846755

maybe

during the summer most girls will walk around in swimsuit since the city has plenty water to swim in

>>15845817

>How come most of you guys build an engine before attempting to make a video game?

Because you got no fucking choice either way.

I am using Unity, but 90% of my time is engine coding.

No matter what you will engine code a fucking lot unless your game is mass produced asset flip fps shit

>>15846766

>not making a dynamically changing tan skin shader

>>15846787

After release maybe

>>15846755

Kagura-san is a nigger.

NIGGER

NIGGER

NIGGER

NIGGER

NIGGER

NIGGER

NIGGER

NIGGER

NIGGER

NIGGER

NIGGER

NIGGER

NIGGER

NIGGER

NIGGER

NIGGER

NIGGER

NIGGER

>>15846784

>By unique faces, I mean faces that never share a normal no matter what combination of in/out you pick.

Like I said before, these faces don't have normals, since you cannot construct a single normal vector perpendicular to a 4D plane.

The face/plane {0~1,0~1,0,0} can have both {0,0,1,0} and {0,0,0,1} as vectors perpendicular to it.

Instead of asking vague questions with wrong definitions, why don't you just write out all the faces of a tesseract, determine their "normals" and figure it out yourself?

>>15846920

Those are the xy/xz/xw/yz/yw/zw planes I mentioned earlier. And there's 6 sets of 4 faces of them.

>>15846998

Yeah, in hindsight it makes sense considering the three "unique faces" as I called them on a cube are effectively the xy/xz/yz planes as well.

I think I can progress with my game now, though first it'll just be a lot of "programmer rages at Blender interface".

>>15845817

No engine is sufficient. They're all either closed source, Windows only, resource hogs, or don't have sufficient functionality. Making a custom engine shows in the final product. There's always cruft left over from using a prebuilt one. Even when the devs go the extra mile to decorate their game, it's usually easy to tell which engine it's made in.

>>15845869

>These people completely fail to understand that their prescribed methodology would force all coding and scripting workload for all games using a given engine onto just one person.

Programming is done best when programmers write software to interface with each other, not working on the same code.

>>15845940

I've had some of my assets used but nobody has gone so far as to use my game idea.

>>15848583

If you want to keep using a state machine, you just add another state for "attack while locked on".

>>15848652

I honestly don't know what a better way to do it than a state machine would be in terms of performance. I need performance because I am targeting something similar to Dwarf Fortress. Where the units can do 100s of tasks and combat is one of them. Also with +1,000 dorfs instead of 50-100.

>>15848768

Youtube won't help unless it either explains WHY you're doing what you're doing as well as how. That's where most tutorials fail.

The best way to learn to code is to get your hands dirty. You have the bare basics down, do you not? Think of specific things you need, Google them and try them out on your own. You'll pick up on it over time.

>>15848773

I'm bad at that, I don't know what I wanna do

"just like make pong" helped me a lot because it gave me a sense of direction

>>15848784

Just like make Super Mario Bros.

>>15848768

i'm not even a dev but i think what you're looking for is creativity

>>15848789

alright

>>15848768

What do you have to lose by starting your game?

>>15848985

>Why the hell does C++ generate all this garbage syntax?

That's the price to pay for generics.

>>15849037

>What do you have to lose by starting your game?

seems like a really bad idea considering it would be a fighting game, would turn into a huge unfixable mess after a few months

>>15849050

Then just start again. You'll be exactly where you'd be if you never started.

>>15845940

It's generally the execution that matters, not the ideas themselves

>making a RPG

>game mechanics are on the way

>can't think of a setting, plot, or anything related to narrative to fit mechanics

Goddammit.

>>15849673

I'll be your idea guy, bud

>>15849068

>only good speaker at CppCon admits he only uses the language because it's popular

>>15849673

>have a finished engine

>can't make a story, art, or sound

>spend my days optimizing my engine

Every time I do something other than programming, it's garbage.

>>15850116

That might mean you're destined to be a wageslavery cog or worse, part of a team

>>15850204

No way, I won't let that happen. I'll find a way.

>>15849868

No, anon. I only just started yesterday, so I'm a nodev in the extreme, unlike you who's at least working in Twine coming from some coding experience. I could try bouncing off some story ideas in the thread, but I'm afraid I'll just get billied.

>>15850364

generally, you get two or three chances to fix shit before people actually start bullying you.

I'll trade some of my pajeetery for your coldsteel

        m.onButton = function(code)
            if (m.locked = false)
                real_delta = m.acc_time - m.last_held_time
                m.last_held_time = m.acc_time
                held_inc = 60 * real_delta
                
                if (code = 6) 
                    m.game.map.nudge(m.pos_x, m.pos_y)
                end if

                if (code = 105) : m.r_held = 0 : end if
                if (code = 5) : m.r_held = m.r_held + 10 : end if
                if (code = 1005) : m.r_held = m.r_held + held_inc : end if

                if (code = 104) : m.l_held = 0 : end if
                if (code = 4) : m.l_held = m.l_held + 10 : end if
                if (code = 1004) : m.l_held = m.l_held + held_inc : end if

                if (code = 102) : m.u_held = 0 : end if
                if (code = 2) : m.u_held = m.u_held + 10 : end if
                if (code = 1002) : m.u_held = m.u_held + held_inc : end if

                if (code = 103) : m.d_held = 0 : end if
                if (code = 3) : m.d_held = m.d_held + 10 : end if
                if (code = 1003) : m.d_held = m.d_held + held_inc : end if

                if (m.r_held >= 10) 
                    m.r_held = 0
                    m.pos_x = m.pos_x + 1
                end if

                if (m.l_held >= 10) 
                    m.l_held = 0
                    m.pos_x = m.pos_x - 1
                end if
                if (m.u_held >= 10) 
                    m.u_held = 0
                    m.pos_y = m.pos_y - 1
                end if

                if (m.d_held >= 10) 
                    m.d_held = 0
                    m.pos_y = m.pos_y + 1
                end if

            map = m.game.map
            if (m.pos_x > m.game.ROOM_WIDTH) : m.pos_x = 0 : map.tryMoveChunk(1,0) : end if
            if (m.pos_y > m.game.ROOM_HEIGHT) : m.pos_y = 0 : map.tryMoveChunk(0,1) : end if
            if (m.pos_y < 0) : m.pos_y = m.game.ROOM_HEIGHT: map.tryMoveChunk(0,-1) : end if
            if (m.pos_x < 0) : m.pos_x = m.game.ROOM_WIDTH: map.tryMoveChunk(-1,0) : end if

                m.x = m.pos_x * 64
                m.y = m.pos_y * 64 + m.game.UI_OFFSET.y
            end if
        end function  

>>15850405

Alright, if people are willing to be patient with my 'tism then I'll go for it.

I suddenly got the urge to make a RPG with a heavy element of love.

Your character lives in a machine world where order is heavily prescribed to by society. However, great gremlins roam about disrupting the functions of the world. Such things are accepted as natural, though society responds by limiting themselves to maintaining key features of the world-machine where its great settlements are founded.

Suddenly, your character receives a message in the dark of night. It is delivered in the voice of your character's former lover, asking that her soul be freed such that she may pass on. You must go to all the places you have previously visited with her and collect the mementos of your time together, which have mysteriously become objects of interest to the gremlins for unknown reasons.

That's what I've got so far after after getting the urge to make a love story. It's cheesy, but I'm hoping people will become interested for the same reasons people like love songs and romantic movies.

>>15843388

ah i found it, apparently the vertex being scaled in the vertex function instead of the setup function of the shader was causing the shadow to not get scaled and appear invisible

also shadows weren't there at all with Lightmode = Deferred

if anyone's ever curious, i found out about this+other errors from this guy

https://github.com/tiiago11/Unity-InstancedIndirectExamples

got the engine refactor to same functionality as the previous revision, with a few things improved due to being able to store more things in the map format

dialog gets cached to resource manager now instead of hitting the disk every time to read the file and preprocessing it again, character sprites can be defined in the character file instead of separate sprite files and also everything is loaded on demand instead of startup and everything can use string names instead of numeric ID's, which was a pain to work with

RIP in peace 2500 lines of editor and map format code ;_;

>>15848985

C++ was originally marketed as C compatible, so they always tacked on new things while keeping it visually similar and compatible with C. Ironically, they failed at that (there is a lot of valid C that isn't valid C++), but as a result the syntax is overloaded to hell and back. C++ is what happens when you have no design beyond "Did you say $FEATURE? Of course we have that too! Bjarne, get the duct tape!". It's a terrible language that survives because of inertia. There's also this weird meme that games HAVE to be written in C++. I have no idea where it comes from but it's wrong, especially for aggydaggy type stuff.

>Also is modern OpenGL basically defunct because of Vulkan?

No, OpenGL support isn't going away any time soon. OpenGL 1 is old as shit, but at this point you should make your fucking game instead of procrastinating with continuous engine rewrites.

>>15845817

A lot more effective for learning, not locked to someone else's bloated botnet nor platform nor language nor have to deal with any of their bullshit or learn their method of development, more control in terms of game mechanics, easier to do things because you know exactly how everything works, better performance and smaller filesize for small/medium sized games, that "daunting time consuming" thing only happens for the first time you do it (unless you change 2d engine into 3d engine later).

>>15848768

>waah wah I won't do it until I'm good enough!

You'll just keep swimming around it like a faggot and taking 50 times longer to get there than you would if you just worked on the actual thing from the start. You'll have to fail trying as you learn anyway whether you do it with the actual game or something else.

>>15848768

The thing about games is that they all build on one another, more advanced games are just a compilation of multiple "smaller games" in one bigger game. The reason to start with beginner games of different genres is that you'll accumulate these basic software building blocks that are used in almost all advanced games and allow you to do pretty much any game later on. It doesn't have to be pong, just search for simple games that you think are fun and have at least one tutorial available.

Anyways, if you want to break down fighting games to the very basics all you really need to know about is hitboxes and frames, see video related. If you want to learn decent quick networking then check out first person shooter networking tutorials as they're typically the games that need the fastest and most optimized netcode, then just replace raycasts with hitboxes and gut anything that's not needed when sending information over the newtork.

>>15848768

build a command interpreter

>>15850486

Have you ever heard of A Chinese Ghost Story? I get the feeling that you would love it.

>>15848985

>>15849068

C++ is a legitimately good language.

>>15848663

Use Unity Entity Component System thing

If you use it right, you get insane performance

Just be advised, it's UNDER DEVELOPMENT ALPHA

That thing is SHIT to use right now, but the devs are doing pretty impressive stuff with it, so eventually when it releases, it will be great.

I am using it right now, mostly to get used to the paradigm. Most of my code will not be able to benefit from most of it's features for quite a long while, probably, maybe it already does and I don't know.

>>15848768

Well, if you want to learn more before you start your game.

Make your pong multiplayer, you are going to need this knowledge to make the fighting game anyways

>>15850341

get yourself a 1800~ fantasy novel that's public domain and use it as the story for your RPG

Like Baron Trump’s Marvelous Underground Journey: http://www.gutenberg.org/ebooks/57426

At the end of the html version there is a licence document. I am no layer, but it seems free to use on whatever you wish.

imagine render culling happening irl

>>15851915

What do you think your eyelids are for

>>15846786

>>15846342

Seeing these posts about the 4th dimension reminds me of something I wanted to do a while back. How can I do for 4d what Doom and Wolfenstein did for 3d?

>>15852741

NIce to know that, anon. Reminds me of the times when devs actually had to figure out things to squeeze the limited power of their consoles

Are you using "Project Base" as a base for your romhack, or just plain Vanilla Super Metroid?

>>15851635

>for your RPG

It's not an RPG. The reason I can't make a story is because I want it to be good. If I were to copy or throw in a generic story, it would be done by now. I've had a vague outline for the setting and story for a long time now. The problem is implementing it.

>>15843876

What's the story behind this RTS?

>>15851510

Even the guy who made C++ bashes it.

>>15852900

I'm not making a hack per se, but learning about romhacking in general. Thus, I'm making tools and parsers and the like. I've almost got a complete room parsing thing done up

>>15853209

Oh right but I do have two eventual goals (even if they'll take a year or two)

>Random shit like replacing item cancel with a beam selector (eg only using one beam at a time, Prime style)

This would be part of a larger hack; but it would open design space into hatches opening or blocks breaking only with a certain beam (similar to how super blocks work). Enemies could be rebalanced to have resistances against certain weapons, since you'd only use one beam at a time

>Roguelike randomizer

As far as I know, there's absolutely nothing like it. There's a few randomizers that replace item locations but that's it. Imagine going through 2, 3, or 4 zones, fighting maybe two minibosses and a main boss, and then taking an elevator up to the surface and calling the ending. It'd be like a minigame but good bait for streamers and racers because it might be interesting to watch.

There'd have to be a huge pool of rooms to use, or else a way to construct them randomly ahead of time. I looked at how HoMM3's mapgen works and I have a decent idea of how it'd translate down into general logic, and then into snes code, but it's still a very large and elaborate goal

>>15851465

I'm watching it right now, thanks for the rec.

I'll shelf the current plot for now, since I got feedback from a friend saying that it was too sappy. I'm also shelfing the machine-world idea for now, since my game system demands magic to be involved even when I haven't figured how magic fits into the setting.

>>15854701

Gambling with money to get random shit, with chance of rare seeds

>>15854701

Decorating. Let the player paint their house, place sculptures, paintings, and potted plants, and buy new paints and decorations from the store.

>>15854734

I only have 16 KB of storage so everything that has to be saved makes the arable land smaller

>>15854740

Yes, that's 16,384 bytes.

Are you going to use all 256 possible values per byte? You might be able to shrink it down to 5 bits per crop instance (32 types) and then use a combination of compression and area trees to determine crop health and growth. 16 kb is a huge amount for a save file

>>15854740

>>15854769

for better space usage it might be better to use a separate LUT that stores the health and crop values, since the player can build similar crops in a row and the tiledata could be RLE compressed fairly easily then, the tile count could be further increased with some tileset specifier in the header (ex starting from u use tileset 0, from v use tileset 3 etc)

or it might not work at all, just tossing ideas

>>15854740

also if you have a lot of EEPROM then you could just stuff a minified zlib implementation(ex. miniz or uzlib) and it should provide better compression than RLE.

>>15854769

> 16 kb is a huge amount for a save file

It really depends on what you're saving, but I'm glad you read something online to tell you what useless shit to say.

>>15854813

>>15854790

There are no libraries and the max binary size is 4 MB. I also don't want someone to be suddenly able to save the game because of a particularly retarded plant layout that can't be compressed enough. Designing a save system that will fail because of completely valid player behavior is a god awful idea.

>>15854769

>16 kb is a huge amount for a save file

Maybe you should wait until you have some idea of what needs to be saved before you start suggesting that something is a "huge amount" for a save file. More things being saved means less room for land. There's no argument against that.

>>15854067

>magic

>machine world

>sappy love stories

You sound like my best friend who I gamedev with, although he does have an idea already.

>>15854067

>Machine world

>Magic

Instead of powering stuff with electricity or fuel you could power machines with crystals or some magic liquid. You don't need the middle ages to make magic reasonable.

>>15856592

Seconding this. Modern settings and magic are not mutually exclusive. Medieval settings are overused.

>>15854067

>I'm also shelfing the machine-world idea for now, since my game system demands magic to be involved even when I haven't figured how magic fits into the setting.

Thief 2 mixed magic and machines and it was really good.

Thanks for all the feedback. I'll keep the machine-world idea then.

I guess I was kind of apprehensive for going techno-magic for two reasons. First, as an amateur writer, I thought it would've been best to just stick to genre basics before moving to more advanced settings. Second, I actually see techno-magic as overused, especially by the Japanese.

As for the former lover idea, for now it's easy to replace that with mentor, friend, boss, business partner or whatever placeholder for the time being.

>>15855093

Don't tyrannize yourself, anon. Negotiate with your inner sloth and schedule a balanced day.

>>15858783

How does someone who doesn't know about objects have anything critical to say about C++?

>>15858783

Less reasonable is C#'s winforms. Out of the box if you want a custom user control, it inherits 4 classes deep and has over 40 variables and events. So wasteful.

I'm reading an interview book out of morbid curiosity. It spends ~200 pages on surface-level understanding to help you hold a conversation, and then has a short chapter on "knowledge" base which teaches you what programming is.

>>15859321

Autists have very strong opinions on C++, but virtually zero have any personal experience to explain why and are just repeating whatever idea infected them first. You have to find the head vampire and kill him. C++ and C# are entirely different languages that are implemented in entirely different ways though

>>15851718

>someone's making a new game?

>oh, it's just fairydev

>>15859321

C isn't perfect either and lacks a lot of things that could make it a whole lot more convenient to use without sacrificing any of the low level straightforwardness that makes it good. C++ fixes some of those things but you get a truckload of shit along with it.

Ignoring the excessive amount of bloat and bad features in C++ that you can for the most part ignore, the bigger problem is how pajeet-tier OOP spaghetti programmers use it and promote their own bad practices as if it's the hottest shit that everyone should do, and make all online guides using those shit features and practices (thus why you can only "mostly" ignore them).

>>15859495

>pajeet-tier OOP spaghetti programmers use it and promote their own bad practices as if it's the hottest shit that everyone should do

This is where a lot of my contempt for programmers comes from. Procedural fags don't behave much better, they just happened to pick the less retarded ideology.

>write simple and fast switch statement with 30 cases for map importer

>each has a line or two at the most

>hear a rumbling and a foul smell

>pajeet bursts through wall behind me

>creates a tile base class and factory

>every case is now its own separate class that inherits from it

>turns 50 lines of code into ~34 classes

>>15859524

>problem that would take 1 or 2 simple functions to solve

>some fucking retard makes a class with 7 spaghetti methods out of it

I've literally seen it before.

How does one learn to get a sense of good vidya music? For instance what is it with the music in monkey island that feels like pirate music (whatever that means), what makes the chimey ice world music feel cold or the soundtrack of sotn hit so many shades of melancholy? How does the above differ when dealing with modern audio hardware vs the gameboy's 4 channels.

Original music is rarely discussed in these threads. Does everyone just use libraries now?

>>15859321

C code corresponds more closely to what you get than C++ does, which makes it nicer for low-level work. It's much easier to learn the language it its entirety, which means less stupid surprises. This is especially relevant once libraries enter the picture: "just use a subset lol" is a nice meme and all, but unfortunately everybody uses a different subset, so this doesn't work in practice. God have mercy on your soul if something you want requires Boost.

There's something I really want to understand though: Why are so many people attracted to C++? Is it this (false) promise that you can do everything for free if you just become enough of an EXPERT PROGRAMMER? I've seen C++ fetishism in /agdg/ since way back in 2012 or so and it's puzzling me.

>>15859706

Your best bet to learn how to make good BGM is to read up and learn about "program music", which was a musical fad centuries ago all about creating music to fit a particular theme or narrative, called a program, hence the name.

Stuff like Vivaldi's Four Seasons is program music, each one designed to bring to mind the spirit of that season.

>>15859321

I messed around with Eigen and Xtensor for a few weeks. The problem with C++ is templates, or the improper over-use of templates. For example, Xtensor is a header-only library that only requires Xlt to compile. In theory. It actually only compiles on a select few versions of gcc and MS C++ using C++14. Even then there are some very very annoying issues to get it to work. C++ is fragmented. C++ has many odd and sometimes obscure features that are not properly implemented across compilers. The standard is simply far too complex with too many things left undefined or up to the people writing the compilers. The result is an absolute minefield. You essentially have to re-create the dev environment of the library developers or risk dealing with some show-stopping bugs.

>>15859836

It's probably because you're still faster doing tons of under performant shit in C++ than you are in most other languages. In some ways, it reminds me of the javascript of the systems world, where it is just everywhere, shit tons of money has been sank into it, pajeet programmers are plentiful, and it has a vast ecosystem of libraries and build tools to make whatever heap of shit you construct run.

>>15860651

FOREVER

>>15859852

That's a starting point, thanks, but isn't the concept a staple in game development by this point? Most games i've ever played had BGM that matched the atmosphere yet I find no reverence to program music in gaming anywhere.

any ideasfags have any cool ideas for things to do with a text console? it can be roguelike related but doesn't have to be.

>>15860815

puzzle mechanic where npcs' past dialogues have to be interpreted on riddles to enter tombs and whatnot.

>>15841545

I think this is pretty cool. Nice work. I like your function pointer use and wouldn't mind seeing more of the system. I tried something similar but much simpler, then punted and now I just use Lua.

>>15843365

>Those are just window libraries. They don't give you 3d just a window to draw on.

No. Qt 5.9 and higher provides a full 3D interface at somewhat higher level than OpenGL and/or direct access to the OpenGL layer, your choice. You can also mix and match, so call OpenGL functions mixed with the Qt 3D functions.

People new to OpenGL 4: I recommend "OpenGL 4 Shading Language Cookbook" as an introduction to OpenGL, GLSL in a sane way. The book also provides a streamlined, sane C++ framework that is logical and quite easy to use for your own programs.

>>15850486

Stop posting.

Start writing the story in a notepad document. Save that document. Don't touch it anymore.

Buy/download a book about programming in C.

Then start deciding about what type of fucking game you want to make. 2D, 3D, text based, hex based, grid based, whatever the fuck. Then start building shit. You will not get to implement a story for a LONG ASS TIME. We are talking years if you don't have any programming experience, and many months even if you do.

>>15861499

basically if it has 50+ upvotes on stack overflow and was written by a white man it's probably a good solution. if there is a 1000 line solution posted by pajeet and has 0 upvotes, you should just skip it entirely. If you use stack overflow for any amount of time you will learn this quickly.

>>15861596

Can you make a 3d game on C? I knew about making 3d graphics but didn't look so good.

>>15861630

This makes me want to run a statistical analysis on stackoverflow answers comparing answers by white names vs pooinloo

Recently, I finally started working on my game.

I want to make a middle age Slavic village simulator. I want to simulate occupation, personality, mood and add as much context as possible.

The end result should be randomly generated RPG with emphasis on dialogs. Players would be given only choices that make sense for this type of character in this conditions.

I have goal similar to dwarf fortress. To generate interesting and unique story. But I don't want to stimulate history and whole damn planet. But concentrate on drama.

What so you think, AGDG. Maybe someone has already made something similar, but I hadn't found it yet?

>>15848784

Open AGDG wiki it has tones of useful resources including ideas for simple game mechanics.

Mostly I implement shitty versions of classic and flash games. With squares instead of real resources.

>>15861993

I think projects under this general idea are not rare, but they never reach a finished state, doing something DF-like is much more work than it seems, even if you keep the graphics as simple as possible.

Still, good luck i would enjoy this kind of game.

Shmup anon here.

After finishing the different UI part (text, sprite/icons, bars), I regrouped everything and… Had to re-arrange framebuffer and VAO binding to unfuck rendering.

Now to link this part with actual data and I should be done for now with the UI.

After that I should have some background, units and projectile, UI rendered to different framebuffers and rendered correctly, I'll have choice between making the background have some panning for semblance of movement, making enemies rotate correctly when following a path, level design, or checking for sound support (music/sound effect)

How can you make topdown exploration fun without combat? The only restrictions the player has is time, energy, and their resources to regain energy. The only real benefits in the game are opening up shortcuts to explore further, and gathering unique resources. Something like a puzzle game but with minimal mechanics. Maybe I can take a few pages from earlier zelda games.

>>15863379

Worked for pikmin

What's a decent amount to pay for amateur vidya music per minute? I could probably get some for peanuts from a desperate starving artist but I'd feel bad. I also have some very tight filesize limits for the console (roughly all the music should fit in a megabyte). Worst case scenario I can just host it online and fetch it from a server every time the cache gets cleared, but I'd rather have it all self contained and not need internet access to have background music

Consider:

>RL/Open world game

>Every item has a weight and bulk associated with it

>STR increases your carrying capacity

>DEX or AGI increases your max carried bulk

>The idea is that you have to raise both to carry huge loads

>Otherwise if you're weak or clumsy you just can't manage

>>15863496

While we're making shit up, can it also be multiplayer?

>>15863496

Consider:

Does this kind of constant elaborate inventory weight management make the gameplay fun?

>>15863496

>>15863522

I can't think of a single game where having to waste skill points on encumbrance or dropping shit on the ground was fun. Maybe diablo-style inventory management, but that was based on size/shape rather than weight.

>>15863522

It would put emphasis on containers, so perhaps there's an enchantment to reduce a container's weight by an amount, or increase its volume limits. It would just be a list of shit, and if it can get added to the container, it will. This way, you don't have to necessarily shuffle things constantly.

Weight/encumbrance is a resource, and RLs are all about resource management. Fight me.

>>15863552

In 99% of cases inventory management only exists to waste the players time. MMOs are the worst case of this, because most of their goal is to waste the player's time to milk more monthly fees out of them. They also serve to limit the amount of resources a player can use/convert at once with minimal effort, or use at once to stop people from progressing too fast or trivializing the economy. Unless you're an extremely rare exception like an autism sim I'd just be extremely generous with inventory space, or at least allow the player to just buy more space.

>>15861596

I'm sorry, anon. I'm on KingJew-mas break, so now is one the only times for me to shitpost. Thankfully, I'll be learning how to code eventually since I'm shifting to Computer Engineering. I also already have a textbook to go over, Deitel and Deitel C++.

>>15863496

Instead of a RL/Open World, why not make it a Platformer with Fetch Quests? You want to keep your inventory as light as possible, but you also want to trade in items for ability upgrades, and you're forced to carry really big/encumbering stuff for certain quests. You also can't take too many quests at the same time, since those might force the player to carry too much stuff and render their movement into shit. Strength will also affect max jumping distance, but Agility would affect movement speed.

>>15863995

That was my least favorite part of college programming classes. Before you could write anything you had to put together a comprehensive UML diagram which would be worth most of the points. The "Is a" and "has a" or "uses a" shit is the most formal, overly complicated waste of time I've ever seen. Luckily, immediately after leaving college, those diagrams only exist in my nightmares.

>>15864175

As s professional Dev I draw those diagrams once in a few months. Those are good when you have read the code and still don't understand what's going on. And when you need to make a presentation. You can't just dump the code in this case. But mostly, yes it's unnecessary.

>>15864363

what are your thoughts on loos?

>>15863522

If it can be automated, yes

Inventory management can be fun if it's about planning and not "juggling shit in windows using the mouse"

>>15863496

It depends on the execution as opposed to ideas alone. You're going to get several mixed answers here because they have a different vision of what it should be like.

>>15863522

People didn't mind it in STALKER or Divinity OS 2.

>>15864175

>>15864363

I don't know about you both, but I can say with pure confidence that 90% of what I've learned in CS was, is and will forever be useless. Everything that's useful and works in real life could be taught in a semester and is readily available online.

I even had teachers tell me that it's perfectly valid to write documentation for a system before implementing it, that I should use one those horrible UML code generators. I even had to go through 6 physics courses, just to "prove" that I can understand how electricity flows in circuits so I could take a course on digital circuitry and answer one question on the last test of the course.

>>15866581

University is mostly an introduction to a discipline. Useful for people who learn best with an actual teacher. That being said, there is merit in the system, since the curriculum is created by people with established knowledge. I can teach myself programming but I wouldn't even know what discreet mathematics are. Some fundamentals are very good to know.

>>15866621

Established knowledge from decades ago from people that have never stepped in the private sector and just fucked around in academic endeavors. It's rare to find actual teachers that have something useful to teach.

Worst of all, the same teachers that teach useless shit try their hardest to make it as hard as possible to pass their courses.

If there's any merit in the system, I experienced almost nothing of it. I can remember about 5 courses that I actually would say that were indispensable and truly made me a better professional, the rest was just reinventing the wheel for no reason at all, or trying to make a square work as a wheel or worse even, trying to explain how we should only show the wheel to the client after 5 years of work without any input from them.

At least it wasn't humanities.

>>15866621

>established knowledge

Have you read any of these CS papers that build off established knowledge? They're all lies with intentionally distorted numbers that nobody checks because the entire point of college is to churn out interchangeable poo in loos or to produce a single contrived paper so you can join the elite club of people who don't fucking do anything. At the very least, if they're just stupid instead of liars, their work is completely useless outside of theory.

>>15859836

>Why are so many people attracted to C++?

For me the reasons are -

1) really good libraries available in native C++. Bullet3D, Qt, GLM: all excellent.

2) not having to pass pointers around quite as much due to the implicit "this" and member functions.

3) standard template library for data containers has it's issues but they're good enough (for me). std::string sucks tho fuck std::string. Not having to re-invent the wheel is the biggest reason for me.

4) well, you know, now that I think about it, I could really use C more and be alright…

>>15866739

Was going to say experts but that was silly. "Already adequately educated people", maybe?

>>15866694

My point is some people (most, probably) need a guiding hand for the first big steps. I do agree that university is overrated, and stupidly expensive in the age of free information.

>>15859321

C++ is versatile, although it can be overtly verbose. It's not a problem if you're competent. There aren't that many alternatives for compiled languages (that aren't too new to have credibility).

>>15866287

So it turns out my problem is that most data in Super Metroid is compressed. There are tools to handle it but they aren't FOSS and I also want to understand the compression algorithm, but I'm fucking retarded. I've seen two different implementations (three if you count the ASM) and I still can't make heads or tails of it:

https://hastebin.com/raw/fepumabopu (CPP)

https://hastebin.com/cazewacoxe.py (Python)

If anyone can offer some additional insight or spoonfeed me, it would be very appreciated, I've been stuck on this shit for literally days. Here's a sample of a 7F length color palette both compressed and decompressed:

https://pastebin.com/raw/qSayMRQ5

>>15866888

Got it working now.

This is really cool! I have no idea how it really works, but it was able to transform 155 bytes of compressed level data into 1,282 bytes of raw tile data.

>>15866835

>C++ is versatile, although it can be overtly verbose. It's not a problem if you're competent

C++ can rape cache, bottleneck with too many syscalls, and destroy compile times if you don't know all the gotcha's the language has.

>>15868397

>not from europe

>english is not your first language

So everyone from mainland europe?

>>15868418

>be a LGBTQIAWTFBBQ disabled woman of color with mental illness feminism

>get money and support just for existing

Sure is white privilege around here

>>15868898

Something is wrong with that hat. It has more texture than everything else combined.

>>15868918

The hat is not celshaded and is using a very detailed leather texture with normal and height maps, we still haven't decided on normal shading or cel shading, when we do i will change all textures to match.

The art style of the game overall is still developing as i get better at making assets.

>>15868931

How far do you have to go on the game? How long do you have before you go homeless?

>>15869008

>if the game fails

The game failing is one possibility, but that's after it exists. Will you really be willing to pack up and wageslave before you have something you can even release?

>>15869012

If we ran out of money before we can release something we will work on it during weekends like most anons do, it will take much longer, but will be released one day before Star Citizen and Bannerlord at least

>>15869021

So, how far do you have to go on the game? And how much time before wage slaving? Your brother is much smarter, you should wage slave for the both of you while he works

>>15869037

He is a better programmer, but he is a lazy bum and only works when pressured, he also has zero creativity so we work well together because we can do what the other hates doing.

And stop feeding his ego, when he gets smug he becomes and lazier.

>>15869059

>And stop feeding his ego

No no, I think he should keep doing that, I like it

>>15869005

Ontario…YES!

So much concern about inclusiveness and teaching race-based-shame. So little concern about teaching critical reasoning and preparing people for the real world.

How To Ruin a Society 101

>>15869183

They're just adding any sexual quirk or fetish in there. Fancy your teacher? Edusexual! Witholding genital contact momentarily to prolong masturbation a sexual identity? You betcha'; Selfcontrosexual!

That list will be 3 miles long by 2020 I guarantee you.

>stuck on an issue inside a simple nested loop for ~30 minutes

>search online based on a hunch

>the OC donut steel language doesn't support nested for loops

>instead just operates on the first object in the list twice and everything else once

>associative arrays don't have indexes so I can't use a regular for loop

holy shit, I hate this

>>15869352

I meant nested "for each" loops, it's not quite that retarded

>>15869439

>new line before the opening brackets

I think you should just kill yourself

>>15869439

Doing them separately is a pain, but it helps with compilation times, apparently. But yeah, your bracket habits deserve suicide.

>>15869515

Forgot to link original post https://stackoverflow.com/a/6647683/8685682

>>15868397

>need only one to be eligible

I have autism, give me money.

>>15868397

just say you have autism

>>15869439

Prototypes go in headers, code goes in .cpp's.

The reason you separate them into different files is to reduce compilation time.

When compiling, you compile each .cpp separately. (These are compilation units.) In order to make the .cpp's interface correctly with each-other, they do need the function prototypes, for which they include .h files.

>>15868397

>>15869538

>>15869553

>people with mental health issues

Well considering how they're always trying to classify opposing views as literal mental illness, I guess I do technically qualify.

>>15868397

Just say you identify as a nonbinary, presenting male

>>15869578

If you play their game you submit your soul though.

>>15869439

To explain more about what >>15869466 and >>15869573 said, it keeps compilation times down because every time you change a header file, everything that includes said header has to be recompiled as well. This is why you should try to write out your header entirely in one go rather than adding a little to it every time.

It's also why you have to keep includes in header files to a minimum as it can cascade into massive recompiles. That's why forward declaring is a thing. For those that don't know: imagine you have a custom class, for instance a vector3 implementation. If you want to declare a vector3 in some header, you could #include "vector3.h", but any changes to vector3.h would cause every header that includes it to recompile. Instead, you just add the line

>class vector3;

to the header where you would want to include it, and add the actual include to the cpp. It's also helpful for solving circular dependencies, where class A needs class B as a variable but B also needs class A as a variable.

>>15869621

You only need to have one of the conditions apply to you, so every non-english European can apply. It's not a completely redundant condition since there's some parts of Africa where they speak English as well

>>15869621

I know you want to spot kike shills but you're gonna need to learn to read first bub.

>>15869515

What exactly aren't you understanding?

He's calculating the displacement along each axis using the equation of motion with an uniform acceleration and then using the Pythagorean theorem to calculate the distance between its original position and its current position. Not sure if this will help but here https://en.wikipedia.org/wiki/Equations_of_motion#Uniform_acceleration

>>15869793

or atelast over the average acceleration.

Meh, who cares.

Hope you understand the concept behind it now a bit better.

>>15869793

I understand the theory now, thanks, I'm still having trouble putting it into C# though.

I'm going to suppose "n" is the number of acceleration events, even though it's not really explained. DT is DeltaTime for the specific event.

It just doesn't seem to be working.

    void Update()
    {
        
        Vector3 acceleration = Vector3.zero;
        Vector3 distance = Vector3.zero;

        for(int i = 1; i < Input.accelerationEventCount; i++) {
            acceleration += (Input.accelerationEvents[i-1].acceleration + Input.accelerationEvents[i].acceleration) * Input.accelerationEvents[i-1].deltaTime;
            distance += acceleration * Input.accelerationEvents[i-1].deltaTime;
        }

        currentPosition += distance;

    }

>>15869951

Make an old game then.

They aren't so bad. Most games I play are old.

>>15869951

Can you make classes? You could use the booleans as bits and build equivalent of ints with methods. Without classes, maybe an array of booleans and functions to interpret and write to them? Without arrays, you could manually make a bunch of bools to constitute your ints (assuming you have a fixed amount of necessary ints).

>>15869889

Sorry for the late answer.

Yes, you have gotten that right.

n is the Number of measurement events, dt is the time between those measurement events.

Without knowing c# I will give you some hints to think about:

Since it seems you are initialising a Vector of Size 3:

- Are you sure your acceleration Events are also Stored as an Array or List of Vectors (or however that is called in this case) and that * behaves like you want it to (should be an element wise operator, but is also for scalar multiplikation)?

- You also forgot to half the Sum over the Acceleration Events.

- And technically you are calculating velocity with the accelerationEvents and not acceleration

- When do you call Update and how many Acceleration Samples are present at that time?

- What is the readout of the accellerometer or your test set?

- Is your Program resetting all your Variables/Measurements before or after Update? or at all?

>>15870388

That'd be almost overkill for normal applications. In reality almost no programming language actually has a single bit for a bool, it's generally an entire byte, sometimes even the same size as an int (ie 4 bytes).

Notably, some implementations of std::vector<bool> are specialized to actually use a single bit per bool, which is one of the few times the C++ STL is actually useful.

>>15870507

He can only use bools though. I was proposing a solution.

>>15841367

Thanks for the response. I found a pretty good tutorial on it and I'm making some progress. TY

>>15870543

For some reason I interpreted his post as constant booleans instead of boolean constant, assumed it meant any bool variable can only be constant instead of all constants having to be booleans, and misread your post as a result of that wrong assumption.

That still begs the question of whether or not his language actually has bools as single bits, trying to wrangle a const int out of bools an entire byte in size would be stupid. And added overhead of having to interpret those homemade constants might be enough of a bother as well to defeat the benefit of not hardcoding all values.

>>15869889

>>15870478

To clarify the last questions:

Because the code only works if at least 2 acceleration Measurements have taken place (accelerationEvent[0] and accelerationEvent[1])

And if you reset those eventcounts and events, else you will always repeat ALL prior events.

You could also think about using a timer to trigger or atleast delay measurements to not lock up your programm with Interrupts, and you should define some deadspace where measurements might be discarded, oversampling can also help and what should happen in case of hitting an obstacle or if the state velocity = 0 and acceleration = 0 can be reached with your setup.

>>15869951

I don't remember taking that screenshot

>>15871264

Use namespaces, faggot.

>>15840140

Start with Unity, there's a lot of documentation about it.

>>15868397

>tfw you fulfill 3 of those categories

I need to sign up for this shit lmao

>>15870227

>>15870388

>>15870543

Do you goym not know what a constant is? It's not a variable I'll tell you what

>>15871809

I think Forge encourages you to design around network commands liberally. In single player mode, it skips the serializing/deserialized/network step and just "receives" the data as if it had been received over the network instantly. In practice, a single player is the server and client. The advantage is setting up your event and animation system in single player mode in a way that's already formatted for online play.

Do you actually need Odin Serializer? Forge remastered has a Unity editor window that lets you assign parameters to remote calls, and then compiles it into a class that can be send or deserialized automatically.

>>15872161

Oh shit you're right. Wait, so all values have to be hardcoded bools that can't change? So there's no interactivity at all with the program?

>>15872683

nigger, go learn C or something

>>15873005

I know C.

I've got a moving circle moving into a static one. I want the moving circle to slide around the edge of the obstacle. I did this by calculating where it'd be if I pushed it away, then adjusting only its x or y based on the direction its moving. This seems to work, but it moves very, very quickly along the edge of objects because your movement is essentially being adjusted by a sin or cos wave. Maybe I just need to make the primary movement alongside the circle slower based on what part of the co function you'd be on. I don't know much math aside from sin and cos

>>15872392

Hosting source online only isn't too wise, yeah. I self-host my services so (((somebody))) can't shut us down or cause problems. Was it a similar case to what we were doing?

>>15872365

For the record, I'm still fairly new to Forge. I did the basic ball rolling tutorial just to see something working and that was about it.

I know about the Network Contract Wizard, but I haven't quite figured out if it's able to work with non-Unity base classes or how I can make use of it to simplify my architecture (though most of that is not network related).

I don't know if Forge's built in serializer supports this, but I kind of wanted all of the data to be easily serialized and loggable to a file so the entire battle can be recreated (assuming the build is of the same version), just because I want either A) The potential for replays and moreso B) the ability to look through history for balancing and playtesting easily. Odin is pretty powerful (not sure about Forge's) and I'd assume it could support serializing individual classes inheriting from abstract command classes separately, and also does not require Forge to write this data to a file. Note I still need to test this though and maybe I'm building on something I don't know. I would like to simplify that so it's not so long and complex, but I have no idea if I can easily remove or reduce classes without adding a bunch of extra work elsewhere.

>The advantage is setting up your event and animation system in single player mode in a way that's already formatted for online play.

I figured we'd just send commands that tell the client's setup to animate x or call event y (for local and visual stuff like UI), and let the math and data management be done by the server. I can't tell if that's meant more for data being in the inspector or if it can access raw classes. Though it sounds like you're saying maybe I could do something like this?

public class Combatant : NetworkWhateverForgeHas {
    List<ConditionInstance> conditions = blah;
    public void AddCondition(int id, int turns) {
        // create new condition instance (raw class) and give it id and turns
        // load condition from database
        // call OnCondiApplied from combatant
        // call OnApplied from the new condition, in case it has code for that.
    }
}

… and just set it up the Combatant class in NCW to have AddCondition with two int arguments? I guess my only concern with direct calling like that is it sounds like you're sending instant commands and also not encapsulating those commands so they can be saved and reloaded.

Man, I probably need to get more familiar with Forge. Getting the exact info I need had been a pain from the documentation, but I'm probably looking at the wrong parts.

>>15874309

Last i checked the forge documentation was shit, wrong, and unfinished

good Luck

e l e v e n y e a r s

Why are so many never ever fags making these 2d RPGs? Like bob's game or handmade hero as well. I mean this guy seems to know stuff but never gets anywhere despite endless time spent.

>>15876139

Prob has a job. Most 2d games back in the day where made by a team of employees working full time, even overtime. These are just dudes doing shit in what little freetime they have.

>>15876723

pythagorean theorem. You know the apparent "height" of z, and the real length of z. Use that to get the third side, and once you know all three sides use the trig functions to get all angles, and then alpha will be the compliment to the one between z and z's "real" length.

>>15876743

>>15876758

Surely that would only work if the observer was orthographic or otherwise doesn't take perspective into account

The plane bisected by y (your green line) would have to pass through the cameras origin when extrapolated, meaning the inscribed triangle wouldn't necessarily have a right angle

>>15876758

i think you want the pythagorean theorem: a^2 + b^2 = c^2

so if you know the height of orange and the depth of green, orange^2 + green^2 = black^2

accordingly, if you know c and need to find the value of a or b, you can run a reversal of the theorem: c^2 - b^2 = a^2

>>15871234

Can I get more pics on your waifu stalker? Is it like that guys pic related.

Have we heard back from the Aethyr dev?

>>15876139

I watched the first part of this, and he had an issue with file icons on the desktop causing lag, complaining that they were dynamic images instead of static; what exactly does this mean? That the raster image is being recalculated every frame, or something?

>>15871201

>>15872315

It great to see progress on your engine man.

>>15877558

Add flight controls and a catch-fish mechanic.

>>15877635

Better vid. They've gone from youth to middle age and they're still in the engine dev trap. They have all these people and all this kickstarter money and there basically is no game. They just added the dialogue system. A bland and ugly generic 2d RPG will be this dude's life's work and no one will care. What's even the point. Sorry to be so obsessed with this but I'm just morbidly fascinated by this stuff. I was already freaking out over this years ago. What are they all so busy with? Mind boggling.

All of the decent/technical GDC talks are behind a $500 paywall. Is there somewhere I can just pirate them?

>>15877828

>engine dev trap

They already explained what they're doing. Game engines are shit and they're never going to get any better if no one actually knows how to make them. If the world were infested with shitters like you, everyone would be using Unity for the next 50 years.

>>15877807

That's way too easy. But okay will do.

>>15877871

Exactly what part of what they have made looks good or ground breaking to you? If they used an engine they would have been done 10 years ago. I know they didn't do that because they're infatuated with the dreamcast but there's basically no reason for that either, it's like how bob decided his game HAD to be on the DS and then chimped out when nintendo didn't want to talk to him. I think they're mainly using engine dev as an excuse for why they can't just make the game already.

>>15877852

>tfw trying to push my Super Metroid editor into a randomizer that functions as a roguelike

>>15877958

>>15877871

Ive learned a lot recently. How to manage a UI application better, invoke other exes via Process.Start(), and wrote a SNES rom decompressor. Just like make whatever and never stop learning

>>15877958

>If they used an engine they would have been done 10 years ago.

>>15877958

Robert Pelloni was actually accepted by Nintendo, provided he'd take a team of theirs and make changes to the game, which would also include changes to the story. He didn't like that, declined, but the protest was mostly staged as marketing.

He eventually did start work on a PC port, but by that time the game had been ported from GBA to DS to PC and the source code had become such a mess that he got stuck and eventually dropped the project when his mental state deteriorated and he became homeless.

And now he dislikes games after his "enlightenment", aside from simple entertainment toys, which is why he now just focused on developing the puzzle game which is available for free. Not that I've heard from him in over a year now.

tl;dr don't talk about "bob's game" if you don't know shit about it.

>>15878028

you're trying to reason with a jungle nigger with a slave mentality

Any decent IDE's for c/c++? I was using JetBrain's IDEs and they are fucking amazing but I don't want to deal with that subscription fee. I tried emacs but I was spending more time configuring it than actually coding.

>>15878076

I like code::blocks

>shared_ptr<Piece> Grid::dontPutSameColorDiagonalOrNextToEachOtherReturnNull(shared_ptr<Piece> p, int x, int y, ArrayList<shared_ptr<PieceType>> &pieceTypes, ArrayList<shared_ptr<BlockType>> &blockTypes)

t-thanks bob

>>15878076

>I tried emacs but I was spending more time configuring it than actually coding.

If you don't want to do a lot of configuring then Emacs isn't for you. Are you looking for an IDE or a text editor? There's a big difference.

>>15878082

I used it a long time ago. I don't remember it having anything like jump to deceleration, find all usages of, or auto complete. It was a very long time ago though.

>>15878091

I've been using a text editor for years. I finally started using a real IDE and realized how much time was wasted without basic stuff. I think you can implement all the things I want in emacs, but everything script I used felt super janky and it was really slow for no real reason.

>>15878092

It got a lot better and I've even managed to compile vulkan shit with it. They've been feature creeping like crazy at least on the linux version though. It literally has valgrind built into it now.

>>15878096

I kinda avoided IDEs because a lot of them felt like they just crammed whatever gadgets they could into it. Valgrind for example you can just run from the command line. I really just want something simple with the navigation shortcuts that isn't vim

>>15878093

Here's a guide on setting Emacs up for C/C++: https://tuhdo.github.io/c-ide.html

There's starter packs for Emacs if you don't want to config too much. Emacs is archaic and has a lot of absurd defaults. It's slow, prone to crashing, has a horrible API, and takes a long time to use well. It's an investment.

>>15878110

When I got into emacs I already messed around with starter kits and stuff. I rebound everything to make sense, like ctrl+f for find. What ultimately turned me off is how janky it felt and how ridiculously slow it was. You can't configure those things away.

>>15878099

I'm using C++ and SFML for my game. Using the command line for compiling on both Linux and Windows. I've been using Emacs since like 2007 and have a massive config that I'm content with so it works for me. If I need IDE features, I edit using Emacs and switch to the IDE. I do this at work where I need to use Visual Studio.

Getting into Emacs is daunting which is a shame. If a few minor issues were fixed, the userbase would explode in a short amount of time. The developers are completely out of touch. Unfortunately, there's no replacement. If I wasn't swamped, I'd be working on one.

>>15878118

It still loads faster than an IDE and it saves time in the end by editing text more efficiently. IDE features don't suffice when the user is stuck with a notepad style text editor.

>>15878140

>Unfortunately, there's no replacement

Not 100% sure about that. JetBrain's IDEs have god-tier shortcuts. They're just slow as shit.

>>15878099

Qt Creator is pretty good and can be integrated well with gcc/gdb/valgrind. I like the keyboard shortcuts. Having 21st century text editing and code editing features make leaving emacs or vim behind fairly logical. Code completion and reasonable views of the code structure, as well as highlighting .. I really have no idea why people use vim still besides being stuck in their ways.

Typing code in isn't the bottleneck. It's quickly and easily understanding what is going on. IDEs bring tools to do this that console text editors simply cannot.

>>15878076

4coder. You don't need an IDE, just a text editor and compile from the command line. Or with 4coder you press alt+m and it runs any batch file called 'build.bat' in the current directory.

>>15878193

>Code completion and reasonable views of the code structure, as well as highlighting

Emacs has all of this.

>Typing code in isn't the bottleneck.

Lacking in text editing functionality is a huge bottleneck. Having to constantly switch between the mouse and keyboard is a huge drain, especially on a trackpad.

>IDEs bring tools to do this that console text editors simply cannot.

Emacs is not just a console text editor. It has a graphical version which is how it's intended to be used when the environment supports it. Being able to use both a graphical and console version is an advantage over most IDEs. It can be used for all text editing with the same config seamlessly as opposed to IDEs which are specialized. Emacs can be extended without recompiling which most IDEs don't support.

>>15849673

How can you not make some generic shit out of your ass.

Alternatively, read some books, it will come to you then.

>>15878238

Didn't know that about Emacs. Cool.

I agree, not having good keyboard-only navigation while writing texts sucks. Qt Creator is pretty good at this though. I don't really need the mouse all that much.

I like to see class hierarchy lists and members listed with their types, as well as my debug breakpoints and custom variables in sidebars.

I'll give Emacs another chance and check out the graphical version.

Right now I'm checking out sublime. It's pretty amazing. It has a license fee, but there isn't really anything in place to force you to pay it (kinda like factorio). If it's good I'll still give the devs my sheckles because a good text editor needs to be supported

>>15878360

Something as fundamental as a text editor shouldn't be proprietary.

>>15878375

>Fundamental

You say that, but it's near impossible to find a decent one. Sublime looks like it has pretty much every navigation feature I want without being slow or having bloat out the ass.

>>15878397

I mean you have your superior foss text editor, so why would you be care if there's an inferior proprietary one? Like I said. I didn't want to spend 15-30% of my coding time configuring emacs to work exactly how I want, and just want jump-to and various other navigation methods. I also have a really shit computer so I feel the slowness a lot more than you might.

>>15878302

The lazy way is to have two same objects on two different light layers/masks, one invisible that casts a shadow and one visible that casts no shadow. For the not so lazy way you have to learn shader magic.

>>15878405

>I mean you have your superior foss text editor, so why would you be care if there's an inferior proprietary one?

We must secure the existence of free software and a future for the technologically literate. Encouraging people to use free alternatives to proprietary software is the ethical thing to do. If you've ever been coerced to use/make proprietary software, you know what I mean. Having a bigger user base indirectly leads to support which benefits me and other users. Proprietary software should not be getting that support.

Emacs is my operating system. The only 2 programs I ever have running are Emacs and Firefox with Keysnail. I know its flaws, I'm looking for an alternative too, and I can't blame you if you go with something else. Back when I was getting into it, I quit several times out of frustration. I kept coming back to it because I craved its functionality. Writable dired is just one of the things you can't live without once you get used to it. The learning curve is painful and it takes a lot of time to get it working properly. There's no getting over that but you can eventually get to the point where you don't have to worry about it.

>I didn't want to spend 15-30% of my coding time configuring emacs to work exactly how I want

As opposed to not even having the option?

>I also have a really shit computer so I feel the slowness a lot more than you might.

Having a faster computer doesn't make Emacs much faster. Emacs isn't like VI where you're supposed to have it installed on every computer. The idea is that you start Emacs when you start your computer and only close it when you're shutting it down. If you need another instance of it, just make a new window. C-x 5 2 by default.

>Sublime has very slight things off about it

Fuck it. I'll give emacs another shot.

>>15878522

I'm going to try out emacs again because I ran out of options but you have one big flaw with your problem:

>Having a bigger user base indirectly leads to support which benefits me and other users. Proprietary software should not be getting that support.

Yeah, it gets flooded with retards who can't code and bloat the fuck out of the codebase. This is a problem plaguing free software, and it will innevitably kill it.

>The learning curve is painful and it takes a lot of time to get it working properly.

It's not the learning curve. I don't want to switch back and forth between emacs and other IDEs I'm required to use and the little oddities with emacs.

>As opposed to not even having the option?

Probably. If emacs still becomes a time hole eating up productivity I will drop it again. I don't want to be a no-dev forever because I'm constantly tinkering with my editor.

What's a good engine for messing around with a 3d platformer? I don't have a specific idea for a platformer in mind, I just want to be able to set something up and futz around with it.

Side question: if I knew exactly what kind of 3d platformer game I wanted to make, how much trouble would it be to create my own engine? Would I be better off saving myself some work and using something like unreal or unity anyways?

>>15878594

>What's a good engine for messing around with a 3d platformer? I don't have a specific idea for a platformer in mind, I just want to be able to set something up and futz around with it.

Unity

>if I knew exactly what kind of 3d platformer game I wanted to make, how much trouble would it be to create my own engine?

More than you could possibly imagine.

>>15878607

He said 'good', Anon.

>>15878617

There are no good engines that let you fire up and screw around. Unity is perfect for that.

Someday I'll make a replacement for Emacs. It'll be lighting fast, minimal, and have a custom extension language with a solid API. The user will build it from the bottom up, not the other way around. It'll have sane defaults so that new users don't have to spend ages configuring just to begin using it. It'll focus on the features that are used daily as opposed to being an endlessly expanding bloated mess.

>>15878630

You'd be doing gods work tbh. Emacs is a lot like Dwarf Fortress. It kinda sits there a bit neglected but has insane potential, then some people borrow a lot of its ideas and blow up (Minecraft, Rimworld, Rust, etc). I think this is why I gave up on Emacs before. Because I learned Dwarf Fortress isn't actually that fun

>>15878630

You better do this fag, and in a language other than (((Lisp))).

>>15878656

tell us everything that's wrong with Lisp

>>15878617

Unity isn't bad, 99% of the retards that use it are.

>>15878661

See >>>tech/952400

>>15878670

Unity is a shitty engine using a shitty and slow language. Shit attracts shit-eaters.

>>15878670

anon, using reason and logic doesn't work; they're ideologically driven and only hear what they want to hear.

>>15878685

>Opinions are logic and reason

But I do agree

>>15878685

My dude, Unity is a poorly made engine developed by SJWs and faggots, using a shitty (((garbage-collected))) bastardization of C/C++, made by pajeet-tier coders at Microshaft. That is not opinion, that is fact.

>>15878432

>>15878459

The lazy way is not gonna work in the long run.

Guess I'll learn shaders. If it's stuff like using that #ifdef it's probably simpler than I though. Thanks.

>>15878594

I think some dude remade mario 64 in Unity and released the code under "super character controller" if i remember correctly which is a pretty good base to use if you want to make a basic 3d platformer.

>>15878647

>>15878656

The extension language already exists. I made it to make stages in my game. The challenge is setting up a GUI for the text editor portion. I'll also have to write a ton of string manipulation functions. The language supports strings but it doesn't have a lot of functions. The syntax is like (((Lisp))) but the language isn't a (((Lisp))). It doesn't use a linked list as its primary data type and has no car, cons, or cdr. Here's some code:

(loop (= q 0) (< q 10) (++ q)
   (when (odd q)
     (printn q)))
1
3
5
7
9

>>15878661

Here's a pasta that describes some of my gripes with it.

Pundits of Lisp like to say that syntactical purity is one of its perks. If that is so, why is it normal for Lisp implementations to have so much syntax? Why is it that * in C is wrong but ' in Lisp is correct. Lisp has sigils(reader macros?) for quote('), unquote(, or ~), splice(,@), function(#), char(?) and so on. There's special syntax for defining functions with optional arguments or variadic arity. Cons cells are usually defined with (a . b). Why not (. a b)? Even values have syntax. Consider this list: ("string", 123, 1.23). Is "string" a string or a symbol called "string"? Is 123 an integer or a symbol called 123? How do we know 1.23 is a float? With pure syntax, wouldn't those values have to be defined explicitly? In other words, they would be evaluated as symbols and converted to a proper type. Lisp users are clearly not above "syntactical sugar". In that case, what is the extent that it is appropriate to use? If you keep tacking on more syntax, you'll end up with something that doesn't resemble Lisp. The languages in this category tend to be more popular. Can it be, in that case, that programmers in general enjoy syntax? On a scale from Perl(dense syntax) to Lisp(sparse syntax), most popular languages are in the middle. The typical programmer doesn't like Perl because of its abuse of arcane sigils. The same programmer doesn't like Lisp because of the overbearing amount of parentheses. Programmers feel comfortable with a language that is light enough on syntax that it's easy to pick up and read without memorizing symbols but heavy enough not to be forced into dealing with an AST. Someone with experience in C-style languages can pick up PHP and Ruby easily, but Perl is incomprehensible to anyone that hasn't gone out of their way to study it. Likewise, Lisp is hard to read due to its lack of syntax. 1 + 1 is easier on the untrained eye than + 1 1. What is the ideal balance between syntax and syntactical purity? In my opinion, a true Lisp has never been tried, but most programmers wouldn't like it. Why are macros a feature when the explicit nature of the first element of a list being a function and the rest being its arguments is so fundamental? Isn't that something that a Lisp programmer would avoid?

>>15878724

Well, opinions can contain reason, and logic (f.e. an argument from a subjective opinion).

However, an opinion doesn't have to contain reason or logic, but it can also include either/or.

They're not mutually exclusive concepts.

>>15878741

>spouts mostly opinions

>that is fact

sure thing kiddo

>>15878791

Not planning on doing it in Lisp. Still deciding if I'll do the whole thing in the language or instead implement it in C++ and use the language as an extension.

>>15878830

Theres always godot. Get in early before someone makes something really cool with it and all the shitters flock to it and make it a meme.

>>15878830

>Unity is a poorly made engine

But that is true. It's horribly slow and incredibly bloated. It certainly doesn't help that newbies and retards use it to maek gaem, but I doubt competent users could do well either.

>developed by SJWs and faggots

But this is true, you faggot. Look at the devs' profiles and see what they look like.

>shitty (((garbage-collected))) bastardization of C/C++

But that's exactly what C# is. Garbage collection is the last thing you need when you make a game, unless you really like playing at 10 FPS.

>made by pajeet-tier coders at Microshaft

I hope don't need to explain this to you.

>>15878963

You seem pretty invested in this, sorry bud, don't really care.

>>15878963

>10 fps

That shit freezes games most of the time lol. 0 fps.

>>15878976

>You seem pretty invested in this, sorry bud, don't really care.

Why'd you reply then, faggot? >:3

>>15878991

You get my point, my dude.

>>15878963

>poorly made engine, horribly slow and incredibly bloated

Unity isn't horribly slow and bloated, that's Unreal you're thinking of. Especially in the latest versions ever since Mike Acton got to call the shots regarding optimization, Unity is pretty damn fast and becoming more so with the upcoming features such as the finalized version of the ECS.

>developed by SJW faggots

Quite a few of them are, yes. Luckily they have competent people up top, with the faggots being in charge of documentation, tutorials, customer support, the (((asset store))), and the like. That's what pretty much every tech company looks like now, a hardcore core of turbonerd programmers that talk to almost no one in the company but make most of the contributions, and a big swath of fags around them.

>garbage collected bastardization of C/C++

Garbage collection doesn't matter in 99% of all games, only the ones that are so performance critical that they try to wring out every last bit of performance out of a PC. It certainly isn't going to matter for (((you))) though. Hell, it doesn't matter for the anons with 10 gorillion blades of procedural growing grass, which even while not nearly as optimized as they'd like still runs at more than acceptable framerates.