/agdg/ + /vm/ ~ Amateur Game Development General

Damn you and damn your mustard rice

Lel

I haven't been able to post in several /agdg/ threads in a long time.

It's still very sloppy (and I suck at action games, why I am making one is a big fat question mark), but I'm trying to get as much feature complete before I get stuck in an endless tweaking rut.

Todo list: I plan to have boss monsters with dismemberment and homing missiles and other EX skills.

Have fun getting support or updates for your game for all those who still use the first game maker studio.

There's more than one way to use the depth buffer.

Fuck, you're right.
Those cunts are gonna stop supporting the first GameMakerStudio, aren't they? Fuckers.

Reported, you degenerate.

You should do something about the firing noise. It sounds a bit flimsy.

So is this being done in two passes or something? Draw the wall geometry first, and then do the tree decal?

Like, my understanding is that you can't change render settings mid draw call, and that if the depth buffer was fucked up, then you'd see hidden geometry drawn on top. Or… was the depth buffer disabled by default and it's up to the user to use it? Again, I'm using XNA and using all these effects is still pretty new to me

just upgrade to unity3d already.

What's a good open source 3D engine with good netcode. Is godot 3D good now?

Well yes, the decal is another object.
It's using Z offset to fix Z-fighting, then in the fragment shader, it clips if the underlying Z is too far from the fragment being rendered.
That requires support for reading depth in the fragment shader, not sure how common that is. This was made in Unity, using the old OpenGL 2 renderer. Should work in D3D9 as well, if that's what you're using.

That's what would make sense. So if you ever plan on selling your GMS game, you have to buy GMS 2, or else no support for you in case of problems. For example if you need to integrate steam on your game, I don't think they will bother supporting old game maker versions in the future.

I pushed version 0.0.1 of a visual novel I've been writing for a while. All artwork is either non-existent or a placeholder image. The main thing I'm working on now is getting Act 1 completely written. Then I hope to get some real assets, design the GUI well and have a functioning Act 1 demo around 6-8 months from now.
windows link: lit.statseven.com/bedaiowin.zip
mac link: lit.statseven.com/bedaio.zip

I'm just gonna be the first one to say, who gives a shit?

Game Maker Studio is technically terrible and the fact they have the gall to ask $100 for another undoubtedly badly designed and programmed engine is just ludicrous. There are various other completely free/open source game engines out there that the even idea of using proprietary crap like game maker is pushing on the absurd.

Pics related: All FOSS, all 3D capable, all better than Game Maker.

Yeah, its 3D is quite stable now. They recently released a new high level networking framework which is apparently quite good, too.

botnet + trojan

The music is really nice. And that room you started in looks pretty good, better than how it looks outside. One suggestion that I have that's bothering me though, I don't think the game needs dynamic hair and clothing because it looks very weird. Animation for the hair and clothes probably look better and more professional.

It's not at all. Is there anything I can do to prove it's legit?

phaser and godot are quire good. not like i've released anything with either but they're quite good. guite gud.

post your tiddies

...

Yeah, whatever. I did shit on them for bringing up GMS 2, but GMS is not terrible. Everyone who I asked why game maker is bad did not make any game with it. Until you prove to me that you used it's features I don't have to consider your "arguments."

just post them tiddies

...

You can play my game and ogle the placeholder characters. They have nice 2d tits.

Are your tiddies in your botnet?

Back to cuckchan with your shitposting. That's your argument, a greentext. Fuck you, and your wife's son.

There's no botnet. You want to see a script of the game to prove it?
lit.statseven.com/literature/VisualNovelScript.pdf
This contains about the first half of the game.

Yeah, not downloading your trojan, Shlomo.

But Gorski isn't a Jewish name at all. It's polish

We all know how much the poles loved jews in the past, kike.

Just a jape friend, just a jape!

But it's not a jape!! It's a real game. I really would like some good criticism on the writing style. I'll play somebody else's shitty game if that's what I gotta do get some critique.

I was implying the anons calling it a botnet/virus/dolphin-porn/Win10-upgrade-key were merely pretending

I really need to find a way to dynamically mask out specified polygons on a mesh in Godot, like I can do in Blender. Otherwise rigging thus is going to be like a repeated kick in the nuts.

Can you please stop shitting up random threads with your false flagging shit Holla Forums?

Hey my favorite fellow goys,

What screen capture / video recording program do you guys recommend?

I'll post my shitty progress but I don't have anything to record videos with.

pref free cause no $$

none

I use Kazam but it's Linux only. Apparently Open Broadcaster is good though.

alternativeto.net/software/open-broadcaster-software/?platform=windows

OBS is probably your best bet
then just use ffmpeg for converting to webm

Noted. Will get a collection of gunshots, normalize them to sound like the same gun, then randomly choose the right noise (with criticals being better pitched.)

Thanks. Current music is royalty free. Though I modelled 80% of the environment assets, the bedroom used stock UE4 textures (high-res). Outside I had to improvise from concept art.

I actually don't have any budget for animation.(remember the 2000 dollarydoo edition thread, yep I've been soured on using "professionals" since.) Been using mocap plus the help of a guy here. I don't mind the cloth physics.

The teleporting is bugging out back to the totally original did not steal pokemon town.

The goblins don't attack the player yet they just forage around and the attack animation isn't direction based (yet) nor does it render properly (the animation loops more than once) But I did factor in delays so it can't be spammed although it needs to be fine tuned.

But here you go guys.

Also to

Thanks a bunch!

Probably not the best use of my dev time, but since I didn't really feel like coding tonight, I managed to squeeze a ton of decal data into a custom 8-byte format.

So basically, while the dungeon is made of tiles, there are 10 locations where a texture can be drawn on a quad within that tile. I managed to condense data or use lookup tables to keep things simple, but expansive. Basically, every rotation and scaling has 16 degrees of happening, or 32 if I flip the appropriate flag.

Pic related is how I might construct walls/quads from the data given, and how it might look in the game.

why not have a pointer to a list of decals as a field? That way its more scaleable.

Actually, it should be scaleable the way it is now. Maps can (technically) be as large as 256x256 tiles, but realistically will probably only be 32x32 or so. The texture is just pulling from the texture map/atlas that I set up. Otherwise, decals are just an arbitrary list of "additional textures" that get drawn; nothing will reference them, unlike map cells or textures. They themselves aren't indexed; map tile is rather which map cell it gets drawn in.

I could probably jam like 100 decals into a tile if I really wanted to, but that wouldn't be a very good idea in general.

Hey, that was actually pretty easy.

ShareX, it's like a very advanced version of GifCam and you can record directly into mp4 or webm. Might need to tinker with the settings a bit though, for example it uses mp4 and too high bitrate and starts recording immediately by default.

so i'm trying to fix my city map thing from yesterday
someone told me my array was what was causing objects to get duplicated
how do i make a dynamic array of pointers?
is this right, because right now the program is crashing and i'm not sure if this is the reason why
template class List{protected: //array of items T **items; //size of array int size;public: //constructor List() :items(nullptr) {} //add item function void AddItem(T *_item) { //creates temporary array, 1 element more than the current item array T **temp = new T*[size + 1]; //adds current items to temp array for (int i = 0; i < size; i++) { temp[i] = items[i]; } //last item is the new item temp[size] = _item; //increase size size++; //set items array items = temp; } //remove item from item array void RemoveItem(T *_item) { //check if item is in items array if (Contains(_item)) { //index of item int index = 0; //get index of item for (int i = 0; i < size; i++) { if (GetItem(i) == _item) { index = i; break; } } //temp array T **_temp = new T*[size - 1]; //copy items from items array to temp array, excluding the item for (int i = 0; i < index; i++) { _temp[i] = items[i]; } for (int i = index; i < size - 1; i++) { _temp[i] = items[i + 1]; } //set items array items = _temp; //decrease items size size--; } } //get item from items array by index T *GetItem(int i) { return items[i]; } //return size of items array int Count() { return size; } //check if item is contained in items array bool Contains(T *item) { for (int i = 0; i < Count(); i++) { T *_t = GetItem(i); if (_t == item) { return true; } } return false; } //reverse the order of the list void Reverse() { //make temporary array of the same size T **_temp = new T*[size]; for (int i = size - 1; i >= 0; i--) {//add the elements backwards _temp[size - i - 1] = items[i]; } //set items array items = _temp; }};

...

They're not serious are they

I don't know about your problem, but AFAIK you shouldn't create a new array every time you add an item. A common method is to double the array size whenever it goes over capacity, and only move the contents when that happens.

std::vector

the issue last time was that i was trying to compare pointers, but because of the array, they were getting duplicated, so even if i think i'm comparing the same objects, they're actually different
and i'm supposed to fix that with an array of pointers
and naturally i want it to be a dynamic array, so it has to be a pointer to an array of pointers
ignoring what you said about changing the size, i still need to figure out why it crashes when i do it
right now, if i add a few items to a list, and then try to print something about them, the program crashes. if i just add them it doesn't crash

hm, actually it looks like the list is working fine
it's the crossroad class that seems to somehow be fucking up

For 50 doodoos it better support Windows, Mac and Loonix out of the box if it wants to compete with Unity and UE4.

I use Dxtory and Fraps. Neither of them are free, though, which I suppose beats the purpose of the question.

Don't feel like posting the character model I'm working on until it's in better shape, but just wanted to say thanks anons, I've had a lot of hard stuff going on in my life lately and seeing you all hard at work in this thread really helped me get back to work myself.

If you've seen Game Maker 1-8, then you'd know they are dead serious.

I did that 2.5D jumping which was a problem for me. I just added a variable yflat which would be the movement for normal and for depth and y= yflat during jumping y could simply change and the depth wouldn't change. Pretty easy.

What changes did they made? Anything major?

From what I've seen so far the UI is significantly improved, the sprite editor is much better, rooms can inherit from one another, and the scene editor has new features.

I haven't seen anything that'll revolutionize your game itself, but the actual game making process is much improved.

alright, i fixed that, one of the classes was somehow fucking with the pointers

now that shit works, how do i get the shortest path in a graph of linked nodes? i started working on this with the idea of A* until i realized i don't have coordinates
how do i change the code to work?
List * Path(CrossRoad *_start, CrossRoad *_end) { List storeIn; //start at the starting node CrossRoad *cur = _start; //list of trios TrioList CameFrom; //add the starting node, it comes from itself, 0 cost CameFrom.AddItem(new CameFromTrio(_start, _start, 0)); //frontier, contains all the nodes we're about to check List frontier; //add starting node frontier.AddItem(cur); //do cycle until frontier is empty while (frontier.Count() > 0) { //remove cur item //if we had coordinates, we could use a priority que to remove the item closest to the goal, would make the process a lot faster cur = frontier.GetItem(0); frontier.RemoveItem(cur); //if we've reached the end, reconstruct the path if (cur == _end) { //add end storeIn.AddItem(cur); //get number of connections; //we're not necessarily going to check everyone, but we need a limit int cost = CameFrom.Count(); for (int i = 0; i < cost; i++) { //next is the element that cur came from CrossRoad *next = CameFrom.GetKey(cur); //add next to the path storeIn.AddItem(next); //if we've reached the start, reverse the path //now start is at the begining of the path, end is at the end if (next == _start) { storeIn.Reverse(); return &storeIn; } cur = next; } } //get neighbors List *neighbors = cur->Neighbors(); for (int i = 0; i < neighbors->Count(); i++) { Neighbor *_next = neighbors->GetItem(i); CrossRoad *next = _next->key1; int newCost = CameFrom.GetCostSoFar(cur) + _next->cost; if (!CameFrom.Contains(next) || CameFrom.GetCostSoFar(cur) < newCost) { //add next to frontier frontier.AddItem(next); //get distance to next crossroad //add next to camefrom CameFrom.AddItem(new CameFromTrio(next, cur, newCost)); } } } return &storeIn;}
and i think there's also an issue with the frontier, i seem to be checking some nodes more than once

actually nevermind, i fixed that, apparently i had to remove the item from the frontier with the shortest path, not at random

Yeah, ok, the editor was shit but was it SO SHIT it deserves 50 bucks? And that's because I already have studio, if you're a newcomer you need to pay 100 bucks.
The new collision system (which is allegedly tenfolds faster) and the new room editor are cool, but it doesn't really help in making things less cluttered. No objects without useless, bloating default variables, no custom events\methods, no UI system, no variable types, no pointers so that I can instantly know if a variable is holding 15 or if that 15 is actually a data structure.

Now, let's say Jimbo McNoDev decides he wants to make videogames, what are you going to do, use GMS2's "limited resources" version because obviously you're not wasting 100 bucks on a software you don't know, or do you go to Unity and get 100x the amount of content for free?

ok, why isn't this working
void testTextFileInput(List *streets) { ifstream text_file("Map.txt"); char line[100]; if (text_file.is_open()) { CrossRoad *tempCR; CrossRoad *tempCR2; while (text_file.getline(line, 100)) { string SourceName = ""; int dist = 0; text_file >> SourceName; char endline = ' '; tempCR = new CrossRoad(SourceName); while (endline != '\n') { endline = text_file.peek(); text_file >> SourceName >> dist; if (!ContainsCrossRoad(streets, SourceName)) { tempCR2 = new CrossRoad(SourceName); } else { tempCR2 = FindInList(streets, SourceName); } tempCR->AddNeighbor(tempCR2, dist, false); } streets->AddItem(&tempCR); } } text_file.close(); for (int i = 0; i < streets->Count(); i++) { CrossRoad *t = *streets->GetItem(i); cout Name()

and even outside the infinite cycle, it doesn't seem to be reading it right
SourceName is "" for the first crossroad, even though it should be "Попа"

You already read the fucking line with the getline function. What the fuck are you doing with the >> and the other shit after that, work on the string you got from the stream with getline.
Also why the fuck aren't you using std::string?

Not any sort of expert, but while loops always fuck me. Maybe add another condition ( != EOF or something).

can i read variables from a string and if so, how?

As long as you don't mind mixing C with your C++ (or finding a C++ equiv), look into scanf/sscanf. I think those can do that.

Here's an example how i usually do with a stringstream.

void World::LoadTexCoords() { std::string path = "./Content/Sprites/AtlasCoords.txt"; std::ifstream atlas(path, std::ios::binary); if (atlas.fail()) { Log(path + " Failed To Load", Debug::Error, true); return; } std::string line; while(std::getline(atlas, line)) { std::istringstream iss(line); std::string type, t; int X, Y, XDim, YDim; if (!(iss >> type >> t >> X >> Y >> XDim >> YDim)) { break; } sf::IntRect rect(X, Y, XDim, YDim); _TexCoords.insert_or_assign(type, rect); }}

ok so what happens when i don't know how many variables there will be in a line?

I don't have any progress to show, but I got muh pathfinding working properly last week, and I've worked out ways to make things move around nicely.
On the agenda is more basic shit, like updating my pathfinding to detect collisions/obstacles, figuring out how to do multi-click shit properly (like left-click on a thing to select it, left-click on nothing to deselect it, left click on something else while it's selected to enter a sub-level of commands rather than deselecting or simply selecting the new thing) and so on.
I'm using an engine, but I still feel like I have to do a lot of basic engine-level work myself. Hopefully in a few months I'll have all of this behind-the-scenes make-shit-work stuff done, and I can actually start building game mechanics.

just fucking neck me now

For what possible purpose?

For lines that i don't know how long they will be, i use this
void split(const std::string &s, char delim, std::vector &elems) { std::stringstream ss; ss.str(s); std::string item; while (std::getline(ss, item, delim)) { elems.push_back(item); }}std::vector split(const std::string &s, char delim) { std::vector elems; split(s, delim, elems); return elems;}

You will need to convert the strings to int's or you adjust it. As example make a vector of some struct.
This method also has the advantage to split based on a delimiter.

This looks neat. Good luck.

I made some minor fixes to muh upgrade items yesterday :^(

I switched to stolen sprites, until I get a feel for what the player character should be able to do, and it made progress a lot faster. Now I'm having too much fun scripting moves to actually fix issues.

...

Is there room for two space pirate games made on old freetard engines?

Fucking do it, user. I'd love to have a pirate bro.

>for (int i = 0; i < size; i++)
Breh you need to delete the old items array before you overwrite it or you'll leak memory.

i know, but i haven't even gotten to optimization yet, i need to get it to work first

it's not about the optimization it's about not breaking your fucking program

well i mean it works right now, so it technically isn't broken
it's just that there's an array of pointers stuck somewhere in the memory

It's all cracked software from here on out.

You probably shouldn't do separate functions like that. It'll get optimized better as one function, and you don't really need the one that takes the vector by reference, because RVO will almost always save you there, and even if you're assigning to a previously-constructed object, modern C++ will move it instead of copy, making it almost free.

std::vector split(const std::string &s, const char delim){ std::istringstream ss(s); std::string item; std::vector items; while (std::getline(ss, item, delim)) items.push_back(item); return items;}

Though the original guy shouldn't be using a text file like that, but rather exporting and parsing through a library. Hand-serialization is dildos. Even using Lua would be preferable.

True, i actually just copied that snipped from stackoverflow. I used that as base for my own version (which should look similar to your solution) but i was to lazy to look it up. Has been a while since i used the function and i still haven't got around to make some utility library or something that contains all my helper functions.

I don't even understand why you bothered buying it knowing that there's plenty of alternative and that there's Enigma that's the exact same thing.

Mini-progress mini-report:
Pathfinding now respects obstacles properly. It took about 2 minutes of thought followed by 2 lines of code to make the adjustment.

I love when shit just works. Best feeling in the world.

Thanks to the user who suggested ShareX for screencapping. I was already using it but didn't know it had a video feature.
H265 support when?

GMS isnt the same software as GMS2
You work in gamedev you should know how sequals work :) :p

Is there an app I can use to make games on my smartphone?

I'm traveling a lot by bus so I can't bring my desktop with me but I have a pretty good phone so I was wondering if I could make games with it.

I haven't been posting. sry.

touchdevelop.com

Pretty simple, but it works.

Cool as always.

Demo when

That's Exactly what I wanted, thanks man.

Time to abandon my current project again.

My space pirate game is singleplayer, but the engine does support netplay–not very well, but it's there. I dummied out all of the multiplayer functions, but in case anyone tries to force it, here's an easter egg of what happens.

I'll need to replace those sprites down the line with the main characters, but hey that's all that's needed now, all the code/animation is implemented.

i am at peace

Got EQS in UE4 sort of working in c++ now. Don't know how to get the queries from c++ so I created a bp instead which has variables for all relevant queries. Unfortunately this introduced a bug I am trying to fix now that crashes the game on exit due to my AI brain class now being loaded dynamically instead of on construct.

Looking cool. I really enjoy your spritework, as it's entirely the opposite of the usual indie shit pixel art you see today.

Looks a LOT like links awakening.

I finished transferring the Unreal project from a blueprints based project Blueprints to a C++ based project. Now I just need to make some C++ classes and re-parent some Blueprints so that I can clean up and speed up the blueprint code.

Also I need to check out the Unreal Engine C++ documentation because this is a programming environment I'm not entirely familiar with. Luckily I already know C++ so it shouldn't take long.

Feel free to ask questions in the thread for retarded shit. I might be able to help out.

Implemented Pasta, Noodles and Sausages today, raw prepared foods are much easier to implement
Multi-step cooking soon

How do I make that 90's/00's look and feel for textures? They have a limited palette, but I don't really understand the color ranges they're using.

You want to make sure that you've got a pretty high contrast without getting into the black. Adjust your levels so that your colors aren't too saturated and you get a nice dark 90's tone. Use lots of earthtone grays, browns, greens, and reds.

old games could change their palette, the only thing was, no more than 256 colors (8-bit) could be shown. They also used things like nearest neighbor texture filtering. certain graphics cards also couldnt work with textures larger than certain sizes, so for example the biggest single texture you might be able to have on one card would be 256x256 (i think this was a voodoo card)

make sure you use nearest neighbor and have it so you can see the pixels when you get close enough to the texture

I love all of you.

t-thanks

Keep in mind they didn't have bump maps, so they had to do a lot of shading on the actual texture. This is why they look so dark in comparison to modern textures which use bump maps and look fullbright.

Obviously Zelda inspired. Looks like a GBC/GBA game. The overworld is pretty nice. Too much samey grass. Maybe add 1-2 more tall grass sprites. Also the guys legs look pretty bad. They're too skinny compared to rest of the body, too much gap there. Looks like his legs will crumble under his weight at any minute. The animation and light effect is wonderful.

is it possible to make a world engaging with just 2d squares that have just one color

No.

Yes

Maybe

...

I don't know, can you repeat the question?

Well the textures in the second image are 128x128 and the whole thing is 64 colors on a palette.

Is there a particular reason for that? Like what about outdoor textures like grass or water?

Is it easy to generate bump maps dynamically? I remember some normal mapping generator took a texture and assumed that black was 0 and white was 1, and interpolated all colors based on this; it also shifted the texture in 4 different directions.

is it possible to make a world bright, shiny and fun with just cute little squares colored like your big boy shirts?

Yes, so long as it's 3D and the shapes have lively animations as if they were next level programmer graphics.
But you REALLY have to go all in on animation and make your shapes expressive(or do experiments with their colors and textures).
Environments could be pretty fun depending on what you decide to do.

Theres a bunch of tools to do this. I was planning on using a gimp add-on, but I'm still not entirely sure if I want to do bump maps personally so I haven't tested it yet. I think if you use bump maps your feel of a 00s/90s texture will basically vanish. My plan (if I skip bump maps) is to generate a normal map, take it, do some wizardry in gimp where I filter out the light and dark shading of the normal map, and then overlay it on the texture and probably up the contrast so it looks dark. You could probably do the same.

...

I have to make a 2.5d shmup with 1-button movement (à la Jetpack Joyride) in Unity as an assignment for my school.
Almost done with it, works really well and is pretty fun so far. The only things I still have to do are aggressive enemies, a bossfight, and a good amount of visual polish on the environment.

This leaves me wondering: How can I make it as close as I can to a bullet hell during the bossfights without making the 1-axis restricted movement a pain in the ass?

Finally got around to punching out the Swim-back and Swim-side animations

This is almost done. This should have been ready months ago. I will post this on /agdg/ and /furry/. I'm so ready to move on with my life.

do you actually know how to swim?

Absolutely not. I almost drowned as a child and have never made any real progress since.

The animations were based off of random videos from youtube.

Haha, same!
Except I drowned twice, not once.

Don't love me user. I'm a hack fraud who'll never get a project past the prototyping stage.

Is this a /furry/ game?

Yes

I expect a flop since it's very limited.

trying to get a simple shopping app to work

me too except I can float now

Alright, I'll keep looking into it and probably go in more of a 6DOF direction even if it means probably not using an id Tech engine and working with a genre I barely know.

What OS do you use?
What language do you write your games in? (C or C++ ?)
Do you use Blender or something else?
Photoshop or Gimp?
I included some music to listen to for inspiration.

I use windows and write my games in C

I use paint.net for image editing when i need to edit images

image vaugely related

Win 10.
x86 assembly.
ms paint.

Do you use win32 or something else like glfw?

I use win32 to create the openGL context, and then draw the graphics with openGL

I was lying.

I really use win7 with UE4. Code in c++ mostly, blender for 3d work which I very seldom do and typically paint.net for images but I have photoshop too.

Is that a tree dating sim?

Dude trees lmao

Question for UE4 devs, has anyone ever been able to build a game that runs on cheap (and even not-so-cheap) laptops?

I've been using close to tablet settings in my games just exporting to desktop and they still run like shit.

I use Linux, write in Python GDScript for Godot but effectively the same, and use Blender and GIMP.

Recursion is bad, kids. Each planet has a field that references what colonies are on it, each colony has a reference to what planet it's on.

Awww

Re-posting progress

I hadn't seen these. I'm curious to watch your progress, a buddy of mine and I messed around with a similar mocap process several years ago and didn't get very far. I'm hoping you'll have better results.
To be fair, our issues had more to do with us than the software, I think.

It interested me because it's not the most expensive thing ever, also because I am not at all good at classic animation. I can clean this up, learn IK and script mecanim behaviours, but making animatons from scratch would be impossible for me. and paying for animations? Not at this point

I think we did nice progress in the past week. You might remember the freaky motoko, morphing and lurching in agony, while it was supposed to be a silly dance…

the iPi mocap guys developed their software quite extensively from what I can tell, they even added motion controller support for attaching to annoying bones hands are especially shaky. Pic related is what I'll do when the PS move wand arrives.

The results are already quite satisfactory, although there is still much to improve and learn

especially in the lighting department
and simple backgrounds
and clothing colour- we mostly wear black, ffs

Hey /agdg/, I've been thinking seriously about a game I've been wanting to do for a while, and I'd like some advice from devfags who might know what they're doing before I realize I wasted 2 years and fucked up by choosing a shitty engine for the task or something similar.

Basically, I want to make a turn-based RPG with gameplay like Valkyrie Profile's (player binds skills to keys, attacks in turns are quick and involve the player to avoid them from getting bored, timing your attacks is important) with a bit of PXZone (integrating combos and following up skills, if you time stuff gud you can attack more) and and SMT's press turn (play smart, abuse weaknesses and you'll get more attacks, mindlessly spamming skills will be heavily punished). I was thinking of a low-poly 3D art style (akin to most PS1 3D games' style, PSP's if I git gud in my lifetime) to give it a rather comfy feeling. I'm not very sure about soundtrack and other such specifics yet but I'll think about it better when I start doing serious progress.

What I'd like to know is if there's an engine that'd be best suited for what I had in mind, as well as what programs I should use for modeling or even advice to make my shitty idea not so shitty. I've got a bit of experience programming on C, C# and Java(lel), and I'm willing to learn more and take my time in order to make gud shit™. I'm well aware that I should start with simple games because doing your dream vidyo first is always a trainwreck, but I want to start working on concrete milestones towards that goal instead of just fantasizing. Thanks in advance.

Pretty much any engine that's 3D can handle that. There's no such thing as "engine limitations" when it comes to gameplay despite what bullshit bethesda likes to pull. Especially not if you have source access to the engine.

Use blender to make your model and anims unless you've got maya or 3dsmax. Unity, UE4, or any variety of smaller free engines could handle it. Godot for example. It's up to you to find an engine that you enjoy using. Make gamejams or small projects first to get used to the engine and dev in general as skills you'll learn will carry on to later projects always.

Make simple clone games for every single of your major inspirations. Will bring you closer to your dream because you learn how to implement & understand the key points while still being achievable targets. Ask someone else about 3D engines, I don't know shit about them.

Yeah neither of us could model or animate for shit but he had a friend wiling to spend some time modelling for us, so we had some similar reasons for exploring it. We were actually looking to go the mid 90's route of using 3D models to generate sprite sheets, so we weren't even that worried about perfect animations coming out of the process because we could pick and choose frames. We never really got it up and running, we made some progress, but we stopped working together before we got to the point of having a really smooth process going.
I wouldn't be surprised if the tech has come a long way since then, we were trying this in like 2009 or 10. I'll be rooting for you.
I've been nodev for a long while, but I'm thinking about getting back into it and making a twinstick, I'd forgotten all about my time with mocap and ipi until your posts. I might try the same process again.

I use both windows 7 and linux
I program in C++
I use gimp, paint.net and blender

can someone help me optimize this function.
it fills a triangle, defined by the the points xa through yc, with the color r g b.
voiddraw_triangle(float xa, float ya, float xb, float yb, float xc, float yc, byte r, byte g, byte b){ // to make fmaxf work you need to compile with the flag -lm float mostLeft = fminf(fminf(xa, xb), xc); float mostRight = fmaxf(fmaxf(xa, xb), xc); float highest = fmaxf(fmaxf(ya, yb), yc); float lowest = fminf(fminf(ya, yb), yc); float fAlpha = lineFunction(xb, yb, xc, yc, xa, ya); float fBeta = lineFunction(xc, yc, xa, ya, xb, yb); float fGamma = lineFunction(xa, ya, xb, yb, xc, yc); float fAf12 = fAlpha * lineFunction(xb, yb, xc, yc, - 1, - 1); float fBf20 = fBeta * lineFunction(xc, yc, xa, ya, - 1, - 1); float fGf01 = fGamma * lineFunction(xa, ya, xb, yb, - 1, - 1); for (int x = mostLeft; x < mostRight; ++x) { for (int y = lowest; y < highest; ++y) { float alpha = lineFunction(xb, yb, xc, yc, x, y) / fAlpha; float beta = lineFunction(xc, yc, xa, ya, x, y) / fBeta; float gamma = lineFunction(xa, ya, xb, yb, x, y) / fGamma; if (alpha >= 0 && beta >= 0 && gamma >= 0) { if ((alpha > 0 || fAf12 > 0) && (beta > 0 || fBf20 > 0) && (gamma > 0 || fGf01 > 0)) { PutPixel(x, y, r, g, b); } } } }}

Win7
Javascript
Krita or my own drawing program depending on what I want to do

Which game engine is easier to program, blender or unity? I know that blender game runs like shit, but when I make low poly games it won't run like shit right? Pic related, the poly count I'm aiming for.

Literally who?

Not even Blender's creators think Blender is a valid game engine option.

As long as you don't blatantly rip the other guy off, it's fine. Besides:]

Modern GPUs are REALLY good at rendering a lot of polygons. Polycount is probably the least of your concerns if you want performance.

As said, I don't think anyone in their right minds would recommend Blender as a game engine. If I recall correctly, I think there was even talk about abandoning the whole Blender game engine idea and making it focus more on motion capture or something like that.

What about game maker studio, with 3D environments and 2D models. I'm most experienced with gms, but never made a 3D game. Does it run well?? And any good tutorial from scratch?

Last time I cared about GM was 7 or 8, and back then trying to use it for 3D was a sure fire way to make you depressed to the point you want to commit suicide. So unless that magically changed and they support stuff like bone animations now, that's a no.

so i've still got issues with my text file reading from
i got it to correctly read when the line ends by using a stringstream
my issue now is that somehow the pointers overwrite whatever was in them the last time, even though there shouldn't be anything there to begin with since the pointer gets declared within a cycle
example:
Crossroad1 Crossroad2 100 Crossroad3 200 Crossroad4 200
Crossroad3 Crossroad1 100 Crossroad2 200
crossroad1's neighbers are crossroad2,crossroad3,crossroad4
crossroad3's neighbors are crossroad1 and crossroad2
what should happen is that first we create crossroad1's object, then we create objects for the other crossroads in the same line. when we reach the end of the line, we repeat the same process for the next line
what actually happens is this
crossroad1 gets created
crossroad2 gets created
crossroad3 is supposed to be created, but for some reason it overwrites whatever was in crossroad2 to be the same as the new object. presumably because they both use the same pointer in the code (tempCR2) even though this shouldn't be happening
so now when you look at all the objects we've got we have this: crossroad1, crossroad3, crossroad3
same thing happens when we try to create crossroad4
we get: crossroad1, crossroad4,crossroad4,crossroad4
then when we move to the next line, the same thing will happen, except that the first element will also be changed
crossroad3,crossroad2,crossroad2,crossroad2,crossroad3,crossroad2,crossroad2
someone got a clue how to fix this? i'm sure my list class works fine, outside of the text reading function it works perfectly.
i checked multiple lines of code, and the issue is almost definitely happening at tempCR2 = new CrossRoad(SourceName); and the same one for tempCR
after it happens, everything after tempCR has been overwriten with whatever new Crossroad i'm making
pastebin.com/AX52BY2G

Not gonna lie, figuring out all the pointer stuff going on in your two code snippets is too annoying for me without proper debugging, so i can just offer some simple advice:

1: Is there a reason you use your own class instead of std::vector or some other stl container? The stl stuff works pretty good and it would mean one less potential error source.

2: Use breakpoints and step through your code, check what each line exactly does and step into the functions that you call es well. If you use visual studio you can trigger breakpoints with F9 and step with F10 or go into a function with F11. I think that's true for pretty much every IDE, if you use gcc or clang you will need to find something else to do that.

2) like i said, i'm 99% sure that the issue is in tempCR2 = new CrossRoad(SourceName); , so i don't think i need breakpoints. the pointer gets initialized in every loop cycle, and because of that, everytime it should be pointing to the address of the newly created object
i have no idea why at that exact point in the code, it overwrites the last object
the even more confusing thing is that it doesn't just overwrite the last object, but all the objects that were made with the same pointer
1) no reason in particular, i just feel more comfortable using my own thing. but i'm sure the list class i made works correctly, i've tried it outside of this function with more than 3 objects and it worked without any issue there

But not 100%, thus breakpoints.
Did you test it against convoluted loops filled with pointers, too?

Your shit don't work, user. Roll up your sleeves and step through it one line at a time if you want to fix it.

you saw what Yandev did?
Lowering the view distance and simplifying the character models? Didn't help much with the inefficient code.

well the issue is definitely the pointer inside the loop, i still don't know why though

you have VS so you can be able to debug and track line by line what is exactly occuring

tracking shit line by lone doesn't really help when the actual issue is only in one line
the matter of fact is that the pointer retains it's address through more than one cycle. everytime i'm adding the same object to the list, except that it's name changes everytime. when that happens, all the objects in the list change their name, since it's the same object
the way to fix this would be to have a different object added to the list everytime
which is what SHOULD be happening
WHY is it not happening?

I played a bit around with your List class and i'm pretty sure that the class doesn't exactly does what you think it does.
To be fair, i don't really like working with pointers too much, i only use them when i need to, so i can't explain whats happening. Anyway, i could fix the problem of your List overwriting the values by removing some of the pointer magic you did.

template class List{protected: T *items; int size;public: List() :items(nullptr) {} void AddItem(T _item) { T *temp = new T[size + 1]; for (int i = 0; i < size; i++) { temp[i] = items[i]; } temp[size] = _item; size++; items = temp; } int Count() { return size; }};List *test = new List(); for(int i = 0; i < 2; ++i) { test->AddItem(i); }

I needed to remove some functions, because ReSharper can't handle templates all that well, and i can't be arsed to fix all the other shit.

As i said, i'm not the most knowledgeable with pointers, but i think you should refactor your List class. Or some smarter user can explain whats actually happening.

except that what you did makes an pointer to an array of Ts
in which case, when i add an item to the list, it's a duplicate item, not the original
and when i want to check if a certain object is in the list, i need to compare the address, so i can't have duplicates in here
aside from changing that, i can't see what else you've changed

I know how to switch case and loop properly though. But to object oriented programming, I totally suck at that.

go to uni :^)

your code is messy, hard to read; anyway in the third picture, your pointer code to copy to the temp array checks out.. for add_item - now the only thing im not sure about is :

"T **temp = new T*[size+1];

because i am not a c++ programmer.. the right way to allocate it in C is:

"T **temp = malloc(size+1 * sizeof(T*));

so just make sure that those are equivalent, i dont know if they are.
if you dont know why all the things in items are the same, then the problem isnt here, items is being constructed wrong before you pass it in.

I made a typo, i meant :

T **temp = malloc((size+1) * sizeof(T*));

You can of course still create a list of pointers
List *test = new List(); for(int i = 0; i < 2; ++i) { int* j = new int(i); test->AddItem(j); }

Pic related, the objects in the list have different values + adresses.

Of course i don't know if it's usable for you, but all i can do is fish around for a solution as i don't have the full code.

fuck me, that reminded me
i'm already using an array of pointers inside the list
but when i make the list in the cpp file, i make it a list of pointers
in other words, i have a pointer to an array of pointers of pointers
i have no clue how it came to this
but yeah, making it a regular list of CrossRoad fixed things
don't i just feel dumb now

I imagine this is why programmers constantly say "don't try to reinvent the wheel".

writing broken code and feeling dumb is a necessary part of learning how pointers work

good job

anyways, shit actually works now

Good one, user! I bet that feels pretty damn good.

tbh i didn't even need to code anything yet
right now, i've just got to write a header file that describes what the classes and methods do, and to write documentation with charts and shit which describes what data structures i'm using, not to actually code them
so in other words, i'm around a month ahead of schedule.
i'll probably end up spending it trying to optimize shit like the blatantly obvious memory leaks, or the fact that i make a new array everytime i add a new object to a list

use a linked list, this way you dont need to allocate temp arrays or anything of the sort to push and pop items

i'll just make it double the array size if the array gets completely populated

I think we can all relate.

Making a shitty, fly-by-night pron game.

Done with most of the actual code, now I'm on to making assets. Just finished modelling a USB key. Thus concludes my pitiful amount of progress; back to playing my DS.

I wish I couldn't, but god damn can I relate.

Somewhat veteran in GMS here, wouldn't recommend. 3D is more a fancy feature than a full-fledged tool, the whole UI is clearly designed to make 2D games, 3D functions doesn't even appear in the drag and drop options.This coupled with the fact that GMS2 will be the only one receiving support soon enough further indicates this incompatibility, as there is no sign of 3D support in GMS2's new features.

Ok, guise. Need money, how do I meme game?

Does GMS2 not even have 3D ds_grid or array support yet?

Wait a sec, let me check.

get a job

I'm not good at anything to get a job.

If you're not good at anything to get a job, you're probably not good at anything to make game, user.
Learn at least one programming language, see if you can handle it, and then try to make game if you can. It'll take you a long time, though, and you'll have many false starts as you try to work out how to do shit.

Good luck, it's a long hard road filled with effort, no fun, and absolutely no guarantee of making a cent at the end of it all!

Game dev is the last thing you should be looking at if you need money.

...

Forgot to say: Guess not.

I wouldn't consider scripting with GML as "programming" but you can make a functional vydia with that kind of knowledge.

That's some realistic advice. Still want to meme game tho.

Example of what you can do only scripting.

It's the same as programming, it just does some of the more math-intensive stuff for you. Unless you're talking about using the drag-and-drop stuff instead of gml, in which case I fully agree.

Not even once.

Amen to that.

...

You should check out sigma engine, the dev posts here frequently:
ika.neocities.org/sigma_engine.html

god damn. i'd expect it to be the other way around.
inb4 >using rpg maker
i'm just practicing sprite work

Moving on to next thing. Don't know what I want to do yet, so started on a ray-casting FPS. Could be fun to do a persistent online one or with peer to peer multiplayer.

You get what you deserve if you use anything later than 7.

I needed muh Linux Subsystem and my Windows 7 install was fucked

Do it anyway.

Do it anyway.

Learn it, and do it anyway.

be sure to use realloc or whatever the (C++ equivalent is) instead of new- it copies the old block of memory over for you

thats me!

If it's made well, I will always play a new roguelike. Hell, I'll always play them even if they're shallow clones.
Angband, Zangband and Oangband were all pretty good.

There is no C++ equivalent to realloc. If you really needed it, you'd use realloc and placement new, but practically here, you'd just use a std::vector or std::list.

Strife succeeded in that ambition. It was basically Deus Ex: 1996 edition. Have you played it?

Have you looked into EasyRPG?

Neat. P2P multiplayer is somewhat difficult to get working right. Existing examples like vanilla Doom, ZDoom and Duke3D are very vulnerable to any kind of lag and packet loss. I'd just go with client-server.

Yeah, probably just go with client-server. At least I'm more familiar with that.

Do I really need to draw 47 versions of every tile if I want them to connect properly? Am I missing something?

So far I've been using a method that draws the corners separately, so I can draw only 1 tile plus the edges, totaling to 17 sprites. Downside is that it requires me to draw up to 5 sprites per tile.

You can reduce the number drastically if you use symmetry (axial and line)

I only have a very rough idea where this is going.

Getting in on those sweet, sweet Haydeebucks? (If there even are any)

A lot of that can be reused if you rotate the sprites.

Amazing.

What?

Well, you can't rotate anything that has even a hint of perspective or shading in them, i.e. pretty much all the tiles I have.

When's it going to fall over?

You know, the video game.

That looks about right, I remember drawing the same kind of tileset for top-down games.
Beware in case you need to add more types of ground to your landscapes, in which case you should really switch to overlays with transparency (e.g. basic grass tile, sand tile on top, road tile on top) instead of everything baked in.

Soon, I'll probably need to think of a counterweight on her back or head.

Maybe you just need to dial back the tilt a tad so I can appreciate the jiggling instead of constantly being reminded of Michael Jackson.

It's like you never heard of pixelshit before.

Yeah if you have those corner "borders" inside it adds a lot of additional tiles. If you can get away with pic related, you can keep it at 16 I believe.

Dat jiggle.

install gentoo

I don't know why people think things like this

Colleague of mine that does crazy shit with the engine.

Message there is not final (obviously), I just tossed it in there to see if he'd find it.

6DOF style games would be really fun.
For research, I'd highly suggest the Descent series.

Do you have anything in mind for the overall gameplay or what the goal will be?

Decided on Godot. Should I do some 2D stuff first to get used to it before jumping into 3D development? Or should I just gun for 3D development if that's what I really want?
(Pic mildly related)

Reminder that gaming journos complained about the robot not having realistic breast physics. They actually fucking complained about the robot with artificial tits not having accurate tit jiggle.

Switch to 5x5 tiles and do it in the center, and it reduces it to 16 sprites. A lot of those are really redundant unless you really need the walls on the edge of the tile.

Rate pixelshit metroidvania protagonist plz

That depends on what you're trying to do, I guess. Have a somewhat relevant infographic.

Fucking EasyRPG nigga. Use it.

Always start simple and ramp up. At first you should focus on getting familiar with all the tools and the UI.

LooksLikeACuck/10
I like the look. Looks like a GBA sprite which is always welcome in my book

Will do, thanks for the advice

Un bon gros roux. Il me rappelle le camionneur de Trucker's Delight.

Note that that infographic is pretty similar to marching squares.
en.wikipedia.org/wiki/Marching_squares

I don't have the slightest idea what you're saying, I just named the pic "mec" because he's a mechanic.

Oops. It also means guy in French.
I was saying he reminded me of the dude from Trucker's Delight a bit. Mostly the shape and the mustache.

ANYONE ONE ELSE MAKING A GAME FOR PROCJAM?!

itch.io/jam/procjam

Sign-up close in like 5 hours. Any one else going to make something? It would be fun if a few anons submitted things. Let's make something cool over this next week.

i hope someone makes a fun evolution game

Are you going to make anything? The submission guidelines are pretty open ended.

that actually sounds really fun

I might make an evolution simulator in UE4 for this

Would anyone else want to join in?

im a BP scrub who doesn't into C++ but I'm decent with BP

Maybe I should recreate a thing I made back in highschool, an interpreter for making simple text adventure games in Brainfuck. It's completely inefficient and stupid, but funny and might help get people interested in esoteric languages.

no time, i'm busy with university projects, homework and algebra homework/lessons
and whatever free time i have left i use to play card games
gamedev is put pretty low on my priority right now

HA.
I'd still like people to try, but HA.

I'd like to make a proof of concept level generator, but I've found that I'm not skilled/fast enough to do game jams.

In Unreal 4 I need to load all of a certain object into an array to pick things out and manipulate

Is it better to search for all of a class even if some aren't what I want to manipulate or is it better to search for all of a tag and tag the ones I want to manipulate?

IE. Is it more expensive to:
1. Dig through an array full of a single class and identify incorrect objects
2. Search every object for a tag

can't you just add them to an array whenever you initialize them?

Pretty sure number 1 would be more expensive

But honestly it sounds like this might be a job for interfaces if you haven't considered them.

I meant number one would be less expensive

KNIHGT

Sounds fun, I'm game. I just gotta read the rules to see if they accept nsfw entries.

There's still around 10 days left. Although if what's ailing you is working under a deadline I guess that doesn't help much.

I haven't because I am unaware of interfaces

I'm still trying to learn at the moment.

If they don't take nsfw entries, you should do it anyway, then release a censored version to them, and then give us the original. :^)

fix that arm topology fam

I lost the model thanks to Blender's buggery anyway. Gotta start over.

How do I learn 3d modeling humans? I have no idea where to start.

Youtube, or pirated paid tutorials if you have a cgpeers account.

how are you going to get faster if you are not even trying?

next time it happens just use the recover last session button
to prevent it from happening again just save your stuff as soon as you begin (if you don't know what to name the just use the date and give them names only after they earn it)

if you can draw it you can sculpt it

I never draw anything my whole life.

You remind me a lot of Peruvians. As soon as they think they see another of their own they burst out their retarded language no one wants to hear.

>>>/loomis/

How did Robot Area 2 allow people add parts to robots?

gunstruction also exist.

Tried that, but it only left me with the bones.
It was really weird.

any other games like this? i downloaded it but the first time it ran with missing sprites, and now when i try to turn it on again it just freezes my toaster

I would aim for for a combination of 6DOF and Quake-style gameplay/level design with weird gravity that varies depending on the ships you loot, and possibly some simplified space simulator mechanics to sell that comfy space pirate lifestyle. It would also make a very nice mechanical testbed for a more ambitious project that I've been reading a lot of engine documentation and research papers to prepare for.
I'll probably pirate the Descent games soon and buy the AquaNox games when they go on sale again, bundled together for under a dollar.

How do you make maps in Godot 3D like in Hammer/GTKradiant? I don't want to do gridmapping

What OS you're using? Try XP

By making maps do you mean environments? Because you do that shit in blender and use import scripts for fancy stuff.

just like make importer

I still love you. Believe in yourself. Set your mind to smaller projects. Don't think about the bigger picture. Just pick 1 aspect of a job and get it done. Then move on to something else. You can do it.

wait, so you niggers saying I just make one big .obj instead of a .map?

make rooms instead
docs.godotengine.org/en/latest/tutorials/3d/importing_3d_scenes.html

Ok so how do I make rooms in godot?

yea nevermind, i got it from another torrent and it worked fine

you do it in blender, I meant to quote as well

As .obj or .dae?

.dae for whole rooms
.obj for very unique, very repeatable or detail objects (pick ups, breakables,etc)

you are going to export the whole scene then add details within godot

more

I don't know how to feel

forgot to ask
is there a way for obs to start recording as soon as it starts?

what did you ask?

It was like 2 years ago now. If I remember, I think their example code had a small typo, so I asked for some clarity in the comments section. I just remember going back the next day and noticed it wasn't there anymore, but noticed one of the newer comments was a more clarified version of my question.

For those of you that use 3D capable frameworks like MonoGame, how do you get around not being able to visually edit your levels?

I think most people still do, they just use external editors. From my experience that makes things more difficult to develop for, but decouples the code and the 'levels' which I can appreciate.

I like this. What is it about?

I'm telling you to import the compiled .bsp, or compile the .map into your own format.
Don't listen to the fags saying to use blender or import as .obj, because you'll be missing out on
and your character will fall through the floor and next thing you know you'll be working for Ubisoft

Sorry, user, I'm just wage slaving for mobile games. Glad you like it, though.

Gotta pay those bills somehow. Stuff looks neat though.

When I started using Godot I played around with 2D first. It's better documented, that's for sure.

How do you manage multiple animations going on at once in Unity? Seems really complicated.

For those of you who code your engines do you see OpenGL as still relevant? Or should Vulkan be the focus going forward? I mean this for the solo hobbyist type.

REEEEEEEEEEEEEE
RRRRRRRRRRREEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEE

Implementing ads into this is going to make me sad

reverted to 2.77, which is officially the last good version of blender until they stop fixing what isn't broken

so I was testing some workflow stuff and somehow made a gondola in the process

I don't like rushing shit just for the sake of rushing it.

I'll join jams when I'm comfortable enough with basic game mechanics that I don't get stuck tinkering with them out of curiosity or uncertainty.

Now make a game for it

As far as I understand it, Vulkan is much more low level, so it's sort of like asking if C++ is obsolete because C exists.

Correction, it's like asking if Game Maker is obsolete because things like Unity and UE4 exist. The answer to which is no, but they're a hell of a lot better.

The way I look at it, on desktops OpenGL will remain viable for two reasons:
1. It's an easier language to learn
2. Everyone already has OpenGL support in their drivers

In the long term, you're going to see more and more people move to Vulkan on desktops for the performance boosts though.

On mobile, performance is much more important. I can't imagine OpenGL ES having much of a future when there's Vulkan, unless all the mobile programmers are just that retarded (which they may very well be).

If you're trying to decide to learn OpenGL or Vulkan, I personally might would go hardcore and just learn Vulkan. In reality I learned OpenGL and I'm now studying Vulkan when I can, although time isn't really allowing for it so my 3D game may release with just OpenGL support.

There's simply not a whole lot of a market for mobile graphically impressive games. Simple games rule the market, and those work fine with OpenGL or even premade engines.

Got textures walls. Rather than going pixel by pixel I'm doing 1px wide quads of textures and then scaling them vertically, so it ends up just being one draw call with all the quads batched together.

...

Good, keep it up. It makes absolutely no sense that energy weapons have recoil.

Depends on the weapon probably. Pure energy Laser/Plasma/Beam weapons shouldn't have them, but if you shoot any kind of energetic projectile, there still should be some sort of recoil, first because of the force that propels the projectile forward, and secondly the air that gets compressed/pushed aside by the projectile. even if it is just light recoil.

ye for things like Gauss/Rail guns I would understand, I have a hazy memory of some gun that is energy based something like laser/plasma and it did made a nice amount of recoil which triggers me tbh fam.

So I made a stupid 2d turrent game in C++/SFML to learn but now I'm compeltely stuck. Every idea I come up with seems to be either a boring copycat game or something unfeseable to do for a single developer.

Well, people can make fun of ideaguys, but actually getting an idea that is unique enough and also feasible for a 1MA, isn't that easy.

You shouldn't worry much about being a copycat though, even the first "videogames" like pong and electronic tic-tac-toe were based on real life games, and from then on games always were build upon previous game ideas with a few unique additions. The tricky part is to make your game similar enough to previous games for players to feel familiar with it, but also set oneself apart enough with a bit of new shit, to be seen as original and not a chinese like copy that can be replaced by any other game of the genre.

Be a copycat. The best way to learn is to reinvent the wheel.

Finished basic stuff for my naked dude. The missing back arm will be holding weapons and stuff.

Run cycle looking good or too cheated?

1. why does he look like a pill
2. whats the orange bit on his belly

1. Because he'd be tiny-ass in the game (32x32px) and has to read well. Also the idea is to have the guy cast into an aggressive world, so he should look rotund and inefficient.
2. Chest hair.

Any musicfags here? What program did you use and how did you learn to use it?

I've used Cubase in the past and it works pretty well, in particular has good MIDI support.

Even now that I'm on Linux, I might still switch back to Windows/Cubase for music work. But nowadays there's at least what looks to be a decent paid solution for Linux, Bitwig Studio. Definitely looking to experiment with it at some point.

I like. Dig the flab on his arm.

Really I'd like a good FOSS solution, but LMMS and Ardour aren't all that great imo…

I hate videogames

Slide post? Should I/somebody report?

You'll fit in well with us then.

Me too, but only the ones you make

I don't even know how to fucking script bcuz i waste my time more on sprites and slow guitar jams.

How do you make a radian lerp not do this when it goes from positive to negative or vice versa

use a regular lerp?

HEY KID, WANNA MAKE A GAME?

That is a regular lerp

I want to make a meme

Wait, so the radian goes negative when it goes past a half-revolution?
wouldn't it just be:
if (rad < 0) rad_correct = rad + (2 * math.pi)

Causes the same thing to happen but at the opposite end