/v/ - Video Games

>>8768673

*some

*space suit

I hate typing on mobile devices.

>want to make a SRPG but dread the fact that I'll have to create a competent AI

>end up doing nothing all day but masturbating

The strifes of being a NEET.

>>8768720

Work out at a gym early in the evening, sit down at your desk with some coffee, and get to work in your game. I've found this to be rather effective. Plus, you'll feel more confident and better about yourself in general.

>>8768764

I already go to the gym.

Tomorrow I'll be reaching lmao2pl8 on my squat, also coffee has no effect on me since I've drank 2-3 cups daily for years(my teeth are yellow as fuck too).

>>8768694

Hm. It depends on the way the character interacts with the world. Is it first person? Perhaps only the walls and objects the character/player physically looks at are recorded on a map.

Third-person could simply be based on getting close to it within a certain distance. Same with Top down.

For C# shit, would it be better to use the Visual Studio Ultimate 2010 that I already have, or to download whatever current version of Express that's available?

>>8768804

Good for you, bro. My small frame and bone structure can't handle that much weight above my neck, but I still stay in rather good shape. Can deadlifts over 200 pounds, so bretty gud.

Yellow teeth sucks. I find Rembrandt toothpaste helps remove plaque buildup and makes my teeth look somewhat better. Try it.

>>8768895

You'll probably be okay if you stick with that version, but I'm sure it's always in your best interest to get the newest version for any bug fixes. Upgrade when you can.

>>8769176

>Good for you, bro. My small frame and bone structure can't handle that much weight above my neck

I'm in the same boat, I'm 5'8-9

Ah! My good friend and colleague, Mr. /agdg/ himself! Long time no see! What have you been working on good sir? Perhaps you have a completed video game to show me?

>>8769403

Should probably elaborate. Doing something like

 Start()
{
direction = transform.TransformDirection(Vector3.forward);
}

FixedUpdate()
{
translate(transform, direction * 10 * Time.FixedDeltaTime);
}

thinking I might try saving the coords as constants in another class or something.

>>8769522

>>8769344

What engine are you using for this?

>>8769577

love, 2d lua. The screen on the right is merely the level maker, let me put this in game for you to show off how it looks.

>>8769403

>I'm using a billboard as a projectile

>The problem is it rotates to match the camera each frame

Yes, that's what a billboard is. Don't use a billboard if you don't want it to do that. Use a plane.

>>8768720

>want to make a SRPG but dread the fact that I'll have to create a competent AI

I thought it was one of the easier genres to make AI for? Regardless masturbation wont help, I tried that approach for years.

>>8769656

>I thought it was one of the easier genres to make AI for?

I've never made an AI at all, so I'm a bit scared.

I want to be a game dev but I'm a fuck up and I dropped out of college. I could have learned C++ and a little bit of C#. Mind you the program was entirely based around financial programming for banks and businesses, so I doubt I could have transfered anything over.

What are some good starting points for learning C# and using either Unity or UE4?

>>8769586

>>8769617

Thanks man, how's Lua anyways?

I feel like I should learn it because muh patriotism, but I'm unsure if it's a good language at all.

>>8769687

I'm not the programmer, I dont know that much about lua itself. I only get to work on the small things like buildings and units, in standardized structures. I do design in general.

>>8769677

thats what google is for

and you can always ask for help here youd be surprised how much some people know here

It happened, /v/…

I've started a second game before finishing the first one

I think i'll get back to it when i learn how to make fog of war and how to implement online multiplayer

>>8769686

looks at the Epic tutorial series for learning UE4. There are tons of resources for learning languages online as well.

Dont fret about dropping out, I did it too and its not like anyone is judging you.

>>8769624

What do you mean? How would using a plane be different?

>>8769718

>fog of war and how to implement online multiplayer

where you making an rts?

im making one to and i haven't implemented either of these

om not to worried about either though i have pretty good idea for fog of war

and online multiplayer i've done before

right now im trying to get pathfinding to work but it's tricky

>>8769734

Because they don't rotate to face the camera at all times, like billboards are supposed to do.

Feels bad that I never have progress to show here. I'm handling programming and writing but I can't do art for shit, and I know drawfags don't like to sign on to vaporware, so I'm trying to get an engine built and some starting levels for my game. Then I'll start looking around for more teammates.

In the meantime though, I barely even have something that can be recognized as a GUI, rather than abstract art that moves.

>>8769687

Lua is a great scripting language. It has a few quirks but overall it's good and Lua-JIT is fast for what it is.

>>8769290

That's not too bad. You can still pass off as average, especially with the right shoes. I'm somewhere between 5'9" and 5'10," but with my shoes I stand perfectly average or just above.

What kind of SRPG were you wanting to make? Have you actually started anything?

>>8769738

Yes i am, i'm using gamemaker, which has a dedicated pathfinding function, it's an algorithm called A*, you should find out if whatever method/engine you're using supports it.

>>8769763

>What kind of SRPG were you wanting to make?

A Disgaea clone pretty much, with a few added features that I'd like to see.

>Have you actually started anything?

All concept so far, so no, I fear starting it then hitting a brick wall with the AI since I've never done that, and scrapping the whole project.

>>8768812

>>8768941

>>8769313

Yeah, the main idea I had in regards to maps were that each explorable area in the game had a map you could find. The map could only be accessed from your inventory and was unfinished at first. At each checkpoint, the map is updated to include more detail on areas you've been to (like >>8768812 said with the third person bit)

And don't worry, I fucking hate quest markers.

How difficult would it be for one to build an engine for a Harvest Moon/Rune Factory clone from scratch in C#?

I know I could use a game engine, but it feels lazy, and I feel like I can put my education that I spent so much money on to good use.

>>8769774

im not using an engine

anyway i already got A* working it's not that complicated

the tricky part is get it to work fast and accurately

which is mostly down to the graph you use

at first i tried a grid based but thats not accurate enough for my taste (you get 8 movement directions at most)

so now im trying to create some sort of navmesh dynamically

i just put nodes on the corner of every rectangle

that is fine i think but at first i connected every node to every other node that didn't intersect with any terrain

but the amount of links would scale like O(n^n) or something stupid like that

so i changed to to 1 link every 90 degrees im just worried about the accuracy

but sacrifices need to be made i need to be able to generate the mesh with 200+ object and call A* on the generated mesh 100+ times each loop and still get a respectable framerate

>>8769819

Acquire a copy of Stardew Valley and then decompile it. It was made in XNA, therefore C#.

>>8769774

Quick reminder, how does A* know a chain of nodes its following wont eg lead into a "pocket" and waste cycles?

>>8769911

>Acquire a copy of Stardew Valley and then decompile it. It was made in XNA, therefore C#.

I'm pretty sure this would be asking for a lawsuit if that anon ever did finish his own game.

>>8769926

To learn from it and see what sorts of techniques they used? Hardly.

Besides, its how mods are made for closed source games, anyways. Terraria is also done with XNA and so many people have decompiled that slutty little .exe for learning purposes on the official forums alone.

>>8769896

i don't know, if you use a 2x2 grid it emulates the movement quite well, but that's for you to decide

>>8769911

it spawned 4 objects, up, down, left and right, and thenthose objects spawned another 4 around them, and so on, until a path is found (or the impossibility of that path)

>>8769977

Yeah, I know the node-based nature of it, but this was a specific optimization of it.

No worries, Im at work, Ill read up more when I get home

>>8769746

Are you saying that planes have a way to show the image to the camera at all times that doesn't require them to rotate, or are you suggesting that I give up on having projectiles face the camera?

>>8769961

And I thought MY flowcharts/info logs were neat and tidy. Nice.

>>8769911

>Quick reminder, how does A* know a chain of nodes its following wont eg lead into a "pocket" and waste cycles?

it doesn't actually it just minimizes wasted cycles by using a priority queue with a heuristic and a weighted graph

>>8769977

the problem with grid is that you get a lot of paths that have the same cost according to the algorithm

but in reality only 1 of those would be the shortest route

and there's really no way around this at least that i have found

for detailed explanation read http://theory.stanford.edu/~amitp/GameProgramming/Heuristics.html#breaking-ties

>>8770029

ever herd of KISS

Keep It Simple, Stupid

Go with minimal, clear stuff. I'm curious, what do you flowcharts look like then?

>>8769911

Anyone here ever use XNA? How different would it be using that over coding everything by hand in Visual Studio?

>>8770046

I usually have a mess of notes of bullet points, but at least they're categorized. So it's not too bad.

>>8770095

usually it's better to keep connections going, and put things in as minimal words as you can. It's the arrows, with darkened ends, that really set things apart. Don't mind it though, what works for you, works for you

https://gateway.ipfs.io/ipfs/QmTmMhRv2nh889JfYBWXdxSvNS6zWnh4QFo4Q2knV7Ei2B

all the programmin you need,gigs of books and shit

>>8770115

I may start doing that. But I expect an SRPG from you in 10 years.

>>8770116

awesome, thanks for the extensive resource anon

for the agdg neets out there,where's your money coming from? this wage slavery is killing me

>>8770127

SRPG? I'm mainly going to make strategy/ rts/ tbs/ grand strategy games. I might do an SRPG….maybe. It depends on what my team wnats. I'm in charge, but I strongly consider what they do.

if any of you are interested in helping out with level design, we're still looking for people

>>8770148

Currently in between real jobs. I have over 10k saved up from working various jobs and from suing someone for almost killing me back in high school. Have a few very small investments, too, so that might help me in the near future.

>>8770148

i'm trying to get a job, until then i only have autismbux :(

>>8770139

np got it from /tech/,seems to from various torrents and sources,becareful some of the stuff will be way out of date, avoid any c++ befire 2003 i think, the bulk of everything is pretty solid tho

>>8770177

nice, i've got -13k :^)

>>8770214

tismbux enough to live with a roommate and pay bills?

>>8770228

I'll consider myself lucky if i can leave the house within the next 6 years

>>8770214

i'm barely making $9 working xboxlive, but working is mentally draining,leaving no reserves for pregaming :l

>>8770003

That's the opposite of what you just said you wanted.

>>8770257

nine?

as in nine dollars a month?

>>8770148

I have a small sum from tuition refunds, but it's not going to last me long. I've been searching for nearly a year now since I graduated, but haven't found any takers. I'm starting to wonder if it's something I'm doing wrong.

Fortunately, my degree is in software development, so whatever job I do end up with (if I ever do find one, that is) is going to mesh really well with my other hobbies.

>>8770293

What I want is for the image to always face the camera without the projectile itself changing course because of camera rotation. That's why I'm trying to store the transform value in Start() in a way so that it won't be updated later. If I could declare constant Vector3s id be golden.

>>8770529

Wait fuck I got that backwards.

>>8770545

Ok now you're adding details.

Anyways Billboards GRAPHICALLY always face the camera, there shouldn't be any changing direction otherwise unless you're doing something incredibly stupid with them.

>>8770363

l o l, an hour,which barely covers everything

>>8770439

if you have a software degree,whats the hold up? if theres no jobs in the area you might need to move

>>8768673

She is too bland to be a main character, maybe with clothes on…

>>8769403

use an empty object (or an invisible cube) and modify that then parent it to the particle you want to show

>>8770598

>whats the hold up?

Money, mostly. I wouldn't be able to move on whatever little I do have unless I just decided to gamble and hope I find one after I moved or knew I was getting a job out wherever I moved to.

>>8770663

I'm at a loss because I'm pretty sure with billboards the only thing that's supposed to match the camera is the 2d image you drop on it.

Dump it on a cube if you can?

>>8770728

whats the holdup on the actual search tho? if you can code + have a 4 year degree theres plenty out there, theres also work at home software dev jobs if you can code

>>8770751

I guess I've been looking in the wrong places? I've just been using Indeed/Monster/CareerBuilder all this time. Where do anons on here go to find jobs?

>>8770749

Work on those torso shapes a bit. Make the abdomen/pectoral divide a bit more pronounced and they'll look a lot less like doodles.

>>8769313

Man i couldn't agree more. I Just tried out the division at a friends house and was just fucking blown away at the amount of handholding in it. You could play the whole thing in the minimal, it even does both the world map and lacal pathfinding for you if you let it. Even the secrets are marked. It's a shame because the level of detail is insane, but you have absolutely no incentive to look around

>>8771029

Thank you, sir.

>>8771063

>the secrets are marked

What's the point then?

>>8770877

those are the usual, but actually upload your resume and make it public if you haven't ,usajobs is there as well, what area are you in?

and what can you actually do?

>>8770732

Jesus Christ I'm a retard.

I had the billboard inside an empty for hours, I'd just never set it as the referenced object in the editor. Six hours for ten seconds of work.

>>8770034

That equal weight problem should be solved with the heuristic component of A*. I think there are a lot of ways to go about speeding things up/ making it look more natural. First of all, you do not need to generate a new path for every actor, nor do you need to do this every frame, or even all in one. Clusters of adjacent actors can be treated as one, and should only need to recalculate paths for individuals if they encounter problems. Next you can distribute your pathfinding between frames, so no more than what your computer can handle to maintain its fps is ever done at once. One way you can make things look more natural with grid is to cast a ray to the next node in your path every frame. You will essentially read "up" your path and walk in straight lines between entrenches or exits instead of getting caught up in moving between tiles.

>>8770034

>the problem with grid is that you get a lot of paths that have the same cost according to the algorithm

What if you used Manhattan distance in the estimate instead of euclidean?

>>8768694

All I ever ask for is a map I can make notes on ala Ultima Underworld.

>>8771095

its the collectables. The game tells its story through them and partly through missions so you need to find them to have an idea of the story.

>>8772020

how do make it to not get lost in that shit?

>LightSource has a Vector3 for Position and Normal

>Only using it for 2D purposes, so position will only change XY values

>Normal is defaulted to Normal.Up, which is (0,1,0)

Will that fuck anything up, or will the shader I make "know" that the normal's Y value corresponds to the position's depth (Z)? Or is that too convoluted, and should the normal be changed to (0,0,1) instead?

Not sure if you guys have heard about this yet, but this is very relevant to you amateurs using gamemaker.

>>>/tech/547899

>Yoyogames, the creators of Game Maker, are deciding to delete some 150000 tools and games after being bought out by PlayTech.

Save whatever you can before they can nuke everything.

>>8772663

Archived and farted to Tumblr

http://archive.is/FEAPY

>>8772910

That model looks really good. Did you sculpt it yourself? If so, what did you use as a reference? I really like the pseudorealistic anime look.

Keep making these thread.

I'm in the works on getting this loan application started, but when it's all said and done I will be looking for employees, and I'd like to start here.

>>8773021

> I will be looking for employees

for what?

>>8772910

I gotta say anon, your work looks really good for a solo dev working with 3D.

>>8773051

WHOOOO MONEY

But seriously, what he said. For what?

>>8771445

I live near Philly, but I think that's the problem. I guess I've always been hesitant to get a job there because Philly is a shithole and the commute is depressing.

I have a well-rounded experience with most common languages (C++/C#, Java, HTML & CSS, Python, etc.), as well as general tech knowledge they teach all the IT kids. I.E. there's absolutely nothing special about me.

>>8773320

>Philly is a shithole and the commute is depressing.

I know that feel all too well, I have lived in PA my whole life. I had a job offer in Philly a while ago but didn't take it and luckily found something closer that doesn't require driving into 2 hours of gridlock both ways in the pavement ape jungle. Trying to move west as soon as possible because this place is beyond saving.

You can try craiglist, angel.co, and watch stack overflow and hacker news for job postings as well.

>>8773431

No shame in using a reference for inspiration or assistance. It's your own thing now. That's impressive. Do you use Blender? And, if I may ask, how long did it take you to not only get to the point where you could model this well, but also how long did it take you to do this particular model?

Any tutorials you might link me to? The only really good one I know of is this guy: https://www.youtube.com/watch?v=F1CTfis1TEg

>>8773021

hold on a fucking moment, do you even know what kind of game you wanna make? the last thing an aspiring dev needs its crippling debt for the next 5 to 10 years.

>>8773431

whose boobs are these from then?

Anybody need another pair of eyes to run over their story? I'm incapacitated for a while and have some free time to do so. I've found it difficult to get my own work looked over without hearing "Yeah, post it. I need a good laugh." So, I know it can be difficult to get honest criticism from someone that doesn't want to ridicule you.

>>8773021

>>8773051

>>8773058

>>8773671

Yeah, this isn't the sort of thing you can just casually drop cryptic hints about. I wouldn't even consider getting involved unless you spelled out clearly and plainly what you are planning to do, how you're going to do it, why you're doing it, who you need to get it done, and what you need to get there. I'm sure many other anons would agree; autists we may be, but we ain't stupid.

And you really should do it before you get into anything as serious as a loan. If your plans fall through for any reason, you better be prepared to pay them back somehow.

That's not to discourage you (or anyone else for that matter). Just be smart about it; plan things in advance.

>>8773737

Pretty much this.

>>8773465

that guy is a fucking wizard

>Any tutorials you might link me to

http://www.blenderguru.com/tutorials/introduction-to-rigging/

if you want into rigging

>>8773465

I've been dabbling in 3D for years.

Took my time with this particular sculpt, spent days getting it right.

As for the model that's in the game, I can't give you an estimate. She went through several iterations, and I reused as much as possible from each previous one.

>>8773465

And I use Blender, Sculptris and Silo.

>>8768694

As well as Thief, give the King's Field approach a thought too - you find maps by exploring and use them from the inventory but their accuracy is variable from one to the other.

>>8771710

It would be hard to limit which units do and do not get pathfinding without fucking over the gameplay but ill give it try

>>8771736

the problem with that is that when you introduce obstacles it doesn't always give you the right path at least i think so

the other problem with grid based is that i would need to interpret path

i want my units to use omni directional movement so i would have to detect certain patterns in the path and change go directly to the position where the pattern changes

but there is really no way to distinquish between a diagonal right left

and a right left that actually goes around an obstacle

>>8775755

>i want my units to use omni directional movement

That shouldn't be a problem, just evaluate on all 8 neighbor tiles instead of 4.

>>8776189

that would just give me 8 directional movement

>>8776253

My bad, I thought you were doing tile based shit. You have to analyze the map and place the nodes at the corners of obstacles, then.

>>8776267

yeah that's what im doing now

can someone who knows some vector math help me

this function is supposed to return true if 2 lines intersect

inline bool linesIntersect(const f2& start1, const f2& end1, const f2& start2, const f2& end2)
{
	f2 direction1 = end1 - start1;
	f2 direction2 = end2 - start2;

	// this is the determinant of the matrix
	float d = direction2.x * direction1.y - direction1.x * direction2.y;

	float s = (1 / d) * ((start1.x - start2.x) * direction1.y - (start1.y - start2.y) * direction1.x);
	float t = (1 / d) * ( - (start1.x - start2.x) * direction2.y - (start1.y - start2.y) * direction2.x);
	return s >= 0 && s <= 1 && t >= 0 && t <= 1;
}

but when i call it like this

  if (linesIntersect(f2(100, 100), f2(200, 100), f2(150, 150), f2(150, 250)))
  {
  	cout<<"something"<<endl;
  }

i get nothing

it seems to be working in some cases though and i can't figure out why

>>8777090

shit you are im actually retarded

>>8777090

>>8777116

  if (linesIntersect(f2(100, 100), f2(200, 100), f2(150, 50), f2(150, 150)))
  {
  	cout<<"something"<<endl;
  }

fixed it still nothing

>>8777123

>>8777116

>>8777090

ok i fixed it

line before return was supposed to be this

float t = - (1 / d) * ( - (start1.x - start2.x) * direction2.y + (start1.y - start2.y) * direction2.x);

>float s = (1 / d) * ((start1.x - start2.x) * direction1.y - (start1.y - start2.y) * direction1.x);

You sure it's not supposed to be direction2?

>>8777165

Fuck, I'm slow.

>>8777066

By the way, you probably want to check if d s close to zero or parallel lines will crash your shit.

>>8778197

you don't need to learn to do everything, only just enough for people to get interested to contribute.

while i was bored at work i made an implementation of the A* algorithm (in python that's all i have at work)…

it just makes a path between to static points, but for some reason it sometimes makes extra steps, and the path gets longer

maybe someone would like to take a look

or use as reference to make his own not shit code

I use a square grid, all the movement costs are the same, and you can't go diagonally

    def find_path(self, start, end):
        evaled = set()
        notEvaled = set([start])
        start.g_score(start)
        start.h_score(end)
        start.f_score()

        while len(notEvaled) != 0:
            current = min(notEvaled, key=attrgetter('f'))
            #print "."
            
            notEvaled.remove(current)
            
            for neighbour in current.neighbours:
                if neighbour.full: continue

                if neighbour in evaled:
                    if (neighbour.g - 10) > current.g:
                        neighbour.parent = current
                        neighbour.g_score(start)
                        neighbour.f_score()

                else:
                    neighbour.parent = current
                    #if neighbour != end:
                        #neighbour.canvas.itemconfigure(neighbour.idC, fill="cyan")

                    neighbour.g_score(start)
                    neighbour.h_score(end)
                    neighbour.f_score()
                    notEvaled.add(neighbour)

                    if neighbour == end:
                        path = [end]
                        while end.parent != None:
                            path.insert(0, end.parent)
                            end = end.parent
                        #print "found path"
                        return path

                evaled.add(current)        
                #if current != start: current.canvas.itemconfigure(current.idC, fill="yellow")
             
        return None

    def g_score(self, start):
        if self == start:
            self.g = 0
        else:
            self.g = self.parent.g + 10

    def h_score(self, end):
        tmp = abs(end.x - self.x)*10 + abs(end.y - self.y)*10
        self.h = tmp

    def f_score(self):
        self.f = self.g + self.h

>>8778515

pretty sure that's a lot more complicated than it needs to be

heres a more clean implementation

def a_star_search(graph, start, goal):
    frontier = PriorityQueue()
    frontier.put(start, 0)
    came_from = {}
    cost_so_far = {}
    came_from[start] = None
    cost_so_far[start] = 0
    
    while not frontier.empty():
        current = frontier.get()
        
        if current == goal:
            break
        
        for next in graph.neighbors(current):
            new_cost = cost_so_far[current] + graph.cost(current, next)
            if next not in cost_so_far or new_cost < cost_so_far[next]:
                cost_so_far[next] = new_cost
                priority = new_cost + heuristic(goal, next)
                frontier.put(next, priority)
                came_from[next] = current
    
    return came_from, cost_so_far

its from this page http://www.redblobgames.com/pathfinding/a-star/implementation.html

worth noting that he defines his own priority queue but its only a few lines long

i need haskell help again

i have a list of numbers

i want to sort them so that the ones with a prime index are in the front of the new list

no errors or anything, but it just doesn't give the right answer

what am i doing wrong?

isPrime n = all ((/= 0) . (n `mod`)) (takeWhile (<= bound) (2:3:5: scanl (+) 7 (cycle [4,2,4,2,4,6,2,6])))
  where
    bound = ceiling (sqrt (fromIntegral (n-1)))
quicksort :: (Ord a) => [a]->[a]->Int-> [a]  
quicksort [] [] counter= []  
quicksort (x:xs) originalList counter
    |counter==length originalList = xs
    |isPrime counter == False = quicksort (smallerSorted++ biggerSorted++ [x]) originalList (counter+1)
    |otherwise = quicksort (smallerSorted  ++ [x]++ biggerSorted) originalList (counter+1)
    where 
	smallerSorted = take (counter) xs
        biggerSorted = drop (counter) xs
    

>>8778515

Use a heapqueue bro. Also, what's that -10 and +10 stuff?

>>8778662

isPrime n = all ((/= 0) . (n `mod`)) (takeWhile (<= bound) (2:3:5: scanl (+) 7 (cycle [4,2,4,2,4,6,2,6])))
  where
    bound = ceiling (sqrt (fromIntegral (n-1)))
quicksort :: (Ord a) => [a]->[a]->Int-> [a]  
quicksort [] [] counter= []  
quicksort xs originalList counter
    |counter==length originalList = xs
    |isPrime counter == False = quicksort (smallerSorted++biggerSorted++targetElement) originalList (counter+1)
    |otherwise = quicksort (smallerSorted++targetElement++biggerSorted) originalList (counter+1)
    where 
	targetElement=[originalList!!counter]
	smallerSorted = take (counter) xs
        biggerSorted = drop (counter) xs
        

still not working

>>8778676

size of grid squares i would assume

>>8778652

i'm having trouble piecing his thing together scrolling up and down that page.. does he give the whole final source somewhere cause i can't find it..

as for cleaner it certainly could be cause I just wrote it all just like that, but i certainly hope you realise that fragment you posted is not the whole thing

>>8778676

>heapqueue

well i'll be… i'd never heard of this. I'll try it out tomorrow maybe, and yes the side of each square is 10px

(if i remove 10 from the "neighbor's" g score i'll get his parent's one to compare cause they always increase by 10 so that's simpler; if i want different ratings for terrain for example i'd have to change that)

been working on my pathfinding for a while

biggest bottleneck is this function

	inline bool straightPath(const f2& start, const f2& goal)
	{
		for (int i = 0; i < staticObstacles_.size(); ++i)
		{
			if (staticObstacles_[i].intersectsLine(start, goal))
			{
				return false;
			}
		}
		for (int i = 0; i < dynamicObstacles_.size(); ++i)
		{
			if (dynamicObstacles_[i].intersectsLine(start, goal))
			{
				return false;
			}
		}
		return true;
	}

it just checks every obstacle for line intersection and returns true if there is a straightpath from start to goal that doesn't intersect any obstacles

I use this to start and end pathfinding since im not using a grid based system and i also use it to connect nodes

intersectsLine looks like this

	inline bool intersectsLine (const f2& start, const f2& end) const
	{
		return linesIntersect(start, end, topLeft(), bottomRight()) || 
						linesIntersect(start, end, bottomLeft(), topRight()) ||	
						contains(start) || contains(end);	
	}

and linesIntersect looks like this

inline bool linesIntersect(const f2& start1, const f2& end1, const f2& start2, const f2& end2)
{
	f2 direction1 = end1 - start1;
	f2 direction2 = end2 - start2;

	// this is the determinant of the matrix
	float d = direction2.x * direction1.y - direction1.x * direction2.y;

	if (d == 0)
	{
		return false;
	}

	f2 startDiff = start1 - start2;

	float s = (1 / d) * ((startDiff.x) * direction1.y - (startDiff.y) * direction1.x);
	float t = - (1 / d) * ( - (startDiff.x) * direction2.y + (startDiff.y) * direction2.x);
	return s >= 0 && s <= 1 && t >= 0 && t <= 1;
}

anyone got any ideas on improving performance in any way

>>8779010

can you give us some visual representation of your map, i think it'd help

reading this

>>8769896

i'd say it's the way you build your nodes/links that force you to use this in the first place, and if you'd revise that you'd cut the whole thing

since you already have the algorithm itself, and it doesn't care about straight lines at all

>>8779205

you know i just realized you asked for my map

well i don't really have set map in fact this whole needs to work wel with changing obstacles

but this map has about 30 obstacles and it's pretty much totally random

also the black nodes are the path A* gives and blue and red are obstacles

>>8779239

i'll check it out when i get home…

otherwise good luck anon

Why does AudioSource have a PlayClipAtPoint method, but require the creation of a separate object for everything else?

>>8779205

hmm that's pretty much what i wanted to see, but i don't get it – are all the nodes themselves visible on this (i'd guess they are where lines cross)

>>8779239

>also the black nodes are the path A* gives and blue and red are obstacles

by nodes do you mean the lines on the pic… if yes you shouldn't be running your A* in this way - it is not meant to calculate the path to and from every node in advance (at least that's what it looks like sorry if i misunderstood)

Or do "90 degrees" mean you have "invisible" nodes at the corners of each rectangle

In all honesty though i'd suggest getting back to a grid map, because it will make you life easier (you can ditch the straight line code for good + the A* algorithm itself is meant to work on this kind of map): if you use hexagons instead of squares i'm sure you'll find it smooth enough,

>>8780389

>by nodes do you mean the lines on the pic

no the lines in the pic are the links between the nodes

the nodes are at every corner of the blue and red objects and are not visible

heres a better picture

the green line is the path A* provides

>>8780389

also i doubt it would be faster with grid based

considering this approach reduces the amount of nodes the algorithm needs to visit by a huge amount

with the added cost of having to generate the nodes everytime the map changes

which for my purposes is not actually all that often

>>8780502

you can usually ignore anyone in this thread

>>8780502

Ah, i think i get it now: so you use the straight line functions to find the links between the nodes since you don't know where the nodes will be exactly before putting in the obstacles?

what are the black squares though?

>>8780693

oh yeah they are obviously the nodes you'll pass trough.. now i think i see what you are trying to do;

i think it's interesting, and will think about it tomorrow cause i'm tired now

>>8781929

Armstrong would be proud.

> absolutely need realtime global illumination for a project

> only preexisting open source solution is a clustered deferred rendering clusterfuck layer on top of Panda3D

> OpenGL 4.3

It looks neat but I'm worried about performance and requiring DirectX 11-level hardware on Linux when Gallum3D doesn't fully support OpenGL 4.3 yet. Are there any alternatives that don't involve using a proprietary engine or implementing my own GI?

Tiling or clustered forward would be nice too, even though nobody but Unvanquished's developers seem interested in implementing it.

>>8782944

How is Panda3d? Would you recommend it to someone who's babby engine was Unity over something like Godot?

As someone whose only experience with programming is a little bit of Python, is Godot a good way to go? I've been going back and forth on which engine I want to start with, and I don't want to default to something like Game Maker, because I want to have at least a little bit of flexibility with my code and an engine-specific language doesn't seem very flexible at all (not to mention I know I'll end up defaulting back to drag 'n drop development, which will severely hamper my progress moving forward).

>>8784135

>defaulting back.

Trust me, once you start implementing things with code you'll be hard pressed to give up the freedom it provides.

Just make some shit in both engines and see which one you like more. Better to start making stuff and figure out your preferences now than stagnate trying to devine what engine will be best.

>>8784056

>he read the eula

I fucking love you anon.

Which engine gives me more freedoms, game maker or godot?

>>8784056

What the fuck is this shit? Is there any other engine that is anywhere near that bad?

Not even unity or UE4 are that fucking retarded

AGDG mumble. We are working on a mech game using godot engine

mumble-us.cleanvoice.ru 50688

>>8781989

>>8780693

it's worth noting that connecting the nodes is done during graph generation and not part of the A* search algorithm

so i can control those things seperately

>>8779010

made a pretty good improvement to straightPath

it looks like this now

obstacles_ is the same as static and dynamic obstacles combined

i realized that the seperating obstacles into dynamic and static wasn't doing anything so it's just obstacles now

	inline bool straightPath(const f2& start, const f2& goal)
	{
		f2 dir = goal - start;

		if (dir.x >= 0)
		{
			if (dir.y >= 0)
			{
				for (int i = 0; i < obstacles_.size(); ++i)
				{
					f2 dir2 = obstacles_[i].bottomRight();
					if (start.x <= dir2.x && start.y <= dir2.y)
					{
						if (obstacles_[i].intersectsLine(start, goal))
						{
							return false;
						}
					}
				}
			}
			else
			{
				for (int i = 0; i < obstacles_.size(); ++i)
				{
					f2 dir2 = obstacles_[i].topRight();
					if (start.x <= dir2.x && start.y > dir2.y)
					{
						if (obstacles_[i].intersectsLine(start, goal))
						{
							return false;
						}
					}
				}
			}
		}
		else
		{
			if (dir.y >= 0)
			{
				for (int i = 0; i < obstacles_.size(); ++i)
				{
					f2 dir2 = obstacles_[i].bottomLeft();
					if (start.x > dir2.x && start.y <= dir2.y)
					{
						if (obstacles_[i].intersectsLine(start, goal))
						{
							return false;
						}
					}
				}
			}
			else
			{
				for (int i = 0; i < obstacles_.size(); ++i)
				{
					f2 dir2 = obstacles_[i].topLeft();
					if (start.x > dir2.x && start.y > dir2.y)
					{
						if (obstacles_[i].intersectsLine(start, goal))
						{
							return false;
						}
					}
				}
			}
		}
		return true;
	}

gives me a good 30 fps extra when generating the graph every loop and it also spees up A*

also made some other improvements to node generation to speed it up even more

last issue is that the complexity of the path has a huge impact on the speed of A*

from over 100 fps down to 9 with the most complicated path i geuss it's back to callgrind for that one

>>8787971

yeah eliminating objects that cannot possible stand in the way helps, but as your path gets longer the time saved tends towards 0 (worst case being from corner to diagonal corner where every object could potentially be in the way) so that's what's slowing it down, and not the A* algorithm itself

one more thing you can do is since you now keep track of the direction of the vector you are checking, you can now target only one diagonal of the object to check against with linesIntersect() instead of both (knowing the directions means you know which of the two diagonals the vector will cross for certain if it passes trough the object)… so that's one less call i guess

Also when you pick a start and end points for your path do you dynamically create nodes for them… How do you bind them to the existing ones..?

>>8788139

>Also when you pick a start and end points for your path do you dynamically create nodes for them

for the start i parse the nodes and use straight path to determine the neighbors

and for the end i do this

if (straightPath(current, goal))
{
	cameFrom[goal] = current;
	return reconstructPath(cameFrom, start, goal);
}

also checking only 1 diagonal is an excellent idea will implement that now

>>8779010

In the linesIntersect function you can take out the division and check s and t against the range [0,d]. It took me an age to realize d being negative was fucking my shit up.

http://pastebin.com/nt4ZejFY

LinesIntersect1 is what you posted, LinesIntersect4 is with the change. It gives me a modest speedup (~450k calls per second compared to ~350k) but I don't know how much of that will translate from Python into C.

>>8788529

that's a good catch right there thx for all the help man

i think the path finding is almost in a usable spot now

>>8789030

But we learn from it, right anon?

R-right?

>>8788172

hmm what exactly do you mean by parse.. aren't the other nodes already there when you start the pathfinding?

>>8789040

anon, i've been gamedeving for at least two years now

the amount of retarded shit i still do can still amaze me

>>8789055

ill just show you the code its a little hard to explain otherwise

inline vector<Node*> neighbors(f2 position)
{
	vector<Node*> results;
	if (nodes_.count(position))
	{
		Node* node = nodes_[position];

		for (int i = 0; i < node->links_.size(); ++i)
		{
			results.push_back(node->links_[i]);
		}
	}
	else
	{
		bool bottomRight = false;
		bool bottomLeft = false;
		bool topLeft = false;
		bool topRight = false;
		
		for (int i = 0; i < nodesV_.size(); ++i)
		{
			Node* otherNode = nodesV_[i];
			f2 vec = position - otherNode->position_;
			if (vec.x <= 0 && vec.y < 0 && !bottomLeft)
			{
				if (straightPath(position, otherNode->position_))
				{
					bottomLeft = true;
					results.push_back(otherNode);
				}
			}
			else if (vec.x > 0 && vec.y <= 0 && !bottomRight)
			{
				if (straightPath(position, otherNode->position_))
				{
					bottomRight = true;
					results.push_back(otherNode);
				}
			}
			else if (vec.x > 0 && vec.y >= 0 && !topRight)
			{
				if (straightPath(position, otherNode->position_))
				{
					topRight = true;
					results.push_back(otherNode);
				}
			}
			else if (vec.x < 0 && vec.y >= 0 && !topLeft)
			{
				if (straightPath(position, otherNode->position_))
				{
					topLeft = true;
					results.push_back(otherNode);
				}
			}
			if (topLeft && topRight && bottomRight && bottomLeft)
			{
				return results;
			}
		}
	}
	return results;
}

the else is triggered if the position is not already a node which it almost never will be

it basically finds 1 node in each direction

It is worth noting that this same code is used for creating nodes but no node is actually created

Where can I learn about how logic gates work? Or is there something else that I need to learn before I can properly understand it?

I mean I get it for example than an AND gate requires 2 separate signals to pass though a signal, but I want to know how that gate is actually accomplished in practice.

I want to add some kind of an electronics building system to my game, and I want it to more or less mirror how real electronics work.

>>8789170

> but I want to know how that gate is actually accomplished in practice.

Transistors.

What engine should I use if I want to make a SRPG?

>>8789188

GMS. Always GMS.

>>8789188

I don't think you really need one if you're doing 2D stuff.

>>8789188

Make your own engine anon. Join us.

Anybody got example code from when they first started trying to develop their game ideas?

Preferably C++

Give me your shitty code please

>>8789170

I'd recommend looking into bitwise math (i.e. you can program operations with it), as it really helps to get the concept across; as you can experiment with it until you understand it (while using the same logic).

It's also extremely fast, and can replace a lot of otherwise, on a microscale, semi-costly operations if the compiler doesn't already optimize said operations

>>8789222

Why, u makin a banner or something?

>>8789230

>Why, u makin a banner or something?

Looking for fap material

Hey, MoM anon here, I wanted to ask you guys about compiling in doom (Before asking in the doom forums, that is)

Allegedly, you are free to compile mods in dooms, which allegedly removes any conflicted copyrighted code in doing so. I was wondering if any of this is true, since I am concerned Bethesda, at some point, will try and monetize on old doom mods and do what blizzard did by saying they own any original idea made in it.

>>8789236

Interasting, but sorry can't help u anon

>>8789292

>doom

>Bethesda

Do they own the rights to doom (the legal entity owning said intellectual property)?

I though idTech did?

>blizzard did by saying they own any original idea made in it.

Only if they own the intellectual rights, and have an EULA to back it up; otherwise it's up in the air as said content creators didn't agree to said terms upon using the software.

>>8789179

Well that seemed promising at the beginning, but I'm starting to get flashbacks from my early days of learning to program. The situation where the answers are baby tier simple 1+1=2 shit, but the people explaining are explaining it in the most convoluted and backwards way imaginable and expecting me to understand a bunch of other shit that they don't bother to mention.

>>8789343

Basically you have this shit called a semiconductor that's not really good at either insulating or conducting. It's got either extra electrons or too few depending on whether you get P or N type. Anyway the idea is if you put a charge near it (but still insulated), it scares the electrons away from that side (or attracts them). Then if you attach wires to that side you basically have a little switch that's controlled by putting positive or negative charge near it.

>>8789334

its the only major concern I have with making a project with the doom engine, is all. I feel like Bethesda is going be a cunt when they release doom.

>>8789292

I'm pretty sure Blizzard only had mod support for WC3, so they can pull that bullshit. Doom is open source licensed under GPL, so Bethesda can't pull such shit.

>>8789170

http://www.kongregate.com/games/krispykrem/kohctpyktop-engineer-of-the-people

You use transistors as PNP and NPN gates. This game by Zachtronics gives you nothing more than those two gates. It takes a little while to figure out how to do them in this game though.

>>8789428

>>8789334

Thanks for reassuring me!

Also, I am working on areas and stuff right now, as well as a little surprise. I don't think I mentioned this before, but the game also has fully 3D characters and 2D sprites, so they are many surprises still to be had!

>>8789089

so it goes into the else for the start point of your pathfinding because it hasn't existed previously I guess?

my point is just make sure that once you have build your map you already have the vectors of neighbours of all your "static" nodes, and when you insert a new one for the start of pathfinding you don't call this for all the nodes, but only attach the new one with new link(s)

>>8789488

i won't be adding nodes to graph at all though

only obstacles

and if an obstacle is added or removed every node does need to be reconstructed

or i could go through every node and reconstruct every link but really wouldn't be much faster since thats most of the work for reconstructing every node

>>8789414

I'm missing the basic concept of how this works in practice. In the beginning there's only one source of electricity, but it needs to translate into a system that can switch things around.

Reading about this it sounds almost like current can loop through itself infinitely, because otherwise there's no way to switch this gate on/off and have it remember it's state when the system no longer receives electricity, since in order for it to be flipped it needs to be receiving a current. Is that how electricity works? I don't know. Maybe I'm missing something about the terminology and basic concepts of electricity itself.

>>8789428

>http://www.kongregate.com/games/krispykrem/kohctpyktop-engineer-of-the-people

It's taking me a little while to understand what the hell I'm supposed to do period. I have no idea what I'm supposed to be doing with any of the anything.

>>8789292

Unless they license states that the own the assets you made in their tools, or loaded into their engine, they don't have rights to it.

Note that you should never ever try to be a smart ass and monetize a mod, since then they can always argue that you're selling their stuff, even if you require other people to buy the game you're modding first. Although you don't TECHNICALLY break the law, courts are shit and technologically retarded, so they wont be able to tell the difference.

But if you take the route of Dota2 and produce a completely separate game when it gets popular then you're probably safe.

>>8789559

I don't plan on monetizing on the mod at all thankfully.

Actually, here's something I should look into; I wonder if I can get the game to run without doom…

>>8789542

Nah, there's latches and flip-flops that can 'remember' their state until they're signaled to update. They're a little quirky but the work. The only other thing you need is a clock to generate that signal every cycle.

>>8789580

Okay so this is what's confusing me the most: NOR gates.

It supposedly outputs a high signal if both of the inputs are off. But if there's no input, where does the signal come from?

>>8789622

with the quake 3 importer, all the cubes in the map maker get turned into cube-like mesh colliders in here

imagine how i feel when my stairs end up having 5 mesh colliders

>>8789695

Its funny you mention the quake III importer, thats what the Zdoom engine uses too.

What are you making your game in?

>>8789803

I don't understand what any of the cubes on the sides are, and I could not figure out how to place the second type of silicon.

>>8789816

>The inputs aren't currents, they're voltages.

Well I think there's the missing key distinction I was looking for. Reading through all this jargon online made me overlook that.

I guess I do need to read more about electricity itself, because I have no idea how current and voltage would interact with each other in the context of these things.

>>8789966

Actually I might be remembering that part wrong. Anyway the gist is all the components have both the data inputs and an input from the battery.

>>8789941

I'd trash the materials if any are brought across and reset options under Window > Lighting if any had been changed.

>>8790029

Oh well, I think I'm going to give up anyway, because I went down the wikipedia rabbit hole where learning about one thing requires me to learn about 5 other things, which all require me to learn about 5 other things, which all require me to learn about the first thing.

I don't even know what electricity is anymore.

>>8790067

Electricity is a binary gender you cis shit

>>8790065

i've tried clearing the old data via the lighting window multiple times

tried changing all the materials in the scene to be the default material

still the same issue when the lights are baked

>>8790140

Wow just wow are you saying there's no such thing as a cisistor?

>>8790291

>So this is what game dev feels like

Pretty much.

>So, I need a cutscene before the fight

>5 hours later elbow deep in writing a lexer

>>8789622

Really?

I can have around 1000+ physics rigidbodies moving at once, doing guidance too (proportional navigation… so it predicts where they're moving based on the missiles velocity relative to the targets velocity, turn rate, and corrects guidance based on those three variables).

TL;DR

I don't think it's the physics system, which is a multithreaded system since Unity 5.0, it's probably your code being inefficient.

F.e. it could be you doing too many fixed updates, doing physics updates in the wrong update timestep (like doing physics stuff in Update(), when you should be doing it in FixedUpdate() => the physics timestep), moving static colliders, etc. etc.

>>8789941

That's what it looks like when you're missing normals… or using cast shadows and have an odd orientation of your directional light + your lightmapindex is set to hard shadows.

Disclaimer: I don't use bake, only realtime, so I dunno if it messes with your normals or not… or if this is the issue. I'd check it out thou

>>8790067

electrons anon, electrons

>>8790291

Simplify your approach, and implement what you can; then refine the system after u get something working.

>>8790433

Setting box velocity in FixedUpdate, having them slide along surfaces.

>>8790464

I don't use 2D physics, and unfortunately I don't have to time to assist atm (studying for a calc exam, and my self-allotted break is over)

Though, I'd use the deep profiler to determine the bottleneck

it could be the GC if you're creating a bunch of new variables that go out of scope every fixedUpdate (which racks up pretty quick for so many GOs/updates), it could be moving static colliders, or it could be that 2D physics isn't as optimized as the 3D physics.

Profiling it is about the only way to tell, and I can help later tonight after my exam if u post what the profiler says

>>8789515

thing is you already adding the nodes to a graph, your nodes, links, and neighbours form a graph, which the A* algorithm works with as it is graph traversal in its essence

I was going to suggest you make adjustments only to nodes affected by changes, but then i realized you'd have to check if the new object "disrupts" an existing link as well so it would hardly be faster…

In all honesty i still believe you'd have been better off with a grid based map, because then you would have had no need for literally every piece extra code you have put here up so far..

my unoptimized mess of an A* evaluates 15k nodes in 1.5 secs yeah it's slow, and that's written in python

now you throw that in c++, and with your maps that have more free spaces (from the looks of it), and it would fly like a bird…

but i know how it is when we start something, and want to stick with it

>>8790684

>I'd rather have the player create the component in pic related with more basic components

How are you going to teach the player that that's what he needs to build?

>>8790654

it really wouldn't be

event though im doing weird stuff with the graph

the A* algorithm is pretty much unchanged

like you said A* is just a graph traversal algorithm

and the less nodes the faster it is

with this approach im reducing the amount of nodes by atleast tenfold

with the added overhead of needing to generate the graph

however having these seperate also gives me more control

i only need to generate a new graph when obstacles change

so this is definitely the faster approach

>>8790758

You're supposed to build whatever you want; use the basic components creatively to make more complicated systems. It's only a part of the game. It was mostly inspired by Minecraft redstone machines, except I wanted it to have more practical application in real life.

It'll be able to interact with things like an audio player, moving "robots", and hopefully a display screen. I have no idea how I'd use 0s and 1s to print an image at this point though so the monitor is less likely to happen.

last post for awhile

>>8790684

>I can program it

I assumed as much, I was meaning the initial implementation of the mechanic when it comes to my line, "simplify your approach".

>This whole problem appeared when I wanted to add a switch that's toggled on/off by electricity

Then just do that, and another system that can be activated/used by said basic mechanic.

Don't over complicate the mechanic at first, just make it simple, and then expand on it later once you have a basic foundation to work from.

I.E. if you try to build a houses walls and bedrooms before you have a foundation, you're gonna have a bad time.

>>8790684

>For example, I'd rather have the player create the component in pic related with more basic components assuming it's not massively more complicated.

What other components would they need to build?

Please give me your shitty C++ noob code.

I'm told the only way to learn a language is by looking at it, and I want to see something at least a little interesting.

>>8790850

I learn best when I understand the subject at hand thoroughly, that's why I'm also an engine dev.

I think this is already very simple, but right now using the component in my mind would be building walls and bedrooms, because I don't have the foundational understanding of how these things are supposed to work. That's how I'm seeing it anyway.

I guess the best way to summarize my problem is that I want the least amount of building blocks possible to give the most amount of possibilities, without going into hyper autistic territory. Logic gates are around the level of complexity that I want to have, but I don't understand how they would work in practice therefore am unable to add them into the game. I almost need a video that shows electricity moving through these things.

>>8790856

I don't really understand the question

I want to give them a bunch of basic building blocks and let them do whatever they want with them. And it must be possible to build just about anything.

>>8791089

here's a maze solver

#include "stdafx.h"
#include <stdio.h>

const int MazeHeight = 9;
const int MazeWidth = 9;

char Maze[MazeHeight][MazeWidth + 1] =
{
    "# #######",
    "#   #   #",
    "# ### # #",
    "# #   # #",
    "# # # ###",
    "#   # # #",
    "# ### # #",
    "#   #   #",
    "####### #",
};

const char Wall = '#';
const char Free = ' ';
const char SomeDude = '*';

class COORD
{
public:
    int X;
    int Y;
    COORD(int x = 0, int y = 0) { X = x, Y = y; }
    COORD(const COORD &coord) { X = coord.X; Y = coord.Y; }
};

COORD StartingPoint(1, 0);
COORD EndingPoint(7, 8);

void PrintDaMaze()
{
    for (int Y = 0; Y < MazeHeight; Y++)
    {
        printf("%s\n", Maze[Y]);
    }
    printf("\n");
}

bool Solve(int X, int Y)
{
    // Make the move (if it's wrong, we will backtrack later.
    Maze[Y][X] = SomeDude;

    // If you want progressive update, uncomment these lines...
    //PrintDaMaze();
    //Sleep(50);

    // Check if we have reached our goal.
    if (X == EndingPoint.X && Y == EndingPoint.Y)
    {
        return true;
    }

    // Recursively search for our goal.
    if (X > 0 && Maze[Y][X - 1] == Free && Solve(X - 1, Y))
    {
        return true;
    }
    if (X < MazeWidth && Maze[Y][X + 1] == Free && Solve(X + 1, Y))
    {
        return true;
    }
    if (Y > 0 && Maze[Y - 1][X] == Free && Solve(X, Y - 1))
    {
        return true;
    }
    if (Y < MazeHeight && Maze[Y + 1][X] == Free && Solve(X, Y + 1))
    {
        return true;
    }

    // Otherwise we need to backtrack and find another solution.
    Maze[Y][X] = Free;

    // If you want progressive update, uncomment these lines...
    //PrintDaMaze();
    //Sleep(50);
    return false;
}

int _tmain(int argc, _TCHAR* argv[])
{
    if (Solve(StartingPoint.X, StartingPoint.Y))
    {
        PrintDaMaze();
    }
    else
    {
        printf("Damn\n");
    }

    return 0;
}

>>8791089

how much shit code you want fam?

>>8791194

>And it must be possible to build just about anything.

You can build just about anything with transistors.

>>8791212

An entire pastebin worth

>>8791228

That might be in the hyper autistic territory I was talking about.

>>8772910

what is your game called again? i knew it once but i forgot.

>>8791264

heres an old abandoned project that barely works

and doesn't really do anything even if it does work

http://pastebin.com/index/6yVGGC5d

>>8791325

Cheers boss.

>>8784056

>>8784436

>>8785334

Reminder that Yoyo Games is being purchased by another company soon (or has it already happened?) and that could force another change in the EULA.

>>8791337

I don't think you're supposed to look at code that doesn't work. I think you're supposed to look at good code.

>>8791429

I can't tell the difference.

>>8791455

The point is you're not going to learn anything from bad code.

>>8784056

No, the eula defines the "Software" as the installation package and the associated media and applicable documentation. So everything that it came with GameMaker it's theirs but not what you create with it.

What you created in the eula is defined as a "Game" and is bound to the "Games" part of the eula if you need to know what really influences your game.

so i remade the gym and added windows for some stupid reason, even though it's underground

still can't quite get the right lighting though

might be my poor choice of textures

also, is there any better way to do reflections other than a reflection probe? my mirrors look like shit

>>8785334

>tfw I saw cuckteam as well

>>8796726

do you think they're going to find out or something

>>8796892

I'm also a bit of a paranoid fuck.

Is Godot good now?

Also if I was going to be making an arena shooter, what idtech-based engine should I use?

>>8793992

>I've wasted over a week on this shit, and I had it working before.

Well hey, surely you take people's advice and set up git or some other form of version control so you can just rollback to your working version, right?

>>8796875

This tune makes me think of Chinese cowboys in the desert.

I made a unity addon but I'm not sure how to shill it

Any ideas?

>>8797497

Subtle of overtly?

I mean, Unity probably has a creator's forum for just that purpose, you can also try Reddit and stuff. Don't misrepresent who you are, say you're the author and it's for sale. Don't pretend to advocate a new product with a new account, you'll hurt their trust.

>>8797578

Well, nothing in bad faith. I just want people to know that it exists.

http://play.pokemonshowdown.com/battle-ou-349026916

>>8797768

>>8797768

Greetings from fullderp thread, our chimp can't contain himself. Sorry for the inconvenience. Polite sage.

>>8795226

Does player actually have x/y members (especially after the newImage call)?

>>8796948

Not like you can't just make it in sepples with SFML.

>>8793992

>I am using different render targets and blend states, but it just won't work.

What exactly is wrong with the result?

>>8784315

The real problem with fusion is that all coding is attached to the level and not the objects…

You end up having to either make a level editor yourself (with it's own format to save to and load) or copy the level over and over to make new ones (possibly bringing along bugs making it hard to change later in development)

http://store.steampowered.com/app/373290/

When I see shit like this being sold I actually get sick to my stomach. How the fuck do these people not have just a tiny bit of shame before releasing anything like this for a price? I just feel fucking sick.

>>8798406

>Cannibalism! Eat the corpses of your enemies to restore health. But beware, unless you find a special item, you can choke to death.

Maybe it's the description but that sounds like shit

>>8796726

as long as you have local copies of your game and you distribute them with a vpn there is nothing they can do

not that they care enough to do anything anyway

>>8798406

Protip: Release your game and do a better job and charge $20

>>8797578

I feel a bit dirty just posting it places. Is that just what people do?

>>8789428

>http://www.kongregate.com/games/krispykrem/kohctpyktop-engineer-of-the-people

>You use transistors as PNP and NPN gates. This game by Zachtronics gives you nothing more than those two gates.

Fuck that game. Just the first level is impossible. Even trying to reproduce the tutorial from the video help is too much.

>>8798716

I see. I created separate media accounts separate from my shitposting, but I'm finding myself caring less and less. Should I be worried?

So does Unity have a map editor that I can fully use to create my game or not? Do I have to make my own map editor?

Also is Unity's C# very similar to the actual C# meaning that if you learn unity you could eventually have C# on your resume?

>>8798839

>Also is Unity's C# very similar to the actual C# meaning that if you learn unity you could eventually have C# on your resume?

It's not "unity C#", or anything like that it's just component based C#. Easy to misunderstand, considering the abomination that is unityscript.

>>8798862

So then I can just move on to other C# projects if I get bored of Unity? That would be cool if I could use that and have it on my resume.

>>8798892

You'll probably get sick of all the fucking bugs that stick around for months after getting reintroduced way before you get bored of it.

I would learn the language fundamentals before learning Unity, though. Makes everything so much easier when you can separate what you're doing and what the engine is.

>>8798915

I am learning C# to use it with unity from here

https://www.youtube.com/watch?v=12Ste_E1Q7g

Is it really a bad idea to start with this and not the fundamentals? I mean if I search for the fundamentals of C# I get the basics of programming which I don't need to know again. Since I know javascript can't I just learn C# from unity related tutorials?

>>8790843

you are missing that you call straightline() between the start and end every time you need a new path, which possibly iterates over all the objects that exist (as i said in >>8788139 ), and then call the A*

So you have that on top of generating a new graph each time an object is added/removed, and that's why i'd wager a premade grid with more nodes would be faster

>>8798406

>CO-OPTIONAL GAMEPLAY

wat

>>8799576

Add perfect blocks, that cost you less energy if you put up your block at the last moment instead of keeping your guard up constantly.

>>8799666

yea satan, i should do that

i think i'll just add a timer for when you start to block

and when you get hit, if the timer is less than 1 or whatever, you don't lose energy and health

otherwise you lose energy and half the damage

if you're not blocking you lose energy and all of the damage

>>8799538

now that straightpath is properly optimized it should be fine

>>8802280

unity :^)

What do you guys think about Death's Gambit? It's made with game maker.

>>8802714

i'm sick and tired of pixel art

otherwise it looks decent

>>8802771

Are you tired of pixel art or just lazy pixel art? Because some pixel art games still look just timeless for me, no matter how old they get they still look good. But lazy pixel art… that's a difference. I don't know though I think death's gambit pixel art looks much better than most new indie releases.

>>8802804

pixel art in general

it's one thing to do it in the 80s/90s, it was a necessary thing then

but now it just feels like everyone is trying to shove it down my throat

>>8802817

I agree and disagree with you, on one hand there are waaaay to many pixel art games out there, on the other hand it's very cheap for indies to make it and with effort it looks really good, without effort it looks worse than a flash games. I've had enough pixel limb games, but I'd like to have another game with good pixel art.

But yeah, more games looking like Rayman Legends wouldn't be a bad thing, it's not pixel art and it looks beautiful. I guess it depends.

>>8803492

Man, you should had at least put the effort to write an introduction/description. Maybe post some webm footage.

>>8803492

Either way your game reminds me of that weird mmorpg Worlds or whatever it was named, looks old school/trippy

>>8778197

Same

>>8804251

Probably localization, assets, and other such things that are selectively chosen.

>>8778197

Same

Let me post.

>>8804804

That makes no sense.

>>8803492

could you tell us what the game is about? after watching the trailer and reading your one blog entry i still have no idea.

>>8800147

im already doing that

>>8799881

thats not a bad idea

>>8805584

I forgot to add, specifically the sorcerer class.

>>8804941

Absolutely, as I mentioned in an earlier comment, I was inspired by demon souls to make a sorcerer-ish dungeon crawler. You move by holding down the shoe button and tilting the screen where you want to walk. The general idea when casting a spell is that you tap where you want to cast it, but some spells are different. Fire lvl 2, for example, requires that the user tap and drag, thus moving a ball of fire around that spits out smaller balls of fire behind it.

Some of the bosses are very difficult, and one should feel very proud indeed if one can finish this game. I myself took more than five tries to kill the last boss. Then I made it ever so slightly easier so as not to put off those of a more n00bish bent.

Some day I will probably port this to pc with better graphics, 3rd person mouse orbit/fps, and with an inventory/equipment system, but for now, I think this is quite an appropriate adaptation of the souls je ne se quois to android devices. I would rather play this than "Slashy Souls" (http://www.please use archive.is/2016/2/27/11125424/slashy-souls-first-look-gameplay-video-dark-souls-mobile-namco-ios-android)

At any rate, I would be positively elated if someone here was able to finish even the second world.

>>8805684

>You move by holding down the shoe button

>Some of the bosses are very difficult, and one should feel very proud indeed if one can finish this game. I myself took more than five tries to kill the last boss. Then I made it ever so slightly easier so as not to put off those of a more n00bish bent.

autism?

>>8805734

Accelerometer input does sound fancier…

>>8805684

why would you make a game like that for mobile? twin sticks or kb+m wseem perfect for your kind of game. Also why are bones taking up more than a quarter of the screen?

>>8805738

What is autistic about that?

>>8805757

I really do regret doing this for mobile. Also, though it's not terribly obvious from the screenshots, the bones are transparent, and are for style. I'll eventually add the option to disable them.

>>8805763

It just reads like your only experience with writing comes from anime

>>8805809

Yikes. I guess all that Kant and Shakespeare went in one eye and out the other.

>>8804598

Duuuuuude, I want to know what game you're talking about.

>>8806538

yeah it's called, "Worlds", that anon seems like an /x/phile, so it's more than likely what they're talking about (that's the only board I've seen that makes threads about it).