/agdg/ + /vm/ - Amateur Game Development and Modding

It would be neat for devs to give more in depth updates on the wiki on a project page if they don't have one elsewhere, since the /agdg/ board is pretty dead. I fucking love doing change logs so it's a motivator for me too.

Youtube tutorials to start, unless you want to bore yourself to death

Usually, I'll just download tutorials for something I think could be useful, change stuff and see how badly it fucks up.

Maybe I'll add my game next time I make non-black triangle progress.

complete the most basic tuto, shooting spaceship did it for me

The very first thing I usually do is make pong in it without reading ANY documentation. Then I look up documentation, see what I did "wrong" and look up some parts of the engine that interest me most for what I want to make.

you will coder for byond

Am I gamedev yet?

You are doing God's will.

not enough if/else chains

Yes, my child

Yes but jesus christ that is one fucking ugly programming environment.

Needs to be cleaned up of all the default shit they give you but it's okay.

BEHEAD ALL WHO INSULT TEMPLE OS

Good, I wouldn't want to get someone's hopes up
;__;

I did this

I like this

nice

I wrote full Pater Noster with it.
I also deligned glyphs for common combinations of letters like st and nt.
I want to add more particles like one that tells the repetition of a vocal.
I think T is common enough to turn into a particle that is atatched to the consonant before and if there are no consonants before the normal T is used.

Tutorials first, documentation second, experimentation third.

...

looks like it would make a cool ancient language to put on ruins

I love it.

...

Nevermind, it just took a gorillion years to process a "small" version of the image

Give it to me straight. If I wanted to make this but my experience with Unity is rather basic(Think end-of-1st-year coding student). How much of a difficult undertaking would making a game like this be?

Nice going, Chode Monkey.

Anyone know anything about netcode? ATM I'm literally appending my packets to a global string then sending that. I need to make some sort of wrapper function but am not sure the best way to do it.

Thoughts?

Is that supposed to be a gun?

I already do, faggot

Y-yea.

needs one line in the center

Like this?

better o was just thinking of moving the line over a bit
and the highlight should widen with the rest of the barrel

5 mirrrion hours in ms paint

Didn't get it, sorry, but your sprite is quite artistic. By the way, is it wrong to overpost progress?

not if its animated

The most time-consuming thing in fighting games is doing the animations and getting the hitboxes good.

Good thing I took 4 years of animation in high school. Granted none of it was 3D, but I have probably the best foundation going into this without going to school for 3D animation. I've made some alright SFM animations, although I never finished any of those projects(usually because of something breaking that I couldn't fix, it was in beta after all).
Is doing hitboxes in Unity easy? From what little I've seen it seems like the easiest part of the engine to do second only to making simple physics objects.

Every. Fucking. Time.

Attach hitboxes to each of the character's bones, turn them on/off for different attacks. ezpz
Watch vid related and think about it as you're designing hitboxes and hurtboxes. It will help you keep things balanced on top of the hours of playtesting and tweaking you should be doing anyway.

you ==will==

Experimenting with props.

That's a big mushroom

...

Very crude inventory menu. Background shooped from something on publicdomainphotos, need to convert it to pixelshit.
Now to put inventory items in it.

Glad to know you are alive, pixelman, what engine are you using?

I finally abstracted out everything necessary for my inventory/equipment system. Going to do player health/stamina etc. shit first before making actual GUI for it though, because I know that's going to suck.

Technically the actor has nothing in their inventory but the sword (rigidbody986) in this image, the armour and boots are just fused to the character model until I actually make my "update_paperdoll()" method change the visual appearance of the actor based on their equipment.

Where is that sprite based game top down thats like some what Indiana Jones? I mean the pixel art and sprites are really good too. Solo dude making it. Seen him here a few times.

The one with the mexican gunslinger, right? There's also that one dev working on the 3D game with a blue-guy with ridiculous prat-falls and gondolas. And then there's also the dev working on the 2D hand-drawn brawler.

Last time I remember it was a white character. Nothing really about mexican or gunslinger. Unless you got pics idk. But I can tell you there were some dungeons and monsters.

I remember these, its two diffren't games. Don't have pics though. Those guys barley post on here.

Just fuck my boards up fam.

"projects on hiatus" page most likely. I have a few as well.

dude, I started the thing yesterday and added like one game (Hellbreaker), while other user (thanks Anycomb) added projects that only appeared in the last thread.

You'll be found, people just have to start contributing. While /agdg/ can't make a game together, I'm sure collaborative efforts will at least lead to the creation of a comprehensive chronicle of all you dudes' hard work.

It's an open wiki. If you are missing something, add it. I'll be sitting down for some research myself soon.

Protips:

I don't trust that shit.

well fuck me sideways.

should tetrachrome be on the wiki? can someone that is in contact with him ask if he wants it on there?

Respect to the guy on one hand, historical value on other.

Where's a good place to start for someone who has a software dev background but doesn't know how to make games?

2D or 3D ?
Also do a roguelike, it's a good beginner project.

I have an idea for a 3d game but I presume I won't be able to make it as a first game. Do most of the skills for writing 2d games transfer over or is it a whole other ballpark?

In a top-down tile-based game (such as a RL), how would you handle having different heights? It's not easy to represent in this view

Same shit but you have to add one more diemension.

Btw, when I say add I don't mean you have to add anything, the third dimension is already there, but you will have to think about it instead of everything just being flat.

Loli hugs!

I can finally live out my dream of being a wooden barrel.

The smile's cute but her eyes not focusing is real creepy.

Not sure If I understand. The eyes are always focusing on you. It's creepier when I remove the small subtle pupil rotational offsets if that's what you are referring to? It also makes her look more alive with them.

Her eyes don't look like they're focusing on the camera.

Not the same guy, but I think it's about the moment she opens her eyes but her head is still turned.

I can't quite put it into words but there's something off about that moment in particular that pushes this into uncanny valley territory.

If you want to enginedev, then making a 3D game is a lot harder and you'll need more math skills.

If you use an engine like Unity or Unreal engine, then they'll mostly handle the 3D physics and maths and graphics for you.

That environment looks comfy as fuck, feels like I'm back in the 80's

It's attention to details that separates the meme from the masterpiece

Huh. Do they do look like to me. I could try manually animating the eyeballs and override the hug animation with them.

...

Over half of that video is spent on talking about irrelevant shit instead of just explaining each phase of an attack in a sentence or two. Explaining that the active part of it is when the hitbox exists and it can damage the opponent shouldn't take that long.

Update:
We have "walls" now. Still no collision detection though. Took me a while since I wanted to use a linked list to have an index of every wall, which I've never done before, so I got hung up on some pointer stuff. At least it's nice and organized and should be easy to modify.

Also I made some fresh Terry OC for anyone interested. Get it while it's hot.

Does this blood spurt out in the right or the exact wrong direction? Also how do i make it look better?

Use particles instead of 2-3 images like you're doing. If you want to get really fancy, figure out where the models intersect, and create a quad/plane based on that, then generate many particles with a point of origin from there

I'm using GZDoom. Recently went GPL 3.0, so it can be used to make games.

You mean solid color particles instead of sprite particles entirely? Right now this is a burst emitter that shoots 2 to 3 particles each of 2 different sprite materials, and its spawned at the impact location and faces the impact normal. Im playing with the saturation and brightness but i think my images suck in the first place, and ideally i would have spritesheet based animations for the spurts.

Well fuck me sideways, am I gonna have to make progress reports again? Not like I stopped working on it, I just kinda took my time, especially since there's nothing to show other than pages of character descriptions and lore stuff. And a puzzle game I made for the in-game diner out of boredom.

What's your game ?

Hand over your source code, neckbeard.

en.wikipedia.org/wiki/GNU_General_Public_License#.22Viral.22_nature

Is the name of your game Last Dawn ?

do you have any tutorials for getting started with making games (not doom mods) with GZDoom?

Depends. In real life, blood spurts are uncommon, if anything it's more likely the blood would spatter in the direction of the blade's movement, so right or wrong is pretty relative, games usually don't have realistic blood mechanics.

How far do you want to go? What you have now can work, just work on the sprites a bit and add gravity acceleration to them, that alone will go a long way. If you want to be fancy, look into mesh particles.

can't quite figure out how to scale the army width/height correctly, but at least i sort of managed to get the cavalry to always stay on the flanks

Im thinking that as well, that would be the opposite of the direction it's going in right now. I have a Rotator (Pitch, Yaw, Roll) for that data and i can't figure out how to get that opposite direction, the native .Inverse() method doesn't seem to be the right thing.

I decreased the saturation and brightness a little, looks like webm related now

That looks pretty overkill to me

it's a dumb solution, but put a object at the tip of the sword, register it's last position.
then just splash the blood in:
lerp(swordTip.position - swordTip.lastPosition, swordTip.position - player.position, Random(0, 0.1))

I might have those vectors inverted, don't want to bother thinking about it

That would work i think, but i already have this impact normal, it can't possibly be this hard to "invert" or whateverthefuck this thing so it points in the opposite direction, i'll have to look it up later

hey, man. I'm currently researching this long gone project that was developed on /agdg/:
8agdg.wikidot.com/general:onhiatus

It's code is actually well documented in the archived threads, you might find it handy
archive.is/xmOZH

kek
yet you know what I mean, code-boy

i am the aoedev, i gave up on that because i can't into efficient pathfinding
doesn't actually have anything i need, in there i was trying to do everything with A* and a grid, here i'm just using rigidbodies and very simple local avoidance.
are we talking about the other one? i can't find the source code on that

VAMPIRE KILL IT

Maybe we need a "project graveyard" page.

That'd honestly be a good idea, just to put things into perspective and to bully people for dropping their projects.

Better, but you probably need to tone down the velocity a bit, when it does fly off 2m into the air with no trail or anything it looks pretty bad. Two ways to do it: either you use bigger sprites/flipbooks that have the shape of a spray, or you emit more particles in a cone with randomized velocity so that they form a more or less consistent spray (also add gravity and drag so they slow down).

If you are getting the impact normal, I assume it is a unit vector. All you need to do is multiply it by -1 and you will get the direction into the surface you hit rather than out of it.

I think I made a post there about there being 4 RTSes in development. You were the only one to post code, Tetra mentioned his project only once.

I'm glad you're still here
I'll have to make a wiki page for you

So "On hiatus" isn't enough?

What about plans for the demo day by the way? I think it would help anons stay focused, on deving in general and on the most important parts for a vertical slice.

I'll try those improvements on the sprites, and i want flipbooks as well.
It's a "Rotator", with Pitch, Yaw and Roll

Hiatus implies that it's going to be worked on again later.

well… you never know, right? ;_;

So what about projects by anons who never said goodbye or that they quit? Stuff like that Doom Wad from the same thread

unless there's something I don't know
one of my motivatios to start the wiki is to fill in the gaps since I have long periods of inactivity

I sort of agree that "Project Limbo" does sound better than Project Graveyard.

www.assetstore.unity3d.com/en/#!/content/17660

On hiatus implies the game is still worked on, but not at the moment. I'd just put those into active projects because it's not that big of a deal.
A project graveyard page would make sense for games that are confirmed to be dropped or dev vanished off the face of the earth. You could set some sort of time limit before you move a game there, like a year or something. And you could always put them back into active when the dev does miraculously come back.

nevermind im fucking dumb, i do have a vector and scaling it with -1.0f does the trick, thanks user

yeah, the old RTS is dead, i've reinstalled my toaster at least twice since then and haven't bothered to save anything
i've got this uploaded, but don't feel like bothering to check what's inside. might be the whole project, but last time it was unusable because of the shitty pathfinding
if anyone wants to pick it up, change A* to navmesh first
drive.google.com/open?id=0B1laio2k4HufN0ZoTTh0Q3VsNk0

www.assetstore.unity3d.com/en/#!/content/30247

but that's just using navmesh
which i was originally trying to avoid

why? you did the project before navmesh became free in 5.0?

because i couldn't figure out how to do decent local avoidance between units
went with A* since i thought i could just make them treat each other as an obstacle bad idea
and now i'm just using this gamedevelopment.tutsplus.com/tutorials/understanding-steering-behaviors-collision-avoidance--gamedev-7777

Okay then, let there be both "limbo" and "on hiatus" pages then.
That's mainly because I personally can't let go of things and have at least 3 projects I'd like to return to and spamming the "Active projects" page (which is what most people will visit) with my stuff would look too much like vanity.

I think adding a page with archive links would be wise too. I am really angry at myself for not saving the few threads I have on my HDD with full-sized images or webms. Time to warm up the old WinHTTTrack.

Still waiting for more dudes to help with this

hot take! This is going to appear on the wiki.

github.com/antonpantev/unity-movement-ai
it's for 2D movement only, but since RTS works on heightmaps, there shouldn't be any problem translating the coordinates

Made no progress at all, but drew a title screen (concept) and gave the project a name. Might be a shitty name. It's the least bad I could think of.

With limbo AND hiatus pages, that's 4 different pages, so going through all of them might be a bit confusing. How about you make one limbo page and within that add a status tab that says stuff like
etc.

I should pick my game back up to get onto the wiki.

Nah, it sounds pretty cool IMO, at least for the theme you chose.

I too plan to add pic related set to Beelzebox

forgive the chingchong but I thought this was useful for any 3D friends

i've been using the images as background this entire time
i feel stupid

player healthbars work
haven't bothered to make those for the actual units though, ideally i'll just do a minecraft like thing where the texture lerps between good and bad

amazing

do it for a round faced cute girl. Man faces are sharp, so it looks easy to do, but what about round girl faces?

No
Do it for Kirby

Damn, this looks excellent. It's pretty fucking close to the style I've been trying and failing to emulate and at about a quarter of the polygons to boot. I'll have to try this out when I finish work.

github.com/coelckers/gzdoom
Here you go, have fun.

I don't, mostly because it's kind of an uncharted ground. Once the engine went GPL you'd think people would be all over it.
At the moment, all I can really say is to edit the iwadinfo in the engine to recognize your game, everything else I'm still discovering down the line myself.

Oh, this is the sketchy beat'em'up?
I'd been following for a while, always excited to see more. Keep it up, I like it.

Upload possibly related too. Someone else posted it a few threads back.

Great stuff. I don't really plan on adding anything directly inspired by the Bible myself (might be a good idea, but I haven't really read the book). Mainly thinking of Dante's Inferno and various historical time periods as inspirations or level theming.

Thank you. It's always nice to be remembered.

That looks great for small non talking characters but it's too low poly and has no good topology.

blatant reminder for anyone who hasn't joined in on the fun.

8agdg.wikidot.com

wikipedos needed

Whats the status on that horror game made by a group of anons?

Adding my game now~

Been lurking these threads for quite a bit. Making this small game for Android, that hopefully will get some play. I can link it here when it's done, so you guys can comment on it!

Gravity is always in effect. You die if you touch the floor. You grab as many burgers as you can. When you level up, it gets harder. You press the right or the left half to add force in that direction.

No sound at the moment.

Should have read your post before watching the webm.

I know normalfags and the casual game audience is brainless so if your game was to be successful, it would have to have good advertisement or be lucky. There's a trillion brainless games like this.

It is nearly impossible to get any sort of attention on mobile stores, so good luck.

The aesthetic is all over the place, it desperately needs some kind of theme.

...

I'm mostly doing it for fun. I don't expect to make any money. Just putting in a portfolio. It's fun, though.

Programmer for the horror project here. It's dead. No, nobody got angry and quit in a blaze of glory. Creative differences didn't push the team apart or anything. It just sorta stopped. The reasons aren't really that interesting so unless you really want to hear them I won't go into it.

Pistol shots and its effects on walls.

What about my laziness?

The current Cuphead game "journalist" thing inspired me
I will add a difficulty "Gamejourno" its like the sandbox mode, but it also triggers an event that says "yay you won congratulations!" every fucking turn

I was reading through some dev explaining how that is perfectly normal when creating tutorials and all devs face it. It is true, I've had my own retards to deal with and continuously ask for even more retarded tutorials (a publisher in this case, not a journo).

But its also true these people are hugely dense motherfuckers and 10% of the murican population has less than 80 points of IQ, and you should consider these tutorials more of an handicap accessibility feature (like settings for colorblind and support to special controllers) than a normal one.

It's true, watching that guy play was just like watching someone test a game I had made before. Except those people I chose to test because they had almost no experience. It wasn't much different from him to be honest.

If you want to learn how to do a tutorial right, look no further than Splatoon.
As an experiment I had someone extremely inexperienced with games play a bunch of them, and the title that clicked with him the best was Splatoon.

99% of games say they're "made so that anyone can enjoy them", only 1% actually achieve this, because the level of retardation you have to deal with for anyone to be able to enjoy a game is borderline insanity.

Nintendo is better than most anybody when it comes to teaching people how to play games.

Personally I'd axe that whole-room lighting effect because it creates a very undesirable "strobe light' effect when you rapid fire the gun I mean it's your game but I'm just offering constructive criticism

You'd be shocked by the number of Nintendo games I had him try that didn't go well for him at all.
Nintendo goes further than many, but most of their games too fail to be something that even a retarded game journalist could pick up.

Splatoon is different- study how it differs from other Nintendo titles.

if you're an enginedev I'll have to forgive you

I'm planning on making the muzzle flash better, somewhat like doom, almost invisible but it adds to the feeling. Thanks for the heads up anyway.

Sorry user, I'm using Gamemaker.

Still shilling for demo day, no one up for it?

Add it

Sounds like loss of interest.

Wana work for me for free?

just kidding, but i am interested on the failure of your project that i may avoid similar situation

I'll take it for now, I suppose.

Inventory system is now in and works, has its first inventory item. Pills are gonna be the generic "small healing" pickup, might go with the standard first-aid kit for big health pickups.
Unfortunately, there's a big ol' caveat in that you can use items even without needing them and they still get taken from the inventory. For example, even if your health is full, Using the item still takes away from the inventory.
Not sure how to fix that, yet.

Turn your

I suppose before I include more details i should provide proof that I am who I say I am. This arena should look familiar to anyone who was following the project.

Yeah, basically.

I joined the project because I wanted the experience working with a team. I wasn't getting paid so I needed to rely on my own discipline and motivation to keep working on it. The most glaring problem with that is I don't play or enjoy horror games, so what the game wold be in the end wouldn't be something that I could really get much enjoyment out of and as I'm working on it I can't pull from my own experience. I liked programming the mechanics because I enjoy programming in general but I knew I wouldn't like the game. It was my fault joining a project I should have known I wouldn't like when it was done, and also my fault for not telling anyone about my gross inexperience with horror games.

Concept artist had some issues in his personal life. I hope he's OK.

Our leader, the guy who organized the whole thing, was a glorified Idea guy, and near the end he was running out of ideas. That didn't help me when I needed some clear direction as to where to take things. He, made it clear he was also ok with us not updating, which I think led to us eventually not working on the project. Near the end he talked about learning how to program. RE7 was kind of a nail in the coffin for him and where I think the cracks started to show. Much of what we were talking about doing showed up in RE7. He wasn't angry, just kinda bummed out.

The modeler just stopped updating like the rest of us. I can't say why other than loss of interest.

You should make it so the less HP you have the more likely you'll get hallucinations or something. Shadows moving at the sides of your vision that disappear when you look at them, incoherent whispering, fake enemy sounds coming towards you. etc

Funny thing, I thought the opposite, I think when the character too many pills he could begin hallucinating forcing the player to only use the item in careful intervals.

take too many*

Hi Hello I'm the guy behind GMOTA, this hexen like mod you posted here and on your wiki, it is definitely NOT on hiatus.I just don't post here on Holla Forums all that often, and I apologize for that. Have some skeletons to make up for it. Along with an upgrade item for your subweapon pouch

and have the edgelord too

...

Would Ron Gilbert being perpetually salty about Pirates ripping off Monkey Island be just bed-shitting like the case around Nier?

Looks sick.

I laughed, you've done some good enemy design.

sry to hear your project fell apart, you start up someting new?

You better hope (((Zenimax))) doesn't sue you.

I'm doing my damnedest to make this a good update, and to unfuck my project.

Would you believe I convinced myself I wasn't good at monster design until someone told me to look at some of the other monsters I've made in the past? Now I'm having fun with it.
Probably too much fun. Have some more webms of the two special attacks for the arm cannon, the pavise guard, and the bonelord.

Unless they start suing every other doom modder, I don't have anything to worry about, because GMOTA is just a Doom mod, something that's 100% non-profit. Though I do plan on taking the characters I've made here with me in future projects.

Am I the only one who devs with a game running in the background? I found having something where there's shit going on even if you're idle makes it easier to focus. Like in Gmod I would just afk in an apt watching other players do shit or in spellforce I would afk grind and watch how my base handles onslaughts of enemies.

Done, thanks.

That'd be interesting. The pills are specifically unmarked and unlabelled, so who knows what they actually are? Maybe they've got some side effects.
Though I'd done that just so I don't have to explain how a kid isn't ODing by guzzling down pills to regain health. They're spooky ghost magic pills!

You're only a useless faggot if you do nothing. Even just writing down your ideas and figuring out how you'd implement them is progress. Maek game.

I fixed a bug with her eyeball angles. The roll was at a constant 0 which made her look retarded. And the pitch was never the same for both eyeballs.

This looks fucking metal

well, at least post screenshots or write a summary of what happened on the wiki please

woo, thanks my man. Post from time to time. I'll add the ZDoom forum link to the project page.

NOW I remember the project. Sorry to see it ended with a whimper. I'll add it to the wiki asap. You deserve to be remembered, at leats.

Is Demo Day happening or what?

Im all for it, we just need a date and everyone to be in on it

TETRA MAN COME HOME

Is there a reason people can't just show a very basic method involving files and folders, and doesn't require some special software? Like just show me how to make everything find each other and what is required to get it to work without installing some stupid bullshit and going through Trusted™ Windows© Installer Wizards®.

TFW no mentally challanged vampire loli waifu

Learn how to link libraries. Then it will never matter what IDE they're using.

Using msys2 mingw you can compile via gcc and keep it simple on windows. That and a text editor is all I use.

That's exactly the kind of stuff I'm trying to learn.

You can follow handmade hero without using an IDE, he only uses it for debugging.

You can even follow it on other platforms by looking up the equivalent of each system call, they all map across surprisingly well.

Anyone got any hints on how to get past the graphics wall?
Every time I try to learn how to actually display something on a screen I start looking into graphics libraries/game engines/etc but I slam into a wall of "memorise these 50 confusingly named functions, and then write 300 lines of abstract boilerplate" or "download, and install this gigantic program, configure these 70 other toolchain programs, learn these seven scripting languages, and then you too can have an entire fps engine embedded in your 2d platformer" and I make it about an hour in before all desire to continue is drained out of me and everything falls apart into gibberish.

I just want something as easy and self-contained as writing non-graphical code, where I have my entire project in one folder, and my "toolchain" is gcc, make, and a text editor. I really feel like this is an option, but everything I can find is just so damn complicated, everything is made to be as "easy" as possible, which only seems to make it needlessly complicated and confusing.

Holy fuck, I've gotta try that. I'm more motivated to git gud at drawing now, too.

I'm no expert, but wouldn't it still work for Mega Man Legends style characters, where the mouth is just part of the texture, which changes as they speak?

would be useful to look at even when drawing 2d

My own skills on that are fairly limited so don't take what I say to heart, but I'd say just keep trying.

I had a similar problem some time ago and I was stuck for a stupidly long time because graphics related things were so confusing. But after I finally got all the GL shit working with a reasonable understanding of what it all does, it hasn't been a big problem since. I can just tweak it as I need and build my own abstraction layers if I want to make it easier for me.

For example I made functions like this for myself;
fillSquare(x, y, width, height, r, g, b)
Which just draws a square on the current frame. I use them for prototyping since it's easier to just slap that somewhere than create buffers and matrices and shaders and shit and manage where they go.

Someone explain, in very simple english, how I can use quadtrees to see which objects are closest? Distance squared (or cubes) has been very reliable, and I just don't understand how to prune objects. I get how the tree works, but like, consider:

expand from the player outwards, if you find an object stop

attacking works, both range and melee
at this point i think i just have to add unit groups, and then start transitioning the project to multiplayer

I asked about spatial pruning, not pathfinding

what's the difference

Is this an ironic question?

This isnt a pathfinding problem. Again, I understand what a quadtree does, I just dont know how to solve problens with it

quadtree is a tree. learn how to handle and traverse trees.

You have a quadtree that contains all objects. You know your position (128,128). The root node contains 4 subnodes of size 64*64. If a node has more than X objects, it splits into more nodes of 32*32, then 16*16, etc.
[128,128] is in the 4th node, so you enter that node. This node either contains a list of objects, or 4 nodes. Then you either check which object is closest like you'd do without a quadtree, or you see in which subnode [128,128] lies, repeat until you get a list of objects. You always start at root and drill down, there's no going up from the player.

Check all points that are in some properly sized square around the point you're querying. Properly sized is the tricky part - should be large enough that the real answer is actually in the square but small enough that you're not querying too much. (This would be a lot more natural if you used a circle-based tree.) One way is to find a square centered on the query point that only has one other point in it (by taking a square of arbitrary length, selecting a random point inside that square, constructing a new square with that point on the edge, and repeating until there's only one point in the square) then take double the size of that square. Then you just check the distance of all objects in that square.
I'm only paraphrasing a stack overflow answer here that I found in 5 seconds of searching for "quadtree structure closest", so please try to use that resource in the future.

What about the in case of the picture I posted? The red and blue squares are in separate buckets, so I'd have to basically check everything, no?

That doesn't involve quadtree pruning, though?

I've been posting about this problem here in some form or another for months, if not years, and I'm still no closer to figuring it out. 5 minutes on SO isn't going to magically solve it for me

You query the quadtree every time you ask for those squares. Start from the root, and recursively check each child whether its area intersects the square you're querying. If it does, go to the children of that child, if it doesn't, discard it. Record all leaves you come across, you then only need to check against all objects in that leaf to see if they're in the square or not. That prunes a lot of area.

Now that I think about it, that routine can be optimized a lot, but it shouldn't be extremely terrible under normal conditions. You can definitely stop as soon as you find any point in the square, instead of trying to list them all, that's the important speedup.
What kind of checks do you commonly use your quadtree for? Maybe you might want to use a data structure that's more optimized for the tasks you need.

You search for an area which is smaller than your smallest node, so you do 4 point queries and check the smallest nodes those wind up in. In this quick edit (I got lazy drawing more squares, hence the large areas in some quadrants) the gray area is the query, and the checkboxed nodes are the ones that you wind up checking collisions in. This means you check about 25% of the objects in this case, 6.25% in an optimal case with this particular depth of quadtree (where all 4 queries are in the same quadrant).
I've also heard people say something about making buckets larger so they overlap, but have no experience with that.

Fug, forgot image.

Alright added myself. I hope no one minds that I didn't put myself at the bottom of the list because of an inflated perception of self worth.

Also check out this song:
clyp.it/uqivyfpr?token=8013dbc6aa0a79034b126df741387fc1

Woah fuck this looks awesome

I wanna fuck that thing.

Thanks I have a bunch. I haven't been posting for a while, but I've still been working on various bits and bobs.

Here's an unclipped gif of the anim. weirdo.

You should post more often. Still contributing to patreon.com/Labyrinth

You're pretty good.

Christ almighty your presentation is slick, tell us more about your game.

You're my hero

How long's that been the case? The blobs bothered me before, so I'd be surprised if it's been in for a long time but I didn't notice until now.

Yes, sorry about that. I have a post planned, which is just a large excuse, but I was very sick for about 5 months. Probably due to the internalized stress of trying to be a pennyless whodev. I really didn't have the hours in the day between being bedridden and trying to get as much dev done while not being bedridden to post. It feels real bad letting people down like that, but I'm glad some people either forgot they're still giving me money or stuck through my unexplained absence.

ty

Here's this little pre-written blurb:
Aethyr is an Open World 2D Action Adventure game set in a world ruined by an arcane apocalypse. As humanity faces it’s final days the player seeks to uncover the mysteries of the past, delving into forlorn temples in monument to eldritch gods. Aethyr features 21 skills such as Axes, Spears, Elementalism, Auramancy, and Lore to fit your personal play style, randomly generated items, and dungeons full of secrets to discover. Aethyr’s combat features no “Real-Time RNG”, complex position based combat, stealth, and a host of unique and interesting enemies to face.

I basically wanted to take a SNES Action-RPG like Secret of Mana, make the combat all about positioning and fast decision making, then jam in Morrowind's skill system and Diablo's loot generations, and finally throw it all in a fantasy version of STALKER's world and see how it pans out.

It's called a deadly concoction of stubbornness and autism.

They still are blobs, but those lights are from sunlight peaking through the "foliage" so they have a leafy texture to them. I plan on adding dithering to the lights to make them less blobby, but that'll come later down the line.

No hard feelings. I saw your name on my patreon invoices and never regretted it - especially that it's a rather meager amount. sorry for that

Was the illness caused by your move to Prague? Were you forced to do it or was it a decision based on some rationale?
Are you better now?
wanna visit Poland perhaps?

Well if you're gonna reach for the stars, might as well reach as far as possible, shine on you crazy bastard, I'm rooting for you.

unit groups work, shift+1 to 9
i'm not gonna bother showing which unit belongs to which group though, especially since a single unit can belong to multiple groups
example:
all units bound to 1
cavalry to the left bound to 2
cavalry to the right bound to 3
all units are in snibbeti snab formation, but individually the cavalry is in a line formation
so when i'm moving 1 they're moving in snibbeti snab, but then when i tell the cavalry to charge they do it in a line
i still haven't actually implemented trample damage
so thats next, and then multiplayer conversion

No need to apologize for giving me money, haha. Being sick was nothing to do with Prague, it was just a long string of illnesses. I got food poisoning which I believe lowered my immune system to the point where I developed bronchitis, but I also think that the food poisoning caused some sort of bacterial infection in my blood, so for a long time I was getting fevers every day and inescapable lethargy. Later I got what I think was mononucleosis, at which point I went to the doctor and he didn't care so I just weathered it out for the course of a couple months. I'm feeling a lot better now, so I've been able to dev a lot more. And once I have the money to leave my apartment, I'd love to visit Poland.

Thanks a bunch, I'll try not to disappoint you. B)

Finally, that's the last angle done, all that's left is pain frames.
I feel kinda foolish for going into such detail for an enemy that a player might not even potentially see in a playthrough, but fuck it.

...

I'm weird like that.
These guys replace wolfenstein SS, which are used very sparingly in most Doom maps. Though I could always make a monster summon these guys at some point, if I feel they're underused. I also have plans on making shitty bootleg skeleton mimics of the other heroes too.

and then for Knightmare mode actual dangerous Shadow versions of the heroes will have a chance to replace basic monsters and those guys will push your shit in

If it's uncommon and doesn't have similar care put into it as other enemies, then it may just feel like forgettable padding. A well made uncommon enemy will stand out among the rest and adds important variation to the game and how it feels to play through.

it's not about the rarity. It's that Wolfensein levels are secrets/rarely opened in Doom.

They do that to make the setup retard-proof. Also, compilers are platform-specific: The C run-time has to start up very differently on windows than it does on Linux. If you want a cancer-free compiler for windows get: bellard.org/tcc/

...

finally got around to replacing the broken physics system and just animating the breasts and hair manually. It's coming along.

...

I have a list of dream features I would like my dream game to have and this is one of them

I like what you're doing user but that looks as unnatural as a peanut butter and ketchup sandwich. I'm all for breast physics but at least make them make sense. For instance have them drag and bounce when you rotate the camera and then jiggle a bit when you stop. What you have now just looks absolutely ghastly. Like literally ghastly. Like her tits are possessed.

if some nigger doesn't want to play your game then let them? adding in content for post game and the like rather than hand holding the player and dragging their nose threw all the stuff you made. someone can always decide your game is to long and quit someone might just want to finish your game and be done. but if someone likes your game they will explore it to find more content.

Another batch added. Done for today.

Please add info if you have any

Okay I'm into it.

The animation is off for a few reasons. The tiddies bounce too late after the movements and don't really move in a way that makes sense with the movement itself.

Secondary animation is all about delays. The parent bone moves and the inertia of the child bones cause them to delay behind the parent. See vid related.

I want to do the Revelations: Persona thing where there's half a game's worth of content locked behind an obscure, difficult and specific gameplay path. That's way harder than the normal game path but has the better story and gameplay of the two paths.

That's what I'm advocating for. Games that are a solid length for normalfag bitch babies who'll complain anything longer than 20 hours is too lon. But players who replay the game and read between the lines will be treated to a plethora of content that's unlocked through their own competency.

Oh, wrong video evidently

Everyone needs to be developing their games for GNU/Linux and not Windows. Especially now that the tech companies are censoring everything. Game will attract normalfags and normalfags will attract developers.

Doesn't Unity and UE4 build trivially to Linux?

This video is really good if you like pixelshit. Goes over a lot of animation techniques cleanly and concisely, and you get to see the workflow.

What makes you think freetards don't like censorship too? I mean look at Mozilla.

Magnificent.

I skimmed through the video. The stuff he seems to go over is good advice, but I really dislike the sprites in Duelyst. I feel like they don't read as well as they should, and they look extremely out of place in the actual game.
I'd post a tutorial in response, but I don't know any because I'm self taught.

What other option is there? Windows 10?

Custom bitmap fonts = nice

I think duelyst has really good pixels, but they look very garbage in game because they're loaded with bloom.

We've been over this.

The greatest tragedy of our time.

Is there something fundamentally wrong with 640x480?

It's 4:3

You tell me

Take it up with God, not me.

I think they're too busy and it's already hard to make sprites without an outline stand out. The bloom makes it a lot worse.

At least at 480p is enough to emulate all the games that are good.

...

I don't see how you're going to play a PS2 game on an operating system with only 16 colours.

Dithering.

As far as I'm aware, there's no function to clear all data from an array in GameMaker, similar to ds_list_clear, nor is there a way to destroy it, like ds_list_destroy. I need all the spaces which were previously filled to be undefined, not just zeroes. I managed to get what I want by simply reinitializing the variable to be another variable type and then immediately swapping it back to an array, guessing that GameMaker might handle the rest and delete all of the values therein, but I don't actually know what goes on in the back end. Are there any unforeseen consequences that this would cause? Is this how memory leaks happen? Is there a more correct way to do this?

How else will we avoid the CIA niggers

Use a better engine

Let's say you have an array of 500 elements. Each element is a reference to another object. You've basically just shoved those 500 references aside, and made a new bucket for a bunch more references to fill up the array.

This means the original array is still intact, somewhere in memory, pointing to the objects. Depending on how GM's GC works, it might ignore collecting any referenced object, or it might collect orphaned references that are no longer accessible from the main code. You should always assume it won't clean up after you, which means you should clear the array out manually.

Why not just do a for loop, and for each element in the array, kill the reference yourself?

Fucking love how arbitrary computers are some times. Obviously I fucked something somewhere, but man i sure do miss those feelings of what in the fuck is this thing even doing right now.
FML, I'm going to be another fail-dev turned artist.

Mark of a good programmer

I intend to, but I'm sticking with this project for now to prove to myself that I have the tenacity to complete a game at all.

Well shit.
My code is dependent upon the functions array_height_2d and array_length_2d, which return how many elements there are in a 2D array vertically or horizontally. If I simply use a for loop to make all of the elements 0, those functions will still detect the 0s the next time the code is used. Therefore, I need a way to make all of the previously defined elements undefined once again. Or am I not understanding what you mean by "kill the reference?"

What exactly are you doing with it? Obviously you can't resize an array dynamically because of how memory is allocated, but there are some tricks to get around that. For example, you could just have a 1D array, and keep track of the "width" and "height" yourself. Under the hood, most languages treat 2D arrays this way because it's faster, or at worst, as an array of arrays; 2D is just a convention to make it easier for you to work with grid-like data

It should be fine, GM will probably take care of it

I don't know GM but it sounds like your array is static, meaning you can't shrink it. If it is dynamic there must be a function to actually shrink it.
Setting anything to undefined should never be the solution.

If nothing else works you will have to keep track of the size yourself.

I can understand it.
I myself have an autistic devotion to 320x200.

Basically, it's poker. The game detects what possible card combinations the player currently has, and stores the possible combinations, as well as information about them, within the array.
Fuck me, I forgot that ds_grids (basically 2d arrays with a unique set of functions for handling them) exist because they wouldn't work for the last thing I wanted to use them for. I think that'll be the proper solution here, which'll make things a ton easier.

Would this accomplish what you need?

That actually sounds pretty good.

If I'm just learning Unity, using mere primitives in 3D just to do basic gameplay mechanics. How well would it run on a 2.0ghz single-core CPU? Asking because I often have to spend a lot of time away from my desktop and don't currently have enough money to get a better laptop.

Best part? It was nothing in the script, whatever problem I was having was somewhere between my Blender model and the Unity import. So either I fucked that up somehow, or Unity did.
I'm guessing me but fuck you Unity anyway, it is easier to blame the other

Yes, that's what I'm doing right now, but I didn't realize it was an intended method. I skimmed through that page, but somehow missed those lines. In my case, it's:
That's great, since now I know that it'll still work properly even if the ds_grids don't do what I want them to. Thanks.

Passing assets (especially 3d) around between different toolsets and frameworks is actually a really common problem.

Importing Blender objects into Unity is a pain. Was following a good tutorial series until a long streak of model errors that I couldn't fix no matter what I tried to do just made me say "fuck it" and abandon that series.

Do you guys like John Woo?

Yeah, shit sucks. I've been meaning to get a Blender -> Unity workflow hammered out and documented well enough to publish a proper tutorial but in all honesty, it is a huge pain. In part, the issues make sense, like the +Z is up thing, due to the dual history of vidya/movie production from one angle and AutoCAD coming from another. But on the other hand, FFS, decide on some standards for the industry so we can all move on with our professional lives (queue appropriate XKCD about standards comic from a plebbitor).
All the info I can find on the web is out of date and apparently this shit changes significantly enough bit by bit to break almost anything in a tutorial that is older than a year or two. Most people never even bother to mention what version of what they are using, either (how fucking hard is it to say something like "Using Blender 2.78 and Unity 5"?!)

…and this is why I want to write that workflow tutorial finished. This has been preddy gud so far: youtube.com/watch?v=ocLZNRasU64&list=PLFt_AvWsXl0djuNM22htmz3BUtHHtOh7v

Did you ever get around to making her face cuter? I remember the nose was too low and too snubby. It could be that you've become used to her face. I've noticed that whenever I model something, the more I look at it, the better it looks even if it isn't finished. For animating, it's the opposite for me. The more I look at an animation, the more off it looks.

That FOV is kind of small.

look up raylib

This triggers my autism as bad, if not worse than mixels and rixels combined

As opposed to?

underage b&

Holy shit, I learned something today.
en.wikipedia.org/wiki/Mode_13h
en.wikipedia.org/wiki/Pixel_aspect_ratio

But why does that matter? An image is an image

I just saw this a couple days back, and this kinda relates. Should explain it.

mostly because the lowest tier of source ports don't realize this and make all the sprites from old games look like manlets.

en.wikipedia.org/wiki/Aspect_ratio_(image)#Distinctions
SAR × PAR = DAR.
PAR = DAR/SAR.

So it's basically a bunch of autism that doesn't really matter for modern gamedev? I mean if you were emulating, it makes sense, but most PC hardware uses square pixels and

Still interesting topics

Literally no different than using shitty crt filters to be "retro".

I think I found the problem. That's not doom, I'm just using the textures while I experiment with godot.

Well here's the gist. I don't know how much you know so if I'm stating the obvious it's just because I'm touching as much as I can.

Static libraries are compiled straight into the code. Dynamic libraries are loaded when the program starts. You probably don't have to worry too much about which one you use when you're starting out because you can change them easily.

When you want to use a library you have to download the binaries or compile it (if it's open source). The openGL/direct libraries might possibly be on your computer already, although I forget where they are on windows and I'm not 100% sure. Linking is compiler specific, so you just have to google "how do I link libraries with this compiler".

When the libraries linked you have header files. The header files are basically what tell your code what functions exist, while the libraries have definitions for these functions. You probably already know how to include headers.

From what I understand if you link the libraries and include the right headers the IDE shouldn't matter beyond that. You could use straight GCC if you wanted to.

How to you guys do localization? im putting all text into files in these folders, so to add a new language you just create a new folder, fill it with the translated text files (they must have the same filenames) and add the language entry to the options.txt

The player choose the language option in the options panel and the game changes the folder it takes the text files from

I know that feel. 7 years ago when I wanted to start programming, I picked up C++. On my first project, I quit because libraries were giving me a fuck. Didn't go back to it for another 2 years.

Post inspirational videos.

[audio only]

Just going with some /loomis/ at the time (that board is comfy as hell, it needs more love)

Bumpan with the /loomis/ resource hub for those wanting to draw some lewds
8ch.net/loomis/hub.html

I added the programming page. Feel free to screech at me for getting something wrong.

8agdg.wikidot.com/programming

I'm not gonna edit since I don't want to join the botnet, but that's not correct. C# and Java run in a VM basically.

I should have expected this.

What are you inputting that sound data directly into your source code? How does music get compiled?

Fuck wat?

Woops, reading more I see you make the distinction later on. Still, it's not exactly accurate to group them with C/C++.

I'm pretty sure it has something to do with how GameMaker converts sound files. I didn't specify how it should do that, so it's probably doing some really dumb default. That said, I suppose it's not literally the compilation, but it's something that happens as a part of the simultaneous processes that GameMaker undertakes when you tell it to compile. It always takes longer the first time you compile after adding sound files for this reason, and I suspect that when you use the "create application" option, it re-converts everything again. Also, the application size is now around 44 MB, as opposed to the previous 4 MB, when the sound files added were only about 12 MB total.

So apparently there is no native way to make a shallow copy of a list in gdscript. You actually have to manually copy the variables through a for loop. After figuring that out my Health/Stamina/Energy class is now working.

Please tell me some kind of Zelda H-parody.

The character controller is very similar to the N64 Zeldas, yeah. Semen meter is just japes, it's actually an Energy Meter which Stamina and Health draw from to recharge over time (health only recharges to half, or a quarter if Exhausted). Once your Energy drops below 10% your character will become Exhausted and need to go sleep in town or something.

You can copy arrays like this, it's kind of a hack.

Oh, so you're adding the original to a null array? That's clever. I wonder if it works with dictionaries too. Is that Godot v3 btw? Looks quite different to what I'm using.

Yes it is 3.0. The alpha1 build is here: godotengine.org/article/dev-snapshot-godot-3-0-alpha-1

Yes.

MAGA.

But what if you're so autistic that you insist on supporting computers from 20 years ago?

godotengine.org/article/we-should-all-use-gltf-20-export-3d-assets-game-engines

They compile to a VM, and needs to be done ahead of time. Makes more sense than sticking them with eg Lua

This is a fantastic (but somewhat hard to skim) page about FOV in RL games.

adammil.net/blog/v125_Roguelike_Vision_Algorithms.html#colored

One thing I was considering was the way colored lighting should work. I had a tile-based colored floodfill that looked kind of okay before, but I wonder how good their recommended method is? He says that for each light source, add up the colors on the light, then scale down/average them out.

Unreal so desperately needs a "Recompile all Blueprints" button

Yeah, I knew there were errors with it, I was actually scrambling to get some kind of animation out this morning before I went to work just to get some kind of feedback.

So here's a fixed version, though rather than try and fix the animation and timing on the breasts themselves, instead I just exaggerated the animation on the body to match the breast movement.

I'm not sure, I may have adjusted the face but as it is I think it looks good.

Note that for the face I have a small light source directly shining onto her, tracking with her face, so that the lighting is even. Without it, the result is usually the main light in the scene giving her face a dramatic high contrast lighting that just makes her look creepy.

I used four sets of lights and tried method 3 and 4, using PS. The left results in a more washed-out color (average), and the right results in a more luminous value (scaling down). The gradient is interpolation between the two.

It might be interesting to calculate it twice and then interpolate it to get a color among the gradient, whether the gradient depends on distance or what, I don't know.

I'd fuck her even if she still has that ayy lmao head shape.

God damn it. For the rest of you, how common is it that you can't make progress on one system of your game until you've built up a little more of another system in a way that you couldn't possibly have anticipated, and find yourself having to slowly piece every system together based on the current iteration of every other system, often forgetting things about each system along the way? It feels like there must be a better way to do this.

Every project for the last 6 years

I dont know what to do with the skeletons arms when its running.
Also I imported firelink shrine because I was curious to see what my shaders would look like on complex geometry. I dont like it

Speaking of Dark Souls and shaders, how do they achieve such a grayed, washed-out look? Specifically, if you look at geometry from a far distance, it tends to have its color clamped to fit a particular palette. What is this technique?

I often have a situation where making progress means tearing makeshift solutions apart to the point of making the game unplayable until the new system is complete. This leads to me stalling that for as long as possible because im a lazy fuck.

fog?

Maybe?

Like in pic related, cliffs are basically two-toned blotches

I think that's a combination of fog, tonemapping, depth of field blur, and darkness.

I was wondering in the last thread whether it's possible to create a shader that actually looked like a CRT. All the ones I ever saw are so shit it's nauseating.

are there ZDoom-like Build ports?

thx mang.

can you post the cool lightgem .webm from last thread and a screenshot? I wanna add your game to the wiki.

Only eDuke32, to my knowledge. AMT TC is a fantastic demonstration of what you can do with the engine, but working in CON is a nightmare.

What is so bad about it?

...

Thanks a bunch! Really impressively done

Actually, I'm jelous your stuff already surpasses my earlier project in the L/S stealth department

What's your approach? I'll need to implement the same thing in Unreal at some point. I already have an idea for an algorithm, but not having spent a huge amount of time thinking about it, it'll probably have its problems. Does your algorithm play nice with colored and moving lights?

while you are at it, add mine as dead.

Add it yourselves you lazy fucks

Admitting defeat is hard enough already.

Currently it only supports omni-directional lights, spotlights will be more complex. Moving lights should be fine, not sure about colors, but brightness (energy in Godot) and attenuation are both taken into account. Each light is a scene set up as pic related, then just instanced into the main scene. A great feature of Godot is that I can change the energy, range, attenuation, and other properties inside the editor without affecting other instances. Here's the script for the light with some quick comments:
extends OmniLightvar player# player has a function called light(), which affects an array of lightsfunc _ready(): # make detection area the size of the range var area_s = SphereShape.new() area_s.radius = self.omni_range get_node("Area/CollisionShape").shape = area_s set_fixed_process(false)# player enters detection areafunc _on_Area_body_entered( body ): if body.is_in_group("player"): player = body get_node("RayCast").enabled = true set_fixed_process(true)# player exits detection areafunc _on_Area_body_exited( body ): if body.is_in_group("player"): player.light_remove(self.get_instance_ID()) get_node("RayCast").enabled = false set_fixed_process(false)func _fixed_process(dt): # cast the ray from the light origin to the player get_node("RayCast").cast_to = player.translation + Vector3(0,1,0) - self.translation if get_node("RayCast").is_colliding(): if get_node("RayCast").get_collider().is_in_group("player"): # the light can "see" the player, calculate appropriate light level and update self in the player's light list var amount = omni_range - self.translation.distance_to(player.translation+Vector3(0,1,0)) if amount > 0: amount /= omni_range amount = pow( amount, omni_attenuation ) amount *= light_energy else: amount = 0 player.light(self.get_instance_ID(), amount) else: # there's an obstruction b/t the player and the light, remove self from the player's light list player.light(self.get_instance_ID(), 0)

And here's the relevant parts of the player script:
# there is a dictionary called lights in the player class# lights are stored by their unique instance ID, and updated when neededfunc light(id, amount): # change lights[id] = amount update_light_level()func light_remove(id): lights.erase(id) update_light_level()func update_light_level(): light_level = 0 for l in lights: light_level += lights[l] # this is probably physically inaccurate # hud updating below

He asked me.

RIP, friend. I hate the fact that we have a project graveyard at all. Still, don't give up hope and keep like making games. We believe in you.

Reposting my shilling & rather cryptic advice for the stealth user

when the character steps within the light's radius cast rays from the light to the sensors to check whether the light actually reaches the character.

He did it a bit differently but the overall idea of raycasting and calculating atenuation at the given distance remains.

Yeah, maybe I should encourage people to add their stuff more. We need contributors. But that botnet excuse is compelling.

It might feel cathartic, you know. A chance to begin anew.

That's exactly the algorithm I thought of and I can tell you why it won't cut it in my case: it won't work with global illumination (indirect lighting).

However, I might be able to work it out using this:
answers.unrealengine.com/questions/2235/accessing-indrect-lighting-with-c.html
So apparently you can access the data for the indirect lighting.

Even If i wanted to keep working on it, my last working backup is a few months old, back when it was going to be a phone game.
You are right i will give it another go, but next time I will roll my own characters however right now i would like to work on something else.

That's exactly what I'm doing with my current pixelshit project.

My last project (which is already on the wiki under Active, funnily enough!) hit a huge setback that I'm not sure how to address yet and just thinking about it makes me depressed.
So I'm doing pixelshit horror in order to keep active and figure out some possible workarounds in the meantime.

It's easy to get into a rut, but you've gotta keep on plowing ahead with SOMETHING.

Would there be a reason to move it to "on hiatus"? You sound like you want to continue working on it. We believe in you.

is it the Zscript problem with random frames?

how about a jam game or a small project for fun? I start small, weekend or week-long projects to reset myself and bring things into perspective.
the wiki is one of those

In my Unity project I set that to zero/black when there is no directional light in the scene (moon/sun). When moon/sun appear, global illumination is added.

I haven't implemented this to the light meter script yet but with some tinkering, I think adding ((r+g+b)/3)*a of the global illumination colour to the lightmeter value should do it.

Nah, it's not on haitus, I'm still working on it. I'm just not really happy with how the code is working and I'm not sure how to fix it.
Not quite, but it's a symptom of a bigger issue. ZScript is causing a memory leak somefuckingwhere, so having the game open longer than half an hour drags everything to a crawl. I can't identify where, and converting the code back keeps the slow leak in. It's driving me mad.

Are mod projects welcome there?

Finally I am getting close to a baking result I want, I just need to find a good default material or modify one whatever so that I can use it for all my other models that needs baking, it kinda sucks though that the Ambient Occlusion is fucked up somehow all it does right now that it makes my model more bright instead of making certain parts more dark despite I used a few color nodes with multiplier effect which did not fixed it at all, welp. Hmm setting up demn baking stuff in Cycles doesn't seem to be any more difficult then in Internal rendering as I thought but it kinda sucks that I need to get rid of other materials else Cycles refuse to bake for some stupid reason.

Once I finally have a decent material then I can start working on making more rocks, ingots and what else is needed for ore stuff. The Internal rendering is driving me crazy with its fucked up baking result because either the lightning is not uniform or there is not enough details baked on.

I would say >>>/vm/ is for that, modding threads get bumplocked on /agdg/ and rightfully so

Seeing as GMOTA is on there, I don't see why they wouldn't be.

But we're not on /agdg/, these threads are named /agdg/ + /vm/ for a reason, and it's our wiki instead of the /agdg/ boardowner's.

I agree.

Please, feel free to add your project

I hope it will end well. Doom loading this long feels surreal.
I expect it you have some form of contact with ZDoom devs?
I remember listening to Terminus' podcast (or is it you?) and he was quite excited about Zscript. Perhaps the enthusiasm was premature. Or perhaps I remember that wrong.

I mean it is called the /agdg/ wiki. But i don't want to be an autist about, maybe make a separate section for mod projects and post on /vm/ so they will get in on the fun as well.

Ah right I forget it was a mod for some *Doom source port. I was confused since the wiki just mentions /agdg/ and not /vm/ at all, there is another Doom based project there too.

Well If I recall it correctly before the new BO took over this /agdg/ board /vm/ project threads were allowed to be posted there since it was done so in the past. In my opinion I find it a bit stupid since Holla Forums /vm/ hardly gets new post, though on the other hand this /agdg/ board doesn't get that many posters either, meh whatever.

Yeah I will do it right now then.

I'm no longer entirely sure whether I like Lua.

Welp I am done adding my 3 projects but how the fug do I unfug the formatting?

see the source for the active projects page and try to recreate how it's written. Don't let there be any gaps between entries, for one.

fixed it, there are still images and names missing though

There, that's all the sprites made, cropped, added, alligned, and renamed. All that's left is the behavior for this guy.
Still so much left to do though

Thanks m8.
I just uploaded it to imgur since I have no idea where the "upload file attachment" is or however it is called in this wiki.

It's at the top

Progress report!
fablesoflaetus.com/report15/

I uploaded the images now and replaced all my imgur link with the attached files. At least this site has a old upload method since the new one is broken for me

the arms should counter the twisting placed upon the body caused by the hips not being under the center of gravity. the extreems of motion lining up with the opposite leg.

I'd have the game list all folders, do some sanity tests on the folder (something like: Does the folder have a text file named 'jobdescriptions') and fill the language combo box with the folder names. Do only once when the game starts, and add a 'refresh button' to the right of the language combo box to reload it

It's very exciting. It's also still in its infancy, so shit's breaking left and right.
Nobody's really figured out the "do this" and "don't do this" things yet, and because it's only only a couple months out none of the stuff has been really hammered down and fixed.
I'm sure it will be in time, and I do have contact with the devs so it'll be narrowed down bit by bit over time, but fuuuuuuuuuuuuuuck is this shit crushing to deal with in the meantime.

if unity, unreal, and godot are allowed, then other types of modding are fine by me

to be fair, godot is like another shitty open source game with good map editor

Rustfag here.

I'm learning C# by writing plugins for Rust before continuing work on my own shit (I knew basically fucking nothing a few months ago)

Right now I'm trying to bring back some fun from the original concept arts, and now I'm working on the "Caretakers" - the huge fucking robots that provide some PvE in an otherwise nearly PvE-free game. They're modular as of now, and each body part is it's own entity which provides me with an opportunity to give them sick-ass randomized loadouts. They'll be largely passive until aggroed and will be roaming around the map performing mundane tasks. I ultimately want to make them seem more like animalistic AI than robots with a purpose.

I decided that I'm gonna give em 16 different weapons, one under each "shoulder" or one instead of the head depending on their type. Basically there's no limit to how varied I can make them so I need some original fucking ideas up in here boys. - So far I've settled on a gatling gun, a long-range flamethrower and a railgun. I'm gonna need 13 more weapon ideas so fire away.

shoulda clarified that it's not 16 weapons on every robot but rather a pool of 16 different weapons for the loadout for a total of 256 different combinations

Is there anyone playing this game nowadays ?

A long-range mortar with explosive shells.

local avoidance is not the same as pathfinding and should be treated as an entirely separate problem

want some shit tier models? are they going to be fully modular? (broken legs or randomized parts pushed together catipeler centipede along with your quadruped spider legs guys)

Finished a dummy, apparently the bullet collisions on enemies are working fine. I think this is enough as a prototype, I'm going to stop programming for now and start the pre-production phase and elaborate the game design.

Too lazy to post a video so here's a screenshot.

I'll consider it.

I wanted to add my game to the wiki but it seems to have it's own retarded syntax that i don't feel like figuring out.

in case someone else wants to do it for me

name:
the Rebellion
RTS game inspired by the rebellion citizen uprising in half life 2

Tools:
sfml

Status:
Inactive

Author:
anonymous

Is this supposed to be on hiatus or discontinued (aka graveyard)?

hiatus i guess i mean i might return to it at some point but i might not

Done after retarding for 10 minutes.

Why would you want to choose just a few files instead of changing all text files at once?

Not bad, mate, not bad at all. It's been fun watching your work develop, I hope you keep this up!

Nice, man, that looks neat.

It's getting pretty dead tbh but I'm riding this train until it is.

That's pretty fucking neat

They're fully modular. As a matter of fact in the video every leg is independent and gets an update every n seconds depending on how many legs there are.

They do a raycast from their default position + momentum of the caretaker to find their footing in the next step.

I'd like to have some fucking models but unfortunately it's a serverside mod for a VAC-protected game so I can't import anything new. The robot in the video is made of existing models, mostly low walls and beams.

I hope you learned your lesson about being an asspie. Our track record of having devs throw hissy-fits is getting it of hand.

Nice stuff

What the fuck did you just fucking say about me, you little bitch? I’ll have you know I graduated top of my class in the Game Design course, and I’ve been involved in numerous fundraising gamejams, and I have over 300 finished flash games

Even when I do throw hissy fits they're generally over pretty soon. I'm not a woman I don't keep grudges.

On an unrelated note - I'm trying to get into sound design. Rate the salvaged sniper sound out of 10 and tell me how to make it better.

soundcloud.com/rottenhuman/salvagedsniper

thanks user

not familiar with it but if you wanted to make a standalone clone or the like i usually lurk here and i lack faith in any of my projects, and modular robots sounds fun just by itself. as long as you don't want to make a fucking robocraft clone

6/10, kinda average. If the gun is supposed to be like a fuck-you power weapon, beef up the shot sound and make it a bit longer. The bolt action sounds kinda dry too.

I mean to lesser or greater degrees we're all autists. But yeah, it's important that even if/when we act like a sperg we let shit go- on both sides. Saying stupid shit and getting bullied for it are prone to happen so in the end you can only learn from your mistakes and not let it drag you down in the future.

It's an EAC and VAC protected game which means that clientside mods will get you banned.

It's supposed to be a classy (as classy as salvaged weapons get) intelligent weapon. It's meant to be a restrained, dry shot. That being said, it is meant to sound powerful. what do you mean about "beefing it up"?

The length of the shot will be extended with the ingame reverb. This is just a bare sound.

Amen

Not him, but I'm pretty sure he means increase the volume of the lower frequencies, or if there isn't much in the way of low frequencies, add some, them pump them up.
Also watch vid related. He goes over gun sounds among other things.

So i wanna start developing a 3D game but I don't wanna make a 3D renderer myself
So is there anything better than EU4 for this
All I really need is the 3D renderer, something like i stick in models and it just renders them for me
Needs to be opensource

EU4 is fine it just kinda feels like overkill.

Thanks dad

Engine Unreal 4 is much more than a renderer, if you really want just that consider barebones C++ libraries or maybe simple library-like game engines like Irrlicht.

Use Irrlicht or Ogre3D with a physics library.
Irrlicht is easier but old and Bullet Physics is the most used physics library.

Basically at the moment it sounds like a loud paper bag explosion. Also for comparison listen to the Halo sniper shot, it sounds nice and satisfying to fire because of the second beat of the shot.

Just adding that you can spend a good year or more trying to get a fully featured 3D character controller to work in Bullet. Its one of the big arguments for comprehensive engines.

I don't need any physics

I don't know why, but knowing that games that look like this have gone on to sell several hundred thousand copies gives me hope.

I guess I should have specified im making a rome clone so i won't need any physics or a character controller or anything like that

Anyone know where I can read on the traditional structure of a packet for a video game's netcode?

Did three animations today, two of them revisits of older ones. Not implementing them before the new animedit is done though.

...

...

Okay, fine. I don't really like recording anims outside of the engine, since the entire timing will be different and it's almost more of a pain in the ass to setup, but here's a preview.

the thing is I don't there are any C++ libraries

animemaru.com/next-installment-of-when-they-cry-series-to-be-drawn-entirely-by-child/

YOU ARE FAKE NEWS
Actually, Ryuukishi07 has said that for the next entry in the WTC series, after experimenting with having professional artists do the work for him for titles such as "Rose Guns Days", he's going to go back to drawing all of the character sprites himself as he did for Higurashi and Umineko.

Would it be possible to merge Victoria 2,EU,and HOI into a mod? for either Vic or EU?

I really want a super autistic grand Strategy game.

Anyone in here learn to code through codeacademy? Can anyone here vouch as to the quality of the teachings?

If I was his publisher I would not fucking allow him to.

Mechanics question for a roguelike or game with a TES-style skill system

Since I want to let you organically grow a character, what if every skill had increased xp gain (or reduced cost, whatever) based on your base stats? Eg, if you had a 16 points in INT, intelligence-based skills might receive a +6% or +12% bonus to experience gain, while strength-based combat skills wouldn't get any additional improvement, making it guide you towards using magic

Isn't that just web development?

why not just learn how to program, by making programs?

it says on the box you can learn python

...

I can't wait to start implementing mechanics and actually make something fun out of this. Captcha unrelated.

what is your question? If is it okay to raise stats based on actions? or if it is okay for stats to not influence each other?
both are pretty subjective questions that can be answered by "depends on the implementation".

Engine?

Just Godot, I'm not autistic enough to make an engine. I've tried, it's a real pain in the ass.

When a recipe produces more then 1 item and it has a probability it won't show as percentage anymore and that triggers my autism.

...

I'll revv up the archiver.

you really have gotten better

Sorry to see your project abandoned. I remember you from the Tic Tac wars thread. I think I even have your youtube channel subscribed

Well, I too kinda feel a different method for adding projects might be needed but what we have now just works, so we'll see.

will add your mod to the wiki soon

I'm using more reference animations, that's at least half of it.

thanks dad

new bread since this one is already on page 14