6935e9 No.14828498
Thread's dead baby, thread's dead edition
Resources
>>>/agdg/
>>>/vm/
>#8/agdg/ via irc.rizon.net
>Dev resources: http://8agdg.wikidot.com/resources
Links
>Wiki: http://8agdg.wikidot.com/
>Beginner's guide: >>>/agdg/29080
>Previous thread: >>14794392
Announcements
>QUARTERLY DEMO DAY SCHEDULED FOR August 8th
>Please contribute to the wiki if you can
c51f91 No.14828544
Why haven't you finished your game yet ? (why your game is shit ?)
bbeb12 No.14828575
>>14828544
Haven't finished my game engine yet
a5404f No.14828584
>>14828544
I'm working in it. Give me time.
9b84f6 No.14828589
I wanna fuck your video game character.
ab19e6 No.14828598
PROGRESS REPORT ANCHOR
What have you done lately, anons?
Any projects you're starting up?
Any new skills you've learned?
Any significant milestones you've made?
Any particular difficulties you're having?
Post a picture of your progress!
This post also provides you with a license to blogpost about your dev struggles! So long as you spoiler that shit my guy
d8e980 No.14828664
>>14828544
My butthole is clenched way too tight from a life of shit-tier anxiety management and I can't sit properly without my ass either tearing or prolapsing a little. Practicing meditation would help but I don't do it enough meditating while laying on your side is much harder than meditating while sitting up and I'll probably need to go on meds soon because this shit has been going on for seven fucking months.
Please, if you ever talk about fixing up your life, do it before you reach this state.
d8e980 No.14828674
>>14828664
No, I am not rustfag or that huenigger with the broken laptop monitor.
d14655 No.14828691
>>14828598
I finished translating my game and now I'm working on polishing the main menu to make the game look good for a demo I'm doing on thursday. After that I'm going back to doing gameplay.
>>14828664
Jesus Christ.
e841b5 No.14828701
>>14828598
>What have you done lately, anons?
Messed up with shaders, I gave up, just going to use Uber Shader I had bought a year ago and give up on selling my terrain system on the asset store
>Any new skills you've learned?
Copying pasting parts of shaders from stack overflow
>Any significant milestones you've made?
I had everything I want but parallax working
>Any particular difficulties you're having?
Going to have to edit Uber Shader to use TextureArray, although It's probably a better idea to give up on the notion of more than one texture per terrain face. I wonder how feasible is to put meshes at almost the same height of another mesh to add those details rather than letting the user paint on the ground. How does decal works anyways?
>Post a picture of your progress!
c51f91 No.14828713
>>14828575
Why are you still here ? Less shitposting and more developing. Take example>>14828691
>>14828584
>give me time
No
>>14828664
Stop putting things in your ass
7260ca No.14828738
>>14828725
Then start putting things in your ass. It should reverse the process or whatever.
a5404f No.14828739
>>14828713
Well too bad. I'm taking all the time I want.
ab19e6 No.14828744
>>14828664
I'm sorry to hear that, anon. Hope you get medical attention for these issues. That sounds like hell from what I've heard so far from people who have similar conditions.
>>14828674
in case you weren't aware it was decided a few threads ago to scrub any and all memory of rustfag/RR from the face of this board and the wiki because every mention of him derails threads without fail. He's been successfully bullied into giving up devving, so lets not give him any reasons to suddenly pop up here and start another holywar
>>14828691
That's excellent! How many languages have you translated it into? What languages in particular? Is that a picture of your game? In case it is, it's looking pretty damn good.
>>14828701
Is that a level editor you're working on there?
>>14828738
By god, you're revolutionizing medicine, anon.
bbeb12 No.14828757
>>14828713
I AM developing, and eventually it will be a game . So what is your point? Talking about programming is shitposting?
f2002d No.14828795
>>14828544
>Why haven't you finished your game yet ?
Because i haven't gotten out of my depression yet
d14655 No.14828801
>>14828744
It's currently in english, swedish, russian and greek. That is indeed my game, thanks anon.
a5404f No.14828810
>>14828795
Maybe you have the order wrong. Wouldn't finishing your game get you out of your depression?
ab19e6 No.14828818
>>14828795
dedicated blogpost anchor is just above, anon. also use spoilers for it.
>>14828801
How'd you get translations for those languages? Do you speak that many?
d14655 No.14828831
>>14828818
No I had friends that could help me out. Some of it is just google translate as well, for now.
ab19e6 No.14828841
>>14828831
that's good! Just make sure to never get your translations done by randoms - those never end well.
86c59d No.14828858
>>14828598
Progress is taking a step back due to personal stuff (about a month of stress followed by a month and a half with no access to my gamedev rig, or a PC that can run much more than a word processor). So gonna get all the modeling I can in this month, then plan as much out on paper (literally) as I can while I can't use my PC. Gonna post progress later, as well, not much to show that I haven't yet.
046b90 No.14828863
>>14828598
Made some decent looking (2D) clouds, based on this shader:
https://www.shadertoy.com/view/XsjSRt
ab19e6 No.14828874
>>14828858
What game are you going to be working on, anon? What genre?
>>14828863
Those clouds are looking wonderful, anon. Maybe slow em down a little, though.
527cfa No.14828922
>>14828664
wat, how?
>meditating
I'd recommend to listening to something that induces asmr, like rain, or the sound of a snow storm is good for me.
It helps to get enter into the meditation state.
After you get in the meditation state those sounds aren't noticeable.
>>14828701
If you want good advice on that type of stuff look at gdc/siggraph presentations; so many gold nuggets there for shaders.
F.e. the witcher 3 devs have some good stuff from witcher for triplanar and texture arrays (presentation name: Landscape creation and rendering in REDengine 3).
>How does decal works anyways?
If I recall correctly it's a g-buffer based technique, but it doesn't play well with shadows and has other various issues.
>wonder how feasible is to put meshes at almost the same height of another mesh to add those details
How about using a heightmap with a tessellation shader.
So you'd be using your base mesh as the low-poly mesh for colliders, base shape, and the gpu side tessellated mesh as details.
You can even paint the heightmap if you wanted.
>>14828598
I decided to put a huge passion project I had going for a few years on hold, and produce a much… MUCH smaller game.
As working, going to uni, and also doing gamedev while not having "a released game" after years being a bit demotivational; so I feel extremely driven to release a game. It'll still be a game I want to play, but much smaller/tighter in scope.
I'll definitely get back to the passion project once I've got a few games under my belt, and have more funding to just dev…. but at this point I need a few years to finish that game.
e841b5 No.14828940
>>14828744
>Is that a level editor you're working on there?
Yes
Takes a lot more effort than initially thought.
It's main purpose was to automatically add props like stones and trees to zones based on configuration, but it seems I should have just made those zones and used Unity's default terrain solution for the terrain, rather than trying to solve both problems Although mine can support caves
ab19e6 No.14828957
>>14828940
I can relate, albeit my project is a tad different. I'm currently dealing with voxel-based terrain with a DC implementation slapped on top and it's a lot of work. Mesh generation can be pretty fucky.
mine also supports caves. heightmaps are for fags
9697c0 No.14829250
Any anons know anything about coding the animation backend? Specifically how the skeleton relates to the mesh? Everyone seems to use an inhouse file format.
From what I understand you can just map verticies to a specific bone. But what do you do about parts on joints that might need to stretch? For example shoulders? I remember in some source mods had weird ass shoulder physics because the shoulder mesh didn't stretch evenly.
bbeb12 No.14829343
>>14829250
>Everyone seems to use an inhouse file format.
Consider developing your own inhouse file format
527cfa No.14829358
>>14829250
>shoulder mesh didn't stretch evenly.
there's many potential causes there: low poly count, bad weight painting, bad topology, retargeted animations to a generic type of rig setup causing inaccuracies, not enough bone contributions to vertex supported in engine for made animation in modeling program, animations not being made for the specific topology so not accounting for that case in weight painting/etc, or the animator wanted to just like make animation.
>what do you do about parts on joints that might need to stretch
usually comes down to how you do the model/rig/animation itself (topo, weight painting, etc).
The logic on the backend of animation is relatively straightforward, but getting everything else right can be complex; or just requiring a trained hand.
e841b5 No.14829383
>>14829250
As far as I understand, the bones are actually just transformation matrices. Each vertice is set to one-to-four of these matrices.
If you don't know what transformation matrices are, just think of a thing that stores position, rotation, scale and some-other-stuff-nobody-uses-for-anything in a convenient-to-multiply format. It's about as much magic as quaternions.
The skeleton is a hierarchical organization of these matrices, meaning that if a parent matrice changes, all children changes together (hence why they need to be convenient to multiply, as far as I know, moving relative to 15 parent objects is simply multiplying a identity transformation matrix to each one of the 15 matrices in order).
A vertice will respect one to four matrices, with a weight associated with each matrice, so it probably calculates where each matrice wants to move it and then interpolates the one to four positions into the final result
1c83bc No.14829616
>>14829250
https://github.com/guillaumeblanc/ozz-animation
also AssImp library has a very simple implementation that uses standard file formats for meshes and animations, and example code to animate some example characters.
Each bone has a set of transformation matrices that describe the keyframes of the animation. You interpolate between the transformation matrix for each bone. Each bone is associated with a vertex list and weight, describing how much it affects each particular vertex.
My understanding is that inside (concave) creases in meshes can look really bad. Also if not enough polys cover an area that undergoes large movements relative to adjacent vertices, it can look "stretchy" and weird.
I'm just starting the road on implementing Ozz with my stuff. I've used the AssImp code though and can vouch for it, although it's messy and not easy to use alone, if you do not want to import all of AssImp into your project (I don't).
f97735 No.14830007
>>14828544
I'm a poorfag who's been saving up for a new HDD to get my desktop running again. I'll have enough money to order it tomorrow and I should be back on by next week after I torrent all the necessary software again.
f0ef21 No.14830084
>>14830007
i hope you are penniless because you are too young to get a formal job, otherwise you really should work on improving your situation first, because not being able to buy an HDD on demand is too poor even for a poorfag.
e1218f No.14830400
>>14828544
Because >>14828575 and also Factorio
ab19e6 No.14830445
>>14830084
I have forty starved ethiopian children writing my data onto clay tablets, fuck getting a job dude
13b091 No.14830475
>>14828598
I started work on the Options Menu and started working with save data. Currently the only thing saved is the last difficulty you set the game at, but I'm going to obviously add more stuff in the future, such as custom key bindings.
35cfd1 No.14830727
I've made some progress with modding F.E.A.R.
Right now I'm figuring out how to do little things in the game's engine. Like the reflections. The game used a ton of really nice looking reflections that felt seamless in levels so I wanted to figure out how to implement them.
Trouble is the documentation that's included with the editor explains none of this. So I had to do some detective work. I dug through the premade maps the devs included with the sdk, which were all multiplayer. And I found one where a dev snuck in a reflection in the map off in a corner. Which meant I hit jackpot.
So I extracted just this room and removed everything through a process of elimination until I figured out the bare minimum of getting the reflection to work. And finally I did. Essentially the texture is part of a special shader that contains a map for the parts that reflect light, which are the parts colored in green. After that you need two separate entities. One that affects just this texture and tells the engine that it functions like a mirror, and another global one that affects all entities of this type. I'd imagine this was done this way incase you want to have multiple entities. After that it just works. It's hard to see in the last screenshot but it is working and can be seen if I move around.
I might make a tutorial in a few days that I'll upload to moddb on how to get reflections working. Mostly for my own benefit so I don't forget, but I'll post it publicly incase some other sperg also wants to map for this game. This MP map also showcased a camera/view monitor like Half-Life 2 so I'll try getting that working as well.
86024b No.14830800
>>14828544
I'm trying to git gud at art first.
>>14828598
I think I'm done with this. What do you think? I made it wallpaper size too.
For some reason ps lowers the image quality in a way that the shadows don't look as good as the original, why is this?
ab19e6 No.14830803
Okay boys I'm on suicide watch. I implemented a system where chunks are loaded in batches and can look at each other's voxel data to avoid having to create seam geometry and now for whatever fucking reason my contouring is detecting a sign change at chunk boundary despite there being none, resulting in a flat wall on one side. The biggest problem is that I honestly have no fucking idea how this could possibly happen. Beats me. It's retrieving an air voxel from somewhere but it's searching for an intersection within a solid fucking space of material. What the fuck.
>>14830475
Looking great, anon! It's a good thing that you're not being ableist and are implementing special needs options for the people who need it most.
>>14830727
Anon, your autism is the best kind of autism and I appreciate the fuck out of your investigation of the way FEAR did it's magic. To this day it remains my main inspiration for whenever I need to do lighting in my levels.
ab19e6 No.14830813
>>14830800
I think it looks pretty great, maybe make the letters just a tad smaller so they're framed better by the buttons! Sage for doubleposting.
ab19e6 No.14830815
>>14830813
awgh I clicked the sage box twice and there it is, I look like a fucking tard
bbeb12 No.14830818
>>14830803
why not stitch them together by averaging the vertices on the edges? That sounds simpler to implement.
>>14830400
I also very recently started playing factorio. It really eats into my dev time as well.
To be more on topic, i'm trying to concentrate on porting the latest version of my engine back to Linux. I'm right now porting the networking code. Luckily both windows and unix use the same networking API.
35cfd1 No.14830820
>>14830803
Most of the people that mod this game don't speak English. Like I talked to someone who knows a lot about the game's engine however he only speaks Russian and used Google translate to talk back. So I mostly have to figure this shit out myself.
The engine itself is much more moddable than a lot of engines I've dabbled in. (STALKER's is definitely more obtuse). It's just so little of it is documented, so I'm unsurprised few people made shit in it.
ab19e6 No.14830844
>>14830818
Because if there's a sharp edge there I'd be losing it to the average. I turned it off for now for quicker iteration on the code, but there's a function that derives a sharp edge from Hermite data of the surface and averaging a vertex it produces between two separate chunks would result in noticeable distortion.
>>14830820
I could not imagine putting this much work into modding, but I appreciate the fact that folks like you out there exist, digging through old code of cult games and producing patches/mods/new maps for us to enjoy.
What exactly are you planning to produce?
35cfd1 No.14830870
>>14830844
>What exactly are you planning to produce?
I have a lot of ideas, some ambitious some mundane. Right now a lot of my goals are to try and reproduce things from the vanilla game.
Eventually I'd like to make a series of singleplayer maps. I remember in the console versions of F.E.A.R there were these "instant action" maps where they'd essentially drop you into an exclusive singleplayer map that contained no horror elements where your goal was to kill everything and reach the exit. So I might try to make something like that.
Eventually I'd like to do something more ambitious like attempt to continue a previous Monolith game. Like see if it would be possible to actually make Shogo 2 in it as a mod.
9697c0 No.14830874
>>14829358
I think the weight painting was the part I was missing in all of this and what I have to read more into.
ab19e6 No.14830883
>>14830870
I hope everything goes well in your FEAR study project, anon. From what I've seen in the previous threads, you're dedicated and diligent and as such I'm sure you'll get to the point where you want to see yourself get with this.
046b90 No.14830894
>>14830820
>anon ventures into slav territory to JLMG
Now this is amateur gamedev.
bbeb12 No.14830963
>>14830894
>implying that /agdg/ isn't already slav territory
5425ed No.14831207
trying out this thing https://forum.unity.com/threads/navmesh-links-generator-for-navmeshcomponents.515143/
and i'm getting the issue that it generates links through ome surfaces simply because there's an edge in the navmesh there
anyone got ideas on how to fix this other than manually checking if every single link is correct?
raycasting isn't an option since that would break the links that are actually supposed to work
871ae4 No.14831235
>>14830963
I'd like to think that it's all the same man.
Will of solid steel, with a noggin just as dense. Maybe more so with that additional lead. Life just can't keep him down.
58b8d4 No.14831310
>>14828598
>Figured out how appdomains work in practice after like a week of tinkering
>closed 361 tabs
>now have proper mod support
>still don't have a game
>>14828544
It's no use, man. You finish one game, you start on the next. You're here forever.
e8a43e No.14831362
>>14828598
>Any new skills you've learned?
Using ray tracing renderers. Considering buying a Blender VRay license to do my stuff for the Shinobu project .
>Significant milestones you've made?
I've been lead Technical Artist on another game at work and it might get funded by outside investors which would be really good. Company might start up its own game studio. Don't bother asking about the game it's just for baiting investors (good graphics, mobile, etc.) Vid related.
>Post a picture of your progress!
Here's some videos of some new look at logic I'm working on. Target switching.
79daa1 No.14831415
>>14831235
So like Florida Man, except dumber
30a7c0 No.14831449
>>14828598
Doing some webdev for a portfolio site, mostly frankensteining together https://thebestmotherfucking.website/ and http://motherfuckingcms.tk together with a design I made myself based on some template and some custom basic js. Hopefully by the end of it I'll have an easily editable site that looks like a professional template one but with some extra features and without the bloat of jquery.
Pictures coming soon when I finish it. Probably shouldn't link h8chan to my portfolio site since it's literally [my full name].com
>>14831362
Work-project looks comfy if nothing else. Was about to say "oh now I get why your Shinobu is so qt, you're a pro" but then I remembered you got this job by showing that studio your Shinobu wedding simulator you absolute madman.
>already being lead tech artist months after joining the company with technically no prior experience
Nice. Kinda goes to show that not being able to find work due to lack of experience is a meme, if you work hard in your own time and show good results you can get anywhere.
62eb8d No.14831708
>make a program
>it automatically opens a console window
Why is this and how do I stop it?
cfb43d No.14831760
Is it necessarily bad to want to take bits and pieces from other works, be they game mechanics, settings, or story concepts to create your own? Every time I read something cool or play a game that has a neat way of doing something I think about how I could implement it into something of my own.
It does feel like I don't have any creativity of my own and would just be aping others, but I also realize it's near impossible to be completely original. Thoughts?
5425ed No.14831783
YouTube embed. Click thumbnail to play.
>>14831760
it's never stopped anyone so far
62eb8d No.14831786
>>14831760
You can either copy ideas and be blamed for copying it, or you can forcibly try to avoid any similarities with any other game and then still be blamed for copying because there's something out there that already did it but you just didn't know about it. The difference is that in latter case your game will be shit because you avoided all the good ideas just because someone else also did it.
7c4539 No.14831793
>>14831760
Everyone does it m8, look at how many people copied doom or gta.
58b8d4 No.14831794
>>14831708
Depends what you made it in. If it's c++ you'd put a WinMain instead of main and fuck with some compiler options.
871ae4 No.14831810
>>14831760
You just have to steal enough things from enough places that people can't tell what the fuck you're ripping off anymore.
But only steal the cool things.
62eb8d No.14831814
>>14831794
I use WinMain to open a real window but the console also appears.
>fuck with some compiler options
Didn't think about that, I'll look into it.
58b8d4 No.14831831
>>14831760
>Is it necessarily bad to want to take bits and pieces from other works, be they game mechanics, settings, or story concepts to create your own?
No, as long as you steal from enough people that cobbling it together involves some creative work on your own. Also, don't settle for the first version, keep iterating on it.
e45b19 No.14831871
>>14831760
That's totally fair game. Practically every existing story used something else as a base, be it a different game, movie, books or things like biblical stories and legends/fairy tales.
Use existing things as a base, but change them up to suit what you want to go for and combine it with bits from other sources and ideas of your own. That's where the difference between inspiration and straight up copying something lies. You can't copyright something like a character personality or a game mechanic, but if you don't change it up people will absolutely accuse you of being creatively bankrupt.
>>14831831
Also this.
938a1d No.14831898
f97735 No.14832292
>>14830084
I'm penniless because I'm in education. My money has all been going towards a CS degree (By that I mean living expenses while attending). I have an in for a "formal" job right out of college since my friend's dad is in-charge of a team at a software development company.
It gets better from here. I just have to suck up a few more years.
f97735 No.14832295
>>14832292
inb4 sysadmin cuck gets all riled up about CS students again
58b8d4 No.14832298
>>14832292
>my dads friend works for nintendo
f97735 No.14832327
>>14832298
It's not really like that. It's just some boring software company that makes software for whoever's willing to pay for it (Usually other companies who want specifically constructed software for internal use). But I am more likely to get a job since I'll have a recommendation from someone in a management position.
5425ed No.14832423
>>14831207
well it more or less works
still need to manually remove links in some places where you can't go, like where there's a ledge but no wall underneath (unless you want to switch it to a jump instead of climbing)
ignore the jitteryness, it's partly because of the gizmos and partly because of the recording
2a973b No.14832472
>>14832423
I like the comfy Oblivion music in these, keep up the good work.
5425ed No.14832761
>>14832719
i added that at least a month ago, i only just now got it to automatically generate the links instead of having to place them there manually
56acf8 No.14832777
>>14831760
To bake an apple pie from scratch, first you must invent the universe.
I'm not a fan of Carl Sagan, but it's a good quote. You can not make anything that is "entirely yours" in every way. Everything you make will be based on assumptions and ideas from your surroundings, even as far back as your upbringing and environment, and games that you grew up playing. Everything you make is dependent on other things. Even when it comes to art history, absolutely every change and movement is based on inspiration from others, even if the inspiration is trying to explicitly do the opposite of somebody else, it's still inspiration from somebody else. Thinking of some of the original stories, most of them evolved from aping off something else. Mario was aping off King Kong and evolved incorporating Japanese folklore. Metroid was a ripoff of Alien in a lot of ways. Even as far back as the Hobbit and Lord of the Rings, Tolkein invented a lot of things, but the majority of his foundations were borrowed from old folklore. You can not find a good, creative, original story that didn't take a lot of ideas from other things.
As others have said, which would you rather have, a creatively destitute game that takes a lot of fun ideas and story concepts and ends up being a really good game, or a really creative and original game with no discernible foundations that is boring and shitty? If you take a lot of ideas from other games and stories and merge them together in a way that is cohesive and fun, isn't that also creativity?
tl;dr: don't worry too much about it. Just make a fun game. If you really need to, you can twist the story ideas and some mechanics after the fact, as long as the game is fun.
92d6cc No.14832804
>>14832719
>>14832423
see that building there?
d14655 No.14832850
Just bought a tablet to test my game on, which puts at around €1000 spent on developing this shit so far. I'm feeling the pressure to actually finish shit this time around.
86024b No.14832854
>>14830800
I see what you meant, the letters were way too big, just needed a slight shrink to fit better. tbh I don't know if anyone is going to use this image for anything, but it was good practice.
>>14831362
>I've been lead Technical Artist on another game at work
Are you guys hiring? Because damn, I am the living stereotype of a starving artist right now and my problem is my portfolio only consists of cooking anon's tiny icon illustrations
5425ed No.14832866
>>14832850
>buying garbage to test shit meant for garbage
don't you have a phone to test it on?
or if android you could emulate it
seems like a waste of money for an educational game
86024b No.14832868
>>14832854
woops, first is meant for >>14830813
d14655 No.14832878
>>14832866
It's not really developed with phones in mind and you can't beat testing on a live device.
9781bd No.14832909
>>14828598
I was experimenting with blender and found out that it can procedurally generate trees. Not the style that I need for my game, but still pretty cool.
Other than that I've been improving my engine and added a scripting system to it. I'll probably make some assets and environments next, and finally settle on the art style.
ab19e6 No.14833064
>>14832777
trips of truth
all art is stealing, but you gotta make sure you're stealing basic elements, not complex combinations of things. It's like in writing you learn shit from other authors, various literary devices and ways to talk about stuff, but if you steal entire soul of the work you're a dirty fucking plagiarist.
>>14832909
Any idea what it uses? I'm going to be procedurally generating trees for my game later on, it'd be neat to know how that shit's done in other applications. hopefully it's fast enough because my shit is meant to be generated at runtime unless it's saved to disk prior to load
9781bd No.14833119
>>14833064
It's probably a python script like every other addon in Blender. This one is official but disabled by default. Here's the doc: https://wiki.blender.org/index.php/Extensions:2.6/Py/Scripts/Curve/Sapling_Tree
ab19e6 No.14833371
>>14833119
Thanks, anon. I'm gonna look into it.
6838b5 No.14833386
>>14832854
You need to learn 3D applications like Maya, Substance Painter/Designer, Modo, Arnold, Vray. I only use Blender because its good for quickly mocking and hacking together shader assets.
955ff3 No.14833579
>>14828664
it's time for prozac, anon
45c3a8 No.14833899
Rate this UI. Was told it's "too confusing and blends too much with the background" but wasn't told how I could make it better.
>>14828544
Crippling Warframe addiction.
5e0c2c No.14833939
>>14833899
The UI borders are entirely too bright, that's what makes it blend in with the game view. It might look better if your borders were much more contrasting. If you look at older city builders, command and conquer, and other games with a UI that's vaguely similar, you'll notice they employ contrast at the borders. They use a really bright color or a very dark color.
ba02bc No.14833940
>>14833899
too confusing and blends too much with the background
d14655 No.14833941
>>14833899
Have you tried making it darker? I don't really like that it's so heavily textured but I can kinda see the look you're going for. Probably would help if you could add some actual GUI elements in there as well.
Performance turned out to be fucking ass on the tablet, so I guess it's a good thing I bought one but jesus christ I have no idea why it's so bad.
58b8d4 No.14833949
>>14833899
>Was told it's "too confusing and blends too much with the background"
Yeah. The colors aren't different enough from what's going on in the main display, the structure is also similar (divided into bricks vs divided into hexagons) and the borders between them also need some work to make them stand out.
45c3a8 No.14833977
>>14833939
>>14833941
>>14833949
Trying brighter colors. I'm thinking I'll have to do something about the wood ring effect, it's just not doing what I want it to do.
58b8d4 No.14833989
>>14833977
Try giving the border some shadow like it's 3D.
871ae4 No.14834008
>>14833977
It's not just the borders, it's the general texture and value. You want the delineation to be clear even in extreme circumstances.
45c3a8 No.14834014
>>14833989
I was actually trying to figure out a way to do that. I know that you want to basically paint on a gradient in a certain way I can visualize but can't really describe, but I don't know *how* to paint it in that way.
>>14834008
How much/what kind of blur is that?
046b90 No.14834027
>>14833977
That stone has too much detail.
58b8d4 No.14834051
>>14834014
Not sure if you need a gradient. I'm not any kind of artist.
e72d43 No.14834059
Anyone else staring into the deep abyss of neural nets to generate content? Shit's amazing although quite a lot a work on it's own to make usable.
pic related, a few drawn lines are enough to generate the map. If you want the paper look for "Interactive Example-Based Terrain Authoring with
Conditional Generative Adversarial Networks"
871ae4 No.14834069
>>14834014
Just some standard blur at around a 20 point radius I pulled out of my ass to make a point. You could just step back a bit and squint at a thumbnail to get the picture. Also using lightness to desaturate, rather than luminosity.
Cultures gets away with not being that contrasting in value by having those UI elements being just one big ol' block of relatively solid gray in an otherwise very colorful layout.
Right now, your HUD just looks like it belongs in the same frame, so to speak. Like you were looking at the lower-left corner of some diagram on the wall and you could see the wall behind it, in early CGI vision.
871ae4 No.14834116
>>14834069
Also, you might want to postpone tweaking the bar itself until you get some ideas for how the buttons or whatever should look like, or how many there are going to be and how much of the backdrop would actually be visible. It still needs some tweaking no matter what you slap on there, but what you put on it will pretty drastically impact what you need to make everything distinguishable.
Design choices in general might be something to consider. Do you want the land to look drab, or would more color help? What would you need to make everything read well? You can do a lot with just texture and patterns as well. Maybe something that won't somehow end up joining with the hexagonal lines, as at that particular zoom level, the dark gaps kind of line up here and there.
Maybe it'd be easier to just edit a screenshot in some program to see what works and what doesn't, if changing parameters in the game itself isn't extremely convenient.
45c3a8 No.14834171
>>14834116
This is a city view for a 4x-ish game, side bar is supposed to contain all of the units/buildings/whatever you can build, bottom bar is supposed to have info about the thing you want to build and the button to actually build the thing.
Although, now that you mention it, I haven't decided where I want to put the build queue. I might just turn the entire bottom bar into the build queue.
I'll take a break and do a shitty mockup of what elements I want to go where, roughly.
e841b5 No.14834181
>>14834171
first list all ui elements that must exist, then focus on deciding where they go
f5a861 No.14834205
>fighting the same bug for two straight fucking days
it's so fucking stupid but I cannot find the cause.
45c3a8 No.14834256
>>14834205
Don't give up, Anon! I believe in you!
11a1fb No.14834272
>>14834205
Make it a feature then.
d8e980 No.14834281
>>14834272
>the credits screen is a functional email client and I don't know why
e1218f No.14834308
>>14834281
Please explain.
11a1fb No.14834375
>>14834281
Simple:
Prevent the player from altering the send address and make an email account for the game and make the account it send to that address.
Congratulate the player on completing the game, and ask them for any feedback at the address.
Then as the dev check the email every once and a while and maybe post responses to them online.
And now that glitch is a neat feature.
f368c8 No.14834376
Yo NIGGAZ,
I'm going to develop a physics-heavy, 3d game engine in C with the SDL graphics library these holidays. Currently, the objective is just a functioning engine where every object can be broken down into smaller components and is subject to laws of physics.
Already got some good ideas from the resources list of what to read: (i.e. Physics in 3D, lazyfoo's tutorials, game engine foundations, etc…). Any particular resources that would benefit me?
3b74fc No.14834377
>>14834205
What's the bug?
f5a861 No.14834607
>>14834377
I'm this guy: >>14830803 my IP just changed, I had to reboot my router and it ended up causing that.
The bug is that my contouring algorithm isn't reading voxel data properly resulting in the contour thinking there's a surface at chunk boundaries. No idea where the bug is yet. Still going through my code with a fine toothed comb.
ed5f8d No.14834614
>>14834607
dick in ethernet port
e8a43e No.14834622
>>14834376
Slow down there Speedy Gonzales. You are about to embark on a project that will take years to accomplish. Have you ever even done anything similar to what you want to do?
6137ee No.14834662
>>14834622
Are you fucking retarded? Years?
This stuff is pretty straightforward and I'm not doing anything massively complicated. I'm creating a universe of objects comprised of vectors with unique properties that will move around in accordance to physical laws. The math already exists for simulating this stuff in Euler angles which is perfect for vectors. The difficulty is in understanding what works most efficiently, proper game development practices and translating this to graphics on a screen via SDL.
f5a861 No.14834663
>>14834662
not to start any shit anon but you don't sound like you understand what you're committing to.
aaf512 No.14835167
Any tips on making music if I have zero experience?
Preferrably interested in making NES/SNES-sample sounding music, doing it all on PC.
Obviously I need to understand music theory first and loads of other shit, so if you have any suggestions or resources that would be wildly appreciated.
bbeb12 No.14835172
This is the progress i've made porting the latest version of my engine back over to Linux. It's working okay, not sure why its not able to open some of the textures but I will figure that out later. The whole system around mouse input with linux is nonexistent so I can't really do anything in the engine right now.
>>14834662
>>14834376
My 3D engine has been in development since February 2017. So 16 months to get to a 3D world that you can collide against with a custom map editor. Not counting that this is me learning from my mistakes with my last engine (14 months). In theory it shouldn't take you as long if you don't write your own map editor, and you don't use an API like Vulkan, but you still need to know a great deal and you still need to make a great deal of mistakes to actually be able to do it right. Maybe like a year if you do things correctly and move very quickly.
Maybe to consider the problems that 3D game engines have to solve, like how do you do fast collision detection against an arbitrary 3D world? How do you do occlusion culling? It's not trivial at all and you do have a LOT to learn. Yes the physics part where you simulate some cans falling off of a shelf AFTER all of the hardest problems have been solved is a bit easier. Also SDL isn't a 3D graphics library (you probably want to use OpenGL for your graphics).
OK, so consider these resources to get started:
https://www.cs.utah.edu/~jsnider/SeniorProj/BSP1/default.html
https://www.cs.utah.edu/~jsnider/SeniorProj/BSP/default.htm
http://www.alsprogrammingresource.com/csg_tutorial.html
http://www.glprogramming.com/red/
https://learnopengl.com/
http://nehe.gamedev.net/
https://pybullet.org/wordpress/
f5a861 No.14835190
>>14835172
Why Sigma? Is there any particular backstory to why you decided to name your engine Sigma?
5425ed No.14835192
>>14835190
it strikes fear into the eyes of algebra students
45c3a8 No.14835205
>>14835172
I fear and respect your autism, Sigma Anon.
f5a861 No.14835212
>>14835192
I'm still confused by Sigma notation myself. Could never figure it out in school, and now reading math papers on various CS shit - they fucking love that Sigma don't they?
One day I'll understand it.
45c3a8 No.14835225
>>14835212
Fuck math, this is why I cuck to Jewnity.
Godot was too buggy in 3d when I first started it and now it's too late to switch.
bbeb12 No.14835232
>>14835190
Yes, I thought of this when I was making maps with valve hammer editor. (This was back when I was working on a mod for half-life 1) The "missing texture" texture is a lambda. And, I was thinking that if you were going to name your game engine after something, you should name it after a greek letter because they have cool names and you get a cool logo. I couldn't use lambda because goldsource already uses it, and "Sigma" is a pretty cool sounding greek letter. And so, that's why my "missing texture" texture is a Sigma.
A lot of people have said that my map editor looks like hammer editor. That isn't a mistake, since it's the only map editor that I am any good at using. So, I took a lot of design choices from it.
>>14835205
Thanks! I'm hoping to have a playable demo by the end of the summer where you can walk around in a level.
>>14835212
It's literally just a kind of for loop for regular mathematics.
f5a861 No.14835240
>>14835225
In my particular case even if I use a premade engine (which I do - UE4) I have to do a lion's share of math myself because you know, when you do something even remotely original/uncommon you have to do it yourself.
>>14835232
That's cool. And you're right, Sigma is a cool-looking letter.
>literally just a loop
But like, how do you read it?
5c1760 No.14835243
>>14835172
>16 months of deving
I wish I had your level of commitment, man. How do you keep yourself motivated to keep moving forward?
f5a861 No.14835257
>>14835243
I'm not Sigma, and I've certainly haven't done anything of his scale, but I think this sorta thing comes down to why you're doing what you're doing.
In my own case, I've just decided that games are the one thing I love the most, and so gamedev is the only route I have. If gamedev is the only route I have, the game I'm working on now is the only route I have. And once you're there, the decisions you've made push you forward, giving you no opportunity to slack off.
bbeb12 No.14835270
>>14835240
Ok, so this code is a for loop to calculate the average of an array:
int average(int *array,size_t size){
int ret = 0;
for(int i = 0; i < size; i++){
ret += array[i];
}
return ret / size;
}
And pic related is how you do it in sigma notation. So the number at the top is the maximum, the number at the bottom is the variable, like "int i = 0" at the beginning of a for loop. Then the result is every value of x[i] (in the mathematical notation, x is a set ranging from x[0] to x[n], so its like an array) Sigma is a lot less flexible than a for loop because you have to add everything up together.
I also found a much simpler picture about it on google images. So maybe that makes more sense.
>>14835243
I don't really know, I guess I just focus on smaller parts of the task in the short term, until eventually it just starts accumulating into what it is today. So, right now I am only really thinking about making the linux version work for all of today and yesterday. And the "reward" in my head for me is that I get to work on the next part, which is a problem that am usually more interested in solving when it is just out of reach and I have the current problem to get out of the way first. So i'm kind of looking forward to implementing occlusion culling (maybe I will do it before Godot does it! then I can brag about it on here in a tounge-and-cheek way) And I eventually forget about how long the project is, I don't spend a lot of time thinking about that beyond day-dreaming about how I want to implement the systems that are months away from me starting on, like lightmaps.
f5a861 No.14835283
>>14835270
Fuck yeah, anon! Thanks!
aaf512 No.14835290
holy shit I don't even know how to start
how to weed out all the options and be left with one straight-forward way and do that instead of constantly thinking 'maybe I should do something else'
after much pontification I started an online c# course some time ago, but started to have some problems with supposedly easy task (just in terms of logic)
thinking I lack in this department, I now started to freshen up my math/algebra/etc through online courses in order to supposedly hone my critical thinking skills (haven't done any math in 7 years)
but now I again start doubting myself whether that is something I should even waste my time on
jesus
45c3a8 No.14835295
>>14835290
What happened, anon?
5c1760 No.14835313
>>14835270
Honestly, that's probably the best way to approach a large project, care about the little victories that make up the big one. I think I'll try to remind myself of this during the times I feel like it's a bit much to finish my projects.
>>14835257
I think I can see where a single-minded motivation like that can help, but I'm not sure if I could handle the times when I hit a brick wall with that kind of mindset. "Wow, I suck at the only route I've given myself. Why do I even live?"
aaf512 No.14835329
>>14835295
nothing really, outside of what I described
58b8d4 No.14835332
>>14835290
>how to weed out all the options and be left with one straight-forward way and do that instead of constantly thinking 'maybe I should do something else'
Do one thing and see how it works out in practice. Just think of coding as thinking out loud, not making a commitment to whatever you're typing. If it works, it works, if it doesn't you've got a better idea of what the real problem is.
45c3a8 No.14835341
>>14835329
What was the specific task that got you thinking you should go back and study math again?
f5a861 No.14835351
>>14835313
I feel like that right now. But the thing is - just like everything else before that, I know I can do it. Any one person can do anything they set in front of themselves with sufficient motivation. You can do anything you want, anon. Anything at all, but you just gotta really make sure you decide on it and never turn away no matter how hard it gets.
It gets really hard, but at the times when it does, you learn the most. You'll hit lots of brick walls, you'll be afraid of lots of things, and when you're confronted with them, you'll feel hopeless and as though you can't win. But you just keep pushing, digging into it as much as you can even if you only have enough strength left to peel away at it, one atom at a time.
That's what separates a yesdev from a nodev. It's what separates any one "yes" person from a "no" person. Not talent - that shit doesn't exist. Not better brains. Not better conditions. It's an ability to set a goal, and push yourself into it, until you as a specialist fit the hole the goal is hidden inside of.
>I'm not sure if I could handle the times when I hit a brick wall with that kind of mindset. "Wow, I suck at the only route I've given myself. Why do I even live?"
You can't suck at working to achieve a goal. It's something you either do or don't - there's really no qualifier for how good you're at it, because it's all learning and subsequent use of what you've learned to get yourself a step closer to the goal.
I know for a fact, anon, that if you do this, and set your mind to achieving the goal you've set and really, really stick to it the way I laid out, you'll achieve it. I believe in you, and you should believe in yourself just the same, because it's not something specific to anybody in particular, it's something everyone can do; And that includes you!
13b091 No.14835479
>>14835290
I am not great at math and I do fine in gamedev. Hell I am not even very knowledgeable in programming. 99% of the time the most complex math I use is just trigonometry, and anything more complex that I need I just read up online and write into my language. Understanding the concepts of programming logic and how to write good code is much more important, unless you're writing your own 3D game engine I guess.
1148a7 No.14835487
>>14835172
I don't know anything about openGL but the little bit of research I did led me to believe that I would have to learn to C++ to take advantage of it.
I doubt I could get by with just C.
5425ed No.14835493
i got the link generator to check if there's a wall near the bottom of the link, if not it gets converted to a fall instead of a climb, main difference being the animation and that the fall is a curve instead of linear interpolation between the waypoints + the end point is placed further
45c3a8 No.14835494
>>14835480
Addendum:
The way I've got my game designed at this current time, you can only build one copy of each building. Once it's built, it doesn't get removed from the build options, but it changes visibly (probably just a color change) and adds a button to destroy said building.
bbeb12 No.14835537
>>14835313
If I'm hitting a wall sometimes I will stop what part I am working on, and work on another part that I can make progress on, until I can figure that part out.
>>14835487
I'm only using C, and doing fine with it. If you just want really basic graphics then use OpenGL 1.1 or something, that will allow you to make a 3D engine while seriously reducing the amount of time you spend actually programming 3D graphics. (not that you cant make it look good)
367427 No.14835578
>>14835487
OpenGL's API is in C. You don't need C++.
Sure, you can use C++ to objectify many OpenGL objects, but then you run into the issue that the entire thing is this big ugly state machine. Not to mention that the things you can split up into many pieces (models, shaders, etc.) are things you shouldn't actually split up because you want to batch as much work together in a single draw call as possible if you want to optimize your performance.
9697c0 No.14835606
>>14835487
>I don't know anything about openGL but the little bit of research I did led me to believe that I would have to learn to C++ to take advantage of it.
What you read is full of shit, and what you posted is opening a huge can of worms. I'll break it down for you simply.
OpenGL is almost entirely written in C with no C++. Most high performance games or console games are written in C, or C++ that is basically C. They ignore everything C++ adds like, inheritence, templates, operator overloading, exceptions, etc. I honestly still to this day do not understand what they actually use from C++ other than the name to find programmers easier.
Almost all indie PC/Mobile games are made with an engine that uses C#, java, etc and pay a performance penalty for it, but they can make their game much much faster. The main trade-off between C and a high level language (C++ is in the middle, but it has the worst of both worlds) is either build everything yourself from scratch and make it faster, which will burn you out. Or just have most of the grunt-work done for you anyways. Honestly. If you're going to use C++ you're probably better using C# or another high level language.
bbeb12 No.14835644
>>14835606
>Most high performance games or console games are written in C, or C++ that is basically C. They ignore everything C++ adds like, inheritence, templates, operator overloading, exceptions, etc. I honestly still to this day do not understand what they actually use from C++ other than the name to find programmers easier.
Relevant webms to this, I don't know if it actually answers the question but maybe it helps, you've probably seen this though.
bbeb12 No.14835879
I have progressed on the linux version to the point where you can look around using the mouse, and you can move around by typing the "+forward" command into the console. The linux keyboard input isnt integrated totally into the keybinds system, since the +forward -forward commands require it to know when you started and stopped pressing that key. Maybe X11 has an event that says when someone stops pressing a key? That would be helpful for this. Also I still have broken textures, it doesn't bother me though it's not the end of the world. I will fix that once the more important issues about keyboard input have been solved. At least all of the networking code is working correctly, and the FLIF code sort-of works… it's probably some OpenGL rule that is breaking the textures, and not FLIF decoding, though.
Its kind of late, so i'll work on it tomorrow. Then after this I can finally start working on the occlusion culling.
938a1d No.14835937
>>14834607
I have no idea how feasible the following is to do, but w/e. Make your chunks overlap. So instead of a 16³ covering 0-15, 16-31, have it cover 0-15, 15-30, 30-45 etc. So each chunk shares it's boundary layer with the adjacent one, storing duplicate data. As a first pass, just contour normally. The false border will be hidden inside adjacent geometry. Or not, if the adjacent chunk hasn't finished meshing yet, which is good. You don't want holes in the world. Then, look into culling the hidden geometry when adjacent chunks load. I don't know how the resulting geometry is ordered, but there might be some simple way of selecting all geometry inside the boundary voxel and just deleting all that spans beyond the midpoint of the boundary voxel plane, from the perspective of a particular chunk. Keep the closest half. Alternatively, just select a 4 voxel thick boundary set, 2 from each chunk and mesh the 2 thick middle set, i.e. run the algorithm over the full 4 wide set, but only emit geometry for the middle 2 voxels.
dcbee1 No.14835973
>>14835879
>FLIF code
Is it real-time or does the engine generate a bitmap from it?
938a1d No.14835976
>>14835172
>>14834376
I should add a bit to sigmadevs answer.
The first two links are about binary space partitioning. That is an awesome data structure for a ton of shit, but it's a pretty common opinion that it is not suitable for representing data that should mutate at runtime. So while it's good for indestructible static level geometry (collision, line of sight, culling, pathfinding, the list goes on), it might be less so for your application, i.e. deformable dynamic bodies. This is mainly because the time complexity of constructing a BSP from a polygon soup is O(n²log n) and they're difficult to edit post-construction (removal in particular is hairy). I'm personally convinced it can be done and I'm gonna try at some point, but it's not the common application, so it's treading deep water.
Another issue is that it's a surface representation. It simply distinguishes between open and solid space. It doesn't say anything about what that solid space actually contains. So if you want your objects to consist of anything other than generic metal (or at most a per-object substance), you need a volume representation. That's why stuff like voxels with surface construction, like what the dual contouring anon itt is doing is common. It can represent the contents of bodies, not just the boundary between the body and the universe.
A thing you could look into (f368c8 that is) is voronoi partitioning. It's a way of splitting up surfaces or volumes into chunks that kinda look like the shards of a shattered crystal. It's pretty commonly used when you want to make it look like something broke.
b29018 No.14835986
>>14835281
breddy good. Needs lighting and texture work but the geometry looks promising
>>14831362
>I've been lead Technical Artist on another game at work
this kills the "get help" argument
Happy for ya.
>>14833899
try using a simple marble background instead of brichs, with a distinct border between game space and UI
>>14832909
>Not the style that I need for my game
They look real nice, though
938a1d No.14836073
>>14835487
>>14835578
It's a known problem that it's difficult to model OpenGL with OOP. There's even an article on it.
https://www.khronos.org/opengl/wiki/Common_Mistakes#The_Object_Oriented_Language_Problem
Take it from someone who has written GL code in C++. It's a bitch getting it nice. Sure, you can always hack until it works, but you'll want more than that from your engine code. It's supposed to make development easier, where you can just compose high-level behavior from a set of smaller components. Not harder, where you constantly have to extend your engine, hunt down weird bugs and add case-specific fixes everywhere. I still don't quite know how to model GL with OOP well and if you think "well, obviously a buffer is an object, array and index are subclasses of that, a VAO is an object, so is a shader, a uniform, a vertex format etc.", trust me when I say that major problems start popping up when you do it like that, particularly in the interaction between these objects. Feel free to try and please report back if you find the perfect abstraction so I can copy it.
637c86 No.14836193
just spotted this in a game I picked up recently. kudos to the dev
5425ed No.14836215
>>14836193
>pixelshit
wot gaem
b29018 No.14836220
>>14836215
>pixelshit
confirmed cuckchan agdg
e4de97 No.14836222
>>14836193
Was this advertised in some /agdg/ or did you stumble upon it by a chance? Or maybe it's your game?
637c86 No.14836225
e4de97 No.14836253
>>14836225
I wasn't asking about the name of the game, ansdor.
637c86 No.14836260
>>14836253
someone in a group of people I play games with bought it and recommended it, so I got it myself
172289 No.14836579
>mfw I left my project files on another computer
I'm too comfy to go back now just to get them, and I can't really do anything useful with my old files. What can I do for the rest of the day that isn't a waste of time?
83695e No.14836590
>>14836579
Reverse masturbation
30a7c0 No.14836630
>>14836579
Look up how to use version control properly so you can always access your latest files anywhere.
836898 No.14836647
>>14828598
I have nothing to show besides that I've made significant progress for 4 days straight, which is (sadly) a record. If when I don't give up, I will have something to show.
Seems like this summer is very hectic, I have family visiting or events to attend nearly every weekend, and shit has piled up at work recently as well. Despite all this I will just like make game by August, even if it kills me.
b29018 No.14836661
>>14836630
this.
also
>>14836579
>I'm too comfy to go back now just to get them
>for the rest of the day
How… how far away are they? Cause if it's a 15 minute drive you probably should get off your comfy ass.
86024b No.14836692
>>14836590
>Reverse masturbation
Excuse you?
046b90 No.14837316
>>14836692
That's when you take a syringe full of your semen and shoot it back up your urethra.
bbeb12 No.14837321
>>14835973
It's just the code that decodes a FLIF image using libflif_dec. You can learn more about it at http://flif.info/
>>14835976
Great stuff anon… you're right, its one way of doing things, but not the only way, and there are lots of ways to get to your own 3D engine.
5425ed No.14837326
dumb question
how viable is it to outsource myself as a codemonkey for vidya on something like fiverr? my country doesn't have gamedev businesses other than ubisoft and i have too much free time as is
one thing i'm noticing is that nearly everyone is from pakistan or some other 3rd world shithole
cfb43d No.14837472
>>14837326
>my country doesn't have gamedev businesses other than ubisoft
Then go work at Ubisoft
5425ed No.14837488
>>14837472
if i wanted to work at ubisoft, i mind as well go work at a supermarket instead, i'd get the same or more job satisfaction
367427 No.14837510
>>14837326
The main problem with sites like that is because in order to be chosen for the job, you need good credentials, which you get by doing jobs.
The reason why you mostly see pajeets there is because that's the only decent paying thing available to them. On that note, if you live in France, then your wage will be pretty shit compared to the cost of living, and I would have to recommend against it.
cfb43d No.14837519
>>14837488
What makes you think that? Are you familiar with their inner workings? Would you really rather work through some garbage service like fiverr than take an opportunity to work at Ubisoft? You can rag on the crap they deliver all day, but that's not an indication of how it would be like to work there. You'd certainly get more experience with them is what I'm thinking.
e841b5 No.14837537
if (inTris.Count == 2) {
for (int tri = 0; tri < borderTris.Length / 3; tri += 3) {
if (borderTris[tri] == inTris[0]) {
if (borderTris[tri+1] == inTris[1]) {
inTris.Add(borderTris[tri + 2]);
}
if (borderTris[tri+2] == inTris[1]) {
inTris.Add(borderTris[tri + 1]);
}
}
if (borderTris[tri + 1] == inTris[0]) {
if (borderTris[tri ] == inTris[1]) {
inTris.Add(borderTris[tri + 2]);
}
if (borderTris[tri + 2] == inTris[1]) {
inTris.Add(borderTris[tri ]);
}
}
if (borderTris[tri + 2] == inTris[0]) {
if (borderTris[tri + 1] == inTris[1]) {
inTris.Add(borderTris[tri ]);
}
if (borderTris[tri] == inTris[1]) {
inTris.Add(borderTris[tri + 1]);
}
}
}
}
The problem: If you got only two values from the last algorithm, go through all triangles in the mesh and find the triangle which contains these two values, then add the last vector of the triangle that is not either two values previously found.
Is this the smartest way to go about it? I am sorta allergic to IFs, especially a bunch of similar looking ones one after another, sounds very prone to error
I know I am missing the breaks
367427 No.14837606
>>14837537
Instead of checking whether borderTris[tri+(0 or 1 or 2)] matches inTris[0] and if so, then if borderTris[tri+(0 or 1), tri+(0 or 2), tri+(0 or 1)] matches inTris[1],
only check whether borderTris[tri] matches inTris[0], and if so, then if borderTris[tri+(-2 or -1 or 1 or 2)] matches inTris[1].
The only downside is that you have to do a little bit of bounds checking, but it's a lot more efficient than that trainwreck.
30a7c0 No.14837613
>>14837510
Ubisoft has offices in just about every non-shithole country on the planet and some shithole ones anon, no need to assume the guy's from France.
>>14837519
Have you not seen their team-pictures? I agree that those aren't the kinds of people I'd like to work with, which is why I don't work for them despite them literally asking me to work for them.
5d4843 No.14837654
Anyone know good resources to learn how to write procedurally generated dungeons in Unity/ in general?
e841b5 No.14837682
>>14837606
won't the bounds checks actually make it make more ifs total than mine?
In worse case, mine is 5 checks per triangle.
Also, 0, -1, -2 is not a valid triangle, as you are taking two points from triangle A and one point from triangle B
cfb43d No.14837716
>>14837613
Not really, no. Though I get the concern, that isn't really enough to know that you'd actually be working with those type of people. I just think it's bad form to pull your nose up at an opportunity to do up valuable experience as if you're somehow 'too good' for it.
367427 No.14837761
>>14837613
>which is why I don't work for them despite them literally asking me to work for them.
You could always ask for a high wage. Worst case you'll get the job but at least then you can be a lazy piece of shit and milk it until they fire you.
>>14837682
I'm an idiot and read the "tri += 3" as "tri++". Disregard my post.
Anyway, you can try to loop it with some logic, possibly by %3 or a pre-computed %3, but the result of that wont be more efficient than what you got, since that is essentially the unrolled loop. The only thing you can do to improve it is to add breaks or change those if's into else-if's.
30a7c0 No.14837791
>>14837716
The reason they asked me to work for them is because I already have worked with them in the past. I have the experience, and it was not an enjoyable one, largely because of the aforementioned reason of who you work with. It's simply not recommendable to the type of person that visits this site. The other reason is that the games they put out aren't shit purely because of ideological bullshit, they're just an amalgamation of mismanagement and stupid workers, with occasional competent people working their asses off to pick up everyone else's slack.
Sage and spoiler since I'm pretty much blogposting.
>>14837761
>Worst case you'll get the job but at least then you can be a lazy piece of shit and milk it until they fire you
Would be detrimental to my future ability to get jobs, and is frankly not my style.
efeb5e No.14837926
>>14828598
Back in the saddle after about six weeks of getting nothing done on my game partially because of stomach troubles, in my defense but I'm freshly re-motivated and excited to get back at it.
Right now I'm replacing the convoluted sprite clipping logic I had for when an entity is partially in multiple sectors with depth testing/buffers and need to re-implementing culling to work with the camera changes I've made. I've rat-holed on the rendering side of things for way too long I feel like (my own fault for not having a fixed idea in mind) so my next goal is to work broadly on all the features I need so there's a playable game loop (besides the multiplayer stuff) and flesh things out from there.
aa2453 No.14837959
Well I though of another idea for a game yesterday, guess I need to pick up learning love2d where I left off.
afc7e3 No.14838178
f2002d No.14838268
File: 595cd7c7065885e⋯.png (Spoiler Image, 388.67 KB, 1068x1463, 1068:1463, 3826efa13296066f0e2cfd9a32….png)
afc7e3 No.14838297
>>14838268
isn't godot shit?
172289 No.14838298
>>14838268
That shitty character actually makes Godot look less appealing.
f5a861 No.14838303
>>14838298
tbh it's about the ugliest fucking mascot I've seen in a long time. She looks like she smells like spit and piss.
f2002d No.14838317
>>14838303
>>14838298
i think agdg made a better one, but i don't have it on hand.
58b8d4 No.14838331
>>14838297
It's godotful :^)
046b90 No.14838332
>>14838297
For 2D it's great, 3D is promising, but still a ways from being feature-complete. I really like the Scene-Node hierarchy system.
f5a861 No.14838334
>>14838317
Yeah but even so, I believe that's the official mascot and I doubt they'll switch it anytime soon.
93296d No.14838352
Why haven't you written your own raycasting engine yet, anon? Are you a retard?
5425ed No.14838362
>>14838352
>Are you a retard?
i'm using unity, do you have to ask?
f5a861 No.14838364
>>14838352
Why haven't you built your own computer from scratch out of raw earth materials yet, anon? Why haven't you dug them up yourself, anon? Why haven't you created your own pocket universe to make everything from scratch in, anon? Are you a retard?
bbeb12 No.14838392
>>14838352
Been there, done that
93296d No.14838424
>>14838392
Did you use the actual command line icon as a wall texture?
i'm the anon that asked you for tutorials that you used, like 5 threads ago, you are very inspiring to me
bbeb12 No.14838482
>>14838424
This is the texture, I think a friend made it for me, or something like that. I'm not really sure where it's from. Thanks anon! It'll be cool seeing the kinds of stuff you make
That game was drawn with OpenGL so maybe it's not as impressive because it's not a real raycaster.
046b90 No.14838496
>>14838334
It's an open source project, there isn't really an "official" mascot. The robot has been featured in multiple demos and is the icon, Godot-tan is just one artist's interpretation. I like this version better. Pic 2 is the 8agdg anon's version.
172289 No.14838498
>>14838364
I've unironically considered making my own OS.
5425ed No.14838503
>>14838498
do it mayne
make terry proud
f5a861 No.14838520
>>14838498
I understand doing that for fun, or for learning purposes; I just think that if you plan to ship something worthwhile, crippling yourself and forcing yourself to make shit from scratch is the last thing you should do.
a5404f No.14838560
>>14838503
But terry already made the perfect OS.
e841b5 No.14838638
>>14838297
>isn't godot shit?
less shit than being a nodev
d14655 No.14838642
>unity light map bake doesn't use the GPU at all
What the actual fuck, how can such hacks get a job in this day and age? No wonder it takes fucking ages to do anything is dumbass engine.
f5a861 No.14838672
>day 3 of fixing a bug
>still no idea what exactly causes the bug
5425ed No.14838703
>>14838672
>after months realize out of the blue that i forgot to implement serialization of story-related variables, kind of important in an rpg
>add it in about 10 minutes
>everything works fine
>will never have issues with it
get on my level todd
871ae4 No.14838707
>>14838672
What kind of bug? If you're writing in something that has a compiler that can optimize, see what setting the optimizer higher or lower gets you. That can help you with some varieties of memory fuckery by making it actually die instead of limping along until mysterious circumstances cause data corruption.
f5a861 No.14838742
>>14838707
something about reading voxel data of one chunk from another. Somehow something there results in a chunk returning material index of 0 (air) instead of 1 (solid) resulting in an intersection at chunk boundaries, even when I set terrain generation to fill chunks with solid mass instead of noise.
Still checking and rechecking my code, I have no idea how that could be happening. It's meant to return 0 when trying to check voxeldata of an unloaded chunk, but I've added a debug check for that, and it's not triggering it when trying to get a voxel from a boundary between chunks.
It's like a fucking phantom. I'm losing me fucking mind over here boys.
1defc2 No.14838790
Any advice for getting started with the unity engine? I'm assembling a team to make an MMORPG and I'm interested in using unity to do it. Not much experience in gamedev beyond modding, for me, but if I'm going to head a project I need to learn it.
871ae4 No.14838793
>>14838742
Sounds like it might be an off-by-one error somewhere, or something just being left at the default value.
Try adding some offsets or skipping some values here and there, see if you can make the bug bigger or make the processing break in a similar way.
5425ed No.14838827
>>14838790
>Not much experience in gamedev beyond modding
>I'm assembling a team to make an MMORPG
what the fuck is wrong with you
make something simpler
f5a861 No.14838829
>>14838793
It could very well be, thanks anon. I'm gonna keep going through it until it works
d14655 No.14838835
>>14838790
>I'm assembling a team to make an MMORPG
1defc2 No.14838909
>>14838827
Eh, it's business. I don't have the experience, but I'm in the management position. I consider it unethical if not retarded to manage a team doing something that I am not competent in myself. As such, I consider it my duty to at least reach competency so that I can fully understand all concerns… and honestly so I can jump in and fill gaps when shit needs done.
7d115c No.14838917
>>14838790
>I'm assembling a team to make an MMORPG
5425ed No.14838923
>>14838909
make something simpler
cfb43d No.14838975
>>14838790
>Not much experience in gamedev beyond modding, for me
>I'm going to head a project
>MMORPG
This has to be bait.
>>14838909
>I consider it… retarded
That's exactly what it is.
>reach competency so that I can fully understand all concerns… and honestly so I can jump in and fill gaps when shit needs done.
Try again in 5 years.
1defc2 No.14838989
>>14838975
>5years
Hahaha, for a retard.
76bf34 No.14839008
>>14838989
>>14838790
>retard overestimates his abilities
>gets defensive when called out
Every fucking time
e841b5 No.14839042
b8f124 No.14839257
Looking to make a game but have no experience with art. Here's my first attempt at a main character, can someone give me some feedback?
b8f124 No.14839263
>>14839257
Here's an image that isn't blurry
f5a861 No.14839282
>>14838989
If you're planning to become a new target for the local bullies, you're on the right track.
1c83bc No.14839292
>>14834376
Bullet physics has a demo as part of the distribution that demonstrates convex decomposition and rigid body fracturing. Maybe take a look at their C++ demo for this and build a game with it. Raw OpenGL and GLUT if I'm not mistaken.
1c83bc No.14839299
>>14834607
build a simple test case.
use a debugger.
step through the program line by line watching the variables and flow control.
f5a861 No.14839314
>>14839299
Yeah, I'm currently trying to narrow the algorithm down to a single intersection so I can step through that.
1c83bc No.14839318
>>14836215
hahah, raw envy summarized in a single post. good work. It's a finished game. Post yours.
700662 No.14839341
>>14839318
Hey buddy I think you got the wrong door, 4chan AGDG is two blocks down,
Most people here avoid pixels in favor of proper art for sprites or 3D. Further, your wording combined with suddenly poor grammar and writing makes it clear you rushed this post out as fast as possible. You're buttmad, nigga.
9cd1d2 No.14839365
>>14839263
>>14839257
You should make characters you can animate without wanting to neck yourself. Having to think about a lot of motion in every sprite is an easy way to get things wrong and kill motivation as a new guy. If you want critique on the style, it's pretty good I guess, it's very cartoonish which may be an insult if that wasn't your intention. Colors are alright, it looks a bit too saturated, although that might be what you were going for. If you are interested in how other people do pixelshit, there are some neat pdfs on the wiki, this page near the bottom: http://8agdg.wikidot.com/graphics
keep it up
e841b5 No.14839440
Finished the algorithm that creates a copy of the mesh border for each connected map face (rather than a border for all map faces at once).
Now I have to figure out how to draw them on top of the original border without ZFighting. Especially on corners, where there will be at least 3 of them at once that needs to be mixed equally
fa5161 No.14839587
>>14839257
>>14839263
This is actually a pretty decent start, although the anatomy still needs a little work. I suggest you read up on some tutorials, since the most glaring issues are fairly common beginner mistakes, like using pure white/black, jaggies, and oversaturated colors.
Here's some resources covering these topics.
http://pixeljoint.com/forum/forum_posts.asp?TID=11299
http://8agdg.wikidot.com/graphics (you want the PixelLogic PDF series, under "Tutorials")
893302 No.14839651
>>14838790
I've been lurking game development forums since I was 12, I'm 24 now, I can tell this is bait but due to poe's law I should inform you that you are retarded.
You couldn't even write a MUD (but that's what you should try instead).
b30bfb No.14839668
>>14836193
Great game, but fuck those cars that wont get out of your way.
Goering is a nigger
bbeb12 No.14839700
I finished porting all of the new features I worked on into the linux version of the engine. So, while the linux version doesn't have as much polish it does have all of the features that the windows version has. Now I can work on other stuff like occlusion culling, the engine runs really slowly on the GPU I was testing the linux version with, so hopefully that will be fixed soon.
f5a861 No.14840229
How do you guys get your dev names? Do you get em given to you for the first game you're working on? Or do you make yourself a wiki page and give yourself a name? Just wondering.
On an unrelated note, I made a step forward with my bug. I got a new bug, which means that I probably pushed the cause of the old bug from the area it was in with the refactoring I did and now it's popping up there.
I did try stepping through the code but found nothing out of the ordinary. Time to bring out the big guns and refactor shit until it works.
829fc4 No.14840230
>>14840229
>Or do you make yourself a wiki page and give yourself a name?
that would be gay lol
f5a861 No.14840235
>>14840230
I'm just wondering, pls no bully anon
bbeb12 No.14840256
>>14840229
I started posting and after a while people started calling me "sigmadev". The first time was when someone called me sigmafag once. Now people also sometimes call me by my username (ika), I think it's because on all other sites I use that name. Maybe you'll get one soon, Dual Contouring Anon!
f5a861 No.14840278
>>14840256
thanks for clearing that up, anon! I was feeling confused and a little left out but now I'm all caught up.
785ede No.14840286
>>14840229
Congrats your dev name is bugman.
86024b No.14840287
>>14840229
Usually it's made up for you depending on what you do.
f5a861 No.14840305
e8a43e No.14840332
>>14840229
Take a wild guess, bugman.
7260ca No.14840336
>>14840305
Fine, I found a name generator you might find useful
Rapper Names for [Replace With Own Name]
A Performer of Autistic Music Deserves The Perfect Name
Spike OJ
CooloMann
Astolfo Astolfo
Inspectah Autistic
Markastic M M
Spike Simpson
Markadonna
Ol Mann
Mark Arc
Mark-M
Da Real Mark
Markulous M
AynalVohr Simpson
Dark Mark
AynalVohr OJ
Marka Marka M.
Tots-Simpson
Autisticdoc
Mark Park
Mann Man
EmEm
Autisticman
Oj Day
Markormous M
M.M. Simpson
EmAyEnEn
Abs-Autistic
Tots-OJ
Mann Plan
Tots-Autistic
EmAyAreKay
M.M. OJ
Autisticface Mark
Big Spike
Mann Can
f5a861 No.14840350
>>14840286
>>14840332
picrelated
>>14840336
i had a genuine question that I now realize was very stupid and im sorry now please stop the bullying i didn't mean to be this retarded
30004f No.14840369
>>14840350
I don't know why you think you need a name
It's always blatantly obvious who you are
f5a861 No.14840390
>>14840369
well I didnt think I needed a name, I was just thinking about how many anons on here are known by their handles and i thought it was neat, and then while waiting on my shit to recompile i was like "i wonder what mine will be a few months down the line".
and of all names this shit landed me with fucking "bugman". this is what i get for even thinking of being a borderline namefag
700662 No.14840413
>>14840256
>>14840229
You don't really try to get a namefag thing going, bugman, that's generally frowned upon.
Eventually someone will call you something and then you'll roll with it, such as bugman, though it's usually related to your project (I.E Ika/Sigmadev was called Redskydev from time to time, people called me Kowloon anon for the project, Sena! anon is named after his waifu/porn games main character).
f5a861 No.14840434
>>14840413
im sorry, looking back i think I just had a momentary retard moment. I'm not trying to deliberately namefag, honest.
I had another neat thing I wanted to bounce off you guys - since I'm doing procgen that is tied to a seed, what'd be a funny secret seed to sneak in? im already settled on "1055" and exactly how it's gonna look
5425ed No.14840526
>>14840229
long story short, i was playing wow years ago, traded accounts with some guy, one of his characters had a cool name and i've been using it for literally everything ever since then
although in the wiki i'm stuck with my youtube name
e8a43e No.14840683
Please do not feed the loli.
f5a861 No.14840756
>>14840683
That's amazing, anon. Very expressive animation, especially considering that it's procedural.
700662 No.14840779
Progress, managed to get basic animated movement going (no anims for the model yet, working on rig/model for main character and using the Widebrim as a placeholder).
Ambience worked out great, testing ambience from multiple sources.
I found a great program for making maps in that can be easily exported to Godot. It's open source and works a lot like Source Engine Hammer, it exports natively to collada and .obj as well which is great. The .obj exporter even exports texture data which godot can read and auto apply.
Planning on making a first test environment now.
We've also finally set up our group model/asset repository and we're setting up git soon. Turns out we had over 100 minutes of audio for the soundtrack already.
5425ed No.14840787
>>14840779
is this skelly game? i remember something just like it but it was UE4
http://8agdg.wikidot.com/skeleton-game
046b90 No.14840804
>>14840779
Oh so kowloon is going to be on Godot now? Neat. What map editor?
238d08 No.14840816
7083ed No.14840824
>>14840787
>>14840804
Wrong games. It's Die Totenmaske 2.0
700662 No.14840827
>>14840787
>>14840804
Nah, it's Totenmaske 2, too few colours to be Kowloon:
http://8agdg.wikidot.com/die-totenmaske
We wanted to remake D.T because the first game is hampered by RPGmaker as well was the fact that it was made in a month.
D.T 2 is the collaborative effort of me and quite a few others but I am working on maps, having something nicely close to Hammer but usable in Godot was necessary for me. http://www.delgine.com/index.php?filename=product_deled
This works well enough for our needs though nothing can beat the comfiness of Hammer.
I have relapsed from time to time and worked a bit on Kowloon but for now it is on hold, there is a good chance it will be in either Godot or Ue4 when all the maps are done.
>>14840816
Sorry about that, the recording got a bit fucky.
Here's a better one.
9d070f No.14840835
>>14840434
>secret seed of 4 characters
1488, which generates a giant swastika near spawn
>>14840683
HNNNNNNNNNNNNNGGGGGG
938a1d No.14840862
>>14838742
This might be a stupid suggestion, but have you tried just stepping through the relevant codepath? I assume you fetch the voxel (material index you call it?) somewhere and under some circumstances, you fetch from a chunk other than the one currently being worked on. Could you add a check for this case, put a print statement or something in it (just to have some code at all inside the if block) then set a breakpoint on that line. Then step through until you get to the voxel fetch and see how that call behaves.
938a1d No.14840907
>>14840683
When the ball goes down and she squints, it looks like she's afraid you're gonna slap her.
58b8d4 No.14840934
>>14840907
Treat em mean
keep em keen
f5a861 No.14840968
>>14840835
not necessarily 4, or just numbers. It can be a sequence of any characters. And I would put in 1488, but since I'm planning to sell the game that'd be sabotaging my own game.
I'm all for putting in an homage to various boards on 8ch, and that does include /pol/, but there are cheeky things and then there are straight-up stupid ones. That right there would be shooting myself in the foot, I think, especially in the current political climate.
>>14840862
yeah, I have been trying to step through it up to a relevant bit, but I'm not seeing anything anomalous. It's weird - I'm using recursion in the voxel getter function to make it be able to seamlessly traverse various chunks no matter what position you specify so that makes debugging just a tad more difficult.
Thanks for the suggestion though, I appreciate the thought, anon.
e8a43e No.14840998
>>14840907
Oh she gets treated rough. But it will become smoother over time.
>>14840756
It's half procedural. There's a base animation beneath all the logic.
f5a861 No.14841039
>>14840998
oh yeah, I'm aware. Still, though - a lot of procedural animation looks like garbage, and yours does not. so you did a great job, anon. Keep doing what you do - it's very impressive.
5425ed No.14841098
>>14841039
speaking of procedural animation, i found this yesterday on reddit, pls don't lynch me
700662 No.14841110
>>14841098
>Mentioning that you got it from reddit
What sort of attention whore are you? Why would you even mention you got it from reddit, do you want to be bullied?
Why don't you go back there.
5425ed No.14841120
>>14841110
i knew this would happen
i'm sorry everyone
58b8d4 No.14841123
>>14841098
THE UNENLIGHTENED MASSES
58b8d4 No.14841140
>>14841135
>calling others rotten to throw suspicion off yourself
I'm onto you, mother fucker.
028fb6 No.14841146
>>14841123
THEY CANNOT MAKE THE JUDGEMENT CALL
bbeb12 No.14841154
>>14841098
The filename is obvious enough for us to know where you got it from.
30004f No.14841157
>>14841135
no, bugman is rotten.
you need to keep up with the latest /agdg/ lore.
9d070f No.14841165
>>14841157
He said he's just another anon who got Rotten's project files after he ragequit, and if he's actually Rotten himself he's at least become less of a retarded sperg and he'd finally have realized Saeglopur with some shitty accents I'm too lazy to add is a terrible name.
f5a861 No.14841186
>>14841157
anon, them's fightin' words
f5a861 No.14841232
>>14841165
that's the situation, but honestly if just the fact that im working with this leftovers is enough to cause drama, I'll take it elsewhere - just say the word.
There's always cuckchan /agdg/ general but theirs doesn't feel as comfy.
5425ed No.14841268
this is why i never do the modelling or animation
f5a861 No.14841275
>>14841268
well, it's offcenter so I imagine that makes it look a little odd - but if you center it and make the entrance fit the actual membrane thing you got going, I'm sure it'll be just fine! The more you do the animations and models and stuff, the better you'll get at it. And honestly - that door doesn't look that bad - it just needs a little bit of polish on the frames, and a texture on top!
172289 No.14841291
Does this count as a game?
f5a861 No.14841304
>>14841291
but what is the ruleset? like, how do you win or lose?
b29018 No.14841369
>>14839282
>local bullies
more like the clean up crew
f5a861 No.14841371
>>14841369
never said they weren't! but they scrub the cancer via bullying, so that's what I'm referring to.
fdc1f9 No.14841436
>>14828598
this anchor thing is a great idea. and so that my post isn't in vain:
pic related is volumetric lighting used by unity in their adam demo
https://github.com/Unity-Technologies/VolumetricLighting
the main cool thing i'm interested in here are the pcss soft shadows these lights use.
why this isn't implemented in the main engine, i have no fucking idea
>>14840229
I often flagfag using the picture of a certain AI. I think tetradev was the first to call me shodanon because of that and i really liked that name so now i use it for all my dev related endeavours. But as >>14840413 said, people call you whatever they associate you with, so I react to diablodev and beelzebox anon as well.
some anons that shared their social media /youtube accounts so i used those for their wiki profiles. most don't seem to mind
>>14840526
>although in the wiki i'm stuck with my youtube name
jazz jackrabbit 2 is a good game, makes for a fine name. and you ain't stuck with it, doctard. you were AoEanon for a while after all.
cfb43d No.14841442
>>14841098
That's pretty cool.
b29018 No.14841628
>>14841371
so what's your story, again? I think I missed it. How did you and rotten meet? How did you get in possession of the priceless work he never showed anyone (outside of the UI which looked ok and the godawful model)? because lemme tell ya, you sure as hell inherited his posting frequency
e4de97 No.14841728
>>14841268
What's wrong with it exactly? When the character goes outside it works and looks pretty good. Is there some problem with going inside?
5425ed No.14841732
>>14841728
it looks ugly
and that's about as far as i can bother doing something out of my normal field
172289 No.14841751
>>14841304
I was thinking of making some kind of esoteric game that uses windows as the game elements.
For example you have a main base window, and you can click it to spawn bullets, and then you click the bullets to send them flying. And the objective is to block enemy bullets and hit their base window. You can resize bullets which makes them stronger against the enemy base, but they're more likely to collide with enemy bullets. And maybe you can spawn walls and resize/position them however you want.
5425ed No.14841756
YouTube embed. Click thumbnail to play.
>>14841751
maybe something like this shit?
2a973b No.14841815
>>14841756
>2007
Goddamn I'm old.
f485d1 No.14841842
Not entirely satisfied with this, but I'm going to move onto something else.
Pretty standard gimmick with this guy, make him run into a wall and attack him while he's stunned. He'll block attacks when he's not stunned.
5425ed No.14841850
YouTube embed. Click thumbnail to play.
>>14841842
you could just ripoff some earthbender moves
pillar of earth rising from the ground and dashing towarsd you is a classic
f485d1 No.14841854
>>14841850
More dissatisfied with the visuals and number balance than the move set.
I have other elementals I'm gonna use similar type moves for, specifically the ice one.
f5a861 No.14842040
>>14841628
added him on Steam at first for bullying purposes, then ended up just talking to him to see where it goes, which went on for a bit.
then someone has found his girlfriend's account around a month ago, and he nuked the entirety of his online presence save for steam, and asked me if I want his project files since he can't work on it, because "the project is compromised".
>How did you get in possession of the priceless work he never showed anyone
thats the thing though, there was barely anything of substance - just UI, the model, a barebones chunk system, and a completely botched octree where every single node was an individual UObject, which caused the app to run out of memory instantly. Also a wikidpad GDD.
so I got to work and now i'm chugging along. There wasn't much inherently wrong with the GDD beyond the obvious cursed runes bit, so I decided fuck it why not
>because lemme tell ya, you sure as hell inherited his posting frequency
guilty as charged - I'm gonna try and tone it down a bit, sorry anon
>>14841732
It looks a few notches better than I'd expect someone who has little animation/modeling experience to do. You did a solid job, just keep iterating on it and you'll do even better.
>>14841751
sounds interesting, i look forward to checking it out on whichever demoday you post it. Just make sure it supports multimonitor setups.
>>14841842
The pixelart is extremely comfy! I like it. As for the enemy - I'd add a predictable window in the pattern that you can squeeze through if you're cornered. Seems a tad difficult to avoid the volley.
>Pretty standard gimmick with this guy
Don't be afraid to do cliche things if they work - I don't think 2D games have done much in the way of originality for quite some time now.
>>14841854
Do you have any previous projects I can check out? The elementals are looking excellent, you're probably the best pixel artist I've seen on 8ch /agdg/ so far.
f485d1 No.14842121
>>14842040
The volley is pretty easy to dodge with abilities, I was just running around not using any though.
I've only really got this game, because I've been working on it for a long time. Here's a bunch of stuff I've done in the last couple months that I haven't posted.
f54c81 No.14842189
>>14842121
It still looks beautiful. I'm kinda wondering why you haven't released it already. It looks certainly good but I don't know much else.
f485d1 No.14842213
>>14842189
Too many features, and they're all kinda essential to how the game works now. I'm in the last leg of it, and I finally can work on it full time now without having to survive off commissions.
About to implement this junk which will be one of the final ticks off the feature list. Still have a lot of content to add as well, but that's all fun stuff.
5425ed No.14842713
>>14842040
>>14841268
i think this one worked out well enough
1aa426 No.14842786
>>14835644
From when are those videos? All those aspects he names have improved considerably, especially the STL
bbeb12 No.14842814
>>14842786
They're from 2014. But I don't think that " improved" C++17 STL would actually change the fundamental reasons that he doesn't use the STL.
1aa426 No.14842832
>>14842814
What are the fundamental reasons to avoid the STL?
bbeb12 No.14842864
HookTube embed. Click on thumbnail to play.
>>14842832
I embedded the whole talk if you want to listen to that part with a little more context. Basically a few reasons:
>He uses custom allocators for everything, STL uses standard allocators
>STL data structures allocate to the heap in an unpredictable way, which is bad
>STL data structures are "general purpose" for everyone, and not optimized with information you get from the context of the problem being solved
So in the end, using the STL makes it really hard to reason about the performance of your program. If you use a std::vector for example, the entire point of using such a thing is that:
>It expands the buffer (allocating to the heap) without the programmer having to think about it
This is good if you don't want to reason about the performance costs, etc, you just want an array that gets bigger so you don't have to worry about making mistakes or writing routines for everything. This is BAD if you need to have absolute control over when and how memory is being allocated, when you need to be able to avoid cache misses, etc. It's fundamentally a slower data structure compared to the non-STL way of doing it, just making an array, because of how it works and the problems that it is designed to solve. The guy giving the talk has problems like "how do I minimize and consolidate allocations on the heap, and how do I minimize cache misses each frame?". So the STL doesn't solve that problem, it is designed to solve a whole other set of problems.
http://assoc.tumblr.com/post/411601680/performance-of-stl-vector-vs-plain-c-arrays
1aa426 No.14842891
>>14842864
How does C deal with memory fragmentation?
Why did they choose to write the firmware for the F-35 in C++?
d8e980 No.14842923
>>14842891
Because it's a notoriously overengineered clusterfuck in every way, so of course they chose C++ and of course the F-35's software is widely regarded as terrible and horribly buggy.
367427 No.14842933
>>14842891
It doesn't.
You use malloc, which takes a size in bytes and returns a pointer to the allocated memory. The underlying memory allocation algorithm is determined by the compiler.
Either that, or you live in no-malloc-land like embedded environments, in which case you have to live with no dynamic allocation or writing your own allocator.
bbeb12 No.14842961
>>14842891
>How does C deal with memory fragmentation?
In C you just have to allocate large buffers so that the memory is continuous. So, you have to design your program in a way that reduces it. It's up to you.
>Why did they choose to write the firmware for the F-35 in C++?
F35 firmware has a completely diffrent set of requirements to follow than a video game. So really it isn't very comparable. Speed of the software is not important, but that it has no errors. It's a machine that costs many millions of dollars, so they can just give it a faster CPU.
Even then the C++ the the F35 uses is very different from the "modern C++" that is advocated at a place like cppcon, you could read the way they write it here:
http://www.stroustrup.com/JSF-AV-rules.pdf
f5a861 No.14842968
>>14842713
Yeah, it looks pretty good! The textures really brought it up to par. Are you planning to rework the doorway, as well?
>>14842121
I'm repeating myself here, but your artwork is stunning, anon.
bbeb12 No.14842969
>>14842961
Also to add to this most of the F35 code is written in C, not C++. So its kind of a bad example to give as someone choosing C++ over C since most of the firmware is written in C.
Source of this image is on page 38:
https://www.csiac.org/wp-content/uploads/2016/02/2010_03_30_SoftwareQualityReliabilityandErrorPrediction.pdf
1aa426 No.14842997
>>14842961
How exactly is speed not important for software that controls military aircraft?
8225e5 No.14843018
HookTube embed. Click on thumbnail to play.
>>14842814
Unironically pure C.
367427 No.14843019
>>14842997
As long as it runs fast enough its good enough. And if it turns out that it's running just a tad bit too slow, then putting in a faster CPU is a significantly cheaper solution than to actually try and make the software faster.
See also this ancient forum post about military just putting more memory in a missile, rather than fixing a memory leak: https://groups.google.com/forum/message/raw?msg=comp.lang.ada/E9bNCvDQ12k/1tezW24ZxdAJ
If you think people write good software in any sector, whether its video games, military or medicine, then you're sorely mistaken.
Everything is ducttaped together till its good enough. It's a miracle shit even fucking works as well as it does.
317a71 No.14843038
>>14842997
Probably because they can afford faster cpus, since the machine is so expensive in the first place, the cpu price won't make that much of a difference.
871ae4 No.14843052
>>14843019
Ah yes, one of the harder problems in aeronautical engineering: how to optimize memory capacity to minimize the temporal discrepancy between OOM and BOOM.
bbeb12 No.14843072
>>14842997
It costs many millions of dollars. They can just put in a faster CPU for a few thousand dollars if its not fast enough. This is pretty different from a console game, where the CPU can never be faster.
Speed is important, sure, but not as important as it is in a video game. C++ is still "fast enough" for this aircraft, just not as fast as C. You'll notice that they write most of it in C and some assembly, so I imagine the code that needs to be fast is the C and assembly code.
f5a861 No.14843097
>>14843072
That's hard for me to imagine being the case. Surely it's cheaper to get a squad of programmers to optimize code than buy CPUs for every single unit? It's hard for me to imagine an army of pajeets working for the military.
367427 No.14843110
>>14843097
They generally finish a product before putting into mass production. And when you put it into mass production, you have to buy a CPU for every single unit anyway.
e841b5 No.14843113
>>14843097
>Surely it's cheaper to get a squad of programmers to optimize code than buy CPUs for every single unit
not when you have to go through 300 hoops to make sure all of them have the needed clearances and that they are not traitors that will sell it all to Israel
bbeb12 No.14843134
HookTube embed. Click on thumbnail to play.
>>14843097
It IS optimized, most of it is C and you can tell that all of the really important stuff is written in assembly. Its not really clear what the C++ is for, but I seriously doubt it's the actually critical parts of the aircraft that need to run fast.
And yes, it costs them say $4000 to buy some expensive intel chip that can execute their C++ fast enough. Optimizing all of the shitty C++ code, in a codebase that is probably hundreds of thousands of lines long, is a lot more expensive.
>>14843113
Video related
f5a861 No.14843162
>>14843110
>>14843113
>>14843134
thanks for explaining, I love military stuff
527cfa No.14843314
>>14842961
I have a friend who does contract work for a branch of military, and he's pretty much said nearly the same: it's more about having no errors, and having redundancies; while being less about "as fast as possible", and more "just fast enough". They also have a few "practices" that are similar to gamedev, like pre-allocating all memory, but also differences like working with extremely limited resources (which could be a similar constraint if say you're developing for a switch or a handheld).
When I asked what he was working on, he said he was working on a team doing some low-level coding for a missile control system; which involved a lot of C/C++ (w/o the fluff of C++), and working with things like microcontrollers (embedded system programming).
9697c0 No.14843544
>>14842891
>How does C deal with memory fragmentation?
You allocate one giant continuous memory buffer, so you control fragmentation.
>Why did they choose to write the firmware for the F-35 in C++?
Mars rover and the vid the guy posted also use C++. Except they ignore pretty much every C++ feature and program in it as if it where C. They prob do the same with the F-35.
e841b5 No.14843579
Uber shader uses the R vertex color to decide which texture goes where.
Why the fuck this is happening?
Some of the generated meshes have 0.5 R in the corners, where I only ever write either 0 or 1
for (int v = 0; v < finalBorderVecs.Count; v++) {
tmp[v] = new Color(results[v] ? 1 : 0, 0, 0);
}
border.colors = tmp;
What could possibly be causing this thing?
e841b5 No.14843707
>>14843579
forget this post, it was the shading. I removed that and it now shows the correct values
86024b No.14844286
>>14843134
Woah, the outrageous thing here is not the bird blowing up an expensive jet, it's that the jet it's so ridiculously expensive to begin with.
046b90 No.14844574
>>14844286
Public sector + defense budget + lobbying by weapons manufacturers = inflated prices.
Remember early on when trump complained about the price of fighter jets on Twitter and was able to get a lower price?
bbeb12 No.14844647
>>14844286
The jet wasn't blown up, it was damaged. It's also supposed to be able to hit birds without any problems. The point of the video is that it was probably damaged by the SADF
b44860 No.14844882
Hey, can you cool pro indie devs give me advice on making this worse? I'm trying to illustrate the concept of pixelshit to a friend.
sorry for subjecting you to this
b954aa No.14844914
>>14844882
Gotta oversaturate the colors a bit more, also make the pixel aspect of the legs thinner than the rest of him.
6d2de7 No.14844923
35cfd1 No.14845720
I figured out how to spawn AI and get them to actually attack the player properly. I was confused about this since the documentation included with the editor doesn't explain any of this. Thankfully someone a few years ago posted a series of steps on how to get it to work.
Essentially the world needs a Navmesh to indicate where the AI is allowed to move around. If the navmesh isn't there the AI stands there braindead. Once the navmesh is there, the AI will aggro when it sees the player and will attack. To make it do more interesting things you can give it pathnodes to walk toward and set it to "patrol" so it'll walk around. Doing this allows one to make a fairly rudimentary level pretty easily. You can also change its allegiances like make it so it wants to help the player or make it attack other enemies. And you can change the weapons it's using. I was successfully able to spawn an ATC security enemy to attack the player.
I might play around with this a little and set up a few experiments like spawn 5 ATC security guys and see how they square off against 5 replicas.
35cfd1 No.14845793
>>14845720
I'm also trying to compile a list of possible locations for F.E.A.R singleplayer maps. I have a few in mind (mostly stuff like a library or something) and I've been compiling visual references of locations I'd like to remake in the level editor. I'm wondering what you guys would think would be a cool location to fight replicas in.
5425ed No.14846005
what's good software for painting textures onto 3d models that doesn't need a decent gpu?
i've tried substance painter before but it didn't work because my gpu is old shit
dd3331 No.14846041
I've been working on a 2D project for some time and am considering moving to 3D. How big of a leap is that? What should I expect if I do decide to change?
35cfd1 No.14846069
>>14846041
>How big of a leap is that?
It's very big Anon
>What should I expect if I do decide to change?
Expect your workload to become enormous and expect having to learn a ton of new software. There's a reason why most one-man indie games are 2D.
58b8d4 No.14846083
>>14846041
>How big of a leap is that?
It's more work creating assets*, it takes more math and the bugs are funnier. 3D engines can also feel kind of autistic in insisting that things be done in a certain way.
*not really true for high quality ones.
5425ed No.14846112
>>14846041
the good: you can reuse a lot of stuff due to skeletons, saves you the work of having to redraw nearly the same thing in order to animate a sprite
the bad: it's harder to do it right and make it look good
046b90 No.14846161
>>14846041
In terms of programming, not much. Your engine probably handles most of the logic for 3D. The real hurdle is 2D to 3D art. Modeling, UV unwrapping, texturing, animating, These are your new challenges.
d14655 No.14846342
>spend 2 days on trying out various settings for baking light maps
>they're all shit
>finally realize import settings for meshes is missing "Generate lightmap UVs"
ec4a08 No.14846474
>>14842814
>goto listed as a plus
what a shitty meme image
>c is always faster than c++
>c++ is bad mkay
i should have expected /agdg/ to be full of retards but seeing all this still suprises me
06225d No.14846496
Guys sorry for the retarded question, but how are 2d sprites or 3d models + textures usually made?
Let's say I'm fucking bad at drawing by hand, can I still make okay sprites?
I kinda want to make a game, but I'm unsure whether I can make something that doesn't look like shit or not.
434581 No.14847049
>>14846496
Pixel art is easier to make but you still require similar fundamental abilities to make decent art with it. Same thing with 3D, and it may even require more skill depending on what you're making, not to mention 3D animation is a very technical and specialized skill. Pixel art is easier to accept even if it's kinda shitty though, and the larger the pixels the less detail you need to put in.
>I kinda want to make a game
Give it a try first, the game comes before the art.
574a8d No.14847225
>>14846496
Like >>14847049 said, art comes second because games are firstly interactive.
Code it first. You can always find someone to do your art for you, or you learn to do it yourself.
But first, hack something together.
>>14835606
>OpenGL is almost entirely written in C with no C++
OpenGL isn't "written" in anything. OpenGL is a C API specified by Khronos. Drivers and software implementations implement this API however they see fit. You even have Python and Haskell binds for OpenGL!
OpenGL is nothing more than a set of function calls your videocard provides a hardware implementation for, feel free to use a software implementation of OpenGL if you don't care about speed or have specialized needs.
So I don't know where you get this "they ignore everything C++ adds" bullshit?
Also: I have no idea why people aren't using Khronos, but I've been out of the graphics games for a while now while I was never truly in it. I admit
>>14835487
My advice for you is to learn how to code anything at all. If you've ever given any programming language a shot less than a year ago? Go use that same language to build something – anything at all! If you have no experience what-so-ever? Pick an engine, learn its language. Don't roll your own engine, stand on the shoulders of giants instead. I'd personally avoid Unity because licensing, community, royalties and not being able to fix your engine if shit breaks.
Stop thinking about C++ as merely "a superset of C" and start thinking of C++ as a different language. If anyone in this thread is ever using C++? Remember to never again write C-style code with pointers until you have 90% of your features and are optimizing for the platform. C++ remains a very pragmatic language with a lot of cruft. You better learn how to handle a debugger if you ever start using C++.
>Why should I use C++ then?
There's one thing that C++ is good at that other languages aren't.
Whatever crusty-ass code you wrote back in 2006 with performance hacks? It still works.
Whatever performance demands you require? It enables you to optimize effectively.
My god, I gotta write a C++ course for you guys. Every couple months I come in here and start rambling about C++ and why you're all doing it wrong. Would you guys pay to watch a pragmatic/hands-on C++ course and perhaps pay me to tutor you?
I might also do small commissions. I dunno.
5425ed No.14847236
>>14846496
if you're bad, you're probably gonna end up doing placeholders like me
t. anus house
ignore the stuttering, i swear to christ this immediately goes away when i stop recording. smooth as silk when obs isn't doing its thing
046b90 No.14848740
>>14847236
Shabeen shaboom, you are now known as the man with the anal room.
ec4a08 No.14849664
>>14847225
no one here wants to learn proper c++
they want to spout memes they once heard about how c is the best language
note how nearly everyone here is using unity/c# anyways so they are uniquely unqalified to say anything about c++
>inb4 ten replies saying "im an embedded systems dev working in c"
056064 No.14849690
>>14849664
no one here wants to learn proper c
they want to spout memes they once heard about how c++ is the best language
note how nearly everyone here is using unity/c# anyways so they are uniquely unqalified to say anything about c
>inb4 ten replies saying "im an embedded systems dev working in c++"
:^)
9697c0 No.14849722
>>14847225
>OpenGL isn't "written" in anything. OpenGL is a C API specified by Khronos.
[citation needed]
6935e9 No.14849775
>>14849725
>>14849725
>>14849725
>>14849725
>>14849725
i may of forgotten the header, and don't have pass to delete it
574a8d No.14850831
>>14849722
You are such a dumb fuck with zero reading comprehension. I guess WebGL is implemented in Javascript, amirite?
574a8d No.14850851
>>14849722
You know what, let me just post this:
OpenGL isn't implemented in anything. It's an abstract concept, a perfect machine. There are drivers and software implementations that give you a valid OpenGL 'context' in which those C function calls make sense. The provider of this OpenGL context effectively implements an OpenGL 'model' that works according to the specification and selected 'profile'. Yes, the 'standard' OpenGL API is C, but there's C++ bindings and even Haskell bindings available as well. Including Javascript bindings (AKA 'WebGL'). Does that mean WebGL is 'implemented' in Javascript? No it doesn't. An API isn't implemented. An implementation FOR an API is "implemented".
My god, you need to learn some proper terminology.
Here, go read this: section 1.2 through 1.2.2:
https://www.khronos.org/registry/OpenGL/specs/gl/glspec46.core.pdf