94480b No.14739428
panic time, it's demo day tomorrow 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: >>14713738
QUARTERLY DEMO DAY SCHEDULED FOR MAY 5TH
Polite reminder that the wiki exists, you are encouraged to contribute to it if you can (even if it's just your game page)
e14015 No.14739439
>>14739428
Threadly reminder that Source is the only engine worth developing in.
2817ee No.14739446
>panic time, it's demo day tomorrow edition
I sure am feeling it.
4c6052 No.14739457
>>14739439
Academic reviewed evidence?
5618ff No.14739476
Crunching stuff for Demo Day. Somewhat appearing playable, although it's still very simple in concept. Dunno what I'll get finished today since I still have a slight backlog. Hardcoding AI and other things just to make it on time, but at least I can go back and clean it up later.
2817ee No.14739494
What's procedure for tomorrow? New bread for the demo? What time does it start?
eb4ad8 No.14739500
Remember: The deadlines we impose on ourselves are less directly stressful than the ones that would be imposed on us in a corporate setting, but no amount of corporate stress can match the sleep-ruining existential dread you'll get later from missing your own deadlines.
5618ff No.14739503
>>14739494
http://8agdg.wikidot.com/demo-days
I don't know if anyone's organized for it this time. It was pretty rushed in February. Perhaps we should start an organization bread in /agdg/ to keep it organized? Then prepare some demos who get in before the actual thread post time.
e32e52 No.14739506
Well, I have a goal of sorts.
It is to finish a font for my game
by demo day
on demo day
by the end of the week. I only have lowercase characters, lowercase accents (which is just copypaste work, really), and cyrilic letters, half of which are just Latin letters shuffled around. And now that I know what style I'm going for and all the intricacies of the software, three days seems doable, even if I'll be out of the house for half of tomorrow.
95bc71 No.14739519
POST PROGRESS
POST DEMO STATUS
>>14739476
reminds me to do one of these
2817ee No.14739539
>>14739503
I'd do it if I wasn't crunching like a madman.
>>14739519
Trying to put some final touches on shit that might be showstoppers for people testing. Going to try and deploy a demo this evening and have some friends test to make sure it works on a fresh install.
30866a No.14739555
>>14739457
>someone still uses proofs based on my shitty logo design
5dbf59 No.14739564
How many computers need software rendering? Is it worth it to add a software renderer in your engine in case OpenGL isn't usable?
I'm not trying to make a big game or anything, I'd just like if my games were playable for everyone no matter what kind of toaster they're toasting from.
2817ee No.14739567
>>14739564
>2k18
>software rendering
4c6052 No.14739580
>>14739555 (checked)
yoru bird is of good, makes good eggs da
>>14739567
>2018
>pulling a Graf Zahl
2817ee No.14739598
>>14739580
>removing a fully functioning software renderer is equivalent to implementing one in a new game to support 20 year old computers
Just stick with OpenGL 3.3, it's fine.
210dcd No.14739666
>>14739494
just updated the dd page with a bit more concise info. it should be clearer now
02d23f No.14739691
>>14739580
What a cuntbag. Also doesn't know a thing about Linux apparently. It's people like that that hold software back, ironically.
dd0b00 No.14739816
>>14739564
If it's easy to add, just dew it tbh. Otherwise it's probably a waste of time, almost everything that would be used for games has at least a shitty integrated graphics card. Funny that you mention it, I had a stupid crackpot and eternal engine dev idea that might require software rendering.
868622 No.14739917
>>14739428
Erni's Iceberg will be there niggas
03563b No.14740295
whats step 1 to just like make game
e32e52 No.14740308
>>14740295
Come up with an idea for a game that you want to make, preferably something that hasn't been done, hasn't been done in a long time or isn't being done often enough in your opinion. It provides great motivation to actually make a game.
Then come back, lurk, learn through osmosis and direct questions, and just fucking do it.
e86382 No.14740311
>>14740295
Figure out what you want to make. Then drop that and make something simple. Make several simple things before you start trying to make want you want.
03563b No.14740318
>>14740311
whats considered simple
mario 64?
e32e52 No.14740325
>>14740318
Pong, usually. A simple top down shooter. Simple one screen platformer. Arkanoid.
I personally don't agree with this way of doing things, but most other anons do so I won't talk you out of it. I just find it hard to learn from a project I don't care about.
d1a460 No.14740351
>>14740318
>mario 64?
Haha no. NES Super Mario Bros. would be a good second or third project. If you are a newfag definitely start with 2D.
>>14740325
Motivation to make really stupid basic shit may be in short supply, but if you try to dive into something complex you're setting yourself up for failure. If your goal is to "make your dream game" and you don't find devving in and of itself entertaining, you're probably just going to burn out 5% into your complex project anyways.
03563b No.14740375
>>14740351
>>14740325
yeah okay now how do i learn programming
>attempted c++ via reading book but i gave up, mostly because it taught visual studio over an engine and i didn't understand much
>followed guides but once i finished something i realised i knew jack shit
programming may not be for me though, I'm way too bad at math
e32e52 No.14740389
>>14740375
Once again, I'm sure anons will disagree with me, but I just type into a search engine [Engine/Language] how to implement [feature] and read.
I assume you'll use some sort of existing engine, so you can start with their existing tutorials and documentation, or download free example games and read the code, fuck about with it and see what changes.
>programming may not be for me though, I'm way too bad at math
It's more logic than math, really. You just need problem solving. Figure out how to implement a feature in the most efficient way possible, and if that's too hard, just implement it in a way that it would function. Math is useful, and at times you can't escape it, but once again, most times, you'll find tutorials that already tell you what to do, step by step, so the math will be minimal.
83dddf No.14740390
>>14740375
most of what you learn in math in college (all the way up to physics) is just a formal process, you can even teach a nigger if you're dedicated enough. Try taking some actual math classes and get a feel for it.
72cfff No.14740395
Good luck for tomorrow anons. I've got no updated demo for this demo day because I fell down the hole of re-implementing things and feature creep.
03563b No.14740401
>>14740390
>Try taking some actual math classes
83dddf No.14740406
>>14740401
They're great for making people comfortable with math if they're too stupid to self-teach
03563b No.14740411
>>14740406
the issue is that I fall asleep too fast when doing math
83dddf No.14740419
>>14740411
If you can't focus on something difficult for even 50 minutes at a time, you need to take better care of your body
d1a460 No.14740421
HookTube embed. Click on thumbnail to play.
>>14740375
You really don't need to be amazing at math to do programming, it's mostly about logic and being able to visualize how things work– then translating that mental logic to code logic. If you aren't a dumb motherfucker and know how to use a calculator you'll be fine. If you are a dumbfuck like me start with something simple like Python or Lua. Godot is a great place for beginners to start because it has a robust 2D toolset that uses a scripting language nearly identical to python so you can have fun seeing results immediately with simple code.
Vid related is what I started with ages ago.
2817ee No.14740422
>>14739666
Nice trips.
I can't do with any more crunching, I've built my demo and uploaded it to mega, bring on the demo day I say.
03563b No.14740424
>>14740419
I draw for 4 hours straight whenever I decide to git gud
1e4948 No.14740668
>>14740424
people over here sometimes dev for 12-16 hours straight.
5dbf59 No.14740687
>>14740668
That's me, plus doing it again immediately after waking up, and repeating for several days until giving up completely for a month
>>14740424
4 hours of drawing isn't really that much.
Also in regards to learning, you might want to consider C# instead since that's supported by Unity and Godot.
03563b No.14740695
5618ff No.14740699
>>14740687
>>14740668
I have a pizza in the fridge, but I'm too busy to eat it. I'm so close to a break in which I can take some time to watch some bad anime and eat, though. Been working mostly nonstop for about 8 hours now.
7b488d No.14740703
One of the best ways to learn is to loosely recreate something that already exists. If you don't know how to implement a certain feature, study up on it until you do. Recreating something also means you don't get bogged down in the creative phase and it's always nice to have a relatively clear picture of what you're end product will be.
9ec4c5 No.14740721
03563b No.14740730
>>14740721
yeah both are memes
1e4948 No.14740732
>>14740699
>>14740687
doing good, bros. Take notes, 03563b. I guess the answer to your question
>>14740295
>whats step 1 to just like make game
is to become passionate about making vidya, do all in your power to learn and spend all the time you can on it.
eb4ad8 No.14740764
The colors used for the demo day art are:
>Blue Winter
>Green Spring
>Yellow Summer
>Brown Fall
Why not use red for Fall, which is the color of changing leaves, and also the final color of the /agdg/ logo?
71159f No.14740785
>>14740764
>brown fall
Looks like the joke went over your head.
188428 No.14740788
>>14739519
LOOKIN_GOOD.jpg
eb4ad8 No.14740792
>>14740785
I don't know anything about Gundam, sorry.
210dcd No.14740793
>>14740695
>says the newbie asking for advice
any engine has its flaws, and it's strengths.
unity is just meme'd on bcs it's so easy to use the assets on the asset store to just shit out a game, so it gets a bad rep bcs of the damn pajeets, but it's actually a really solid engine.
For example, unity has the best docs of any mainstream engine, while UE has some of the worst docs of any mainstream engine.
UE has some of the best visual programming tools for C++/shaders of any mainstream engine, and unity has [experimental] visual programming for only shaders.
>>14740703
definitely this
>>14740422
that got me thinking.
maybe we should just make the demo day thread on /agdg/ early; as I'm definitely not going to be up at 12am tonight.
This is so anons don't need to stay up to post their demos, and also since that thread will be linked in the /v/ demo day thread; the sleepy devs won't have to worry either.
>>14740424
i dev'd for 36 hours mostly straight the other night (took time to eat/coffee), 4 hours is nothing.
slept for a good 10 hours after that though
>>14740764
good idea, u should do it
9a3db4 No.14740801
>>14739500
You're killing me.
5618ff No.14740806
>>14740793
>maybe we should just make the demo day thread on /agdg/ early; as I'm definitely not going to be up at 12am tonight.
Yeah. I'm not sure what exact time Demo Day will be posted in /v/, so it'd be really wise to have some anons organize it now. If no one else does, I guess I'll take a crack at it. Any takers?
1e4948 No.14740843
>>14740806
go for it. Can't wait to play all your games
188428 No.14740864
>>14740793
What if we made the thread now on /agdg/, and then just have Mark make a locked sticky pointing to that thread in particular?
Dunno
1e4948 No.14740872
>>14740864
I think we tried that 11/11 and failed somewhat. your call. I can sticky the demo day thread on /agdg/ but don't have such powers here.
5618ff No.14740877
>>14740806
Alright. I'll start an organization topic in /agdg/ once I finish off this leftover pizza of game development achievement.
210dcd No.14740882
>>14740806
as u know, i'm down, adding another (you)
make sure and read the updated guide b4 making
http://8agdg.wikidot.com/demo-days
>>14740864
pretty good idea, but as this anon mentioned >>14740872 it doesn't quite work as envisioned.
>>14740872
legit
>>14740877
kek, sounds good
1e4948 No.14740900
>>14740864
>What if we made the thread now on /agdg/, and then just have Mark make a locked sticky pointing to that thread in particular?
asked Mark, he'd prefer us to make the thread on 05/05 US time but he'll sticky whenever
210dcd No.14740910
>>14740900 (checked)
no worries
the only early part was making the demo day thread on /agdg/. as it'll be less of a rush, and hopefully more organized than past threads.
the /v/ demo day thread stuff will go as scheduled (on 5/5 at 12am).
1e4948 No.14740924
>>14740910
wow, we already have an early entry in the progress general
210dcd No.14741001
>>14740924
yeah, they seem like one of those "advertise my game only" anons (forgot the term).
not to mention they're happily endorsing reddit, fuckin' wew, newfaggotry at it's maximal height.
5618ff No.14741007
>>/agdg/31455
Prep thread is up.
210dcd No.14741012
5618ff No.14741018
>>14741012
Soory, I don't crosslink often.
13fde6 No.14741025
I managed to get levelling up working in my game for the most part. I've still gotta tweak some numbers, reorganize the code and implement stuff for HP and MP, but it's working like it should and I did it all by ripping off FF7's calculations because I have no fucking idea how to do any of this shit.
9ec4c5 No.14741033
>>14740923
>proper jews
>>14741018
use >>>/test/ to avoid bullying
1e4948 No.14741049
>>14741001
>shoot downvotes
It'll be a bannin if they don't participate in the demoday thread
210dcd No.14741066
>>14741049
kek. i'm pretty sure the bullying will make them leave, and especially so if they post in the /v/ DD thread.
oh boy will that be a laugh
71159f No.14741071
>started developing a shitpost Demo Day entry on a whim yesterday
>realized I didn't have enough time to finish it properly
Sorry guys. I'll try to finish it sometime but I have to practice artfagging a little more first.
188428 No.14741082
>>14741025
The great thing about FF games (5, 7, 12, and Tactics in particular) is that they've been meticulously combed through on Gamefaqs and usually have indepth mechanic guides that has all the formulas.
It's pretty easy to get a hang of how your stats should interact, once you're more familiar with math. Path of Exile is also a good source, sometimes.
fcce08 No.14741085
>>14740923
>Good jew
There is no such thing and have fun in the death camps when /pol/ leads the incels to victory.
13fde6 No.14741095
>>14741082
Yeah, JRPG autism is a powerful tool. It always shocks me to see just how ludicrously indepth the FF7 guides are.
5618ff No.14741137
>>14741025
It took me some time to wrap my head around how to make a nice exp curve for leveling amongst other things. One resource that was kind of nice for getting started for me was https://www.davideaversa.it/gamedesign-math-rpg-level-based-progression/
I'm still pretty new to the concept, but I was able to make an exp curve and a diminishing return stat investment algorithm with this kind of stuff.
13fde6 No.14741169
>>14741137
Oh shit, thanks Anon.
d1a460 No.14741180
>>14741085
> a rat snitch who basically ratted out Eddie for doing what Best Buy, Walmart and Amazon did before they became a thing.
> Almost from the beginning, Crazy Eddie's management was engaged in various forms of fraud. The Antars deliberately falsified their books to reduce (or eliminate) their taxable income.[8] They also paid employees off the books, and regularly skimmed thousands of dollars (in cash) earned at the shops. For every $5 Crazy Eddie reported as income, $1 was taken by the Antars. In 1979, the Antars began depositing much of this money - hundreds of thousands of dollars - in Israeli bank accounts. The Antar family skimmed an estimated $3 to $4 million per year at the height of their fraud. In one offshore bank account, the family deposited more than $6 million between 1980 and 1983.[3]
>more than $6 million
And Eddie Antar is supposed to be the good jew here?
eb4ad8 No.14741183
>>14741180
Remember the six million in tax fraud, goyim.
188428 No.14741265
>>14741137
One curve I like is X/(X+N). It ranges between 0..1 and never quite reaches 1. You can basically use it as a normalized function / multiplier against something else. It also starts pretty quick then levels off pretty smoothly.
188428 No.14741275
Holy shit, getting a working replay function was literally three lines of code.
So basically there are only 6 player actions possible - up, down, left, right, and two modifiers (item and push). This fits nicely into an enum of type byte. Every update, I increase the current tick by 1, and if it overflows, it just resets and acts as a rolling counter. However, if the tick matches the current value of the replay data, it reads the next index of the pair, interprets that byte as player input, and then jumps to the next pair.
In short, I can keep my replays as a byte[] and quickly validate level success/failure
028aab No.14741293
>>14739519
No demo this time unfortunately. Mainly because I don't have much "new" playable progress. Mostly just been playing around with some different art stuff for a new level and getting things moving on sound assets.
210dcd No.14741323
>there's a wireframe shader on the asset store
>$15
>wew
Shader "Debug/Wireframe"
{
Properties
{
_WireColor("WireColor", Color) = (1,0,0,1)
_Color("Color", Color) = (1,1,1,1)
}
SubShader
{
Tags { "Queue"="Transparent" "IgnoreProjector"="True" "RenderType"="Transparent" }
Blend SrcAlpha OneMinusSrcAlpha
Pass
{
CGPROGRAM
#include "UnityCG.cginc"
#pragma target 4.0
#pragma vertex vert
#pragma geometry geom
#pragma fragment frag
half4 _WireColor, _Color;
struct v2g
{
float4 pos : SV_POSITION;
float2 uv : TEXCOORD0;
};
struct g2f
{
float4 pos : SV_POSITION;
float2 uv : TEXCOORD0;
float3 dist : TEXCOORD1;
};
v2g vert(appdata_base v)
{
v2g OUT;
OUT.pos = UnityObjectToClipPos(v.vertex);
OUT.uv = v.texcoord; //the uv's arent used in this shader but are included in case you want to use them
return OUT;
}
[maxvertexcount(3)]
void geom(triangle v2g IN[3], inout TriangleStream<g2f> triStream)
{
float2 WIN_SCALE = float2(_ScreenParams.x/2.0, _ScreenParams.y/2.0);
//frag position
float2 p0 = WIN_SCALE * IN[0].pos.xy / IN[0].pos.w;
float2 p1 = WIN_SCALE * IN[1].pos.xy / IN[1].pos.w;
float2 p2 = WIN_SCALE * IN[2].pos.xy / IN[2].pos.w;
//barycentric position
float2 v0 = p2-p1;
float2 v1 = p2-p0;
float2 v2 = p1-p0;
//triangles area
float area = abs(v1.x*v2.y - v1.y * v2.x);
g2f OUT;
OUT.pos = IN[0].pos;
OUT.uv = IN[0].uv;
OUT.dist = float3(area/length(v0),0,0);
triStream.Append(OUT);
OUT.pos = IN[1].pos;
OUT.uv = IN[1].uv;
OUT.dist = float3(0,area/length(v1),0);
triStream.Append(OUT);
OUT.pos = IN[2].pos;
OUT.uv = IN[2].uv;
OUT.dist = float3(0,0,area/length(v2));
triStream.Append(OUT);
}
half4 frag(g2f IN) : COLOR
{
//distance of frag from triangles center
float d = min(IN.dist.x, min(IN.dist.y, IN.dist.z));
//fade based on dist from center
float I = exp2(-4.0*d*d);
return lerp(_Color, _WireColor, I);
}
ENDCG
}
}
}
028aab No.14741356
>>14741323
I always wonder how many sales those things actually get. It can't be much but if doing that makes money I can't really blame them.
210dcd No.14741370
>>14741356
it would be reasonable at something like $1 to $5, but that's about it; $15 for the equivalent of this code >>14741323 is just being a scumbag
2724a3 No.14741440
>tfw failed demoday deadline
I'll make sure to try your lads' projects
388c31 No.14741756
>>14740375
Because you don't learn programming, you do programming and learn by accomplishing tasks you set in front of yourself. I didn't learn jack shit as a teen because I wanted to grasp a skill and then use it, but it doesn't work like that. Then I started working on my game and learned animation, coding, sound design, etc. because I was interested and I knew what things in particular I had to learn.
5618ff No.14741766
Less than three hours until Demo Day topic goes up. Reminder you can drop your game at >>>/agdg/31455 if you're in a timezone that makes it really late and don't want to stick around to post it as soon as possible. I think my demo is Okay Enough™ and now I'm just touching up a few extra things here or there.
e6aca0 No.14741962
Got combined BlendShapes working. Video shows random animations to test it. The workflow from Blender is going to be a little bit wonky for animations but I think I finally nailed it down. I'm aiming to have a demo for 8/8/18. Good luck from me and Shinobu.
77a4ee No.14741975
6cc62a No.14741976
>>14741962
>those brief glimpses of horror when she momentarily realized her life is a lie
77a4ee No.14742006
>>14741976
kind of looks like "here comes the rape" to me
5618ff No.14742025
I'm uploading my buggy builds. It's a big primitive, but hey whatever. It's a build.
>>14741962
Blender bone + shapekey animation? If you have any resources on that, I'd like to see. Some of our models use shapekeys for face stuff, but doing animations in Unity for that while doing bone animations in Blender is kind of tedious and annoying and ideally avoidable entirely.
72cfff No.14742030
>>14741975
Don't try and replicate 2.5D FPSes through software rendering. Just use OpenGL and fake the no-pitch distortion. You'll get much better performance, save time and won't be designing features around outdated ideals.
e6aca0 No.14742042
>>14742025
I don't have any resources on it. I simply Googled how stuff works and read the docs. For me animating bone animations with shape key animations is not tedious. Just make an action for the skeleton and a shape key editor action for the head mesh. I like to work on one part of the animation at a time so I usually animate the skeleton first and then hop over to the blend shape animation. Never both at the same time. But I was referring to the wonkyness of the animation workflow for how Unity binds animations to meshes. I'll be writing up in my thread diary later about the specific technical aspects.
9d8442 No.14742061
Happy Demo Day everyone! Show me the games.
5618ff No.14742094
>>14742061
The official demo day topic will go live within the next 20 minutes. From there, hopefully, people will post their demos. There wasn't much coming in on the prep topic for that.
bbf526 No.14742125
>>14741974
Check Dood and Douk Nouk 2.5D source codes, fam, for the later see software rendering in EDuke32, because OG Build has it written in ASM. http://dukeworld.duke4.net/eduke32/synthesis/
Dood is 99% C so you do not need looking into source ports for that.
>>14742030
Fuck off Graf Zahl.
72cfff No.14742165
>>14742125
>killing support is the same as not designing your game around 90s limitations
>implying Graf has ever cared about performance
1f0ad8 No.14742383
>>14741293
That's a shame, I was really looking forward to playing.
000000 No.14742398
Can I be a part of agdg if i jus post what page of concepts, techniques and models of computer prog by peter n haridi my niggi in bread?
agdg you are my family. you give support, i tell you game is shit. fair. feed me (You), i feed you (Shit). fair. may b gud additional vices too
1e4948 No.14742633
>>14742126
>forgetting the anchor post
2817ee No.14742789
>>14742771
This looks amazingly comfy.
91a6e2 No.14742801
>Only have freetime to dev on the weekends
>Laptop harddrive fucking dies Friday
1814f4 No.14742831
>>14739567
>>14739564
I don't have a graphics card while using a potato pc. I'm likely just an outlier among a small group these days. Don't know for sure about that, though.
I'm developing my own pixelshit game that uses software rendering for obvious reasons.
51bcaa No.14742856
>>14741962
to make animated expressions, you just had to create a new animation asset and make blendshape keyframes, right?
5618ff No.14743561
>>14742633
Soory. I misread that the anchor post was for the prep topic but didn't get it was for /v/'s too until it was too late. Hopefully someone a little more attentive can handle creating this stuff next time. I only really took it since no one else did.
03563b No.14744088
>>14740687
>4 hours of drawing isn't really that much
I'd like to think it is because I'm not a NEET and construction work fucks my shit up
2817ee No.14744101
Working on that demo feedback. Feels good to have the game in the hands of nerds trying it out.
4c6052 No.14744288
Some progress on the first moonman doom map, I still need to work on the starting area since it looks a bit too plain for my taste, the village section is good enough for now. I have a bit different plan for the first map normally the first map tends to be easy but for this one it should be instead a "trial of fire" in order to test the players patience. I do play test it often in order too see how it "plays out".
028aab No.14744300
>>14742383
Past builds are always there. But on the plus side, I'm getting closer and closer to being done, and the next build will have things like music and a lot more sound. Its going to feel MUCH mire complete. So the next time you see new content it should be pretty exciting. Consider the lack of demo this time to be making the next one even better
203d3b No.14744364
What's the best emitter shape for dust clouds while walking?
Cone looks weird as shit
26e711 No.14744484
How do I find a partner to make my game with? I can't stay interested when I'm doing it by myself.
203d3b No.14744554
>>14744484
What can you do?
2817ee No.14744601
>>14744484
I wouldn't team up with anyone that is outside of punching distance.
03563b No.14744606
360fdf No.14744610
>>14744484
>How do I find a partner to make my game with?
its easier to make their game. then guilt them into helping you.
26e711 No.14744697
>>14744554
Everything. I'm a sysadmin, so I can handle website infrastructure/setting up git repo/etc. I can do the programming for anything that would be in the scope of a video game. I'm a relatively talented musician. My weakest point is art but I still manage; just takes me longer to create assets equivalent to that of a dedicated artist.
203d3b No.14744771
>>14744697
I suggested this to another anon in the last thread, but I might as well just repeat myself cause it's always useful.
You can find a lot of artists on twitter. Your best bets for that is to check #pixel_dailies (https://twitter.com/hashtag/pixel_dailies?f=tweets&vertical=default&src=hash) or #zbrush (https://twitter.com/search?f=tweets&q=%23zbrush&src=typd) hashtags.
In both of these you will find talented artists. At that point it's just really a question of looking at their profiles to see if they're fags or not and contacting them with an offer to collaborate 50%-50%.
If they ask you money for assets, GET THE FUCK OUT. If a non-professional artist asks you for money that's a huge red flag, and it will most likely not be worth it. Find someone that is clearly interested in working in games and create a mutually beneficial collaboration, there's enough artists that you will eventually find someone that will agree to that.
a6eec5 No.14744815
>>14744484
Friends. If that's not an option, do game game jams with people and see what comes of it. Don't bother trying to work with strangers otherwise, for most "passion" isn't a good enough motivator. Money is the glue that keeps a team running, but I'm assuming you don't have the means to pay someone.
Basically, if you haven't proven yourself, or try to work with someone who hasn't, it's not going to work out. I've wasted years trying to work with coders (as an artfag) and was left hanging every time. Same can go for artists however.
c5147b No.14744865
>>14744697
>setting up git repo
How is that worth mentioning? You sound like a pajeet trying to pad his portfolio. typical sysadmins
ab9b9f No.14745007
>>14742856
No. I animated it in Blender and imported that animation. Manually doing blendshape animations in Unity is for very specific purposes.
26e711 No.14745158
>>14744865
I mean self-hosting the site and server for the git repo; Github is a cancer.
26e711 No.14745186
>>14744865
half the people working on games in this thread dont even know what git or version control is let alone know how to manage a project
188428 No.14746844
Holy fuck goddamn am I tired and demoralized
388c31 No.14747196
>>14746844
Tell us what's troubling you, anon. Maybe I can give you advice.
2817ee No.14747215
>>14745158
>Github is a cancer.
>yfw
91a6e2 No.14747929
>>14745186
>half the people working on games in this thread dont even know what git or version control is let alone know how to manage a project
I'm sure your knowledge of a handful of git commands and webshit expertise heavily outclasses that of the anons in here who talk about such trivial things like managing CPU cache.
9370af No.14747975
>>14747215
Github is cancer, hence why anons tend to use gitgud or set up their own server. Which, I'm fairly sure the setting up a git server well and securely is a bit more of an involved process than half the people sperging out at the "I can set up git" anon realize. Unless you use gitlab, that's pretty retard-proof.
69a6d8 No.14748034
>>14744606
>>14744815
So if someone has the money to put together an amateur team, where would he find people who can help?
91a6e2 No.14748123
>>14748034
Gotta actually go out and meet people and see who's worth picking up.
2817ee No.14748140
>>14747975
>Github is cancer
[citation needed]
91a6e2 No.14748165
>>14748140
>>14747215
Don't you dipshits realize git != github? And that github is hard left, just like every other large software corporation?
2817ee No.14748180
>>14748165
>And that github is hard left
In what way does that impact using github to host your software? The moment they remove any software for any reason whatosever people will be leaving in droves, so it will never ever happen. I don't even use github but your autistic sperging is retarded.
91a6e2 No.14748182
>>14748180
>1 comment on the matter = autistic sperging
Projection at its finest.
90f7ca No.14748185
>>14748180
>The moment they remove any software for any reason whatosever people will be leaving in droves, so it will never ever happen.
It already did reterd.
a34c81 No.14748190
>>14748180
>>14748180
This happened many times.
2817ee No.14748192
90f7ca No.14748198
ed236f No.14748202
>Might get my desktop fixed soon
>Might not have to wait 4-5 fucking months to get back to developing Stalkan game and other useful shit like video games that aren't on my bing-bing-wahoo machines. Painting software, video editors, etc.
First thing I'm gonna do is start digitizing my concept sketches for the game. Then I'm gonna pump all my free time into development (Which isn't a lot but it'll have to do). Feels good man
>>14748034
There are websites for gamedev classifieds in which you can put out an ad or find an ad for a fellow gamedev looking for work. Additionally, if you look up gamedev meetups in your area, you'd probably be surprised how many there are. If you value working with someone you can have actual meetings with, this is also a good resource.
>>14744288
>Those digits
>Making Moonman Doom map
>Where you play as Hitler
Fucking heiled.
>>14748180
They've implemented a "hate speech" detection system and banned accounts using that tool. It's the reason why so many github alternatives popped up so suddenly.
2817ee No.14748206
>>14748198
>>14748202
Well fuck me, good thing I never used it.
a34c81 No.14748207
a34c81 No.14748214
But honestly, even with Github out of the picture git on its own is pure cancer. Source control should be something mundane and simple yet git turns it into an autistic over engineered clusterfuck. Plastic is SCM made specifically for game development and it works far better.
26e711 No.14749014
>>14748207
Thanks dad, I have a job and couldn't be here to defend my anonymous internet rep from the autistic hordes.
>>14748214
This is true. Very, very few people are willing to learn a new VCS just to work on a new game project though.
26e711 No.14749049
>>14747929
>implying basic programming or resource handling is difficult
>trivializing system administration
Write us an essay on how you'd go about installing and securing a Debian or CentOS server so that it's safe to be public-facing on ports 22, 80, and 443 at the minimum. Detail what git server software with a functioning and actually usable frontend you'd choose and why as well as the steps you'd have to take to install it. Then tell us how you'd secure it while not putting undue stress on those that have to access it by requiring something as extreme as VPN-only access.
Bonus points for picking one that actually gets used in enterprise development environments and not Gogs.
26e711 No.14749078
>>14749049
Pro-tip: you're going to have to have an nginx reverse proxy in front of the git server, so don't forget to talk about how you're going to set that up.
71159f No.14749087
>>14748214
>Source control should be something mundane and simple yet git turns it into an autistic over engineered clusterfuck
>recommends some proprietary shit instead
Nigger, Git is anything but overengineered. If you want something simpler there's always Mercurial and if you even think of suggesting the clusterfuck that is SVN, you deserve a paddling.
188428 No.14749105
>>14749087
What does svn do differently?
71159f No.14749167
91a6e2 No.14749400
>>14749049
>>14749078
Nobody is impressed with your ability to do grunt work and configure software.
26e711 No.14749428
>>14749400
There's a reason I have a job and you don't.
91a6e2 No.14749465
>>14749428
I work with people like you everyday. It's my main motivation for gamedev.
d58fb1 No.14749469
>>14749428
>asks for advice on looking for a team
>gets some bantz
>as is common with anything here
>gets all defensive and whiny
>can't leave well enough alone
>can't take a joke
>hyper autism about trivial shit
>anon wonders why he can't find a team
WEW
203d3b No.14749493
I really wish you guys would leave this shit out of /agdg/. Instead we have to deal with the threads becoming trash around every demoday
2817ee No.14749501
>>14749428
>sysadmin thinking he's hot shit
Top kek
>>14749493
Well then, how about you
POST PROGRESS
I have none because of MotoGP
26e711 No.14749507
>>14749469
>gets some bantz
>bantz right back
>somebody not in the conversation gets triggered
okay dad, please give me my allowance back
42e717 No.14749513
>>14749501
>amateur game maker studio "developer" thinks he's hot shit
Top kek
2817ee No.14749560
>>14749513
You should probably try harder. Maybe even post your game you (1) faggot piece of shit
f8bf6e No.14749681
>>14749507
no, fuck you joel
188428 No.14750060
>>14747196
implementing basic sprite sorting and batching in sfml
it doesn't come with any out of the box, and while i can do immediate draw calls, it starts to eat processor if I do more than 100 calls/frame, so I need to figure out a good way to batch stuff by texture, and even though I have XNA's source, it's still tricky to work through
4798c3 No.14750709
Finals are over so i'm able to post again. Still working on figuring out the BSP compilation bug but I have more ideas about how to find it again. Shouldn't be too long to figure it out.
72cfff No.14751149
>>14749049
>search "set up server for A+ security rating" and follow very basic steps
>worth bragging about
26e711 No.14751379
>>14751149
>A+ security rating
Not sure what this even means.
188428 No.14751398
>>14751379
A+ certification is a Microsoft standard
72cfff No.14751477
>>14751372
I literally posted the search term, you retard.
0189ad No.14751494
>>14751149
>>14751477
Ok ladies. You're both beautiful. Now knock it the fuck off.
2724a3 No.14751872
I'm going to implement dialogue localization with (semi) randomly generated lines.
>ex. "Go speak to the (count/mayor) in the (castle/cave)"
My idea was to register each variable in the dialogue as a table that way multiple dialogues can share the same variables. This will allow me to add and localize more dialogue variables if needed.
4e92df No.14751883
>>14751494
>using CuckCuckGo
Anon, really?
26e711 No.14751887
>>14751477
That has literally nothing to do with what I fucking asked you to provide you stupid piece of shit. Microsoft Security standards? Fuck off you silly niggermonkey. Answer the fucking question or accept that your dumbass has no fucking idea what goes into system administration - because you don't. You're a freshman in community college who is hoping to get a CS degree and is a prime example of Dunning Kruger. Literally hang yourself.
4798c3 No.14751943
>>14751933
good, we don't want you here
26e711 No.14751944
>>14751937
>tfw you can't keep track of IDs
autism
210dcd No.14751954
>>14751933
and
>>14751944
>using a vpn
>not even trying to act different
>it's not me guise i swear-zeys!
26e711 No.14751961
>>14751954
>>14751953
I've been here for years, you entry-level programmer shitters aren't going to scare me away.
26e711 No.14751973
>>14751963
I guess you missed the entire rest of the post where I said I could program, do music, and and some of the art.
Or the part where I never said anything about HTML and was talking about setting up a website and dedicated git server so we don't have to get assfucked by Github's shit policies.
Are you triggered because you can't do any of these things? Or let me guess, you know some Python?
388c31 No.14751975
>>14750060
sorry, friend, low-level stuff is out of my reach. I'm mostly a game design guy used to working with premade engines. Respect to you for putting in the work yourself, though.
26e711 No.14751979
>>14751974
>trying to imply samefagging
>my id has been the only one posting 1-2 sentence paragraphs
So you're totally braindead. Nice.
388c31 No.14751980
>>14751974
do not mention his name lest you summon the reddit demon
26e711 No.14751985
210dcd No.14751987
>>14751963
don't forget he's an official sysadmin!
i don't see why anyone wouldn't want to team up with such a cheery fellow!
>>14751979
>>14751973
>I've been here for years
>doesn't know a semi-recent meme
shoo sho back to reddit with u
388c31 No.14751993
>>14751983
legit though I believe it was decided in the demo thread that he's not worth mentioning anymore. exiled and erased from history, etc.
26e711 No.14751994
>>14751987
>here
>8ch
Not your thread full of official college freshmen.
26e711 No.14752002
>>14752001
None of course, dad.
210dcd No.14752013
>>14751994
>/v/ wide word filter
>common meme on many boards, for years
>i browse 8ch like the cool kids i swearzies!
i would just stop responding but i'm having a laugh
ed236f No.14752245
>>14751918
Most first year CS students also usually learn Python and/or Java and get an introduction to some form of C, too.
>>14751994
College freshman would be better than you. At least they have the willingness to learn and the humility to understand that they aren't god because they can setup a git repository. But please, keep bragging about being an OFFICIAL sysadmin like that's some big deal and shitting all over people who don't know ever little caveat of git and VCS systems.
203d3b No.14752502
>try out Unity hub
>I don't have to download an EXE for each separate beta version of Unity I want to try out
I can't believe it took them this long to actually implement this
203d3b No.14752663
>>14752581
That's just unlucky
Have you tried just deleting the installation folders?
203d3b No.14752674
So I baked a light over my scene, and that's working just fine. My character, not being static, wasn't included in the bake, which is also fine.
After that was done I added another light that was set as real time and was masked only to my character but while my character is being lit as he should he won't cast a shadow on the rest of my scene. This is even after setting the light to cast shadows and the ground to receive them.
Anybody know how to fix this?
203d3b No.14752703
>>14752674
Nevermind I rebaked and now it's fine even without the second light.
Fuck if I know why it wasn't working.
dc9967 No.14752720
>>14752674
>>14752703
Are you using Unity? Because light bakes in Unity are fucking fucked.
203d3b No.14752722
>>14752720
I am.
Dunno seems fine now
57f47b No.14753074
New model, 982 tris. Old on the right for comparison. I also made some props like the houses, the factory and the lamp post in my spare time. Finally got my shader working with shadows as well, and made another shader to handle the transparent leaves.
>>14742789
thanks
c61541 No.14753194
>>14753074
Is that second one dabbing?
72cfff No.14753211
>>14753074
The hollow eyes are bit unsettling and the legs a little spindly for gondolas in my opinion. If I remember right your old model had some trouble looking natural in motion when the legs were moving. The new one looks like it might be better in that regard, from those examples.
>>14753194
He's checking his watch.
>>14751887
>that I asked you for
Learn how to check IDs newfag and my post had nothing to do with Microsoft. >>14751494 even gave you a screenshot of results, which you still missed. Your need to wave your dick around is really pathetic.
79b2f4 No.14753654
>>14753074
I like the structures of your Gondolas, but I'd rework the eyes of the new model, they look like they're of the netherrealm. It might be the small highlight on them, but they look slightly unsettling. Far better than the old model though, but it can still use a mouth.
>>>/gondola/3 is perfect for reference.
8fd860 No.14753662
>>14753074
Shouldn't the eyes be protuberances instead of holes?
9a3db4 No.14754338
51bcaa No.14754581
>>14740295
just
Does anyone know why the GPU profiler is useless unless I am recording? (It just randomly decides if it will accurately GPU load or not on a whim, unless I am recording, then it will behave)
Or how to make OBS recording size match the stuff I am actually recording?
4798c3 No.14754787
I decided that there is absolutely no way that I can figure out the bugs in my BSP compiler without being able to visualize the compilation, so I have been working on a "Debug View" that renders the polygon lists as the BSP compiler splits them into subtrees, so that I can fly around a 3d model of what the compiler is doing and identify what is making it do the wrong thing.
2817ee No.14754861
Working on polishing my levels, currently trying to make the first level a bit faster and smoother. It now also uses the player's avatar instead of just a random character. Going to add some voice acting as well to make it more fun.
188428 No.14754893
>>14754787
The BSP stage is just to optimize the raw geometry by pruning away impossible faces, etc, right? What bugs are you even hitting with that?
188428 No.14754936
http://www.arl.army.mil/arlreports/2015/ARL-TR-7333.pdf
Found a really in-depth article on sphere and disc point picking
87af78 No.14754961
>>14753194
how dare you
>>14753211
>>14753654
>>14753662
>>14754338
>tfw you've been working with a version with your own creative liberty for so long you forget what the original was
The hollow eyes thing was me trying to make the gondolas as creepy as possible for that stupid ocean-island gondola video. I somehow forgot that was my choice.
But don't worry though, if my current ideas happen, there will be swappable parts and body sliders and also gameplay
Have a temporary vanilla gondola for your time.
>>14754787
neat
4798c3 No.14755013
>>14754893
No, that's usually called the VIS stage (in source engine at least), which happens after. The geometry needs to be organized into a BSP tree to do fast visibility checking in order to compute what is and is not visible for pruning to occur.
The issue has to do with mitigating invalid geometry, which cant be easily explained unless you have a deeper understanding of how a BSP tree is constructed. Short answer is "my polygons are being sorted into the wrong subtrees"
cc55ef No.14755278
>>14742398
I've just read the Preface of Concepts, Techniques and Models of Computer Programming.
How far do I have to read and do the exercises until I can make Cave Story 2?
cc55ef No.14755285
>>14754961
Changing the color or texture on the inside of the hollowed eyes would probably be good. Like a gradient from the brown to black rather than just brown then black at the back. Grey saturation could also be good.
210dcd No.14755359
>>14754961
>me trying to make the gondolas as creepy as possible
spot on with the whole primal fear thing. hitting on the points of deep inset eyes, with a lack of being able to see the whites of the eyes; quite creepy.
188428 No.14756693
>Try to generate random points on a quad
>Left is trying to use barycentric coordinates with 3 targets, which makes it obviously clumped
>Right is splitting the quad into two triangles and using barycentric within the single triangle, which works
My only concern is how to define the 4 points within the quad. Right now I have it as rand(p0,p1,p2,p3) which works, but it is sensitive to the order. If I don't define the points in a specific order, the bottom happens. But at the same time, I don't think my random function should card about triangle winding order
584b9b No.14756712
>>14739580
Graf did nothing wrong. If you wanted to have a modern source port you shouldn't whine about not having a software renderer.
188428 No.14756716
>>14756712
The last time I saw software rendering as an option was in Runescape like 6 years ago.
Pardon my ignorance, but is it using the game to manipulate the texture data, rather than the graphics card/
188428 No.14756724
>>14756693
Oh nevermind, instead of picking two triangles, I made it pick between 3 random points from the specified quad
188428 No.14756752
ofuck, never mind. While it works per triangle, it's not uniform because the triangles have different area
e6aca0 No.14756788
Finished pass 2 on her eyes. I'm very satisfied with how it turned out. Hey pupils stay constrained within her eye socket view and when she squints her lower and upper eyelids. Except when she is blinking. First pass for random eye movements and blinking. I think I need to synchronize the random eye movements.
>>14753074
Woah that turned spooky real fast.
9a3db4 No.14756793
>>14756788
anon is this a porn game? Your work is becoming really good and, sorry if this feels judgmental, I dn't want to see it wasted for such lowbrow ends.
90f7ca No.14756796
>>14756793
Every game is a porn game if you mod it.
e6aca0 No.14756867
>>14756793
Eventually it might have porn but no not primarily a porn game. Which will be difficult to differentiate in the weeb crowd. But I don't really care. It's going to be free anyways.
188428 No.14757138
There. Made a few random helpers.
>Random.NextPointOnUnitSquare()
>Random.NextPointInUnitSquare()
>Random.NextPointOnUnitCircle()
>Random.NextPointInUnitCircle()
>Random.NextTriangle(a,b,c)
dead74 No.14757156
>>14756788
blinking distorts her cheeks, but the blink motion is pretty good, your bottom and top eye lids shouldn't move separately in that same way in the first vid, but it mostly just looks strange how they drop when her eyes hit the top.
>>14744101
I didn't see this for demo day, what's with the hipster low poly, or nupoly shit.
dc9967 No.14757231
>>14757156
>I didn't see this for demo day, what's with the hipster low poly, or nupoly shit.
Game for kids and I buy all my art.
dead74 No.14757265
>>14757231
>games for kids
What like flash games?
>i buy all my art
I hope it's cheap then, at least it's consistent.
dc9967 No.14757287
>>14757265
>What like flash games?
No just a (primarily) mobile game for kids around 5-8.
>I hope it's cheap then, at least it's consistent.
It is. It's commonly used assets in a lot of shovelware out there. I just don't have the time or skills to produce art of my own at the pace I need. Also, even if a bit cynical, kids don't care, they just want it to look fun.
03563b No.14758071
>>14756788
how do i make anime models like you :(
188428 No.14758266
Managed to declump the point picking for quads. Basically instead of flipping a coin to look at each sub triangle, you calculate its area and then roll a weighted dice.
The only problem here is that I need to calculate the area each time, which is fairly cumbersome. I can use Hero's Formula to do so, but this requires the length of each side, so at best I'm looking at 6-8 sqrt calls per point picked.
f5eafc No.14758757
>>14757156
Good catch for the eyelid blendshape. Also the eyelids is just a debug animation. Its not an actual animation id use.
>>14758071
Pick up Blender and practice every day for a year.
188428 No.14759021
>>14758266
So I think I found the fastest way to generate a point from 3 points, given my case.
Triangle point picking is pretty simple. Using barycentric coordinates, you pick one point as your origin, and then treat the other two points as axes, then generate values U and V, such that (U+V < 1.00) and create a vector from that.
Quad point picking is basically the same, except to keep it uniform, you want to weight it so that ABC and BCD are chosen based on their areas, and then do the triangle point picking above. The hard part was finding the triangle's area from 3 arbitrary points, in a fast way that didn't involve sqrt calls. Since a triangle's area is defined as (1/2 * BH), you have to find the base and height. The base is any line segment, but then you can use the dot product to determine the height (a point perpendicular to the line segment).
https://en.wikipedia.org/wiki/Distance_from_a_point_to_a_line#Line_defined_by_two_points
What's really cool is that the top part of this formula is actually the BH part already - you just need to take the numerator and divide it by 2 and you get the triangle area. It was not intuitive and the whole issue took me about an hour to solve by pen and paper. But the plus side is that it's about 16x faster for me to do quad point picking now.
My game doesn't even need it, but I did want to put it into my engine so there's that
ea9ee3 No.14760077
>>14757156
>I didn't see this for demo day
>>14742672
hurrrrrrrrrrrr
97a3b8 No.14762322
How do I do this without using ugly blocks of if statements? Is there another way to do this?
f5eafc No.14762360
>>14762322
Looks fine to me. The logic looks too simple to further simplify. I dislike your coding stance. Far too many new lines for my tastes. I make my code more compact and I don't new line before and after my squigglies. I feel like its a very pajeet thing.
90f7ca No.14762367
>>14762322
>Is there another way to do this?
Temperature ? temperatureSample.r; < 0.7 ? (temperatureSample.r; < 0.3 ? Biome.temperature.Cold : Biome.temperature.Temperate) : Biome.temperature.Hot;
But really, it's not that hideous in the first place and not really worth worrying about after it's written.
97a3b8 No.14762397
>>14762367
I've just seen too many YandereDev "if statement" memes to the point that I'm paranoid about using if statements.
6e661a No.14762437
>>14762322
it's fine. I personally would use even more new lines and more braces just to make it more readable.
4798c3 No.14762443
>>14762322
Nothing wrong with the if statements, I am more concerned with the magic numbers. Maybe #define them to some more descriptive names like HOTNESS_THRESHOLD or TEMPERATE_THRESHOLD so that you can more easily change the values.
dc9967 No.14762450
>>14762443
I'll second this. Personally I don't use curly brackets on single line if-statements but that's just a question of style.
2973b5 No.14762456
>>14762397
nothing wrong with if statements, but if you're checking a single variable for multiple cases you should just use a switch statement.
And pre-mature optimisation is cancer, just write the fucking thing first and get it working, then make it look nice and remove redundant steps.
97a3b8 No.14762460
>>14762443
C# can't do that, but I'll do something similar.
4798c3 No.14762469
>>14762456
switch statements only work when checking one variable- you cant use it to check if a variable is within a range of values. In this case the most optimized code is what was written.
>>14762460
>he doesn't run his C# code through the C preprocessor
You can use global const variables to do this in C# I believe. Although I'm not being that serious about running your C# code through the C preprocessor, there isn't any reason why that wouldn't work- after all, the C preprocessor should be thought of as a general purpose text processing tool, and not something that can only be run on C programs.
97a3b8 No.14762489
>>14762469
The C preprocessor can be run on any kind of text?
Could I get this working in a word document to implement a really shitty resume autocompletion?
dc9967 No.14762499
>>14762469
>>he doesn't run his C# code through the C preprocessor
Never heard of doing this, can you share something, some example maybe? Sounds pretty interesting.
4798c3 No.14762541
>>14762489
>>14762499
I don't know how it's done off the top of my head on all compilers, but in gcc using the -E option will only run the C preprocessor on that file.
https://gcc.gnu.org/onlinedocs/gcc-8.1.0/gcc/Preprocessor-Options.html#Preprocessor-Options
You can of course script this into your build process, I mean as long as when you hit "compile" on your IDE you can control what commands are being run, and so you can use gcc to preprocess all of your C# files before they actually get compiled into bytecode or whatever.
Maybe this sounds like a kind of weird idea but it is realistic and it will work, its not really something that people actually do but it's something that people know you can do, I heard about this idea from a friend after I was complaining that Java has no preprocessing tool like C does.
dc9967 No.14762555
>>14762541
Cool anon, thanks. Always interesting to hear about new stuff like this.
9a3db4 No.14762567
>>14762322
Guy you don't need brackets for single statements.
if (true) Temperature=whatever;
else if (true) Temperature=meh;
else Temperature=0;
At least it works on C++. Can't imagine why it shouldn't on C#.
90f7ca No.14762597
>>14762567
Some places having coding standards that mandate curly braces everywhere because at some point pajeet might turnif(something)
doAThing();
intoif(something)
doAThing();
doAnotherThing();
so a lot of programming books perpetuate the style.
9a3db4 No.14762646
>>14762597
If you're working for yourself and you're not a moron you'll be fine though. Anyway, I was just suggesting a shorter way of coding it, since he asked.
6001ef No.14762827
I need some help with c# and Unity, this kind of goes beyond what I currently know and I can't find much info about it.
At the moment I'm initializing a bunch of data from my animations every time my game is run, these are saved into a Dictionary that pairs a Hash (an int) and a CurrentAnimationState, which is an enum.
I tried building my project, unfortunately all of this data is currently being taken from "UnityEditor" and because of that it can't be exported with building.
I understand that Unity's Dictionaries cannot be serialized.
What should I do instead? How do I export this data to an external file I can use after building my game?
d43f42 No.14762900
>>14741323
These things exist to punish artfags for not being able to program
2817ee No.14762928
>>14762827
Post code.
>>14762646
Always assume future you is a moron. Maybe not applicable in this case because if you're a pajeet-level moron then you're already screw. But designing APIs and stuff, always assume that even future you is a moron.
ef3d6a No.14762932
If I printed my game's source code on paper, and sold it along with my game, would my game count as open source?
6001ef No.14762964
>>14762928
I found this (https://forum.unity.com/threads/finally-a-serializable-dictionary-for-unity-extracted-from-system-collections-generic.335797/) and I'm not using one of these rather than a normal dictionary.
What part of the code? Is this enough?
[SerializeField]
public class MyDictionary : SerializableDictionary<int, AnimationList.CurrentAnimationState> { }
public static AnimationList AnimationDictionary = new AnimationList();
public class AnimationList
{
public MyDictionary animationList = new MyDictionary();
public void AddAnimation(int hash, CurrentAnimationState type)
{
animationList[hash] = type;
}
public static CurrentAnimationState currentAnimationState = CurrentAnimationState.Idle;
public enum CurrentAnimationState
{
Idle,
Locomotion,
Default,
Pivot,
Dash,
Jump,
Land,
PivotStop,
Grab,
GrabJump,
Slide,
Crouch,
LongJump
}
public int currentAnimationHash;
}
This is what I'm currently using to add to the dictionary.
And I add to it like this
AnimationDictionary.AddAnimation(Animator.StringToHash("Base Layer.Idle"), AnimationList.CurrentAnimationState.Idle);
7b15e4 No.14762991
>>14762322
The alternative to that is switch case statements, but in C they can only be driven by absolute values, not a range, so there's no easy way to implement them in this case.
The best you can do if you really want an alternative is go with this for now and later do something like
switch (Temperature)
{
case Biome.temperature.Hot:
// Code
break;
case Biome.temperature.Cold:
// Code
break;
default:
// Code
break;
}
90f7ca No.14763007
>>14762932
I think so.
>>14762960
>what is life like as a programmer nerd?
Basically you type a lot
1b742d No.14763009
>>14762322
typeHere temp = temperatureSample.r;
if (temp <= 0.7) {
Temperature = (temp < 0.3) ? Cold : Temperate;
} else {
Temperature = Biome.temperature.Hot;
}
i hope this is formatted correctly
1b742d No.14763014
6e661a No.14763023
>>14762960
post "portfolio"
2817ee No.14763025
>>14762964
I don't see how this would impact building your game (or why it would use UnityEditor namespace), are you sure this is the actual problem?
>>14762960
>i doubt any of you enjoy life very much
At least I can make actual games and not just faggy VNs and walking sims.
6001ef No.14763033
>>14763025
It is, it's just not in that.
The problem comes into how I'm adding stuff to that dictionary.
// Find the Animation Controller
UnityEditor.Animations.AnimatorController ac = GetComponent<Animator>().runtimeAnimatorController as UnityEditor.Animations.AnimatorController;
RuntimeAnimatorController ac2 = GetComponent<Animator>().runtimeAnimatorController;
// Find how many layers it has
int layerCount = animator.layerCount;
Debug.Log(string.Format("Layer Count: {0}", layerCount));
// List all states on Layer 0 and add them to a list
UnityEditor.Animations.AnimatorStateMachine sm = ac.layers[0].stateMachine;
//List<Animations> animations = new List<Animations>();
for (int i = 0; i < sm.states.Length; i++)
{
UnityEditor.Animations.ChildAnimatorState state = sm.states[i];
string animationName = "Base Layer." + state.state.name;;
string animationStateToAdd = state.state.tag.ToString();
if (animationStateToAdd == "")
{
animationStateToAdd = "Default";
}
AnimationDictionary.AddAnimation(Animator.StringToHash(animationName), (AnimationList.CurrentAnimationState)System.Enum.Parse(typeof(AnimationList.CurrentAnimationState), animationStateToAdd));
}
Essentially I'm getting all the information I'm adding from the UnityEditor.Animations.AnimatorController, however RuntimeAnimatorController which is what I can use outside of the Editor, doesn't contain any of the same information.
Can I put my hash/enum data into an XML? I can't understand most of the stuff I'm reading online and I can't seem to find a straight answer.
1b742d No.14763037
>>14763023
can i work for ubisoft now
2817ee No.14763087
>>14763033
Right. What are you going to do with the dictionary later on? You might actually be better off using tags for animation states and then checking the tag of the current state? The other option would be making an editor plugin that writes your data to a file, I'd just serialize it to binary but you can use XML/JSON/YAML whatever you want if you want it to be human readable.
6001ef No.14763109
>>14763087
I'm already using the dictionary to check if my current Animation hash
animator.GetCurrentAnimatorStateInfo(0).fullPathHash
is contained in the dictionary and returning its state.
I need this done as quick as possible so using an enum is probably faster than using strings.
Where can I find more info about putting all of this into an external file? Anything would work honestly but I can't make any of the stuff I've found on google work with my script
2817ee No.14763129
>>14763109
I doubt using tags would significantly impact your performance but that's your call. This should point you in the right direction.
https://stackoverflow.com/questions/41709325/retrieve-all-animator-states-and-set-them-manually-in-code
6001ef No.14763137
>>14763129
That's pretty much what I'm already doing, it's still telling me to use AnimatorController from UnityEditor.Animations namespace, which I can't use.
2817ee No.14763146
>>1476313
Read the answers, it shows you how to make it to an editor plugin that writes to a file. Or you just #ifdef your code to write to a file when in editor mode and read that file when not. Or use animation events to push states instead.
9a3db4 No.14763264
>>14762928
>Always assume future you is a moron.
That's shockingly good advice.
388c31 No.14763295
hey dual contouring anon, if you're still there, after an embarassingly long time I figured out how to do everything, except the second to last bit where I use a quadratic error function to approximate the final vertex for every cell. Would you mind breaking it down a little bit? I am ashamed to say that I do not understand mathematical notation and every place I've looked so far uses scary math terms that my brainlet self does not understand.
If anyone else knows how to properly do it, you're very welcome to explain it even if you're not the person in question. Basically I got 4~ isosurface intersections with a normal of the surface, and I need to use them with something called a QEF function to approximate the final vertex for the mesh.
Lots of thanks in advance, you guys are saints for getting me this far, especially the DC anon.
6001ef No.14763305
XmlSerializer keySerializer = new XmlSerializer(typeof(int));
XmlSerializer valueSerializer = new XmlSerializer(typeof(AnimationList.CurrentAnimationState));
var stream = new FileStream(Application.dataPath + "/test.xml", FileMode.Create);
foreach (KeyValuePair<int, CharacterControllerLogic.AnimationList.CurrentAnimationState> animation in AnimationDictionary.animationList)
{
keySerializer.Serialize(stream, animation.Key);
valueSerializer.Serialize(stream, animation.Value);
}
stream.Close();
How do I make this look like it's not a complete mess of broken tags?
515c69 No.14764242
YumeScript v0.01 officially works!
Basically the engine has a Scenario manager which hotloads and parses common .ys files containing data for things like items, actors, zones, etc. Each zone currently consists of a single map and a script file. A zone script file consists of event listeners at the highest level - I can listen to global game events (e.g. "zone_enter") as well as individual entity events (not implemented yet) with conditions (also not implemented yet). Each listener has a sequence of commands, which can be nested.
The script is "compiled" into a hierarchy of command blocks, which are stored as objects in memory and are ready to be used by the engine. When an interesting event is detected a Runner instance is created for that block, which executes each command in sequence. The interesting part is that each command decides on its own when to advance the runner, this way I can display NPC dialogues and wait for user's input before advancing to the next command. Another interesting part is that each Runner is separate and runs in parallel, so multiple "threads" can run concurrently.
Right now only two basic commands are supported (see screenshot), but since a large chunk of the system is already implemented it's simple to add more. The next step is to implement listener conditionals and entity interaction listeners. I'm going to write a detailed post about the entire thing later.
My favorite part about this is how it contributes to my workflow. I can create and change models, textures, maps, data, scripts while the game is running, then press a key to reload everything, without recompiling or restarting the game.
188428 No.14764243
>>14763295
You know the correct formulas though, right? Can you link to one, maybe an anon can help interpret it
188428 No.14764264
>>14762460
>C# can't do that
The fuck are you talking about? You can declare consts in the class namespace, you can even declare consts within a function body and the compiler will replace the numbers/fields where appropriate.
One caveat to this is that consts only work with compile-time constants, which means you either use primitive types, or default(T) for your custom types. To get around this, the pattern is to define things as a (private) static readonly field. Eg:
>public static readonly float Pi = 3.14159262f;
388c31 No.14764315
>>14764243
There you go! I got the function I'm just too uneducated to read interpret it properly. Thanks for taking interest
2817ee No.14764372
>>14764264
He means you can't use #define MyConst 1.0f for example.
>>14764332
Grapes, fox, yadda yadda.
188428 No.14764388
>>14763295
Oh, I was hoping for a page link to it or something, since it looks like its from wiki.
>E[x] is the evaluation function
>Σ is sigma, which means sum of
>Function E is defined as the sum of i elements
>element n * (x - element p)) squared
So basically this looks like line of best fit. I remember seeing this in high school math. Basically for every point, you compare it to the line and get the area(?) between it and the line to see how close the line intersects the data. I believe that falls under more of statistical analysis sort of math stuffs.
Anyways, I'm mostly just taking a guess in the dark, here.
388c31 No.14764397
>>14764388
There's this paper http://www.cs.wustl.edu/~taoju/research/dualContour.pdf and they explain this a bit in section 2.3 but honestly it's all fuckin greek to me. I'm stupid. I ate playdoh and snorted glue in the back of the class back in school, and im just now catching up.
nice hitler dubs by the way
1b742d No.14764753
>>14764315
The squared dot product of nⁱ and (x-pⁱ)?
>minus pⁱ from x
>dot it with nⁱ
>square it
I hope I'm remembering this correctly from because the last time I did something like this was school.
388c31 No.14764788
>>14764753
Yeah but what is x, though? It seems to be on both sides of the equation. I feel like there's 20 anons reading this right now and laughing but i dun get it. Isnt' x the unknown that I wanna get?
03563b No.14764798
Can i animate in Blender and then put the animations in a game engine?
388c31 No.14764811
>>14764753
like, the n is the normal of the isosurface at the intersection point, and p is the intersection point. Okay, that much is clear.
what the fuck is x?
6f8871 No.14764973
>>14764798
You should just be able to export the object to a file that supports animations and put it in the engine.
Hell, last I checked, Unity supports .blends so you don't even need to export.
51bcaa No.14765294
Testing some fake shadows "tech"
It's taking a sliced screenshot of the tree and having that cast shadow, pretty boring process to set up, but the gains are substantial
I plan on using this for far away trees (which some might claim should have no shadow to begin with)
X and Z slices would probably make this almost indistinguishable, but I don't need to care about this so soon, gotta go back to making game not engine
0189ad No.14765356
>>14765294
At some angles it looks bretty gud but at other it looks absolutely ridiculous. I suppose if you'll only use it for distant objects it should be fine if you're willing to allow that performance hit.
c4cd95 No.14765406
Followed some of the advice I got last thread in regards to mixing, spent more time on mixing, still attempting to improve.
I thank you all for your advice on this stuff, I realize I'm quite amateurish at this time.
http://picosong.com/wP3an/
c1daa0 No.14765411
Good news everyone, Microsoft added Unix line endings to Notepad.
https://blogs.msdn.microsoft.com/commandline/2018/05/08/extended-eol-in-notepad/
Better late than never, as they say.
1e4948 No.14765419
>>14765406
sounds like Kowloon OST
that a coincidence or you the same guy?
sounds like late 90s Monolith soundtracks, SHOGO comes to mind. I like it. it's kinda cute
c1daa0 No.14765487
>>14765440
I mean, it is possible to use Bash on Windows.
4c6052 No.14765498
>>14765411 (checked)
I find it a bit "surprising" that they are doing it now, they had more than enough budget to add support for it in the past but did not do so. I wonder if they are doing that because of the PC market change which has been stalling for decades? But then it took also a Windows 7 release in order for them to make changes to MS Paint.
>>14748202
>Where you play as Hitler
ayy I was actually playing as the SS class, but I don't have any mugshots for the 2 other classes, so it is stuck using Hitla one. Pls gif free gibs man, just do think about the doom players who have to feed their families.
c4cd95 No.14765540
>>14765419
Yeah.
Kowloon is on hold right now as I am working on Die Totenmaske with a few other people from here (My baller friend who was my Kowloon co-dev, Sena! anon, an anon who did some designs for On Cosmic Tides, etc)
I've improved my music a lot since the days of my Kowloon webms, particularly in regards to repetition and weird off beat garbage that would make it in.
Mainly, AGDG has helped a lot.
In regards to it resembling Kowloon, it's an alternate style I've been working on for D.T which is more focused on classical instruments and drum beats. Regardless there's a lot of similarities.
I'm surprised you could point it out, it's actually somewhat cool.
7ca62e No.14765560
I know this thread is going to die soon but I'd like some advice on my code.
occc.co/share/ralphie_quest.zip
The way I planned on handling all nodes/entities on the map was via polymorphic objects. However, now that I know more about how cache misses can seriously fuck with performance I am debating whether I should make all entities "the same", the only difference being their "active"ness. For example, a static node like a chair won't need to be updated alongside nodes that move around, so I was thinking about having two pools of nodes that are updated on separate threads. Please give me feedback if you're bored. Thanks.
b8c7cf No.14765570
>>14765498
They also added the ribbon which makes finding things difficult, while also removing certain features (no invert color anymore??) and making things harder to use. Seriously, in old versions you want to draw a rectangle with a frame, you left or right click and it uses primary or secondary colors accordingly. Now you have to click the color indicator then choose the color, then pick the right tool. Right clicking is no longer functional since it just opens a cut/paste context menu
Why would you willingly download a MS tool to update Notepad when you could just use Notepad++
7ca62e No.14765575
>>14765560
By the way the pertinent code I'm referring to is in nodeContainer.cpp/h. I'm not finished with importing .tsx files from Tiled yet…
03563b No.14765599
i've been procastinating about game dev like the whole day
idk what to actually do though, i'm past the ideafagging stage
4c6052 No.14765603
>>14765570
>Why would you willingly download a MS tool to update Notepad when you could just use Notepad++
I don't know man, running MS Paint under Wine sounds like a good idea to me.
500b1c No.14765608
>>14765599 (Checked)
If you don't know what to do, how are you past the ideafagging stage?
03563b No.14765633
>>14765608
I have too much I can choose from and I can't choose, (learn how to) program,make character model (hard since i cant draw outside of front view and its agonizing without references) or i could draw characters like a faggot
210dcd No.14765655
>>14765643
oh, and for iteration[0] the dual vertex position is the middle of the cube (octant)
500b1c No.14765662
>>14765633
Then your first priority is to learn to program, since everything else is dependent upon it. You can make the best assets in the world, but you won't be able to do anything with them unless you can put them in a game. Start by making copies of simple games that you don't need art for, like Breakout and Asteroids. Then move up in complexity from there.
210dcd No.14765673
>>14765655
i'm retarded, lemmi re-do that post; some of that doesn't make sense
210dcd No.14765693
>>14763295
>>14764315
>>14764811
x -> dual vertex position error (will by used in determining final vert position)
∑ -> summation (between 0 to n, n = intersection count, so a for loop)
n -> normal for i
p -> intersection point for i
for (int i = 0; i < intersectionCount; i++)
{
dualVertPosiError = Vector3.dot(normal[i], (dualVertPosi - intersectPosi[i]));
dualVertPosiError *= dualVertPosi;
}
Unfortunately, it's not nearly as simple as the equation says right there; that's only a partial step of the QEF unfortunately.
If you want to fully understand how2 solve it, then I'd recommend learning calculus (mostly matrix math); even with my background of taking a few calc courses, I still don't fully understand how to implement it myself, but i know what it's doing.
Or… you could use a library: https://github.com/nickgildea/qef
Also, look at source code: https://github.com/nickgildea/DualContouringSample
210dcd No.14765721
>>14765693
attached is the part of the paper where they're explaining how to solve the QEF.
last part was just finding the error, which can be called the "minimizer" (i.e. we minimize the error of the qef, so distance from surface, or alignment to the gradient of the noise/density/level set).
188428 No.14766034
>>14764788
>E[x] = x * (…)
The left side is just formally naming the math function. The evaluation function is named E, and it has one parameter that it evaluates: x
You can ignore it entirely: substitute 0 into the left side and rearrange to solve for any of the variables (x, p, n). For example, when the function evaluates to 0, you can determine which X value to use. You can plug any number into it. But as far as programming it goes, you usually ignore the left side and just look at the meaty chunk.
https://en.wikipedia.org/wiki/Distance_from_a_point_to_a_line#Line_defined_by_two_points
For example, in the above, this is the most friendly expression I could find. Surely, you can make sense of it? It's basically a programming function call at this point. Looking at how the numbers interacted, I was able to, for example, adapt it to only use the top part and figure out a distance/area configuration by rearranging the formula, to help me solve the random point picking distribution I used here: >>14758266
210dcd No.14766162
>>14763305
There's definitely some type of settings you should be able to edit for the xmlSerializer.
Here, found it:
https://msdn.microsoft.com/en-us/library/system.xml.xmlwritersettings.aspx
Look at "omitxmldeclaration".
>>14762322
looks fine to me, I'd maybe come back in a pass to crunch the size of the code down though, and add some constants/static vars so it's easy to change/maintain.
210dcd No.14766251
>>14765693
Also, if you're really into learning this stuff, look at these terms on khan academy n such:
>basics of calculus
Search terms: summation properties/rules, partial decomposition/derivative, matrix operations and their applications (matrix add, subtract, multiply, divide), matrix forms (square, identity), matrix solutions for linear systems, matrix row-echelon form (gauss-jordan method), Jacobian matrix & determinants & cramer's rule
>introduction to matrix calculus (will cover types of matricies w/more detail)
important: symmetric matrix, and transpose matrix
>single value decomposition (SVD)
SVD pseudoinverse (approach of SVD used)
If you learn all of that, then you'll be good, but implementing it yourself is still a pia.
Though, you can also understand a lot of machine learning formulas with those skills; so that's a nice perk.
388c31 No.14766816
>>14766251
Not as much that I'm into that stuff per-se, but I keep getting myself stuck in math way out of my league because I was cocky enough to start trying to make a game with good-looking destructible terrain and surprise-surprise the reason there aren't many out there is that it's pretty fucking hard and has a lot of complicated math concepts involved in it. Since I already started, giving up is not an option. If you think this stuff is what I need to learn to properly understand the topics I'll be running into there, I'll definitely start learning it.
>>14766034
Thanks, anon! The function looks much less daunting now.
>>14765693
Okay, seems like it's not as simple as just breaking it down for the layman. I'm going to use the library for now, but I feel like I'm going to have to learn this eventually so I'll put smacking my brain against it on my bucket list. I really appreciate the time your put in to put this explanation together.
388c31 No.14767397
hey rate my dual contouring algorithm out of ten. hope im not making you guys insecure, I know this is next level stuff.
47fbde No.14767462
So here's a question for somebody that's mildly interesting in /agd/ but doesn't want to quit his day job and start getting invested:
I enjoy taking online classes with CourseEra, and find it pretty useful. It's also absolutely free, which means not wasting any money on the bullshit. There's a few that are related to game design:
https://www.coursera.org/courses?query=game+design
Has anyone taken any of these classes, and can give comments on their quality? I imagine that, at the absolute worst, it's going to be better than college courses, for the simple reason of not having to deal with political bullshit.
9a3db4 No.14767483
>>14767462
My paranoia tells me you're a shill. Also, there's plenty of books floating around, no need for casualized courses.
72f142 No.14767525
>>14767462
Assuming you dont have a team you will have to 1ma gamedev and do you think game design is going to be the first thing you need?
Briefly looking at the course descriptions feels like they are "how to ideaguy" courses.
Dont get me wrong i think they might help you, but you should probably develop other skills first, because even if you master game design how are you going to finish anything or get others on board? (without money, because if you have money you should probably be learning how to manage teams and budget the whole thing properly)
9a3db4 No.14767647
>>14767525
> feels like they are "how to ideaguy" courses.
You make the point better than I could. If you have any potential as a gamedev then you already intuitively know game design. What you do need is programming skills and relevant art skills like 3d modelling/animating.
6e661a No.14767753
>>14765540
It's great you guys collaborate so well. It seems that /agdg/ anons over here are much more compatible and willing to help each other than the spergs on cuckchan.
388c31 No.14767781
>>14767753
the /agdg/ thread on /vg/ is aggressive as fuck. I thought the opposite would be true, but they tend to be sarcastic, unhelpful and uncooperative. 8ch /agdg/ is fucking magical.
b8aef2 No.14768688
So I just spent the last 4 hours failing to load an .obj's triangles. It turns out that some retard at Wavefront decided that the index starts at 1 instead of zero.
188428 No.14768754
>>14767781
We also have a cyclical sister thread on 8 /vg/:
>>>/vg/27335
e4480b No.14768760
>>14768754
Big sister or younger sister?
188428 No.14768817
One idea I've had for a future game is a Terraria clone. Each world would actually consist of 25 or so map cells which are about the size of a small world. The cells would be arranged in a hexagon, and each cell would loop on itself. How far along you are in the cell determines where you end up when you travel to another cell. Thus, you could mapgen it and have distinct biomes, heightmaps, etc., and it would be a pseudo 3d world
One thing I'd wanted to do was have a weather system that was manipulated by spells and terrain in a predictable manner, but still being a simulation rather than being arbitrary.
I think the solution I found that works with the map cell system (imagine being on one part of the game world island and seeing a storm front rolling in) is to use particles per map cell. For example, each cell could have 60-100 or so "weather nodes" associated with it (hard limit 256), and each one contains weather data, such as pressure, humidity, elevation of the node, etc. Thus, you could get an aggregate of the nodes and determine the major 2-3 weather systems in a particular region, which would let you model hot/cold fronts, and determine rainfall and wind. The player could then get weather related spells that eg, release more temporary "hot particles" into the air, which would reduce the chance of rain
d22652 No.14768930
>>14768817
>magic terraforming
47fbde No.14769520
>>14767525
>>14767647
Yeah, I really don't need a 'how to idea guy' course. I've been putting ideas into scrapbooks since I was 10. The problem is I don't actually know how to implement anything. The best I've ever managed is working with the Morrowind CS and learning to script, but that's just incomparably simple.
The thing that bothers me the most is modeling, honestly. The moment I try to do anything 'arts' it immediately collapses into a mess.
72f142 No.14769675
>>14769520
Well you will have to keep at it if you want your art to not be shit (or anything else). It really hurts to be shit at something, but its the only way to get better.
4798c3 No.14771227
I added camera controls to the debug view by copying the code for moving around in the 3D view in the editor (at least, most of it. I didn't include the scroll wheel or the up-down keys). So, now I can move around the debug view. Now that this is done, I can work on the actual reason for the debug view- analysis of the BSP tree's polygon lists and splitter planes at every step of compilation.
9a3db4 No.14771288
>>14771227
So what is the whole BSP thing for? I always thought it was just an extension for hl1 maps and that hammer's calculations were about baking lightmaps.
71159f No.14771300
>>14771288
It's a satanic ritual Ika performs to avoid modeling his maps in Blender.
188428 No.14771366
>page 14 and no bread
worst timeline
2724a3 No.14771383
>>14771366
It's the last /agdg/. Ever.
Go home everyone, it was fun while it lasted.
4798c3 No.14771392
>>14771366
>>14771383
Dont worry, I saved the day!
New thread: >>14771391