/agdg/ + /vm/ Thread

Don't quit your day job for game dev edition

Post your progress, ask any doubts you have about development, or ask for feedback for your game or mod.

Remember, there's #8/agdg/ in Rizon for discussion, we also have resources and FAQs in >>>/agdg/ and >>>/vm/

Other urls found in this thread:

my.mixtape.moe/fmareg.mp4
my.mixtape.moe/vwcasp.webm
programmers.stackexchange.com/questions/60544/why-do-game-developers-prefer-windows
shiny-rice.itch.io/distantbreakoff
steamcommunity.com/id/amigarules/
ivanovitch.neocities.org/index.html
social.guhnoo.org/
jabb.in/
shitposter.club/boris
sealion.club/group/gamedev
store.steampowered.com/app/392080/
youtu.be/2RynIEZJctw?t=27s
twitter.com/NSFWRedditImage

Aspect Ratio is 16:9 now.
Made a database of sorts (to store and import sprites and other data to the character object).
Made a mockup of the dialog system.

slide it moishe
my.mixtape.moe/fmareg.mp4

That portrait is 10/10.

>first webm I find my.mixtape.moe/vwcasp.webm to my suprise


nice, looks gud
The text sound is a bit jarring to my ears though, maybe adjust it to be a slightly lower frequency.

oh wait it's an mp4, I'm retarded, well anyways the mp4 doesn't display for me user.
I did test a different mp4 hosted on there just in case it was my browser.

So that's why Holla Forums bitched at me that I need H.264 encoding to be able to upload it.

at my site it just werks, palemoon + loonix.

it just shows a nigga constantly sliding left/right when close to a player.

ah, codec issues, makes sense

Is there a way to extract textures and models currently in use by a game on screen? I haven't found a way to crack open the models and textures for the game I want to extract the content from and port it to another game.

Try 3D Ripper DX.

How does this thing look now?

Wondering if any of you know what the deal is with directx 9 being the directx of choice for people making a game engine from scratch. Even that one Metroid 2 fan remake uses D3DX9_43.dll . Just curious why its always number 9. Not 10+ not -8, but always ⑨?

9 is widely supported, and runs on winXP.
10 requires windows vista and higher.
11 requires a modern GPU, and above OS+.
12 requires win10.

In general DirectX 10 doesn't justify the higher system requirements needed to use it, unlike 11+. There's also still a large amount of 9-level GPUs floating around given the large gap between 9 and 10, whereas 8 came out only two years before 9 and is missing useful shit like HLSL.

...

Figures.
One more question. I always see directX and OpenGL mentioned as if they are competing things, yet both are generally used in a game and probably at the same time, I think. Why would you use one over the other?

What happend if you try to use 3dfx's glide API? Also is Vulkan suppose to be a glide?

Compatibility as well. DirectX is proprietary, and only runs on Windows and related Microsoft products. It has great driver support, and in some cases performs better than opengl. OpenGL, however, can run on Windows, OSX, and Linux, and most other platforms. It tends to be a bit slower than directx, but that's because they have different approaches when it comes to implementing the rendering pipeline. Although with OpenGL4, I bet that its gotten a lot faster. I wouldn't know, I'm devving on a toaster that only supports 2.1.

It sounds like you have a lot of reading to do.

...

In the old day, you could use a slow cpu chip paired with a voodoo card and still get good performance. The GPU did a lot of the heavy work.

Vulkan task is similar and advance.

If you want a thorough, but quality overview for a directX vs openGL; check out this stackoverflow post:
programmers.stackexchange.com/questions/60544/why-do-game-developers-prefer-windows

They can both be supported, sure, and generally are automatically chosen for best compatibility depending on your GPU/OS if you're using an engine like Unreal or Unity.

Nah, they're not directly compatible, nor do they run at the same time; it's one or the other.
Generally, they have nearly the same functionality for graphics.
As OpenGL is purely a graphics API, while DirectX can do a bit of everything.

Compatibility, usability (HLSL vs GLSL… the language used to code), API requirements (mentioned above), open standard vs proprietary, etc etc.

Shame it's a placeholder, but hey thanks.


Done and done.

Made a somewhat functional dialog system:

- Added a database of sorts that contains the text and effects (like changing music, moving the portraits, maybe even a screenshake) for each enter press
- Added the possibility of small wait times between the text, using and improvised "mark up language"
- Added portrait system whick makes easy to insert them in the scene, manipulate them and remove 'em.

Nice, it's gentle on the ears and sounds nice on top of that.
Dialogue system looks good, and gotta love that smug.

...

I got the same feeling with Mighty No 9/100.

Depends in the game.

Do you have to fork out a lot of money to have real guns featured in your game?

Depends on brand, they have their own licensing terms. Same as cars.

I believe that is the reason Call of Duty is being set in the future lately, and Battlefield in the past. If you're setting your game in the future, I believe there is a good workaround, apart from having fully original weaponry. It's just to "modernize" current weaponry.

It's what Ace Combat 3 had. Namco literally took real world aircraft, they slapped a COFFIN and few other gadgets, and called it something else. Take for example the F-22 Raptor, they called it the F/A-22B Raptor II. I think the same can be done with weapons, it's what I'm going to do. You can also take weapons which aren't licensed, like say, every AK rifle made up until Kalashnikov went private, the AR-15/AR-10/M16/M4, and old as shit weapons, like the Mosin Nagant, the SMLE, and many others.

My bad, it was called the F-22C Raptor II.

...

Not like the F-22 was a cheap or easily maintainable craft. Your DoD can't stop sucking Lockheed Martin's cock.

It's looking like it would have legitimately been cheaper than the F-35 at this rate.

Robert Gates belongs on a pike.

help when can i get patreon dollars :(

Don't worry. The Israelis will sell the secrets to the Chinese like they planned to with the 22 then they'll have no choice but to make something newer.

I did a shitload of work for some guy and he didn't pay me and then tried to close his account and now I have to go through a month long arbitration process

fuck this I'm going to go work at mcdonalds

on what project did you worked on? care to tell more?

If you aren't working with someone as a close team, never do work without getting paid first.

Ayy, I haven't released a build since last october. Come grab it, faggots. You need Steam installed and running to play it, though. shiny-rice.itch.io/distantbreakoff

...

Fine, tell me how to get a proper online subsystem that isn't Steam running in my game inside my time constraints, please.

maybe make it atleast work so that steam is only needed for multiplayer.

It's multiplayer only.

meh

heh

Mighty No 9 is an easy fucking thing to make. Inafune is a hack and the whole project was horribly mismanaged. You'd be surprised that having someone who's not a hack at the helm makes a difference.

I think Null subsystem with certain programs works but I only know shit ones, and if you forwarded ports it should work with direct IP connection through the console, with that port after the IP.

Now I'm using position and color in the text to identify who the hell is speaking.

I redone the past dialog in the same fashion as well.

I'm working with Unreal Engine 4 here. Null does work. I was actually looking for a comfy server browsing experience here, I've always thought that direct IP connections were a bit of a hassle, mainly because you have to depend on other resources to know which servers are up. I'll discuss this topic with my pillow later tonight.

...

Wew

Want to do a Kaiju fighter similar to Atari Godzilla games. I'm about a fifth through this C++ book and was looking to use UE4 as it's destructable mesh system would be highly beneficial to the project.
I'm gonna start reasonable and small by prototyping with just two characters(a robot and a kaiju), one stage(But unlike Kaijuland Battles it'll be a city stage with destructibles and not a bland arena). Of course it's only a smart idea to ask for input before going into any project so lay it on me.

bro
take it form me
your drawings are fucking cancer

like, i can respect that you want to make your own game
but you need practice
i mean i draw like, once a fucking month and i can still make shit marginally better than this with no references

It's harder to write quality, performance code with opengl, even when just using opengl 4+ core profile (no pre-4.0 compatibility functions). There are just so many ways to do things in OpenGL and only 1 or 2 of them will give good performance. Also AMD OpenGL drivers are still shit, not really opengl's fault but it will affect your code.

Try reading the thread nigger.
Hell, I'm fairly sure they're traced, at the very least the main character looks like that one chick from Madoka.
Also your goblin doesn't look particularly good either, you need more anatomy practice.

I'm working on a 256 colours only game. Do you think 4 shades of each colour is enough ? I don't know if I should go with 32 colours in 8 shades or 64 in 4 shades.

woops

but yeah that's why i said marginally better
i don't draw nearly enough

still that male character is hideous and if that's anything close to gamedev's drawing ability…

add a little dither to make the leather look less like plastic

At the very least, you need to represent the RGB color space (so red, green, blue), plus the CMYK color space (cyan, magenta, yellow - all which are combinations of RGB), as well as white/black.

Honestly, even a 16 or 32 color palette could give you a lot to work with. I don't understand the reason why you'd use such a large palette because it would imply you don't know how to use colors effectively, which means why even bother with a palette at all? Otherwise stick to a smaller one.

Pic related is a pretty solid 16 color palette, for example. You could easily expand it to have more color ramps between colors and tones, without much effort.

Maybe it's latex?


256 colours is 8-bit. Using a limited amount of colours on-screen at any time is a secondary limitation, as well as a limited amount of colours per sprite.

I'm moving to UE4 from Unity because I cant stand not being able to do anything without coding knowledge. I'm not buying playmaker

I didn't decide on 256 colours on a whim, the game is coded to work on 8bit colours, having each pixel as one byte is much faster.

Are you an artist? If not, stop being useless and learn to program.

Only on machines that have real hardware sprites. Nowadays sprites is synonymous with 2d graphics. In my case o have available 256 colours (255 actually as one is for transparency) whether it's "sprite" or background or whatever else.

But if you're going to be that pedantic, then why not just use 16 colors per sprite, then the sprite data could be serialized and condensed down to 0-F per pixel. That is, one byte could represent 2 pixels.

Wont have time for that until next year faggot.
My art skills arent quite at art fag level to drop it and learn coding. I have friends for coding

My code can support 256 colours and you tell me to only use 16 because what ? Sure I could make it work with only 16 colours, but then I'd need to bit shift when reading the screen data and it would be slower than 256

It's for when you want to accurately emulate the aesthetic of an old console, and thus not be a hipster douchebag slinging the millionth indie pixelshit shovelware onto the internet.


Half a byte is called a nibble, or nybble. Depends on who you ask.

Why are you emulating the slow scrolling text meme? It doesn't matter if a downy 10 year old is playing the game, he can read the text after it's all appeared. Slow text crawls are the work of the devil.

Yeah youre right, I didn't think about that.
What's your favourite 8 bit system ? Mine is undoubtedly c64

I didn't have an 8bit system myself during my youth, but did play NES at a friend's house. If my parents hadn't sold their c64 before I was old enough to use those things it might have been my favourite, but nostalgia goggles make me answer NES.
My actual favourite technology-wise would be the Vectrex but that doesn't have sprites so it's not really related.

I feel you. I never had c64 as a kid, but I do now, it's fucking fantastic. As a kid I had many Chinese nes clones. Funny that we didn't know it was Nintendo, nobody knew there was such thing as Sega… Just "TV game".

My neighbour's mom went to Italy once and brought him back Sega master system. Of course he couldn't play anything since we all had pirated Famicom carts. We thought it was different cuz it was from Italy.

Nah


Are you planning on making something like a competitive Rampage or just a traditional fighting game with big characters?

THEY ARE PLACEHOLDERS

That's why I code, m8.


If you click all the text shows up instantly, pic related.

Both girls are traced from Madoka, you have a nice perception.

One-man designerfags using Blueprints > Enginedevs.

You can publish on PS4 now with only scripting knowledge.

You might be able to pump shit out faster but you'll never be pushing the boundaries of what can be done with technology or implementing your own cool features.

Well I currently have two concepts in my head. Both are inspired by Pacific Rim/Godzilla/Evangelion type stuff. Both are to be weighty in their gameplay and be focused on things like the feel of combat and the destruction of the city(Which is where UE4's destructible mesh system comes in). And both are focused on Robot vs Kaiju combat.
Example A is a smaller-scale more in-depth combat system with typical mech game camera system. The idea is to bring the kind of Pacific Rim or Evangelion scenarios to life. With the smaller scale being there deliberately to make it as to generate a sort of hide-and-seek gameplay element of being able to dip out of combat, reload weapons or setup an ambush, by hiding behind buildings.
Example B is straight up Atari Godzilla style but still with the differences between robots and kaiju being there. With robots using a heat meter and kaiju having stamina. And with robots having some powerful attacks with limited charges.
I'll probably end up going with example B just for the simplicity of it. With example A being for the next game once I've established that I know how to make game(Prototype of a game/proof of concept, probably will end up that way) and can get a team together

Well is there any gude books how to become a professional modeller? I would like to learn about UV-Mapping, Retopology, Texture Drawing and doing stuff like Bump/Normap maps, Animation and what else is needed nowadays.

I am using mainly Blender and Loonix.

get a subscription to www.cgpeers.com (they open on the first and the 14th of every month for registration I believe)

Has tons of books, tutorials etc

Well alright I gotta keep my eye on it then.

At what point have I surpassed scripting and truly entered "programming?" How is one not a subset of the other?

Where do I start if I'm wanting to actually apply myself to programming? I've had a lick at C and Python in the past but it was more out of curiosity than having any end goal in mind. I feel stupid asking but with the sheer number of languages and guides, as well as all the shitposting I've read on Holla Forums I just don't know what to do.

On one hand people I've talked to in games and other sites say doing anything in C takes way too long. But Holla Forums shits all over C++, Java, C#, etc. Which seem to be the most common languages used in development.

switch to maya and learn zbrush. Mainly because zbrush is kinda easy to learn. Its an easy 2,5d sculpture program. So most of the shit you create in maya is easy to transfer to zbrush and vice versa.

Also there is allot of tutorials for zbrush and you can get amazing results with it.
There is sculpture program from Autodesk but I've not dabbled with it. So I don't know how that one works.

Key distinction here - you're not a team.
Don't use C++ - it isn't much of an abstraction improvement over C.

Learn something like C# for Unity or Lua for Love.

Holla Forums types have a tendency to evangelize their own preferred language like it's the gospel from on high.
The truth is, in the long run, the languages all do pretty much the same thing, and while there are definite differences in how they get there, and each has it's own strength and weaknesses, those differences aren't going to matter much to you until after you get past a certain point. Later on, you'll better understand the reasons to use different languages. For now just start with anything commonly used, aka well-documented with an active community.
If you're wanting to focus on something you can apply to gamedev right out the gate, C# is a good place to start. As for me personally, I love working in C, and it's easy to learn, but you probably won't get to gamedev straight away with it.
It's hard for me to recommend C++ to anyone, anymore. But that just be my own bias showing.

My understanding was that C++ is basically C with OOP. Isn't that why the industry uses it?

Reminds me of vid related.

In theory, yes. But in reality, there are substantial differences which people smarter than me have written about.

And even if that's the case, if C++ is just C with OOP (btw you can write OOP-like code in C, so technically it would just be C with inheritence), there isn't really anything that OOP adds to make it a lot easier to write code - you're still dealing with fairly low level shit a lot of the time. That's what slows you down in C - not a lack of inheritance.

They use C++ in the industry because it's a fast OOP language and they need low-level details because they're writing very large and performance-sensitive games. And you'll also find that most companies don't have pure C++ codebases - they'll write their game engine in C++ and write the actual game part in a scripting language like Lua.

I mean, if you want to be an enginedev, then go ahead, but otherwise I wouldn't recommend it.

Guess I might go for Lua, if I ever become interested in making my own engine I might look at C again.

To be clear, there's nothing wrong with C++ for enginedev. I just personally prefer C.

Also, if you do end up going with C, keep in mind that K&R is a meme and nobody actually writes C code like that anymore because it's incredibly unsafe.

Alright, I've already found some books on Lua that are straight from its developers so I'm going to poke at those for now. Do you know of any good resources for practice? Something like the roulette threads on tech but slowly scaling up rather than being entirely random.

First time working with OpenAL, implemented a basic 3d sound system in my game engine.

Behold - authentic stereo screaming experience.

That was the idea everyone got on board with originally, yes. Man, I didn't really wanted to go down the language debate road, but I think this is worth discussing a little. Sorry if it runs long.
I'm not that wild about OOP either, to be frank, especially on smaller projects, and by smaller, I mean not extremely massive. The chief benefit of OOP is that with heavy encapsulation, different parts of the team can work on different modules that are almost completely isolated from each other but can be put together to make a final product. This makes the project as a whole less likely to go off the rails when four people quit, or something has to be redesigned because the publisher really wanted a particular feature added that no one planned for, and so on. Basically OOP can push a project through a lot of hurdles.
The thing is, the more and more you abstract out the design, the less and less you retain of any performance benefits of using a somewhat lower level language. Inefficient code is inefficient in any language. But you sure as hell retain the potential for frustrating and subtle bugs.
(C#/Python/Other shit Holla Forums hates) basically give you all the same OO related benefits, in a safer, easier to work with package, at the cost of potential low-level performance optimization that you probably weren't going to take proper advantage of anyway if you're working with higher-level abstraction and attempting to decouple everything. On the low-level side, C is only difficult for some people to work with because you can't ever "just" focus on the logical reasoning of your program, that is to say, you pretty much ALWAYS have to be aware to some extent about what is actually going on in the machine. But the thing is, if you're trying to write efficient code, you really kinda have to do that anyway, and I find I actually work much faster in non-OO C than OO C++.

Basically this is all just a very long way of saying what the other user said

Both of those being critical. What you really have is a small handful of people writing tools in C++ for the rest of the team, who don't (and shouldn't have to) know that much about that level of code. But a smaller team (or one person) working on a smaller codebase will get more done, faster, working in C, because it's very simple and powerful, or in something like C# or python or whatever, because they're relatively safe, well suited to putting together modules/libraries with, and allow you to focus on the logic of your game over what's going on in the machine.

All this is just my two cents, of course.

(checked)
what advances has maya over blender?

You should have "placeholder art mode" in the options if you ever finish.

Let me recommend LuaJIT and C together, in glorious harmony.

LuaJIT is incredibly fast and includes a very simple to use and direct interface to C and C++.

I've moved to a strategy that if a piece of code doesn't need to execute more than once or twice a second I write it in Lua. Lua is also great for quickly defining game objects, material definitions for rendering, high-level AI strategies, loot generation logic, battle flow logic, etc. It's so easy to prototype a change to a Lua script and just re-load it on the fly, without hassling with recompiling C code.

The guys who made Grimrock and Grimrock 2 turned me on to it in their development blogs. It's a very powerful and easy to use combination for a single person or a small team.

I like C and C++ a lot because how low level it is. People here might say that's a bad thing, cuz it's harder, but when you git gud with it, it's very beneficial. Whatever you code will be translated to machine code (in most cases) and of course it's faster. I like the fact that I can write a function in C and then see it in assembly and it still makes sense.

Learn c++ you won't regret it.

what I mean is that the generated asm code will be closer to what you wrote in C than in any other language.

How would you make walls in a perfect top down perspective ? As it is now it's only black lines, but there must be a better way, right ? Any ideas ?

Being an artfag trying to get into coding is more difficult than I thought. 50/375 pages through this C++ book and I'm sweating a little bit. Any artfags in here who successfully learned C++(or just any programming language) got any tips on sticking it out till the end or advice on making life with C++ easier?

Stop selling out to tumblr, you fucking faggots.

Maybe you should learn to fucking draw first before you make graphics.
>>>/loomis/

You could always be the artfag for a programmer. Aren't there usually more programmers in need of an artist than the other way around?

Who is?


Yes, but collaborations are fucking hard. Few people, artists and coders alike, have the grit to stick with it to the end.

Okay, I'm fucking retarded. How do I do data-based ballistics in UE4 and in Blueprint, again? no time for C++ fam

Just to clear things up, data-based ballistics means that instead of having an actor being spawned every pull of the trigger or every fraction of a second, I have a raycast that travels a short distance and then having another spawning from the endpoint of that cast, and a ballistic trajectory is emulated with each cast.

I've just gotten my feet wet with this shit, and I found a roadblock instantly. So, I can easily cast rays from the muzzle once, but the moment I make more than one raycast in the player character blueprint, to emulate ballistics, I realise that that part of the logic is busy already with one ongoing shot. This means that, once I've done one shot, you can't fire another one until the previous one has hit something or has expired.

Help?

I'm just a coder.

I never read any programming book. You can find all your stuff on the internet. It's more about getting in the mindset of programming than learning the language itself. You must learn to think like a machine, then any language will be much easier to learn. I started with AutoIT. Very easy scripting language and it has neat GUI editor. Then I started with C/C++ when AutoIT wasn't enough.

BTW. I'm actually looking for graphician. If you want to talk, even not serious shit then you can add me on steam

steamcommunity.com/id/amigarules/

Just like in art: reading a book is not going to make you a programmer, it's only actually programming that will. You need to write simple programs and work your way up as you piece the concepts together. Think of a program that you could conceivably write, or want to learn how to write, and read about the topic until you understand how to approach it. Or pick a library that sounds interesting and play with it. You need to know that it will take a lot of programming before you are good at it, so program things that interest you and that you find fun or exciting.

Life with C++ will only get easier if you understand C and object orientation. Languages are created at a specific time for specific reasons, if you can put a language into its context it becomes that much easier to grasp. Also if you're not using some windows exclusive API like cinder, directx, or the winapi itself then I'd recommend you try a linux. They've got a much more comfy environment for C and C++. It can be frustrating spending a lot of energy on your programming environment though, so if you don't have a linux install and the thought of setting it up is intimidating then worry about it when you do have the energy to spare.

Yeah, I'd rather do the art and programming by (mostly) myself. I'd rather have the only outside help I need be for specific things I can't do and for when I get to needing an audio guy.

So I should basically spent more time with exercises once I learn the basics and develop my knowledge from that? Where can I find good game making exercises?

This ! I can't agree more. You learn to code all the time. And the linux bit is also true. I was using Windows all my life, but tried Linux about a year ago. Was frustrating at first, but now I got used to it and I love it. For cross platform coding, especially games or anything graphical I highly recommend SDL.

Anyone have recommendations for 3d modelling tutorials? Preferably for the latest/closest to latest version of blender.

The excercises should be self-directed, it's the best way to maintain motivation. You need to decide for yourself which aspects of game dev interest you the most, and learn those topics through your own research and programming attempts.

yeah I did that and stopped posting here or making progress. Just waiting for everything to get settled… I will be back.

There's lists of general programming excercise projects at >>>Holla Forums634156 which includes a few games too, but any really old game like pacman would also be viable. My first test of any engine is to see if I can make a working Pong game without reading any documentation, that's sort of the most basic game possible.

Keep it as is as placeholder graphics and get an artfag to do a better job later in the process

That's how it is now, but still. I just don't know how to make walls in perfect top down perspective. Is there a better way than just one color lines ?

So if I were to make a kaiju fighter, I should be attempting to program and researching the individual components of what makes that up?

That's probably a good place to start, isn't it? Making simple games first and learning from them.

Also, people tend to abuse OOP because that's all they're taught. Whether online or in academia, that's what's being masquaraded as the be-all-end-all way to program. Now all you see people doing is writing endless boiler-plate abstraction schemes for everything whether it's needed or not.

progress

and just need hair

Then to the collision and damage stuff

hold on, what genre are you going for? I thought you were going to make an rpg

Yeah, I'll have to make it semi-turn-based somehow. Still figuring out the battle system.

You could try making it similar to Valkyria Chronicles.

It'll be more like Resonance of Fate, I guess. I don't have the stamina / resources to make all the characters for a full roster like in VC.

GET YOUR FUCKING GAME ON BOYS
make something of yourselves

you, i don't like

no
i'm gonna waste my life away playing old vidya instead

what are you, some kind of yandaredev?
proper recognition is a bad thing

I'M TRYING

Found the sellout tumblrinas. It's not about recognition, but having principles. There have been /agdg/ project that started out simple, but then received massive attention, which caused the authors to completely renounce their ties with Holla Forums and /agdg/ often demonizing the former, and fully embrace their new audience, consisting mostly of teenage tumblrinas. All of that in the name of dosh.

Fucking don't be like that. Have some balls.

People who sell out and demonize the people that initially supported them would've done so anyway once they see success, whether they post on tumblr or not. It's not a community's fault but rather the developer who turned into an asshole.
The majority of the people who use tumblr are retarded, but it's a useful tool for a dev to keep track of your progress and get it out there. Lots of devs use it without pandering to anyone. You could use something like blogspot instead, it's just not as comfortable.

Well on my part, I never liked the idea of hosting my shitposting on tumblr, for me it is a place which is filled with infected persons who any new member unfortunate enough will fall victims and grasp them on their claws, turning the victim into one of their one with bizarre perception of reality and reality bending toughts.


I only have this website which I really need to update it and adding some dank graphics that shouldn't burn out the toaster, I would like to have some kind of a blog posting functionality as well so that I can just use the new post button and informing the users about stuff.
ivanovitch.neocities.org/index.html

My VN demo is getting closer to being ready. Can't wait to finish this so I can not worry about it anymore. Except in the low chance it gets popular and the patreon bucks start flowing in.

help i need a game designer/storyfag

We should be more vary of such potential traitors. Also, out of all the potential (micro)blogging platforms you chose tumblr? Come on, son! At least use GNU social or something.


How's Skeleton CSS? I've been thinking about setting a little website for myself, but I'm to lazy to write my own CSS. Dealing with all the "responsive" shit for mobile is a pain in the ass.

seems to be nice so far, I haven't used it extensively yet, so I cannot tell how it performs overall.

I will later make a concept model/render of the new website design I have in mind.

You won't know until they start whining about how 'toxic' and 'mean' imageboards are. By then they'll already have jumped ship to reddit.
That exists?
Look, I'm not a fan of tumblr myself but it's the thing that works best right now to make the masses aware of your product. Twitter is even worse and doesn't allow for larger posts, legitimate content is overlooked on Facebook and most other options nobody will bother to look at. I rather tumblr died and people switched over to itch.io or something, but that's not gonna happen for a while.

wops I forget to mention that part.
Well I'd say just rip it from another site or try to find some templates thats what I would do anyway since I am not experienced at making a HTML 1.0/2.0 site at all.

There is shitload of website templates, HTML and CSS ones. I have to admit wading through all the shit till to find a fitting one is tedious, I tried it myself a bit and I haven't found a gud one yet.

hmm I heard 'bout it but I dunno man.

Found someone who was willing to model a gun for Space Pirate to use. It's just one gun, but it's better than nothing. First is the actual model, second is how it (should) look in-game.

Will need to touch it up in Photoshoop, add some details and shading and such, but it's an interesting start.

Yup. Either pick an instance listed on social.guhnoo.org/ now or wait for jabb.in/ to get configured correctly (which should be soon). It's kinda like email applied to social media: make account on one server, follow and message people from other instances. Avoid Quitter like the plague.


Come on man, don't be like that. We have free software and waifus.

Welp, which one of those GNU+Social server should I pick then?

Where?

So Space Pirate in Space is going to use non-pixelart in the future, or do you mean turning it into pixelart by touch-ups?

shitposter.club/boris
Well here it is buddy.

Good choice, @moonman's a chill admin. Try remotely joining this !gamedev group: sealion.club/group/gamedev

Yeah, likely converting it into pixel art with touch-ups.
How it was done in ye olde days was to take either photographs or 3D models and draw over them. Doom was pretty much entirely photographs and photomanips of stuff (the BFG is a very impressive-at-the-time editing of a raygun toy), Shadow Warrior was almost entirely drawn-over 3D models, Duke Nukem 3D was a mix of the two.

Once I start getting serious about replacing art, my preferred direction for artists would be probably be doing a combination of both, leaving the "entirely original" art to things like conversation portraits and character designs.
If it has to be drawn from scratch, it loses a certain something, I think.

Mind, I'm not an artist, so my opinion on art ultimately means shit all. But I want to keep a consistent visual style so I hope having a direction of "do it like this" is good?

Alright I am in, so how do I remote shitpost targeted to a specific group?

Anyone know where I can find the source code for released game-maker studio games?

I'm trying to see how other people organize and code for bigger projects, along with messing around to see exactly what does what, but so far the games I found (iji plus his other projects, spelunky) all have obsolete code.

Cock it into an in-between position about 3 quarters of the way between B and A.

oh bother

Anyone here with some experience modeling want to team up and try to make some Dota 2 items? I figure that it's a pretty good way to get some experience, and I've done it once before in the past. Since I'm admittedly a hack when it comes to modeling, I'm willing to split the potential income 60/40 to you.

Why do most of you talk just about the money ? This is supposed to be amateur board/thread, most of you probably never even shipped a game before. I guess nowadays in the age of patreon bux you can't do something just for fun.

There's 150 posts in this thread, excluding yours and mine. How exactly are most of us talking about money?

Well for my part all my projects will be either GPL2, CC-BY-SA or something else, so basically close to libre as possible. I don't care if I can make sheqels with my projects or not.

I might get off my ass someday and upload shit that I have made to OpenGameArt, I have quite a few models and some textures.

speaking of money, I'd sell out all of you in a heartbeat if it could fund my addiction to not working on a game

Are you still going to sell it?

(checked)
I won't even charge for a cent.

Webdev here. While tedious, it's really not hard once you know what you're doing. I can see how it'd be a drag to learn it if you don't have a specific reason, though.

just my opinion but
fuck that
this looks awesome as it is
you should leave the battle system real time but maybe with some kind of quick access menu that slows things down, akin to ni no kuni

This would look great with WASD control, a dynamic camera angle and RMB to draw/aim weapon/focus camera.

I want to make a first person CRPG/Dungeon-Crawler in the style of Daggerfall, Ultima Underworld, and Wizardry 8, I have everything, dungeons, plot, characters, weapons, rpg system, everything has been planned out, but I don't know where to start. I don't want to use 3D models, because that complicates a lot of the development. What's the best, commercial friendly engine for a first person game with sprite-based enemies?

You can use "sprites" in real 3D engine too, just make the models flat and rotate them. There's shit ton of free for commercial purposes FPS 3D engines. Oldschool 2.5D is a problem, since not many of them were created, we just moved to real 3D too fast.

Just use pick a decent engine that supports billboards.

i can't tell who's talking.

there should be some graphical effect that emits from the character's sprite for each line spoken by someone different.

Would GLOOME work?

Didn't the developer of GLOOME stop development?

Yes due to lack of interest despite the MoM user used it mainly and constantly developing his game.

You're looking at an old version, already been fixed later on in the thread.


Doesn't sound like a good idea to use then.

Has anyone from 8ch agdg actually done that?

using the Gloome engine? Yes Memoirs of Magic user used it all the time

Godot entered 2.1 stable a few days ago. Finally my 3D game doesn't shit itself when I add more than one player character object.

It's still too slow. Having to constantly click each new text window just because I read faster than a child is annoying.

I'm a retard, that post was meant for

That's actually kind of related to his gimmick, too.

It also has live script reloading now, just how fucking rad is that?And also multiple upgrades to the previously lacking UI.

They shit on everything user, it's the same as here, everyone has individual preferences.

First, props for deciding to learn to program.
What I did first was decide which language I was going to actually use; thus the first decisions is which engine, and hence which language.
This will give you a good starting point, and will give you a direct purpose for learning everything you can from here on in.
F.e. Unity uses C#, UE uses C++, and other engines like love and such use LUA… GM uses some amalgamation of a few languages.

During the next steps, remember to not overwhelm yourself. Take it slow, and keep in mind that each concept of programming you take the time to learn will help you avoid defunk code, and hours of wasted time which could've been avoided; if you'd only known about the relating concepts (w/o having to reinvent the wheel from some perspective of not knowing how to approach the problem).
As being a competent programming is mixture of knowing your stuff, and being resourceful.

So, what I mean when I say, "don't overwhelm yourself", is in the sense of pacing what you're learning, and to take the time to learn all of the mundane, simple, and fundamental concepts without skipping anything.
Even if it seems trivial, just typing it out and getting the output desired helps you learn the concepts, adds some motivation, and how to approach a similar problem because you've put it into code before.

Once you know which language you're going to use, you'd be smart to at least learn the basic foundation of all programming languages; in addition to implementing into code or on paper while learning them.
That is, what is a pseudo-algorithm (how do you write one, and write a few out, implement them into code), what is data types, conditionals, loops, functions, methods, and it would be advisable to also be familiar with the most common design pattern; OOP (object-oriented programming).
Through this, you'll understand how to approach a problem, and how to put a proposed solution into code.

NOTE:
If you cannot find good resources, and are not too confident in learning on your own: use a site that has courses with an instructor (linda, coursera, etc), and the focus being an introduction into the language; they will go over all of these concepts.
Practice everything they give you, and do some practice on your own to see how you can break it, and how you can fix it, and how you can do it in another way.
If they do not go over OOP, I recommend doing some course on this; as you're learning how to use a design pattern, which is fundamental to having organized code, and learning how to approach a problem using said design pattern.
This is extremely fundamental, and will help you in many ways I'm not mentioning.

After you've learned this, and have become confident in your ability to code basic to intermediate programs, and how to break them + fix them; you can start learning things relating to your engine, and using the language you've been learning in-engine.

The design pattern you're more than likely going to be diving into with the engine, is ECS (entity-component system); which afaik is used in Unity and UE. So, it would be best to familiarize yourself with this design pattern, and since you have experience with OOP; you have experience in using a design pattern, and can adjust your thinking accordingly.

When doing coding in-engine, and using the language; there will more than likely be an abundant amount of guides.
Use those resources, and note that you will be ahead of the curve since you have practiced beforehand.
Learn each of the systems (going back to ECS), and how to use their components (ECS) in code with an entity; then make a practice project using these different systems.

After this, you can start, but even with this abundant preparation be prepared for defunkt code, and rethinking your approach; as it's apart of being a programmer.
Although, that foundation of knowledge will help you even recognize there is a better approach, and how to catch potential errors/bugs/etc before you move on and create a chain of ghost like bugs); in addition to knowing how to put imagined mechanics into code, and seeing those results we all want.

Best of luck user!

How? I'm the opposite: I've got most of the technical aspects down but don't know how to design the actual RPG, in terms of races, classes, monsters, spells, skills, stats, combat mechanics, etc.

Blueprints would be great if I understood it.

I cant even get the stupid engine to paint the terrain the color I want it. UE4 keep switching between amazing and terrible in my opinion

amazing

Sounds like a match made in heaven

Well I quit. Someone beat me to it, and this was the only unique idea I've ever had. Shouldn't have been so lazy and put it off for so many years.

It's very simple: get a setting, then make stuff up. Keep doing this till you have a good setting, some example dungeons, and a basic idea of what the main quest will be like. The early mechanics can be as simple as coinflips, but you need to develop it, define what skills do. Races aren't necessary, nor are spells.

So what is it?

Wut game?

WOT. C++ is the industry standard in game backends (and possibly programming in general) and most of linux is written in C++. What is their elitist alternative? x86 Assembly.

*C

store.steampowered.com/app/392080/

It's early access, and there's never enough decent games. Why not continue with what you were gonna do? My game is hardly unique from the premise, especially after having to narrow the scope to make it realistic for a solo faggot like myself to make.

Look at the negative reviews, improve upon it, and stop crying. Yours looks more interesting already, this game's battles look rather weak-feeling.

...

What are you working on m8y?

I was previously trying to redo my engine for my main project (terraria-like), because I've learned a lot since I last worked on it.

I feel like it would be easier to just start from scratch.

My shit game is progressing smoothly but code looks more and more like some eldricht incanation with every passing day and I throw inner shitfits every time I look at it. Send help.

And that's, dear Anons, the reason why you refactor as soon as you notice something wrong with your code.

It's missing "realize they are dumb, incompetent and won't be able to accomplish anything no matter how hard they try"

Wrong infographic. That one falls under "Why people think they gave up."

That's literally the second one in the pic, though.

Are there any multiplayer incremental game libraries?

...

I think that falls under "Believe in their weakness"

What language?

If it's GML then that's expected.

YOU'RE THE BEST AROUND, NO ONE'S EVER GONNA KEEP YOU DOWN

I've recently redone ballistics by implementing what another user had extracted from the Realistic Weapon Blueprints pack. Turns out that the creator of the pack had posted a picture of his projectile logic, and I just copied it. I wasn't too sure how to implement data based projectiles without any actors being spawned, so I just went along with the second best option, which happened to be this one. This one needs an actor to be spawned, but this actor doesn't handle physics nor collision through components or C++ classes. Instead, it's just a blank slate which holds the code for raycasting a short distance in each tick. The actor's real value comes from the fact that it can be used as a reference point, as the ray's start point. Obviously, it can be done in a more sophisticated way, but apparently this method does work well enough in multiplayer. I will be optimizing my current code, since casting a ray each tick hundreds of times on the server's side doesn't seem too cheap.

I've also implemented input movement based on recoil, which means that your aim will be thrown around when doing the shooty, and lastly I've also done an item swap animation. Nothing too serious.

Might not be as bad as you think. At the speed bullets are moving, they generally don't hang around long, so while I guess it depend on how many simultaneous players you expect to have, and the distances they'll be shooty across, you shouldn't have hundreds of casts going per tick.
Given a projectile speed and number of ticks in a second, you can put a hard cap on number of times to cast for a single actor.

nice spark gun

The map's scale is fairly large, and I expect the game to hold its weight with 64 players connected. Still, it's good to have some leeway. I was thinking of reducing the amount of rays casted, simply by placing a short delay. This is entirely made in Blueprints, so tick events don't work as well as in C++. I think I'll just wait and see how this plays out in a real scenario.

Ceci n'est pas definitif, ceci est un placeholder.

What language? I wanna help someone with code.

Why the fuck do your bullets just suddenly start dropping like rocks? Are they hitting the edge of the map?

added a proper minimap today
It feels nice working on features again
instead of just banging my head against pathfinding for several months.

hey /agdg/ I wanna make a game, but I'm an ideafag. Halp.

give up before you even start
t. soon to be university dropout

Refer to the second image in the OP.

Pay others to do it. That's what I do.

hmm strange it seems if i make screenshots of my game in fullscreen mode it just ends up blank
Any idea what might cause this im using sfml
anyway heres the actual image

It looked strange to me too, then I started to run towards the end of the trail to check out the traced path. It turns out it's just a problem of perspective. It curves downward just like it should, tracing a parabola.

That's exactly how my bullets work.
Neat.

Thought some kind of weaponized GPS/PDA tool would be really cool. Enhances your vision in the dark, zooms in, scans for dangers, and makes the gun instant-hit because that's what they do in the future I guess.

Feels like I could add something more to this, though. What could be some useful little features for a little spherical data assistant?

That's a good way to do it. Best you can via blueprint. Data based can only be done on C++. My recoil also works in the same way, it's the only way I really like. Same way STALKER does it.

Does anyone know a workaround for UE4's problem with scaling simple colliders at the slowest possible speed?

Hey guys, my first time posting in an agdg thread.

I've been learning for a bit, but I'm starting to get serious on working on my game now. I've been using blueprint lately and I've found it to be very comfy. I created a small personal tech demo for an idea I'm working on and was able to pull it off well enough, but it's starting to get a bit too involved to program how I originally just slapped it all together. So today I'm going to completely rewrite the entire thing and revamp the design to make it faster and cleaner, wish me luck! Ill post a picture once I'm done.

By the looks of it the two most important things, pickups and enemies, are highlighted using this vision. The only other thing I could think of would be to inform the player of things that they can't see when using normal vision, like enemies/pickups behind walls or off screen, but that would probably throw things out of balance.
Maybe it could be done by just having it notify the player when a pickup or enemy is nearby, rather than pointing them at where the pickup is off screen or through a wall.

Turning your machine gun into a rail gun doesn't make a lot of sense and may be a bit too powerfull of a buff, but whatever.

Other than that I can't think of anything that would make it more useful without adding in extra things just to make it useful, like forcing the player to hold the data assistant to open keycard doors or something, which just ends up makes the game-play less streamlined.


Good luck.

this is fun

How'd you get the water to look so nice? I'd love to steal it.

u get em

I got unlucky getting polarizing opinions, but what are Gamemaker's pros and cons?

Okay so I want to make a zombie survival game but I don't know shit about coding.

I was thinking I could probably use the code for Space Station 13, take out atmospherics and all I would have to do is learn a bash hybrid of C and spriting.

Please tell me how wrong I am.

very wrong.
That code is specific to that game. It just literally wouldn't work like that

The market is completely oversaturated in that shit, what makes you think people will want to play yours?

anons here are going to shit on you even if you were making your game in C++/sdl
stick to what you're using, try not to hop back and forth between shit

To be fair, a hilariously small amount of those "survival" games actually focus on the survival. Most of them are just shooters. In fact, Cataclysm Dark Days Ahead is the only proper zombie survival game I can think of. That and Fort Zombie, which I liked.

I was making a dystopian neotokyo vaporwave map for the Gmod server and now the thread has disappeared. I put so much effort and thought into the level now I feel it will never be fully realized and played on by my rp bros. I haven't been this legitimately depressed in a while.

It's not like I'm trying to make a commercial product out of this. Just something for my friends and I to enjoy.

I just want to play an ss13 server like colonial marines except with zombies instead of aliens, and that server doesn't really need atmos code so I don't see why I can't just do without it to make things run faster.

If you can make the meshes and textures for the level I can reuse them in my game in UE4 since that's the exact setting I'm going for. I just can't do art.

Email me at [email protected]/* */ and let's talk

Pros:
Easy to learn 2D engine
Netcode functions aren't hard to learn either
Nips make lots of hentai games with it

Cons:
Unoptimized shit engine that cannot handle complex scenes (fe: rts with 2000 units- should be easy for a modern 2D engine, right? No, fuck you.)
Made by jews (read the EULA)
Has a built-in "asset store" (see above)
Is a non-free 2D only engine (Fifty American dollars?! Are you fucking kidding me?)
GML (the scripting language) is slow and otherwise shit for a variety of reasons
Drag'n'Drop programming is completely unusable for anything serious and is only there to allow children to make pong and angry birds. Often cannot even be used in conjunction with GML (see: Box 2D physics)
Game scenes consist of finite, walled off "rooms" that you must specify a limited size for rather than figuratively infinite spaces with an origin point in the middle
Sprite editor is trash and shouldn't even exist
Missing lots of obvious useful functions/libraries most 2D games would use
Cannot even support proper transparency in 2D sprites iirc
3D toolset is so hilariously bad/underdeveloped even Godot v1.0 would do a better job
Was used to make Tumblrtale


Let me stop you right there user

Some of the cons in that outweigh the pros for me (useless features, unnecessary limitations, taking your money if you do sell anything I imagine all engine makers would want their "cut", and it'd be a pretty big cut).

The big points I want are ease of learning (though it doesn't have to be fast learning), and able to do some fancy graphical/optical effects. Not so much "accurate lighting systems" like Terraria though I tottaly would if I could, but more like screen shaking/changing the hue/saturation of everything on screen, warping everything, etc. Or complex shit, like debris that'll be made of a bank of X randomized parts that bounce around randomly when they hit the ground (reacting to collision they hit, and bouncing less each time).

Anything that can do that?

In terms of the closest thing I can think of for comparison to projects I want to do are Fin Fantasy Tactics: Advanced, or something like a classic turn based FF, but with animated battle sprites and no chibis (and animations for attacking and special moves. Rather than just raising your hands and then an effect appears over an enemy).

I don't know why this sentence completes the picture, but it does.

Gib systems are generally something you would write yourself as they tend to be rather project-specific for how they work, as well as being rather easy to write. Complex stuff like warp shaders obviously has to be supported by the engine though.

If you're looking for an easy to learn engine I would strongly recommend Godot. It's FOSS and has a lot of great features, especially in its 2D toolset. They have a lot of example projects you can muck around in, and they even have some examples for complex lighting. I'm using it for a 3D game, but just using the 2D toolset for my gui I can already tell it is a hundred times better than GameMaker.

If you've never done programming before which I'm guessing is the case you'll first want to watch some tutorials on learning the basics of Python Godot uses a language called GDScript which is effectively Python with one or two differences. Stuff like variables and their scope, functions and arguments, data types such as arrays, dictionaries, working with strings, and what Object Oriented Programming is. It might take you a day or two to learn; Python is a pretty straightforward language.

After that start watching/reading the Tutorials for Godot on GamesFromScratch.

IIRC, from what I touched of Python looked good.
Looked more "logical" than C++.
Bare in mind this was YEARS ago, so I might be miss remembering.

Thanks man,

It's worth noting Python is a much higher level language than C++, which entails it being easier to use/more human readable at the cost of execution speed.

Python being the exact opposite of that is why I dislike it so much. It's far less uniform, which makes it much more confusing to me. How the fuck is indention-based scope better than curly brackets?

Unless they made a shitload of progress in the past year or two, this is simply not true. But then again, looking at how Krita went from garbage to literally the best digital painting software available in such a short time shows that it's at least possible.

I don't really like it to be honest, but it forces you to keep your code readable.


That's precisely what they've done. GameMaker is shit and will always be shit, because unlike Godot it is designed from the ground up to be shit.

Shit. I guess I should have waited for Godot.

Nice try Carlos, but that's precisely what the engine is named after the endless wait for new features to be added

Not visldeogames. Fuck off gook.

every modern C++ IDE auto-indents for you.

Python is shit for retards.

I really wish people would code their own games from the ground up, at least the ones doing simple 2d shit like VNs, side-scrolling platformers/shooters/whatever, top-down RPGs, etc. The thing about engines is, they really only do about 80% of the coding work for you. Which 80% varies by engine, but you really can't make a game without some programming skills. Artfags trying to hack prefab objects to do shit they were never intended to do, and engines making their prefab objects and interfaces ever more bloated and retard-proof are why game performance is declining, bugs abound, and resource consumption is at an all-time high despite the final game looking like utter ass (see nu male's sky for the latest and most public example of this).

Nobody said to code your own generic and reusable engine ala Unity or GameMaker, just use a library like SDL/SFML/GLFW/etc and write the few more functions and objects your shitty little game will actually use. You'd be surprised at how easy writing your own ClickableWidget or AnimatedSprite can be, and what you'll learn in the process.


Python is trash. The few months it saves you in time to learn/master over an actually useful language like C++ will never make up for the shit code you squeeze out for the rest of your life. Low and high level languages aren't even mutually exclusive either, see: C/C++ and Lua/LuaJIT, you know, shit used commonly in real games.

Wow, cutting corners and expending minimum effort already. Link to your kikestarter please?

I'm working on a 3D flying game in Unity. I already have some ideas on the intended flight model, but I haven't implemented any code yet. If I want to have the aircraft line up with the cursor, is that gunna require me to understand quaternions, or are there easier methods? Also, should the crosshairs be along the same forward axis as the aircraft, or vertically offset? I currently have the crosshair appear to line up with where bullets may land in the distance, is this something I should wait to worry about until after I get a working flight model in place?

On the topic of programming, and just out of curiosity, has anyone here tried making a game in LISP or tried integrating it?

Entity-component systems are neat. That is all.

It seems cool at first, but if you're making an engine I read it fucks you in the end. Although I'm buying hard into the data oriented design meme.

it's like a nigger rorschach test

???

I am going to fucking kill whoever did this.

I only ever used Python as a scripted language together with normal program in C++ and it was very good for that, same with LUA. Never had an occasion to write something just in either of those languages. I know python (at least that's what i've heard) for webdev and now whenb thinking about it, I'd rather write a website in Python than PHP, but I'm not webdev so don't listen to me

It's pretty clear what it means.

A company with no understanding of the market is hoping that it can get unpaid interns to make a hit for free.

Gais I got it! I'll make my game in RPG maker!

I went braindead for a second and mixed it up with something else I heard which was basically an entity object and pointers to components within the objects. The OOP version of component system. So rather than doing a giant array of each component you would go through each entity and use their component.

Okay.
I was just curious, as data oriented design and ECS would typically fit together quite well, since entities don't really contain their data, leaving you reasonably free to organize it as best suits optimization.
Actually, it sounds like what you're describing is pretty much standard OOP.

Lua because Love is just comfy. I just lost the ability to write nice looking code after years of using nothing but Fortran.

I'm trying to replicate this lighting effect from a source mod, I managed to use it in a TTT map ages ago, and it seems that you need to make an inside box in the wall with a reflectivity texture on the inside, with a light inside of that inset, but not immediately visible to the player.

I tried it with unity recently with a standard texture, but I can't get anything along the same lines. It's like the light will literally pass through fucking walls. What the dicks.

i couldn't even get it to used baked lighting correctly

Well shit. Are there any unity asset store lights that are good?

That's why we need more people to help with ENIGMA/LateralGM and free ourselves from the Joos.

is ENIGMA/LateralGM gud for non programmers?

It has the usual drag-and-drop features of GM plus EML, which is essentially GML mixed with C++. You'll be fine.

Sounds like you're thinking of just plain ol' composition. Which can get hectic if there's a cobweb of message passing going on.

Then again, just about everything can fuck you in the end if you take enough wrong turns somewhere along the way.


Do people still play Zombie Master? That map was a pretty good one, if I remember correctly.

Sadly no

;_;

user, look at pic #2 and think about why it seems like the light is leaking in from the outside. You are using planes for the walls. The difference between a plane and a cube/cuboid is that a plane has no depth and therefore is only one-sided. Of course your directional light (sun) is going to leak in, because there is nothing stopping it. Why do you think you can literally look through the walls in the editor? Because there aren't any on the outside.
You need to either remove the directional light, make your walls have depth, create an artificial shadow using another object or surround it by a bigger object altogether.

For those stripes you want to use a mixture of emissive material and area lights. Area lights need to be baked and are quite slow to do so. They look great though. Point lights are a stupid choice for that effect, and highly inefficient at that.

The problem isn't with the directional light leaking in. It's with the point lights leaking in.

The point lights which happen to be within the bounds of the level geometry. They're phasing through the walls, rather than actually bouncing to reach those walls.

Shit, didn't read this part until just now. I actually looked through this beautify plugin, which makes emissive lights really nice looking, and I'm actually going to give that a shot. But what are area lights? I don't believe I've ever heard of them.

Thanks

does anyone with a head or two knows how the fucking animationplayer work in godot?
its fucking stupid, I cant even set the duration for each of my animations?

you should have just used unity, stupid goi

unity is dying and is a sick meme

but how the fugg do i use godot

use unreal :^)

i dont trust chinks as I know what chinks are because I am one

also 5% royalty

but muh superior engine
it's good goy, trust me

and I also want it to run on actual toasters, with opengl

good luck with your game, i'll be happy to play it when it comes out in a decade or two :^)

If i ported my text, sprites, objects… etc from a gamemaker project, would it work?

If you want free low end FPS 3D engine you can use CUBE 2. One of the nice features is the in-game level editor that works in coop too.

well I'm not making a shooter, I want to make a ninja gaiden-like game

I'm sure you could make any game using it with some small changes.

That triggered my nastolgia

It's a nice game, yes. And it's still active online, at lest the CUBE 2 a.k.a Sauerbraten
Good arena shooter, they have nice insta gib mode.

Okay, I gave emissive lighting a try, it's not the effect I wanted, but it looks cool

Okay, I think I've nailed it now.

Fuuuuuuuck, Cube. I thought that thing was dead.
I was looking forward to Death Illustrated, back in the day.

This could be the basis for an interesting idea, actually.
The first thing that comes to mind is small text displays labelling important areas/doors with text/pointers that can only be seen when the visor is on.
Like a variation of the mission objective marker, except you have to actively go out of your way to use it so it's not so handholdy.

Thanks, user.

I never actually played cube myself. I played a knockoff called assault cube all the time and remember just sitting around in the map editor all day sometimes co-op. It was like minecraft before minecraft

Point lights are light sources which are represented by points.
We're talking about points in the mathematical sense here. Points in space have no size. They are infinitely small. This makes the lighting equations simple and thus blazing fast, but also inaccurate, as light sources in real life do take up space.
Imagine that you're standing on a flat surface and there are an object and a point light floating around. If you look at the point light and you see it, you'll be standing on a lit part of the surface. As you move underneath the object, it will obstruct the point light at some point. At that point you'll be in the shadows. As a result you get something called "hard shadows", in which means that the shadow dropped by an object is either there or not. "Soft shadows" are a workaround for this, in which you blur the hard shadows you created. They're still far away from the real thing.

Area lights use a more realistic lighting function, in which your light is emitted not from a point in space, but from a surface (as is the case in real life).
If you replace that point light from earlier with a fluorescent tube, the situation will differ. As long as you see the light source in its entirety, you'll be standing on a lit surface. As you move underneath it, your object will start to obscure more and more of it. That means that your shadows will get darker and darker gradually. You'll be standing in the "penumbra". At some point you won't see your light at all anymore and you will have reached maximum shadow intensity. You'll be standing in "umbra".

The problem is that you can't do this sort of lighting in realtime, so you need to bake it.

Keep in mind that emissive materials only create indirect light (realtime GI -> only static geometry). Also make sure to adjust the ambient light and reflection intensity in the lighting settings of your scene. That white haze towards the end of the corridor doesn't look good.

There's ton of games or mods made for it.

Thanks! I've removed the directional light, and tuned the settings a bit, removed the fluorescent tube, and added an area light, and baked it until I got the settings I desired.

I'm really loving the glow coming out of the inset from the wall - and that is what I was aiming for all along, But why is the inside of the inset still riddiculously dark?

I kept one of the inset unlit for comparison, strangely enough the light hits the other side and almost makes both sides equal.

I really appreciate your help by the way.

Area lights only emit light in one direction. What you want to do is add the emissive stripes again, but set their GI mode to none. That way they will glow without lighting the surrounding objects up. Make sure that your glowing stripes do not cast shadows, as you need to place the area lights slightly behind them (5 cm should suffice).
Oh, and depending on how many of those lights you have, you might have to use deferred rendering, preferably with linear color space, as it looks better. Both can be changed in your player settings in the "other settings" tab.
I quickly put together a test scene to make a screenshot.

So I have a account at cgpeers now, well for what should I am looking for when I want to git gud at blender and texturing?

Game maker is as powerful as much time and effort you put into it. You can write your own custom .Dll's to get by its limitations.

the asset store has some nice tools that help you build the game.

its bad but its good for making basic sprite, you should edit the sprite in a 3rd party app like gimp or Photoshop.

i dont know what you are talking about, it already supports transparency.

i got mine for like 6$ on humble bundle.

Awesome, I see now!

Why does pálinka look like sperm in the pic?

I…I… don't know I din't made the pic I swear man.

I'm starting to be a bit angry. Would you please kindly change the colour to water-like?

Alright alright here…

i don't like it, change it back to white

Imposztor!
Don't believe his lying ass
Change it more like waterish
You can see from the pic

...

wops missed a few spots

Yeah, feels a lot better, thanks
Now let's not derail the thread any further
Many thanks

t. a now calm Hungarian

damn magyars

.t polfag

Convince the black magicians of Resuource and Chorus to release new demo god damnit !

>that wasn't the bug
>fixing it made things worse

Damn? Are you sure you are from pol?


Can't, I drank all the pálinka I had

tak kurwa, co wątpisz, suko jebana ?

Chorus released demo last year, but Resource does only coops :( and Focus is dead since the 90s.

A google fordító erejével felfegyverkezve, te manók által seggenbaszott hímkurva, megértem a faszcibálásodat, és méltóképpen válaszolok neked, te gennyes trágyahalom. Anyádba szart egy tehén, amikor meg akarta dugni, mert imádja a tehéngecit, és megszülettél te. 3 bölcs bögöly jött étged üdvözölni, és ajándékként láthatólag agyon és faszon csíptek, mert kevesebb értelmed van, mint egy teherautó által kilapított sünnek, te mocskos ogregecit nyeldeklő, körbebaszarintott kacskafarkú tajnigger köcsög.

Látod, én is tudok káromkodni a magam nyelvén, nem csak te vagy ilyen ügyes.

Wait, does it have better 3D capabilities nowadays? First time I used it it still crashed non stop and missed a bunch of functionality GM had at the time.

Ez csak egy vicc, én még mindig szeretlek: *

Hozom a kolbászt, paprikát, hozd a pálinkát, és összeborulunk, felem ;-)

Hungarian is to weird for google translate and I'm opening 6th beer.

"I bring the sausage, red pepper, you bring the pálinka, and we get along well together, brother"

BTW. Can you translate this ? It's one of my favourite songs

Sorsod olyan akár a szikla
Rideg és kemény
Pengeként hatol előre
Az idők tengerén

Tépj szét minden láncot
Mely benned él
Őrizd a Metál lángot
A holnapért

Kevés volt a hangom a szóra
Kell az üvöltés
Kicsiny volt szívem a jóra
A harchoz túl kevés

Tépj szét minden láncot
Mely benned él
Őrizd a Metál lángot
A holnapért

Űzött vadként járom az utcát
Senki nem figyel
Lökdösnek jobbra balra
Ez rég nem érdekel

Tépj szét minden láncot
Mely benned él
Őrizd a Metál lángot
A holnapért

Sorsod olyan akár a szikla
Nincs már visszaút
Pengeként hatol előre
De kísér még a múlt

Tépj szét minden láncot
Mely benned él
Őrizd a Metál lángot
A holnapért

Whats the song name? I would like to listen it.

Here's full album, it's the first song. BTW, this is my channel, I upload rips from my vinyls, tapes and CDs, mostly polish, czech, russian etc. metal

Ah thanks, looks interesting. bookmarked it because I don't have a ZOGtube account.

Sure

Your fate is like a rock
Rigid and hard
Penetrating forward as a blade
On the sea of time

Rip all chains
Which live in you
Guard the Metal flame
For the tomorrow

Slim was my voice for word
The roar is needed
Small was my heart for good
For the fight it's too small

Rip all chains
Which live in you
Guard the Metal flame
For the tomorrow

I walk the street like a hunted boar
No one minds me
They shove me left and right
In this I'm not interested long ago anymore

Rip all chains
Which live in you
Guard the Metal flame
For the tomorrow

Your fate is like a rock
Rigid and hard
Penetrating forward as a blade
On the sea of time

Rip all chains
Which live in you
Guard the Metal flame
For the tomorrow

Mind me, this is on the go, so a few parts might be not exactly correct, but they bear the same meaning

That's fine, it's way cleaner than whatever google translate says. Many thanks. There's many songs that I'd like to understand, not only Hungarian, but I'm not gonna shower you now in lyrics.

Yeah, thanks for not showeing ;^)

And no problem

repostan this from muh shitpost account:

Hmm I just have a idea, I think I am going to add APC like vehicle, such as the Kangaroo or w/e it was used by the canucks.

The APC functions like this, it has maybe 10-20 troopers or so which spawns various types such as Assault Rifle, SMG, and AT-Rifle squads. And after it has spawned a few it won't spawn anymore. I am not sure if its also possible that the APC can carry injured troopers and heal them then later release after a minute or so.

That also means I won't add any Arch-Vile like enemy and none of the corpses are resurrectable.

Regarding the flashlight, I hope it doesn't cause any major lag during Multiplayer because it is controlled via ACS which I try to avoid at all cost because when the ACS is updated it WILL break save games. So when I do a simple update even just changing a few values it will break due to the way ZDoom handles ACS.

And no I couldn't find a way to make a Flashlight system purely based on Decorate.

Also some of the enemy vehicles will have special ability, the SU-100 for example will have a limited amount of APCR shots which causes more damage and after it is depleted it has a cooldown period of about 5 minutes before it can use the round type again. You better destroy it quick before it fucks your shit up.

At the very least I can give you another kickass Hungarian tune, but this time in english. Tell me how Hungarian is his accent hehe.

Well, there was a migrant Hungarian mathematics professor, teached on some American uni
Even I couldn't understand his english

The guy in the vid is better than some of my old classmates, even while he is rattleing

I guess that's excusable considering this is from 1988. IDK if you listen to metal or not, but if you do, you might recognise this voice

he did the vocals on Mayhem's De Mysteriis Dom Sathanas

Give me something to fucking do, nerds. I need to texture some shit. Like I said before, let's make something. Even if it's just a Dota 2 item or something.

Nah, my friend listens old hun metal
I'm more of a jazz person

I have a untextured tank, want to do this? Through the UV Mapping is shit. I used auto-uv unwrap

On the topic of accents and professors:
Fuck my life.
I have to learn for an exam and YouTube isn't a big help. Almost all the videos are made by Pajeets. Now, I don't give two shits where people are from. What I do give a shit about though, is whether I can understand a word they're saying or not.
I've heard some pretty thick accents in my life, but strong Indian ones are the worst. They're borderline impossible to understand at times.

Clean up your UV and I'll do the textures + all the other maps.

Like this?

Alright I will try to do a proper uv-mapping tomorrow as today I am busy with something else.

same im too lazy to finish it.

Man, this reminds me my levels are hopelessly incomplete and they need serious work.

Pretty much.

I hear devlepper far too many times a day, working in tech.
All curryniggers seem to pronounce developer and development as devlepper and devlepment, stress on the first syllable.

Some accents are hard to penetrate and understand, but curryhouse creole is less clear than bix nood, the intonation and pacing doesn't make any goddamned sense, sporadic rolled R-s come and go, even in the same words, and the very sound of it is as flattering as a crying baby.

ayy lmao. I never interacted with poo in a loo, but it sounds hilarious

Hard to say. Out of all games by Daniel Remar that I've tested only one compiled successfully without me having to do any modifying the scripts.

Thank you for saying this, I can confirm that Indians cannot say the word "Development". There was a board game meetup I would go to and we had an Indian guy there and in one of the games we played there was an upgrade called "Develop", and he would always say "Devlop". At first I thought maybe he was just a little retarded or something, but I asked him what he does and of course, no joke, "Software devlopment".

Since then, hearing him or any other Indian say it drives me a little nuts, kind of a pet peeve of mine.

:30

youtu.be/2RynIEZJctw?t=27s

I got better gun flashes and finished the suppressed ruger that someone here suggested, but because it's such a small bullet moving so slow it takes a lot of shots to kill one of the robots. This will be fixed when I add areas on the robot that crit.

Next on the list is a Revolver and Suppressed Sniper Rifle.

fugg

why can't I hear the trigger

That's nothing fam.
For some reason this guy's videos were always recommended to me by youtube even though they usually have under 2000 views and are usually only really vaguely related to things I watch. Eventually I gave in and watched one, and was fascinated by how utterly incomprehensible he is, and how much content he shits out anyway.
Now sometimes I let his videos run as background noise while I do other stuff.
He says English is his native language, by the way.

The warping from the muzzle flash is a bit distracting tbh. Maybe tone it down just a bit

Easy enough to do. I made it visible if you are looking for it, but not terribly noticeable if you aren't. Turned out I needed to turn it's prominence down by a factor of 10. A little more than 'just a bit' but I think it's better now.

I made things. Sorry for the absence.

Nice to see you back

Welcome back, dude.

No apologies necessary… but you can post lots of progress if it will make you feel better

bump

We're past the bump limit, boy. Here there be monsters.

Nice to see her backside.
Did you finally get internet again, or was there some other problem?

oh jesus, i remember seeing this pajeet claiming that adobe lumberyard was the next big game engine

your waifu is really stuttery when she's moving on stairs
isn't their just an invisible plane collider there to work as a slope rather than multiple box colliders?

New OP forgot to link new thread in here, I guess I'll do his job for him