[a / b / c / d / e / f / g / gif / h / hr / k / m / o / p / r / s / t / u / v / vg / vm / vmg / vr / vrpg / vst / w / wg] [i / ic] [r9k / s4s / vip / qa] [cm / hm / lgbt / y] [3 / aco / adv / an / bant / biz / cgl / ck / co / diy / fa / fit / gd / hc / his / int / jp / lit / mlp / mu / n / news / out / po / pol / pw / qst / sci / soc / sp / tg / toy / trv / tv / vp / vt / wsg / wsr / x / xs] [Settings] [Search] [Mobile] [Home]
Board
Settings Mobile Home
/g/ - Technology


Thread archived.
You cannot reply anymore.


[Advertise on 4chan]


Super Mario 128 edition
/gedg/ Wiki: wiki.installgentoo.com/wiki/Gedg
IRC: irc.rizon.net #/g/gedg
Progress Day: rentry.org/gedg-jams
/gedg/ Compendium: rentry.org/gedg
/agdg/: >>>/vg/agdg

Requesting Help
-Problem Description: Clearly explain the issue you're facing, providing context and relevant background information.
-Relevant Code or Content: If applicable, include relevant code, configuration, or content related to your question. Use code tags

Previous: >>101337514
>>
>>101359645
MinGW (not w64)
By modern I don't necessarily mean the very newest one and I don't use c++ anyway so lacking new features doesn't matter
>>
>>101359693
i don't get why soo much hate for FXAA is far better than TAA and has no ghosting
>>
File: GSKJjcCaUAQpglp.jpg (495 KB, 1390x2048)
495 KB
495 KB JPG
Hullo. Absolute beginner here. Are there any books that go in-depth on SDL using just C (I have one using C++ that's perfectly readable so far but it still warns about learning polymorphism et al)? Or should I learn C++ anyways even though I won't use any of its features?
I'd also appreciate any resources not already mentioned on the wiki. I just want to start fucking around instead of being stuck on a bunch of books forever. I can look shit up.
>>
Would it be possible to make something like AI Dungeon but ran locally and with more RPG elements?
>>
How does one develop a 2D engine using C#? Can I use Vulkan or DirectX?
>>
>>101360275
C# has some good 2d game libraries like mono game or XNA (FNA)
>>
>>101360160
We can't you just adapt what you're learning in the book that uses C++ to C?
>>
>>101360275
monogame
>>
>>101360160
>books
lol

https://lazyfoo.net/tutorials/SDL/
>>
>>101360286
>>101360434
Those are great frameworks but I want to build it from scratch.
>>
>>101360160
>>101360433
Like what I mean is how you learn SDL should not be language specific as it's designed to work with many languages.
>>
My end goal is to create something akin to counter strike. Obviously not as grand but at least playable. I have fond memories from playing cs 1.6 and always wanted to know the how's and why's.
I started OpenGL, have been steadily learning it and once I have some good knowledge about it, will start on the actual game.
For 3D what is a good framework/library that doesn't abstract as much?
>>
>>101360538
CS was just a Half Life mod, why don't you just make a mod?
>>
>>101360460
use Silk.NET in that case
>>
>>101360538
raylib is very simple and handle everything 3d related and is faster than SDL
>>
File: GSF3KMlaUAI0i8P.jpg (126 KB, 832x1216)
126 KB
126 KB JPG
>>101360433
>>101360469
That's definitely the plan. I was just wondering if there were any out there I missed, just in case.
>>101360444
On the list, though thanks.

Time to learn I guess.
>>
>>101360616
Why is SDL so slow? It gets a lower uncapped framerate by a few hundred than GLFW even when both are just used to make a window and GL context
>>
>>101360697
>lower uncapped framerate by a few hundred
This is an absolutely meaningless measurement
>>
>>101360697
https://www.youtube.com/watch?v=QO4zLdAWJsw
no idea why, here are some test made by a user
>>
Should I be multiplying vertices by the MVP matrix on the CPU or GPU?
>>
>>101360721
any per-vertex or per-pixel operation should be on the GPU
>>
>>101360706
It still proves that SDL somehow has lower performance.
>>
>>101360748
No it doesn't, you have no clue what the fuck you're talking about
>>
>>101360796
>take game that uses glfw to create a window and opengl context
>replace the part that uses glfw to create a window and opengl context with sdl creating a window and opengl context
>uncap framerate on both
>sdl gets a few hundred fps lower than glfw despite rendering the exact same thing
stfu you stupid nigger, YOU have no clue what the fuck you're talking about
the glfw version also had input btw i didn't even enable sdl's input system
>>
Is it true that Epic doesn't care about Unreal Engine anymore? And that they're devoting all their focus to UEFN?
>>
>>101360857
The behaviour of a graphics window with an uncapped framerate and nothing going on is completely undeterministic
>>
>>101360886
Then why was it consistent that the SDL build had a 300+ lower fps than GLFW every single time I tried it
And this wasn't a title screen doing nothing, it spawned into the same 3d test level in the same exact spot
>>
>>101360910
>300+ lower fps
First of all, FPS isn't linear
400 vs 700 is completey different to 5000 vs 5300
>>
>>101360275
OpenTK
>>
>>101360160
>I can look shit up.
You can't look up something you never heard about. At least read a brief overview of algos and data structures, because without understanding big O you will write cringe that will be painful to look at in the future.
>>
>>101360629
eli?
>>
>>101360721
People like to do model-view on the cpu
>>
>>101360868
That’s not how it works. Unreal is their main product, the company doesn’t exist without it. However it’s my guess that they want to shepherd indies and inexperienced people into UFEN and their “metaverse” ecosystem. Timmy wants to be in the same position as Gabe.
>>
File: 1.jpg (59 KB, 932x396)
59 KB
59 KB JPG
>>101361689
Can't be assed to dig for recent statistics, I doubt that something changed significantly unless proven otherwise.
>>
>>101358764
what's the problem with this?
>>
>>101363455
As a rule in life in general, ignore twitter screencaps, and twitter posts themselves, unless they are made by someone you explicitly know about and care about the opinion of.
>>
>when what you want to do requires multiple integrals and projecting large semitransparent meshes
>>
>>101359693
how many lines of code can I expect to write roughly for a simple exploration game that only has a handful of key items, something like babbdi or yume nikki (except 3d)?
just want to gauge the size of the endeavor so I can estimate how long this might take me

similar question if that question is too stupid: how long do you reckon it took to make babbdi?
>>
>>101361034
Can you suggest something to read?
>>
>>101364004
in what engine?
>>
>>101364012
The easiest book is Algorithm design manual. One of the early chapters has some math, but you can skim it and the rest of the book is practical. Provides good coverage, so I recommend starting with it.
The advanced books are Algorithms by Sedgwik and Algorithms by sir Sanjoy Dasgupta (jokes aside it's ok).
The in-depth option is CLRS, but it's not a light read. Useful to have as a reference book anyway since it provides pseudocode that's easy to adapt into actual code regardless of language.
>>
>>101364055
godot
>>
>>101364074
10,000
>>
File: IMG_9552.jpg (1.04 MB, 1170x1551)
1.04 MB
1.04 MB JPG
>>101364004
found my answer:
>>
>>101364082
10000 LoC is honestly not bad
doable in 2 months for me Id say
>>
>>101364112
If you already know what to write, that's it.
>>
>>101360134
temporal niggers only look at pretty static screenshots
>>
>>101360134
FXAA works with less information, it's objectively inferior
>>
There's no Thinkpad General atm, I hope this isn't too offtopic? related to gamedev.

TLDR: I'm broke and want a decent laptop for gamedev -- I'm interested in 2D but want to be able to switch to 3D if I change my mind (will most probably only use Godot).

Is this good enough for 400€ ?
Thinkpad P51 Workstation:

Intel i7 7820HQ
DDR4 32 GB
SSD M.2 512 GB NVMe
NVIDIA Quadro M2200 4 GB
IPS 15.6" 3840x2160 4K X-Rite PANTONE
>>
>>101364748
define gamedev, running gcc from nvim and working with UE5 need slightly different hardware
>>
>>101364790
A game probably a tad more complex than Hollow Knight, and like I said, I will probably not be using Unreal...

What do you think is lacking for UE5 use here, specifically? The graphics card?
>>
>>101360160
>I just want to start fucking around instead of being stuck on a bunch of books forever
then why are still looking for books lol?
pick up and work on a project and you'll find out how ez or hard it is. simple.
If you spend just enough time to get comfortable with SDL docs, you won't be needing any books. Its not about doing whatever the exercise the book tells you to do step by step. Its more about understanding how SDL library functions and using it to get your job done. Because books can only get you so far, you'll ultimately have to get your head around whatever API you are dealing with, be it SDL today or Raylib tomorrow.
You can start with something as simple as drawing a cricle or a square on the screen. Then move them around the screen and finally take things as far as your time allows you to, like making UI widgets or a small game.
>>
File: cheer-up-lets-go.gif (373 KB, 370x300)
373 KB
373 KB GIF
>>101364112
noice.
don't give up if you get stuck, anon-san.
Keep at it.
>>
>>101364832
>What do you think is lacking for UE5 use here
Non-laptop CPU, because you'll get tired of compile times.
>>
>>101360718
> hey look at my benchmark !
> things doesn't seems to run exactly the same way
> do not release the code so people can check if it's properly implemented

>
>>
openGL version shitposting and sdl vs anything should be a bannable offence at this point.
>>
>>101366030
U mad about SDL2chads and OGL1.1chads?
>>
>>101366077
I wish those "chads" would post actual progress
>>
File: totally-fair.png (66 KB, 831x1528)
66 KB
66 KB PNG
>>101359693
Gaming rn
In my limited experience I find forth to be best used with pen and paper
>>
>>101366381
please no more roleplayers
someone post a real game they're making
>>
>>101366381
f#
thats a first i think
>>
>>101364012
>Can you suggest something to read?
Tranime
>>
>>101366401
I was, in fact, making a capital G game until life happened
For now, enjoy a small slice of what could have been something
My experiment with forth was to assess whether I could use it as a scripting language for which I could write some kernel for and fuck off to in peace
>>
>>101366401
i will stop playing dota and dev for you anon
>>
>>101364012
>>101351548
>>
>>101360857
Don't measure by drawing a simple primitive. Take a look on the frame-time, not FPS. Check the results.
>>
>>101366985
To get meaningful measurements you need to give it a significant workload, like draw 10,000 sprites then measure it
>>
https://medium.com/@david.dehaene/delta-rollback-new-optimizations-for-rollback-netcode-7d283d56e54b
>>
I only care about developing 2D games for Windows right now.
Win32 (GDI and or DirectX), or SDL2?
>>
>>101367239
sdl2 https://github.com/JIghtuse/cavestory-sdl2
>>
File: trash tesselation.png (30 KB, 370x370)
30 KB
30 KB PNG
progress? made a uv sphere and a very retarded subdivision function
>>
>>101360160
read https://wiki.libsdl.org/SDL2/Introduction
keep this link somewhere : https://wiki.libsdl.org/SDL2/APIByCategory
install it following https://lazyfoo.net/tutorials/SDL/01_hello_SDL/index.php
check https://www.parallelrealities.co.uk/tutorials/
from time to time check the API to see functions and structs in details.
practice a lot.
>>
I’m thinking of making something with OpenGL 1.1 to see how much I can bend it.
>>
What's the status of SDL3 3d rendering? I remember it was one of the big features when it was announced.
>>
>>101367828
iirc Ryan or someone started working on it but it's nowhere near completion. I imagine it'll be added in a minor release of SDL3 sometime later.
>>
>>101366246
kek
>>
>TIL:the only sane way of handling DLCs is by having lua support for the main menu
That's one way of showing DLCs, but how do you actually handle multiple DLCs code structure wise? you can only do so much by replacing the main executable and load new assets.
>>
>>101368255
Don't all devs just have a single codebase for all DLC, and everyone has the DLC already on the computer, all "buying" the DLC does is turn on some logic that enables the content

you don't actually have to maintain separate versions of the game
>>
>>101367828
they really want to support 3D rendering ?
I was reading about it few hours ago and thought it was mostly a version that would add more device support, camera support, HDPI, and a new audio API.
>>
>>101368522
do they? last time I've checked there was one guy working on something called SDL_GPU which was very similar to raylib
>>
File: 1698452236296825.png (63 KB, 1045x624)
63 KB
63 KB PNG
>>101367828
>>
>>101368522
github.com/libsdl-org/SDL_shader_tools/blob/main/docs/README-SDL_gpu.md
>>
I would like to ask something ridiculous. If there is something within an engine I find unique enough to patent, but I still want to let people use and modify it. Should I bother to register?
>>
>>101369384
>>101368545
So it will be a replacement of SDL_Renderer for 3D libraries like Vulkan or openGL if I understand correctly. Nice.
>>
the fuck is the difference between a StorageTexture and just a regular Texture binding? How do I use one?
>>
>>101371026
saves you from binding. there are more functions that does the same with other objects, it lets you have a bindless renderer outside of drawing.
https://github.com/fendevel/Guide-to-Modern-OpenGL-Functions
>>
>>101359693
Got a few questions for you guys
1. I hear people saying ready made engines have a certain "feel" to them and I kinda agree. Unity games are unity games, unreal games are unreal games.
But what is the feel, what inside the engine give s it its feel, and how do I make sure my engine has a unique feel to stand out?
2. Getting filtered really hard by Vulkan. Any suggestions? I did the tutorial, the vkguide, some rando youtubers tutorial and I still don't understand half of it.
3. About C++. I'm new to it. Do you guys use the debugger for debugging? I'm used from other languages to just use print for debugging, is learning the debugger really necessary for CPP or is print enough?
>>
>>101367750
That's all I use, but I'm just making a 2d game.
>>
>>101371636
>3. About C++. I'm new to it. Do you guys use the debugger for debugging? I'm used from other languages to just use print for debugging, is learning the debugger really necessary for CPP or is print enough?
i use both. for simple debugging printing is enough, but when shit hits the fan gotta bring out the big guns
>>
>>101370472
If you can't afford to pay to defend your pattern, you shouldn't bother getting one. Software patterns also only work in some parts of the world.
Just license it under a license you like.
>>
Really want to dip my toes into C++ game dev. I primarily cut my teeth on C# and haven't touched C++ since college. Would a Tour of C++ be a good way to get up to speed and then just pick a library/framework to work with?
>>
>>101372181
grandpa c++ os old and defunct
>>
>>101372181
listen to >>101372209
Go straight to C and then have fun with whatever library you want. SDL, Raylib...
Just try getting shit done unlike us
>>
Will this be worth picking up?

https://youtube.com/watch?v=84Lu1kQ-NU8
>>
>>101371636
>what inside the engine give s it its feel
Lots of trivial things like input latency, stuttering, specific graphical effects, the UI. I don't think it's something you should worry about, if you make your own engine it's gonna feel different regardless and it's not something anyone with a brain cares about
> Getting filtered really hard by Vulkan. Any suggestions?
Use OpenGL instead
>is print enough?
Step through debuggers are very helpful when your're debugging large, complex systems. If I've written something that's several thousand lines of code and I'm running it for the first time I always use a debugger to catch the initial mistakes I've made
>>
>>101370472
The entire point of patenting something is so that aren't allowed allowed to use it
>>
>>101373604
>Use OpenGL instead
No sorry, I will stick with Vulkan
I don't even care about making a game anymore
I don't even care about a making an engine anymore
All I care about is figuring out Vulkan
I will do the tutorial 1000 times if I have to
I will learn the documentation by heart
I will dissect every FOSS Vulkan engine ever made
I WILL master it or die trying
>>
>>101373106
Raylib actually added optional SDL support, so I'll fiddle around with the cpp wrapper for a bit.
https://github.com/robloach/raylib-cpp?tab=readme-ov-file
>>
>>101373677
That's retarded
>>
>>101371636
>But what is the feel, what inside the engine give s it its feel
That is entirely up to the design of the game. Same game made in different engines can give the game feel.
>and how do I make sure my engine has a unique feel to stand out?
its not ez. All existing engines probably already inherit everything that is worth so if you want to stand out, the only way you could do that is by providing more than you competitors. As in your engine having superior user experience than your competitors. Imaging having a better procedural generation of land and possible assets, rigging and animation with easy to understand and edit UI, lighting, PBR, volumetric fog and all than currently existing major engine. That's simply a lot of work and demands a lot of prior experience to pull it off.
>Getting filtered really hard by Vulkan. Any suggestions?
Just practice triangle over and over again. Only after you are comfortable enough with it, go for handling descriptor sets, textures, depth and lighting. Don't take a step further until you are able to do a triangle without having to look at the docs. This will clear a lot of things early on. I'm currently making a UI framework and I'm doing it in two draw call. I'm drawing all the UI into a texture first and simply displaying that texture onto the screen, similar to how wayland does it it. Vulkan isn't easy but the amount of control it give you is worth it. Don't give up and keep at it, anon. Its only a matter of time.
>3. About C++. I'm new to it. Do you guys use the debugger for debugging?
Sorry, I don't do C++ anymore but when I used it, I always used my own logger and profiler. If you are on windows, visual studio is the best you got.
>>
>>101373937
>Sorry, I don't do C++ anymore but when I used it, I always used my own logger and profiler. If you are on windows, visual studio is the best you got.
what are you using if not C++?
>Just practice triangle over and over again.
Yeah good advice I will do this
>>
File: logo-slim.png (73 KB, 500x260)
73 KB
73 KB PNG
>>101373954
>what are you using if not C++?
I tried out a few langs and settled on Odin. Modern C++ is a mess. Rust is annoyingly painful. Zig is a cleaner version of C++ but nothing like C.
I like C but its a tedious language for 2024. I always wished C spared me from header files and came with lots more stuff out of the box. Odin lang was exactly it for me.
>>
>>101360198
Yeah definitely. No idea what AI Dungeon is though.
>>
File: 2024-07-11_22:23:49.png (140 KB, 1920x1080)
140 KB
140 KB PNG
I should wrap some of these function arguments in a struct. But I won't ...
>>
>>101374018
>Modern C++ is a mess
lol okay
>>
>>101374751
That's not a controversial opinion, pretty much everyone agrees with that
>>
>>101373677
Best way to learn vulkan is to learn opengl yah doofus
>>
>>101374821
The best way to learn Vulkan is to learn Vulkan, but nobody in their right mind should be using Vulkan
>>
>>101371636
Don't start with Vulkan, especially if you're new. It's aimed at AAA game and framework/emulator/DXVK etc. devs. OpenGL should be more than enough for your usecase.
Debugger is more convenient than print once you get a hang of it. You can check any of dozens objects state without having to write and then delete prints and recompiling multiple times, and do it at any part of code execution, and not only during the print statement execution. The only use for print is checking out how some value changes in real time, but then you can just show it on screen once you implement text rendering.
>>
File: IMG_4933.png (14 KB, 600x371)
14 KB
14 KB PNG
What went wrong?
>>
>>101375964
I don't think Vulkan is used by any consoles, DX12 gives you Windows and XBox
>>
>>101375994
I believe it's the Nintendo switch's "native" API.
>>
>>101367357
noice. Trash but still, looking good.
>>
>>101364102
Maybe even less than 10k lines in Godot.
>>
File: c0c8yclcovq71.jpg (205 KB, 2442x1016)
205 KB
205 KB JPG
I'VE ABANDONED MY ENGINE
>I'VE ABANDONED MY ENGINE
I'VE ABANDONED MY ENGINE
>I'VE ABANDONED MY ENGINE
I'VE ABANDONED MY ENGINE
>>
I'm a 3D modeler looking to mingle with game devs, do you have tips for finding them?
lately I've been lazy and working with someone might help me get off my ass and keep modeling
ideally it would be someone who knows what they want, can communicate what they want and doesn't want fucking ugly LP models, someone with a vision you know
>>
how fast are your incremental builds? I was using c++ and it was over 10s switched to c and now its under 100ms. Shits bananas what I put up with...
>>
>>101377015
post your work
>>
>>101377079
I mean C++ is a lot more complicated of a language than C so I don't think it's unreasonable for it to take longer to compile.
>>
>>101377079
Yeah, C with a proper build system is basically instantaneous.
>>
File: its over.jpg (68 KB, 1024x1024)
68 KB
68 KB JPG
>>101376741
rewriting a new engine when?
>>
File: images.jpg (13 KB, 299x168)
13 KB
13 KB JPG
>>101377200
Probably this weekend.
>>
>>101377125
Fuck off with your garbage. C++ and Rust build times are in no way acceptable
>>
>>101377247
how old is that? Can't be recent clang isn't ten times slower than gcc
>>
>>101375964
Literally every A, AA, and AAA graphics dev I've seen express their opinion on graphics API, said they hated Vulkan.
Something along the lines of, they took everything that sucked about OpenGL, amplified it tenfold, but without the one thing that made OpenGL worth using - simplicity.
>>
>>101377280
Do you know anyone who said that who isn't Jon Blow or his gay buddies
>>
>>101377247
I don't understand how C can be so simple yet take so much time to compile, is this with optimizations on?
>>
I learnt basic OpenGL, where do I go from here?
>>
File: joker_surprised.gif (3.77 MB, 636x640)
3.77 MB
3.77 MB GIF
>remove one (1) vertex->fragment vec4 variable
>fps increases by 50%
WHAT THE FUCK
Is bandwidth REALLY THIS EXPENSIVE?
>>
>>101375964
Vulkan (and DX12) was sold on the promise that “this is how modern graphics cards work”. In particular the idea of handing the burden of Pipeline State Objects to the developer. With OpenGL and DX11 PSOs were handled by the driver and you just tell it what you want to do and it changes the state accordingly. Khronos overshot it with Vulkan and the api was too unintuitive and awkward to use. They know this and have since released a ton of extensions (dynamic rendering, shader objects) to alleviate the complexity to the point where it’s come back full circle to the old APIs and people are like “wtf was the point?”. Vulkan-tutorial.com teaches barebones Vulkan but everybody learning should seek to use stuff like dynamic rendering early on. I wonder where the API would be if not for Doom. There’s a rumor that Starfield was developed using Vulkan and months before release Bethesda asked AMD for help who just threw it out in favor of DX12. Not trying to scare anyone off, just laying out the details from what I’ve researched. Vulkan 1.3 supposedly has the API in a pretty good state albeit “extension hell” which I don’t think is a big deal.
>>
>>101377353
I'm guessing you're probably measuring a very simple scene with nothing else going on
>>
>>101377362
Really sounds like people should learn DirectX or OpenGL going by what you're saying
>>
File: AAAA.jpg (87 KB, 900x900)
87 KB
87 KB JPG
>>101377209
LETSS GOOO!!
>>
>>101377369
I'm writing a voxel splatter, I'm drawing quads and then cropping the quad by raycasting the voxel in the fragment shader.
Not passing the color of the quad from the vertex shader increased my voxels/second from 2.2 billion to 3.5 billion, using a scene of 11 million voxels, which is absolutely silly.
>>
>>101377441
Well if you're drawing that many voxels then memory bandwidth is very important
>>
>>101377079
you shouldn't have long build times with a toy project, and you wouldn't be able to switch with a non-toy project, so you're either lying or are comically incompetent
>>
>>101377388
If you want to get something done yes. I don’t think learning Vulkan is bad, that knowledge carries to every other graphics api. But it’s important to know what you’re getting into. If you want to learn Vulkan you should have a firm reason for doing so other than “it’s new”. If you’re like me and want to use stuff like hardware raytracing you’re forced to use Vulkan/DX12. I’m also a masochist.
>>
>>101377272
Read it again
>gcc -> 5kloc/s w/gcc and 4.6kloc/s with clang
>clang -> 380loc/s w/gcc and 400loc/s with clang
They're about the same. Clang probably is written fully in C++ style while GCC is more in a C style as it was written in C for early versions so Clang itself takes forever to compile

>>101377338
GCC and Clang are just not fast and iirc he didn't use any uneccessary flags for the builds. TCC shows it can be way faster
>>
D3D12 has some pretty nice documentation and sane feature levels compared to vulkan.
>>
>>101377524
If nobody is using Vulkan over DX12 what's the point, use DX
>>
>>101377540
>D3D12 has some pretty nice documentation and sane feature levels compared to vulkan.
That's interesting, anon. Now, tell us how many platforms D3D12 targets
>>
>>101377539
TCC doesnt do optimization
>>
>>101377547
Linux. Which is pretty valuable if you’re an indie dev because those guys are starved for games.
>>
>>101377557
No shit
I swear 90% of /g/ is illiterate. Fucking read what I posted. There is no optimization added
>>
>>101377540
>pretty nice documentation
Lmao the DX12 docs are practically nonexistent. All you have is the Microsoft examples and that’s it. Vulkan at least has their spec.
>>
>>101377569
GCC / Clang optimize the output regardless, it's not equivalent
>>
>>101377566
>Linux. Which is pretty valuable if you’re an indie dev
Please don't post anything this ridiculously wrong again
<1% market share
>>
>>101377540
too bad it's stuck on a dead end platform
>>
File: suyu.png (27 KB, 960x159)
27 KB
27 KB PNG
>>101377596
>muh market share
Housewives playing Sims and chinese net cafes won't be a target audience for your game, and hardcore gamers are moving to Linux.
>>
>>101377583
>Source: my ass
https://linux.die.net/man/1/gcc#:~:text=Reduce%20compilation%20time%20and%20make%20debugging%20produce%20the%20expected%20results.%20This%20is%20the%20default.
>>
>>101377651
-O0 still does not produce the same results as TCC
I'm not disagreeing with your point but TCC doesn't really prove anything because it doesn't have the same constraints
>>
>>101377647
The number of Steam users running Linux is 1.6%
Sorry, not less than 1%
>>
>>101377675
Go look at the chart. TCC built with GCC is only a tiny amount faster.
The point of the benchmark is to show actual order-of-magnitude differences.
Small differences like Clang vs GCC are not relevant as they could have been caused by other things, but C vs C++/Rust is a very obvious 10-100x difference
>>
>>101377691
cool
>>
>>101377710
Do your fucking research
>>
>>101377738
why are you replying to me schizo?
>>
File: file.png (61 KB, 1147x175)
61 KB
61 KB PNG
>>101377539
doesn't the column `compiled with` refer to which `compiler` the compiler was compiled with? Or did I mix it up?
>>
>>101377769
I didn't know debunking misconceptions with publically available facts made me schizophrenic
>>
>>101377782
Yeah, so there is <10% difference and then with building Clang it's similar but switched with Clang slightly faster than GCC.
So, I'd say GCC and Clang are roughly the same speed
If you want to read into it, maybe GCC is faster for C and Clang is faster for C++, but it's not a big difference and not reliable
>>
>>101377782
nvm I get it now shit table
>>
>>101377785
well now you know, so the next time you have sudden urge to creep out strangers by sharing your delusions take your meds
>>
>>101377870
>delusions
It's hard data taken from the biggest PC gaming platform by far
>>
>>101377892
What about China?
>>
>>101377910
what about them
>>
>>101377917
How representative is steam there?
>>
>>101377928
No idea, I don't know if there's any reason to assume everyone in China uses Linux
>>
>>101377937
Maybe things are changing now https://www.neowin.net/news/chinese-government-to-dump-windows-in-favor-of-linux/
>>
>>101377946
You're arguing hard data vs a news article about the chinese government thinking about doing something
>>
>>101377962
What hard data? Most popular in steam doesn't mean most popular in China.
>>
>>101377981
Steam stats are hard data
You're arguing facts vs "well what if"
How can you be this stupid?
>>
>>101377998
I don't have steam. Pretty much the majority of people who play games don't have steam. Not having steam is not indication of not having a computer. Steam can only measure data of people who use steam. And I'm asking you what about China. Do they use steam or not?
>>
>>101378045
Most people who play core PC games do it through Steam
>>
>>101378117
Define "core PC game". Anyway, if you dislike money, is up to you.
>>
>>101378146
The games people would talk about on /v/
Talk to any developer who sells on Linux, they make up <1% of the sales and 50% of the bug reports
It's a waste of time to even support Linux
>>
>>101377079
use a proper build system like xmake and you won't even notice it
>>
>>101378173
ok shill
>>
>>101378194
Think rationally instead of emotionally
>>
File: linux gamedev.png (153 KB, 831x698)
153 KB
153 KB PNG
>>101378200
Apparently they are the best bug reporters.
>>
>>101378320
That's also true
Different reports will give less flattering figures
>>
>>101378320
>pic rel
well linux is mostly used by devs so not surprising
>>
>>101378320
Despite being 5.8% of the user base Linux makes up 38% of bug reports.

kek.
>>
>>101378388
Part of it is that if you go to a random open source project and give a shitty not even half assed bug report, they'll just close it and tell you to fuck off, so people know to do better.
>>
>>101376289
thanks, its just a plain triangle subdivision, not quad or anything fancy. one day ill go back and displace it
>>
>>101377362
Once you've got Vulkan 1.3 set up with dynamic rendering, all dynamic states, push descriptors, buffer device addresses and timeline semaphores, it's actually a pretty lean and mean graphics API with complexity comparable to OpenGL 3+ but without the API quirks.
>>
>>101379169
>Once you perform a bunch of ritualistic cope Vulkan is just as good as the thing it was designed to replace
This is some webshitter logic
>>
Hello time wasters. How is the new reinvented wheel going?
>>
File: snek_case.png (82 KB, 1080x395)
82 KB
82 KB PNG
>>101359693
Give it to me straight anons, how much suffering am I in for if I were to start a new project with bevy? I hear a lot of "ECS bad" thrown around, but surely that could be made managable through an abstraction layer? Am I missing anything else?
t. cnile + raylib fan
>>
>>101379462
ECS is bad, Rust is bad, you can do it but why would you
>>
>>101367828
I would also be interested in this. It would be nice to have a library that's just for rendering, no game engine attached, just as SDL and SFML already do for 2D.
>>
>>101379462
>Give it to me straight anons, how much suffering am I in for if I were to start a new project with bevy?
Lots of gaslighting you whenever you even hint at something being bad/dumb
>>
>>101379462
ecs is a pattern, you can use it in any language, c++ has the flecs library for example, you can create a raylib game with ecs if you want
bevy is an engine built on top of an ecs library (bevy_ecs duh), technically bevy should be simpler than raylib since it's a full engine and it has a plugin system for stuff that you'd need to implement from scratch in raylib
>>
>>101379395
Great. I was finally able to play one of my favorite games on modern hardware thanks to my reinvented wheel. Feels good.
>>
>>101379315
OpenGL will be gone in 10 years anon. Its time to move on.
>>
>>101379652
I've heard that one before
>>
>>101379462
You should be using a battle tested engine/framework instead of “rust, the game engine”
>>
Do I really need 2 frames in flight? Why can’t I just use 1 and call it a day?
>>
>>101379848
You can
>>
>>101377388
DirectX12 is just as bad as Vulkan
Khronos didn't do anything wrong, they just gave you more power, same as DX12 did
That's the new trend
Just seems like all the devs got filtered
But since Doom exists and it has proven well written Vulkan is phenomenal, they have no excuse other than admitting they were filtered
>>
>>101379893
What does "filtered" mean in this context? You think Vulkan is too hard for AAA developers to use despite retards in this thread being able to use it?
>>
>>101379906
>You think Vulkan is too hard for AAA developers to use
Yes it is, several developers have admitted it and several companies have tried to make engines and failed or are subpar.
>despite retards in this thread being able to use it?
retards in this thread don't make AAA game engines. you can get a "functional" engine just by following a tutorial but it won't be good.
>>
>>101379980
If you can follow a tutorial and get a functional engine you've overcome the hard part
Nobody in AAA dev doesn't do something because it's "too hard", they aren't amateurs
>>
>>101379652
Considering all the programs that needs OpenGL to run, it is crazy to think it will "be gone".
Sure, there will be a new standard, but GL will remain a thing
>>
>>101380001
>If you can follow a tutorial and get a functional engine you've overcome the hard part
What no? Lmao
You can follow the tutorial without understanding most of it, but the "engine" you made is useless, since you can't optimize it.
You need to understand everything perfectly to be able to improve that code.
>Nobody in AAA dev doesn't do something because it's "too hard", they aren't amateurs
But they are. Never before did they have such low level control. This stuff is completely new to them, therefore they are like amateurs.
DX12 is the same, the only reason it has caught up a bit more, is because of the bigger Microsoft ecosystem and therefore more shit to mindlessly copy paste.

There is literally nothing on Vulkan out there. There is the official tutorial which makes a triangle. There is VkGuide which takes it a bit further and makes an unoptimized, barebones "engine". There's a couple of completely useless tutorials on youtube, and there is one serious FOSS game engine.
That's it.
If devs were smart they would study that engine since it is based on IdTech 4 and there are a couple of Doom devs on it as well.
>>
>>101380101
>But they are. Never before did they have such low level control. This stuff is completely new to them, therefore they are like amateurs.
Literally everything about this statement is untrue

There's too many retards in here lately
>>
>people debating compile speed and languages
forever a nodev ngmi
>>
>>101380129
How is it untrue? Explain.
DX <= 11 didn't have this and neither did OpenGL or any other API.
Low level access was introduced with DX12 and Vulkan, that's why people say Vulkan is equivalent to DX12 and OpenGL to DX<=11
>>
>>101380129
nice cope
>>
>>101380187
Vulkan is a cleaner API that makes less assumptions and is more explicit. Pretty much everything Vulkan does is already exposed in OpenGL 4, and even if it wasn't it's not a big deal to learn anyway

>>101380191
This isn't the thread for speculative gamedev fiction, fuck off
>>
File: 1710613345784274.jpg (164 KB, 1920x890)
164 KB
164 KB JPG
What are some good game concepts for learning languages? I can get funding from my government for voice actors and I want to develop a game with heavy narration that would allow you to hover over text to see it in English. I was thinking about making a card RPG like Voice of the Cards but maybe something simpler where you explore a basic world completing simple everyday tasks around a house or town would be better to learn actually useful vocabulary but then it would be too easy if you were familiar with the language already. What games actually do this well, all games for learning real languages seem to get bad reviews
>>
>>101380231
Complete lies. Where you called 1 function do a thing in OpenGL you now have to write thousands of loc in Vulkan and manage everything explicitly, that OpenGL did automatically and you didn't need to know about.

People fail even in the most basic stuff like memory allocation. That's why there was a memory allocation library written recently and just by using it people are reporting 30% increased frameworks.
>>
>>101380371
*frameworks
kek
>>
>>101380371
>Where you called 1 function do a thing in OpenGL you now have to write thousands of loc in Vulkan and manage everything explicitly
Can you give an example?
>>
Thinking about making a Quake clone with raylib, am I going to kms?
>>
>>101380493
This guy got this far using Odin + Raylib. Seems capable.
https://youtu.be/4DKa01rcJPY?si=UxNw3ar-p11bd9QI
>>
>>101380101
>there is one serious FOSS game engine.
Where
>>
>>101379462
At this point I don't care anymore, go on and tell us how it went.
>>
>>101380231
>glbegin(gl_quads)
okay
>>
>>101380640
That's OpenGL 1 you mongoloid
>>
>>101380414
just Google a triangle example in opengl Vs Vulkan
>>
>>101380685
You think AAA developers are somehow incapable of this?
>>
>>101380231
Wow, you sound like an experienced gamedev who knows what he's talking about. Show me your openGL raytracing implementation.
>>
>>101380704
Why the fuck would I implement raytracing in OpenGL
>>
>>101379462
I just paused my bevy project I've been working on for 8 months to try out some other things. Honestly it's not terrible, and you get used to it, but it's almost like Stockholm syndrome. I forgot how much quicker you could add features in other languages/architectures. Also tons and tons of refactoring. Handling character state and doing different things when e.g. a human attacks vs an orc attacks can get messy quick if you have a lot of conditions. Just read the loglog games article if you haven't. that echoed a lot of my sentiments working with bevy and rust. If you read it and don't care, then go all in, you'll be fine. It's just pretty slow.
>>
>>101380702
>dont use that example
>>
>>101380685
>>101380704
https://github.com/SaschaWillems/Vulkan/blob/master/examples/triangle/triangle.cpp
https://github.com/okaram/opengl/blob/master/triangle.cpp
>>
>>101380740
I'm running with your example (even though it's completely non-specific), do you think AAA developers would be incapable of doing this, even though people in this thread can do it?
>>
>>101380750
>give me an example of vulkan takes more functions than opengl
>hello world
>nooo not that one AAA devs are better than that
>>
>>101380767
I just told you I'm running with your example. Do you think AAA developers would be unable to do hello triangle? That it's too hard for them?
>>
>>101380777
what takes more function calls? vulkan or opengl
>>
>>101380788
Vulkan, how is that relevant?
>>
>>101380802
>>101380414
example given
>>
>>101380830
And the example aligns perfectly with what I was saying - No new concepts are introduced, it's the same thing but more verbose
If you want to claim that Vulkan is somehow a paradigm shift from what came before it, you need to back that statement up
>>
>>101380777
it's not about being "unable" to you fucking retard it's "why would they bother"
gamedev is already notorious for insane crunch and deadlines and if it takes 1200 lines to draw a triangle for negligible performance gains on an API that runs on less platforms than OpenGL why would they bother? The triangle example is just the minimum, the complexity doesn't just stop once you draw one thing to the screen.
>>
>>101380876
>it's not about being "unable" to you fucking retard
Someone said Vulkan was literally too hard for AAA game devs
>>
>>101380888
so you're just arguing semantics
>>
where my (((lisp niggers))) at
>>
>>101380912
No
The reasons you provided make sense
Saying it's too hard for AAA devs does not make sense and is absolutely retarded
>>
C++ is the same as Python, just more verbose. You can see it if you compare Hello world written in both languages. I'm a FAANG coder btw.
>>
>>101381064
Retard
>>
>>101381064
can you get me a faang job? i wear programming socks
>>
>>101381064
kek
>>
I'm surprised none of you is using C#. As boring as it is, it's probably the best lang to make a game due to a combination of features, tools and libraries. Terraria and Stardew Valley were made in C#.
>>
>>101381132
It's far too practical for this thread
>>
Considering that octree is, essentially, three binary trees - one for each dimension - why not just use three binaries and common part the result?
>>
File: itsover.png (433 KB, 674x656)
433 KB
433 KB PNG
>Graphics bug
fml
>>
File: 1590524137-0.png (80 KB, 326x277)
80 KB
80 KB PNG
>>101381221
>>
Alright lads,
Got my OpenGL window open. Set up the rendering context, started basic transformations.
Screen shot shows a viewport that is centred on the client window. The world window has a fixed aspect ratio.
Windows programming and OpenGL coding is, ngl, comfy as fuck.
Gonna start work on the 2D/3D viewing systems next.
>>
>>101381158
what are the most popular frameworks used here?
>>
>>101381420
only shit ones
>>
File: IvyMen_71I9wfo5Kg.webm (2.14 MB, 1008x728)
2.14 MB
2.14 MB WEBM
C++ and DirectX, been working on this for a couple months
>>
>>101381420
Unfinished Vulkan engines in C
>>
>>101381565
based dx bro
>>
>>101381565
looks cool, much more readable than most of these boomer shooters
>>
>>101381631
what do you mean by "readable"?
>>
>>101381565
Sweet.
Are you doing this from the ground up or are you using some library on top of Direct3D?
>>
I just realized I suck at algorithms and data structures so I'm gonna make a library to visualize that.
Not exactly gamedev but I will have to do some graphics programming. I think that'll be a good learning project. Also I lack creativity for making games.
>>
>>101381635
graphical clarity
>>
>>101381641
>Also I lack creativity for making games.
Welcome abroad.
>>
File: machineGuni.gif (76 KB, 512x512)
76 KB
76 KB GIF
>>101381638
No extra rendering libraries. These are the libraries I'm using for other stuff:

>directxtk
Mostly for input handling and texture loading
>imgui
For debug UI, but I don't really use it much anymore
>nlohmann-json
For reading some configuration stuff in my map file format
>yaml-cpp
For reading other configuration stuff related to assets in the engine
>ms-gltf
For loading GLTF files. Going to have more of these later
>physx
Physics/collision
>tinyobjloader
For loading .obj files, don't have any yet though
>>
>>101381132
>gc language
>best to make a game
yeah no
>>
>>101381673
Most of that I'll be writing myself.
>For reading other configuration stuff related to assets in the engine
It's going to be a while before I start looking at asset management and the like, but how are you storing and searching for assets in your game?
>>
>>101381673
Are you following some DirectX book/tutorial?
>>
>>101381717
You didn't learn a thing from last thread did you?
>>
File: shock.gif (5 KB, 256x256)
5 KB
5 KB GIF
>>101381132
After I release my current project, my next thing is going to be a graphics framework in DirectX which manages infrastructure and view allocations (i.e: all the DirectX 12 stuff), but which will allow you to write your own shaders still and define the parameters of your textures, and then put them in a command list and have them execute. So you're still writing shaders and binding resources, but the execution order is handled for you and you can more easily chain together shaders into sets of techniques. Sort of like the old .fx files, but I guess using some imperative language instead.

Anyways, since Windows forms can pass an HWND without marshalling from .Net 8 to native C++, I figured I could provide some C# bindings to the framework so that window management and game logic can be handled in C#/WinForms and rendering is handled in C++. This would also allow quick prototyping of UI using winforms. C# wouldn't be required, obviously; it's a C++-first library but the Windows development platform would make this kind of interop trivial (though tedious) to implement
>>
>>101381772
Not currently. I started out with Frank Luna's DirectX 11 book and also used Rastertek sometimes as well, but at this point I just kind of do these things myself and use the book as a reference if needed. This kind of graphics isn't hard at all IMO. The real hard shit is getting into PBR and trying to optimize visibility determination in large scenes.

>rastertek
https://rastertek.com/tutdx11win10.html
>>
>>101381779
what happened last thread?
>>
>>101381851
control + f "reference count"
>>
>>101381851
ARC/GC debate that derailed the thread
>>
File: output.webm (1.89 MB, 1280x720)
1.89 MB
1.89 MB WEBM
Added undo to the terrain tool.
>>
File: mahiru scream [tk3qh5].jpg (101 KB, 521x593)
101 KB
101 KB JPG
>>101382279
AAAAAAAAMAAAAAAAAAAZINGGGGGGGG!!!!!!!
>>
>>101381717
you'd be surprised. depends on the lang I guess. In my case I can turn off the GC and run it manually, so I just do so in in-gameplay moments like when the player dies/goes to sleep/etc. so far seems to be working. the GC wasn't even an issue I'm just curious if something like this works.
>>
>>101381673
>json, yaml
>gltf, obj
why not just one from each?
>>
>>101382623
Also some languages offer sub-ms GC times, which is totally acceptable for games. Unless you're making nothing noteworthy the GPU would probably not even let your game hit 1000 FPS (1ms).
>>
>>101381673
holy bloat
>>
>>101380252
that one camera picture game was interesting, I'll look up the name for you when I'm not lazy.

But the most interesting idea in language learning I've seen recently is that employed by jpdb and renshuu, which are basically just better more complete anki. Also this https://www.youtube.com/watch?v=7t58Y8YrfCk

renshuu also has some good games, doing a crossword in japanese was pretty cool.
>>
>>101382467
thank you saar!
>>
jeets in my gedg???
>>
>>101381990
>debate that derailed the thread
those are the best threads though
>>
does anyone know where I can get royalty free/creative commons/copyright free high quality photos and/or photo textures?
I know creative commons has a bunch of wikimedia photos, but where do I find collections of hundreds of photos of random walls and grounds and photos of buildings, etc?
I guess the obvious option is "take the photos yourself you lazy bastard" but I dont want to spend an eternity taking photos and I also might not be able to find everything I need irl
>>
Hello, is this where I get my GED?
>>
>>101384036
photos that I can UV map, not HDRIs necessarily but I'll take those too I suppose
>>
>>101384036
AMD has a huge free texture library in the materialx format
https://matlib.gpuopen.com/main/materials/all
otherwise i think there's one of those "awesome list" things on github with free gamedev resources
>>
File: MTLX.png (39 KB, 500x898)
39 KB
39 KB PNG
>>101384757
the licensing info is right here btw, i think most if not all of them are MIT like this
>>
do you have a blog where you post progress? i'm thinking of creating one
>>
>>101377566
Just target wine/proton, it's good enough now
>>
File: 1693931108398524.png (2 KB, 364x33)
2 KB
2 KB PNG
>>101377691
It's 2.08%
>>
>>101385585
not as good as native
>>
>>101385653
They'll fix it for free in time, why would you spend your own time and money when there's wine jannies already doing it for free?
>>
>>101360734
good thing MVP matrices arent per vertex or per pixel. hey they're even statically sized for you!
>>
>>101377547
imagine wanting to run on something other than windows/xbox, but not wanting to learn at least 2 more APIs?

Switch? Mac? Linux? iOS? Android?
>>
>>101382279
Do you use OBS for screen capture?
If so, what are your settings?
>>
>>101387113
#!/bin/sh


./urban-tycoons &
game_pid=$!

get_property_value() {
local property_name="$1"
local output="$2"
echo "$output" | grep "$property_name" | awk '{print $NF}'
}

i=10
while [ $i -gt 0 ]; do
echo "$i"
i=$(($i - 1))
sleep 1
done

wid="$(xdotool search --pid "$game_pid")"
info="$(xwininfo -id "$wid")"
x="$(get_property_value "Absolute upper-left X" "$info")"
y="$(get_property_value "Absolute upper-left Y" "$info")"
width="$(get_property_value "Width" "$info")"
height="$(get_property_value "Height" "$info")"

ffmpeg -f x11grab -video_size "$width"x"$height" -framerate 60 -i :0.0+"$x","$y" -an -y output.mp4 &
ffmpeg_pid=$!

while true; do
if ! ps -p $game_pid > /dev/null; then
kill $ffmpeg_pid
break
fi

sleep 1
done

while true; do
if ! ps -p $ffmpeg_pid > /dev/null; then
len="$(ffprobe -v error -show_entries format=duration -of default=noprint_wrappers=1:nokey=1 output.mp4)"
trim_s=1
len="$(echo "$len $trim_s" | awk '{print $1-$2}')"
ffmpeg -i output.mp4 -ss 0 -t "$len" -c:v libvpx-vp9 -an -b:v 1M -fs 3M -pass 1 -f null /dev/null && \
ffmpeg -i output.mp4 -ss 0 -t "$len" -c:v libvpx-vp9 -an -b:v 1M -fs 3M -pass 2 -y output.webm
break
fi

sleep 1
done
>>
Hello is froggy sneed anywhere around here? Tell him he forgot his toothbrush over on the other place
>>
>>101387547
Sorry but after watching his streams, I don't believe that frosch formely froggy ever leave his house.
>>
>>101387576
he streams?
>>
is multiplayer a no go for beginners? do I just need to go ahead and accept that I have to focus on single player games for now?
what exactly does it entail? how complicated and difficult is it?
>>
>>101388420
turn based multiplayer isn't that hard, I wrote shitty multiplayer code in college and it was easier than I expected it to be. TCP makes it pretty straightforward, you just send out basic packets with game state changes, and as long as you don't have any stupid bugs, the clients on both ends should stay in sync.
real time is harder, not so bad if you are making a cooperative game. real time competitive is incredibly difficult, because now you have to think about latency compensation and things not feeling bullshit (like getting shot behind cover because you weren't in cover yet on the other guy's screen)
>>
>>101388420
Multiplayer difficulties are usually centered around controller configuration when playing locally or syncing data over a network connection. I think a beginner could pull off local multiplayer, but I really wouldn't try to make a network game as a beginner.
>>
>>101388420
Multiplayer means for every single thing that happens in your game you have to think about how it's going to be synchronized to other players
This is only easy for a turn-based game, any real-time game it's complicated because you have to compensate for lag which means you're dealing with multiple world states
>>
>>101388420
>turn based multiplayer isn't that hard
>>101388478
>This is only easy for a turn-based game
hm, well I'm specifically talking about a strategy trading card game. so I guess it can't be anymore turned based than that. This is good news. Thank you
>>
>>101377555
the only one that matters for games?
oh, and xbox I guess
>>
>>101381132
pros
>you get modding for free
cons
>c# disassembly is so easy you might as well publish the source code
>>
>>101390976
>muh intellectual "property"
shalom, you should have a game before concerning yourself with whether someone can see how bad of a programmer you are
>>
>>101391084
save your freetard seething
>>
>>101391175
i only see a rabbi seething at a language for not sufficiently protecting his intellectual "property"
>>
>>101391272
>I'm entitled to the source code of others
>>
File: 816ww3WTGgL.png (310 KB, 1920x1080)
310 KB
310 KB PNG
>>101390976
>c# disassembly is so easy you might as well publish the source code
I don't think this is a big deal. What's the reason people wouldn't like their code to be published? Piracy, but piracy will happen anyway. And as far as I'm concerned, piracy never stopped popular games from becoming popular.
One may also fear that, by publishing the source code, a third party will learn from it and get an "edge" to that sort of game, eventually surpassing the original. But that too won't look likely if experience is of any indication. It may help for others to make mods, which would boost your popularity.
>>
Sorry, but here's a not-so-technical question. Some greentext context:
>making a top-down 2D online RPG
>about to release on steam
>game looks like a roguelike with graphics
>very UI heavy
>some particle effects and stuff like that, but in general the visual presentation isn't extravagant
My steam page is going to need some kind of video. How the hell do I make a good video presenting a game like this? I've made promotional videos for 3D action games before, which is easy. But this seems impossible. The steam guidelines even say that you should avoid too much UI - which is funny since my game is 50% UI. Lots of games seem to have videos that are basically cutscenes without any gameplay, which is kinda lame. Also no way I have time or money for that.

Anyone with suggestions? Any examples of similar games with cool steam page videos I should check out?
>>
>>101380958
(where-at here)
>>
>>101366487
how many lines of code did you write?
>>
>>101391428
>What's the reason people wouldn't like their code to be published?
There are several valid reasons unrelated to piracy, mainly avoiding datamining the game mechanics and secrets, datamining assets is unavoidable, but mechanics can be protected somewhat.
Let's look at RNG, humans are so trash at statistics almost every game cheats in favor of the player, either by showing the wrong statistic (Darkest dungeon having a hidden extra 5% to hit, to +10% on some other game I forgot), or by weighting the dice the longer the player fails to the point of guaranteeing the positive outcome. The thing is, while people love this, they cannot be aware of it or else they'll straight up trashtalk the devs (or at least the elitists of players) or the whole "illusion" crumbles.
Think how the original Resident Evil 4 had dynamic difficulty but never actually told anyone about it, while other games can't shut up about their "genius" game design and exposes every little cheat they pulled. Now think about all kind of secret mechanics you want to have without actually exposing it to the player.
>a third party will learn from it and get an "edge" to that sort of game, eventually surpassing the original
Also a valid concern, especially for those shitty cheat outsourced companies, the problem is that they never try to surpass the original, they only bloat the market by millions of mobile tier asset flips. At least make them actually work for it.
>>
>>101391576
People who are only programmers and not creators don't understand why you'd want to conceal the source code from someone else because to them, the code is the product
>>
New thread:
>>101391816
>>101391816
>>101391816
>>
>>101390976
I am making a .Net based engine. My first releases will be AOT-compiled, but later I will just ship the jit-code because I don't care much.



[Advertise on 4chan]

Delete Post: [File Only] Style:
[Disable Mobile View / Use Desktop Site]

[Enable Mobile View / Use Mobile Site]

All trademarks and copyrights on this page are owned by their respective parties. Images uploaded are the responsibility of the Poster. Comments are owned by the Poster.