[a / b / c / d / e / f / g / gif / h / hr / k / m / o / p / s / t / u / v / vg / vm / vmg / vr / vrpg / vst / w / wg] [i / ic] [r9k / s4s / vip] [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

Name
Options
Comment
Verification
4chan Pass users can bypass this verification. [Learn More] [Login]
File
  • Please read the Rules and FAQ before posting.
  • You may highlight syntax and preserve whitespace by using [code] tags.

08/21/20New boards added: /vrpg/, /vmg/, /vst/ and /vm/
05/04/17New trial board added: /bant/ - International/Random
10/04/16New board for 4chan Pass users: /vip/ - Very Important Posts
[Hide] [Show All]


Janitor acceptance emails will be sent out over the coming weeks. Make sure to check your spam folder!


[Advertise on 4chan]


File: IMG_5005.jpg (48 KB, 560x350)
48 KB JPG
What Would John Carmack Do? Edition

/gedg/ Wiki: https://igwiki.lyci.de/wiki//gedg/_-_Game_and_Engine_Dev_General
IRC: irc.rizon.net #/g/gedg
Progress Day: https://rentry.org/gedg-jams
/gedg/ Compendium: https://rentry.org/gedg
/agdg/: >>>/vg/agdg
Graphics Debugger: https://renderdoc.org/

Requesting Help
-Problem Description: Clearly explain your issue, 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: >>109306361
>>
First for malicious malloc
>>
>>109357191
What would John Carmack do if he were here right now
He'd make a plan and follow through
That's what John Carmack'd do
So what would John Carmack do
If he were here today?
I'm sure he'd kick an ass or two
That's what John Carmack'd do
>>
File: 1722245174776232.gif (3.63 MB, 211x374)
3.63 MB GIF
>>109357412
what?
>>
Been reading about CC Engine from Airfix Dogfighter game, I wish they freaking leaked the source code
>>
File: IMG_5007.jpg (70 KB, 1179x1000)
70 KB JPG
>>109357191
Can you make a game engine using Glide API? Is there an SDK floating somewhere?
>>
>>109357552
John Carmack is an olympic figure skater who won gold.
>>
Where's all the DirectX documentation/guides/etc? Why is it still the hardest API to learn
>>
>>109357760
Even if you could, why would you? You'll just make life a lot harder for yourself, better do it in DX/GL/Vulkan and try to ape the look there.
>>
>>109357760
Unlike DirectX which Microsoft unofficially continues to support the deprecated versions there is no drivers for glide you would have to wrap in dgVoodoo which translates everything in to DX anyways, so what's the point, really?
>>
>>109358219
nta but
>Microsoft unofficially continues to support the deprecated versions
Wait seriously? You mean I can just make a game in DirectX7 and get away with it?
>>
>>109358087
You learn Vulkan first and then sidestep to DX12. They are nearly identical.
>>
>>109358346
And where's the guide to sidestep to DX12? Besides, I want to learn about DX11 too
>>
File: dx7snp.png (74 KB, 461x786)
74 KB PNG
>>109358305
yes basically you can force directx7 compatibility and you link with directx9 and it just works, who knows how it actually works but it's legit directx7 code taken right out of the directx7 sdk example projects
>>
>>109358441
On 64-bit?
>>
Hey friends, I made myself a native C (and C++) dependency build and website. It builds several libraries that I used or want to try to use for static and dynamic linking across windows/linux (X64/Arm64), macos (arm64), ios, android, wasm (emscripten). It's pretty cool. Either use the website/binaries directly, or clone the repo, and get your own builds.

https://deps.morew4rd.com/
>>
>>109358452
Uh no, directx7 is 32 bit only
>>
>>109358524
Curses.
>>
Worked a little bit on my game, flamethrowers are a pain, you need a lot of separate stuff to make them work
>>
Man.
I wish there was a proper computer equivalent to the Nintendo DS local wireless multiplayer thingy.
LAN play just isn't the same.
>>
>>109359608
local co-op died 25 years ago
>>
>>109359632
>t. actual unironic child
>>
>>109359957
im 40 years old, i remember publishing a local co-op game in the 2000s from a guy who hadn't got the memo and nobody bought it
>>
My maps are straight dookie
>>
>>109360399
that's exactly what John Carmack would do
>>
>>109360433
He’d hire Romero to do the maps for him
>>
>>109357191
>>109357412
>>109360433
Now i have the song stuck in my head
>>
POST WEBMs OF YOUR GAMES/ENGINES IN PROGRESS

I am only interested in and want to see vids. Only post vids. I am not interested in pointless /gedg/ conversations.
>>
>>109360473
????
>>
>>109360473
I'll go years and then something will remind me and one of those old songs will get stuck in my head
>And also this one: https://www.youtube.com/watch?v=pdRAyG-BuFE
>>
>>109360784
Fuck Gregory
>>
>Bang! Engine
this is what Age of Mythology was built on, wierd name huh?
>>
>>109360549
My project is currently 4 refactors deep.

Once they are done, I will.
>>
File: 2026-07-24 22-22-59.mp4 (3.67 MB, 1280x720)
3.67 MB
3.67 MB MP4
>>109360549
I can't showcase scene because it's so noisy video compression completely removes most colors, but have some perf graphs.
>>
File: file.png (169 KB, 1920x1080)
169 KB PNG
I feel like the jump between knowing how to program and how to start drawing stuff on screen and slowly writing a game is a huge leap.

I've been asking Claude to guide me to using raylib's DrawRectangle to slowly build a terraria-like world, but it has been handing me out way too much code. Then I told it to not ive me so much code and now it is just explaining concepts but I am completely unsure about what to do.
Is there something I should read before trying to do this? Except I don't want to re-enter tutorial hell again, of course.
>>
I am going to build an entity component system for my game engine.
Ie instead of an array of structs I will have I will have arrays of contiguous data referenced by an ID number.
>>
>>109362392
no one understands all the code, all code bases become unreadable, tech debt always happens, rewriting is usually a bad idea, just create shit and make it work no matter what
>>
Anyone here have experience making a srpg engine? I was retarded and installed Linux so I have to make things manually so if anyone has any tips on it I'd appreciate it, was going to make something similar to how fire emblem awakening/fates does things but with more 3d cutscenes, but I have no clue where to start from scratch, I was just gonna use unity until I fucked things up for myself.
>>
>>109357191
He would do some programming until he got tired of arguing instead of programming and then disappear to do some more programming.
>>
>>109362392
John Carmack would tell Claude to shut the fuck up, then go back to programming.
>>
>>109362810
I guess that's right.
Thank you for that.
Except...
>>109362593
The problem is I don't really have any idea about what to do. I know how to program but trying to build a world and a game makes me feel like I actually don't know any programming at all. I'm just staring at my keyboard.
But oh well, I'll keep trying.
>>
>>109362810
>John Carmack
...has been posting on X for months about how amazing AI is and how it's a generational leap forward.

And he's right, as long as you use it properly it's an amazing tool.
>>
File: 1784935580498307.png (336 KB, 2560x1414)
336 KB PNG
>>109362456
Behold.
>>
>>109362865
As a solo dev you won't know how to build a game engine, game engines are developed by teams of experienced engineers collaborating on a design. Thus the only practical way for a solo dev to build a game engine is to first study an existing designs then to use that knowledge to build their own. Design is an evolution you can't just start from scratch without understanding the history.
>>
>>109358509
This is basically already done with vcpkg export. It just packages vcpkg so that you can just copy the zip to another PC.
But if you want a runner to download libraries that are exactly matching the current environment, vcpkg has binary caching (but vcpkg export is probably faster, and I think you must use manifest mode for binary caching, and you probably need to lock the version numbers of the libraries to prevent constant updates).
But the problem with both options above is that you are missing the buildtree and source files, which matters if you want to view the source code of a library while debugging (you still get the file+line without the source files if you have debug info), or on windows+msvc if you want your .pdb files to work without needing to manually copy the .pdb file beside your exe (easy fix, don't use pdb files, just use static libraries or custom /Z7 build for dll's).
>>
>>109362987
I think John Carmack would tell John Carmack to shut the fuck up, then go back to programming.
>>
>>109363768
did you make a carmack tulpa
>>
>>109357412
include me in the screenshot
>>
>>109362392
>I feel like the jump between knowing how to program and how to start drawing stuff on screen and slowly writing a game is a huge leap.
It really isn't.
>get input
>update player position
>update enemy positions
>draw
>repeat
>>
>>109362392
Do you actually have the world represented in code in a way you understand? Because actually drawing it with some rectangles should be the easy part.
>>
>>109363962
Right until this point, yes. I do understand what is going on. I mean I've been playing around raylib for some weeks now, I wrote a breakout clone. But for whatever reason the jump between that and trying to build a terraria-like clone (or at least the world first for now) feels huge. It took me a lot of time to understand what was going on with some of the loops. And still now, I have no idea how the camera comes into place, since my breakout clone didn't even use a camera at all.

But oh well, step by step, I suppose.

>>109363934
Thanks
>>
File: izin on eck.jpg (38 KB, 1000x435)
38 KB JPG
>>
>>109363378
cmake is just as easy. also if you want sources, you can clone/submodule the repo. the idea is not having to rebuild as often
>>
Carmack took the Jab..
>>
File: 1784948964011957.png (220 KB, 2544x1400)
220 KB PNG
The entity system progresses.
>>
>>109364393
Oh dear....If only John Carmack had told John Carmack not to take the jab then maybe John Carmack would still be John Carmack.
>>
>>109364026
>And still now, I have no idea how the camera comes into place
1. get player position, this will be the center point of the visible area
2. subtract half the screen size to get the position of the top-left corner of the visible area
3. when drawing each object, take its position minus the top-left camera position, and draw it there
4. (optional) add some logic to skip drawing entities that are outside the visible area
>>
>>109364043
vcpkg is made for cmake as a package manager for making patches to non-portable libraries to make them portable / conforming (such as libraries that don't use cmake, like gn, or automake, or .bat/.sh scripts), it will work with projects that don't use cmake (you just point to the /lib, /include and copy the /bin DLL's), but you need to use cmake/msbuild if you want vcpkg to copy the DLL's into your build directory so that your application can run (on linux, the default triplet is static).
So you basically just have 2 ways of using vcpkg, you got classic mode which is just a folder that holds all your built libraries, and manifest mode which specifies libraries in a json file per cmake project. I like classic mode because it uses less disk space/won't check for library updates (a flag fixes this), but manifest mode should be more "retard proof", and the classic mode needs a relative or absolute path to your vcpkg folder... not everyone has the same vcpkg folder location.
>>
Why is there always a guy dumping walls of text about build systems?
>>
File: 1784951893443269.png (208 KB, 2560x1440)
208 KB PNG
YESSSSSSS!!!! MORE MORE MORE
>>
>>109364522
ok cool. I just use cmake install, a lot of the projects support this anyway. the ones that don't I made shims. it all just works. it'll save me a lot of time I hope
>>
>>109357191
I made a video game but I am not so sure about the price. I'd buy it for $1.00.
sort of casual game targeting all smartphone platforms but it is also playable on PC in a 512x1024 window.
it is actually fun to play and kind of addictive to...
How many people do you think will actually buy it for $1.00?
>>
>>109364621
cmake install includes too much trash in my opinion, that's how I started building my projects, but the install target does not understand which dll's or exe's your project uses.
The purpose of install is to just copy your binaries into the global linux pitchfork paths (which is a bad idea anyways since there is no "uninstall" target, it creates an absolute mess).
The way how I like to test my projects is by running my programs inside of the build folder, and the install directory should be the staging directory where I can just zip the folder and upload.
This doesn't matter if you only use static libraries, but from my experience, trying to build a static library in cmake is where a lot of things start to go wrong.... vcpkg fixes those issues, the most important fix is that you do not need to link to every library dependency in a static build (AKA, if you use freetype, it may link to zlib (libz?) or whatever, on a dynamic build, you don't need to link directly to zlib in cmake just because freetype uses it, but on a static build, you must link to the static version of zlib before freetype). I don't remember if this is specific to specific platforms, it could be, but it is nice to be able to switch from dynamic linking to static without needing to worry about this.
>>
>>109364718
Cool. But yeah, I can do all that with manual install steps in cmake, without introducing another new thing. Zip files are just fine. If there are missing stuff in them, I'll update the build scripts a bit until it's correct. The important thing about this is whole thing is in a single place. And reusable.
>>
Should be good enough camera system for a simple 2d game?
void CameraCreate(Camera *camera, vec2 position, vec2 size, float hud_offset)
{
camera->box.x = position.x;
camera->box.y = position.y;
camera->box.w = size.x;
camera->box.h = size.y;

// Y offset from hud
camera->hud_offset = hud_offset;
}

// Always keep the camera centered on the player
void CameraUpdate(Camera *camera, vec2 position, vec2 bounds)
{
camera->box.x = CLAMP(position.x - (camera->box.w * 0.5f), 0, bounds.x - camera->box.w);
camera->box.y = CLAMP(position.y - (camera->box.h * 0.5f), 0, camera->hud_offset + bounds.y - camera->box.h);
}

vec2 CameraScreenPosFromWorld(Camera *camera, vec2 world_pos)
{
return (vec2){world_pos.x - camera->box.x, world_pos.y - camera->box.y};
}

vec2 CameraGetPos(Camera *camera)
{
return (vec2){camera->box.x,camera->box.y};
}
>>
>>109365006
You are using the dynamic VC++ runtime for your static libraries (C libraries still use VC++ for memcpy and some string functions). You should be using /MT, since that's what vcpkg defaults (there is a x64-windows-static-md for people who want it).
Technically if you still want DLL's, and you don't want the VC++ you can build with something called the "hybrid UCRT" to avoid the VC++ but still keep the binary size down + avoid potential /MT errors like signal() not working across DLL's.
The UCRT is preinstalled on windows 10, and it's officially supported on windows 7. And it's forward and backwards compatible.
Then you can add a vcpkg triplet with those linker flags, and you could add other flags like sanitizers, or cfguard (fun fact: llvm-mingw's libc++ is built with cfguard, and soon intel nova and next AMD will release with ChkTag, which is a very fast way to check if memory access is invalid for release binaries! AND it will work on older x86 cpu's! unfortunately none of the compilers have a flag for it even though it's theoretically possible to use right now, because it's not finalized... It wouldn't work on an old CPU, but maybe we can simulate it in a VM?)
>>
>>109365337
are you a bot?
>>
>>109357352
I just saw the last thread. Does /gedg/ still have constant arguments about malloc?
>>
>>109365418
no people just love to invoke and hopefully summon the malloc schizo but he stopped posting here ages ago
>>
>>109365418
yes
>>
File: video.webm (2.72 MB, 1280x720)
2.72 MB
2.72 MB WEBM
GPU frustum culling.
>>
>>109366059
>fuckton of objects on screen
>470 fps
>no objects on screen
>480 fps
good job, I guess
>>
>>109366105
There is no lighting/shading/etc yet. Just a bunch of meshes (all drawn using a single MultiDrawElementsIndirectCount function call) and textures. So I would say it makes sense that FPS doesn't change much.
>>
>>109357191
Daikatana
>>
>>109364614
>can afford 742 clouds
I think billionaires should be taxed more
>>
>>109364039
>no ick
it's not the same :(
>>
>>109366105
Maybe he's got a 480hz monitor with vsync on.
>>
>>109366105
480fps with an empty scene and 470fps with a crowded scene is preferable to 1500fps with an empty scene and 60fps with a crowded scene.
>>
>>109357760
Yes but you will either need a 3dfx card or a "glide wrapper".
The sdk used to be available through web archive here,
https://web.archive.org/web/20000302023735fw_/http://www.3dfx.com/developers/glidesdk/glsdk-intro.html
but like all other good things it looks like the download links have been nuked.
Just google g2sdk.exe or g3sdk.exe and you should be able to find some download links...
>>
>>109364450
Thanks anon. I'll try to work on that today.
>>
We're all gonna make it.
>>
>>109368523
Nice, what API? Also have you used PCem or something or you just happen to have windows95/98 machine?
>>
>>109357760
>Laughs in ProSavage
>>
>>109368523
not with such shitty screen caps
>>
>>109368996
This. Can’t see shit capt’n
>>
>>109368523
Anon, are you devving on a PC from the 90s?
>>
File: 1785035775598947.png (416 KB, 2560x1440)
416 KB PNG
I have overhauled my entity system.
Biggest change is no more function pointers calls in loops which was obliterating the CPU cache and I am keeping track highest entity that exists..
Delta starts to fumble at >3000 entities however if you look at my laptops specifications in the top left this is to be expected?
>>
>>109372470
>function pointers calls in loops which was obliterating the CPU cache
this doesn't matter if you only have 3000 entities
>>
File: 1785036705920723.png (283 KB, 2560x1440)
283 KB PNG
>>109372485
The issue right now is rendering on this poxy laptop. If I disable rendering delta does not budge even at 80,000 entities.
>>
>>109372519
the actual work you will be doing when you iterate over entities like rendering will be an order of magnitude slower than the cost of a function pointer call
>>
>>109372522
The issue I thought was when you call a function like that the CPU throws out it's cache and when you return from the function it has to hunt down the memory again.
This is literal rape.
>>
>>109372537
>The issue I thought was when you call a function like that the CPU throws out it's cache and when you return from the function it has to hunt down the memory again.
It does
But it's an order of magntitute slower than any actual work you will be doing like SDL_RenderTexture
So it only matters for tiny function calls being called a lot
>>
>>109372549
I have over scoped.
>>
>>109362392
If you have no idea how to program, don't tell the AI specifics. Because you told it to 'use raylib's DrawRectangle' it has to try and make things work with that constraint. Better to let it take the wheel from the start.
>>
>>109368618
>>109370235
It's just a virtualbox vm running NT4. I want to keep everything software rendered so the api i use is just Win32/GDI for outputting to a window, everything else (sprite/line drawing, rotation, etc) will be implemented by hand. Ultimately i want the code optimized enough that it is able to run on a ~30 year old shitbox.
>>
>>109373417
>rotation
you wont be able to do this on a 30 year old pc
>>
Thoughts on VK_EXT_shader_object?
>>
>>109373502
Quake was released in 1996 so yes, you you can absolutely do rotation on a 30 year old pc. I'm not really aiming to make a 3d game however.
>>
>>109373532
you can if you keep the resolution and drawcount low like 320x240
not at 640x480
>>
>>109373590
a rotation is just a sin/cos function call and a rotation matrix. it doesnt matter what resolution your screen is. the only thing that's gonna increase the cost of a rotation is if you increase the number of vertices you operate/rotate on. A larger resolution will only affect the cost of rasterizing the final frame which will depend on things like polycount, texture size/count, shading used etc.
>>
>>109373691
>just a sin/cos function call
do you know how expensive that in on 90s hardware
you also need to do texture blending on sprites if you dont want it to look like ass which you cant really do without hardware graphics
>>
File: 1785052479807552.png (5 KB, 1280x720)
5 KB PNG
So I've discovered that layers are something my entity system will have to deal with.

It's all so tiresome.
>>
>>109373709
surely an ecs is well-suited to storing an extra int per entity?
>>
>>109373700
>do you know how expensive that in on 90s hardware
You could optimize it if necessary by using lookup tables.
>you also need to do texture blending on sprites if you dont want it to look like ass
Kinda depends on what type of style you're going for though? A 8-bit like game such as Shovel Knight doesnt really use any blending I think? But yes, having hundreds of sprites with blending is probably not gonna be doable unless you use it very sparingly or stick with 8-bit sprites and a color-palette that has been carefully designed around blending for your particular game/style.
>>
>>109373417
Software rendering is awesome
>>
>>109373754
I have to change my whole rendering system.
>>
>>109373709
>>109373754
>>109373938
How can your ECS not just take an extra component with a few bytes in it? How did you design it?
>>
File: oracle33.webm (2.97 MB, 480x432)
2.97 MB
2.97 MB WEBM
My Oracle game clone is really starting to make progress. Turns out the system for where you have a 4 adjacent rooms next the current room is called the Neumann neighborhood, of course Von was the one to invent it... What did he not invent?
>>109373709
ECS doesn't really make a lot of sense for smaller games. Just use a Fat struct for the Entity, which is what I'm doing.
>>
>>109363213
Now hook up Lua scripting for the entities and you will have Löve 2D in 15 years.
>>
>making engines in anal domino's papa john's 2027-1
>>
>>109373844
Shovel Knight mimicked 16-bit, not 8-bit.
>>
>>109375320
this but making games
>>
>>109375327
nawww, making games is easy

opus 5, make me some game dev money. Make no mistakes
>>
>>109375337
>game dev money
what year do you think this is?
>>
>>109372537
>The issue I thought was when you call a function like that the CPU throws out it's cache
>>109372549
>It does
What the fuck are you guys talking about, it does nothing of the sort. At worst, if it fails to predict the destination, you might get a pipeline stall for a few cycles. If it does predict then there's literally no penalty. In any case it's not going to flush the cache, that makes literally no sense

Aside from CPU behavior, the other thing to worry about with virtual calls is the fact that they prevent inlining. So if you're going to do virtual calls, it's better to do one big virtual call instead of many small ones.
>>
>>109375274
Looking good, anon. What's going on inside the dungeon that makes it look like your hitbox gets detached from your character? I guess it's actually in the right place since you can push the block normally, but gets rendered wrong?
>>
>>109375373
Thanks! That's actually a visual bug since Im drawing the players collision rectangle using the world space position instead of the screen position. I haven't bothered with it since I need to redo how I handle sprites so I can have y sorted sprites for multiple rooms at a time.
>>
>>109375341
it's the singularity. That's all years from now on, gramps
>>
>>109375417
the pringularity
>>
>>109375423
my ween fits perfectly into the can!
>>
>>109375322
Ok, and?
>>
>>109375274
Jonathan Blow btfo
>>
File: IMG_5011.jpg (22 KB, 447x447)
22 KB JPG
Something pushes me to do it.. to just simply whip out my Windows 95 PCem and do some DirectX 3.0 fuckery.
All purely for educational purposes
>>
>hows game development coming along?
>development is...yeah Im working on the physics of the engine
>oh wow game physics!
>uh developing some physics uhh
doombros...
>>
>>109358365
Don’t brother with anything less than Vulkan/DX12
>>
>>109377051
So where's the guide for DX12?
>>
https://strawpoll.com/bVg8BNOwbyY
>>
File: IMG_5012.jpg (206 KB, 1179x905)
206 KB JPG
>>109377026
Holy golly gee
>>
>>109373512
Nvidia encourages you to use it. I use it with vk_ext_descriptor_heap which deprecates pipeline layouts so it’s basically OpenGL’s shader system. Don’t see myself using pipelines again outside of raytracing where they’re required. Can’t say anything about AMD, I know they technically support it but I don’t know if they literally support it. They’re still dragging their feet with unified image layouts and descriptor heap from what I’ve seen.
>>
File: floatingspheresrc.png (300 KB, 927x432)
300 KB PNG
>>109377050
im not even up to this level yet
>sadly neither is 99% this industry
>hello engine department?
>we have fallout the frontier instead
>>
File: hq720 (5).jpg (84 KB, 686x386)
84 KB JPG
>>109377127
>bro why dont you use unity or godot they have assets you can just plug in
>because look at this
>whats wrong with it
ground control, to major tom....
>>
>>109377153
conductor we have a problem
>>
File: mugi.jpg (43 KB, 564x619)
43 KB JPG
anyone participate in GMTK game jam?
>>
>>109377423
Honestly never saw the point of gamejams, lots of time and effort that could go into my real game instead.
>>
>>109376999
I find it interesting that so many Japanese games copied the sakoban block pushing. But honestly Zelda wouldnt be zelda without them.
>>
>>109377977
they are social events. autistic shut-ins wouldn't understand.
>>
>>109378937
Right, so wastes of time for normalfags to socialize (and for people who watch CGDCT to cruise for cock).
>>
>>109364039
nice
>>
>>109378937
Yeah I hate social media, never cared for your gay popularity contests or clout chasing
>>
>>109378978
>>109379121
after the last jam I participated in, the first place team got contacted by an investor and got funding for their project (it wasn't part of the rules or prizing - they were just monitoring the event). stay poor, losers.
>>
>>109379154
So they had to keep working on the dumb jam project? Sounds gay.
>>
>>109377977
The point is make dumb gimmick "theme" games instead of good games.
>>
>>109379161
not as gay as staying poor.
>>
>>109373754
>>109373983
>>109375274
I have implemented render layers, I can replace the visibility variable with this by not rendering layer 0. 1-255 will be my render layers.
I just want to be able to spawn self terminating entities in a somewhat autonomous fashion.
>>
File: 1785108691375901.png (293 KB, 2540x1397)
293 KB PNG
>>109379282
>>
>>109379282
So even for static things such as backgrounds it's a entity? Seems kinda wasteful. I would only have things that are dynamic be an entity. Far more efficient imo.
>>
File: 1785124248196442.png (92 KB, 1277x1320)
92 KB PNG
>>109380398
It doesn't have to be.
>>
>>109380505
I mean I guess that works for your use case. For me I need to construct the background from multiple tiles.
// Each sector is using the Von Neumann neighborhood: The center room plus its N/E/S/W neighbors.
static void WorldDrawSector(World *world, SDL_Renderer *renderer, SDL_Texture *tileset, vec2 offset)
{
Room *north = &world->rooms[world->curr_room->adj_room_ids[0]];
Room *south = &world->rooms[world->curr_room->adj_room_ids[1]];
Room *west = &world->rooms[world->curr_room->adj_room_ids[2]];
Room *east = &world->rooms[world->curr_room->adj_room_ids[3]];

vec2 north_offset = { offset.x, (world->curr_room->bounds.y + offset.y) };
vec2 south_offset = { offset.x, -(world->curr_room->bounds.y - offset.y) };
vec2 west_offset = { (world->curr_room->bounds.x + offset.x), offset.y };
vec2 east_offset = { -(world->curr_room->bounds.x - offset.x), offset.y };

RoomDrawTiles(north, world->canvas, tileset, renderer, north_offset);
RoomDrawEntities(north, renderer, north_offset);

RoomDrawTiles(south, world->canvas, tileset, renderer, south_offset);
RoomDrawEntities(south, renderer, south_offset);

RoomDrawTiles(west, world->canvas, tileset, renderer, west_offset);
RoomDrawEntities(west, renderer, west_offset);

RoomDrawTiles(east, world->canvas, tileset, renderer, east_offset);
RoomDrawEntities(east, renderer, east_offset);

RoomDrawTiles(world->curr_room, world->canvas, tileset, renderer, offset);
RoomDrawEntities(world->curr_room, renderer, offset);
}
>>
File: force.mp4 (2.16 MB, 640x480)
2.16 MB
2.16 MB MP4
;_;
Some ragdoll stuff
>>
>>109378978
>>109379169
Seething nogame retards
>>
gamejams are for gamedev secondaries who don't make anything good and need constant feedback to stay motivated
>>
>>109381109
>says the guy who hasn't maid anything
sure thing bro spend 8 month on your own engine and rendering setup
eventually you'll make a good game
>>
>>109381118
I've released 3 games
>>
File: anotei.gif (219 KB, 100x100)
219 KB GIF
>>109381118
>says the guy who hasn't maid anything
>maid
Not everyone wants to publish a generic slop hoping to win over the public. Some people really care about the craft and want to do everything meticulously. It's not a matter of right and wrong, just different methods.
What I can guarantee you is that very few successful gamedevs participate/have participated in gamejams, you can count them on one hand. After all, gamejams are more for socializing with other needy people than for actually making progress. That's why you almost never see anything of value coming out of gamejams, just generic slop.
>>
>>109381151
sure
>>109381182
>sly. It's not a matter of right and wrong, just different methods.
>generic slop hoping to win over the public.
Yeah yeah
>NOOOO DON'T ASK ME TO FINISH A SMALL INTERESTING GAME I HAVE TO WORK ON MY SHADERS! I AM CREATING VALUE UNLIKE THOSE PESKY JAM DEVS MAKING FUN SMALL GAMES
>>
File: file.png (371 KB, 1200x675)
371 KB PNG
>>109381210
>JAM DEVS MAKING FUN SMALL GAMES
Mind naming 1 (one) fun small game from jams? Just one.
>>
>>109381238
https://deepnight.itch.io/nightclub-showdown
>>
>>109379154
Ok but I don't want an investor? You think I'd be working on a game AND engine if I didn't already have a well paying job?

Sounds dumb, why wouldn't I just work on my game instead?
>>
File: Bronn laughing.gif (1.79 MB, 355x343)
1.79 MB GIF
>>109381252
>a generic sidescroller is a fun game
>>
>>109381267
>Didn't play the game
you are worthless
>>
>>109381260
>No don't ask me to finish a game
>>
>>109381210
Name one fun game jam game
>>
>>109381118
>sure thing bro spend 8 month on your own engine and rendering setup
At least 2-3 years, though not consecutive.
>maid
See, your jeet "hustler" mentality where you "gotta make it" as fast as possible because you're living in a slum in Mumbai and your dream in life is to escape your life doesn't really mesh with creative pursuits.

I am fine with my game taking 10 years and not paying itself back (though realistically this one will probably take ~5), if I were looking for money and normalfag acceptance I'd have jumped into Unity/Unreal and made an "action roguelike" or whatever the fuck they call them.
>>
>>109381285
see >>109381252
>>
File: fdhhr (49).png (2.69 MB, 1665x1665)
2.69 MB PNG
>>109381287
>At least 2-3 years, though not consecutive.
>>
File: type_syntax_comparison.png (84 KB, 1080x589)
84 KB PNG
Help me decide on some syntax for my programming language
Left is more obvious but more verbose
Leaning towards the right but u^ for unique_ptr and such is kinda ugly
>>
>>109381298
3D game engines take time.
>>
>>109381426
making your own 3D engine is retarded
>>
>>109357552
Way back in the late 90s, an early form of blogging took the form of .plan files uploaded to a central web server. People could browse a developer's .plan file and see what they were working on lately.

The sign is a subtle joke, etc.
>>
>>109381468
john carmack was the only person who did this
>>
>>109381447
Only if your objective is commercial success. Even then it's not retarded, just lower ROI than using Unreal or Unity.
>>
File: lu.png (50 KB, 1217x203)
50 KB PNG
>>109381426
>>109381447
Any developer who has been working on an engine from the ground up to eventually make a game is passionate about system architecture and solving a monolithic task, more than they are about making video games
>>
>>109381479
It's retarded from pretty much any standpoint, it's a ton of work unless you want to use a lot of middleware in which case why not use Unity or Unreal
>>
>>109381482
>Any developer who has been working on an engine from the ground up to eventually make a game is passionate about system architecture
Are they? Everyone I see is just following Vulkan tutorials, never seen anyone who knows anything about system architecture in here
>>
>>109381486
Most people here seem more obsessed with problem solving and building the perfect engine than they are with game design and creating an interesting gameplay loop, which is much more essential to a game
>>
>>109381499
the point is they're noobs just tinkering with beginner stuff not system architects
>>
>>109381482
Not true at all, my engine is purely a part of my game and would not work for other games, it's not a generic engine. If my game doesn't need it, I don't add it, that includes an editor window which I have no intention of adding.
>>109381483
Did you ever think that if you made your own engine you'd later be in a much better position to switch to Unity/Unreal if you needed to? There's a reason the only well performing Unreal games are those that had actual engine people on the team. There's also your own experience and skills which is far more important than anything else, as long as you gained (valuable) experience nothing is wasted.

Again this is just the "hustler" mindset applied to game development. I got news for you: if your objective is to just shit out products and make money on your time and effort investment you're in the wrong place. The odds of you doing that as an indie game developer are near zero, especially if you're going in with that mindset since a few dozen games made with that mindset are released on Steam every single day.
>>
>>109381537
>Did you ever think that if you made your own engine you'd later be in a much better position to switch to Unity/Unreal if you needed to?
No, that's some deep cope
So is calling making intelligent decisions about what to do with your time being a "hustler"
Using tools that get the job done more efficiently doesn't make you only care about making money
>>
>>109381537
It's not about hustling
it's about your passion, If you spend three years building an engine but release zero games, you're simply more passionate about engine development and problem solving than actual game design.

Real game designers are always building things, big and small, rather than waiting around for the perfect engine
>>
>>109381537
>>109381543
So this is /g/amedev version of artist grinding perspective and anatomy for month before making any real world of art
>>
File: keys.png (614 KB, 1190x682)
614 KB PNG
Asset generation is kinda crazy now if you're willing to embrace AI and be relatively discerning about it's output. Take this example - I needed some placeholder 2D content for door keys in my horror gothic FPS mod. The very first prompt produced this.
>>
>>109381543
>building skills is cope
Also you're not getting shit done more efficiently, just faster. There are countless examples of Unreal games having awful performance because the dev thought he'd just use an engine and it would handle everything for him.
>>109381551
The engine is part of the game, stop trying to separate the two to suit your argument. It's like saying someone is passionate about stock making, not cooking, because he made his own stock instead of buying it at the store.

If you're fine with being a mediocre indie dev who released a lot of games, fine go ahead nobody is going to stop you, but don't expect me to nod and go
>oh yeah you're totally right, knowledge and skills are useless you just use Unity/Unreal and it does it for you!
>>
File: gold_key_model.mp4 (1.43 MB, 1280x720)
1.43 MB
1.43 MB MP4
>>109381572
And running that in turn through a diffusion model that can cast up to 3d produced frankly excellent output.
>>
File: master_key_model.mp4 (2.63 MB, 1280x720)
2.63 MB
2.63 MB MP4
>>109381581
And you might be thinking 'sure, anon. Now show us how it did on the master key model. I bet that's complete shit.'

Well... ok.
>>
>>109381580
>>building skills is cope
The idea that you're learning skills for A by doing B instead of just doing A is cope
>Also you're not getting shit done more efficiently, just faster
That's what efficiently means

You don't need to program your own engine to use Unity or Unreal well, it's completely uneccessary and I'd even say a waste of time
>>
>>109381592
>>109381581
>>109381572
Using AI to make art assets is going to make your game tremendously unappealing even if objectively looks good
>>
>>109381596
Yeah cool. I genuinely don't care. I'm not selling anything, nor am I going to be basking in any significant audience if I don't use AI for asset generation.

These looks great, and completely blow away anything I could have produced myself.
>>
>>109381580
>>building skills is cope
Why not build your game design skills by making smaller games first?

A game engine is just a tool
It sounds like you're more in love with the tool than the act of creating.
And what mediocrity are you referring to? Some of the best indie games I've ever played used pre built engines, and those that custom built theirs could have easily used one too
>>
>>109381603
That's true but it's still slop
>>
>>109381596
>>109381603
It's fine to use ai
Just make sure all things have your tint on it
Come up with your design philosophy and idea and appy it all things you genrate so they look less generic and more coherent
>>
>>109380921
I have come to the realization that you're right. Trying to create a universal anything is just wasteful. Literally only 2 of my entities need more than a source and a destination rect.
>>
>>109381603
>Yeah cool. I genuinely don't care.
yeah, no shit. that's the point. if you don't care, why should anyone else give a shit about your game?
what else don't you care about in this game you're making? programming is solved, you can get infinite flavor text, even music. are you going to design the levels yourself or are you going to ask claude to do it too? will you slop it up all the way? then what? what's the point? what do you even get out of it?
the moment you use AI art you signal to everyone that whatever you're making is shit
so thanks for not wasting our time, I guess. make sure to be honest if you ever put it up on steam
>>
File: 1785149518536440.png (354 KB, 2560x1440)
354 KB PNG
I'VE ABANDONED MY ENTITY SYSTEM... again.
>>
>>109382055
>
>>
>>109381581
>>109381592
what do you use to make these models?
>>
>>109382181
2d was either copilot or grok. From there to 3D was Tripo3d. It did a pretty good job until I took the target verts below about 3k, then shit started looking like ein potato, ya know.
>>
>>109382349
>Tripo3d
ah, I was hoping it was from something local
>>
File: qEwS5N.png (530 KB, 630x500)
530 KB PNG
>>109377423
I liked the theme this year. Ended up being happy enough with what I came up with, wbu anon?
>>
>>109382055
good call
>>
>>109381592
>>109381581
that nice but how will you modify it? Just prompt again?
>>
File: time.mp4 (3.69 MB, 640x480)
3.69 MB
3.69 MB MP4
>>109381072
Za Wardo
>>
File: mugi.jpg (50 KB, 598x701)
50 KB JPG
>>109382554
I had fun, made a quite simple game. My first game jam. I couldn't get the web build to work tho :< Also 4 days to make a game (without a game engine) is torture
>>
>>109383519
Depends if I need to modify it. These are destined to be static models. I have the output in glf format and have been tweaking the meshes in blender, so there’s no real limitation on editing other than my own skill. Same for the 2d, it’s an editable image at a good enough resolution to modify if necessary.
Likewise, editing diffusion models are quite good now. If I wanted to put that first image back through for some changes (‘change the key shape’ or ‘swap the skull for a sigil’) I could, then cast up to 3d again via the same process.

What I don’t get via this pipeline is editable layers as you would with full masters. So things like making the flames on the master key a different material/shader that moves, or animating the hanging skull charms, are a bit more difficult. Still quite possible though.
>>
>>109381478
Him and the rest of the id team were the most well-known, yeah. But it wasn’t limited to them. They just got the most attention.
>>
>>109381468
OP he was replying to here
I was just referencing a song from the 1999 south park movie lol
"What would brian boitano do"
>>
>>109381543
This is the fallacy known as begging the question. When posted as a throwaway shitpost It can be funny-- at least, in the internet retard sense of being funny, and that's assuming you have correctly identified the targets as being deserving of careless derision. When your targets honestly disqualify themselves and it becomes clear you were dead serious about the fallacious meme you posted, it's just fucking sad.
>>
I wasted time posting this. Precious time that could have gone into making my game
>>
>>109364614
too many entities bro
>>
Should my retro CRPG be 2D or 3D
>>
>>109387024
prerendered raymarched backgrounds in isometric multistage 2.5D because KINO
>>
File: 1785186187193571.png (367 KB, 2560x1440)
367 KB PNG
>>109386566
>>109382709
The system I built could handle millions of entities but it does not matter when SDL_RenderTexture is bottle necking on this laptop at 1700+.

It's no ones fault I am just demanding too much from this laptop with a Intel UHD Graphics 600. I may as well be software rendering.

Systems without massive scale are bloat anyway.
>>
>>109387754
>intel graphics
that piece of crap can't even run HL1, which runs on a nokia n95 although at a lower res of course https://www.tomshardware.com/video-games/handheld-gaming/developer-gets-half-life-running-at-30-fps-on-a-2007-nokia-n95
>>
>>109387775
Pieces of crap is what I want to develop for.
I'm sick of this just buy more ram so you can run more slop meme.
>>
>>109387754
Sounds like retarded overdraw you could literally draw all that as one texture (or a grid).
>>
>>109385964
what?
>>
>>109387754
It could probably be done with instanced rendering, and ideally some kind of culling, but that's might be going beyond the limits of SDL's super basic and shitty 2D api.
>>
File: 1676599457024196.jpg (42 KB, 827x694)
42 KB JPG
>>109357191
Carmack would build a program to prove that you should be fired.
>>
>>109388909
>>109389193
It's beyond the scope of what I want to do at this time.
I just want to make a few simple games in SDL3 for now, in the future I could end up using modern openGL directly and then I will look at such optimizations and systems.
>>
>>109389265
The main takeaway are that GPUs really don't like redrawing the same pixel over and over and over.
Even powerful GPUs framerate tank doing that.
>>
>>109387754
You should use SDL's render targets. For static things such as backgrounds they are perfect.
You draw to the render target once on init, and you can use that same render target over and over until it needs to be updated. It can turn multiple draw calls into one.
>>
>>109390278
completely unnecessary for a 2D game
>>
File: 1785219957347900.png (264 KB, 2560x1440)
264 KB PNG
We have pipage.
>>
damn, that multithreaded by default paradigm is really nice

I wish I knew about this sooner
>>
>>109392019
It's a retarded high-concept impractical idea
>>
>>109392019
>multithreaded by default paradigm is really nice
And how are you going to synchronize it all without losing more performance than you gain?
>>
Hi guys please check out my Youtube channel :)
https://www.youtube.com/watch?v=bcWWP6-_g1Y
>>
i see lots of ppl using sdl here, what are you guys thoughts on raylib?
>>
>>109394303
Good for quick prototypes and gamejams.

Too simple and high level for a real game.
>>
I can't access the wiki... did it go down?
>>
>>109394444
what a waste
>>
>>109394529
A new eager learner wanting to get started with game engine dev is never a waste
>>
how do I land a job at mihoyo?
>>
>>109394656
>working for the bad guys
>>
>>109394656
Same as any place, build a solid portfolio and send an application.
>>
File: Untitled.jpg (74 KB, 800x600)
74 KB JPG
>>109392019
>what if we coded CPUs like GPUs for no reason whatsoever
>>
File: bruh.gif (38 KB, 220x247)
38 KB GIF
all Carmack did was add edges to Wolfenstein

you might as well go from Wolfenstein to Quake 3 OpenGL/SDL3 but Carmack is a pageant queen media darling
>>
File: smoking-my-god.gif (640 KB, 320x213)
640 KB GIF
>doom being a good programming lesson means it's a good game
>>
>>109395039
Wolfenstein is just adding raycasting to chips challenge
>>
File: forces-2.jpg (40 KB, 620x348)
40 KB JPG
>yeah dude forget drag and lift you have to manually recreate the Wright brothers original wood and canvas design before you graduate as a commercial airline pilot
>bro forget about the boeing and blackbird come back to the cessna bro its low level indie
>>
>>109395196
If this is an attempt at making an analogy with game development, a pilot makes no sense.

A better one is an aerospace engineer. How many aerospace engineers do you know that graduated without knowing how to recreate the Wright brothers original wood and canvas design? Would you fly in an aircraft designed by someone who couldn't do that?
>>
>>109395196
I don't really understand this analogy. You have to have a minimum of 1500 hours to get to ATP and grinding that in cessnas is the most cost effective way to get there. No one is touching a jet to get type rated until an employer is paying for it.
>>
File: file.webm (1.16 MB, 1280x920)
1.16 MB
1.16 MB WEBM
>>109360549
>POST WEBMs OF YOUR GAMES/ENGINES IN PROGRESS
As per your request.
>>
What shader language/framework do you guys use for more complex projects?
Writing shader in single GLSL file is fine, but I feel like if you want code reuse and especially GPGPU stuff something more sophisticated might be better.

Right now I'm using wgsl with naga-oil for rudimentary module system and tree shaking, but I'm considering moving to rust-gpu for sake of better modules, ide integration/autocomplete and generics. I have seen some graphical libraries do this after facing similar issue, but it's all emerging and I worry spirv target for Rust might not be as optimized as dedicated shader language.
>>
>>109397017
If you don't know the answer it means you still don't need it.
>>
>>109397174
I don't know the answer because I never needed anything more than few glsl files. Now I do.
>>
>>109396150
making a game in 2026 has you switching hats between engine mechanic and set design art director game loop design if you choose to use an off the shelf engine or a tweaked engine or a brand new from scratch engine.
it's hard to make a clean break from piloting an engine and populating it with game world assets vs making the engine.
Some concrete real world analogy is desired but none quite fit.
>>
>>109396150
>>109396059
I now appreciate why people use Doom as a teaching lesson
>>
which model is best at one shotting roght now? I am seeing opus 5 is doing some crazy thing right now
>>
>>109397365
games that use BSP but dont look like doodoo butt brown mush the beige box shooter Doom or Wolfenstein:
>Tony Hawk Pro Skater, Tomb Raider, The Sims, Crash Bandicoot
>>
File: 1L4Izs.gif (849 KB, 320x180)
849 KB GIF
>>109397365
>there are doom mod ports to the amiga prettier than all doom mod ports
>>
What's the difference between these threads and /agdg/ on /vg/?
>>
>>109397493
a few people in this thread actually program
>>
>>109397557
(I'm not one of those people)
I'm just here because of the vibe coded thread is next door
>>
>>109397493
People on /adgd/ want to make games
People here just want to program.
>>
>>109397603
people on /agdg/ do not want to make games
>>
>>109397017
I use Slang
>>
>>109396918
I have always wanted a game like this but you write instructions with a script or code blocks. ie
left 10
down 5
jump right
and then play out the script.

Like a Legend of Zelda dungeon but it's done with a list of instructions.
>>
File: batman-arkham-asylum.jpg (269 KB, 1500x1000)
269 KB JPG
>i can simulate all this via engineering
>i can evoke all this ethos via theatrical suspension of disbelief
>but we play games for kinetic wank not for Ultima DOS text adventure imagination
Carmack "Story in games is like story in porn. It's meant to be there but that's not why we're here"
>>
File: dither.png (6 KB, 1024x1024)
6 KB PNG
Does dithering have a place in the year of our Lord 1990 + 36?
>>
>>109398908
yes, for correctly ordered transparency
>>
>>109398467
I think Carmack would have some authority on the subject if he'd been successful when facing competition; alas, once other people started making shooters, his games turned into a series of flops.
>>
>>109399004
there were other competing engines to his like Unreal engine and the Build engine for Duke and his were better
>>
>>109395196
Gottem
>>
File: feage.png (307 KB, 550x386)
307 KB PNG
Khan Demands offering!
And acceptance of the rule of JAI
>>
>>109399019
There wasn't really much competition in the field of first person shooters, once the chads at Bungie entered the arena who actually understood how to make a fun game and not just optimize graphics, carmack's goose was cooked.
>>
>>109399395
I just told you who the competition you was you stupid console zoomer
>>
>>109396059
>If this is an attempt
Fucking autist
>>
>>109360549
there you go >>109383858
>>
Making an engine isn't the problem, especially for anything 2D.
But most of you seem obsessed with engines, with literally zero discussion about any of the other disciplines required to make a game, like game design or art
>>
>>109399435
read the thread title genius
>>
>>109399436
Yeah it say
"GAME and engine development general"
but i don't see any games at all
>>
>>109399435
wouldnt it be a nightmare to release a vibe coded art project that glitches in an epic crash out ruining your reputation your one shot at fame and fortune? nooo buy a macbook and more tokens! the world needs more furry fiction!
>>
>>109399385
>guys i have been learning solid state physics trying to redesign my transistor for my sokobon game that blows
>>
>>109399456
How many more months are you going to spend solving problems that were solved 20 years ago just because you don't want to deal with game design?
>>
>>109399464
there are 3 game dev threads on 4chan, if you dont like programming engines you can go to those ones instead
>>
>>109399464
im not reinventing the wheel
im learning how to drive
and not hiring godot uber every mile
>>
>>109399471
I think your distaste for game design and actually making games is amusing

Also, where's the third one? I only know about /v/ and /g/
>>
>>109399481
who said I had a distaste for game design and making games
>>
>>109399477
Are you getting a bachelor's in mechanical engineering just to drive a car?
>>
>>109399491
you're not as smart as you think you are
>>
>>109399491
to sell a car to a factory yes (thats what engineering is)
>>
>>109399485
Seems like it
>>
>>109399502
I just think you're dumb for going into a thread and complaining about the thread topic
But barely anyone talks about programming in here either it's just all just AI slop and posts about nothing
>>
>>109399497
Okay
>>109399496
Sure
>>
>>109399512
I said first and foremost that making an engine is cool and all, but it's not the same as making games

You can't pass one off as the other
and there's a severe lack of actual games in this thread
>>
>>109399520
an engine is part of a game
>>
>>109399522
It's a tool and a tool doesn't have much point if you're not using it for anything, or if you spend years just building the workshop while producing no output
>>
File: DEC_VT100_terminal.jpg (2.1 MB, 2776x2464)
2.1 MB JPG
>>109399512
If you think about game dev what we know from interviews from Andy Gavin vs Playstation Sony is that both sides of the bargain are evey secretive and talking vague technicalities until the IP is rock solid and basically patent ready to shop for a distributor. I can tell you're not an artist not a programmer not a man of craft but less than investor level heckling.
Like we can talk about Pong having 0 lines of code as means of learning the craft. We can talk about Doom as a geology of game dev history that all circuits felt the seismic impact of. We can't post until it's advertiser ready and by that point, buy an ad. Otherwise share your pedagogical journey.
pic rel I learned C is backwards compatibility baggage hell and ironically Apple is setting the trends. Gamedev is really tech first IP later. Movie tie in games prove this to be the rule. Amiga games had amazing art but are lost to time and honestly I have no interest ressurecting them. However I am enamored with Ecstatica with compression artifacts with VGA era constraints as the high art film stage equivalent of gaming over our grey bland shiny rain unreal simulator Alan Wake cringeathons (thank you Kojima)
>>
>>109399547
a little editing with grammarly and perhaps editorial charity would have done me good
>>
File: 1744796485787f.png (742 KB, 800x797)
742 KB PNG
Tourist here. Is this a good thread to "hypothetically" get in contact with talented 3D artists? I need "advice" from someone with experience with Blender and GLB to VRM exporting. On a totally unrelated note, I also have money.
>>
>>109399547
I see your point, but I doubt most people in this thread are being secretive about their IP

It seems like they just want to work on the engine all day and (some) have this assumption that a good game will somehow manifest itself if the engine is good.
>>
>>109399595
Probably not, Go to 3DCD, reddit or join one of those 3D artist discords
>>
>>109399601
thanks I'll try that.
>>
>>109399595
indiedb
>>
>>109362392
handmade hero
>>
File: df.png (175 KB, 565x965)
175 KB PNG
anybody knows what is actually happening with this industry? it's been like 3-4 years of this shit already, and there's no end in sight. even established studios are closing down left and right. like wtf, aren't people buying games anymore?
>>
>>109400753
who the fuck is making games? short development cycles are 10 years now
>>
>>109400753
I wish I knew, games definitely aren't selling less, I haven't worked in the game industry since the 2000s and I have no idea
Maybe just a general symptom of the economy being in the shitter thanks to covid / wars
>>
>>109400753
Industry was bloated from malinvestments driven by agenda. Most of these companies have not been profitable because they make shitty games no one wants but they were still getting funding to do so. Double fine was owned by Microsoft and while Phil Spencer was CEO of Xbox he was basically using the massive profits from Minecraft to keep most of their other game studios afloat. Microsoft restructured last month and made double fine independent again so they're having to slim down to stay operational. t's probably not gonna be over for a while.
>>
>>109400820
>Industry was bloated from malinvestments driven by agenda. Most of these companies have not been profitable because they make shitty games no one wants but they were still getting funding to do so
this implies that people have been buying less games and thats not true
>>
>>109400753
Game studios are ruining themselves from inside by adding predatory monetization systems to keep investors happy or hollowing themselves from inside to lower their own price enough to get some retarded investor to buy it whole and hold the bags.
They do not know how to make games anymore. They do on first when they are small and have ideas, but once they hit big they feel the obligation to just increase production value to somehow guarantee increased return because that's what investors want, but that's not how art works.
At the same time indie developers do not give a fuck about any of that and just make games with low cost in high volumes and when they hit big they get a lot in return. It's much more natural business model for an artistic medium, but investors and big company management hate it. They want guaranteed returns even if it cost a lot.
>>
>>109400860
adding predatory monetization systems doesn't explain why people are losing their jobs
>>
>>109400844
>and thats not true
right...
>>
>>109400870
that doesnt disprove what i said
>>
File: 1785323861627996.png (373 KB, 2560x1440)
373 KB PNG
Added spinning collectable coins.
>>
>>109400874
if you think people are buying new games for full price and then backlogging them, you're as crazy as she is
>>
>>109400887
I'd be happy to be proven wrong, but that fact doesn't prove anything, most games people play are the same multiplayer titles they've been playing for decades like CS and Dota2
>>
>>109400868
The cause is the same, increasing company revenue at all cost.
People are losing jobs because the management rather fire expensive long term workers and hire some jeets that will vibe code next version of their high paying franchise.
>>
File: file.png (545 KB, 800x600)
545 KB PNG
>>109398344
colobot
>>
>>109398344
So like every Zachtronics game? And 7 billions humans. And >>109400909
>>
>>109400909
>>109400924
Yeah but you're in a Legend of Zelda dungeon and you need to push blocks, collect keys and attack enemies, etc.
>>
>>109400903
Firing people doesn't increase revenue
>>
>>109400955
It reduces expenses relative to revenue which makes the whole company more attractive to potential buyers.
>>
>>109400964
it reduces expenses and revenue, they can't make games without staff, and they're not being replaced by AI or Indians
>>
>>109400844
You're taking an average and applying it to everyone. People only buy the games they want. Successful companies aren't the ones having layoffs.
>>
>>109400943
(You) could make one right now!
make a 2d ass room to room game
make up syntax and parser/lexer (easier than you think)
turn it into something your "interpreter pattern" can consume and act on and voila
>>
>>109400999
>Successful companies aren't the ones having layoffs.
Bethesda
Bungie
>>
>>109400969
They don't need to make any games in order to have the entire company bought. You are thinking like a developer, not business owner/investor. If you want some retard to buy your bags you do not care if they are actually good bags, you just want to sell them to some naive idiot, get all the money in cash and then fly to Thailand fuck underage prostitutes in penthouse for few $ a day for your well deserve vacations.

None of these things are made with making games in mind, it's all about milking it dry at low cost and then selling it all as a package at high cost. That's all that matters to investors.
>>
>>109401011
They aren't selling these companies though
Double Fine got cut loose, nobody bought them
>>
>>109401021
Just wait for it
>>
>>109401029
Microsoft cut Double Fine loose
They have not been sold to anyone, Microsoft gains nothing from this and will not gain anything if Double Fine gets bought by someone else
>>
>>109401037
>They have not been sold to anyone
Yet

>will not gain anything if Double Fine gets bought by someone else
They will gain money
>>
>>109401076
If you no longer own something you can't sell it you fucking idiot
Double Fine is independent, no longer owned by Microsoft
>>
>>109401080
Yeah, but the current owners can still sell it.
>>
>>109401091
An investor isn't gonna look at your game company and say "wow you fired everyone because you're going through tough times, this looks like a good opportunity to buy"
>>
>>109401101
No, he will look at "oh, so this is your revenue, these are your expenses, and these are all intellectual properties you own that we can repackage and sell someone else for a markup if needed. Seems like a good deal"
Going through tough times is when companies are mostly likely to be purchased by someone else.
>>
File: 1785326864271370.png (17 KB, 885x890)
17 KB PNG
Tomorrow I will add menu and UI.
So close and I will be don't with this piece of shit for every.
>>
>>109401134
I can't even type straight my brain is so fried.
>>
>>109401121
>Going through tough times is when companies are mostly likely to be purchased by someone else.
Second most likely, most likely time is when they're doing really well
And no buyers do not merely look at two numbers and an inventory, they're not stupid, purchasing a company in trouble isn't that appealing
These moves are purely to decrease expenses at Microsoft and you can't really spin it as greed
>>
https://github.com/SeraphYuki/ragdolls-like-me
i really wanna program a physics engine but i'm not smart enough.
>>
>>109401315
do you want to do anything different that other physics engines don't do?
>>
>>109401371
no i just want to feel smart.
>>
>>109401010
Starfield came out years ago an probably didn't make back enough money to justify it's costs.

>Bungie
lol
>>
>>109401393
what do you think about this >>109383858
?
>>
>>109401441
i think they are smarter than me.
>>
>>109401457
I am that guy and i am a dumbass

Making cool looking game physics isn't actually that hard
A lot of it is just faked, not simulated
You should start with very basic AABB collision and see how far it takes you
Honestly, it can take you pretty damn far before you ever have to deal with rotations and all that fun stuff
>>
>>109401509
didn't you have to program jacobians with lagrange multipliers to have joint constraints?
i've been trying to reverse engineer this https://github.com/id-Software/DOOM-3/blob/master/neo/game/physics/Physics_AF.cpp
but i can't do it ;_;
>>
>>109401431
Starfield was a flop, but Bethesda has a massive string of successes before that
>>
>>109401521
>didn't you have to program jacobians with lagrange multipliers to have joint constraints?
Fuck no
You just use very simple math to clamp the X, Y, and Z angles depending on the joint
Also the trick to solving ragdolls is this little thing called verlet integration It’s not perfect but it looks great most of the time
>>
>>109401587
ok i'll try it like that



[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.