/gedg/ Wiki: https://igwiki.lyci.de/wiki//gedg/_-_Game_and_Engine_Dev_GeneralIRC: irc.rizon.net #/g/gedgProgress Day: https://rentry.org/gedg-jams/gedg/ Compendium: https://rentry.org/gedg/agdg/: >>>/vg/agdgGraphics 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 thread >>108752425
>>108777370>>108780057>I don't wanna shit on your effort but this doesn't look appealing, I don't think its worth pursuing furtherThanks but I will continue.>>108780842>As others have pointed out, your main problem is that the trailer is just a list of mechanics. Stuff like that was appealing 20 years ago when things were newI started the project 15 years ago. I should have publish earlier.>but it doesn't work anymore.What does work now? (if anything)>What's the point of your game? What's the objective? What's the gameplay loop? Mechanics just enrich that.There is supposed to be a career/story mode where you take part in various events to win them, then you advance to next event.
>>108780119>It is not a bad project for a resume.>(Of course, that's assuming they used Vulkan and not OpenShitEl)It's OpenGL - OpenGL 1.>theyIt's not they, it's one guy.>>108780175>resume>New to getting jobs?Right. For a job you need nepotism, not resume.
>>108780958Reminder that GTA 1 would have flopped if not for the police car bug.
>>108780872>Genre.So no racing game?What about GTA game (in 2D)?
>>108781257>What about GTA game (in 2D)?They exist, GTA1 and 2, that's the OP image
>>108781231>What does work now? (if anything)The trailer just has to catch people's attention, it should show off some features but not seem like it's just going through a list. If the trailer catches their attention, they'll go to the Steam page to learn more.>There is supposed to be a career/story mode where you take part in various events to win them, then you advance to next event.Yeah so this kinda stuff could also be shown.Again it's important you don't just make a list of clips for each feature then play them sequentially. Movie trailers aren't made of 10 second clips for each actor, or every big battle in it.
https://www.youtube.com/watch?v=695CiqtPEVEOpenGL fags take another L kek
>>108781242>It's not they, it's one guy."They" is used as a gender neutral pronoun, and before any poltard comes seething about trannies, I am Catholic. Its use as a gender neutral pronoun predates the industrial revolution let alone trannies.
>>108781404NTA but I always default to "he" because somewhere there's a tranny getting mad at it.
>>108780958who needs gta6 when you got GTA1
>>108781412Yeah well, trannies don't live rent free in my mind, so I don't base my vocabulary around them. Though the tranny that frequents this general does piss me off sometimes.
Not either anon and idk what you fellas exactly talkin' about, but this reminds me: The guy who made Wall-E (Andrew Stanton) is one of the last remaining good OG creatives left at Pixar. He wrote Toy Story 4 (which didn't need to exist and it shows) and probably got the flak for all that was wrong with that movie, maybe even accused of being woke. But you know what's funny? He's actually a devout catholic
>>108781257>>108781298Also GTA Advance, the weird 1&2 port, and the DS/PSP versions of Chinatown Wars
>>108781501You can't be traditionalist and woke at the same time, he's probably a prot.
>>108782025aye, fair enough
>>108781254>zoomer fantasies that they actually believe
>>108781385He's right that graphics APIs just suck
>>108781385love when blow doesn't know how to give an actual answer so he just prattles and whines
>>108781385>spend 10 years to make a new language>language is explicitly made with game developers in mind>the only thing resembling a game in those 10 years was a remaster>it isn't developed in the language you created for this purpose
>>108781385>Shits on "OpenGL" in general but only complains about OpenGL 1.0 being designed for 1990's hardware>Never suggests that he's used any modern version of OpenGL before because it's a "waste of time">Shits on Vulkan without any reason either>Praises DirectX (as if isn't just as ancient as OpenGL)>Suggests Metal as a good API>Then admits that he never used it before it's just what he heard>"Here read this article" doesn't elaborate further
>>108777370>>108781231Your game looks promising anon. I like that you've found a niche somewhere between Micro Machines and GTA 1. There's definitely a market for a new-generation Micro Machines type game in the world.It would benefit from a lot more decoration around the environments, to introduce a sensation of depth (see the tall buildings and elevated railways in the GTA screenshot, for example).
im vibecoding my engine and its honestly coming out okay.
I'm going back to C
>>108782268>>Suggests Metal as a good API>>Then admits that he never used it before it's just what he heard>>"Here read this article" doesn't elaborate furtherHe's claimed something like it's impossible to know if a language is good or not without using it for years (doubtless to stave off ambushes by the Rust Evangelism Strike Force.) But I guess a graphics APIs just aren't as complex as a language, despite (speaking loosely) them all containing one.>>108782025Aye, he be reckoned one of them Protestant Catholics.
>>108782025>You can't be traditionalist and woke at the same time, he's probably a prot.Have you looked at the Catholic church lately?
>>108783767from what?
>gamedev Youtube video>some blurb about always loving games and being inspired by them>reel is just a bunch of the most surface level shit possible>Zelda, Super Mario, some scrimblo platformer>newer inspirations are Cuphead, Ori, Hollow Knight, etc>using Unity or Godot>game is some unoriginal Zelda clone or platformer that plays more or less exactly like they did 30 years ago but worse and looks worse too>27 Steam reviewsI may have been going at it for 7 years without a release, but at least I'm not these guys.
>>108782015ctw is peak game
>>108781385He’s being a little disingenuous with the 1990 thing but is still completely right. There’s currently nothing to gain from OpenGL anymore. You either take the deep dive with modern Graphics APIs or you don’t, and then use the “no graphics api” blogpost as an abstraction guide.
>>108785534opengl's still easier to use than vulkan
>>108782268You've never used Metal. It's genuinely the best graphics API out there, because Apple isn't afraid to drop compatibility so they have been improving it without having to worry about some autistic neckbeard running a specific Linux distro version from 2015.>>108785557And Python is easier to use than C++, so what? Learning Vulkan is trivial in the grand scheme of things, how did you figure you're make a whole game if learning Vulkan is too much for you?
>>108785557Also this isn't true anyway and just shows you've only done trivial projects, probably yet another 2D platformer to pollute itch.io. Trying to do anything even remotely modern or optimized in OpenGL forces you to create your own structures and abstractions, which run on top of OpenGL which in turn runs on top of very similar structures implemented at a lower level.Maintenance hell is what it is.
>>108785570So making things easier to reduce your workload is a good idea
>>108785575You create your own abstraction over Vulkan too buddy
>>108785582Ok so we agree you create them in both cases, and these abstractions make up the bulk of the graphics API work.So why do it on top of a janky, buggy API that was outdated in 2010 let alone 2026?
>>108785587>So why do it on top of a janky, buggy API that was outdated in 2010 let alone 2026?because it's easieralso janky yes, buggy not really
>>108785578Yeah but you don't program in Python, do you? Why is that?>because something this complex in Python would be hell to create and then maintain or expand onBingo, there you go, exact same pattern as OpenGL vs Vulkan. >>108785591>because it's easierIt's only less boilerplate at the start, it's not easier. The API's difficulty itself barely registers, the difficulty comes from maintaining and developing everything you build on top of it, which is much harder in OpenGL (also because a lot of the time you're working over a black box). >buggy not reallyThere are multiple known OpenGL driver bugs that will most likely never be fixed and you have to work around them.
>>108785610>Yeah but you don't program in Python, do you? Why is that?because it's way too slow>It's only less boilerplate at the start, it's not easier. that's why it's easierthe abstractions you build ontop of vulkan are literally the same as the ones you'd build ontop of opengl or any other graphics API
>>108785570>You've never used Metal. It's genuinely the best graphics APII agree. It's a joy to use anything Apple, but its docs are not beginner friendly by any means, and you're vendor locked to a system where no one plays games.
After getting the DX12 renderer up and running with an initial basic demo scene, it runs worse (2500-3000 fps) than the equivalent DX11 test case (8000 fps).Wonder why that is.
>Today, yesterdayWater-work. I was initially going to have the source play a spreading animation, but I think it will be much more efficient to recreate it from the individual voxels. Any change would require scanning the world for potential sources, and there could be multiple. Also I've been able to ditch a HashMap<VoxelId, LiquidFlow>. Instead the different states can be found from the VoxelId and its parent VoxelId.
>>108786514I'm not a low level engine developer but I've heard multiple times that DX11 has a lot of optimizations that DX12/Vk don't have so you should expect considerably far worse performance all across the board. DX12 and Vk require a lot of finetuning to even match DX11Not entirely sure if it's true
How the fuck people in the 2000s did game engines? There were literally zero tutorials onlineBooks?Better yet, how the fuck have they done it in the 90s
>>108787921I don't know, how did people do anything before the internet? It truly is a mystery
>>108787921>90s>no books no internet on engine deving>yet devs gonna devHow
>>108777370>>108781231cool shit anon. i don't think this is marketable as a single player sandbox-y game, but it's cool, fun tech.
>>108786514Even D3D11 has relatively expensive drawcalls compared to D3D9. What I would do in your case is check your app in tools like Pix in order to identify bottlenecks or things to improve with your setup. Also make sure your not doing something dumb like bechmarking a debug build.
>>108787921Same way anyone making anything decent does it nowadays.>try to do thing>encounter problem>see if you can find information or solution to problem>if you do, great>if no, solve it yourself>repeat until you have a game
>>108787921>Books?Yep. Look at something like "Black Art of 3D Game Programming", back in 1995 it had everything you wanted to roll your own 3d software renderer.
>>108787978it was mostly instiutional knowledge in game companies
>>108787921Before copying stuff from stack overflow or asking a dumb ai to do it, people often used books (e.g. game programming gems), SDKs, and documentation sets like MSDN (it came on CDs!). Starting from the late 90's, there were also tons of tutorials and forums on places like gamedev.net and flipcode.
>>108787990You’re right, theres a copy on archive
oops downloaded android studio again ooooops might write something from my github again but nobody pays attention to my plan because they arent activists.probably dont even ctlr+shift+tab enter for every captcha to skip
>code is a bit messy>ok time to add some metaprogram code standards check>1265 errors found>I have to go through them one by one and fix them now
Considering making a Sims clone that uses regular controller controls instead of mouse controls.
>>108780958Made this today:
>>108788921Trust me, it's not slow, I nanosleep for 160ms every frame so it doesn't flashbang me...
>>108787921wouldn't it just depend on what hardware they had and whatever api it surfaced for showing pixels on the screen? they just made games for specific hardware back then things are only so complex now b/c of the hardware variety. Game engines themselves aren't really that complicated just managing state, resources, and displaying stuff. All your asking is how they did each of those things which would be it depends on the hardwarelike it wasn't easy but its still just writing pixels to the screen and updating them based on input there is nothing that intrinsically complex about that. Most shit people have in game engines nowadays isn't necessary for an actual game
>>108786514usecase for rendering at anything higher than what a monitor can output?
>>108788994Pixel density. Great for screenshots and fixes aliasing problems. Overkill? Yes but it's not hard to implement.
fuck /agdg/ is completely unusable but I feel too stupid to post my shit little prototypes here since I don't know any real coding and just use unreal blueprints
>>108789060Eh, developer threads are pretty ass to post any real progress. Nobody is interested in playing other people’s games in dev threads and if you do something particularly good half the posters will become defensive and insecure and will hate on it
>>108789040wym a screenshot? A screenshot is always 1 frame it doesnt matter if your game is running at 8000fps or 30
>>108789091Holy fucking retard NTA but upsampling reduces noise and aliasing artifacts. One of the best AA techniques is just rendering at 2x the resolution (although it does have some moire issues with high frequency texture aliasing due to filtering)
>>108789100Bro...>>108789091You're absolutely right!
>>108789100ok you dont have to be so mean i was not being facetious just genuinely curious
>>108780958God i love raylib
>>108789090but I like seeing everyone's progress...
progress
>What a concept — this is genuinely ambitious and creativeThis is AI speak for saying my idea is totally retarded, isnt it?
John Blow says use raylib.https://www.youtube.com/watch?v=695CiqtPEVE
>>108789712Sometimes AI tells me my game wont work, so maybe not.Try giving it an obviously shit idea and see what it says
>>108789752raylib uses opengl
>>108789873Congratulations on knowing that, timmy! Good job!
>jon blow clipwatchers
best begginers tutorials for blender? Blender is where you make your models for you game right?
>>108789933You got a problem with my channel, bubba?>>108789951Entirely depends on what you want to model. There's tons and tons of tutorials for blender I don't think you can go wrong just using youtube and searching. I don't think there's a real definitive starting point.
>>108789962I guess character models.
>>108789951grant abbitts stuff
>>108789933
quite frankly netradiant is better than any level editor I could put out in a reasonable timeframe so why should I make all my levels in it
>>108789460Looks cool.
I have a confession... whenever I see progress posts I don't respond to them.
>>108789460>white people would actually still be living like this and seeing nothing wrong with it if blacks didn't carry them through the industrial and technological revolutions
>>108790350stop flexing that you got a local jamal.
>>108790340i dont respond to them but i like seeing them
>>108790313quake style map editors are only good for making quake style mapslow-poly, blocky hallwaysyou can see the limitations of this in games like Half Life 2
>>108790459NTA but that should be good enough, desuI use tessellated curves/patches from Q3 map files and it helps prevents everything looking so blocky but honestly as a solo dev I don't see why a little blockiness is all that bad. half-life 2 had loads of people working on that game I'm not trying to overscope to try and compete with games made by huge teams
>>108790530The alternative is you make maps by placing pre-made models and tiles which looks better in most cases
>>108790540looks better but quick iteration and grayboxing is the most ideal way of working. just add models on top of your map for detail. this is solo work you don't need crimson deserts graphics
>>108790557>quick iteration and grayboxing is the most ideal way of workingyeah but that's got nothing to do with quake editors
>>108790560it does because quake editors are extremely fast at iteration and grayboxing and takes very little skill or time. that's the entire appeal of using netradiant and the likes
>>108790570>quake editors are extremely fast at iteration and grayboxingso are non-quake editors
>>108790573>ignores the restok
>>108790576Everything you said applies to any good editor
>>108780958Sure, why not?
>>108790587Clearly you're always right and don't like to hear anything so sure! You're absolutely right! I'm sure your AAA quality game is right around the corner!
>>108790595you dont need to get mad about it
>>108790597You're absolutely right! Let me fix that for you!
>>108787921You had to be a straight legit hardcore nerd. Gabe Newell, John Carmack, Tim Sweeney, etc.
>>108790893yes in the 2000s only celebrities made games
>>108790920yes thats how daiktana got made
porting native project to Android is a nightmare. I haven't touched Android in ages and forgot how bad it was.
Should I make a sims clone or just make an engine remake of the sims games
>>108790600ai psychosis
your game has a denylist for not-so-good words hardcoded into the binary, right?
>>108792545no. you should get a bonus points multiplier for using non repeating slurs
>>108792545>damienWhat did Damien do to piss you off? Niger is a country. Wtf at all that shit on the bottom. Iron????
>>108792018I think people underestimate how complicated the sims is. Take a crack at it I love the sims and you'll probably learn quite a bit from the project regardless if you finish it.
>>108792545>double nigger>can't say dong>wtf did damien do
>>108788994There's not much use except for knowing how fast you could go
