dd9 edition>Demo Days Next: https://itch.io/jam/agdg-demo-day-71Previous: https://itch.io/jam/agdg-demo-day-70>AGDG Resources https://hackmd.io/SNd2YQugRG-36CLYc8Og2w>Progress Posting Guide https://pastebin.com/0xEt5atn>Helpful Links New Threads:>>>/bant/agdg>>>/vg/agdgArchives:https://archived.moe/bant/search/subject/agdg/https://arch.b4k.dev/vg/search/subject/agdg//agdg/ Steam Games: https://made-by-agdg.vercel.app/>How to WebM https://obsproject.comhttps://github.com/argorar/WebMConvertersize limit too small? >>>/wsg/other+boards>Previous Threadhttps://archived.moe/bant/thread/24917433/
proggy anchor
Bump
I wanted to make a medic robot; and I really like Tachikomas from Ghost in the Shell Stand Alone Complex.So I looked at some images of them, then saved an edited screenshot of one for my concept art a while ago. Just recently, I built from my concept art - and then kind of realised as built that it is just a Tachikoma without legs. Going to animate it; stick it hanging from a rail in the ceiling and then see how I feel about it still.
>>25008542Now I can't unsee that Tachikomas are walking toilets. Thanks asshole.
>>25002383Working on maps. This one's way too huge so I'm gonna shrink it.
>>25002383trying to make some Agua ardiente (a classic alcoholic beverage of the region) sprites, but it looks like cum.
>>25011237The rum on the other hand, now I think that looks betterWhat do you think agdgies
mira 0n https://github.com/untermischling/gamblingpoly
>>25002383hello all, i have been hard at work on my RPGM game. i now have a bunch of equipment, various time loop systems, and a custom UI with hotkeys.https://platonicsolidgames-ahm.itch.io/save-everyone-demo-3
>>25011237I made the "highlight" band of pixels match your rum one to see how it looked. It still looks off since it is a grey liquid.It being on a grey background I think is throwing things off, is it actually transparant in your game? I recommend putting across several backgrounds to see. And if it still looks shit, try a simple "blue for clear" visual shorthand like you see in many video games, where water is simply blue.
>>25002383not much to show right now, did:>refactor my shaders, super organized and easy to make changes to them now>decided on art style/aesthetic for the game>changing a few models, their normals, atlas texture, etc to fit into that styleand more gameplay>more parkour/traversal, instead of just ladders, the player can now climb through windows and shimmy across pipes/some ledges>my NPC director is a little better nowalso doing the story stuff and deciding on some of the levels. mines the splinter cell style game. kind of debating putting in disguises like in hitman, but i'm not sure if i want to do that yet.
>>25008542looks good, are you doing vertex colors or using a texture for the colors?
>>25013422Actually you're right, I should make it transparent; the beverage is straight up like vodka, fully clear and transparent but you can kinda discern it.I did what you did with the details in the upper part but also toned down the liquid to be more transparent.
>>25002383made some streetlights, the lamps sway in the wind too
>>25015088Looks good and creepy.
>>25015088What's the gameplay like? Wish you posted a webm.
>>25019556>>25019556 #I post videos quite a bit in the agdg stuff so I didnt wanna take the time to make webms, here's a couple ones from last week>>6236744>>6236745
>>25019702Lmao I hate this website>>>/wsg/6236745
>>25019715>"2d" game with 3d killcamsoriginal, to say the least
>>25019715Neat. Is navigation full 2d or can you take a turn where camera would rotate 90 degrees? Without that the stealth elements would suck unless you work on level complexity.
>>25019973No turns but there are elevator platforms that will take you forwards and backwards in that space so I can create mazes of platforms that all still go left to right
>>25020736Sounds like a wasted opportunity, it would be kino with corner peeking and flanking, as long as you would figure out how not to make levels confusing.
>>25020801Yee turns just get two chaotic. I was thinking about maybe doing spline paths so he moves along curved levels at least but that would get really complicated trying to make the bullets move in a satisfying way.If you check this out this kinda shows what im talking about with depth in the levels.In example one, the player needs to dispatch or pass this enemy that is watching down the hallway, he takes elevator one to path two then takes elevator two back to path one to flank the enemyIn example two theres an obstacle in the path, the player takes path elevator one to path two, elevator two to path three, then elevator three back to path one.In this example the elevators are going forward and back but they also go up and down so I can build some pretty complex layers just off this. Im gonna also code the ai so they can switch from 3d fighting to the 2d planes so popping around a corner cover would be more to keep them from advancing to the players layer
>>25020929Forgot pic related lol
>>25020929Not as engaging, unless you need to watch out where elevators take you, but good enough. Hopefully you can snipe things across planes too.
>>25021157Yeah its a compromise, plus the further I go along its becoming less metroid and more contra so stealth will probably only be a main focus for a few of the spookier missions.
>>25021193So I'm guessing it has some shadow complex dna?
>>25021317Yee id say the main influences are Shadow Complex, Metroid Dread, and Contra Shattered Soldier and maybe some Deadlight thrown in there too
>>25002383>>>/wsg/6237096Moar rock textures + caching works better now so they're not all running live and lagging the game out
fixed my memory leak issues with a rendering thing I've been working on for the past few months :)
>>25022681Looks great, what are the pink and white objects supposed to be though?
>>25013965Not sure if it is considered vertex colours, but I manually shade each part of the model with custom materials and never do a full UV wrap texture. That's a screenshot from Blender before my real materials go on, but the game is pretty much the same flat-colours with some lighter/darker bands.If I use textures, it's sparingly and usually an alpha mask. I do fairly well making shaders and stuff though for particles and things. All my UI elements are quads and elaborate shader magic stacked.
Almost out of tileset jail
Neat seeing this stuff. I stopped gaming years ago because my Xbox broke and I didn't have a PC at the time. Just fell out of it mostly. I was big into the early days of steam greenlight and even before that I was pretty big into indie games. Started with project zomboid, way back before they started with their new engine, which I think it still runs on today. Before it was a heavily pixelated look that was really comfy, but the devs basically made the choice to scrap working on that engine after they lost a ton of progress. I can't remember if it was that a computer was stolen, which had their master files on it or if it was a flood. It might have been both. But they had some extremely precarious days back in the early 2010s. Now they're worth like 45 million dollars. Same deal with the guy who made kenshi. He was living in a shitty house driving a beat up car eating canned beans for a while there. I didn't think kenshi would make it honestly, the vision and scope was so massive for a single individual. He had always planned on bringing on extra devs when money came in, but same deal there. When it first came out, it was incredibly bare bones.There were a whole fuckload of terrible games that came out of that era but out of the chaos, there also arose a class of developer that we were previously denied, with the shift to 3d and massive development costs, for a long time there people didn't think guys in a garage could make a game and sell it and hit it big.And really, all that paved the way for this here today. If I were ever to get back into gaming, I wouldn't be going back to play the 10,000th assassin's creed, I wouldn't go back to play call of duty. I'd be going back to check out the new interesting ideas that indie devs have put out in that time.The gap in quality is just that vast, and there is something that shines through in these kind of titles, despite their roughness and jank, and it's something that mainstream companies simply can't capture.
So best of luck to you all, I hope that you find your niche, that you are able to translate the vision in your mind to the real world in tangible fashion, and that you find success doing it. Whatever that might mean to you. It might not be a world I interact with directly anymore but it's still something I take an interest in. Mostly about the history of the industry nowadays. It is very cool seeing you guys post your stuff in agdg as well. Process is the longest part of the journey and it's one I have never actually gotten to see firsthand in that manner, so thank you all for sharing your work. It's quite amazing the variety of different ideas, genres, gameplay concepts. In my mind it's testament to the versatility of the medium.I often think about what drove the split, for my own experience. Obviously the wonder of childhood plays a part, but old games just felt different. A big part of it was that the "magic" behind games wasn't popular public knowledge at that point, and the other was that I came up in childhood at a time where people were still grappling with what a video game even was.For example, you would have games that were a platformers for like 3 levels. Then it would suddenly turn into a shooter. And then back to a platformer for a level. Then a rail shooter. Then you'd be doing something else completely different. I first started noticing those differences while watching a documentary about the medium of video games. They had a guy on there who was talking about the evolution of gameplay. He pulled up some rpg sort of game. And he showed how there was a score constantly visible in the GUI, as well as other things you'd expect to find in a game like Pacman, asteroids or any other old arcade game. He noted how this was a relic, how it didn't really make any sense to put it there and the way the developers had integrated it, didn't communicate what the criteria for gaining points was. They put it there, because that's what was in video games...
Whether that is a result of a corporate suit sticking their finger in the soup (anyone remember the everything must have multiplayer days?), or the developers simply being used to having that as a feature of a sort in games, who knows. Probably a bit of both.Anyways, not trying to detail the thread. Just thought you guys might find some of that stuff interesting, and perhaps glean something from it for your own designs and ideas. Love what you guys are doing here, and wish you all the best in your creative endeavors.
>>25026092Mystical Ninja Starring Goemon on the n64 was like this. It's like, Zelda-like, but then all of a sudden you're in this Gundan/Power Ranger Robot that looks like a little boy (the size of a sky scraper) and it's just like, ok I guess I'm doing this now. It also had like, Japanese gameshow like elements. Then it would switch back to being an exploration rpg, then it threw mario party like mini games at you, and eventually you became a super sayin. One of my favorite games growing up.
>>25026072Back to making RTSs?
>>25026072ramp looks wrong
>>25026072>>25026228I think it's looks really good for 99% of it, it's just, at the intersection of the flat top, and the top of the ramp, it kinda blends together so it can mess with the eyes a bit and then look flat all the way. I think that's the issue but I don't have an artists eye really, I'm just trying to think what that 1% off feeling is. Love it otherwise, looks really cool
>>25026072Yeah make the ramp stairs instead, the texture is doing you a disservice with the perspective, feels like looking at one of those optical illusions where the table has a leg that doesnt go anywhere. The walls and the metal eith the bones on the ground is excellent tho, easy fix
Only have positive thoughts, okay? Stop having negative thoughts. You're casting dark spells on yourself. You're hexing yourself. There's no benefit to that. That doesn't mean that you are delusional. You should be objective about yourself. You can say "my art is trash". I say that to myself all the time. I always think that my art is garbage, okay? Or everything that I do, I think it's trash, but I sandwich that with the thought that I know why it's trash, I will get better, I know what I can do to get better, and I will do it. That's how you have to think. Always have an ultimate positive framework to any thoughts that you have, even while you are being extremely critical on yourself. You wanna be extremely critical, you are gonna be the most critical on yourself of any person, but you're also gonna be insanely, 110% faith positive. And combine that together, and you will be able to achieve anything.
>>25026609>>25026599>>25026228Escher’s Ramp, guys. I had a fix all ready.>>25026120Motor Spirit is kinda just waiting for the artist. Maybe 2-3 more features.
>>25026699https://www.youtube.com/watch?v=EbunGcnfDq8
>>25026699
>>25026902Interesting, so it will have more than UI management or is it just visuals?
>>25026902>Escher’s Ramplooks way better, you should throw one more line at the top between the columns where the slant begins
I don't have proggy to post today.Thanks for reading my blog (my blog.)
>>25023110Granite. Matches the reference image, idk why it doesn't work as well as the rest.
>>25027588Works for me, but maybe some aren't used to the grain size being this big.
>>25027588yeah id maybe make it less contrasty and make the splotches a bit smaller. If you look at big granite chunks like that the variation is on a much smaller scale, a boulder would be more uniform in color
Hospital map cameras are being rendered for initial ingame walkthrough.Time to read the news and watch clips of local trash streamers using drugs rectally.
>>25026902better, but still wrong; lookup some references
th-thanks...
>>25002383thanks to everyone who played my games during demo daymade a lot of progress and fixes on song of the pearl since DD thanks to your feedback.https://vita-brevis.itch.io/dream-and-void>enemy spawner villages are now tied to dungeons. if you complete the dungeon, the spawners will spawn no more.>added multi level dungeons with persistence and progression mechanics.>added consumables system with addiction and tolerance mechanics.>add karma system which ties into the spiritual themes of the story.>added mount system with inventory integration.fixed various traversal bugs, inventory bugs, minimap bugs - dungeons now show on minimap as conquered or unconquered>a lot of work on the quest system. added quest trackers, quest items you can see and use, as well as readable books and scrolls.>added thematic music system which changes per dungeon. recorded a bunch of new music to go with it.>>25027494i look forward to playing this>>25026902kino. your progress and vision for your project are inspiring and i hope i get to play this soon
>>25008542>>25025918Here is my robot in-engine, animated.As you approach, it activates the robot and it shunts along a rail on the ceiling to the counter where it will be interactable as a shop-keeper that sells consumables.It has a few bubbly welcome and thank you animations I'm excited to implement. This robot (like the Tachikomas) can be so cute when they wave their little arms around.
>>25030675Phantasy Star Online Episode 1
>>25030704>PHOTON DROPS!>MESETA!>I'M GONNA CUM!
>>25002383I'm already working on a new project now.Experimenting with a mechanic in the new project.Also need to learn how to make 3D environments.Beautiful & realistic 3D environment.For the old game, I need to make a new trailer & some new screenshots for the upcoming steam next fest.
at certain angles, the flat shading is making it hard to read;maybe you could make each face a slightly different shade and not even use a light source
>>25002383been doing backend mission work and ui stuff, also added a tutorial.
>>25002383oh nice, /bant/ threads are still going. the /vg/ thread is mostly shitposting so i'll stick around here.Trying out multi-level wall placement with models+textures that should look more presentable. I wanted to stack and connect my wall segments in a flexible way to control the height of the dungeon structures rather than hand crafting each building on the map. The friendly deer is just a placeholder pay him no mind
>>25033624Oh, if this is talking about the buildings - yeah I can see the issue. Detail of the.building is being hidden because it's all folding into one colour. I might have to do what you've suggested to "fudge" it where the lighting isn't giving me exactly what I want.>>25034580Looking very consistent, I know what game this is despite not seeing this screen before.>>25034720Nice clean wall texture. Is it flat? It actually looks like it has depth on the vertical girdir type parts but I think that is just good texture work.
>>25034720Oh I'm fucking stupid it is 3D, zoomed into the top where it's more obvious.
I should stop hanging out in the /vg/ thread and just post here, there is nothing of value there anymore.
>>25035675Textures do a lot of the "popping". I discovered the joys of 3d modeling a few months ago and had all that excitement brutally stomped into a pulp the moment I opened the UV Editing nightmare tab in blender. Now I try my best to keep the model as basic as possible and hope that the textures carry most of the weight
>>25036069I have literally never not just placed a couple seams then just used UV smart project, take 2 minutes, idk why everyone makes shit harder for themselves then it has to be just because some other faggot on a forum told them "its the right way to do it"
>>25036125looking back the biggest time sink was learning to navigate blender's UI, or asking "what the heck is texel density and why do I need an addon to make sense of it"
>>25036125I always have troubles when it come to UV, it would be great if there's a tool to help out.How does smart UV works?Do you just press a button and it just works?Do you then draw on the 3D model, or do you draw on a flat texture?
>>25036556youtu.be/MkT4xIhoaLM?is=hlimkugMmlZSMZ2-The answer to life's mysterious are a mere google search away
Want to add a "barrier" layer to my player character meshes and animations. Means potentially republishing from the hacky-workarounds to the export-settings I decided I wanted to do long term. Some growing pains, very cool.
>>25039505I just wanted to socialize with people instead of isolating myself with technologies.Anyway, anybody creating their own levels with assets?Where can I find assets to use?I tried getting some free assets from Unity, but they always have that pink material (shader issue) problem.Where can I get asset that just works consistently without needing me to fix shits?
>>25040992
>>25041014I build everything from scratch basically, sorry.Sounds like the pink shader issue is because of either (a) the materials are not imported or possibly not compatible with Unity; or (b) the shaders are out-of-date with your version of Unity; or (c) your project is in the wrong Render pipeline, where you have Universal Render Pipeline (URP) for indie type stuff, cartoony and mobile games; and High Detail Render Pipeline (HDRP) for going more for that AAA photo-realism look. There may also be a default Render Pipeline that is separate to URP, depending on your version of Unity.I've run into mass pink shaders when I tried switching Unity version once and I ended up spending like two days on the issue before rolling back and deciding it wasn't worth the setback. That was in like my second year of using Unity and I got into the learned habit of just not upgrading for the sake of upgrading if you know how to make your version of today work for you.
>>25041061Yeah...I hate how easy it is for things to just break in Unity.That's why I want to find a way to get assets that don't break
>>25041014There's few ways to go about it.You can make the simple models yourself. As time goes by and you keep using blender, you will learn to make basic models even if you're a brainlet.You can download assets online. Just google that shit. If you're too dumb to google it, try sketchfab. The first issue you will face is that your overall assets quality and style don't match. The second issue is that they often come with licence bullshit. ie: commercial use not allowed or requires you to include them in the credits. It quickly becomes an asset management nightmare.The third option is to use AI. Again, use google. But something like Tripo3D is alright. You provide the software an image of what you want. It doesn't matter if you drew or just downloaded the image. And it will generate you a model based on the parameters you've given it. You can download the model using inspector and unpack the .GLB file.Animated 3d models require you to rig the models. This can be done manually in blender, where you assign piece of the model to a bone. Or you can use something like Mixamo. Mixamo is a web app which allows you to rig models. It's also a blender addon, which does the same thing. As an added benefit you can download animations from the Mixamo library which are compatible with the mixamo rig.
>>25041167Sketchfab(asset repository):http://sketchfab.com/Tripo3d(image-to-model AI service):http://tripo3d.ai/Mixamo(3d model rigging service and animation library):https://www.mixamo.com/Vheer(multiple AI tools, but primarily for modifying shitty concept images):https://vheer.com/Suno(AI music service):https://suno.com/
>>25041182Blender(3d modeling software):https://www.blender.org/Paint.net(Very basic image editing software. Can be used for concepts and blender materials.)https://paint.net/VS code(code editing software):https://code.visualstudio.com/Python(Programming language, contains Pip. Generic conversions such as unpacking models, automation in various softwares, bulk editing, etc):https://www.python.org/Audacity(Sound editing, recording, adding fx, etc)https://www.audacityteam.org/ChatGPT(Dumb questions, coding assist, algorithm assist, automation):https://chatgpt.com/
>>25041167>>25041182>>25041194Thanks for all the links. I'm already using most of them.But I need assets because I'm trying to use realistic style assets, and it's out of my ability to do so.I also don't have years to spend on learning how to do realistic modelling & texturing.Here's the screenshot of the result of spending some time to fix the retarded problem.It's a free asset from Unity here.https://assetstore.unity.com/packages/3d/vegetation/trees/coconut-palm-tree-pack-7888
>>25041254Why are you making a game which is out of your abilities, budget or skills?Realism in 3d comes from few things.Generally you would have a model with photo realistic but flat textures. Then you would have bump textures, these are textures which behave in specific way with the lighting effects. You could have variations of bump textures which actually make the 3d model more complex without making the model more complex. You would also have rigging which is used by the engine. For example, clothing or whatever may hang. There might be light sources on the model itself. And finally you would have a game engine which can support those functionalities.If you're making a model from scratch, it's a lot of work. For AAA-tier graphics you would need a team. AI can be exploited here in creative ways, but ultimately it still requires a lot of time and effort. But I think you're falling into the AAA-studio trap. A game isn't it's assets. It's the game logic and the gameplay loop.
>>25002293>>25002383New progress: realized Marvelous Designer is a trap, and remade my protag's model here for the vid bros: https://www.youtube.com/watch?v=0LgZvFu0d7E
>>25040992chad
>>25042325I realise the video is focused on the clothing, but I want to say instead: That's very good fight animations. In a sense one of those combo strings is what I wanted to convey with some of my unarmed/first attacks but my first two attempts have been jank. Looking forward to iterating on my animations to get them right.Is this Unity, Unreal, Godot? For my attacks I break them down into a staged series of steps with half-seconds between them such as:Enter states to disable other inputsAnimation beginsStart rotation to target (if applicable)Start slight position shifting (if applicable)Check for held button pressActivate hitboxActivate weapon trailDeactivate hitboxReactive weapon trailCheck for comboEnd states (if no combo)
>>25043073Thanks anon, I'm in Godot.What do you mean by half-seconds?For my case I divided the root motion animation from the skeleton's animation: root motion is read from the AnimationPlayer at blend 0.0 (so zero blending, happens instantaneously), while the skeleton animation is read from the AnimationTree at variable blending speeds (usually 0.2).This keeps the character looking smooth, but the actual movement in the world stays exact and snappy. Never blend root motion or you'll get very odd behavior.Also I'm posting from my phone, hence the different ID, but it is me.Eventually I'll do a proper animation tutorial so you might want to subscribe if you're interested
>>25002383hello agdgiesJust updated the demo with some cute fireflies, and the enemies are harder now, in theory.I should really really really start working on the boring stuff like settings page and finishing the story scripthttps://mangoware.itch.io/silbon-alpha-demo/devlog/1672730/fireflies-and-fixes
>>25043778Forget "half-seconds", more accurate of me to say milliseconds.Interesting idea of separating root motion from skeleton animation. I don't think I need to rely on that trick for my project but I'll keep that idea in mind. I haven't had problems with blending into animations containing root motion, but I'm also probably conservative with it - and no complicated jumps or vertical root motion involved.At the moment I'm trying to unfuck my main character rig / skeleton workflow. I have 40+ animations I don't want to remake and there is a lot of nuance when it comes to exporting from Blender and setting up in Unity where I work. My settings in the past were inaccurate, but "good enough" until now.
>>25046138Oh yeah the export stuff you want to get working properly ASAP.I recommend AutoRigPro. The rig itself I have some pros and cons with, compared to Rigify, but the export options blow Rigify out of the water completely. Rigify really doesn't like being taken outside of Blender, but ARP was made just for that and it shows
>>25027141Motor Spirit? It’s a menus game.
>>25048328are there any isometric games that aren't? I'd even consider CRPGs and diablo-likes menu games
lost all motivation for everything again, i realize i dont play games and i keep circling back to trying to remake some old version of minecraft or slender and thats kind of about it :((((( so i've just been playing tetris for the gameboy now
I implemented a multispawn system which allows traveling between maps and spawning in designated positions. This means that if there was puzzles which would require traveling between maps this would now be possible. Saving entity states remains to be done. I didn't have the patience to do that because I noticed that in multiplayer the remote player's walking animations don't loop. I tried fixing it and it was a lot of work but it does work now. Idle, walking and such are in locomotion layer, which mixes with other animations. It's still a bit hacky on idle and walking animations because they're called through a bunch of proxies in the update(), but it's better and the transitions got really smooth. I need to fix all animations though. I'm also making some other fixes like fixing the freezing attack distances and animating zombies turning on pathfinding.
>>25048328>>25026902Menu game connoisseurs rise up
also >>25029218>i look forward to playing thisthxu2
>>25049978I wish I could think of menu games cause that's what's demanded and sells well on steam, but alas I'm a parkour-brained consolebab. Everything I can think of is at the root about physically moving a dude through space like you're controlling an action figure, like a kid playing with toys. Menu games are too abstract to me, don't really get the appeal.
>>25047621I have something where you do Rigify and then a another third-party script to "fix" it, makes it match the bone settings for Unity in much the same way I imagine.I was really struggling to figure out the right combination of export settings in general. I tend to go for making sure my coordinate system coming OUT of Blender is accurate; and then for the IN part within Unity I need it to "Bake Axis Conversion" because otherwise it was this real guess-work where one day, my coordinates would be fine and then the next everything would be 90 degrees leant over. I'm about 10 months into this project now and I'm pretty sure its all settled now. The other big drama was exporting animations - originally it was "project wide" animations which was disgusting; so then I switched to using the NLA Strip in Blender but that kind of seems like I'm doing a kind of workaround rather than the intended use of the NLA strip; for example the NLA strip would have different usage if I was making a CGI film.
>want to be able to crouch+jump to fall through platforms>spend days iterating>rewritten code dozens of times>completely revamp collisions in the process>can't get them to work like I want them>decide to omit that mechanic altogether>everything works nowDevving in a nutshell.
>>25048821The pic I posted isn’t from Motor Spirit, it’s a draft for the next Hypercoven installment or whatever it will be… I think RTS are very much parkour games, at least when played with some vigor
Redid how movement works in an evening. It's a lot more readable playerside. As a consequence, I moved attacking to the skill menu and because shove was so esoteric to access earlier I moved it to a tap button.https://files.catbox.moe/azomox.mp4I also added more enemies and skills. From here, it's time to make the first area and intro sequence. Always start with the second level.
>>25051591anon that's such a basic thing though
>>25053918sickthe pacing of this video… you got a trailer all ready
>>25053918Absolute kino. Demo when?
>>25050096I hope you're right. I'm making a menu game now, and I have a really good idea for a second menu game I'll be making next.
>>25053918Very good, it has been a long time since i felt this jealous of an agdg game.
>>25053977>>25054159>>25055311thanks friendsDemo is gonna come after I get all the little pieces in place, gotta do a lot of art for the protagonist's starting squad and various skills and enemies, but there's a lot to love here already.really excited for how it's been coming together, check out the level up screen. Music isn't mine, and afterwards I just futz around with some more mobs and debug cheats:https://files.catbox.moe/yqnm9y.mp4
>>25051591Call me a retard but>no collision boolean with timer and let the gravity deal with it.And also a helper function to figure out if you're at the lowest level, which only purpose is to refuse service.I'm improving the freezing attack code right now. I'm nearly done with it, just dealing with overlapping actions. The freezing attack direction logic has been moved to the client(and needs to be fixed so animations match your position/rotation). The zombies now get to the correct fixed distance when attacking. Zombies have animations when turning, which should make them slightly more realistic. Zombies which kill you take joy in your death and start eating you, after which they deaggro and start roaming again.Although it does seem there's shit to be fixed with the pathfinding and physics engine. All the functionalities are MVP, which is to say they're shit. The sliding functionality is buggy, which allows players to slide backwards when hitting edges of the navigation mesh. Pathfinding has some weird ass issues with certain parts of the navigation mesh, which causes the zombies to oscillate and turn rapidly. I would love to do some actual game development of functionalities which bring value to the gameplay, but this is the path I chose.
>>25056702is this inspired by beetle adventure racing?fucking loved that game as a kidwhat's the gameplay like?car combat?
>>25056889Never heard of beetle adventure racing, but looks neat. Car battle minigames and car race battle game.
>>25053929I'm retarded. That's my only defense.>>25056618>no collision boolean with timer and let the gravity deal with it.Tried that. Even reduced the effective collision mask for the platforms down to a single pixel in addition to having a whole separate routine specifically for the literal bottom row of pixels to be the only thing that matters, told the engine to disable ALL collisions for like 6 frames (literally a tenth of a second) before re-enabling them...End result the same. If there's a platform under the platform you're falling down from, you'll pass right through it. No fucking clue what I'm doing wrong but it's probably equal parts lack of skill and overthinking the problem.
>>25057107Most physics engines operate with simplified models of those seen in the game. The simplest form is a bounding box: a box which should contain the model in 3d. The most complex form is just that exact model but simplified version of it. Or a combination of simple shape models which imitate a more complex shape. To make things faster the bounding box is applied before going in for the more detailed physics testing. If the bounding box test catches no collisions just return, otherwise keep testing in more detail. Either way. Disabling the collision should be simple. Either you use an existing no collision mode, or you make the exception for it.You know, along these lines:// Collision detection occurs hereif(!object.noCollission)...But you got to remember that this has to catch the detailed testing as well. This is probably where you fucked up.Alternatively you can do this:Physics engines go through the physics models as an array. It's just a list which is processed through with a for-loop for every frame. So. Remove it from the array. Alter it's position to where you want it. Add back to the array.Here's another alternative.All physics engines keep positions somewhere. So fuck it. Raycast below your level to figure out the next valid position, and mathematically force the physics model through.If you want me to help you more, you got to tell me more about the technologies you use.
>>25057402This stuff which I just said is a bad example of how to do things. This is how I do stuff, but don't follow my example because I'm retarded. If you have a real physics engine and not some crap built in a shed, you should have a noclip mode somewhere. It just might be under a different name. If you use VSCode just do CTRL+Shift+F and search possible keywords for it.
>>25057438
>>25002383Finally working on the enemy AI. I wasn't going to use a FSM for regular enemies, but I know I'll need it later on for more complex enemies or bosses, so I might as well set everything up now.
>>25058741This is actually kind of neat. Has there been a game like Fire Emblem that is partially real time combat?
>>25060050It's just mega man Battle Network
They took me out of tileset jail and put me into stairs jailIt’s much more grim in here>>25028336>>25027435
>>25060658Well, again I don't have an artists eye but I think it looks pretty cool. And I think it's better than the previous ones.
>>25060050>Fire Emblem maps/theme but MM Battle Network combat/movementpotentially amazing concept. Might be too overwhelming though, but that could probably be mitigated by making enemy movement slower or having a pause button.
Improving my map editor. Can see what's placed on any selected tile now, set which tiles are walkable rather than relying on placing floor textures, added a bunch of decorative assets so my buildings look like actual buildings now. All using babby's first art project textures & 3d models some of which look like complete ass but I'll try to get better.This was meant to be an Etrian dungeon crawler but making nice organic stuff is hard so it accidentally took a hard right turn headed towards SMT1 which is fine too
>>25060658reminds me of the Fallout 1/2
>>25056618Freezing attack now forced fear animation on the player on a different layer and upon approaching will play the biting animation at correct angle. Player will now bleed upon being hurt. Fear animation fades slowly on its own layer for 24 seconds. Zombies are now faster, animations have been updated to be more aggressive, players have been slowed down. Zombie pathfinding has been fixed, nav mesh reader no longer reads some border cases of triangles as neighbors. Funneling and string pulling have been fixed. Pathfinding has been mostly fixed, so that zombies create an additional point before turning towards new target.Began working on the worm monster. The engine doesn't support worm or snake models and animating them would be shitty, so instead I'll just create a bunch of client side entities(models) following each other which form the worm. They register the previous position and pass it down to the next node, which will then do a lookAt towards the node ahead. This should result in a smooth action.Or at least I think that's the right way to approach this.
>>25060658now it looks right
>>25064965Given the opportunity...Would you?
>>25066202Yes, I would high-five the little hand on the end.
>>25065966This looks good with a lot of character. It might be a little busy, a lot is going on, I assume this is an important NPC with the note on his chest. If it is a regular enemy maybe decide if a chest-pinned piece of paper AND an exposed heart is more important as they see m to be features competing for attention. If this is a main character for the player, which I assume it isn't, the eyes should be visible I suggest.
>>25065966very nice>>25056702neat
Are these fancy railings too fancy? Worried it's getting away from the chunky texel look.
>>25067107Whenever I have to ask myself if my models are getting too complex I always refer to quake maps to keep me in line. My personal view is good textures > good lighting > detailed meshes
Just noticed a pattern. If you use an art style that's popular and well-liked, but not commonly done, it seems to be a huge advantage. Mouse P.I. for Hire did old cartoons, and Roboquest did Borderlands. Bomb Rush Cyberfunk did Jet Set Radio. I think the gaming market will generally punish you for trying to innovate your own unique style, and most of the indies who -have- given us relatively fresh styles took like 10+ years of experience to get to that point. Which actually makes me wonder, do you guys think the unique style of Lethal Company played much of a role in its success? Or is it more a case of, Zeekers had just spent the effort and time to build up a fanbase, and the style might have helped it out a little but ultimately it was more of a snowball effect type deal>>25060658This is dope
>>25060658Looks cool as shit, good work nonny
Now that you mention it, somewhat. Maybe because they are rounder in a world that is very square. Maybe don't stress about it also and just keep going with further elements in your world, come back later after time away.
Just a thought but maybe the OP should have a link to /gedg/ as well. It only makes sense ...
Scrolls are gone, an improvement imo
>>25067408It takes a specific kind of mind to really put all those disparate pieces together in a cohesive way that speaks to people I think. It's all so abstracted, and you're trying to take this totally abstract concept of the style, and give it form in a way that has a visual language that "makes sense". It's like speaking without words almost. It's the fabric of the visual information being delivered.Probably a combination of personal taste, experience in the field, and a vision and understanding of what you're trying to deliver with that vision.Neat thing to consider.
>make cave walls>look shit>ignore them and make the rest of the level>go back and they look fine>>2506740899% of solo devs would fail to make LC because the scope would be too big (at least pre AI). Zeekerss was ruthless in only including the assets absolutely necessary. The filter was a smart way to not make it look Asset Flip #99527349 - it doesn't sell the game itself, but it stops people tuning out long enough for the gameplay to hook them.
i think i finally have a prototype i don't have to be extremely embarrassed of i like nonograms, touhou and anime girls, and so i made this. the plan is to have a game that will quickly become difficult(100x+ sizes). right now i can play the entire loop (load, choose, play, win, load next etc). board state is automatically saved into json on every click, there's undo, dragging, you can click on clues to fade them (no autofade yet), zoom and move the board around, row/column highlighting, coordinate label.backgrounds are placeholder aislop because i can't afford an artist i'd want (kurehito misaki lolololol), so until i get someone it'll have to do.
when you complete the board, there's a menu with some planned image and flavour text, to have some basic "story" for each stage.now the biggest problem around this project is that eventually, i'll have to go and somehow ask ZUN for permission to sell this(hopefully), and i have no idea how to do that. so the entire project depends on one japanese man giving a thumbs up.
and here is the custom editor to edit the nono array. that's pretty much it.
>>25072308if he says no it'd probably be easy to replace it with generic anime
>>25072372yeah that's the plan:1. 2hu2. cute anime girls / anime references nonos3. some cthulu/horror stuff, cause almost every nono game i know is this relaxing cutesy stuff, which is perfectly fine, but i feel there's a lack of other settings
>>25071846looks nicethat's why I focus on the most important things to really sell the scene first and work in reverse order from there. Otherwise, you can get lost in the weeds fixing every rock and wall.
>loved the Tunnels books as a kid>want to set my game in an underground world different generations have discovered>make picrel, old Edwardian-era infrastructure that a group of modern explorers have visited>worried it'll come across as Superliminal Stanley Parable tier 2deep4u metawank
Hello friends. Today I went to an indie game festival in my city. In short, it was pretty grim.In terms of visuals most of the games actually looked pretty good. Definitely a lot of amateur art but still more than passable. There was probably only 1 or 2 genuinely ugly games.However I played close to 20 demos and there was not a single game I could ever see myself playing again, let alone ever pay money for. Pretty much every game could be put into one of two categories:>this game is literally just a reskin of an existing game, doesn't try any new ideas or hooks, and has 1/10th the content as its inspirationor>I have no idea how this game got this far along in development without anyone ever telling the devs "why would anybody want to play this?"This whole experience got me thinking about how much online discourse I see about the "indie game apocalypse". I really don't think things are that bad, there's just way more shit games than there used to be 10 years ago. If this festival had a SINGLE good game it would have been extremely memorable for everyone in attendance. Literally any of the top 10 demos from last demo day could have set up a booth at this event and would have been treated like Bethesda at E3.The point of this post being as long as you can cut through all the garbage and get eyes on your game, I think you'll be alright. Maybe not pulling in hundreds of thousands of wishlists and millions of dollars alright, but you should at least be able to pull in a decent audience. Thanks for reading my blog. I hope you're all doing well, and good luck with your games.
>>25002293>Haven't finished the story script for my story driven game
>>25072672it’s wonderful, keep going
>>25072953>Literally any of the top 10 demos from last demo dayWas there a top 10?
>>25072953Sounds whitepilling but you have to remember it's an indie festival in a city. I think you'd find better games at international events, like that recent Tokyo Game Show that even had some anon from aggy participating.I do not expect to have even 10 good indie games come out of my entire country over the course of a decade, let alone a single city.
>>25072953>>25074487yeah if there was an indie meetup in my city it'd probably look similar. you can't expect to have 10-20 great quality games in one city, that'd be like what, 500 per country?also a good chunk of indie games are just autismo "i wanna play this" projects. just look at aurora
>>25072953That's similar to how I felt seeing AGDG. Like oh, wow, there's like, random genius that pops up and disappears and mostly it's garbage. It gives me a lot of confidence. I mean don't get me wrong, there's at least probably 5 devs in the history of AGDG who I respect. But I mean there's thousands of people trying to make games for demo day, 10's or 100's of entries. It really seems like if you put out an actual real game you kinda have a shot to stand out. Like you can be in the top 1% of demos by just having a real game. I guess I'll find out.Progress - Trying to actually play my game all the way through. So far I found like 10 bugs that need to be fixed. So that's what I'll be doing today.Actually this bant agdg seems to condense the good devs. The quality in this thread mogs the other AGDG thread.
>>25002293>>25002383: Hirdrih Technologic ::progress :: Next chapter is about Untotern (Much like with Nortubel, working on this world during Halloween eve).
>>25075116>The quality in this thread mogs the other AGDG threadthe /vg/ general has been a mentally ill third worlder spamming 90% of the replies in every new thread for at least 2 years. I'm not even sure if he knows he's the only one there. To be honest, I'm not even sure if it's a real human or a bot given the rate at which he posts.
>>25072308You don't have to ask. He's already said yes: https://touhou-project.news/guidelines_en/
nigger
>>25076176yeah i already asked in /jp/ and another anon said the same thing. one one hand it's a relief, on the other i'm a bit paranoid honestly. i really like touhou and i don't want to be a menace to zun-dono.did a bunch of dumb bugs and rearranged the win screen, for the image to be more like a reward, and less like a distraction. the gradient on the right glitches out because it's a hackjob and i fixed it right after recording
>>25002383progress:>split NPC's audio and visual perception into their own classes instead of lumping all perception types together>made ambient sound zones, so in case there is a really loud ambient sound like machinery, NPCs might not react to some distraction/combat sounds>NPCs can finally perceive stuff thats above them, it used to be that their perception was only the height of their collider>Added triggers for progression style events, like going into a specific room might trigger the target NPC and his guards to walk into the next room and start a meeting or something working on doing the meetings/dialogue events between NPCs. Getting filtered by interrupting the conversation and the NPC who investigates walking back and saying something like "...where were we?" and then continuing the dialogue.
>>25076913add innocent pedestrians like in carmageddon
>>25077009Good idea, I'll think on that one for a bit
>>25077118OMG SOUL
>>25075116>It gives me a lot of confidence. I mean don't get me wrong, there's at least probably 5 devs in the history of AGDG who I respect.Be very careful with your hubris. I thought I was heads and shoulders above the other devs too. Ultimately, the market will decide. Be patient, make many games, learn from your mistakes. Don't think, just do and keep doing.
it's an animation
>>25076913>>25077118looking good, are you the same anon whose cars were green colored blocks/programmer art not too long ago?
FYGOON!!!are you here?its mei am here to talk with you, in the future.if not, can someone tell me where fygoon is? has he left? was he never real?
>>25078781fygoon probably never left /vg/
>>25077834>>25078667Thanks anons. I am not that anon, >>25056702 is the first post actually for this project
>>25078667this one? that was me
>>25077864Well, there's modern AGDG (so excluding games like Your only move is Hustle, Tower of K) that I'm competing with on one level, and I think I can make one of the better games here. But then of course there is Steam, and that's a whole different thing. For my first game I really have no clue if it'll sell anything at all. It's a bit niche. But everyone seems to say I'm in a good genre for Steam.As far as progress this weekend I fixed a ton of bugs. Started implementing a pop up system. Super easy to do, just something I put off til now. Still have a ton of bugs to fix. At this point I'm playing my game and fixing everything that comes up, but most of the gameplay is there. I just need to verify it's actually all working. Which it isn't, but it's not too far off. Like most bugs only took me 5 min to diagnose and fix.
>>25081339One universal trait I've noticed from successful devs is a deeply relaxed attitude. They're fascinated by the process of dev itself, so the monetary incentive is not entirely needed, and even if steam were to shut down tomorrow they might still open up their engine each day and tinker around for fun. This is true now, and it was true back in the 80's when Carmack and Romero were obsessing over their Apple IIs and getting their asses beat with a belt for not doing enough homework. I personally don't have that level of interest in game dev, and I'm convinced that it's nigh-on impossible to be highly successful in game dev until you develop that level of interest.But I also think it's possible for anyone to develop that level of interest. You just need to be deliberate, and connect it back towards something you already care about. I've already successfully done this with language learning. You might not care about learning Korean or German in isolation, but once you start dating a Korean or a German girl then you have a natural and strong motivation to learn these languages. ...I bring this up because I don't believe game dev is a market where "Adderall + 40 hour work weeks" is viable anymore. Even if you make a solidly good game, there are just too many arbitrary factors at play. Games have to be something more than a job or a routine to you, they have to be a fascination, something you can truly get lost in for hours and hours. I believe that's why my first project failed
>>25082015Right, I'm not saying game dev is easy. The opposite actually, I think maybe 5% of Software Engineers could do it. Not because it's technically difficult, I just think most people can't really think. They learn by a sort of osmosis, whatever is shoved down their throat they learn, but they can't really move on their own. So ya, the games I work on are interests of mine, are in genres that exist but are totally unique and new in some way. They also try to hone in on where the fun is. But again, if my first game flops, I don't need people quoting me calling me an idiot. The first game is to teach me the entire process, a - z. The second game I'll try to make a truly marketable game.
>>25082428>I just think most people can't really think. They learn by a sort of osmosis, whatever is shoved down their throat they learn, but they can't really move on their own.but enough about corporate office drones
After almost a month of vibecoding (going from never written proper code in my life) I now have an almost complete gameplay foundation - a weapon system, an inventory system and all the necessary consumables + an almost complete magic powers system (still missing a dedicated UI for it, XP & upgrades and some passive powers like generic health updates, but the foundations work) and it's all working surprisingly well.I guess it helps that I've worked on/with games before and already had a rough understanding of how things (should) work under the hood, but it's still impressive how fast I managed to go from absolute zero to actually something. I thought this would've taken me way longer.
non-coder here, I can't decide on an engine at all to start on. is there one that's specifically good for learning the ropes and doing more as I get better with it
>>25085283unity's probably a good choice for you
>>25002383proggyI've been studying different projections, I'm trying the isometric now but it's a little more tricky than the other ones
>>25066202On a second thought this isn't going to work out.I did some thinking, trying to figure out how to complete this project in a satisfactory manner. Satisfactory manner is the key word here, and it's currently not that.This zombie theme isn't going to work out. It doesn't fit the original story plan and it's a bit dull. Like I've said in the past, it's all MVP so far. I've been recently watching some medieval torture documentaries, some horror gameplay marathons and I figured I could draw some inspiration from them. I'll keep the functionality the same: NPCs have two modes of attack, a freezing attack which is more complex set of animations, a drive-by attack which is just for slapping when players get too near but not near enough for a freezing attack, and they've got death enjoying sequence for the horror of other players. I figured if I could make one more functionality to the base NPC then i'd be all set for most NPCs. That would be ranged AoE, imagine a zombie spitting acid which denies an area. I figured I'd also need to add snake physics for bones in specific use cases.The NPCs behavior could be modified with variables in the main class but the base logic would remain the same.The torture methods which inspire enemies i figured would be:>flaying>broken on the wheel>the rack>hanging>burning aliveThe maps are extremely short. So to make the gameplay longer I'll redesign the town map, and this time I won't use AI generated models because they're very CPU intensive and at times illogical, which hampers the mapping process.I also need to create functionality on the server for saving entity states between maps and restoring entity states. This is a high priority task.
>>25088852I also did some thinking on development methodology and gameplay.As of right now when players die they remain dead. It doesn't really follow the spirit of co-op games, so I'll make a small change so that players respawn when the map changes. If everyone dies then game over and everything resets.Currently the puzzles and the keys operate on assumption that cameras have boolean images. "Anything" can be a puzzle with either a key(item object) or numpad(you get the password by reading shit you find on map), and those open doors. But the puzzle requires fucking around in blender to produce both TRUE and FALSE images for the said door. So maybe I should just create a 3d model which is sent to the client. Maybe just a model with baked in animation of open and closed states. I'll still keep the fading between opening a door, but I can just switch to the next keyframe to display the door object as open. It's going to take some thinking. Maybe some trial and error to see what works out better.But what's clear is that the current maps won't do.Thankfully I've learned to make more complex models by now. So I can actually produce maps without relying much on image-to-model AI services.
>>25088891>Maybe just a model with baked in animation of open and closed states.for a door opening? can't you just move the pivot and rotate the transform on an easing curve?
If your game was alive would it help you finish it? Because these plants constantly need water, I don't delay when they need it, but because my game has a perfected robot body and I don't see it languishing, just waiting, I don't feel any motivation to change things about it.
>>25089000Doors in this context refer to any objects blocking paths which can be disabled either with items or passwords. The ambulance blocking the path to hospital is a perfect example. It's just an invisible blocking object and background attached to a camera, which changes depending on status of the invisible blocking object.But I'm quite unsure about this. Gosh, I wish there was a practical way to automate taking boolean images... I already have scripts which render all the images on the map, but with this boolean background shit I have to go through each puzzle by hand and they only have 2 states which make them too simple.I am unsure would 3d puzzles streamline the process or add value. It feels like it could, but I don't want to commit to a process which adds complexity and no value. I'm trying to brainstorm ways to add value without adding complexity.Here's a background for ambulance door in false state.
>>25090229Here's the same camera view with ambulance door in true state.See how this works?You unlock the door, path blocking object is disabled, camera fades and switches background image.
today's progress:1. clicking on cells in an editor is a pain so i made a "png importer" which turns pixels from a png into a nonogram. now i can make them in aseprite which is much more comfortable.2. i realized that making nonos is actually pretty difficult if you want them to be cool, look good and be interesting and be consistent with what you want to tell in a stage, and i managed to make like 4 simple ones. wew.>>25090229yeah i'm not entirely sure what your problem is.>boolean imagesthe what now>they only have 2 states which make them too simple.so like for example, you want to have a progress blocker, but you want it to have more ways of solving it, like "unlock the door", "sledgehammer the door", "shoot the door lock" or something? seems like you have to make them have states then
>>25090266Technical limitation problems need to be addressed before designing a proper map and proper puzzle flows for it. You can't make a map for a system which doesn't exist. Right now this exists. But with the effort required to use and value gained per puzzle, the question arises if it would be more value to create a 3d entity with animations or whatever.
>>25002383Working on powerups, will hopefully have this one ready for DD
is there a name for this style or does it come from some particular tool? seems somewhat common in indie games so i'm wondering if its an actual stylistic choice or if its because some tool does it this way (or both)see also: SNKRXtalking about the simple sprites and jagged edges (often with the black outline)
>>25093093It's just pixel art, just with a simple/toon lean probably so they can pump more out faster. Pixel art was very popular before 3D became widely adopted for games and continued to see regular use for as far as the PSP/3DS, but it's fallen out of favor with normies because jaggies ruin their immersion or some shit. Aseprite is currently the most popular tool for making and animation pixel art from scratch, and a tool like Pixelover can be used to create simulated pixel art out of 3D models or non-pixel drawings, and you can even make pixel shaders for 3D in-engine if you tried hard enough, similar to Metal Bringer.
progress:>hit backspace on all of my NPC awareness/perception>had a problem where they would not forget about audio sources, and would only respond to sounds despite seeing the player >if an audio source started to play and then stopped, the NPC awareness would continue to "build" up until the audio source's score was high enough to react:(working on re-implementing my backups from when things were working. fuck lol.>>25080326>>25080027ohh. cool, that's two games i wanna pay attention to. love driving games, its one of my fav genres.
>>25093146Pixel art is making a comeback honestly, alongside chiptune.
>>25090435
>>25088852
fully voxel planets in space, caves going all the way to the core. vibe coded. graphical upgrades pending
landing and walking on the surface of voxel planet
>>25095233>>25095235Damn nice anon, my no mans sky slop clone has already been outslopped. Are all the grass, planets, and trees vibecoded too? Clankers do a better job than I expected
>>25096855yea they are. i saw at some point the agent was fetching certain free assets online, not sure what, so who knows which is an asset and which is procedural. i literally did nothing by hand i just prompted the ai repeatedly, mostly the same prompt over and over using goal mode. GPT Astra Medium
>>25096855two instructions i think are really helping the agent is "must look AAA quality" and "must run at least 60fps", it makes the AI often view screenshots and judge if it is at "AAA quality" or not yet. which clearly it isn't, but i don't think it would have even gotten this far without that instruction but who knows
>>25096855what's your made in/with? is it multiplayer?
progress - 10 nonograms total for normal stage 1. stage theme is "fairy guessing game". a whole day to make 10 puzzles.>>25095235>we have no man's sky at homehey looks pretty nice, reminds me of planet crafter, what a terribly optimized and designed game>>25096855is that minecraft in space? also your clouds look broken.
>>25097265I'll have to try that out. Damn.. AI speeds things up a lot but its still increasing my idea backlog due to having even more things I want to mess around with lmao>>25097300Just single player, and pure Opus 5.5 (medium). >>25097372More No Man's Sky + Far Cry/MGSV so far. Minecraft aesthetics to keep it simple. Though it does mean you can do stuff like tunnel under enemy bases to fuck with them. And ya, the clouds are on the list of things to fix, just been low priority>a whole day to make 10 puzzles.Nice. Good puzzles take time anon
>>25097372You're making the anime 2d puzzle game, right?Have you considered making an algorithm for generating puzzles? Could be good, even if it's used by the developer instead of the game itself.I'm trying to finish the flayed man model and resorting to drinking again. Such is the unemployed lifestyle.
>>25097389>More No Man's Sky + Far Cry/MGSVdo what. certainly not a combination i thought about, but then i feel like "assault rifle combat minecraft" could be a great game>>25097430yeah. what do you mean "an algorithm for generating puzzles". from scratch, or based on a picture or what? also don't drink, it's bad for you.
>>25097430Also I saw some ps1 silent hill gameplay and kinda liked the dark rooms with flashlight. The fear of being lost and getting cues of imminent rape felt like a great feature. So I experimented with making lighting effects for the 2d-3d. I kinda got it working by applying 2 3d sprite planes on the player and highlighting the 2d background pixels behind them. I figured I could just blur them afterwards and that would be job well done. The 3d light would be just three.js spotlight. But to further make it work, it would require a second scene for depth images and such. I figured it wasn't the time for such exploration and time would be better spent making the new enemy models.
>>25097459If I recall correctly, your game is a variant of sudoku with anime backgrounds. Or minesweeper. I'm not sure because I didn't catch the idea from the webm.Either way. If that's you, then you can just make algorithm for generating puzzles. You can still add the backgrounds, musics, etc, compose each level. But for the gameplay itself you can make a program which makes the puzzles and hence streamline production.Also I've been planning on drinking for a while. I've not drank for a month and it's my neetbux day so I got myself dark rum, coke and lemon juice. A poor man's cuba libre setup. I don't plan on getting assblasted drunk, just getting cozy and enjoying my personal sauna while I still have it. I've got a rule that I don't drink alone but it's one of those days you know. I've got two drinking buddies in hospital and one's asking me out to a bar, but I really don't want to go. I'm on a tight budget because I'm moving next month.
I am like five years into working with Unity and today I learned about Material Property Blocks.Basically, say you have ten lamps in a scene but want some flickering. I would make one shader and then have a script that turns the shader for some to have their property _FlickerFrequency set to 0.5f or something. This "Instances" the material, which is inefficient and I put up with it. The alternative is copying the material and having a Flicker version and reusing that material.If you want that value to animate and lerp as you bring a source of interference, you HAD to "Instance" it, I thought. I even went to StackOverflow once and they were huge cunts about it. People would be like "Ugh, instead of answering this question ask yourself why you would even want that."Wanting to make a new shader effect, I ran an idea past one of these AI to ask if what I wanted to do was feasible via script or if I should make a Vertex Shader for it and in its explanation example it just dropped this mention of MaterialPropertyBlock and now I realise there has been a more efficient way of updating materials for basically no extra effort the entire time - I just had never seen it mentioned in any tutorials or documentation before. It is possible the feature only started existing on the version above mine but I'll double check that.But anyway, all MaterialPropertyBlock is, is you say GetMaterialPropertyBlock on the MeshRenderer or SkinnedMeshRenderer, update it instead of the material directly, then SetMaterialPropertyBlock and I'm fucking spitting.
>>25097518Their game is Picross or Monograms. Those are extremely easy to create and don't require an algorithm unless you want random puzzles, but most puzzles in this genre want a little creativity, drawing a small image so when the puzzle is solved you get a small pixel art.The two best Nonogram games I've played? Pokemon Picross on 3DS was surprisingly the best. Each puzzle was a new Pokemon you "Catch" and each one had a power you could use to solve future puzzles. Legendary Mew puzzles for example only showed up periodically and you'd bring your Mewtwo or whatever to try and solve two random columns right off the jump to try and get it done within the time limit.Dungeons and Diagrams was a Zachtronics one, which just had great theming. It was Picross of DUNGEONS which had rules for where enemies and treasure could possibly be, such as "Treasure is always in a 9 tile room." And "Rooms only ever have one door."There is an indie game called CiniCross which I wanted to like, it was Slay The Spire with Picross puzzles but it's good at low sizes, but their difficulty was fucked and every comment section agreed. I have feedback to the developer about the issue and a simple fix but they didn't heed it, do only sickos can really play it. Their issue is that their Easy, Medium Hard was laid out like a 5x5, 10x10 and 15x15.Guess what, 5x5 is 30 seconds, 10x10 is like 2 minutes, 15x15 is like 10 minutes!! It's exponential!!! But the game treats it like it's all supposed to be 3 minutes or something, so it was like a hard stop any time you get to a "Hard" puzzle.I said he should have just ramped gradually between 5x5, 6x6, 7x7 etc but I think by this person's state of Nonograms they lost complete sight of what normal people are capable of and built a bubble of sickos for their game.
>>25097637I'm retarded and prone to laziness when it comes to repeated work which increases value. I would just say fuck it after the first 30 levels and make a python program for converting a folder full of pictures into puzzles and adding all the assets/settings for a level. I also see it as a modding thing, that players get access to tools necessary for making their own content.
>>25097518>If I recall correctly, your game is a variant of sudoku with anime backgrounds.a nonogram, specifically. it's basically pixel art that you fill using clues. so i'm not entirely sure how could i make a generator, since you have to have an actual picture that makes sensei mean you theoretically could generate those randomly, but they'd be random and not a picture of something>>25097637>each one had a power you could use to solve future puzzles.that's a pretty cool mechanic
>>25097723You can make a program which increases brightness and contrast. Or just pick the blackest pixels for the puzzle image output.As for hints, you could apply LLM as long as you give it context. It probably knows more trivia about the context than you would like to admit spending time on figuring out.I've been working on some LLM and neuron network experiments in python, small side projects which I can't allocate more time right now... In terms of using existing LLMs I would apply Llama interface to use it's models.I don't know what fits your case. I'm QA & automation guy by profession, so I think like this.
>>25097760>As for hints, you could apply LLM as long as you give it context.no, anon. the hints are the numbers on the sides of the board, they're autogenerated. not like actual "hey do this" hints. >I don't know what fits your case.no problem, every bit helps, outside perspective is always good.
>>25097777That sounds simple. So basically your puzzle is just an image which needs to be revealed. User input is similar to battle ships, am I correct?Would you need help making a program for generating images into puzzles?
>Equipping a Tread (Or Landboost) system to your Assault Frame lets you move quicker in the field while also getting a move range and response (evasion/interrupt) bonus in battle. It's good to have but on models with limited Option slots like the Gouge here it's a bit of a sacrifice.>This especially considering activating your Tread system itself costs a full turn.
Finally got something working for the procedural sky/moon system in my game. Every planet has a random number of moons of different colors and sizes that gradually cycle through each phase over the course of several in game days, which determines how much light is cast from the sky. The sky texture changes based on the weather and time of day so stars are visible on clear nights but only bright moonlight can be seen through clouds.
>>25097804>That sounds simple.yeah, nonograms are conceptually extremely simple. you can try them out - https://www.nonograms.org/nonogramsas for the program, i couldn't tell you right now what would it would have to be able to do, but thanks. maybe i'll come up with something in the future.>>25097813>shinji gets in the robot>hands on joysticks appear in uiKINO
>>25097565material property block has been around for a while, at least since version 2017but yeah it's a cool system, only thing to keep in mind is that it will break SRP Batching if you use that. in that case, it's actually better to use a material instance since the SRP Batcher draw calls works based on shader variant, not material instances, and changing a single float on the instanced material isn't likely to require a unique variant.
>>25097813this is cool, I haven't seen immersive hud elements in a (I'm assuming) trad roguelike before
They're making it illegal to development amateur games in EU.https://www.youtube.com/watch?v=mQRcK9nYg9g
>>25101091EU is a steaming pile of shit that produces mostly corruption and regulation. Member countries engage in full blown beurocratic warfare against each other while leadership tries to become too big to fail while being dragged down by member countries' debt. The heads have confirmed their lack of accountability with court sentences that carry no penalty whatsoever, not even a dismissal so it's full corruption now.
>>25101091That's pretty wild. Not allowing native devs to make online games. Pretty wild.
>>25101091I told SKG this was the exact outcome for inviting the government in, they just called me shills then but my prediction is dead on
Drafting a very simple UI framework, glorious immediate mode and fully known element sizes.>>25101485Collectivists, will they ever learn?
>>25002383/vg/ thread is drowning in shitposts again. I regret not posting proggy here, but honestly, it's been a while since I posted anything anywhere public. Staying active on social media just feels disgusting for reasons I can't articulate.Anyways, I'm getting ready to submit my steam page for review, for what it's worth. Already crapped out some of the most amateur-looking capsule art known to man, spent today editing footage for a trailer to put on the page. Does it look interesting enough? I know I don't have a lot to work with here, but I want to give my game the best chance I can.https://www.youtube.com/watch?v=a4YUM2TUWD8
>>25102819you know how trailers in vidya and movies etc do this thing where they show something on screen, and then they show white text on black screen, and then something on screen, and then the text, and so on? the stereotypical "THIS SUMMER" stuff.that's because when you have a lot of going on the screen(like in your case), people miss the text because they either have to focus on reading, or focus on what's going on. so just do the back and forth with the text or maybe some other method where the viewer doesn't have to focus on multiple things.there's some sort of trumpet sound bee-beep a few times, it's jarring.don't really see any other thing i could be immediately critical of
>>25102819Alright. Well, it does have the bells and whistles.But it doesn't seem like the kind of thing I WOULD personally pay for.When I play a game I think of does this immerse me into this world of fantasy enough for me to not bother with thinking of real life issues. Do I get immersed enough to not even acknowledge their existence. And continuity: if the experience continues as long as I desire it to continue. When those fields are filled, I buy the game.Now get this straight: I'm not telling you you can't make money off this. I'm just giving you an opinion. The reasoning behind my decision is that I don't find the gameplay exciting enough. Not even with the bells and whistles. You may ask why, and I answer that I don't want to play asteroids with fx.
>>25101091[cries in EU "associate membership"]I hate the eurocrats so much
>>25105378fuck, forgot that shit would probably also affect us
>>25103472You're right desu, even if it's just two sentences during the least intense part of the video. A ton of events in my game expect the player to read text while things happen on screen too, but I already knew I had to rework those. In the end, my biggest worry about this trailer is it not being ADHD enough from the get go, i.e. not enough things happening on screen to capture a goymer's attention, and padding the run time with text-only screens would accomplish exactly that. Heavily considering scrapping the slower part with text entirely. >>25103472Fair enough, I'm not exactly expecting a commercial success here anyway. I'll be happy if I make the steam fee back, but I still want to give it my best shot. I value immersion in games above everything else myself, but I don't think I could accomplish it in a 2D space shooter, not with my skillset.
>>25002383Testing rendering and reading depth textures for a project involving terrain deformation, though the specific approach I'll have to throw away
>2mb size limitThats too small man. OBS was breaking so vibecoded a video recorder that lets you cut out parts of the recording video before it save it to a webm and it auto adjusts the file size to whatever limit you give it.Anyone play the Warhawk (PS3)? Was fun times. I'm going to have to figure out how to balance being able to use your ship for ground combat though
progress: easy stage 2 nonos, theme is " Best Scoops of Bunbunmaru Newspaper">>25108558>vibecoded a video recorderanon just WebMConverter that shit, also>webmthat looks pretty fucking funny honestly
>>25108640>progress: easy stage 2 nonos, theme is " Best Scoops of Bunbunmaru Newspaper"Nice!>just WebMConverter that shitIts more than just a converter. It records it, lets you after cut out parts of the video, and finally saves it as a webm of the size you need. Maybe its overkill but I never used a video editor before and this is very simple to use, and took no time to make.
These bots are not playing around
Enemy chase behaviour is working well, but my ffmpeg script has somehow broken and started making bloated blurry webms again so no clip
Basic ass animation that took a long time to get into decent shape.>>25102819Around the 22 seconds mark the framerate appears to drop. Might just be an illusion. Not sure if the average Joe notices or cares.
>>25026072>>25008542>>25011237>>25064965>>25075508>>25113098You can generate that in 30 seconds with AI.
>>25113917but they made it themselves and I think that's really neat
>>25114206Don't answer, that is how /agdg/ fell to the pajeets hands.
>>25002383Doing some new tilesets.Story script is half done so I'm moving forward with actual game progress
>>25112664Shown it to a friend last night and he said the same thing about FPS drops. Fuck my life, I guess I'll have to learn multithreading.
>>25088314more proggyI've been doing some experiments with textures and I love doing themthe coppery tone wasn't really planed, it was more like a happy accident>>25097430>>25113098very cool!
>>25113917Well, anon. There's tasks which you can do with AI and there's tasks which don't benefit from it. It seems like AI can reach a baseline but lacks that something.Say, body horror. It's illogical at times, which means you would have to give the AI multiple reference points in your concept art. Most only accept one image and fill in the gaps. Okay. That could work. But it's also topologically complete shit, which makes making minute details hard. Once you get the model, that's it, there's no making details to it.I may know how to model and how to do atlas textures now, but I don't know how to weight vertices in complex models like >>25113098 . But I am making these models to more personalize the project and turn it into less diluted AI slop. I'll figure out the weighing eventually.
>>25116361Lots of Tibia vibes there.
my NPC behavior/AI is pretty much locked in, now its a lot of refactoring and turning Debug.Logs into actual visuals and logicdid a little eye opening animation for the visible bar/how much the NPCs can see the player, need to work on scanning weights so how fast it fills up will change depending on how lit up is the player, how close to the NPC he is, and how fast he's moving
I wanted to have snake physics because I think it would feel more immersive with specific enemies. I've been working all day with AI to get this working. It does seem to work now.The idea behind this is as follows: i remember seeing diving videos to the M/S Estonia's wreckage. And I remember seeing videos with certain recognizable shapes. Assumed to be drowned people. The videos have been largely censored, but there's certain parts where you just know. And they would have these almost ghost-like unnatural strange physics prior to turning into silt. Almost as if they were made of submerged wet paper or gel (if you're interested check out this playlist: https://www.youtube.com/watch?v=ACpsOTfEplg&list=PLXX-CWidcGgqEXIYZZr-GVL3wwvJbX9Yk ). And I think it would be terrifying to face something like that.This functionality works by making the body practically a no-collide ragdoll and dragging it by the head. It works with mixamo rigs so I can apply snake physics on any characters. I'm going to assume I can use this for some snake/fish people and for the hanged man. I might need to adjust it more.
>>25002383working on a lot of visual updates, got a big vfx asset pack so I'm in the process of upgrading all my particle effects, screenshot of some of the new missile, explosion, and jump thruster vfx. I prob overpaid for a mega bundle and I can find better individual assets but I think it's worth it to just have a coherent style out of the box instead of trying to style match different assets.
>>25119678also filled in name and descriptions for most components and weapons, plus runtime rendered images.
>>25119713Looking cool. Weekend Progress- I fixed like a dozen bugs. Rewrote the background resizing process logic. And got halfway through adding a system for roads and cars to follow the road path.
>>25097777I swore I saw a sprite-to-nonogram maker before and it's relatively trivial to spritefy things. Wouldn't be a bad means to generate a base to work off. At least, I recall it when I had a dumb tetromino-nonogram idea where you have to solve the puzzle with a limited # of black & white pieces. Unfortunately the required information more or less had you solve the nonogram first then solve the puzzle - which was pretty complicated because I'm a jerk and have multiple semi-correct configurations. But it deflated the fun of nono solving.
>>25097565use RSUV instead, it's 32 bits you can set per-renderer and passed to the shader for essentially free, and works with modern SRP batching unlike material property blocks.I'm surprised AI recommended Material Property Blocks instead of RSUV since I had a similar problem and claude basically told me to use RSUV instead.https://docs.unity3d.com/6000.6/Documentation/Manual/renderer-shader-user-value-intro.html
>>25101553Multi-line text edit and IME supportcuz y not
>>25002383>>25106428Doing it in compute shader now and made the heightmap be persistently squished just to demonstrate having some state.
So, do I learn how to animate myself or do I just buy a set of anmiations like UAL? Right now I know almost nothing about animation, but I feel that unlike any other aspect of game dev it is very hard to get even to a serviceable standard?
>>25122361Good question, I went down a similar path of learning this. There's frame by frame, where you draw each part of the animation. That's what a lot of pixel art does. Old school Pokemon. And then Idk what this is called, but you can animate it on a timeline with keyframes, which is what a lot of cartoons do, and theres a built in animator in Unity you could use. Where you like, move your art around and save each pose. There's so much to learn for each, and I haven't really found great resources. Have had to rely on ChatGPT basically.
>>25115474Hay roofs are hard to draw in pixel art from scratchI guess I'll need to get another asset pack
>>25122361>I feel that unlike any other aspect of game dev it is very hard to get even to a serviceable standard?You'd be correct. Especially humanoid movements which we are trained to recognize.Animation, both 2D and 3D is a time-sink that took away years of my life. If you could get away without doing it, I'd definitely recommend that.But you know best what you need for yourself long-term. If you foresee yourself making animations for your games even 10 years later it's definitely a good investment. Like if you want wacky first-person reload animations, you'll have to learn some yourself anyway since there aren't many packs for that. Or non-humanoid monsters.>>25123689Just drop some yellow in there, less is more and all that.
>>25123875Mayhaps
>>25002383>>>/wsg/6247268Enemy chases you, if it loses sight it now searches in a way that's not completely retarded. If you ran down an alley it will search the alley.Game is first person, the birds eye POV and the ?! symbols are only for debugging.
>>25124776First person POV
>>25123875I couldn't help it but add some flair with the spray paint button, got positive feedback from some other fokls.Thoughts, /agdg/?
>>25125353Looks good to me dawg, easy stardew valley money.
>>25118632
>>25127859
>>25125174When close enough, the enemy predicts your path and intercepts you. Looks like a nothingburger but without this you can evade it forever by walking in a circle.
>>25002383spending some time on cute-useless stuffdropped items now rotate
teseting
>>25073041you're finished
>>25002383maked music https://vocaroo.com/179mKrTk8qu1
>>25002383today's prog:nothing to showwritten out like 20 txt files worth of docs trying to make an rpg system with wizards and shoot gun
>>25130858>cute-useless stuffBased, adding fun "useless" details is how you add soul to a game.
>>25002383>>25116361did a little puzzle that I wanted to make, the tiles in the background weren't supposed to be the same but I couldn't use a simple rand() function because of... reasons >>>/vg/583652679>>25117461yep, I'm using tibia's graphics as my main inspiration>>25130800>the enemy predicts your path and intercepts youcool!>>25131589sounds good! just in 03:26 and 03:35 it's a little too fast and sounds weird imo
>>25131209I feel like this has potential. Need more ways to kill enemies without a healthbar/damage system.
I baked all the characters and began importing models into the game as enemy classes.I could use some advice. I can't figure out why my blender made 3d models get fucky in specific conditions. They seem to work just fine when animating them in mixamo. But when I load them in the game, they get stretched in weird ways. The blood eagle model gets pushed in the ground and it's hips are squeezed inwards. The drowned man model doesn't work with snake physics baseline animation and get's scaled massive but squeezed in like a turtle. And the hanged man idle animation fucks things up as well, the character gets stretched. I figured it might be because of the animations but they're based on the regular mixamo fingerless rig, which is suppose to be compatible with all of the mixamo stuff. I've not had this problem before and can't even think where to start unraveling this problem.
Well lads, I've chosen to ignore all the good advice I've gotten in these threads and commit my little dumb space shooter to a Steam release. The page is live now, and there's no turning back.https://store.steampowered.com/app/5252500/Blueshift_Breakdown/Just tell me how NGMI this looks. I think I have the rough idea.
>>25138569It looks good.But sounds bad.If you manage to get enough wishlists to justify the cost, you might want to hire someone to redo the sounds.
