when i render it it has this diagonal line, why is it?Also please rate the model, its fully og i wanted to mess around with an og design, critique anything u want, or if its bad, ik im not the best.
>>1032461It's not a werid haading issue, that shit is just not flat
>>1032465looks like i fixed it by aligning vertex, with snap to vertex mode
>>1032474Anon, I see you fucked up the face in the back in the process. Pay closer attention kek.
>>1032476yeah but thats cuz idk how to fix these little inconsistencies, so im gonna continue creating and learning from it, and if some blender god has any tips here ill be gratefull and keep evolving
every quad gets converted into two triangles that quad is skewed somehow
i am completely broke. i cannot afford virtamate, the cracked version of koikatsu party i have feels like torture to use, and gives shitty results, and i know nothing about 3d modeling so far, the donut tutorial went into one ear and out the other. any suggestions for free software? bills need to be paid, so hentai needs to be made.
>>1032393wdym? if you just wnatl ike game characters and make goonshit out of them you more need to know compositing and find decent poses isn't THAT hard. animation is way harder though if you want to make that.
Learn how to learn, then put in the hours.
>>1032422Didn't you see the other thread, learning how to learn is a false flag operation to gatekeep access to making art
>>1032435
>>1032393it depends on what you want to make. If it is quick and easy money. learn animation, there's already tons of already rigged nsfw models out there for either cheap or free. you may be able to port out some modded source filmaker models if desperate. If you want long term success, first learn anatomy then modeling. selling nsfw models can be a market of its own if you don't plan on monopolizing your own model
Volumetric space layering is writing depth into the prompt itself so the model builds z-space not just x and y, foreground objects have mass and occlusion, midground scales correctly, background dissolves into atmospheric haze, and the result is something you feel you could walk into rather than something you are looking at.
>>1032406>I'm not the one shilling AI here, that's your problem, not mine.I'm basically functionally illiterate without it
>>1032407Good thing AI can literally teach you how to properly create and assemble a beach scene.and/or link you the resources to.
>>1032408AI can't manufacture the feeling of your beach sunset for you.
>>1032409Yeah, it'll just tell you to touch grass, which is probably a good idea.
>>1032410mp4 of webp
It's an inset on my AR I'm working on in blender, topo in replyI'll show the other side of the shape too because idk if it's just that I can't get the light to hit it right but it doesn't seem to have the same issue. It connects differently but it had the same sort of connection before and still only the top had a problem, I'll try to show
>>1031174Looks pretty good I didn't know about bevel modifier really. That would probably be easier to work with you are right. I imagine the real geometry is something similar to what I have though although maybe cleaner?
>>1031175not that clean desu
>>1031173>How would I have 2-4 edge loops for this shape and keep quads only?You should be getting your high detail from something like a subdivision surface mod, not from cramming tons of edge loops trying to force a rounded edge.
Is this the shadows fuck-ups thread? Cus my data transfer modifier looks fine in the preview, but then I go to apply it and not only have the normals seemingly reverted, but now there are all these nasty little creases.
>>1032366Your mark sharps are probably fucked somehow. Data transfer can be finicky like that.
Can we have a productive thread to discuss how to use AI to improve our 3D workflows and become faster and produce professional results, rather than trying to cry about some imaginary 4chan anti AI versus pro AI debate?
Are there any good workflows to create generate low poly meshes out of crudely drawn reference sheets yet?I'm trying to make anthropomorphic Nintendo 64 style character models and it would be so much easier if I could just draw a basic head shape in MS paint, put it into an AI and have that generate me a mesh which I can put in Blender and edit to my liking.I've tried this with multiple frontier LLM's (putting in an image, then instructing it to generate a python script that I can use to spawn it into Blender) but they all either get the shape completely wrong or have dogshit topology to the point where it's better to just make it myself from scratch.The struggle of being a 3D noob is real.
>>1029757sometimes I use it as a base to sculpt from, as a way to iterate ideas on a flat sketch, or to create textures I don't have the funds or skill to have made.
What Comfy setups are you using to run your ZBrush sculpts through for meshing?
>>1030611why would use it for N64 style shit? it's literally all in texture. Ai can't rig or animate for shit.
>>1029757>/g/ here i have no idea how to use any 3D software but my oh my, gemini cooks. it's not really "professional" like how you guys work, but it came out good. https://www.youtube.com/watch?v=PNvFOYe0gEYhttps://www.youtube.com/watch?v=pkPsAYIkw58
What the title says.I'm a total beg.Have a coding background and want to start making my own assets for non commercial game projects.It's likely I'll be sticking to low poly stuff in the long run, but wouldn't want that to keep me from learning "proper" technique. So far the UI looks comfy, but since I'm complete beginner at 3D modelling I'm wary I'll be picking up some bad habits or navigate myself into a dead end.Something like pic related is the level of detail I want to be able to produce.
>>1032171personally godot has been good to me, but you have to be warned it was never built with 3d in mind. anything even slightly off the beaten path is undocumented or worse, implement-it-yourself. with your background im guessing reverse engineering an engine is manageable, but for me it's suffering. i checked out blockbench and it seems great for what it does, but for the model you want i think the geometry would get messy fast. if you're doing low poly shit you have the chance to make something comically performant, so why let blockbench screw you by filling the mesh up with texture weirdness and unnessesary verts?
>>1032168Just use blender. Blockbench can do what you want, but its really impractical to use, and the animation side of things is so bad you will end up having to move the models to a different program to do rigging/animations, at which point just do the whole process there lol. The only reason to use blockbench is the minecraft pipeline, since the program was initially made exclusively to make minecraft assets
>>1032255Thanks man. Playing around with it over the weekend was fun, but I'll give Blender a try next.A few other sources also told me about difficulties with the animation side of things. I also found the way it implements reference images weird, but that might be a beg complaint.>>1032229I'll see how far Godot will get me. Not fixating on low poly yet, but like you said I have a few ideas in mind where I figured I could use the extra performance. Stuff like rendering a ton of 2D sprites from a few 3D models.
>>1032229what features is it missing, in your opinion? My only complaint is that the mesh instancing system was a bit difficult to figure out but that was when I was making an RTS prototype, for what most use mesh instancing for it'd be spartan but ultimately fine.
>>1032263Blenders built in texture painting is doodoo still but it beats out blockbench in terms of overall utility. The compositor alone is worth learning as blender can be used as a makeshift photo editor. If you prefer blockbench's simplicity you can make a model with a rig in blockbench then import to blender for better fx and renders
The free-use OpenAnatomy data, specific example here the skeleton's mandible de-triangulated by myself, has the most beautiful topology I've ever seen. The mandible can be converted entirely to quads except for two tris here, in the mandibular canal, and WHO MODELS THE MANDIBULAR CANAL?!( https://www.openanatomy.org/ )ITT: share beautiful topology.
(interior of OpenAnatomy mandible at mandibular canal end)
(correction, data is from https://anatomytool.org/open3dmodel )
>>1032261Thanks for sharing, that is very nice. I've been meaning to do some anatomic studies.
eyyyy protip (spoiler): the file is fully UV-mapped with normal maps, don't just convert to quads willy-nilly. just be content in knowing that it *can* be converted to quads.
specially the little corners and everything, i cant even do the top area well, any tips on how to proceed plz
>>1031947im trying to do what u did with symetrize and mirror but i cant seem to understand how u did it, can u plz elaborate, ive googled it but symetrize doesent do it nor mirror
>>1032100you need to rotate Empty.004
>>1032109the end result
>>1032221it looks like a crab now
>>1032322yea idk how to texture, onto the next project ig
Previous thread: >>1021734
>>1032625So far so good.
Collage editionPrevious thread: >>1029138
holy epic fail lol
new thread fellas>>1032729>>1032729>>1032729
>>1031630Arigato Mr Roboto
Jita - She doesn’t wear a shirt because wearing shirts is weird where she lives. Only her head is "finished".
>>1032280Here’s the current progress on the game. I’ve gotten quite a bit done since I last posted. I mainly just did work on the battle system, adding things like status effects, character traits (permanent status effects that apply at the beginning of battle, essentially), different ability types (counter-attacks, blocking, follow-up attacks, etc.), spell cast times, ability costs, and some general logic enhancement.The next big thing I plan on doing is implementing enemy behavior, even if it’s just something simple for now. I do believe I’m getting the game into a state that’s “playable”, even if it’s not that fun yet. Of course, I still have a ton of other things to do like enhancing battle feedback, adding visual effects, making the GUI pretty, and all the other things I plan to do with the game.>>1030807Yeah, I’m still working on it. I wanted to try animating by hand just for the experience, but I may ultimately go with physics simulations for things like flapping and jiggling and all those other fun things. Most of the animations I have in the game for my characters look weird at the moment, but as I said earlier, I will redo them all eventually.>>1030199>I'm wondering how these will look shaded and textured. Haha, about that…
>>1032281Here’s another new character, Tache Noire, the sclopetiform (rifle-shaped) fish. I figured I didn’t have many ranged characters planned for my game, and this one, based on the robust ghost pipefish, seemed like a fun idea. I also plan to give it a special ability where it pools all of its elemental power/resistance stats into one, and can shift its color to transmute that stat to the corresponding element.https://files.catbox.moe/9mmvy5.png https://files.catbox.moe/tw5y4r.mp4
>>1032282And as a bonus post, here’s a few renders I made a couple years ago when I decided to try out 3DS Max on a virtual machine of Windows NT 4.0. It was a pain in the ass to get it to work, but it eventually did work and I went through some of the tutorials and made a few of my own things. I haven’t touched it in a while since it was rather time-consuming and I wanted to focus more on my game.https://files.catbox.moe/c4avud.jpg https://files.catbox.moe/0l0j3g.mp4 https://files.catbox.moe/174ma3.jpg https://files.catbox.moe/sbvvlf.mp4 https://files.catbox.moe/jj3roe.jpg https://files.catbox.moe/n89gxx.jpg https://files.catbox.moe/9wf0d2.jpg
>>1032279welcome back! any chance of us seeing that nora swinging tits animation?or a nude version of >>1032278 would be nice
>>1032280>I did get a massive retard boner while working on her.That's just proof of a well-designed character to me!
Feels good to be able to make this.
Sugimori-san learned 3d?!!? he could BTFO gamefreak
kill yourself cris
for 3 reasons> V snap> D move pivot> DV snap pivot
>>1032232The pivot is already there you dingus just use the 3D cursor like a sane person.
>it's another thread where we give awful advices to newbies to gatekeep a hobbyModeling for maya is awful, but if you want somebody to really suffer, tell them that 3dsmax is great for rigging and animation. I hate 3dsmax so fucking much, yet my job requires me to work with it. I don't know if I want tlo handle 7 more years of autodesk code diarrhea
Wait a second, this thread is from...march? Holy shit this board is fuckin dead
>>1032257What exactly is the problem with 3dsmax? I remember using it for vidya modding. Just some simple model swaps for Stalker, but you had to do some unholy ritual to assign bone weights to vertices.
>>1032241It's 2026 AI makes your 10 step workflow into 1 step
how can i make this in 3d? i'm suppose to join a cylinder to an hexagon and vice versa, but how? this is driving me crazy
>>1030013
>>1030014
>>1030009Looks good btw but you need to boolean the two parts together and these edges yeah
easy, peasy
What, nobody likes texturing?
Is it a good idea to just make one base humanoid model and use that if you want to create multiple characters?
>>1032114>>1032142japan devs still use 2013 maya and do low poly techniques they don't give a shit about topology curves
>>1032142No, you should explicitely NOT subdivide without catmull clark. THough subdividing and applying smoothing can be useful if you use it carefully.Basically, a guy who does videos on youtube rather than 'is an artist for real' put out a video where he misunderstands that a real japanese artist suggested that you should not just wholesale apply a catmull clark modifier to your anime character models to smooth things out (yes, don't even fight me this is what he was trying to say, he even applies a modifier to a gacha game model to demonstrate why you don't want to do that), which is true, though he also conflates just straight subdivision (like, he gives an example and it naturally starts to harden the edges around each original face because of course it would, and then pretends like that's a bad outcome rather than just a retarded thing nobody does in the first place), because that style of subdivision requires specific non-optimal topology to get certain features once you most past a certain low level of detail, which starts to defeat it's whole purpose and use. The correct use in this case is to use catmull-clark style modifiers to get the smooth shapes you want without needing to control an unnecessary amount of geometry manually. Once you have those core shapes you can apply it to get the actual mesh out of it, and then start to tweak things manually, no modifier applied on top. There is NO reason to fiddle with 1000 polygons to get an anime girl thigh when you can do it with 40 and then just apply a modifier and cut the straps in after the fact and have a perfectly smooth mesh. OBVIOUSLY final models don't have modifiers applied, relying on edge creasing and shit. I don't know what could possibly get somebody to come to that conclusion other than being sculptingbrained and just knowing nothing about poly modelling. It's like 101 shit.
>>1032152Won't smooth shading do the job in most cases if you're not too picky?
>>1032065Just finish it and pose it, nothing horribly wrong there
>>1032152...or you can just use soft selection/proportional editing and sculpt brushes to have control over it. You don't necessarily need subdivision to do stuff.I say that as someone that almost exclusively does subdivision modelling.
how bad is it, this kinda hard, not cuz hard itself, but cuz i need lots of time, something my min wage slave job wotn allow for, any tips to improve? thanks
>>1032150Why's it upside down?
>>1032159¿uɐǝɯ noʎ op ʇɐɥM
You already have a thread motherfucker>>1031896
>>1032159its not, its just missing legs, wich imma add and drawers
