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

Name
Options
Subject
Comment
Verification
4chan Pass users can bypass this verification. [Learn More] [Login]
File
  • Please read the Rules and FAQ before posting.

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


[Advertise on 4chan]

[Catalog] [Archive]

File: 0.png (2.29 MB, 1080x1920)
2.29 MB
2.29 MB PNG
Welcome to /3/ - 3DCG, 4chan's board dedicated to 3D imagery and modeling.

readme.txt
If you're reading this, you probably got this linked to you because you posted a question that has already been asked many many times. Read ahead, and find your answer.

Scroll down for a useful FAQ and resource links
59 replies and 57 images omitted. Click here to view.
>>
Click here to open this thread.

File: wip 1.png (2.71 MB, 1500x882)
2.71 MB
2.71 MB PNG
/wip/ - Works In Progress - "We're going back to our roots" Edition

Post your work-in-progress projects, recently finished projects, or things you'd like critiqued here.

Previous thread: >>977313

List of free resources: https://pastebin.com/cZLVnNtB (embed) (embed)
/3/ Discord: https://discord.gg/gbYCEBPuK2
292 replies and 117 images omitted. Click here to view.
>>
>>981890
this >>981899 is not >>981638
>>
File: eyerigtest.gif (2.78 MB, 790x676)
2.78 MB
2.78 MB GIF
Successfully reverse engineered an eye rig. Want to make a mouth rig using similar techniques, but a little more expressive.
Still needs textures.
>>
>>981907
Good job including movement in the eye lids when the iris moves
>>
File: TopoGigiology.png (2.56 MB, 2880x1620)
2.56 MB
2.56 MB PNG
>>981363
Ah cool, i was getting a bit worried of making people feel like Sgt Hartman finding that donut in FMJ.

Topo is driving me crazy tho. those fangs have way more polys than they deserve, silly me thinking grabbing some cones and putting a crapload of segments on them was a good idea, well, if people think it's fixable i'll finish the mouth and apply mirror + subdiv/subsurf.

Getting some Muppet vibes from this, maybe i'll go in that direction, giving it some wiry boneless, bendy arms, i'm more of a M.C Hammer than M.C Escher guy so i don't know if i can deal with perspective stuff, trying to align 3d collision boxes with isometric prerendered models in Godot gave me a bit of PTSD, i'll suck it up if push comes to shove tho.
>>
File: file.png (2.32 MB, 1920x1040)
2.32 MB
2.32 MB PNG

File: doughnut.jpg (70 KB, 1920x1080)
70 KB
70 KB JPG
le donut of blender. based!
32 replies and 8 images omitted. Click here to view.
>>
>>981588
Looks like basic anisotropic noise fed into normal and/or roughness maps.
Basically take some perlin noise and scale it along one axis to get the long lines
>>
File: wholesome-chungus_png.png (249 KB, 495x577)
249 KB
249 KB PNG
>>981602
congratulations! you've nailed it!
>>
>>978618
>>
bump
>>
>>981639
lard? based! blender? based!

File: maira.gif (83 KB, 448x448)
83 KB
83 KB GIF
I am now much stronger, and now have mastered AI as well as 3D and 2D.
221 replies and 56 images omitted. Click here to view.
>>
>>981692
i coom
>>
File: kaliina.webm (2.41 MB, 1157x680)
2.41 MB
2.41 MB WEBM
more progress.
>>
https://www.youtube.com/watch?v=p7uThh6mrRo

Transition made.
>>
>>981897
>cris
>transition
uh oh...
>>
>>981891
based

File: Longdog.jpg (1.45 MB, 1500x1500)
1.45 MB
1.45 MB JPG
Previous thread: >>973562
146 replies and 40 images omitted. Click here to view.
>>
>>980304

is there any way to change object scaling from "global" to absolute sizes? am beginner and i just learned that the default cube is actually 2 meters, but the scale says 1.000

turns out that the scaling is relative to itself. it's a nightmare doing all of the calculations to ensure that i change it to an exact size (am making small boltgun, i need the handle to fit a human hand, so small).
>>
>>981921

blender recommends 32 GB ram, i bought a faggy gaming computer covered in LED accents just to do this as a hobby. i had the same issue with a basic PC, i think any 4th generation GPU would be needed for textures.
>>
>>981924
gotcha ill just save a schizophrenic amount
>>
>>981922
Yeah, you can change the units in the scene tab. It's good for when you want to model using real-world scale, but can be kid of a pain in the ass because you'll still have to type m/cm/in/ft when entering values if you want to mix them (like 1 ft, 7in), and you can't specify the units until you confirm the action (or press the = sign during the transformation to type it in).
Like it's fine when you're using one unit like feet, but it's a pain when you want to mix in inches. Though I guess using meters would be less of a pain since centimeters divide by 100, so 1.1m would just be 1 meter 10cm.
Fuck, now I'm starting to lean towards jumping on the metric train.
>>
File: Units.jpg (60 KB, 808x412)
60 KB
60 KB JPG
>>981922
>>981927
Fuck. I forgot to attach the image because I was ranting.

File: wet_glass.jpg (2.25 MB, 3840x2160)
2.25 MB
2.25 MB JPG
Previous: >>968906

Daz youtuber's and general guides:
https://www.daz3d.com/help/help-daz-3d-video-tutorials
Wpguru
Parmy
Travis Davids (Marvelous designer)
Thundornd
CgBytes
https://www.laylo3d.com/
https://thinkdrawart.com/
Kelvin Jin

Bridges: https://www.daz3d.com/daz-bridges
Suggested Blender Bridge: http://diffeomorphic.blogspot.com/

Comment too long. Click here to view the full text.
97 replies and 28 images omitted. Click here to view.
>>
>>981796
yeah, its my custom
>>981799
yep
>>
>>981713
>>981753
would it kill you to use the post denoiser? all you need is 10 frames at the end.
>>
Why is DAZ considered bad for doing animations?
>>
>>
>>981920
Eww

File: 1324401739611.jpg (434 KB, 934x645)
434 KB
434 KB JPG
Welcome to our thread dedicated to honing your topology skills in 3D art! Whether you're a seasoned artist or just starting out, mastering topology is crucial for creating clean and efficient 3D models. In this thread, we'll share and discuss various exercises aimed at improving your understanding and application of topology.
266 replies and 148 images omitted. Click here to view.
>>
>>981024
It's like one big puzzle coming together.
>>
File: GKcngntWIAAdeAg.jpg (1.09 MB, 2800x1512)
1.09 MB
1.09 MB JPG
>>966562
>>
I like how sloppy these wipers look. It just werks and motionblur takes care of everything
>>
I'm working on a gun, and the grip part has some curves obviously. Using subdivision modelling. The problem I'm having is that I like to keep things as low poly as possible to keep them manageable, but I just can't realistically tie in the details I need with a low poly version of parts of it.
Would it be bad practice to model the general shape low poly, apply 1 sibdivision to get the nicely smoothed slightly higher poly version, and then tie the smaller details into that? It feels hacky, but I'm not going to be able to hand-place every single one of the vertices just right as compared to the subd.
>>
>>981915
That is pretty common workflow. Rough out a shape in sub-D collapse/flatten the stack down to that level and then optimize the sub-d generated mesh to fit your target needs by modifying deleting loops etc.

There's nothing hacky about it, any method you can think of to generate the polygons you need is valid if the end result is something useful that matches your target specs.

Which VR device has the most accurate/precise controller tracking for modeling in VR?
120 replies and 87 images omitted. Click here to view.
>>
File: IMG_3064.jpg (319 KB, 1918x1436)
319 KB
319 KB JPG
>>
why are you fucking doing this
>>
>>981665
What you should be asking instead is why the jannies of this board let Cris do this.
>>
File: IMG_3173.jpg (154 KB, 1618x1486)
154 KB
154 KB JPG
>>
File: IMG_3199.jpg (240 KB, 1861x1380)
240 KB
240 KB JPG

File: Untitled.png (1.96 MB, 1920x1080)
1.96 MB
1.96 MB PNG
Another blog post of a guy hoping to make it.
for reference i never complete any projects cause I'm lazy and I lack discipline, also because I have a day job.
Let's see if I can finish this time, goal is to finish a base mesh with game res and textures, as well as eyes and hair cards.
Also, please feel free to destroy me, I value your opinions.
238 replies and 70 images omitted. Click here to view.
>>
>>973420
wtc reference pics like this? any particular repository or search terms?
don't have to be Asian. just decent ref
>>
>>980997
check /ic/ they post them on the artbook threads.
>>
>>981075
gratzi
>>
File: Untitled.png (2.36 MB, 1920x1041)
2.36 MB
2.36 MB PNG
got rid of the redness below the breasts, all things given I think I came pretty close to the intended body shape.
>>
>>981095
Very nice

File: chain reveal.webm (625 KB, 720x864)
625 KB
625 KB WEBM
I made product renderings for a client and it went great but now the client is asking for an animation like vid related.

I tried the (newly updated) particle simulation in C4D to try to create a body of liquid and also tried xParticles but both runs like shit on my M1 MacBook. I tried my best to optimize the simulations following multiple tutorials online but I cant get semirealistic results.

Now that I am thinking about it I might be vastly overcomplicating things since I basically just need to raise the product from a plane that somewhat sticks to the object.

Does anyone have an idea for a method I could use to achieve the effect in a much simpler way? I must be missing something.
17 replies and 5 images omitted. Click here to view.
>>
>>981710
>>981711
or you could just do sdf booleans and never have to render a pass and maintain more control over the geo with less friction

thanks for the essays and the 1980's techniques though
>>
>>981778
VFX just means you do it as compositing in a videoediting software. You render out your mesh and a depth pass and inside your video editing software you'd set up layers to blend it all together to make a transition between your CGI and your other layer using your Z-depth.

I work real-time stuff so I haven't touch any videoeditor in years but 'adobe premiere' and 'after effects' was what I used back in the day and it was really easy to just jump in and use those as it's basically a photoshop for videos and I already knew my way around photoshop inside out.

There's probably better suites available, all the VFX tryhards used different software back then too but I can't recall their name.
>>
>>981685
>>981699

How's that discerning eye of yours now you sad fuck?
>>
>>981818
pretty good, man; feels great to be better than everyone else. how you doing?
>>
>>981610
Is this even particles and not a plane intersecting with a special shader?

File: selfie.png (29 KB, 341x450)
29 KB
29 KB PNG
We have this over in /ic. I'd like to see anons portfolios and works. Post your socials.
12 replies and 3 images omitted. Click here to view.
>>
>>981757
Reminder that I'm a hobbyist and I will post my work if I fucking feel like it.
>>
File: demon render.png (1.58 MB, 1148x1141)
1.58 MB
1.58 MB PNG
>>981605
artstation.com/stvarnov
>>
>>981820
Just when I thought you couldn't get more based
>>
>>981884
Oh no don't get me wrong it's just group of lawyers and surveillance enthusiasts that profile, detect and denounce acts of vandalism by leftards. Wéve also dismantled a couple of groups just by using their 'hate speech' jargain against them.
>>
>>981908
jargon

File: DVjx31UU8AE8OJy.jpg (86 KB, 1198x832)
86 KB
86 KB JPG
Previous >>959310

Discuss and post anime styles in 3d. From figurines to celshaded models, etc.
62 replies and 25 images omitted. Click here to view.
>>
>>981838
What's this supposed to be a tutorial for? I don't really understand what the result is supposed to be?
>>
>>981852
cloth fold
>>
File: 1671819394892944.jpg (131 KB, 1384x817)
131 KB
131 KB JPG
>>
>>981835
To do this. I made a nurbs circle and placed it about where I wanted the ponytail to come out, duplicated it and used maya's "live surface" to move the CVs of the circle onto the scalp in the irregular shape it needed to be.
Then I lofted between the 2 curves, and extracted another loop out of the middle of the surface. I adjusted the CVs of that middle loop to be in a better spot, then lofted another surface on all 3 loops and did the same thing 1 more time.
>>
>>981900
alternatively I've offset a sphere I sculpted to intersect with the head where I wanted and did a boolean subtraction.

File: ilc3y1t0mqdb1.jpg (66 KB, 600x600)
66 KB
66 KB JPG
Do they take a special computer graphics course in Japanese only? Why do we westerners have to be content with a subpar anatomy tutorial with ugly bodies?

My answer: Topology over anything else. After extracting 3d models from multiple japanese games myself, they were simply made by hobbyists with no background at modeling. It's the information they had available at Japan which is something I need to obtain.
21 replies and 1 image omitted. Click here to view.
>>
just hit the retopo button in zbrush a bunch of times you fucking monogloids it's not even an arform

>RETOPO
>RETOPO
>RETOPO
oooo I did it I'm yellow now I'm a little japanese man I'm a SAVANT i'm a genus. eat my shit trump 2024
>>
>>980142
Are you ok anon? You seem in distress
>>
>>980142
Do it with this model: https://3d.si.edu/object/3d/1903-wright-flyer%3Ad8c62e5e-4ebc-11ea-b77f-2e728ce88125

I know you can’t because the scanner messed it up and become impossible to use it.
>>
File: pc88.png (549 KB, 640x400)
549 KB
549 KB PNG
>>979843
same way they mastered pixel art:
by being a country of obsessed pedos
>>
File: 1531464858239.jpg (1.21 MB, 1920x1929)
1.21 MB
1.21 MB JPG
>>979843
It's all in the texture

File: 1691660776687753.jpg (211 KB, 1335x865)
211 KB
211 KB JPG
So....that big AI image or video to 3d mesh never materialized, did it? Its mid april. We should all be free of modelling by now. All this big AI talk, but nothing delivered. No image to model, no video to model, no ai texure, no ai uv, and github copilot stinks on ice.
16 replies and 2 images omitted. Click here to view.
>>
There's this project that applies the ideas of noise/diffusion based image generation, like stable diffusion and combines it with gaussian splatting.

The results are pretty good for what is essentially first gen. It even has browser based demo. https://gsgen3d.github.io/ if you can use this to generate a mesh, you could then use another program to retopo, and use image based ai to generate extra maps, or change texture styles.

It wouldn't surprise me if some dev starts using this kind of workflow for background assets in the next 5 or so years. Especially with nanite allowing unnecessary details.

Things like the fruits in a fruit stand in the next elder scrolls game, which is probably still run a gamebryo derivative...
>>
AI will temporarily die the moment someone wins a trademark case. AI will have to be retrained with data that is not trademarked eventually
>>
>>981250
Not a single strike won because AI training on copyrighted data is legal and will remain legal. And what do you do when AI is trained on AI that is trained on copyrighted data? This shit should have sorted out in 2022 and the fact that it didnt fudge even an inch is sign that it will remain that way forever.
>>
>>981894
>And what do you do when AI is trained on AI that is trained on copyrighted data?
ChatGPT and other AI networks already degraded because it's a cycle of shit, this doesn't work.
>>
>>981895
>>981894
>>981250
Honestly, "AI" (if you really can even remotely call it that) has peaked. There won't be any significant "game-changer" improvements in any capacity for a long time. Fundamentally, AI research the way it is now is pretty flawed and there's always going to be an upper limit of what's capable. Which we're pretty close to reaching. As they are now, all the computing power in the world won't make them smarter or more robust, just train them faster. Even training has diminishing returns, since training something too much ends up making it shit.

AI is going to be stuck in "toyland" for a good while. Don't get me wrong, it's useful as it is, and there's certain applications that are really cool, but it'll never become a "real boy" like Pinocchio. The AI shit that seems like straight up magic won't happen for a really really long time. We'd need a significant breakthrough in computing and creating neural nets for something like that. Like we'd need a literal cyberbrain first.

So yeah, in a sense "AI" is already dead. Or rather it hasn't even begun yet. What we have now is just a pale imitation of it by techbros. Neat and useful? Absolutely. But we're a long way off from something that'd really change the game enough to actually warrant fundamental policy change. Policy makers will probably do shit because they're old retards that don't know how tech works, but it's not really worth it at this stage. Image/video generation aren't something that'll shake the world to its core.

Creating garment assets for game environments.

Seems it's a bit niche?

Pivoting from a Rhino + SolidWorks / Blender (to render) background.

What should I know?
Do you have tutorial/ resource recommendations?
**yes i have checked the intro guide**
>>
program is simple. you can learn 100% of it in a few days.
pattern cutting, depending on what you want to do, is not. you can learn 30% of it in a 3 year university degree.
>>
Does it make more sense to texture in CLO or bring it to blender and texture it there?
>>
>>981883
No. Marvelous has a view to layout uvs, just do that, export and texture wherever you normally do. Marvelous will export stitching at geometry iirc. You're better off building that into the textures with something like substance painter (or redo the stich geo procedurally in Houdini or geo nodes if you need autistic levels of detail)


[Advertise on 4chan]

Delete Post: [File Only] Style:
[1] [2] [3] [4] [5] [6] [7] [8] [9] [10]
[1] [2] [3] [4] [5] [6] [7] [8] [9] [10]
[Disable Mobile View / Use Desktop Site]

[Enable Mobile View / Use Mobile Site]

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