[ 3 / biz / cgl / ck / diy / fa / ic / jp / lit / sci / vr / vt ] [ index / top / reports ] [ become a patron ] [ status ]
2023-11: Warosu is now out of extended maintenance.

/3/ - 3DCG

View post   

File: 1.36 MB, 1500x882, WIPOriginal.jpg [View same] [iqdb] [saucenao] [google]
947917 No.947917 [Reply] [Original]

/wip/ - Works in Progress -True Edition
Post your work-in-progress projects, recently finished projects, or things you'd like critiqued here.

Previous thread: >>947775
List of free resources: https://pastebin.com/cZLVnNtB (embed)

/3/ Discord for those interested: https://discord.gg/gbYCEBPuK2

>> No.947918


>> No.947926


>> No.947939
File: 1.87 MB, 1920x1080, jihad_house.png [View same] [iqdb] [saucenao] [google]

All thats left is to make the balcony, and add better stuff on the roof.

also POOP.

>> No.947949
File: 1.78 MB, 1930x1088, TailsHittinIt.png [View same] [iqdb] [saucenao] [google]

Trying to get the composition and lighting right, and making the facial expressions convey the right emotions.

>> No.947950

can you retards keep it to one thread
>>936854 has been up for 4 months, either use it or get a janny to baleet it

>> No.947984

Stop posting in it, and it will poof. or post too much on it and it will poof. If the thread doesnt have the right imagine, i will hide it.

>> No.947994

The car was created using Blender 3D

>> No.948005

previous thread link is pointing to some guys question lol

>> No.948006


>> No.948016
File: 519 KB, 1920x1080, file.png [View same] [iqdb] [saucenao] [google]

Terastilized Mew (no crown)

>> No.948017
File: 333 KB, 570x692, file.png [View same] [iqdb] [saucenao] [google]

Eevee render

>> No.948018

Tails' face lmao

>> No.948019
File: 1.08 MB, 960x540, LeslieWIPTurnaround2.webm [View same] [iqdb] [saucenao] [google]

Working on some furshit

>> No.948052
File: 252 KB, 494x565, MEME TAIls.png [View same] [iqdb] [saucenao] [google]


>> No.948099
File: 404 KB, 1920x1080, file.png [View same] [iqdb] [saucenao] [google]

Got the shader nailed down:

>> No.948100
File: 392 KB, 1920x1080, file.png [View same] [iqdb] [saucenao] [google]


>> No.948165

>this retard really thinks anyone can use that "wip" thread
Get out of here, tourist. We got traditions n shit. Not many, and not good, but we got em damn it.

>> No.948194

/3/ is die... What happened?

>> No.948208

Some /pol/ posters made a discord and now people post in those instead of here because the sweet dopamine of someone commenting on your work comes faster.

>> No.948222

/3/ has always been slow, cool your boogeyman autism

>> No.948264

It got slower. It's not a boogeyman. I went into the first discord people made on the first day it was posted and someone was talking about how Mein Kampf helped him get in shape and get his life together. That was only one of the egregious things I saw. You make the board worse by contributing to the problem or by pretending it isn't a problem. Please leave.

>> No.948269

>walk into library
>one guy having a schizo episode

i can come up with a million more plausible sounding reasons why the board is "slower" other than "muh /pol/
have you considered maybe its because the rise of AI image generation, or maybe its because of retards like you moaning about unrelated shit and not posting WIPs, or maybe its because someone split the thread for no reason? i could go on
theres plenty of other threads on /3/ that have a relatively healthy rate of posts -- a handful of people posting on discord isnt going to kill a board. the posting there is about as fast as /3/
shut up and post WIPs

>> No.948287
File: 1.61 MB, 1920x1080, rocks textured.jpg [View same] [iqdb] [saucenao] [google]

rocks are fun to make

>> No.948291

retarded on basically every count

>> No.948522

lmao even

>> No.948544

Go back to discord

>> No.948682
File: 3.13 MB, 4240x1856, pixie.png [View same] [iqdb] [saucenao] [google]

How does it look? Too crackwhore-ish?

>> No.948693

The wings being made out of her skin it's disgusting

>> No.948802

Ok maybe one post every few days is a bit *too* slow and raises some eyebrows. I preached that /3/ has always been slow before but now I'm not sure it's natural, it's not this slow on a normal day.

>> No.948834

some retard decided to make a wip thread with a different op pic
so now we have 2 for the next couple months splitting the non existent traffic

>> No.948835
File: 1.19 MB, 1080x1920, scout4.png [View same] [iqdb] [saucenao] [google]

The scout is an alien!

>> No.948906
File: 673 KB, 475x674, eyes.png [View same] [iqdb] [saucenao] [google]

not bad but I'd focus on the eyes, a lot more. look at some better reference of how the eyeballs fit into the head and how the eyelids interact with them
right now it looks like you just cut holes into the head and inflated the edges, even stylized or anime faces can look fucked by doing this
sketchfab can pretty good for reference as you can inspect models from any angle and with wireframes on too, here's a couple I found searching shit like "anime face" or "stylized eyes"

>> No.948907
File: 481 KB, 898x705, wireframe.png [View same] [iqdb] [saucenao] [google]

trying to make my own basemesh that I can work off in future projects
I never seem to be satisfied with the topology as each time I start a new one it gets better, but I don't want to start from scratch every time I make a new character

people tell me to just grab a random free basemesh from somewhere but how the fuck am I supposed to learn that way? I might as well just sell Frankensteined collages of other people's work on booth or gumroad or some shit

damn are those procedural? photogrammetry? hand-painted?
lemmy see the topology man

it's giving me half-life 2 synth vibes, you planning on animating it?

>> No.948909
File: 1.47 MB, 968x1005, 1663978018448228.png [View same] [iqdb] [saucenao] [google]

It took me 3 days and tons of bitching, but I finally got the Painter Prman Preset Browser integration to work for 25.1. No more manually having to wire and convert each and every tex map for every change. The "unfortunate" part is my PrMan render looks much much stronger than my Painter iRay preview

>> No.948935
File: 223 KB, 1920x1050, subby.png [View same] [iqdb] [saucenao] [google]

Best way to do implosions? You have 40 hours.

>> No.948936

You beat me to it. Note: The titan didn't have propeller covers on top.

>> No.948937

Damn, I liked that curved shape.
In Max I'd use the explosion modifier with negative strenght. In Maya, dunno yet.

>> No.948939

In maya I'd assign a displace attribute with a negative cellular map.

>> No.948940

Nope sorry just confirmed it had the top shroud added for this trip.

>> No.948942
File: 268 KB, 1920x1050, subby2.png [View same] [iqdb] [saucenao] [google]


>> No.948952
File: 539 KB, 1578x841, et56Miz[1].png [View same] [iqdb] [saucenao] [google]

Finished with that cat, working on my next creature. I have some regrets.

>> No.948965

messing around with bloom;
do ignore the clipping sword, I'm retarded

>> No.948966
File: 2.38 MB, 816x1000, zungarf2.gif [View same] [iqdb] [saucenao] [google]


>> No.948998
File: 1.12 MB, 1031x956, Capture.png [View same] [iqdb] [saucenao] [google]

Re-watched Tron Uprising and got the itch to recreate some shots and maybe do a basic animation.

Not used to stylized designs though and already struggling with the artstyle & shaders. Haven't even started trying to sculpt some of the weirdly proportion characters

>> No.949055
File: 2.74 MB, 4076x1856, pixie2.png [View same] [iqdb] [saucenao] [google]

More painting, changed face.

>> No.949056

What will these be used for?
Looks even more like a Crack whore now

>> No.949059

>What will these be used for?
Hypnosis fetish stuff

I almost like this but you've gotta figure out something for the eyes, that shading on the bridge of the nose only accentuates that there's a space between them and the nose that shouldn't be there. I also think some tear ducts might help.

>> No.949060

can I see the cat topology? everyone seems to have different methods for hair tufts and I can never settle on which I prefer.

nice! you made a cityscape with those yet?

eyes look a lot better now, nice one. Also why does it look like you're doing a high-poly sculpt in Maya?
I've only ever used Maya's sculpt tools to reshape or make blendshapes

>> No.949065

>Also why does it look like you're doing a high-poly sculpt in Maya?
That's a zRemeshed model with base color, roughness, normal applied.
Should I have been more heavy-handed on the retopology?

>> No.949068
File: 691 KB, 900x711, KFH9C8c[1].png [View same] [iqdb] [saucenao] [google]

The tuft bits on her are a bit shit because I'm slapping enough hair on everything that I don't even have to render them, just use them as emitters. If I ever try to make a VRC version I'll probably weld them on. They're all sculpted and manually retopo'd, except the bottom cheek one where I realized I could just put a decimate modifier on it and save the trouble.

The rest of her topology isn't really something to be proud of, but it deforms well enough that I don't have to scrap the mesh at least.

>> No.949069
File: 1.23 MB, 1171x1835, ezgZRhGEVN.png [View same] [iqdb] [saucenao] [google]

Pic related.
There's a torso under the leotard for now hence the overlapping geo.

>> No.949070

why did you make her so lewd?

>> No.949080
File: 511 KB, 2332x3879, normal transfer cards.png [View same] [iqdb] [saucenao] [google]

oh I was expecting a 3M+poly triangulated mess so you actually exceeded my expectations
I still would hate to rig or animate it or use it in a game though
you could probably get away with rigging it for some poses and renders etc.
but then again I'm unreasonably autistic when it comes to topology for some reason, I always manually retopo my sculpts, zremesh or quadremesh just don't do it for me

I wish I had that mentality, I end up scrapping so much shit just because of minor issues.
looks good though, would you try and keep the fur look in VRC though? I never see passable fur in VR unless it's on some very poor avatar with 100k hair cards or some expensive as fuck shader.
It seems pretty difficult to get stuff like that to blend into the mesh without using flat shading or weird normal transferring tricks (pic related) I've tried it myself but it never seems to look right

>> No.949084

No, I don't think it would be worth trying to keep a finer fur texture in an engine unless you were doing some wizard shit with like Ornatrix. With default blender tools it's going to look weird or be heavy, and plus, you'll just look out of place next to all the other avatars with flatter materials. Something like this would be cool - https://sketchfab.com/3d-models/graograman-fur-study-21b40edc94104acbad85b825ed45733c - but I'm not going to attempt it unless I wind up switching to the whole Maya hair ecosystem someday.

>> No.949094

why did you not give her stirrups? could've been even lewder

>> No.949121

She got dat Persona 5, 1000-yard fish-eye stare.

>> No.949124
File: 1016 KB, 1920x1080, Wip1.jpg [View same] [iqdb] [saucenao] [google]

I'm probably gonna go nuts when I'm able to afford a pc and a gpu with more than 2gb of vram, it is quite jarring to see my poor pc struggle.
gotta love pixie I always keep her until the end game

>> No.949125
File: 2.79 MB, 1617x945, ZBrush Document.png [View same] [iqdb] [saucenao] [google]

Ended up taking that anon's advice and got rid of the crazy eyes on my model, among other small changes.

>> No.949126
File: 618 KB, 1716x1080, Lion2.jpg [View same] [iqdb] [saucenao] [google]

older version.

>> No.949127
File: 537 KB, 1920x1080, image-1.jpg [View same] [iqdb] [saucenao] [google]

and this is how the project started, I'm still working on the texturing and presentation but looking back now I'd like to think that I got at least a bit better.

>> No.949129
File: 484 KB, 1520x926, uvs.png [View same] [iqdb] [saucenao] [google]

This is how the uv's turned out.

>> No.949130
File: 1.63 MB, 1718x1080, Fix.jpg [View same] [iqdb] [saucenao] [google]

v3 vs v4.
Eyes were fixed.

>> No.949133

ive seen an imploding sub, was that yours?

>> No.949139
File: 2.10 MB, 1920x1080, garfieldinsidejob.png [View same] [iqdb] [saucenao] [google]

he can't keep getting away with it

>> No.949140

it doesn't hit the same if sam's not the perpetrator....

>> No.949228
File: 653 KB, 1447x841, JRiIBig[1].png [View same] [iqdb] [saucenao] [google]

Almost done remeshing, just got the annoying parts left. Don't know if I'll try to finish him this weekend or go animate something else to take a break.

>> No.949229

>Almost done remeshing
dios mio...

>> No.949230

That's with the subsurf on, I didn't put all of those points down. Doesn't seem that crazy to me, though I did try to surround more muscle groups with edge loops this time.

>> No.949234
File: 3.35 MB, 557x556, cream.gif [View same] [iqdb] [saucenao] [google]

ice cream

>> No.949235

I would like a iced cream

>> No.949245

turkish ice cream man movements

>> No.949271
File: 1.13 MB, 1920x1080, Untitled-1.png [View same] [iqdb] [saucenao] [google]

>> No.949272

lmao even

>> No.949282

Everything looks somewhat acceptable except the hip area is ABSOLUTELY fucked

>> No.949284

I'm sorry if it's wrong thread to ask, but is it possible to save a selection of edges and load it for similar object, or same object from previous save in 3ds Max? Maybe I'm doing something wrong, I can save the named selection set, but it appears only for the object I was working on and only on it's exact modifier, if I'll change anything it won't be there.

>> No.949285

with raycasts

>> No.949287

>mint ice cream
So this is the kind of degenerate person who does 3D as a hobby...

>> No.949289
File: 128 KB, 344x342, 3D control.webm [View same] [iqdb] [saucenao] [google]

Actually it was supposed to be blue but I screwed up the bit shifting.
Two scoops for you!
I'm still a little fuzzy on how to implement rotation properly. I can get it to rotate in x, or y, or z, but only as if they were a 2D plane relative to another variable. I've had to root the odd one out. The odd movement was supposed to just be a full spin in x and z but y got mixed in.

>> No.949313

That looks like you're getting somewhere, but you might need to look away from it for a few days and come back to it so you can see what's wrong with it. There's a lot of issues going on there.

He's trying to learn and you're laughing at him. Despicable.

>> No.949330

thanks, it's WIP and I know the hands, hips and stomach are fucked. Only had like 3 days to work on this

>> No.949359
File: 1.28 MB, 1232x920, LeslieTUrnaroundColor.webm [View same] [iqdb] [saucenao] [google]

Should I bother with fibermesh for fur and hair if I'm planning to retopo and export to blender for posing/animation? I'm kinda assuming it would be smarter to just do all of that in blender.

>> No.949362
File: 186 KB, 2560x1440, gun.jpg [View same] [iqdb] [saucenao] [google]

Back to trying my hand at guns.

>> No.949370
File: 802 KB, 1587x1033, 2023-06-25_19-16.png [View same] [iqdb] [saucenao] [google]

First time i have gotten this far with making a model

>> No.949376
File: 471 KB, 1061x860, file.png [View same] [iqdb] [saucenao] [google]

a green gnu

>> No.949391

What's Beast Boy up to this time...

>> No.949419

Looks pretty good, I hope to create a character at least this good one day. So far I haven't tried modeling anything organic at all. Did you sculpt at all for this?

>> No.949432

Thanks anon, no sculpting involved

>> No.949450

No, mine was just for fun.

>> No.949455
File: 189 KB, 608x547, kimg.png [View same] [iqdb] [saucenao] [google]

Stylized 'King" from tekken art by gopgap.
will be my 2nd completed model.
still need to add the elbow pads, and fixup the clothing from the waist down. Also the head needs more work

>> No.949458

looking cool so far

>> No.949469
File: 315 KB, 403x447, 1674661876992402.gif [View same] [iqdb] [saucenao] [google]


>> No.949475
File: 4.00 MB, 1920x1080, 1674134684066273.png [View same] [iqdb] [saucenao] [google]

how can i give it more depth? volumetric scattering? some clouds in the middle of the megastructure? adding more variations to the buildings but how do give a better sense of scale?

>> No.949476
File: 362 KB, 481x500, file.png [View same] [iqdb] [saucenao] [google]

quite cool.
the stylisation on pic-related is off-putting tho imo. maybe dial it back a little.

>> No.949492

The sides should probably be dominated more by the lats than the serratus muscles, but good work so far.

DoF and some variation would help. Less flat lighting probably would too.

>> No.949523
File: 2.94 MB, 1920x1080, 1662381155799192.png [View same] [iqdb] [saucenao] [google]

>DoF and some variation would help. Less flat lighting probably would too.

>> No.949525

Don't like the camera blur I 'd seed up the shutter at least twice. Also I wouldnt loose the sun glare. You can have both a depth haze and a nice sunlit background. Another element yo add a sense of scale is identical track with similar trains on the background so the eye can see the scale of the scene. Ir tou can have an ovegrown bit in the background it can be a good way to add scale using tiny trees. Looks promising.

>> No.949531
File: 472 KB, 542x567, asadsa.gif [View same] [iqdb] [saucenao] [google]

>> No.949532
File: 3.01 MB, 1920x1080, 1680836379706432.png [View same] [iqdb] [saucenao] [google]

>Don't like the camera blur I 'd seed up the shutter at least twice
its gonna be an animation hence the blur
>Also I wouldnt loose the sun glare. You can have both a depth haze and a nice sunlit background.
which sun glare?
>Another element yo add a sense of scale is identical track with similar trains on the background so the eye can see the scale of the scene. Ir tou can have an ovegrown bit in the background it can be a good way to add scale using tiny trees. Looks promising.
the choo choo is going all the way so by the end of the shot the scale will be shown hopefully, i thought about trees but i just want to go fully concrete look like blame!

>> No.949541
File: 2.93 MB, 1920x1080, Train test 1.jpg [View same] [iqdb] [saucenao] [google]

>which sun glare?
To compensate for the haze rendering your contrast down I would turn up the direct sun value so it adds more contrast to the lighting. Even with a haze this would definitely produce some glare in some some of the buildings and cool reflections on some of the windows n shit. See picrel.
>its gonna be an animation hence the blur
Ah ok
> i thought about trees but i just want to go fully concrete look like blame!
aH OK.

>> No.949544

that bridge can't support its own weight, not to mention a train on top.

>> No.949547

Water is wet also it's not my WIP.

>> No.949571
File: 3.07 MB, 1920x1080, 1683446054730701.png [View same] [iqdb] [saucenao] [google]

is this too much? i feel like i lost too much gloom of the atmosphere
yup working on it

>> No.949572

This reminds me of the minecart systems I would build in the Nether in Minecraft

>> No.949573
File: 2.93 MB, 1920x1080, 1675352117269744.png [View same] [iqdb] [saucenao] [google]

fuck it looks that bad?

>> No.949574

Naw I didn't mean it like that. I don't think I have any surviving examples of any high-effort Nether railways I've built, but the cavernous 3d noise environment with the long railways and smoggy atmosphere brought back some memories.

>> No.949580
File: 1.42 MB, 1920x1080, glasses.png [View same] [iqdb] [saucenao] [google]

Can someone give me an idea for a simple scene to render?
I wanna throw this character into a proper full setting, maybe do a short animation, work on my render and composition skills mainly.

Criticisms on the character welcome as well.

>> No.949584

Nice little french coffee shop exterior. Some pastels might work well with the light colors on the character.

>> No.949586
File: 944 KB, 1920x1080, Untitled.jpg [View same] [iqdb] [saucenao] [google]

she's shoppin for a hat :3

>> No.949587

also, the skirt needs work. it looks too structured. like it's made out of paper or card.

>> No.949590
File: 249 KB, 1143x692, aquarium.png [View same] [iqdb] [saucenao] [google]

she looks like she's curiously looking up into something, so I instantly thought of an aquarium
I reckon it could be a decent render challenge - a bunch of caustics all over the place, maybe throw in some neon tetras and bubbles too

like other anon said - the skirt could look a little less stiff, but overall she looks pretty good!
did you make that from scratch? I wouldn't mind seeing the wireframe

>> No.949601

Personally I like it much more now. Good work anon irs getting there.

>> No.949628
File: 1.84 MB, 1916x1008, JuneWip.jpg [View same] [iqdb] [saucenao] [google]

Bit of progress on the textures.

>> No.949649

Hehe, the seal is cute.

>> No.949653
File: 7 KB, 194x259, download.jpg [View same] [iqdb] [saucenao] [google]

Looks cool, nice rocks. Are you putting them on an asset store? People will pay for rocks, I've seen it.

>> No.949692

But what if it's a Manatee??

>> No.949776

There was a solid span of like 6-7 years where I just forgot that Manatees were real.
They're just so forgettable. Like an animal made up by a child or something.

>> No.949956
File: 504 KB, 1071x909, file.png [View same] [iqdb] [saucenao] [google]

more green gnu

>> No.949957
File: 2.83 MB, 1280x720, 2023-06-30 22-58-58.webm [View same] [iqdb] [saucenao] [google]

he he

>> No.949961

Looks good anon

>> No.949986

Hypnosis device?

>> No.949993

stared at this for slightly too long and now I'm a sissy

>> No.950001

Looking forward to cumming to the final product.

>> No.950013


Based, I'll probably post the finished prop to sketchfab and such when my animation's done so other degenerates can use it. Need to tweak a few things with the spiral and give it a nicer texture still.

>> No.950016
File: 25 KB, 360x360, hypno.jpg [View same] [iqdb] [saucenao] [google]

This texture would work better imo

>> No.950017
File: 2.41 MB, 1280x720, 2023-07-01 14-31-24.webm [View same] [iqdb] [saucenao] [google]

v1 finished.
v2 needs the following
> procedural inside walls
> procedural staircases

>> No.950018

real good



>> No.950020
File: 955 KB, 753x707, 1652786052394.png [View same] [iqdb] [saucenao] [google]

> boring
I agree mate, it looks unimpressive as fuck but it's the first iteration.

>> No.950026

It's a procedural material, I just need to play with the math to make the center of the spiral "tighter" than the outside with some exponential falloff thing. Not in a big hurry, can do that any time before the final render.

>> No.950038

Impressive, very nice.
I'm always a sucker for procedural buildings. Cities are a right pain in the ass to do by hand.

>> No.950048

Cool stuff anon. I recently grabbed Houdini, mostly for FX stuff but I am also interested in its value for procedural modeling. Any tutorials you'd recommend?

>> No.950064

Thanks man. If you don't need interior, Houdini has released a building generator in its tools. It's very easy to use if you already know Houdini
Thanks. Ngl, the best ones are on their website. If you're already familiar with the software, it's the best.
The good thing about them is that they have a varierty of tutorials that covers all lot of different stuff you can do with it. So you can come up later with your own stuff by mixing all of that. Check the project Titan videos for example.
Also watch some of their talks and webinars. They have industry (and indie) guys explaining how they use proceduralism in their workflow (or other stuff since you're also into VFX).
I have like a ton of downloaded courses and tuts in my hard drive that I don't even watch because I rely on their website. I'm a beginer too.

>> No.950096
File: 1.46 MB, 2552x1038, gfdfdfgdaa.png [View same] [iqdb] [saucenao] [google]

>> No.950097
File: 1.08 MB, 2552x824, gfdfdfgdaa2.png [View same] [iqdb] [saucenao] [google]


>> No.950101

Impressive. How do you make something like this?
Python scripting?

>> No.950134

Thanks m8. No it's more of a mix of the node system in Houdini (visual scripting) and VEX code (the "programin language" of Houdini)).
Altough I think it's definitely possible to do it in Python as well.

>> No.950165

how did your roblox grift end up working out

>> No.950233
File: 264 KB, 940x667, oozie 9milimida.png [View same] [iqdb] [saucenao] [google]


>> No.950234

all you did was bunch of extremely basic logic. This wouldn't even be good for 20 years ago.

>> No.950238

take your insecurities elsewhere and leave the lad alone, weirdo.

>> No.950253

I always wonder what pushes people to act like that

>> No.950254
File: 432 KB, 887x847, Screenshot_2023-07-03_18-22-13.png [View same] [iqdb] [saucenao] [google]

Opinions on the design?

>> No.950303
File: 1.93 MB, 1920x1080, file.png [View same] [iqdb] [saucenao] [google]

>> No.950314

Drag the toes in together, the middle/base parts of them will probably have to clip some but it's worth it to make the tips shape together properly.

>> No.950318
File: 975 KB, 1204x675, file.png [View same] [iqdb] [saucenao] [google]

reanimating the table scene for ballad of buster scruggs

>> No.950481
File: 460 KB, 2560x1440, gunt5.jpg [View same] [iqdb] [saucenao] [google]

making progress of sorts

>> No.950482

Very nice work

>> No.950483

Why do I see so many people modeling guns lately? That seems like something where there's probably already a large supply of usable stock models. Are you doing something in particular with them that needs new ones?

>> No.950485

Maybe, but I'm mostly making these to help myself get used to Blender and Substance Painter. I don't really "do" anything with the stuff I make.

Modeling stuff is fun and guns offer an interesting challenge. Decent variety of shapes, different parts that need to fit together just right, plenty of available reference images and all that.

They're also just right for me in terms of project size. Anything bigger and I risk losing interest. My last project was quite a bit bigger and just about killed me, so back to guns it is.

>> No.950517

This isn't relevant to anything going on, but ages and ages ago I asked if anyone knew what the key was to bring the menu back up after you added an object and then clicked off it before choosing everything you wanted and accidentally closed the menu.
For example you add a cylinder, give it 22 verts when you meant to give it 12, then click somewhere and closed the menu and now you can't change the number of verts without deleting some edges or deleting the cylinder and adding it back in.
At the time I got told it couldn't be done, but I knew it could because I had seen it on a video it was just impossible for me to find the video again. Turns out I've just found it again by complete accident and the key is f9. It works for any closed menu not just after adding objects, but it'll only open the last menu closed. So if you added the cylinder, then closed the menu, and then resized the cylinder, it would only bring up the resize menu.

I'm sure now a lot of people will know exactly what I was talking about, but I'm going to say that a lot of people won't know about it either and it's a really handy little thing to know.

>> No.950537

I'm almost certain someone told you to press f9, and either you or someone else went on a schizo tirade about how that wasn't what you were talking about and Max can do it at all times and that Blender is a piece of shit because it can't. All the while people are telling you what button to press.
If it wasn't you, it definitely was someone. Like a year or so ago.

>> No.950541
File: 1.50 MB, 3840x2160, untitled-logo.jpg [View same] [iqdb] [saucenao] [google]

Making a scene of russian convoy getting stuck/abandoned.

>> No.950554

Nah definitely wasn't me. I don't know anything about Max neither so it must have been someone else.

>> No.950563

It is pretty retarded that you can't access that menu after you've moved the object or done anything else to it, though.

>> No.950568

Technically its ok although contrast does not let you see the subjects (tanks). The framing could be better it would appear that the protagonist of the image is the ditch.

>> No.950569

Looks like a realistic diorama because of the scaling, if that's what you're going for.

>> No.950581

Fuck, right the scaling is off. Somehow didn't notice it myself. Thanks!

>> No.950623

I agree, but apart from it disappearing when you do an action like move or scale (or any of the options that shows up in f9 menu), I kind of understand why you can't. It'd be weird to do all sorts of shit to a model that used to be a cylinder and then change its verts from 32 to 6. That would cause all sorts of shit.
It's not quite the same, but geo nodes kind of does that now.

>> No.950650
File: 155 KB, 2125x1304, Screenshot 2023-07-07 011510.jpg [View same] [iqdb] [saucenao] [google]

Opinions? Red liners? CC please :)
I'll be moving to retopology soon but still need to finish the feet.

>> No.950655

Limbs are too skinny. Mainly forearms, hands, calf, and lower thigh. And the arms might be a tad bit too long

>> No.950742
File: 80 KB, 673x523, the undying.png [View same] [iqdb] [saucenao] [google]

More guns, this time the fabled 1911, possibly my favourite handgun. It's funny that in many videogames it's a starter gun when in reality it hits like a truck, only downside being the small magazine capacity so you gotta make those shots count, not that your foe will survive more than 1 if unarmored.

This one is also low-poly at 1800 faces
>inb4 "that isn't low poly at all!"
Understandable reaction, but considering a gun is right up in the player's face it's arguably the only kind of item worth spending a good couple thousand faces on. Older games like Modern Warfare 2 (2009) optimized their games by making 2 versions of the same gun: 1 for the 1st person player model, and 1 for the guns you see NPCs carrying around. I was mindblown when I found that out

>> No.950743


>> No.950806

with things like this it's hard to critique, human proportions are all over the place in real life, the only thing on your sculpt that feels off is the neck - it looks a little too craned forward...

I'm on my own anatomy journey I have yet to settle on an understanding of what an average human looks like. I always end up staring at my model thinking something looks off

also post your topology when you can, I'm getting pretty alright at retopo so I may be able to give some actually useful input

>> No.950819

IMO the major errors are
>why is the space below the stomach fat so smooth
>the insertions for several points are either missing definition or are wrong (biceps/quads, butt, etc.)
>connection of hand mesh and arm mesh does not look like a wrist
>hips should have more bony definition on the forward side and come out more
Won't go into the face since I think that's more subjective in terms of what works and what doesn't for 3D models but I do think you should try doing a face study where you go for realism sometime.

>> No.950834

very nice, i like it

not everything needs to be for a product anon, this kind of thinking is slopmaxxing

>> No.950835

holy shit major nerd neck

>> No.950889

Very cool. I'm curious as to how you approached the modeling of the city - did you make everything by hand or create modular parts to duplicate and scatter?

>> No.950909

>not everything needs to be for a product
Guns themselves are already soulless products so I just don't really see what anyone would get out of practicing making them when it's not even a skill with a lot of demand. I'm just wondering why so many more people are doing them than other products like, I don't know, a complex office chair design, or a car engine, or anything else along those lines.

>> No.950911
File: 192 KB, 734x1329, sp.jpg [View same] [iqdb] [saucenao] [google]

lol hope you guys don't sperg out on like this on social media, he finally got kicked out for this post.
People made fun of him afterwards, I feel kinda bad for the guy.

>> No.950914
File: 2.25 MB, 1578x1114, HSS.jpg [View same] [iqdb] [saucenao] [google]

Finally took one of my models into Unreal, do you guys think this looks ugly?
My mind is warped and I really don't know if this is up to standards in terms of quality.

I kinda would like to apply for a job soon, do you think a couple of projects with this quality will do? plus some sculpts.

>> No.950915
File: 2.35 MB, 1215x1600, HighresScreenshotUE.jpg [View same] [iqdb] [saucenao] [google]

I don't know much about lighting and color theory so I'm going by eye mostly.

>> No.950917

This still has some weirdness going on with the ao but I'm working on fixing that

>> No.950927
File: 2.31 MB, 1076x1530, HSS_2.jpg [View same] [iqdb] [saucenao] [google]

ok fixed.
here's a better shot.

>> No.950936

Do you live in some sort of parallel universe where the most popular games are Grand Theft Armrest and Call of Carburetor?

>> No.950942

people tend to practice and produce stuff they like or find fun, same reason so many artists are motivated to make cute girls and coomer shit.
not me though I get off on sleek office chair designs and intricate architectural details

that's some good shit right there, 10/10 would hang on wall
did you use a specific reference? cos I'd be even more impressed if you just came up with that out of thin air.
I've no idea what jobs are like at the moment, but I reckon that'd be a good portfolio piece regardless
you got a wireframe pic handy?

>> No.950944

I think it looks fine, but the procedural scratches on the eyes are kinda bothering me

>> No.950952

ah now i understand, you must have autism
that would explain the inability to understand others points of view

>> No.950976
File: 3.19 MB, 2560x1440, testrenderWIP.jpg [View same] [iqdb] [saucenao] [google]

yeah, I'm doing a dumb texturing tutorial. is it worth putting something like this in a portfolio or should I keep it as practice?

>> No.950977

Mentors I've talked to have always said to only publish your own work, stuff like megascans is fine if it's minor stuff like set dressing visible through the window of an interior environment.

>> No.950984

>did you make everything by hand or create modular parts to duplicate and scatter?
created a variety of different buildings for each "structure" then used the blender particle system to scatter them in a vertical way, i feel like it can look better if i figured out how to use geonodes.

>> No.950985

real cool, how do you even uv unwrap something like a mask to even texture it?

>> No.950989

Thank you guys, I'll work on it.

I've never actually retopologized a sculpt before. I always sculpt, get very autistic with the details, burn out and move to a different project.
This one I want to take all the way and learn all the aspects of making a full usable rigged model no matter how bad but aesthetics matter to me a lot.

>> No.950999

if you're as autistic about perfection as I am - then I'd advise you to just learn manual retopo and just stick with that.
I've given basically every automatic retopo tool a go and the inconsistency of the results are just never good enough for me, it's just not there yet, even with perfect settings and guide curves and all that shit I still ended up scrapping everything in the end.
Now my approach is manually dragging verts and loops around to exactly where I need them.

Don't get me wrong, sometimes the results are passable, but I'd only ever use them for non-deforming organic shit, like a rock or tree stump, never a character though.

>> No.951008
File: 2.99 MB, 2696x904, shadingconcept.png [View same] [iqdb] [saucenao] [google]

been trying to figure out how to approach texturing and shading for a game, painting normals is very fun until i move the lights around and realize im a dumbass

>> No.951013
File: 237 KB, 914x976, Untitled.png [View same] [iqdb] [saucenao] [google]

now sure how to proceed here.

>> No.951015

That looks fuckin sick mate! I'm curious to see how it looks with the lights moved around like you said.
I've always wanted to make a model that uses painted normals ever since I watched that guilty gear artstyle GDC talk, but I've no idea how to even begin to approach it. Are you painting the normal map texture, or painting vertex weights that influence the normals? I've heard there's a few ways of doing it but they all sound equally insane...

>extrude arms
>extrude legs
>try to make a head
>realize box-modelling characters is a nightmare unless you know what you're doing
If you're an absolute beginner I'd give sculpting a go, it's a lot more fun way of learning character modelling/anatomy imo, you still have to retopologize after though.

>> No.951016

yeah this seems way too difficult. i tried to model the legs separately and link them but i have no clue how to get the edge flow going properly. i'll try to make something in zbrush then.

>> No.951027
File: 2.48 MB, 1708x1310, shadingconcept2.png [View same] [iqdb] [saucenao] [google]

since i'm trying to both keep the poligon count low and be able to paint small details into the textures i'm settling for just painting the normals.
For now the texturing is very sloppy and quick mostly to see if it would even work at all, also if i stick to this i probably shouldn't mirror the uv's when i do a "clean" version

>> No.951030

Give it boobs. Really big and saggy boobs that flop around and squirt milk everywhere.

>> No.951031

guess its gonna be just practice then

>> No.951063
File: 2.06 MB, 1920x1080, untitled.png [View same] [iqdb] [saucenao] [google]

babies first scene holy shit 2d is easier

>> No.951069

Looks awesome, but I agree with the other guy, scratches on the eyes don't make much sense to me. It doesn't really look bad at a glance but if you want people to look at it closely in a portfolio that could stand out.

>> No.951071
File: 3.15 MB, 2744x1816, yuuka.png [View same] [iqdb] [saucenao] [google]

>> No.951078
File: 673 KB, 1920x1080, __Kame House.jpg [View same] [iqdb] [saucenao] [google]

trying to make kame house based on DBZ Kakarot, no idea how to make the island yet, but i'll get there.. hopefuly

>> No.951079

That looks pretty cool anon

>> No.951080
File: 251 KB, 2000x2000, 1680550101588494.jpg [View same] [iqdb] [saucenao] [google]

WIP. Gotta refine the eyelashes and get to work on retopoing the base mesh. Something about the hair is bugging me. This is my second time ever making hair so I'd appreciate some critiques. Thanks.

>> No.951100

We need more 2hu

>> No.951134
File: 258 KB, 450x450, 1666652910955253.gif [View same] [iqdb] [saucenao] [google]

We do

>> No.951176

way too dark breh, I can't see a thing

>> No.951268
File: 950 KB, 1080x1080, wine glass.png [View same] [iqdb] [saucenao] [google]

Babby's first presentable project? Modelled in Plasticity rendered with Cycles.

>> No.951272

>first presentable project
Nope its bad.. Not being mean. It has the typical noob error of geometry artifacts in the bottom part of the cup.
There can be two causes. Non smoothed shading or bad cylinder capping.

I suppose you are still not familiar with the concept of cylindrical capping or surface smoothing so lets use this opportunity for you to learn a good modelling principle and no deviate like so many faggots.

Post a mesh of the cup from the bottom and we'll go from there.

>> No.951274
File: 1.87 MB, 2409x1086, study.png [View same] [iqdb] [saucenao] [google]

Again with fixed artifacts.

>It has the typical noob error of geometry artifacts in the bottom part of the cup.
I'm not so certain; this was set shade smooth and with subdiv. While the pole is a bit naughty and that tight band of loop cuts is dumb (wtf plasticity?), the rest of the geo looks fine, right? Not sure how relevant, but I fixed the weird shadows with adaptive subdiv. Given https://www.reddit.com/r/blenderhelp/comments/11mna1e/wireframe_artefacts_with_shadow_caustics_any_idea/ , it might be a blender bug. Will check real quick.

>> No.951276
File: 2.38 MB, 2515x948, study 2.png [View same] [iqdb] [saucenao] [google]

Some more information. Maybe I'm retarded. Couldn't recreate the redditor's issue. The rest is trying again with a new exported mesh, this time with cleaner geometry from setting a minimum spacing. Are the normals fucked? It can't be that this is because of the pole topology?? Again, setting to adaptive subdiv fixes the issue.

>> No.951278
File: 86 KB, 1480x800, howtoavoidatriangularnodeforidiots.jpg [View same] [iqdb] [saucenao] [google]

>I'm not so certain
>this was set shade smooth and with subdiv
>While the pole is a bit naughty and that tight band of loop cuts is dumb (wtf plasticity?), the rest of the geo looks fine, right?
> but I fixed the weird shadows with adaptive subdiv.
>it might be a blender bug.
Ok I give up. You just admitted to have a pole in your geometry and the only way you are solving it is with aggressive subdivision and then think that even if you pressed some check buttons that might just be a bug of the program cuz your reasoning must be flawless.
>Not sure how relevant
Yeah you also can just take a piece of crap and dorodango the shit out of it until you get a smooth surface. Do you even know what poles cause to refraction? Did you even bother to look up the terms cylindrical capping and shader smoothing?

>> No.951279

>It can't be that this is because of the pole topology?? Again, setting to adaptive subdiv fixes the issue.
>Oh is the bad geometry the cause of the problem? No it cna't be I just used a tool with a name that has the word 'adaptive' so it must solve the problem for me.
YES DUMMIE! The problem is in your triangular pole. And even if that cup is just a practise object the goal of practicing is bettering your work not pressing buttons to hide problems.
For the love of god do the pole again with a correctly done cylindrical cap and then learn in a practical way what it does to your geometry.

>> No.951285
File: 2.03 MB, 1842x1083, study 3.png [View same] [iqdb] [saucenao] [google]

My understanding was that poles only distort shading in their *proximity* due to the difficulties in calculating local gradients, whereas the shading errors appear on both sides of the foot as well as on the sides of the bowl - quite distant from the poles.
>dorodango the shit out of it until you get a smooth surface
>tool with a name that has the word 'adaptive' so it must solve the problem for me.
My understanding is that topology poles *cannot* be fixed by subdivision, which is why they're annoying. I pointed out that adaptive subdivision fixes it because this is unexpected according to that understanding.
>And even if that cup is just a practise object the goal of practicing is bettering your work not pressing buttons to hide problems.
The goal was to learn plasticity and cycles. Plasticity is CAD so getting a mesh out of it necessarily involves pressing buttons. I'm learning plasticity because polymodelling is a swamp filled with exactly this kind of autism.
>Do you even know what poles cause to refraction? Did you even bother to look up the terms cylindrical capping and shader smoothing?
I honestly thought I knew what those were, but maybe youtube tutorials aren't reliable here.

Since I honestly do want to learn, here's two renders. One with the shown grid fill solution to the pole, and another with the poles straight-up deleted. It looks like there's still errors in both the quad solution and the cap-less render, so is it that "poles" in this case isn't referring to the actual vertex, but the edge flow around it? But then how does this differ from pic in >>951278 for the foot? Does it have to do with the weird topology around the base of the bowl?

>> No.951289
File: 214 KB, 937x411, 57981311.png [View same] [iqdb] [saucenao] [google]

Appears to be a known limitation of the MNEE caustics solver. https://archive.blender.org/developer/D13533

>> No.951294

>Since I honestly do want to learn, here's two renders.
Ok nice insight I'm calm now. As for the geometry it doesn't matter if you have capped and used quads to finish the bottom of the cup. If that circle is too small it will still behave like a pole. It's not about getting quads all around its about eliminating stress point on the geometry that cause differential mesh densities in a surface In picrel 'study3' yeah youvé capped the cyilinder but it's still too small so the surrounding geomtery still has a triangular shape that will definietly generate a diferential mesh density between the center and the border of the base.
Meshes should be equally dense everywhere specially on objects that need it so like translucent or refractive properties.
>But then how does this differ from pic in >>951278 (You) for the foot?
The capping should maintain the flow of curvature you intend to make so it doesn't end up like a flat center. I always support having as little gase geometry as possible befor smoothing because contrary to logic less geometry will give you more control in some situations.
>>951278 Picrel check the quad mark but most importantly it gives control of the whole mesh base center using a single vertex whereas in your study you've backed yourself into a corner trying to control a center with two polygons in the capping. Either you have complete symetry or you don't so the cup should have 'symetrical' subdivisions. This is why most cylinder meshes ar multiples of 2 and 3. Making a cyilinder with 3,6,12,18,24 sides is much better that your 14 sides.
That is why in >>951278 picrel the mesh transitions so smoothly from 8 sides to 4 sides and from 4 sides to a single point.
Ok. Not familiar with blender but doesn't surprise me that they can't do caustics. But I've never talked about caustics. The artifacts I still see in 'study3'I would bet good money that they are not in the caustics but in the refraction.

>> No.951296
File: 191 KB, 892x802, 1682835315637656.png [View same] [iqdb] [saucenao] [google]

almost done with this chair some nuts and bolts are left but other than that pretty much done, what would be a good object to practice modelling?

>> No.951299
File: 382 KB, 659x668, This.png [View same] [iqdb] [saucenao] [google]

If I had to say, I find this part a bit weird, maybe try putting the side hair strand on top of the front one or something like that. It might be the angle too, but that cheek is looking a bit weird to me.
Dont take my advice too seriously as I'm also a beginner at this, keep the good work, anon.

>> No.951302

It's not your fault. Use a real renderer (one that actually works) instead of Cycles and everything will be fine.
And I don't know if you've noticed but the resident topology experts have no clue what topology is and/or how it works.

>> No.951304

I still can't see what's wrong with the glass.
And to be honest if you hadn't told me there was some sort of apparent, glaring, horrible issue with the poles, I never would've known to begin with.
It's a perfectly fine, if entirely boring and mundane, glass composition.

>> No.951305

The only issue I can see with your picture is that the shadow of the glass as rendered by Cycles is possibly the ugliest thing I've ever seen.

>> No.951306

It's not my picture, I'm not that guy.

>> No.951307
File: 52 KB, 498x280, chairs.jpg [View same] [iqdb] [saucenao] [google]

Omfg a chair. I know a thing or two about chairs so I have a few comments.
1. The wheels can be a diferent diameter than the central axis support but in chair design the best choice is to have the same radius for the axis cover. It lets the chair be safe around corners and it catches less dirt on daily use.
2. The hydraulic cylinder must have a receptacle at least as long as the shaft.
3.Armrest seems a bit too thick to have that sharpness. If you are going for a rigid foam arm rest those are usually more curved in the to to prevent localized wear on the front portion.

Otherwise great chair. I love chairs. If you want to post more chairs feel free to post them I love chairs.

If you want the next level in chairs you could try and model a Chesterfield Arm Chair or an LC2. both use extensive leather and could be the opportunity to be an intro to wrinkles and textile sim.

>> No.951308
File: 398 KB, 1156x685, image_2023-07-12_155833630.png [View same] [iqdb] [saucenao] [google]

I've been sculpting a boss from a game but I might need to shrink the human part since the battle sprits are chibish.
Does anyone have any good recourses for baking normal maps from sculpts?

>> No.951310

Follow this.
Never bake high poly sculpts using Substance Painter. Marmoset is the way.

>> No.951317

How did you find your way into a Netflix show? Do you have connections?

>> No.951353
File: 689 KB, 2160x2880, 1674213167210063.png [View same] [iqdb] [saucenao] [google]

>2. The hydraulic cylinder must have a receptacle at least as long as the shaft.
do people go into that much detail for modelling for film and animation? or is it for bim? i usually just look at the reference and do by best approximation for the shapes.
fixing the armrest as well
>If you want the next level in chairs you could try and model a Chesterfield Arm Chair or an LC2. both use extensive leather and could be the opportunity to be an intro to wrinkles and textile sim.
i have already tried a chesterfield arm but i wan to practice the curve parts of models in the sense the back rest of the chair is very hard to model for me, i make extrusions for the front and when i switch to the side view its hard for me to make the curve any tips on that? is it just moving the verticies and hoping the shape comes out right?

>> No.951410
File: 1.69 MB, 2098x846, ilovechairssomuchitsretarded.png [View same] [iqdb] [saucenao] [google]

I'm rich so I pay my own Netflix.
>do people go into that much detail for modelling for film and animation?
Not if the style is more cartoony but take a look at your picrel, the piston it traverses the wheel base and almost reaches the floor. It's a common design feature of office chairs that lets you fit a reclining mechanism.
> switch to the side view its hard for me to make the curve any tips on that?
I'd like to see a wireframe but it seems you have far too much base geometry before the smoothing part. It generally gives you less control because you have to manage more vertices.
> is it just moving the verticies and hoping the shape comes out right?
In practice yes but in theory you should treat vertices almost like beziers and define the curve by its smoothing result. When I model chairs I usually make the whole chair in as little polys as I can to get the overall shape and the I start detailing the mesh where it really needs it.

>> No.951417
File: 2.25 MB, 1000x836, derchairen.webm [View same] [iqdb] [saucenao] [google]

Are you going to texture it?

>> No.951429
File: 361 KB, 1587x855, 1685992035790634.png [View same] [iqdb] [saucenao] [google]

it around 270k tris with smoothing, i can cut a lot of it down especially in the legs which looks really dense with sub div on
not right now, i just want to practice modelling and get that grind on. been thinking about switching to 3dsmax for vray since majority of the jobs in my country uses 3ds and rarely blender pops up.

>> No.951452
File: 635 KB, 1280x720, 0001-0200.webm [View same] [iqdb] [saucenao] [google]

trying to figure out how to rig the back, however i feel like no matter how much i work on it the lines on the character will make it obvious that all the shit is just painted on, i might try blending different textures and see if it could work

>> No.951454
File: 3.25 MB, 1736x844, Dontbeavertexpusher2(2).webm [View same] [iqdb] [saucenao] [google]

Yep. Your modelling suffers from a disease that affects 99% of hard surface modellers when they start modelling. Don''t worry. I had it and it and it has a cure.
I call it 'vertex pushing' and its a consequence of starting your model with garbage geometry that you don't actually need.
For example in your backrest i see a lot of geometry that you must've started using a subdivided plane or a box and you've spent hours trying to drag rows and rows of vertices in order to achieve the curve you wanted. In poly modelling you should always remember: 'Less geometry will give you more control' and 'Only add geometry where its absolutely necessary'.

I think it was better to make a video so you can see what I mean by that.

In the example of bad modelling you can see that I try to imitate what many modellers do and its trying to drag and push rows and rows of vertices witch is not only a bad practice but a horribly inefficient one.
In the example of good modelling you can see that I try to only make the geometry that I need to control the curved surfaces and that I use symmetry and smoothing modifiers to my advantage. Why would y want to drag 100 vertices when the smoothing modifier does that for me?
I added a better example that tries to eliminate the use of loops during the modelling process. Loops are obviously necessary to control the edges in your geometry but if you add them too soon you'll loose the curvature of those edges if you need to modify the overall shape so you should consider advanced smoothing options like Open subdiv in max or trying to stack loops into separate modifiers.

I kid you not if you model efficiently you could model chair like that the Herman Miller Lino in less than one hour and use it as a prop for a photo realistic render.

>> No.951461

holy shit thanks anon, we need more anons like you, what did you change in the better modelling section? it was too fast and compressed to notice what you did differently. What would you recommend to practice? i was thinking of modelling a shoe next

>> No.951462

That's a really nice shoulder rig.
And yeah, I think blending textures might work, but it'll definitely make things more complicated. Especially if you're supporting the whole movement, you've got 360 degrees of rotation for the arms going around, and then whatever the range of motion is for the arms going forward.
I guess you could maybe do a texture for like 6 degrees of motion for the 360 and blend those, and then like 2-4 for the forward motion. In T-pose the 360 and forward would share a texture, so you would only have to do that once.
If done right though it could look neat.

>> No.951464

> what did you change in the better modelling section?
Fuck I just notices I recorded the window but not the full program. What I changed was that I erased the edge loops and changed the turbo smooth for an open subdiv modfier and played with the weight of the edges. This lets you get the shape correctly before commiting to a loop that will complicate the modelling process.
>What would you recommend to practice?
A show seems pretty good practice.It has hard parts like the sole. Soft parts and the laces will be a good insight into parametric modelling.

>> No.951465

>what did you change in the better modelling section?
He used a base mesh that had the bare minimum of polygons required to get the desired shape. Allows you to just slap on subdivisions and then control the shape of the final mesh by manipulating that base mesh.

Like he said, fewer polygons = more control. Let the subdivisions do the heavy lifting.

>> No.951466

Yes this.

>> No.951497

>played with the weight of the edges.
i assume this is same as edge creasing in blender , thanks once again anon
i will give it a try and be conscious of the extrusion i make

>> No.951540
File: 177 KB, 1016x774, 1684591257471003.png [View same] [iqdb] [saucenao] [google]

is this too much for the base? how do i figure out the right amount of base low poly edge loops before i sub divdide ?

>> No.951542

Yeah that’s enough, if I were you I’d just sculpt the shoe in zbrush and retopo and extract normals/disp (or vector disp if you use renderman) and donezo

>> No.951543

i want to practice box/poly modelling right now, picking up sculpting and a new program will be pretty dumb without my modelling fundies right?

>> No.951553

Because I'm and old fag that never picked up sculpting I'd do it all on poly and I'ts not hard. Give me the blueprint and I could show you how. I agree with anon that the best way is to sculpthat but you might learn a thing or two on detailed poly modelling.

>> No.951554

this is what i am using
i am going crazy anon, its hard to find good references that match up, the sole view is not aligning with the top view so its hard to guide vertices.

>> No.951556

Ok I'll give it a go this weekend. Never modelled a shoe but with good thechniques you can model anything.

>> No.951558

>i am going crazy anon, its hard to find good references that match up, the sole view
I used to work in a big corporate interior design firm as the lead 3d modeller. I had to model pieces of furniture with nothing more than a small picture taken from a cellphone. That will be a good part of your work so better get used to lacking refs. Cheers.

>> No.951564

you need help with texturing?
I'm practicing anime style texture

>> No.951566

need folds and wrinkles

>> No.951567

Hey yeah sorry, this is a good thing to do for practice, when I was @ ESMA we did this in my first year and that really forced me to step my hard surface skills up (my stupid ass decided to model a pair of timberlands, needless to say I suffered lol).

>> No.951568

References shouldn’t be your sole guide, 1:1 replica models are usually done by LIDAR because it’s extremely autistic to do it by hand since we’re not engineers looking to print or model structural stress in CAD or some shit. Try to get it to look as good as possible, what I usually do is use maya’s shit sculpting tools on a very Low poly mesh that I subdivided and gently nudging things where I want them once a base mesh is completed, trust me anon, most people can’t see the difference and the texturing done after usually sells the deal. Also listen to the chair guy he knows his shit too

>> No.951570
File: 146 KB, 1573x865, 1689090153164906.png [View same] [iqdb] [saucenao] [google]

thanks anons, so far its 20k tris and i thought after i finish modelling i will use a lattice and get that curved overall shape because bending the shape during modelling is apparently too hard for my brain to comprehend

>> No.951571
File: 292 KB, 1582x860, 1671503646967350.png [View same] [iqdb] [saucenao] [google]

le wireframe

>> No.951573

You keep topology down which is good, that Nike logo part is too much though, I reckon you could reduce twofold before you start curving, will make it more manageable imo
Also I’m not sure a lattice might work here, even if you use a very dense one you might still have issues with the thickness of your mesh and different parts of it spaghettifying or something, it might work though but prepare yourself for trouble.
Keep us posted I wanna see the end result as I don’t have access to my computer sadly :(

>> No.951575

>most people can’t see the difference and the texturing done after usually sells the deal.
Best advice. Don't model anything you are not supposed to. In 99% of the cases that sole is better off as a normal texture or a sculpted mesh. I'm just going on 100% poly modelling because you want to polish skills.
>extremely autistic to do it by hand
It is.

>> No.951577

I would still add a perimetral loop just because you'll need to control the sole profile shen smoothing.
Take a look a this one: https://www.blendernation.com/2020/04/08/behind-the-scenes-nike-shoe-model/

>> No.951599
File: 305 KB, 671x390, image_2023-07-14_133822538.png [View same] [iqdb] [saucenao] [google]

forced perspective is fuckin hard

>> No.951600

>forced perspective is fuckin hard
What do you want to achieve?

>> No.951601
File: 626 KB, 1280x1400, 286246447_606598910478040_374100329010761762_n.jpg [View same] [iqdb] [saucenao] [google]

I'm trying to recreate pic related with some adjustments, including a greater distance between the house and the riverbed, aiming for a pre-rendered CG aesthetic (thus the shit texturing and flat lighting)

>> No.951603
File: 1.57 MB, 2083x843, cameralucida.jpg [View same] [iqdb] [saucenao] [google]

Yep the reference ok. I'd bet good money that it was painted using a camera lucida wich tipically has a moderate zoom effect.
I've calculated the fov and it should be around 33 degrees.
Since you are working in blender try this https://www.youtube.com/watch?v=vrEmdEk_O78


>> No.951604

Damn brother, thanks a ton. Really appreciate it. I'll watch the vid, but do you have any general tips on perspective matching?

>> No.951605

If you need a flat reference for that here's the real location.

>> No.951606

>do you have any general tips on perspective matching
Yes. Don't try to match the details perfectly (the usually never fit) Try to always match the bigger picture specially something prone to distortion like a painting.

>> No.951622
File: 3.69 MB, 3914x1440, 1664401589541557.jpg [View same] [iqdb] [saucenao] [google]

okay I think I'm done with it, now how do I make a decent lighting setup?

>> No.951640

Not really, im trying to get everything done by myself

>> No.951695
File: 2.93 MB, 2000x2000, 1665666205071553.webm [View same] [iqdb] [saucenao] [google]

Done :^)

>> No.951696
File: 2.54 MB, 1920x952, 1663424636044446.png [View same] [iqdb] [saucenao] [google]

how do people go about making brick walls like these? broken ones as well. is it zbrush? then retopo and bake high to low normal maps?

>> No.951699

Baby hands

>> No.951701

nice result. what tuto?

>> No.951709
File: 1.72 MB, 1637x1352, IMG_0616.jpg [View same] [iqdb] [saucenao] [google]

Made with nomad sculpt on iPad. I’m afraid that the only way I can get better is by learning how to draw, but the sole reason i picked up 3dcg was because I wasn’t improving with drawing, and sculpting seemed more intuitive to me.

>> No.951718
File: 379 KB, 852x861, isusu3.jpg [View same] [iqdb] [saucenao] [google]

Wip. Deciding on what to do next, refining more the anatomy or hair hell

>> No.951735

this one

>> No.951754
File: 514 KB, 2359x1029, bricks.jpg [View same] [iqdb] [saucenao] [google]

No fucking clue.
My guess is like you said. Take a bunch of cubes at different sizes/ratios and start sculpting. Bake normals and assemble by hand or lay them out procedurally.

I guess you could also take a trimsheet approach. Bake a bunch of normals edges and bevels of different bricks to a texture, unwrap all of your bricks the same way, and then pick and choose areas to use from the trimsheet (pic related, hopefully it makes sense).
Like say you had a texture broken up into 9 brick edges, then you have a bunch of cubes at different sizes and ratios like before (or even handle that procedurally), you unwrap each face to one of the 9 squares in the UV, and since the seams between each of the 9 squares would match up to make a bevel, they'd all match up no matter which of the squares you pick. Then just use another texture for the diffuse and whatever else, I think stretching wouldn't be too much of a problem for bevels and normals, but if you're stretching and squashing the bricks too much the diffuse texture might look a bit off. So triplanar/object coordinates would be best for the regular texture, but UV for the bevels.
Then as before, assemble the wall by hand, or use some procedural method.

I'm not sure if that's something that people do or would do, or if there's a better way, but that's probably how I'd tackle it.

>> No.951757


>> No.951777
File: 354 KB, 1958x904, fucking around.png [View same] [iqdb] [saucenao] [google]

as long as it 3d prints

>> No.951780
File: 113 KB, 960x540, 1689051090239015.jpg [View same] [iqdb] [saucenao] [google]

Sorry if you don't like the edit. It's just so I can see it.

>> No.951798

Honestly just triplanar the shit out of it, that’s usually what I do when I can’t be fucked. Works great 99% of the time especially for organic matter. I usually bake out the diffuse and correct the spec and rough maps and try to make a bump map that works ok because again, normals are just overkill if your shit is going to sit in the BG and will be comped out anyway

>> No.951830

very cool! would love to see the finish sculpt

>> No.951846
File: 389 KB, 2494x1709, Castle.jpg [View same] [iqdb] [saucenao] [google]

There's an addon that comes with Blender that lets you make brick walls. I used it to try and make a shitty castle (pic related), but I gave up on it.
I'm sure you can fuck around with it to make the blocks bigger or whatever and then you can just adjust them the way you like afterwards.

>> No.951863

Yeah obviously I would do that, but since he was asking about bevels and scuplted shit, I figured I'd offer up some kind of solution.
From a distance, I'd just throw on a bevel shader some triplanar repeating textures, and then do some extra layers for like lichen and shit to break it up. But if it were up close, I think the trimsheet thing is what I would do.
Actually, I'm thinking of picking up that solution and running with it for something of my own. Especially since really all you have to do is sculpt 1-2 bricks to get the faces/bevels you need for the texture, and then you can mix and match them for any number of bricks. And if you keep the layout consistent, you could do all sorts of different textures and swap them out. Like one texture with really distressed bevels, and another with perfect ones, then you can re-use assets. Which I guess is one of the benefits of trimsheets with consistent layouts.

>> No.951900

Never used trim sheets before but honestly I’m more and more interested in video game workflows like these for tedious and annoying shit like these

>> No.951901

What's it like living in a castle?

>> No.951902

But this is perfectly good and sensible advice?

>> No.951924
File: 654 KB, 1055x731, m48patton.png [View same] [iqdb] [saucenao] [google]

Working on a 3D Printable M48 Patton. Tracks will be separate objects from hull and connect via pins.

>> No.951928

Good luck with the turret, actually anon are you planing to model a notch to slot the turret and allow it to rotate?

>> No.951930

No wifi, no tv, no phone chargers, boiled food and crenelations. Basically a zoomer's nightmare.

>> No.951937

I always wanted to live in a castle. I wanted to buy that one from Satan's Chimney, but it turned out 95% of it was a set and it broke my heart.

>> No.951941

>actually anon are you planing to model a notch to slot the turret and allow it to rotate?
Yes, absolutely. Should be pretty easy. I might have to make some changes to the shape of the turret to make sure it clears all the details on hull. Ideally, I would like to have a gun with adjustable elevation, but I'm not sure if I have the skill to implement something like that yet.

>> No.951948
File: 13 KB, 270x313, 1245377-270px_npc_sandal.jpg [View same] [iqdb] [saucenao] [google]

The more work you can get done as a single person, the more you can do for a single project.
Video games are all about efficiency since they don't want to pay people and they want to get shit out the door real quick, so paying attention to GDC talks every now and again for novel workflows and stuff is pretty handy. Even for normal 3d work.
I think the one about the wind and vegetation in Horizon Zero Dawn was a pretty neat one that I've always been meaning to give a try, but never had a reason to. Can't say much about the game since I haven't played it (though the way the MC looks in the game always reminds me of the autistic "Enchantment!" dwarf in your camp in Dragon Age Origins), but some of the devs working on it really know their shit.

>> No.951956
File: 1.44 MB, 1080x1080, CrashFaceWeb.png [View same] [iqdb] [saucenao] [google]

New video worked up in UE5.1
I still need to improve on camera work and take recorder, some of my vehicles came out jittery until I stopped using "reduce keys"

>> No.952015

I’m sure there’s a few tutorials or courses you can rip online. Actually you could probably figure it out by drawing simple schematics. Anyway looking forward to it, if you don’t mind dropping your files in .obj once you’re done I could even texture and light it.

>> No.952018

Dude yes horizon zero dawn blew my tiny mind when I saw it run on a friends ps4, I still don’t understand how these people made it run on (relatively) ancient hardware, do you have links to videos or papers where the devs discuss their workflows?

>> No.952041
File: 50 KB, 640x480, untitled.jpg [View same] [iqdb] [saucenao] [google]

Cannot figure out how to make distant hills. They always look either flat or too close

>> No.952045
File: 85 KB, 702x504, active-negative-space-excerpt-sky-good.jpg [View same] [iqdb] [saucenao] [google]

>Uses flat shader, has flat results.

Apply the concept of foreground, middle ground, far ground and background.
Don't use 2 planes but 3. Blend them into your sky color gradually to simulate haze and focus.
Learn the concept of negative space and rule of planes.

>> No.952047

Wow trippy I love it.

>> No.952064
File: 1.54 MB, 1595x841, pls help.png [View same] [iqdb] [saucenao] [google]

Guys, i need to rename 397 tiles, any reliable and fast way to do it without loosing my mind, and wasting my life? All need to have -col at the end btw, using blender 3.4

>> No.952073
File: 505 KB, 1331x808, Screenshot_2023-07-18_20-41-59.png [View same] [iqdb] [saucenao] [google]

I have a simple object I'm unwrapping, but it turns out to be a complete mess
Any ideas?

>> No.952076

press u, and "smart uv project"

>> No.952078

also, apply transforms(rotation and scale,cntrl+a, in object mode), or try to press shift+n to recalculate normals

>> No.952080

I've tried it but nothing worked
Haven't tried that but I've already done a quick retopo and it works great now

>> No.952089

Make a script to do it. I can't tell you how, since I've got no clue how to script, but this is definitely one of those times to do it.
Though my retarded ass would probably make a keyboard macro of like "F2>end>-col>enter>down" and then just spam the macro a bunch. A script would probably be easier if you know a bit about that 2bh.

There's sites that backup different GDC talks, like this ( https://www.gdcvault.com/free ), barring that, there's the Youtube channel ( https://www.youtube.com/@Gdconf ). Some presentations they only have slides on the websites though, so there might be shit on one but not the other.

>> No.952094
File: 228 KB, 557x819, Screenshot_2023-07-18_22-30-22.png [View same] [iqdb] [saucenao] [google]

opinion on Hatsune Niku?
I still have to do the front hair

>> No.952101

She thicc!

>> No.952113
File: 190 KB, 438x634, C3QyaeqnL8.png [View same] [iqdb] [saucenao] [google]

Was still trying some things out and too lazy to model the belt, is it passable as a texture painted ontop?

>> No.952114
File: 2.54 MB, 1440x1080, nodes_lashes.png [View same] [iqdb] [saucenao] [google]

Using geometry nodes for generating procedural eye lashes.
I'm satisfied how they look from a distance but would like to have it look more real in super closeup.
I guess adding randomness will help and some logic to prevent intersection.
This doesn't have to run during animation but will be baked as a particle system hair so it doesn't have to run fast.

>> No.952123

>have it look more real in super closeup.
Have you actually looked at close up pictures of eyelashes? I'm not shitting on you, but that's where I'd start if I were focused on micro details like that.
Hair doesn't just clip out of your skin, which imo would be something to pay attention to and solve for more realistic close-ups. Darkening the area underneath where they sprout out might also go a long way.

>> No.952129
File: 475 KB, 559x719, druids.png [View same] [iqdb] [saucenao] [google]


>> No.952143
File: 1.06 MB, 3304x1812, yuuka2.jpg [View same] [iqdb] [saucenao] [google]


>> No.952172

The face ruins it all

>> No.952211

You should just mark a seam along top and bottom, no?

>> No.952263

edit -> batch rename -> set name -> suffix

>> No.952325
File: 1.09 MB, 1463x742, m48patton2.png [View same] [iqdb] [saucenao] [google]

Starting to make progress on the turret. Topo for the base shape turned out to be more complex than I thought it would be and I needed to make several attempts. There are shading issues, but I don't need it to shade well since it will be strictly for 3D printing.

>If you don’t mind dropping your files in .obj once you’re done I could even texture and light it.
I'll give that some thought. Keep in mind that almost every detail is booleaned onto a base shape, so the topo and unwrapping will be extremely messy.

>> No.952387
File: 1021 KB, 1376x717, m48patton3.png [View same] [iqdb] [saucenao] [google]

Made some progress on the commander's cupola. It's not totally accurate, but I'm pleased with the level of detail. I would say I'm about 65% done with the turret. I still want to invest a few more hours into it. Using Booleans for most of the details has been fun. It's nice to not have to worry too much about topo.

>> No.952400

Oh shit that would be a problem, I use renderman which is VERY touchy when topology isn’t good, especially with reflections

>> No.952402
File: 550 KB, 1440x1080, Cyberverse_Starscream_03.jpg [View same] [iqdb] [saucenao] [google]

>> No.952405

should loose that shiny thing.. tanks are not covered in glossy paint for obvious reasons.

>> No.952414
File: 201 KB, 1196x402, Untitled.png [View same] [iqdb] [saucenao] [google]

Made this tree with a particle system and am trying to make a small forest with it but it's causing a ton of lag. Is there a better way to go about this?

>> No.952432
File: 490 KB, 1711x718, 1662815462318909.png [View same] [iqdb] [saucenao] [google]

is this the right way of modelling a gun?

>> No.952453
File: 144 KB, 2114x1663, This area.jpg [View same] [iqdb] [saucenao] [google]

Is that gun chad around? I need some advice from him on the grip. I suppose anyone could help me really, but I know he knows how to into guns.
I can't quite figure out how to make the front of the grip (which I think is called the front strap) and behind the trigger rounded. It's square atm and when I try to make it more rounded it just gets fucked up.

>> No.952460
File: 221 KB, 1915x855, 1674939884211571.png [View same] [iqdb] [saucenao] [google]

i tried following a tutorial with hard ops and ngons without the addons and i dont get the workflow, there are so many ngons.
really would love his advice on where to get started in modelling a gun

>> No.952462
File: 127 KB, 1551x1558, Some fixture that I don't know the name of where a bolt goes.jpg [View same] [iqdb] [saucenao] [google]

I'm probably not the best one to give advice considering I'm looking for it myself, but I probably would have blocked it out and then put a subdiv on it to eliminate the ngons. Then just moved the verts to where they needed to be and added any loopcuts that I needed to add.
A bit like I've done here, for some fixture that I don't know the name of where a bolt goes, but more to your specification. I'm sure you know what I'm rambling on about.

>> No.952467

>tanks are not covered in glossy paint for obvious reasons.
I'm aware. I plan to 3D print it with somewhat glossy filament though, so it's a nice way to get an idea of what the final product might look like. Plus I find it helps me understand the shape of the surface better. Keep in mind that I don't plan or doing any sort of serious rendering with it. It's strictly for 3D printing.

>> No.952568
File: 1.25 MB, 1274x2048, Tree_Small_3_Diffuse.jpg [View same] [iqdb] [saucenao] [google]

Billboards. Put your tree as an image texture on a single plane, instead of actual geometry (pic related is one of mine, alpha texture is on the normal map instead of the diffuse).
If it's close up to the camera, use actual geometry, but if it's far, use billboards. Might help to bake a normal map for the tree texture as well to give it a bit of light interaction. Using the viewport render from the camera you make the billboard texture from, and using the normals matcap makes normals just fine without baking.

>> No.952569

should have just used plasticity. It would have taken a small fraction of the time and looked better

>> No.952571

Figures. Thanks I'll try it out!

>> No.952614

I doubt it would have taken a fraction of time, as I would need to learn the software package. I'm fairly content with the rate of progress anyway. It's for fun, not money. Once it's 3D printed, I really don't think there's going to be a substantial difference in how it looks.

>> No.952726
File: 173 KB, 2000x2000, 1671870273998347.jpg [View same] [iqdb] [saucenao] [google]

Stocking WIP

>> No.952734
File: 142 KB, 2000x2000, 1683558123021027.jpg [View same] [iqdb] [saucenao] [google]


>> No.952742
File: 604 KB, 2560x1440, tbrender55.jpg [View same] [iqdb] [saucenao] [google]

I'm the guy who made >>950481

I'm not really an expert on modeling guns, though. This is only my second attempt at one, and the main reason it looks halfway decent is that I have years of experience modeling other crap to varying states of completion.

That being said, I more or less got started doing Tim Bergholz' revolver tutorial a few months ago.

If i had just wanted to model a gun I wouldn't have used a tutorial at all, but I was also looking for an intro to Blender and Substance Painter to learn the whole game asset worklflow thing (I come from Max and used to do mainly arch viz) and it seemed like an alright choice for that. Looking back, I think I'm pretty happy with it.

After I was done with that I immediately started on my first project, which resulted in pic related. I basically just stuck to the workflow I learned in that tutorial, but gathered my own reference material.

Do not underestimate that part of the process. Having good references is critical. Use Google image search. Watch disassembly videos on Youtube. Take notes and screenshots. I spent two or three days gathering several hundred reference images, organized them in PureRef and used them to create a composite image I could use within Blender.

After doing a quick blockout to see if everything fit together I just started modeling and kept going until it sort of looked alright. I know it sounds dumb and probably isn't very helpful if you don't have that much experience, but in the end that's how you really learn. Gotta bang your head against that wall.

>> No.952743
File: 969 KB, 3000x3000, ref2.jpg [View same] [iqdb] [saucenao] [google]

FYI, this is what my reference image in Blender looked like. If the one in that screenshot is all you have it probably won't get you very far.

Try to get shots from multiple angles and adjust them in Photoshop so they match up in terms of position and size.

>> No.952794

damn you modeled the interiors as well?

>> No.952816
File: 198 KB, 2560x1440, yya.jpg [View same] [iqdb] [saucenao] [google]

It's mostly just the parts that would be visible from the outside. I did model some parts that you wouldn't see without disassembling the gun, but that was mostly because that made it easier to check if all the moving parts fit right in terms of size and position.

Based on the stuff I've seen since then I may have gone a little overboard there, but it was my first attempt so I tried to play it safe.

>> No.952834
File: 141 KB, 1577x827, 1669595049852197.png [View same] [iqdb] [saucenao] [google]

very fucking cool dude, i just blocked the 3 main pieces and started modelling it, maybe i should break it down further but i want focus more on environments so not sure if i have model it even more pieces

>> No.952843

You'll have to learn how to model either way. The fundamentals are more or less the same, and if you think you can get away with using simpler shapes or skimping on the details just because it's environment art you'll be in for a rude awakening.

>> No.952854
File: 51 KB, 564x597, 1683584919388693.jpg [View same] [iqdb] [saucenao] [google]

>skimping on the details just because it's environment art you'll be in for a rude awakening.
nono i am definitely focusing on the details. the shot i am working on is just going to show the exterior of the pistol and the bullets, so i just modeled the top, the handle and its cylindrical extrusion and the trigger as block outs. now I am going to model the holes and engravings. should i have also modelled the interior mechanism? i feel like that takes more time and it wont be even visible in the end product

>> No.952878

That anon would probably say "yeah", but in all honesty if it's a static background prop, you could have just as easily taken an image plane of the gun, cut it out and give it some thickness in the right places, and maybe some bits poking out, and no one would give a shit or be the wiser. It'd be textured automatically since you cut it out of an existing image, and it'd come with all the nice little details that your brain would fill in since it's a background prop and doesn't get close enough to the camera for your brain to pick up that it's just a texture.

>> No.952954
File: 2.96 MB, 1920x1080, 1661546947350668.png [View same] [iqdb] [saucenao] [google]

i am trying to avoid that workflow, i spent the majority of the last year making makeshift trimshifts and modelling shit and it does get results but if i bring it to something like unreal engine it will fall apart since most of my scenes are held up by cycles. my modelling skills took a huge dip when i was modelling via images.

>> No.953109 [DELETED] 
File: 392 KB, 1224x285, heds.png [View same] [iqdb] [saucenao] [google]

made some heads

>> No.953337

i know something's good when i have the urge to steal it. great job.

>> No.953423

Don't make me use that other stank ass wip thread.
I am NOT posting on a wip thread without a hobbit hole in it.

>> No.953462
File: 155 KB, 662x619, i want to ride my bicycle.png [View same] [iqdb] [saucenao] [google]

On today's episode of "things that seem easy to model but actually aren't", a bicycle. Low poly, at 1200 faces, with moving parts like the front end, saddle, pedals, wheels. Feels nice


>> No.953466
File: 322 KB, 2355x1181, bren.png [View same] [iqdb] [saucenao] [google]


>> No.953566
File: 732 KB, 1586x866, QLqD8x0[1].png [View same] [iqdb] [saucenao] [google]

Trying to learn how to do cartoony fantasy characters. This is my second attempt at a face; the first looked like an awful Bratz doll. The eyes are probably still too big but I'll finish it anyways.

>> No.953852
File: 386 KB, 908x973, ianPvfs.jpg [View same] [iqdb] [saucenao] [google]

Good morning I like tits, pls no ban sir