/3/ - 3DCG

Created some bookcases and books yesterday. Going through a first iteration, currently playing with lighting and wondering if it is a bit too dark now.

Don't know what to do next, I planned working on an interior of a small modern apartment on which I could test some horror gameplay elements, but I already missed the direction with this old style. Maybe I'll model a library/museum with secret passages to underground caves, so I can continue working on my foliage and rocks from another WIP.

CC always welcome, maybe I should retexture the bookcases, I'm not sure they are realistic enough.

>>581330

>>581331
Maybe its not too dark rather than lit not too well.
Floor breakings repetition too obv.
Post more, tho.

>>581330
I like it.

>>581332
Yeah, I need to fix those, they really are noticeable. I want to have a good contrast between light and shadows, but I still want the bottom of the bookcases and handles to be more visible. Maybe I should use those reflection capture spheres, I don't know how to use them properly yet, though.

But I think I nailed that colder light above the bookcase in the center, maybe it could still be better, but those shadows at the top of the books look so good to me.

Anyway, the scene still feels like it came from some indie title, so I need to refine everything a lot more.

Making some commie stuff.

>>581334
Thanks, I'll try that. Yeah, I'll model some lamps, I can't have a floating light sources.

>>581335
The light source is too intense, like a flashlight. Or the radius is too much. Talking about ghe shadows on the left. Bookshelves could use variety. Right now they look like they've been made from one huge block of wood. Or don't lit them much.

>>581338
I was wondering about that, I did model them from a single piece. How could I fix that now so it looks like it's made from separate parts?

>>581339
>How could I fix that now so it looks like it's made from separate parts?

What I did was carefully both change the direction of the wood grain in places (Vertical parts flow north/south, horizontal parts east/west) and paint subtle seams in-between parts that would logically be made of separate pieces (shelves, side/top/bottom panels, the feet, the crown).

>>581339
AO, curvature, edge breakings, color gradient between parts. Sometimes good lighting will do the trick. I'm not an artist, tho, just saying what ive learned from pbr tuts.

Still from an animation I'm rendering. Render times are rough

>>581349
Houdini?

>>581350
I think hoo can do better. Looks like C4D or smth.

>>581350
Yeah it's Houdini

>>581351
Do what better exactly?

>>581349
wow you twisted a few knobs and hit render,

so impressive

I've decided to recreate a few scenes from this game that I really love, starting with this. I'm not great at animating, and I'm awful at modeling people, but I'll try anyway. I'd love to get some thoughts on this.

http://i.4cdn.org/wsg/1504893568147.webm

>>581355
wow, and you did nothing at all! We're proud of you too.

>>581358
those tiny tapered legs makes it look like an insect

Muh porsh

>>581344
dark tower a shit

God left me unfinished

>>581358
thats cool

Gib environment ideas

>>581396
an old house attached to the side of a huge ravine crag. It's very windy 24/7 so adjust for erosion

Recommend me CPU renderer that is not Cycles /3/.

>>581403
vray

>>581403
arnold

>>581403
>CPU renderer
why?!

>>581418
use huge amounts of memory. far more than can be loaded onto a gpu

Beginner

>>581418
>buying GPUs in this market
>Not getting Treadripper or lower end EPYC
wew

>>581427
>not buying a GPU in any market
>being this poor

>>581427
gpu is like 5 times faster, and the biggest thing is that it eliminates noise

>>581434
interesting

>>581433
>Paying $800 for a mid range GPU

>>581434
>incredible amounts of rules that you have to follow to use the gpu's parallel cores effectively
>memory bound

>>581435
>Paying $800 for a mid range GPU
you can order the gtx 1080 ti for $800 on amazon. Its faster than the Titan X pascal. This is the most powerful card nvidia makes for consumers

>>581438
>I dont know shit but I will still clame otherwise
GP100 is the flagship card dummy.
A card ending with 104 is midrange and lowest being 107 period.

>>581441
>just mashing fingers against the keyboard randomly

>>581444
Not an argument.

>>581445
GP100 is not a consumer card. Consumers _do not buy this card_

>>581444
Are you retarded?

>>581449
>GP100 is not a consumer card.
It is sold in retail.
It is Nvidia flagship most powerful card especially for them sweet CUDA cores and VRAM.
Not for games though you got that right.

>>581452
theoretically I could buy a F1 car. That doesnt make my 2017 Mercedes that I drive to and from work mid range. The F1 is not something that consumers line up and buy same with GP100

>>581419
ok.
i can understand that.
but only if you're rocking a xeon.

otherwise you're wasting time.

>>581453
sorry you don't have basic understanding of GPU architecture nomenclature.
Please stop using stupid analogies.

>>581453
>The F1 is not something that consumers line up and buy

>>581453
>doesnt make my 2017 Mercedes that I drive to and from work mid range
correct!
the fact that's it's a mercedes makes it mid range.

>>581456
Shhhhh dont tell him

>>581453
>I wasted a lot of money on this card
>WHAT ??? IT IS MID RANGE??? DELETE THIS

Does anybody know if it is possible to copy layers from substance painter to another model? Or can I create a smart material from those layers?

>>581389
Thanks :)

>>581362
Awww, I shoulda made it a corny looking stick bug

The hand in the background was supposed to be a doctor but I couldnt frame it right

New outfit in the works. All hair Chibi Caesar!

>>581543

>>581330
It is too dark. It's hard to evaluate anything else because the contrast in your lighting (and post) works so badly.

What's the source of this lighting? What is the point of lighting it this way? The wall trim above the bookshelves looks as bright as a daytime archviz render, while the rest of the bookshelves are clipping to black in most areas.

Granted your environment isn't that interesting yet (you've got the models to make it interesting, but the layout is a bare hallway with bookshelves on the sides. You need to look at references for lighting. Movies, concept art, photography. Anything of spooky or dark (nighttime) libraries and manors. Analyze the lighting - where every single light source comes from (it it ceiling lights? wall sconces? moonlight?), how strong it is, and which areas it illuminates. Figure out some subjects you can frame in your environment - a doorway, a desk, a study/reading area?

Either there's something wrong with your albedo values or the contrast in your post process is way too strong. I'd turn that down (and don't forget to avoid extremely dark or light albedo values, even for white walls or black objects).

My other critique that isn't related to lighting is that your floor tiles very obviously. You can fill it with clutter and decals, but even so I'd double the tiling area, at the very least.

>>581356
What the guy above >>581344 said about the bookshelves applies very well to your clock as well as any more furniture you'd planning to add. You can add more details to the clock model and start introducing panel lines by changing your texture's UVs to match the direction of the wood grain. Also, no clue what program you're using, but you want to look into smoothing the faces of the bottom cylinders of your column, so they're not faceted.

I think the animation and camera work is kinda neat. It's cool to dabble into multiple things, even if you suck at em at first it's the only way to get started.

>>581371
I'm no expert in automotive rendering, but your car paint material is missing the sharp highlights characteristic of the topmost coating. Since these are present on the windshield (and thus the environment itself is sharp, and not a bunch of softboxes), the contrast makes it look like your car is made out of some rough metallic surface.

>>581543

Can we see his chibi shota cock?

>>581420
The white materials in your scene are overexposed. This could be due to your render settings, but if I'm guessing right and you made the diffuse color of those pure white (255), try bringing them down to a very light gray (220 or so) to get some of your information back.

>>581551

What an odd request. But here you go.

>>581537
You are easily one of my favorite users on /3/ now

Building a house. I'm never going to finish it.

>>581566
Finish it you fag

I'm going insane over this, please help.

Should modular walls be made as a plane or as a rectangle? I see both methods being used, but more often the one where they are made as planes. E.g.

https://www.youtube.com/watch?v=6--nQD1No-Y

(Btw, at 13:10 he shows some modular assets from Alien Isolation, are those planes, too?)

The problem is that I don't know how to use them when the players wants to go from one room to another. What happens in between them? Should I have a separate piece that hides the hole between the two walls or what? That seems stupid. I know I could try extruding the edges around the door hole inwards, but then they will be on top of each other when I put 2 walls back-to-back.

On the other hand, having modular pieces have depth creates another problems:

>takes longer to model (not too much, but still)
>alignment issues, because sometimes they will have to align the one way or another (shown in pic related on the left)
>if I need to have many different combos where one side of the walls has an X texture, while the other side has an Y texture, I'd need tons of different variations, instead of just having one plane-like wall with every texture

Pic related explains my question a bit better. Don't know what to do and I'm stuck with probably a pretty stupid thing.

>>581537
This is one of my favourite ones so far. When he clenches his fists by his side is great

Finally finished rendering. Trimmed cos size limit

>>581537
gameable

>>581537
Somehow I can't help but imagine a typical grim 4ch horrible person anon making these sugary cute characters.

>>581618
wow you hit render on a kinect mocap sequence. You should be weta in no time lmfao

>>581591
Make a fucking corner modular piece, retard. Nice vid, tho.

>>581624
I am asking what to do with a space between the two walls. Are you talking about that?? Should I model a separate cap for that hole? Seems pretty shitty way of doing it.

>>581618
Looks good.. i guess? In the way shiny moving things attract the eye.
>>581354
Nothing, I thought it was badly rendered water.

>>581626
What's wrong with you, anon? If you're making planes, make corner pieces. Otherwise make your environment out of 3d shapes. https://youtu.be/J1lnlTbMtKs?t=2135

>>581550

>>581591
Get the Source SDK from Valve and watch some Hammer editor tutorials on youtube. If you want to get into mapping or modeling grid-based playable architecture, that's the right way to do it.

>>581633
>Hammer
>2017
pls kid

>>581626
Do you mean in the doorway? You should make a doorframe for your door and make it part of the same asset.

>>581632
stop spamming this shit, it looks horrible

>>581636
How is it spamming? I've literally posted it just two times in two days, and I posted a more glossy version to fix that which was mentioned the in critique that I got.

Go fuck yourself sideways, Jimbo.

>>581634
Yes. Because if you try to make environemnts with generic modeling software, it's not that you can't do it, but you're not going to be a kid anymore by the time you're done making a single room.

>>581632
The model seems fine but you need to re-open the picture in Photoshop or GIMP, check the histograms and do the necessary adjustments to your tonemapping/exposure setting.

>>581629
You're offensive for no reason. This video still doesn't show the key part I was asking about, and I clearly explained the problem in my original post.

Here's pic related. Ok, I added those corners (which I already done before btw). What about overlapping polygons now? I need that wall on the other side and I need that bridge between them so the player doesn't stare at the void between the two razor thing walls.

>>581635
Do you mean that I should create a separate cap for that hole? Like, to have a fully separate modular piece which I have to manually place inbetween every doorway? Sure, I could do that, but that's why I'm asking here what is the optimal method. I have already tried all those things, to be clear.

>>581629
Let's just say I got legit scared for my mental health. Probably the biggest brain fart in my life over the simplest thing.

>>581632
ignore all the retarded comments.
get a decent HDRI, setup a scene with proper lights.
the lightning thing is annoying but your model is going to waste if you don't learn it

>>581618
i think porn with this "gel slush" anim style is an untapped market.

>>581653
Lighting is the most fun part for me. (Not that guy). Why is it annoying to you?

>>581656
because its hard to manage reflections for glossy materials. hdri can really affect your shading and how the materials are interpreted

>>581653
I've got a few HDRIs and I've used them on the model, that was just messing around.

>>581654

As a matter of fact I am fucking jelly.

Shot from a short film, still a long way to go.

>>581697
get ready for shitstorm

>>581627
Ah well cheers, shiny and attracts the eye is all I wanted!
>>581654
Oh shit, you reckon if it was slopping down on anime tiddies you could get into it?
>>581687
Literally fucking jelly? Maybe this could be a new venture
>>581697
Cool shit dude, I wanna see it in motion and not a Youtube screenshot though

>>581698
Start it up, bucko.

>>581697
>13 ft clearance
i don't know if the character is human or alien, but if human that's more like 33 ft.

>>581735
Thanks for pointing that out.

The water was supposed to be lava.

>>581756
>'Hhggh, hhggh'

I call this "Get ready to be sucked".

>>581756
no shit a bit jelly i struggle badly with super low poly

>>581756
>>581761
This reminds me of a project I worked on about twenty years ago on a modded Quake engine that had Q3A styled multi texturing, back when my Photoshop skills weren’t nearly as good as they are today. This is the only screenshot I have left of that work. About 5 years later, DOOM 3 came out and utterly changed the methods used in texturing for games.

>>581775
That's really cool. My slipgate has decayed a little bit over the years. (I may have lost the textures somehow)

>>581780
>>581775
can you emulate this on phone?

>>581793
It’s Quake 1. You can run it on a damn toaster nowadays.

>>581756
>>581761
>>581775
>>581780
quake or glquake?

>https://quakewiki.org/wiki/Textures
in case you weren't familiar with how they functioned with certain ones.

>>581909
>quake or glquake
My pictures are rendered in Povray actually. The textures I'm using are from the (now defunct) Quake re-texturing project.

some pet project work for a friend.

>>581951
Is he really your friend, tho, or you just don't have many friends so you lowered the bar?

>>581951
Kriss Vector with SCAR stock,SilencerCo Osprey,a fucking eotech and a duck taped glock.


Some fucking ingenuity right there.

>>581909
eQuake. It was a custom rebuild of the engine (not by me) from the ground up running on OpenGL that added 512x512 texturing, colored lighting, radiosity, and Quake 3 Arena style multi texturing (you defined layers/animation/blending modes/cube mapping via special shader files in plaintext). It died a long time ago, I cannot find it anymore, and Doom 3 made it an evolutionary dead end for Quake but it was neat little aside in the engine’s history.

>>581958
you got me :(
>>581959
concept wasn't mine, i was asked to do it 1:1 to the references given, and i didnt complained. While real life parts/contructions are obvious to spot, this is how concepts are done nowadays - just put everything into photoshop and make it look good. I have nothing agains it.

>>581971
._. You deserve to have real friends who will appreciate you, anon.

>>581951

just messing around

JUst messing around

>>581989
I'd mess around with that if you know what i'm saying

>>582005
Like.. give the sculpt even more details, anon?

Testing NetRadiant from Xonotic.

Working on my wrestler girl still. Cleaned up the face a bit and added details to the clothes. Hesitating doing arm muscles and shit but need to get on it.

>>582041
You should work on those arm muscles. Deltoids look like panel lines more than like muscles with actual volume.

Other issues:
Skin pores on shoulders, arms and clavicle are too large and pronounced, and look like disease scars.
Strong, uniform noise on shoes makes it look like a CG material from the 90s. There's a lot wrong with these shoes but that's the first thing I'd get rid of.
Wide, bulging mouth along with a wide and recessed chin make her look like an ape.
Bridge of the nose ends too low.
Brow line seems inexistent, it's like her forehead immediately caves in to make room for her nose and eye sockets.
Masses of hair have really bad definition. If doing stylized hair, you should aim to do fewer masses of greater volume (and place them in a way that reflects real masses of hair rather than thin brushed spaghetti). Look at a mixture of real life hair and cartoony head sculpts as reference for this.

Those are the main things I could spot. Her boobs are way too big for the complete lack of fat on her hips (she looks like an anorexic with trashy implants) but you guys do enjoy doing weird fucked up anatomy. You also really need to read more Burne Hogarth and practice your folds.

>>582054
And what I meant about the bridge of the nose is that it starts too low*
As in, the upper part should be higher, near the top of your eyes rather than the middle/bottom (your eyes are slanted and it looks like you tried to align the bridge with the eye corners rather than their absolute height).
My bad.

I did some detailing, added a ramp for accessibility and got into the crates bandwagon.

>>582054
Thanks senpai

Gonna do some slight texture painting in substance later.

work done for the mod
was responsible for retopo,uv's and texture
animations are generated, not final

any tips on trying to make this head look more believable

>>582311
post wires

>>582311
If you post wires, I'll ban you

>>582041
Nice.
Face is a little ape-ish but i guess it goes for most wrestlers.

>>582311
#NotMyHitler

Just making a few final tweaks to some crappy bits of modelling, and hopefully cure that light-leak on the right and we're done.

>>582325
cool. I'd get the gamma up to 2.2 though, like so.

>>582309
That's really nice man - what do you mean by the animations are 'generated'? I actually really like them.

>>582325
This is really good, there's so much detail going on! Nice man.

>>582335
it means they are done with mixamo. its assigning an auto rig and mocap animation to a character.
so for now im using it to test my character mesh, and we have an animator onboard that will make the final animations.

Was a huge bitch getting 4x slow motion while keeping the simulation characteristics the same

>>582341
Splashes seem to be more important than the character/animation.

anybody have any input on the topology of the areas i've highlighted? i feel like it isn't really ideal

>>581989
that isn't remotely how a bra works. the band of the bra (the horizontal band running around the torso) should be at the base of the breasts and should lie horizontally flat around the torso.you should imagine the band of the bra as a ring at the bottom of the bra/breasts that runs around the torso

>>582325
I've seen this before, someone said it was a tutorial or a workshop. Do you know where I could find that?

>>582389
Maybe this? I don't know for sure, I didn't look at dog anatomy so I don't know for sure how all the muscles should flow between the muzzle and the cheeks so that it can be animated well with usable and human-like emotions. I just wanted to get rid of some areas where it looked like the implied line was cutting through the quad, like the stretched diamond in the quad loop coming down from the eye and going across the side of the nose. The one I changed under the jaw might not be for the best, I thought the shape should flow with a clear curvature from the underside of the jaw to the underside of the head.

>>582396
ah that could be the trick, thanks!

>>582395
Yeah it's a a thing on Gnomon workshop. It's really not a 'tutorial' as such - you're not shown how to use any software really, its more about workflow and how to approach a big project like this. So if you're looking for lessons on how to model or set up lighting or materials etc. it's not really going to help you.
But if you know how to use the software already and you want a big project with specific tasks and goals and you want to get to grips with workflow - it's a great project.

'Project' is probably a better word to describe it than 'tutorial' now that I think about it.

Theres stuff I wanted to fix on this one but it took 14 hours to render with sss..

>>581565
>>581616
>>581619

Thanks :)

Im gonna try to get better about seemless loops.
>>582325
That looks awesome. How did you make all those assets? Is it all photography or do you hand paint? that must have taken months

What kind of environment should I make?

>>582493
are they eating the moth lmao

these are so cute i love how jiggly they all are

>>582493
>That looks awesome.

Thanks man.

>How did you make all those assets?

99.9% just regular box modelling in Max. There's like one or two things that were sculpted that really didn't need to be like the bar-seats and the foam padding around some of the steel bars.The motors above the garage doors were done in Fusion360 (trying to unwrap Fusion meshes is a fucking horrendous task btw)

>Is it all photography or do you hand paint?

It's a mix of everything. 70% hand painted in SP, 20% custom textures put together in Shop and the rest just slapped together out of whatever shit I had lying around. The vast majority of the work was texturing.

>that must have taken months

You have no idea. By far the worst part was getting it to render. It took days and days of pulling it apart, inspecting and optimising individual parts, and then merging it all back together to get it to a state where VRay didn't just flip double birds straight at me every time I opened a frame buffer.
I ended up having to do it light cache + brute force which is just dog sloooooooooow, but there was just too much shit going on for irradiance map to handle.

I've learned a *lot* from doing this scene.

>>582493
did you animate all of the grapes by hand or did you simulate some of them?

>>582493
What are you using to render that?

>>582512
Its blender/cycles. Theres a cool mouse recording feature that saves a lot of time for goofy animations like this. I have 3 bones responsible for moving the whole bunch of grapes and every other grape has a dedicated bone to move the individual character. I tried soft body physics at first but it looked a little weird. The arms are physics and animation driven but not the grape bodies

>>582515
how long are you doing 3d?

>>582522
I've lost track but on and off for over 10 years

>>582515
crazy inefficient

>>582528
how would you do it?

>>582529
not that dude but there is no such thing as inefficiency when rendering a 5s short.

the best efficiency i know of is group instances. but its only mandatory for large scenes

>>582529
I would only create a single grape and instance it

>>582527
did you learn at cgcookie?
im curious because im thinking about subscribing. they have lots of stuff you can't find for free

>>582531
The grape faces and arms are already instances of each other. The animations are all unique tho since they all have to react the butterfly at the same time but still look different. I dont really get how you'd do that and still have them all react to the butterfly at once.
>>582533
I dont do many tutorials. I mainly look up random youtube videos when I dont know how to do something. idk if I'd use it enough to justify 21 dollars a month but it looks like they've got some good stuff. Maybe try it for a month?

>>582534
>The animations are all unique tho since they all have to react the butterfly at the same time but still look different. I dont really get how you'd do that and still have them all react to the butterfly at once.
houdini. This is mid 2017.

>>582534
In what way would grapes react to a butterfly?

>>582509
What exactly did you learn from it?

I'd try it out, but I don't think it would be so beneficial. I know how to optimize the render and modelling boils down to repeated generation of small assets. It seems like a good practice, though.

Btw I also watched that video series. Good stuff, especially the first part. Even though I feel it would be better to work on the original scene. That way nobody would even think that you used any "tutorial", and that series doesn't put a focus on that scene but talks in general anyway.

>>582534
What are physics

>>582329
I remember when you were starting this )))

Re-did the mouth. Used >>581539 for reference. Thanks. Not sure of the jumpsuit. May left a few things from the image to the right.

>>582677
Mouth looks much better.

Grains for softbodies are so fucking quick

This texture set is from sauerbraten.

>>582647
I think I'm one of those dudes who don't know when to quit.

High-poly to Low or Low-poly to High?

>>582724
obviously the first.

making base meshes for sculpting is something noble but they stopped doing that years ago. now you can create very good base meshes in zbrush

>>582725
so whats the optimal way to make the Low-poly from High?

using Quad-Draw currently

>>582727
there are lot of small rules you have to know, so its best you watch a few vids. but generally speaking

>never extrude an entire loop at once, it will completely break, make sure to only extrude polygons that are on your camera angle.
>quad-draw is a good way to prevent the mesh from getting fucked, but you only use it for generating loops quickly, when doing something like faces you want to be slow and manual
>check that you don't have any doubles before you are done
>you can use shrinkwrap (or whatever its called in other programs) to force your mesh to snap into another surface, but its only good if you are working on a simple mesh like a rock
>when doing retopo on several meshes at once, you can join the lowpoly to one object at the end, depends on what you need - and make sure to export polygroups.
>you don't need to join objects for the highpoly mesh, infact i recommend you don't do that at all. you want to keep the shading intact as possible before you bake

>>582728
cont.

>export __highpoly.obj as the highpoly mesh
>export ___lowpoly.obj as the lowpoly mesh
>use xnormal or substance painter, .fbx doesn't work well there.
>always bake all the maps you can, you never know when you gonna need them

can I salvage this lion head?
also how do I model a lion head? How do I model?

Im really not happy about the soft body petals. I gotta rerender tonight

>>582793
Its hard to see whats happening in the mouth area. Just keep pushing verts around. All models look weird at first

>>582802
o fuck I got a real reply
what do you recommend between starting out with a cube and subdividing it, or starting out with a single plane and extruding it?

>>582803
for smooth looking cartoon characters, I always start with a subdivd cube but you can go either way

>>582802
God, it's beautiful, are you famous yet? You should be, you're doing the most adorable shit ever.
Would you consider making a tutorial one day?

>>582802
I just wanna know how long have you been doing 3d and where did you learn how to make something like this.

>>582817
I agree he should be famous, but I really don't think he should do a tutorial. Cos it would literally just be teaching others his animation and character style, which would result in a copypaste of the tutorial everywhere and ultimately it would devalue it.

>>582820
True, but I feel that he could easily teach a few nice tricks when it comes to animation. I'm not talking about the whole process from A to Z, as it would devalue the cute animations market with cheap rip-offs indeed. I'm his fan first and foremost, and I don't want to do him any harm.

>cute animations market

>>582721
Usually better than the reverse.

>>582721
Never quit, never surrender. Fight, Fall, Fight again.

>>582821
Yea, it's true art, the only tumblr page I've ever bookmarked

Started this ages ago and stopped. Hated most of it so stripped back to models.

>>582916
I really like the camera angle and lighting. What's up with the speakers though, why are they octagonal?

>>582710
>First guy lands head first
>Two guys land mouth to mouth
Just...nitpicking

>>582929
there a problem?

>>582509
Quick noobie question. Why does it take so damn long to create something like this? Is it mostly modeling and arranging assets or mainly correcting things and rendering?

>>581321
workin on this scene for a commercial, leaves are flat. trying to work on that shit

>>582916
nice lookin
>>582923
prob just un-smoothed

>>582947
Grass is really even, would be nice to see some breakup in its distribution (along with some other plant meshes if you have the time for it).

What is this rendered in? The sky bloom through the leaves looks like PS2 graphics.

>>582938
Modeling and texturing are by far the biggest chunks of scene creation. Composition (both initial and final) and lighting are thing that require experience, but less time to set up.

>>582950
arnold, just did some random bloom effect in photoshop. and the Zdepth pas i thew in there i rendered at 1/4th res to speed up the test render

im not completely sure how to do clump or distribution maps in Xgen which is what im using. ( first time doing a forest scene )

i agree with what youre saying.

>>582947
Damn, that level of quality is enough to get a professional work? What renderer is that?

>>582959
Nvm, I see it's Arnold. Yeeeah, I feel like your settings are completely fucked.

>>582515
>Theres a cool mouse recording feature that saves a lot of time for goofy animations like this.
What is it called? Didn't hear about it. So like, you can animate those small movements by jerking the mouse left and right and recording those movements or what?

>>582960
like shaders or what? im down for some info. i usually do hard surface and some characters. its one of my first jobs, the scene is set up how they want, but i dont mind making it look better shader wise and shit

>>582947
Shadows too sharp. Sans that looks good.

>>582963
ah, just looked it up. added some shadow angle to the light and they come out softer.

>>582962
Not that guy but what I would criticise is the bloom, the fact that the highlights are clipped but not white(??), shadows on the branch on the ground need to be softer, the grass and tree stumps are much nicer looking than the trees and it throws off the whole scene. The bloom makes it look like a videogame. A bit too much depth of field for that wide angle a shot.

I think your trees need the most work. The branches need more detail, the leaves need more variation and for the love of God use a SSS shader on them.

But if whoever is paying is happy with it, who cares. If you want to improve it, there are my crits.

>>582938
Long post incoming because I'm waiting on a render.

I'm the guy who did the scene. I'm not a good indicator of how long a scene like this should take because I used it as a learning exercise.

I'd never opened 3DS Max before April 2016, let alone Substance or Zbrush or Fusion360 or any of the the other things I learned in doing that scene.

So, I would do certain things like the floor, for example five or six times in all different ways, because for one; I'm super picky but also so I could learn different things.
Only a couple of things were done and scrapped and re-done to that extent. But the floor was definitely the thing I obsessed over the most, and I'm not even sure the final version was the best but c'est la vie...

So, off the top of my head here are the different ways I tried doing the floor;

1. A version that used displacement maps.
2. A version that used huge, HUGE, hand-made textures (like 24k files) that I made out of my own photos which I went round taking of pavements and tiles and roads etc.
3. A version that was sculpted in Zbrush.
4. A version that was sculpted and painted in Mudbox.
5. A version that used multiple tileable textures composited into multiple different UV channels.
6. A version that was divided up into multiple UDIMS and painted in one hit in Substance Painter and used multiple 4k UDIM textures.
7. The final version, which was the road painted in SP on a single set of 4k textures, including the water. All other floor elements were simple modular rectangle slabs which were painted up in SP and had a different look on each of the 6 sides of the slab. So, 1 side was tiled, the other side; same tiles but broken and dirty, the thin edges there would be wet concrete, dirty concrete etc and then I would just rotate and duplicate these slabs around the scene till it looked how I wanted.

>>582980

Also... haha... I just looked at it again. Yeah, there's LOADS of reasons why this scene took me far, far longer than it should.

You see the hi-rise apartment block in the background, which is obscured mostly by the cable trunking?
That's a fully realized model in itself which uses randomized textures on all of its elements. There's laundry and stuff hanging from all the balconies.

You can't see it in the scene at all pretty much, but again, that's not why I made it. I made it to learn and maybe use again in a different scene.

Pic related.

>>582982
Which btw was based on this...

>>582944
add more specular/metalic. There's obviously not enough shine in there

>>582944
use hdr and better materials

>>582982
Did you just make a fucking photograph of you rmonitor?
How retarded are you?

>>583009

>>583009
Not sure if this is what happened but sometimes folks use their smart phone for the Internet while they're actually on their computer.

But whenever I've seen it it's because they're sharing something from a fullscreen game they're currently in...

>>582982
Yeah what the fuck, explain this.

>>583009
Yeah because it was the only image I had of that model, from when I took a photo of my screen to send to my girlfriend when she text me to ask what I was doing.

And to answer your second question; I'm not in *slightest* bit 'retarded.'

You, on the other hand, are.

>>583018
Yeah see my post above. Not that I have to 'explain' anything to a cunt on the Internet.

Fuckity bye.

I'm tired of modeling. Fixed up the face a little and the body. Fuck UV mapping and texturing. I'm going to start rigging and blend shapes.

>>581456
I think he means formula 1 not McLaren F1

>>582341

why not just render at 240 fps and slow in post?

Not animated yet, just doodling

>>582961
yeah, the timeline window has a recording button. If you play back animations while you have that button toggled, you can record stuff in realtime with your mouse or tablet.

>>582818
around 10 years. I mostly just posted to /3/ on and off for the last ~5 years and selectively took feedback from these threads
>>582817
:)

Im not that good at documenting my steps and stuff. I coudl try but it would probably be shit and you wouldnt learn anything. It would be like, "lets make this guy look and move like that because it looks ok I guess". I'm still thinking about it tho

>>583091
This guy reminds me of the jiggler dude from Adventure Time

>>583092
he was a cutie.

While making this one, I kept being reminded of that gross thing with the arms at the bottom of the well in zelda OOT

>>582982
>that taskbar

>>583094
My taskbar is almost completely full. I have a shortcut for every program I use regularly there. What's the problem? I use it pretty much like that thing at the bottom of Macs.

>>583094

What's up with it? Are you one of those fucktards that still has a desktop littered with application shortcuts?

>>583105
>> I use it pretty much like that thing at the bottom of Macs.

It's called the dock, and that's exactly how I use my taskbar. I cannot for the life of me fathom why anyone would keep app shortcuts on the desktop or search the start menu every time when you can just pin them to the taskbar.

>>583094
>Going to your desktop or start menu when you want to open your most used programs

Lol

I decided to make a short video with this fucking car and I'm hating every minute of it (this is the ending)

At least I'm learning. And smoke simulation is a bitch.

>>583110
But it's so clean and le minimal that way! I like my desktop as useless as possible. /s
3 extra steps to open a program master race

>>582982
>taskbar on the bottom
you one of them freaks what default to looking up instead of down?

>>583115
Mac user for last 16 years. Dock at bottom. If the auto-hiding feature worked as quickly and instantly on Win10 as it does in MacOS I'd have it hidden the whole time, but there's like a 1.5 second lag so it's just annoying to use.

So I'll just add that to the never-ending laundry list of UI features that Windows does just that tiny bit worse than MacOS...

But yeah, 'macfag'... 'overpriced'... 'gay'... whatever. Take your pick. Yawn.

>>583113

There's a lot of good stuff in here anon. I love the torque-leaning effect.

It's perfect up to the point where the fish-tailing gets corrected at 04s. I feel like it straightens out or corrects too much, too quickly, and it kinda feels like it's sliding around after that and not really in contact with the floor. Fuck the smoke effects. They're just getting in the way. Ok - the backfire is cool, but you're wasting your time simulating smoke. That's always gonna be done in post.

But yeah, gotta say the body leaning and rolling on the gear-changes is awesome. Love it. Those effects would maybe be more believable if it were a heavy-as-fuck 70s muscle car with leaf-spring suspension.

Congratulations - your vid made me wanna listen to Fu Manchu. (Link related).

https://www.youtube.com/watch?v=_Zko7pBeHkk

>>583113
Why do you hate it? Its not bad.
But you should have made the whole tire a emitter for smoke. The way the smoke appears is a little strange and it should linger longer.
Other than that....its quite nice, the animation of the car seems to me a little exaggerated which makes it appear a little cartoony but i like that.

>>583113
I think it looks pretty good, you clearly have a good eye for animation. Your smoke sim is clipping through the car though, make sure your collision is setup properly

>>583121
Not that guy, but it's not exaggerated beyond realism. Engines with 8 cylinders or more that deliver shitloads of torque can make a car list and lean like crazy:

https://www.youtube.com/watch?v=w0DrcwHYXnk

>>583124

The (what looks like a) Toyota Supra at the start of this vid leans back even harder than the animation. Note: way more flames, way less smoke.

https://www.youtube.com/watch?v=gOtgyaDbj1Q

>>583121
>>583124
At first I wanted it to be more grounded but in the end I said fuck it and went full cartoony. I'm glad you liked it, as I was afraid it was too much.
Fun fact: I ripped off the dimensions (and sounds) of a Dodge Viper

>>583119
I noticed it too as I render it, and the reason for that happening makes me very angry: I did a stupid camera movement but there are no points of reference in the ground as it's just a smooth flat texture, so visually it looks like it's just sliding laterally. I wasted a lot of time making sure the wheels followed a somewhat logic trajectory and fucked up with the camera movement. (You can see what I mean if you look at the background, it starts moving as soon as the car starts "sliding")

>>583123
The smoke got me fucked up real good, It dissipated either too fast or too slow, and supposedly it's colliding with all the geometry of the car (which is really an overkill) but it seemingly clips AND forms ugly clusters of voxels on the more dense zones. I'm tempted of just removing it, but those render hours hurt. Guess I have to learn to let it go.

>>583126
Yeah the camera move is only adding confusion. Just let the car blast out of frame.
You need to make this piece just about getting the car moving right, and leave things like camera movement and smoke effects for another day.

>>583126
Baking things like smoke effects etc. into an animation is a risky move because they're so subjective. For every person who says there's not enough smoke, there's someone who thinks there's way too much.
In industry, there's always gonna be a 50:50 chance that someone above you is gonna not like the effects or want the color changed or whatever.
If those effects are all baked into the beauty pass, you're fucked.

>>583126
What program are you using? Is your collision volume fucked if you preview it? Try upping the resolution and make sure the volume is actually animating. Colliding with the car isn't overkill, it's what needs to happen for a realistic sim

>>583130

Phoenix FD for 3DSmax 2018. First time using it, all my life I've been using FumeFX (mainly because that's what the old studio I worked at had)

The collisions are supposedly correct and do displace the voxels, I could see it accumulate in the wheel well and reacting to the spoiler moving. I lowered the simulation resolution to 8mm, more than that started giving pretty crazy numbers for what I wanted to do (+65 million particles).

With overkill I meant that usually I'd make simplified collision boxes, but for this particular case I used the high poly car mesh, and wasn't too satisfied about it in the end.

>>583129
Yeah, fluid simulations and other "post-y" things always should go on their owh channel. That's the kind of stuff you usually learn the hard way, hah

>>583135
Can't you just make a proxy object for collision? Surely you shouldnt need to use the complex geometry of your car. And can you use volume collisions or is it only surface?

>tfw I thought getting stuck and abandoning my projects was only the case with my music
>started happening with 3DCG as well
Nooooooooooooooooooooooooooooo!!!!! Why did I deserve this? PLEASE STOP!

Anyway, shit, I'm afraid of spending a lot of time on modeling assets while I don't have a clear image in my head of what I exactly want to do, what I want to have in my scene and what design direction I should pursue for every asset. But since I'm kinda unable to do that in a reasonable amount of time, I'm stuck where days pass and I don't make literally anything. Has anyone experienced this before? What do?

>>583135
Woah thats amzing

>>583139

Yeah but then I got diagnosed with ADHD and take just shitloads of Adderall and now I'm a fucking powerhouse.

True story.

>>583149
I've heard that so many times senpai

>>583149
You think you're a fucking powerhouse when the truth is you'll burn out your dopamine system and your liver becomes a toxic swamp.
Seek alternatives, learn and study the topic.
Cannabis and Psychedelics might help suppressing symptoms of ADHD for example.

Adderal just fucks your system long-term.

>>583139
Yeah i have that too sometimes, its complex, could be:

fear of being judged (i don't finish it, so i don't have to show it)

procrastination/laziness, then loosing interest in the project

biting more than i can chew, projects gets to big, amount of
workload grows above ability to handle or isn't fun anymore but grisly hard grinding.

technical problems down the road leads to unknown workflows i have to really study first > project gets old

The strategy against it is brutal efficient time management and planning.
Learn how to organize and limit yourself. Make the projects small, like really small. Make a plan, stick to it as if you work for somebody else. Be really hard on yourself in terms of sticking to the deadline, but keep the content flexible enough so that it is fun.

For example in the next 5 days, you gonna spent 5 hours modelling a lowpoly vehicle, texture it, animate it.
If you like the result spent the next 5 hours on an environment for the scene, or another vehicle or something else entirely.

If you break it down like this you get much further than just drifting.
You also learn more about yourself and your limits.

>>583176
>warns against dangers of methylphenidate
>advocates use of psychedelics

Stop talking.

>>583183
ADHD isnt even real

>>583184
So, a lot like your chances of ever getting a blowjob then?

>>583185
what?

>>583113
https://www.youtube.com/watch?v=atuFSv2bLa8

>>583186
Nonexistent.

>>583188
I-I had a thicc gf for over 6 mo

>>583189
Sure.

>>583183
Psychedelic are taken once every 2-3 weeks, Adderal daily.
Amphetamines have proven to be detrimental to health.
Psychedelics have been proven to be beneficial to health.

Stop being a fucking idiot and repeat decades old anti-drug propaganda and play into the hands of the pharma-cartels.

I did not advocate the use, i did advocate learning and study.

>>583091
progress report.
This is fucking hard and want to kill myself.

>>583238
i feel like id be better if he cheeks bloated from the mesh itself

test

>>583238
awesome. more pls

>>583201

You come across as the kinda fella who got most of his science education from Insane Clown Posse albums.

>>583201
>Amphetamines have proven to be detrimental to health.

Methylphenidate isn't an amphetamine.

Seriously; stop talking.

>>583072
In this case it's the exact same thing as substepping

>>583249
thanks for making me aware of them.
Comedy gold.

>>583250
I never implied it is.
Methylphenidate = Ritalin

I was talking about Adderall, you retard.
I never mentioned Ritalin once.

Whats the point of using maya for modeling if i have ZModeler?

>>583305
good camera options makes a world of difference

Why is this board so fucking dead

>>583307
because of some salty niggers that have to shit on everyones work, or on the contrary suck major dick

>>583307

We're already dealing with a pretty obscure hobby/occupation. Most software has its own community somewhere that caters to most of its users so 4chan is going to really only draw the salty fucks that can't get along with the other communities and the autistic fucks who lack to social skills to get along with anyone.

>>583307
adults with jobs and people preferring more standard forums like polycount.

>>583312
polycount is not any better than this place.

>>583312

ding ding ding ding!

It's almost as though people have got a limited tolerance for being called a 'nigger' for using free software and being confronted with .gifs of video game characters being raped.

whoudathunkit.

>>583307
Cuss I spend over 60h working a week and have trouble with anxiety leading me to drink. Working less would make it worse.

Can't figure out how/where to set the gamma on this.

It was output with Reinhard colour mapping, multiplier = 1, burn = 0.05, no gamma correction or colour mapping embedded.

I have two monitors and it looks wildly different on both.

(Screenshots went over 4mb limit so apologies for links)

1. Here's how it is with no overall gamma / exposure adjustment applied. When I look at this capture on my iPad, this is pretty much exactly how it looks on my crazy expensive UHD display...

https://anonimage.net/image/9AbAhf42J

2. And here I've just dumped an exposure adjust layer over the whole thing and pulled the gamma up a bunch so it looks like it does on the cheap-ass 1080 extension display...

https://anonimage.net/image/sVvqbUO5A0

It's not finished, these are only screenshots that are less than 1/5th of the actual resolution. But which looks most correct so far from where you're standing? 1 or 2?

>>583386
1 looks fine to me.

>>583387
Not too dark / over-contrasted?

>>583388
No, its exactly how i would do it. 2 looks hazy/ overexposed.

I really don't know how to make this look less like a toy and more like a real plane. Maybe I'm missing some camera settings?

>>583394
work on making clean shapes first

hard surface ? not really>>583394

>>583394
Yeah definitely what >>583400 said. Right now your edges and shapes are very blobby. You're going to want to add some edge loops / bevels on edges so that when you turn on subdivision it all sharpens up there.

Beyond the playdough forms your material and textures aren't helping. The cloudy texture you're using on the red areas is too large, try scaling it down a lot so it looks like small changes in texture. Also the surface looks rough and soft rather than hard, you'll want to lower the roughness of your diffuse shader if you had it increased / available at all, and smaller / sharper speculars / reflections.

Apart from that you got the hardest part down: The proportions. You've got the shape of everything basically done, now's just getting everything looking sharp which is a fairly mindless task in comparison I think.

Look up some hardsurface modelling info if you don't know much about it. I personally find all the wire diagrams of how people solved complex shapes with clean topology really interesting and soothing in a way. I love the order of the results and the puzzle-like quality of the problems.

>>583418
>>583400
>>583399
Woah, I didn't expect so much help. Thank you very much! I'll upload another version using your tips

>>583241
I smoothed out the edges a bit to help. It looks more like clay now which is alright with me

>>583506
if you are looking for exposure come join the blender groups on facebook, im on a few of them there and haven't seen you yet

>>583510
which one is your favorite? I dont post on facebook too often

>>583506
Ohh baby, I need more of this in my life

Tried modeling a revolver. Don't have enough experience in modeling non-trivial hard surface stuff.

Anyway, posting here because I guess the modeling is nearing the end, would appreciate some feedback if I have to fix something before unwrapping.

Btw, I have no clue how to UV something like this lol. It is made of separate parts, so do I UV every part separately or attach everything together and UV one giant fucking mesh? I actually tried that, it even worked, but my diffuse map for material IDs got completely fucked after baking, so I can't properly separate pieces by their colors for texturing. And ofc, everything is tiny, so resolution would be horrible.

I've seen some YT tutorials where people UV separate parts, but how do I use all those textures maps if programs like SP and Quixel need just a single map? I don't understand that part very well, so please help.

>>583525
For UVs, you typically unwrap every piece individually, then join everything in one big mesh/UV map and pack it together near the end. In games, a single prop will almost always have all of its mesh components unwrapped on a single texture, while in films it's common to break the UVs into several tiles that are super high res. If you're just starting out, your revolver is simple enough that you could probably pack it into a single map easily.

>>583526
I'd put it in UE4, so I need a single map. Didn't know I could UV separate groups and join them in a single map later on, huh... Got any good tutorial for that?

>>583525
Put a seam under the barrel, under the tuber under the barrel, under the cylinder, along the front of the main body down the handle and under the handle, inside the trigger guard all around, screws just put seams around the base of the head so they're just circles on the UV map, on the unseen backside of the handle insets near the visible edge, along where the handle and the main body meet, along the front side of the hammer that's inside the gun, across the sights horizontally along the thicker bottom part, around each hole set inward a little so you can have an area without a seam to have a clean area for scratches to be textured, probably some more places. Then let it auto unwrap and see if any other places need seams to help it split the model apart in 2D. I'm thinking it could be projected onto the UV map in such a way that it's split down the front and opens up like a butterfly for the big main pieces.

>>583513
blender artists

>>583525
you forgot the cylinder dents, which is ok because i forgot them too. i had to boolean them later.
also pay attention to the buttons on the sides, its easy to miss them.

nice modeling overall

>>583536
oh and i forgot to add.

you can make a good UV maps packed into one texture. but it depends of how good are your unwrapping skills.
if you have enough resolution for all parts it means you have good unwrapping skills

>>583536
Ah yes, I actually know about these dents, but I thought I could maybe create it through normal mapping in quixel/SP? Because it seems not so easy to model - but I will model if I can't do it the other way.

Didn't know about that button on the other side. Didn't see it on my reference. Thanks.

>>583535
Thank you, I will keep this in mind.

Btw, found a video how to merge separate UV maps together, nvm about that.

But yeah, I don't have much experience in proper UVing. I have been using flatten mapping way too often in the past. Finished UV for the main part before >>583535 post, so my seams are not exactly in the right places, but will that be a problem in programs like SP or Quixel anyway?

Maybe this UV sucks, I have no idea lol.

>>583538
do them in sp/ quixel. just watch for the ao effect if its missing

>>583538
the UV is terrible

>>583541
i wouldn't say terrible, but needs more re-arranging and efficient use of space

>>583541
>>583542
I know it's probably not good, it is by far my biggest weakness. But for this part I actually follow one YT tutorial and he did it in a similar way. I don't really know what makes a good UV except for a good use of space. For this part I basically just used quick planar map for the flat surfaces and then pack normalize. Just made sure it doesn't stretch and that seams are not very visible. I'll redo that part when I UV other simpler pieces.

>>583542
No it need rework from ground up.
The seams are bad all around unless he decided to use 4K texture map in which case it doesnt matter.

>>583550
Watching some ChamferZone tut again. Hope I got this part right at least.

>>583554
It's basically just a cylinder, the faces of the little extrusions don't need their own section. Just one seam on the underside, and one around each end, and straighten.

>>583554
Plus how much detail do you really need on the tiniest extrusion on the tiniest part of the gun you will see from the back for 90% of the time? Not much. The AO is basically going to be a smooth black gradation at the edge or pure black in the small groove between the two big extrusions at the end, I think you'll only need a couple pixels for that.

>>583556
Or actually maybe on the top? The barrel hides it most of the time, but it's raised for a reload you might see a badly handled seam in the textures. Maybe on the side away from the player camera.

>>583556
>>583557
>>583558
Kek, true. Fuck it, I'm learning.

Don't wanna blogpost too much, but here goes another part. I wasn't sure what to do with all the chamfered edges and maybe I butchered this again. Enough for tonight though, my fingers hurt like hell and this unwrapping shouldn't have been going so slow. But hopefully this gun is textured and finished tomorrow.

P.S. that seam in the middle should probably be placed at the bottom. Not sure what to do with the sight, though. Btw yeah, I planned to use 4k textures anyway, but I want to do everything properly for (much needed) practice. And thanks for all the help guys.

>>583562
Btw, I know it's stretched on the image, fixed that already.

>>583556
>dat cylinder capping

really makes you think

Oh God, I'm so clueless right now that I really need some more help. I have so many questions that I can't answer. Pic related.

>1.
Baking hp to lp is failing. Don't know why.
>2.
For Quixel I need clean, flat colored islands - I can't bake them like that in 3ds max. They all have shadows, both in 3ds max and when viewed in Quixel. This didn't happen for simpler objects before, so I don't know what is the cause of the problem.
3. I don't know if I should group everything or attach pieces together. Basically what I did was group separate objects, all with their own UVs and then added another UVW modifier on top and manually positioned 9 other UV blocks on my 0..1 space. Now I'm not sure if the islands are too close to each other, or I fucked up my seams, but normals are just not baking properly.
4. I also don't know if I should have those screws on my LP model as well, or should they be just normals (but how can I add a specific material to them, then?), but this question doesn't matter at this point. Thanks.

>tfw unwrapping takes longer than modeling

>>583603
It's from an automatic quad cap it looks like, helps with subdividing

>>583618
don't group together the highpoly objects when you export them.

seems to help

>>582311
Give it a nice SSS shader

>>583622
I tried everything I could think of, ungrouped, grouped, attached, re-smoothed - nothing works. The biggest problem is that I don't know if it is happening due to UVs or bad export or something else. I heard something about normal map not baking properly if the seams are positioned on hard angles or something like that, idk.

P.S: I guess I could cheat and just texture high poly mesh for the beauty render, but I want to use this in UE4, so I need to figure this out.

>>583633
the edges seem to be fucked.

maybe try to give more space in your UV map in between islands.
your UV map islands are too close. leave a little space for dilation to happen

Trying to nut out the look of this one before I do the high res sims and wet maps

>>583634
Did that, but it still happening. Here's the normal map. Is this enough space between the islands?

Is it necessary for every part of the revolver to have a specific name + "_low" or "_high" at the end? I tried both with it and without it already, but they didn't make any difference. Really don't know what to do...

>>583638
oh wtf

the bake itself seem bad. did you bake it in sp2?
i can see the error in right there, the resolution seem to be alright.
you have no dilation turned on ( a remainder) if the size is in pixels. 8 should be fine.

if i had to guess it has nothing to do with dilation and you have to change the ray distance to pic related (works for me) the lower the better, but its already low

>>583641
also try to see what results xnormal gives you
you can import xnormal maps into sp2

>>583638
Omg, I should have baked my normals and AO in xNormal right away. I was doing it in SP... This seems to be getting better results now, I'll test it now, just need to stop Shitxel from crashing. Eh, I have a shitton of programs open, my fault.

>>583641
Yeah, I made a big mistake by trying to bake it in SP I guess. Will update when I try this out.

>>583644
this looks alot better
you should be able to bake
normal,world space normal,curvature,thickness,AO, color

the metallic,roughness and height should come from sp2

Hello! I've been working on this sculpt in Zbrush.

It's my first time posting here, I was hoping I might get some feedback from you guys! :)

>>583645
This thread is almost dead, I'll just say that when I apply that new normal in SP, the result is STILL the same for some reason. I'm tired but I'll maybe create my own thread for this where I could post multiple images so I don't spam this thread. Appreciate all the help.

>>583646
I like it .
Outside of the stubby feet and a lack of detail in horns i dont see anything that could use much improvement in that style.

Still have to learn Zbrush one of these days. Going to have to do something with the ears.
Getting closer to getting it done. Not starting a new 3D project till I finish one.

>>583646
Very cute I like it. Love the chunky shapes the face and the fur skirt. The stubby feet do look weird to me.

>>583656
I see! Thanks for your feedbak! My intention was to make her feet stubby, but I'll look up what to do with the detail on the horns. Thanks!

>>583668
Thanks, I really appreciate it!
I love how cute yours is too! Keep it up!

>>583646
very cute. the tail scared me when it sliced the camera. lewds when?
>>583668
are these shots taken from an orthographic camera or perspective?

>>583646
also what was your method for making that fur pelt?

>>583691
Thanks! And yeah, that happened to me the first time I rendered the video hahaha
Also, no lewds for this character, sorry :P

>>583692
I'm not completely sure what you mean by making the fur pelt, could you explain further please?

>>583646
Yes I used an orthographic camera. Helps if I have to re-work the model sheet from it. Could we see a wire frame of your character?

>>583703
the "skirt" that she is wearing that looks like it's made of hair. i'm curious which brushes you used and whether you pulled the tendrils out from a base shape or if you placed a bunch of separate tendrils in place and then dynameshed them together.

>>583705
if you're mousegirl bro, you shouldn't use orthographic cameras when showing other people your models. it looks weird. use perspective.

>>583714
Will do from now on. Thanks

>>583705
It is a sculpt! It has no wireframe worh talking about.
I'm going to be doing the retopology now and engine ready model now, then I'll be able to show the wires =)

>>583712
I've modeled each hair individually from a very lowpoly cone shaped mesh. Quite tedious, In any case, you could watch me work on Twitch while I stream if you want to know more.

(sorry for the repost)
>>583712
The videos from my streaming doing them are still up in my profile :)

If you are interested you could check them out on twitch.tv my username is cless_aurion

Been sick for a couple of weeks, finally close to wrapping up this building interior, only one room left to go.

>>583834
cool, what is it for?

>>583840
Le meme vidya project.

Seriously though, a survival horror project heavily inspired by SH2/SH3 style art. I basically want to make the Lost Highway of survival horror games.

>>583842
>Lost Highway of survival horror game
I love that shit. Reminds me of good dreams I sometimes have.

>>583764
how'd you dock your brushes where they're docked in your video? i can manage to dock Brushes all the way to the left side but it shows up in addition to the default left panel, that's too much screen real estate.

>>583834
Are you using modular walls or is that a single object?

Unrelated:
>Tfw you spend 20 hours on a 3D model and normies don't care. Nobody cares at all.
Feels VERY bad, man.

>>583868
It’s two separate floors built from modular pieces that are made to snap together; then welded into a single part. All the trim work is modular and just dropped into place.

>>583873
Wires- so I have things like a wall section with a doorway, section with a window opening. Place, then vert snap at corners and weld. Polypill of floors and ceilings, then drop in trims/model unique objects (lights, the stairs/railings).

>>583868
>>Tfw you spend 20 hours on a 3D model and normies don't care. Nobody cares at all.

External validation is a double edged sword, man. It can provide that little uptick in dopamine we all need to elevate and motivate us, or it’s lack can drive one to the depths of despair. I have been on both ends of the spectrum more than I can count, and the only way I escaped the trap was to cease caring if other people actually liked or acknowledged my work. I do this shit for myself anyway, to flush the constant self deprecation out of my head, along with everything else. YMMV but I say if your reason for doing this is because you need people to validate your work as some sort of validation of yourself, you’re probably in for a long period of feelsbadman.jpeg.

This is a retro gas pump I'm working on, any tips?

>>583861
Sorry, I don't completely understand where did you dock yours! I need an image to help you!