[ 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: 1.60 MB, 1297x752, WIP_Deja_Vu.png [View same] [iqdb] [saucenao] [google]
588889 No.588889 [Reply] [Original]

Previous thread
>>586207

I've just been in this place before

>> No.589059
File: 36 KB, 363x647, beginign weightign.jpg [View same] [iqdb] [saucenao] [google]
589059

Began weighting. Can't wait to try out corrective blend shapes.

>> No.589106
File: 414 KB, 1433x912, maya_2017-10-30_06-03-58.png [View same] [iqdb] [saucenao] [google]
589106

Gay psycho anime fish lady, part 3.

>> No.589107
File: 65 KB, 574x869, maya_2017-10-30_06-04-11.png [View same] [iqdb] [saucenao] [google]
589107

>>589106
This is taking forever to finish and it should've been done at this point.

>> No.589116

>>589106
The art shows her hands and feet being bigger. the back hair fins bigger and having more round thighs/butt. I will draw ref after work.

>> No.589166

>>589116
The original hands and feet I made *were* bigger and matched the reference more directly, but after modeling the rest of the body, I thought the size looked a little too ridiculous compared to her other proportions, and it wasn't looking good to me, so I made them smaller.

Even now I think the hands look a bit too looney tunes and could tone them down even more.

>> No.589175

>>589166
Fair enough, didn't get a good chance to look at it for the 5 min I had this morning. Just found out the character is Fairy Leviathan. Looking on Pinterest. Still want me to sketch up a model sheet?

>> No.589211
File: 667 KB, 3000x3000, screenshot005.jpg [View same] [iqdb] [saucenao] [google]
589211

>> No.589212
File: 990 KB, 3000x3000, screenshot003.jpg [View same] [iqdb] [saucenao] [google]
589212

>> No.589214

>>589175
That's okay. I published it already.

I just kinda wanna move on to something else.

>> No.589231

>>589166
I think you made the limbs too thin. In the image they are thicker. Perhaps that's why the concept proportion of the hands/feet don't match either. Also. notice those chubby fingers. You instead made her finders and hands slender.

>> No.589250
File: 809 KB, 900x970, 907662fd6580578a1a22be15879c0f96.png [View same] [iqdb] [saucenao] [google]
589250

>>589231
Yeah, I'm still practicing getting proportions down just right. But again, I'm totally burned out on it. I'll keep it in mind for the next model I make but I'm moving on from this one.

If you'll excuse the rant, the problem isn't the proportions or whatever other mistakes it has. I can fix those. The problem is something more fundamental about my attitude towards projects like this.

I'll model shit I find cool purely on a whim (usually it's stuff like this Leviathan). I always set myself a deadline of one week after I put the first cylinder down on the viewport, as I've read somewhere that it's better to have a finished project eventually than to not have a project at all. But I think I go too hard, I'll keep at it for almost the entire week, in the midst of doing uni shit too. I do it in my free time, but I don't find it unpleasant. I enjoy modeling stuff I find cool in my vacant time.

I think going that hard makes me burn out though. This always happens: I get to the point where the model's "done", it (hopefully) looks okay, but some stuff can clearly be fixed still. At that point though, I'm just so sick of looking at that model at that point, that the idea of having to go back down the chain to fix it during this honeymoon period isn't very romantic. I'll dump it on Sketchfab and move on to the next, and I think THAT's a really wrong attitude to have.

I'm very curious about this: how do you guys deal with this sort of modeling fatigue, if you even have it?

>> No.589253

>>589211
>>589212
The face came a really long way. Good work dude, I'm proud of you

>> No.589269
File: 189 KB, 883x560, 420start.jpg [View same] [iqdb] [saucenao] [google]
589269

First time making a building with anything more than a basic room for rendering objects in. Just a bunch of really quick extrusions and half assed cut guides.. but I guess that's what it's pretty much going to be in the end anyways. Supposed to be an automotive shop, but the work area turned out small after adding random shit.

>> No.589274

>>589250
Burn out is a mental muscle you must train, push yourself an extra hour each time you want to quit a model. Your focus will improve.

>> No.589277

>>589253
thanks man !! ))
but, the thing is. I have not changed the shape of a face at all. I was posting Zbrush screenshots, but now it's textured and rendered.
texturing is the key, how character will look. ( not saying that model doesn't matter)


>>589250
your model is awesome, I love it. I want to try same style someday,

btw, you are right, don't dwell on projects, you'll utilise what you have learned in the next one!!
I have that problem and it slows me down...

>> No.589280
File: 960 KB, 3000x3000, screenshot000.jpg [View same] [iqdb] [saucenao] [google]
589280

>>589277
second variant of skin shader

>> No.589281

>>589277
You haven't changed the shape since your early screenshots like months ago?? I could've sworn she looked way more butch at the beginning

>> No.589286

>>589281
oh, you mean reaaally old ones? yeah I changed a bit after that ))
I thought you've meant screens from couple of weeks ago.

>> No.589288

>>589286
Yeah I've been following it since then. Between those changes and the texturing it's looking super nice

>> No.589310
File: 1.07 MB, 3297x3822, AOC_Bakes_Done.jpg [View same] [iqdb] [saucenao] [google]
589310

Finally got the UV mapping finished, now for the texturing, my favorite freaking part. This building is twice the size of the others, so I had to figure out a way to split it so I could use 4x 2048 maps instead of two to maintain the same texel size compared to the other buildings- thankfully Bonus Tools comes with a surface area HUD that took the guesswork out of it.

>> No.589365
File: 187 KB, 1031x570, 4202.jpg [View same] [iqdb] [saucenao] [google]
589365

Messed with this thing a little more. No shop is complete without a test track out back. I have no idea how to properly engineer buildings, but it sure is fun acting like it for a few minutes a day.

>> No.589369
File: 776 KB, 960x540, bowWow14.png [View same] [iqdb] [saucenao] [google]
589369

Still needs more work. Next step: the fucking clouds.

>>589310
Loos great. I want to see it finished.

>> No.589370

>>589250
Take a break and draw. Draw up your next model you would like to do. The poses, break down mesh and props, face expressions. Then when you have gotten tired of that go back to your 3D model. Don't have more then 2 art projects on the go unless your getting paid for it.

If you cant finish it just use it to learn from. By the time your done with the facial expressions, rigging, texturing, extra, you can carry over that skill to the next project.

So far I wanted to make Liru, Sticks, Papi, Gnoll, Kobold, Stolas and over a dozen OCs. Instead I take a break and use my creativity on something else.

TLDR; If you can finish at least learn something you didn't know before.

Found this https://www.youtube.com/watch?v=KhZHjqHedPI

>> No.589380

>>589369
Gonna be a while, my texturing process is slow because it’s all albedo, and I have a family so I squeeze in time when I can. I’ll post WIP’s as I get major sections done though.

>> No.589445
File: 571 KB, 1279x728, 5c2227c5c2d18ca1c1a07f48df673e1c.png [View same] [iqdb] [saucenao] [google]
589445

Character I'm working on for Uni boi, need some feedback. I struggle so much with keeping consistency in theme, kinda feel like stuff doesn't feel right. Some space suit dude, still to add neck shit and other pieces. Any pieces to change or alter and even advice is gucci my negrumps

>> No.589451

>>589445
Why is there a hole in his neck?
Where is his ass?
Judging by his neck and the fact that a helmet will increase the size of the head, from the side view it seems that the head is a bit too small.

Take this with a grain of salt, I'm not a character artist.

>> No.589453

>>589451
Boi, thanks for the reply, uhm the hole in his neck isn't there anymore actually. He's got a rubber coat going around his neck that acts as his under suit & Head too small, goddit boi. Thanks <3

>> No.589469
File: 2.82 MB, 1280x720, Angry alien_final.webm [View same] [iqdb] [saucenao] [google]
589469

Finally finished this thing

>> No.589474

>>589469
Hell yeah, nicely done. That turned out awesome!

>> No.589476
File: 79 KB, 640x480, test.jpg [View same] [iqdb] [saucenao] [google]
589476

Just another experiment. Modeled in Quest and rendered in Povray.

>> No.589480
File: 2.35 MB, 905x905, Sam.png [View same] [iqdb] [saucenao] [google]
589480

So I am done touching geo, the character is already rigged and weight painted but I am still trying to flesh out the texture as I have time. I am pretty crap at textures this is mostly just color block outs, but if anybody has and criticisms or suggestions on what directions I could go to work in more fine detail I am all ears.

>> No.589499

>>589469
real /3/fags dont get impressed by that

>> No.589510

>>589469
That's pretty great!

If you want critique though, I think the capsule ends (top and bottom) has some weird bevels at their ends. Or atleast it looks like it does. Doesn't really look convincing due to that, the weird bevel and the flat material.

Can't say shit about the sim tho, I'm just a modeller. Looks great to me.

>> No.589518

>>589469
Looks good, but i am not so sure what i think of the editing. Seems unnecessary?
How long was the rendertime?

>> No.589525

>>589474
Cheers! Glad you like it
>>589510
Thanks, I appreciate the feedback! The modeling wasn't paid that much attention so I can see why there are some issue with it. What's wrong with the material though?
>>589499
Am I a real /3/fag?
>>589518
The editing might make more sense with the sound: streamable.com/19obz
Rendertime was about 50 hours give or take, the closeups with the smoke were a nightmare at like 30min/frame

>> No.589553

>>589525
You are right, the editing makes more sense with the sound. Good work.
>50 Hours
Goddammit you should have used Redshift NOT Mantra!

I am pretty sure you could cut that time down by using better settings, splitting the channels/render some AOV's and do some post processing on them. Smoke should render relatively fast and you could have gotten away with less samples and some blurring/denoising.

>> No.589559

>>589553
I do use Redshift! The issue was the smoke, everything else was super quick but once the smoke came in it slowed everything down a shitton. 60 million voxels so it basically just goes out of core straight away even on a 1080+1080ti. I definitely should have optimised the smoke for rendering tho, and rendering passes is something pretty new to me so I didn't exactly know how to implement them. Someone also mentioned I could have made the whole scene proxies so it loaded into GPU memory quicker, will try that next time too. What sort of channel splitting/AOVs would you suggest for rendering and comping the smoke separately?

50 hours isn't thaat bad overall for 500 frames at 1920*1080 though, if I had used Mantra it probably still would have taken 4x the time.

>> No.589563

>>589469
>streamable.com/19obz
fucking loved the thing. great job there anon

>> No.589565

Will I see an improvement in my sculptures if I get better at drawing?

>> No.589572

>>589559
You need to watch/read some tutorials on how to troubleshoot your renders.
You render out all kinds of important AOV (like Albedo, direct/indirect illumination. reflection pass,) in the IPR you can display them.
And then search them for noise.
Only increase corresponding sample accounts in channels you see noise, try to get away with as little as possible.
Noise in shadow means, not enough samples for the lights. Noise on direct illumination is probably material sampling, etc...
You must learn all these relations (don't worry its not too much) then you can analyze your scenes and optimize them.
The same principle applies to a couple of pathtracers like Arnold, Renderman and Clarisse.
That is how you use these AOV's for troubleshooting.
Then you can also render them out and manipulate them in an compositing tool as you like and built the final image from them.

In your particular scene i would have rendered the smoke completely separate with the rest of the scene, to avoid eventual interaction with the reflection/refraction of the glass. That would give me more control to render it cheaper and do something in post with it. I would also render out all kinds of object alpha masks so i could do easy color correction (change the color of the blood or the alien.)

I would suggest to invest the time to look into that topic, it might be daunting at first, but it just gives you a fuckton of flexibility and its fun too.
I can actually not tell you much about Redshift since i just have started with it, i am also a total beginner when it comes to Houdini.
But i love to play around with Fusion and Nuke (also a noob).

I am currently working with UE4 and Eevee, but i also want to built a spaceship, maybe then i have an actual use for Redshift.

>> No.589574

>>589572
Thanks for all the info man! I've actually watched a few, and I did render out AOVs (just the standard ones that make up the beauty), but didn't end up using them for much in Nuke. If you were to render the smoke completely separately how would you go about it? I'm not sure I could have done that since I still wantd it reflecting on the glass.

>> No.589575

>>589563
Thanks a lot man!

>> No.589580

>>589565
yes

>> No.589583
File: 583 KB, 1920x1080, cx10114r12uu.png [View same] [iqdb] [saucenao] [google]
589583

critique?

>> No.589584
File: 641 KB, 1920x1080, 6ddjqxlybo68.png [View same] [iqdb] [saucenao] [google]
589584

>> No.589586

>>589574
Just reuse the same camera and lights but turn the other stuff off and tweak the render settings, render, combine and fine-tune in Nuke.
To be honest i don't really see the reflection of the smoke so i would attempt to fake it.

I also know that 30 min per frame in full hd for volume graphics is pretty hefty for an fast GPU render like Redshift. I think i could do that with Arnold on my CPU which is x times slower than your GPU even in out of core mode.
So that is the place to optimize.
What GI render Mode did you use?

>> No.589587

>>589583

>>589584

experiment with more weathering

>> No.589589

>>589583
>>589584
>putting hate symbols on your 3d work
back to /pol/ you go

>> No.589614

>>589589
elaborate

>> No.589617
File: 52 KB, 656x754, 1ae4856ca9[1].jpg [View same] [iqdb] [saucenao] [google]
589617

Thoughts? Critique?

>> No.589618

>>589565
Its extremely crucial to learn how to draw if you really want to master subtler design techniques. You can learn it all without drawing, but drawings take far less time to do then sculpts and box modeling and they can get your ideas out much faster before you do your work. Example: Dont like the eyes on your character? Erase and redraw, boom done. Vs sculpting or box modeling it takes a bit more effort to change things around. Highly recommend drawing in tandem with your work.

>> No.589622

>>589617
>we wuz kengz

>> No.589625

>>589617
More observation is required, but preliminary studies seem to indicate that it's a gibs me dat, sir.

>> No.589652

>>589617
I'd suggest always try to make slightly different characters, when watching tutorials. that way, you really learn the tools you are using, because your designe will have it's own flaws. solving them with tools you are learning, will force you learn these tools.
oh, and you'll have your character, when you finish the course. not something you copied

thoughts: you've followed redbeard's tutorial's 5% well, keep up the good work. it'll teach you very much!!

>> No.589657
File: 68 KB, 846x884, 52d6f69248[1].jpg [View same] [iqdb] [saucenao] [google]
589657

>>589652
Nice eye, yeah hes a great artist.
My aim was to try and replicate his work closely mainly because I plan to recreate 2D characters into 3D works in the future and learning to replicate now would be best. I was thinking original character designs are something I should make when I have a full understanding of anatomy and stylization techniques I've gathered from emulating artists I like.

But it seems to be its own character regardless of my attempts.

Still though Im trying to train my eye for subtle detail, if you could point out a few things that are making mine different from his I'd appreciate it as I am trying to sculpt and adjust the face but I'm just not figuring it out.

>> No.589659

>>589469
Your hard surface skills are disappointing.
Shame really.

>> No.589666

>>589657
>recreate 2D characters into 3D
every 3d character artis does that. you have 2d concept and you recreate it in 3d.
that is literal go to character modeling workflow. there are people who design in 3d from start to finish, I do that, but it takes too much time. at least untill you'll be very very efficient and get over your main flaws.
in production you won't be designing unless you are concept artist.
but ecery character artist must be able to recreate 2d into 3d..

You can take some other concept art of character you like and create it with workflows from tutorials. if you are not confident that you'll br able to design it yourself.
it may take slightly more time but you'll learn much, much more from tutorial.

>> No.589668

>>589657
you should use a lighter UI if you're going to make dark colored characters.
it'll help you see it more; I don't know how to explain it but it has something to do with values and how they appear (google color theory if you want more)

>> No.589672

>>589668
wow racist much

>> No.589674

>>589666
I did not know that, I just thought that was the way that would work best for me. nNce to know Im on the right track then. Im self taught so I'm really just trying to follow my intuition and tutorials.

>You can take some other concept art of character you like and create it with workflows from tutorials

Yup thats the plan, just doing heads for now though as Ive noticed its seriously a weak point for me.

>>589668
I meant to make her white like the concept art but Im having trouble keep large cartoony facial features and not making them look black at the same time. So I just went with it for now.

>> No.589686
File: 2.98 MB, 6342x3812, Brick_Done_w_Multipass.jpg [View same] [iqdb] [saucenao] [google]
589686

Brickwork is done, this building is twice the size of the other ones I have done so I had to split it into two sections to keep the same apparent texel size without going to 4096 maps (the Vita would just die horribly if I did that, as is I may have to scale down to 1024 for that thing if I don’t do other optomiszations like enviro fogging etc to cheat)- Bonus Tools comes with a surface area calculator that was extremely helpful. The right side shows where the actual object splits are- Things came out pretty seamless. There /is/ a 0.7082% scale difference, but you will never ever notice that in game, and I was careful to clone edges across seams, etc to eliminate their visibility. Good times, glad I figured it out so that when I get around to the church I already have a process.

>> No.589693
File: 282 KB, 1500x1500, render2_1.jpg [View same] [iqdb] [saucenao] [google]
589693

>>589657
It's interesting that, judging by her hairstyle, you're probably following one particular tutorial, but that face and nose look like the motorbike girl from his another tutorial. Are you possibly mixing it up?

Anyway...

>>589652
Tell me honestly, could you tell that pic related was done with the help of Redbeard's tutorial if I removed her hair?

I posted this here a few months ago, I put a lot of additional work on her face and I think it looks completely different from the original, so I could pass it as my own at this point. But people realized that Redbeard's tutorial was my reference, (supposedly) because of her hair. Yes, I could have sculpted a completely different hair for her, but for that I felt intimidated and didn't want to fuck it up so I copied it exactly. Shirt with stripes wasn't actually me being uncreative, it just happened to be exactly the style I'd go for, even if I didn't watch the tutorial.

So tl;dr - if I ever return to this project and completely redo her hair - could I pass it as an original work and possibly put it in my portfolio?

Haven't sculpted anything new after this one, I was focusing on environments... I'm thinking about going through some good photorealistic sculpting course, although I don't think I want to focus on character art.

>> No.589698
File: 1.05 MB, 1280x720, explode_1.webm [View same] [iqdb] [saucenao] [google]
589698

>>589659
It was more a time issue, I didn't have the time to model in detail and fuck around with the UVs and texturing 2 pieces that aren't really the focus, when FX is what my reel is about anyway.

Looking for some feedback on an explosion I'm working on, mainly the characteristics. I already have some notes for myself:
- too much high frequency shredding
-Boring emission - I want to use multiple sources and have secondary impulses/trailed parts flying, as well as a shockwave, if I have the time
- Presses against container, need to make bounds bigger
-Needs to be higher res
-Add sparks with some particles

Anything else I should note down?

>> No.589702

>>589659
Wtf, how did you get to this conclusion? Seriously, I need to know because it looks fine to me.

>> No.589726

>>589686

I have a question about buildings. I am working on some buildings for my game and I'm using a material for each main texture (a material for the bricks, another one for the windows, etcetera..) and do the tiling inside unreal (because I keep my UVs in the 0 to 1 space) but I've seen some guys just placing all of their textures in a single file and moving their uv shells at the desired areas. I think this is a pain in the ass, and can cause UV overlapping. What is your input on this? What is the best way to texture a building?

>> No.589727

>>589702

There's nothing wrong with the model, he's just being a picky elitist asswipe who inspects other people's models with a microscope.

>> No.589733

>>589469
quick rundown on how you got it messy and gooey like that ?

>> No.589734
File: 195 KB, 220x258, pol boogeyman.gif [View same] [iqdb] [saucenao] [google]
589734

>>589589

>> No.589735

>>589698
that looks satisfying to render

>> No.589738

>>589726
Texture atlases are mainly used because the engine only has to call a single texture once for all objects in your scene simultaneously, saving on draw calls, additionally, you can more effectively pack multiple objects into one texture if you have a lot of long polys that can’t fit conveniently.
Texture sizes aren’t a problem on modern GPUs, but draw calls (read: latency) still are, so it’s always better to load one big texture to memory and have everything draw from it than a bunch of small textures being called in all the time.
How appropriate this is, is really up to you, tiling textures are more efficient on large-scale objects, but an atlas lets you add in unique texture details without having to lean on engine features like decals and vertex paint, in cases where they're not supported.
One obvious use of an atlas is for character meshes, where you can pack in all their clothes, items and weapons together. The more you can pack within the RAM limits of your target hardware, the smoother the game will run.

>> No.589743

>>589738

Alright then. It's easy to do on characters and simpler objects, but doing so on buildings with complex structures is really not fun at all. I'll have to get used to it then.

>> No.589753
File: 1.06 MB, 2048x2048, AA9DDB7D-6DF7-4604-8030-3F1028513F9A.jpg [View same] [iqdb] [saucenao] [google]
589753

>>589726
What >>589738 saId, exactly. Especially on mobile hardware, you want to eliminate draw calls. This building is 4x 2048 maps (two for the main building, one for the roofs/chimneys/roofline windows, one for all the other windows and trim doodads). I can compress those down quite small and still have them look good and be performant. As for UV mapping and islanding, do yourself a favor and get something like Headus UVLayout to help with cutting and unfolding and packing- it’s still a somewhat tedious task but it’s a lot less crappy with a dedicated program. Also with buildings, and most hard surfaces, it’s a lot easier to hide UV splits- you just put the seams on a natural corner- on organics you have to work harder to hide them in less visible places or painting them out, neither of which is difficult with hard surface. Good luck. Image is an example of a different building’s exterior.

>> No.589770

>>589743
Atlasing isn't commonplace in environment modeling. When it is, it's gonna be for stuff like having a bunch of building facades or windows to create distant vistas.

Obviously it's gonna be seen more on mobile, but the amount of detail required of current-gen platform games means it's not the best way to approach the creation of a lot of props. A brick house in a FPS/TPS, for example, would use a tileable brick texture, along with decals, vertex painting, or blended materials to break it up.

>> No.589800

Are Modo or Maya indie/steam editions similar to 3ds Max in anyway? Looking for something with tight control over transformations and coordinates in modeling tools, like how you can manually enter coordinates of a vertex or edge or face in 3ds Max, rather than eyeballing a move and then optionally fixing it afterwards like Blender

>> No.589804

>>589800
Also, did they get rid of free fusion 360 for hobbyists?

>> No.589805

>>589800
You can type in the coordinates to set an absolute transform for whatever you've selected in Maya (It's called the Input Box)

>> No.589809

>>589805
Between modo and maya's indie versions which is better?

>> No.589816
File: 78 KB, 1000x1250, halter coat 01.jpg [View same] [iqdb] [saucenao] [google]
589816

Been too busy lately to do any clothes modelling but today I got a little free time to start finishing up a new outfit.

>> No.589817

>>589804

Probably because so many people were producing shit with it and giving the software a bad image.

>> No.589823

>>589817
>autodesk
>image
They realized they were being too non-Jewish?

>> No.589832

>>589733
For the fluid source I made a smaller blob out of the alien to use as a source so that the fluid would be 'inside', and so the source would follow the animation. Then I subtracted everything that was intersecting/inside the alien geometry so there wouldn't be fluid particles inside the mesh, just in the holes. I added noise too so that the particles weren't emitting from every single hole. The viscosity is 85 I think, and stick on collision is on with very high stick settings, reseeding at like 48 samples i think, those are the main things. The forces on the fluid are just from the collision geometry and gravity. (Houdini)

>> No.589833

>>589816
Do you have any tips for making clothes?

>> No.589837
File: 269 KB, 1360x768, quickfix.jpg [View same] [iqdb] [saucenao] [google]
589837

So I had to redo the eye and mouth mesh. they did not deform well. Should i keep the eye wider left or narrow right. should i make the top eye lash bigger. Should the eyes be bigger?

>> No.589840

>>589584
maybe add leafs at the bottom of the fountain. Are you going to add water? It looks great.

>> No.589842

I think I will not add any water atm. but the leaves on the ground are a good idea, thanks!

>> No.589863

>>589833

Grow up a closeted homo who secretly wished to be a fashion designer but lived in a small rural town and knew telling people his secret desire would get the ever loving shit kicked out of him. Then spend years wishing you could sew and make your own clothes and pining away because you know if you ever expressed an interest in learning to do something girly like that you'd lose all your friends and get beat up. Then come out later in life and discover that 3D modelling will let you indulge in your childhood fantasy without actually having to know how to sew in the real world so now you can bring the images in your head to life in a virtual world.

Either that or read Jack Hamm's art tutorial book Drawing the Head and Figures which has a section that covers how different fabrics drape and wrinkle on the human body and then translate that knowledge into 3D.

>> No.589868

>>589657
Ugly neggress

I'd have no motivation to make a model like this

>> No.589874
File: 125 KB, 1680x1050, face-1.jpg [View same] [iqdb] [saucenao] [google]
589874

>>589657
Why are the eyebrows so far up the forehead? You know those things are supposed to sit right in the edge of the eye sockets, right?

>> No.589876

>>589693
hair gives it away )) nothing else.
I was kidna defending you, when you posted that image BTW

>> No.589877

>>589863
or use marvelous designer

>> No.589878

>>589586
Sorry didn't see this post. Used brute force GI, not ideal I know but the GI actually didn't contribute that much to the rendertime compared to everything else. Just realised I also really needed the volumetric lighting on the alien once it was enveloped in smoke, would have been a lot harder to fake that. The extra rendertime wasn't that big a deal honestly, I had the time to wait, but next time I'll do some more optimising.

>> No.589880

>>589877

Well.... yeah.... if you WANT to avoid the whole tortured gay childhood route... I guess that works too.

>> No.589884
File: 70 KB, 1280x720, IMG-20171104-WA0001.jpg [View same] [iqdb] [saucenao] [google]
589884

here is a new version with leaves

>> No.589923

>>589837
never mind figure out what went wrong.

>> No.589950
File: 385 KB, 2297x1816, sculptwip.jpg [View same] [iqdb] [saucenao] [google]
589950

WIP of this crab alien/monster thing I'm working on in ZBrush. Have just blocked in the armour, going to work on that tomorrow.

>> No.589951

>>589950
hands look too human compared to the rest of the design imo

>> No.589955

>>589951
Thanks anon, I'll keep that in mind to see if I could do something about that, his initial design started out as very gorilla-like so I maybe let the primate influence take over a little too much in his hands

>> No.589966
File: 54 KB, 539x660, GoodJob.jpg [View same] [iqdb] [saucenao] [google]
589966

Plenty of Flaws, not done weighting, no blend or corrective blend shapes. Gave up on texturing. Outline doesn't look that clean ether.

">My face when I can pose it."

>> No.589967

>>589950
seriously, a crab without a bad-ass claw?

>> No.589970

>>589950
Give him tiny little crab legs on his body.

>> No.589976

>>589950

Too symmetrical.

>> No.589984

>>589966
anon i like her but can i say that pls her shins and feet look like they get a bit too skinny

>> No.590013

>>589966
I really like this ))
bonus points because you are making full character!!

why did you gave up on texturing?

>> No.590024
File: 146 KB, 986x519, shine_ankel.jpg [View same] [iqdb] [saucenao] [google]
590024

>>589984
The legs are suppose to tapper down but it did look a little to spindly. Hopefully I don't have to re-rig the led. Want to move on.

Unrelated.

Also found this site and model. http://www.cadnav.com/3d-models/model-38661.html Wanted my next one to be the same height but thinner thighs this time and little more Tits and ass. Also more mature face. Maybe make another anthropomorphic. I find Elin goes from cute to creepy very easily.

>> No.590025

>>590013
Getting tired. Can you UV map after you done wieghing. I think there is a way http://forums.cgsociety.org/archive/index.php?t-538778.html

If that is true then I will put it off. Want to do something fun like blend shapes so I can stop looking at a dead puppet.

>> No.590028

>>589967
>>589976
Yeah, you're right. 3rd 3D project, still learning how to not make it look like complete crap. I'll add some asymmetrical details once it's a bit further along and keep your advice in mind next time I do a design, it's not really a strong suit. Thanks Anons.

>> No.590031

>>590025
Do the unwrapping, it has nothing to do with weighting.

>> No.590032

>>590024
your character is a mouse and she has stylised anatomy. she can have legs like that. don't listen to everyone who gives you suggestions.
Don't change them, it'll look bad.

>>590025
just UV map the model. I don't think that it'll have any peroblems.

oh and if you are tired already from this characeter. you can move on to something new.
you'll be able to make new one in much less time. so MOVE ON man!
is this your first model?

the one problem your character has is: that she has no tail

>> No.590042

>>589950
Lol, looks like shit.

>> No.590043

>>590032
forgot to rig the tail it.

>> No.590061

>>589950
I think it looks pretty good. It certainly looks like you've put a lot of effort into it already. The only thing I found kinda funny is that the face looks a bit derpy, it's closed in the reference but it looks like you've made it gape open a bit?

>> No.590066

>>590061
Yeah, the face does look kind of stupid. I was told to sculpt it with the mouth open slightly to make it easier to do the inside/for rigging later as I wanted to make a full project out of it, not so much now, I might just move on to another project. I have a layer that lets me sculpt with it open and closed, but the closed position needs a bit of work.

The torso was too noisy/had too much going on so I brought it back to an earlier stage from a few days ago, I'll redo that at some point. Thanks for the critique, I appreciate it.

>> No.590204
File: 3.78 MB, 5120x2820, HighresScreenshot00013-min-min.png [View same] [iqdb] [saucenao] [google]
590204

How do you think the bed I made looks?

>> No.590205
File: 1.60 MB, 3226x1630, HighresScreenshot00015-min.png [View same] [iqdb] [saucenao] [google]
590205

Another angle

>> No.590206

>>590204
>>590205

That blanket pattern looks like something straight out of an Ukranian love motel. Lighting looks really weird, some shadows look soft while others look way too hard, and some of them are overlapping in weird ways (look at the corner of the room in the second screenshot). The whole room is just way too dark, it feels cramped and weird, and the purple-ish hue isn't doing it any favors. Your antialiasing samples are also really low, the jaggies are super evident in the shadows.

I'd try to not put the bed in the corner or at a weird angle in the middle, put it center-aligned, with the back touching a wall. Same thing for that little chest, don't put it at a funky angle somewhere in the room, align it properly and put it next to the bed or something. The table's also way too big, it doesn't look like something that belongs in a bedroom. The bedsheets just look ugly to me but that may be personal preference, so I'll avoid commenting on it.

>> No.590210

>>589863
lmao I hope this is larping

>> No.590211

>>589469
Please please please turn down that CA

>> No.590213

>>590205
just wanted to say the thumbnail looks like a tounge licking an asshole

>> No.590239

>>590213
this

>> No.590258

>>590213
this !!!
and first also has same feel ))
I really don't know why...

>> No.590297
File: 1.44 MB, 1920x1080, screenshot001.png [View same] [iqdb] [saucenao] [google]
590297

started to work on a new Ice material.
Opinions so far?

>> No.590333

>>590297
Reminds me of slush puppies.

>> No.590345
File: 1.84 MB, 1280x720, SceneT.png [View same] [iqdb] [saucenao] [google]
590345

a scene im working on, would appreciate any kind of critique

>> No.590378
File: 36 KB, 607x605, fk.jpg [View same] [iqdb] [saucenao] [google]
590378

i got scared of failing and closed zbursh ........ should I try again?

>> No.590383

>>590378
Erase and start again. Lose that fear, its only digital clay. Plus you can ask for critiques anonymously if you feel you're stuck. Remember to stack yourself on references.

>> No.590385

>>590378
>>590383

DON'T FUCKING ERASE IT, save it somewhere so you can have a record of your progress.

Jesus, don't throw away your work, no matter how shitty it is. Years from now, you could look at that bust and feel good about how far you've come.

>> No.590386

>>590345
I love the feel and colors. Is this a realtime thing?

I would say the one thing that's really an eyesore are the UVs on the side of that broken slab on the bottom left, my eyes immediately went to that.
And I'm not sure I like the idea of depth based grain, if that's what's going on. Leaves the foreground too clean and the background too noisy. I would at least make it a bit more homogeneous.

>> No.590389

>>590385
Lmao calm your tits, I meant starting a new one. He can keep this very screenshot or the whole model too.

>> No.590392

>>590389
Why the fuck would he start a new one? That looks pretty decent to me. It's still very basic anyway, just keep working on it and adding more and more details, why are you scared? Keep looking at references and you'll progress.

>> No.590394
File: 949 KB, 1920x1080, Ice01.png [View same] [iqdb] [saucenao] [google]
590394

I think I am done here:

>> No.590396

>>590394

Would lick

>> No.590402

>>589657
>already making black women
make it an afro and you are hired

>> No.590406
File: 83 KB, 640x480, test.jpg [View same] [iqdb] [saucenao] [google]
590406

Just another experiment. Modeled in Netradiant, rendered in povray.

>> No.590418
File: 144 KB, 640x400, out.webm [View same] [iqdb] [saucenao] [google]
590418

Programmer learning to do the art stuff.
WIP - main dude for my game. First time sculpting after watching a bunch of videos on a tutorial site. Still a long way to go. Just started him this morning. Made a lot of mistakes along the way. Fun to learn though.

>> No.590440

>>589966
So your gonna make a NSFW version of this right?

Id like to see her getting fucked or shoved into another characters vagina.

>> No.590446

>>590418
neato

>> No.590447
File: 41 KB, 798x766, 16f016e5ca[1].jpg [View same] [iqdb] [saucenao] [google]
590447

Proportions are my weak point, love to hear some critique.

Supposed to be a young girls body. No idea on the age honestly. Its the new LoL hero and I wanted to try and replicate the design.

>> No.590457

>>590440
And the vore furs come out to play.

>> No.590461
File: 71 KB, 640x480, test.jpg [View same] [iqdb] [saucenao] [google]
590461

Some more work and different textures

>> No.590464

>>590440
Wait fucked into a vaginaor shoved into a vagina? Or do you mean fucked, or shoved into a vagina?

>> No.590475

>>590464
I want to see EVERYTHING

>> No.590486
File: 1.72 MB, 3360x1079, PureRef_2017-11-08_06-49-26.jpg [View same] [iqdb] [saucenao] [google]
590486

Bartender Lesbian - Part 1:

- Gathering References

- Setting up a base mesh

- Putting on some comfy music

>> No.590489

>>590464
Why not all three?

>> No.590492

>>590418
Good show dude! I am plugging through sculpting too. Very fun but I lose focus too quickly. Need to keep/find me reffs when doing it to stay focused.

>> No.590494

>>590486
Wow thats pretty damn cool I never thought about trying to set up a mesh over a pic, is that for reference or are you going to use that as a base mesh?

>> No.590495
File: 362 KB, 192x108, 0001-0270.webm [View same] [iqdb] [saucenao] [google]
590495

>> No.590496

>>590495
I don't like it. It's scary and I don't like it.

>> No.590500

>>590492
Thanks, man! Make sure to post some of your stuff when you're ready to show it off.

>> No.590502

>>590496
why?

>> No.590504

>>590502
The colors all change to red all of a sudden and it's scary. Red's a scary color to just shove on some innocent poster out of the blue.

>> No.590515

>>590495
I don't like it either. It's very low-res and that is scary. Low-res's scary to just shove on some innocent poster's eyes out of the blue.

/s

>> No.590570

>>590024
Dat mouse ass

>> No.590576
File: 574 KB, 1280x720, 0001-0270 (1).webm [View same] [iqdb] [saucenao] [google]
590576

>>590504
>>590515
better for you creampuffs?

>> No.590580

>>590486
>designdoll
Why have I never heard of this

>> No.590583
File: 1.66 MB, 1915x910, maya_2017-11-08_22-05-22.png [View same] [iqdb] [saucenao] [google]
590583

Bar Lady - Part 2:

- Plastic surgery'd the face into shape.

- Modeled out a base hair, eyebrows and eyelashes.

>>590494
I match the body to the reference as best as possible and then bring that into Maya to tweak, retopo and model on top of.

>>590580
Because it's advertised as a drawing aide a-la 3D mannequins in CLIP STUDIO PAINT, and not as something to be used by 3D character artists. The fact that it can even export the posed mesh in .obj is a miracle.

>> No.590597

>>590576
Now that is pretty cool. But as always, y'all fuck up your lighting, it could be better.

>> No.590598

>>590576
Why'd you change your funky psychodelic work based on some personal opinions from some nerds on the internet?

You're the creampuff, sunshine.

>> No.590617

>>589589

fucking die normie cuck, where do you think you are

>> No.590629
File: 415 KB, 1264x590, TubeChassisClusterfuck.jpg [View same] [iqdb] [saucenao] [google]
590629

Starting a tube chassis to fit a ferd 5.0 / general BMW guidelines for subframes/diff. It's pretty much just a clusterfuck of different designs mashed into one at the moment, and still needs support to shock towers.

>> No.590631

>>589589

back to /selfhatingwhitemale/

>> No.590653

>>590598
>psychedelic
it was just a water texel for a voxel game.
rainbow wasfor height info.

i did the ice cream version for when the water is milkshake.

>> No.590693

>>589589
lol

>> No.590697
File: 70 KB, 1418x696, 386fecccb0[1].jpg [View same] [iqdb] [saucenao] [google]
590697

Preparing to finalize this before starting on the head and clothes.

>> No.590726

>>589657
Sheeeeeeeiiiiiiit.

>> No.590788
File: 522 KB, 1460x732, Untitled.png [View same] [iqdb] [saucenao] [google]
590788

>> No.590805
File: 2.19 MB, 1920x1080, screenshot000.png [View same] [iqdb] [saucenao] [google]
590805

I am working on some sort of tile in Substance designer. Please let me know your opinion on the shape i have so far...Thanks in advance

>> No.590808

>>590805
It doesn't look realistic from a shading standpoint but more importantly it doesn't look like a design I'd ever see in real life.

Are you looking at references or trying to go by memories of the general idea of a tile? Because our memories suck.

You know what a penny looks like, right? Or whatever common coinage you have in your country?

Can you remember which way Lincoln is facing? Where the date goes? What it says other than In God we trust? What he's wearing? The number of pillars on the building on the back? I can't. But I know what a penny looks *like*.

>> No.590809

>>590805
no place to insert medal/10

>> No.590812

does it has to look realistic? just looking for something that looks pleasing to be honest instead of just copy some standart ornaments

>> No.590818
File: 889 KB, 757x756, Untitled.png [View same] [iqdb] [saucenao] [google]
590818

>>590812
Well, it doesn't have to look realistic, no. But it doesn't really look pleasing either imo. Can you add some AO or collected grunge to the details to make them stand out more, like an outline? That should help.

And the design, let me see if I can analyze why I don't like it.

Red doesn't look right, just this weird rectangle on there is so strange. Pink is too close to another bevel, it's cramped and even distorts a bit. Goldenrod isn't aligned to the center of the corner despite being close to it. Cyan is messy and distorted, possibly overlapping elements. Green is too thin.

http://www.neilblevins.com/cg_education/primary_secondary_and_tertiary_shapes/primary_secondary_and_tertiary_shapes.htm Give this a once-over, general idea is to break elements down by size into primary, secondary, and tertiary, and balance them while also being sort of random.

You should still look at real tiles though because if someone bothered to take a photo of it, chances are it looks pretty good for a tile and has design elements you'd be interested in using.

>> No.590819

I think I agree with you. I just experienced with shapes without a real concept and I somehow tried to make it look good, which was the main mistake in my opinion.

I will start from scratch. Thanks.

>> No.590820
File: 2.36 MB, 1920x1080, too_many_balls.png [View same] [iqdb] [saucenao] [google]
590820

>>590805
I think some procedural aspects are overlapping and it's actually fairly noticeable. I'd get rid of the "excess" if that's even possible.

This sort of thing looks really unnatural in metal too. Try wood, maybe? I think it would fit a lot more.

>>590788
That looks pretty alright. Is that bad boy supposed to be broken down?

>> No.590847

>>590583
Wait that can be exported as an .obj?!

>> No.590856
File: 281 KB, 1476x877, Fub3D-wip1011.jpg [View same] [iqdb] [saucenao] [google]
590856

>> No.590914

>>590856
How many booties do I gotta model to make a booty like this

>> No.590922

>>590856

Shiki Ryougi is that you?

>> No.590923

>>590856
>FULLY MODELLED

>> No.590926

I CAN'T SCULPT ROCKS.

I CAN'T DO IT I JUST CAN'T DO IT NO MATTER HOW MANY TUTORIALS AND COURSES I GO THROUGH. I CAN SCULPT ANYTHING EXCEPT FUCKING ROCKS. WHAT THE FUCK DO I EVEN DO NOW? I CAN'T COMPETE WITH PHOTOGRAMMETRY.

>> No.590934

>>590926
https://m.youtube.com/watch?t=781s&v=UFsVzDXYOKw
If you use blender, check this shit out
(It's also possible that you could remake this in a different software, it would just be a pain in the ass)

>> No.590940

>>590934
Of course I've seen it, it's probably one of the best produceral generators I've seen. But that's the problem - I heard many times people say that rock sculpting is one of the easiest things to do - but when they show their work it looks like trash. Making something that "kinda looks like rock" is not hard (and this generator is perfect for that), but I'm talking about good - fuck good - great rock pieces with interesting details and contours. That is hard as fuck to achieve. Sure, I can use alpha masks or just make some chips and pinches everywhere, but the end result will still suck. Actually, I have yet to see a tutorial where the end result is great, and I've watched dozens of them, anything I could find.

>> No.590941
File: 315 KB, 1422x877, Fub-dress.jpg [View same] [iqdb] [saucenao] [google]
590941

>>590922
No, it's Fubuki from OnePunch Man.

>> No.590944
File: 206 KB, 899x897, 1.png [View same] [iqdb] [saucenao] [google]
590944

Yes, I know the ears are shitty

>> No.590953
File: 248 KB, 1910x912, maya_2017-11-11_14-56-01.png [View same] [iqdb] [saucenao] [google]
590953

Girl With Girlfriend Problems - Part 3:

- Retopo'd body, except for feet.
- Fixed some proportions.
- Finished adding hair in scalp area.
- Added ponytails and a few loose strands of split hair.
- Specular highlights in eyes, dark outline around irises and body.
- Modeled in nose and mouth (temporary) geometry for now.

>>590847
Yeah, but the mesh isn't very good. It's all triangulated for starters, which means it doesn't play well with something like zBrush. There's too much emphasis on belly fat, even on some of the more barbie-esque shape modifiers, and using those sacrifices the mesh's ass, which is already lacking a lot in the volume department. The face shapes are honestly terrible and rarely match reference images, which increases the time you have to spend manually adjusting it after you export it. The models also start in this weird spread-out leg position with their arms sitting neutral next to the torsos, so you have to manually put it into a t-pose after you're done adjusting proportions before/while posing it. Camera controls are also bad and don't let you adjust image references in any way in the viewport.

It beats starting from scratch, but there's a lot of shortcomings and you're still doing 90% of the work at the end of the day. I wouldn't mind switching to a better software specifically designed for this (but again, can't really blame DesignDoll, it's meant to be used as a drawing aide).

>>590941
Sugoi. How can box modeled girls even compete?

>> No.590964

>>590926
>>590940
so just use photogrammetry then? what's the issue? or pick up some actual rocks lying around and use them as reference material and sculpt them manually as close as you can get.

>> No.590976

>>590964
I mean, yeah, that would be ideal. But there are 2 things that are stopping me from doing that. I literally can't find good rocks and cliffs in my area so it's not easy to get the resources on a regular basis. The other thing is that as an environment artist (well, I'm not the one, but maybe one day), I have to know how to sculpt my own rocks. Anyway, like you said, I just to need to keep trying until I start getting good at it. It sucks though that the rocks were the assets I tried modeling and sculpting the most and every one of them still looks like shit.

>> No.590978

>>590976
there's also the possibility of just lowering your standards. if everyone else besides you seems to think they look fine, maybe they really do look fine and you're just being needlessly autistic.

>> No.590979

>>590926
Well, you could post your rocks so we could take a look.

>> No.591053
File: 1.15 MB, 1408x1642, XPR.png [View same] [iqdb] [saucenao] [google]
591053

Light years away

>> No.591306
File: 1.00 MB, 512x512, forest_entrance_wire.png [View same] [iqdb] [saucenao] [google]
591306

>> No.591360
File: 461 KB, 2327x836, Screen Shot 2017-11-13 at 11.30.00684 AM.jpg [View same] [iqdb] [saucenao] [google]
591360

A small update, working on the facade.

>> No.591363

>>591360
You sure about the Coca-Cola brand on the texture? Typically people remove things like that, but I'm not sure why.

>> No.591365

>>591363
It’s a placeholder while i make a mock soda sign that’s similar- I may do something like slip in Cadre Cola (a reference to the soda advertised in ‘The Running Man’, the 80’s Schwarzenegger flick)- generally that sort of stuff is left out or altered because of copyright/trademark issues, especially for a commercial work, although Fair Use laws can be used to get around it.

>> No.591373

>>591360
Oh god, please match the striped pattern on that cover. Think of the poor folk with OCD.

>> No.591378
File: 1.91 MB, 2560x1536, 10F036A6-1168-4547-924C-76663320CFB4.jpg [View same] [iqdb] [saucenao] [google]
591378

>>591373
That is literally how it is on the actual building. If you’re talking about it not lining up near the front, it actually does, the polys for each section dip in like the cloth is sagging on the frame.

>> No.591379
File: 857 KB, 2327x836, fuck.jpg [View same] [iqdb] [saucenao] [google]
591379

>>591378
>each section dip

Oh, the cloth dips downward. I thought this was a hard edge or something.

I have bamboozled myself. Carry on, then.

>> No.591409

>one day accidentally fuck up my (pretty great) lighting
>the next day I open the project and continue working on it, still not being satisfied because it looks a bit off
>2 days pass, finally realize what I did
>fix it, the difference is so mind-blowingly huge that now I'm a bit dissapointed in myself because I don't understand how I managed to work on something, fuck it up, and then not realize when it looked 560% worse. How did my standards suddenly drop again? Woah, I need to be careful of those things...

>> No.591420

>>591409
Open incremental save (you do have it on, don’t you?), export light rig, import into newest version of scene, or delete everything but the lighting rig and import content from newest scene, depending on if you are just talking viewport lights (first option) or lights and render settings (second option).

>> No.591425
File: 3.08 MB, 1920x1080, screenshot001.png [View same] [iqdb] [saucenao] [google]
591425

Hey guys, here is the guy with the weird Tile shape. I did everything again from scratch.
As your last comments were really helpfull (thanks for that), please share your thoughts on this with me.
Thanks in advance.

>> No.591474

>>591378
the top part of the awning that sticks out of the building looks off.

>> No.591494

>>591474
How so?

>> No.591540
File: 2.39 MB, 1832x660, Capture.png [View same] [iqdb] [saucenao] [google]
591540

>>591494
the sections don't match up.

>> No.591542

>>591540
Again, that’s the cloth drape. There is a frame of square steel tubing under it that makes those sections

>> No.591552

>>591542
i don't care what's causing it.
it looks like shit and you should fix it.

>> No.591554

>>591552
>what is perspective
autism speaks

>> No.591555

>>591552
It’s not broken. Unless the material was perfectly taut on the frame, which it isn’t, the only position in which it’s gonna look perfectly straight is from directly above.
>>looks like shit
If you modeled a striped comforter on a bed that was balled up and the stripes didn’t (unrealistically) line up perfectly, would you also say it was shit?

>> No.591591

>>589589
4chan is /pol/

>> No.591611
File: 1.68 MB, 2048x2048, img.png [View same] [iqdb] [saucenao] [google]
591611

Almost finishing. He has no hands and feet yet.

>> No.591618
File: 74 KB, 553x757, megumeme.png [View same] [iqdb] [saucenao] [google]
591618

I especially suck at texturing faces so I guess I`m done.

>> No.591620

>>591554
>>591555
you both clearly can't see the fact that it's 100% unnatural looking.

it should be fixed.

>> No.591622

>>591618
Since it looks like you are already going for a chibi anime thing it might be worth trying to do the face in illustrator. I hate the program but it dose seem to do the job well for people like me who are not actually capable of 2d art. Just pull in some reference you like; a mouth from here, a eyes from there. Vector individual parts, change it enough to be it be its own thing. And just trial and error placement till it looks not derped, also use the reflect/mirror tool.

>> No.591625
File: 110 KB, 634x679, megumeme.png [View same] [iqdb] [saucenao] [google]
591625

i only have enough motivation left to rip textures from source and stick it to map it to planes

>> No.591629
File: 300 KB, 1000x540, original_lighting.jpg [View same] [iqdb] [saucenao] [google]
591629

I hate the fact I have to show this already, but I need your help since I realized I spent 3 days doing nothing except tweaking lighting.

First, I'll post the screenshot from which I'm trying to recreate the lighting. In short, I didn't save it, and the next day it was fucked up, and now I can't get the same look no matter what I do.

So actually it might be good to ask if it is worth recreating, because maybe it's delusional to think this was a perfect setup, and it could still be a lot better.

I'm using UE4, btw.

>> No.591630
File: 398 KB, 1920x1004, other_versions.jpg [View same] [iqdb] [saucenao] [google]
591630

Here are a few other versions I got in the meantime. The main differences are when I switch between scene capture and custom cubemap in my sky light settings, and depending on the intensity of skylight, I'm getting more or less details on the snow.

You can notice every one of them is off by that wood log (it lost color on the bottom, and is now all white-ish, the amount of light getting inside of the spruces, and lighting on the back of the playable character.

So, basically I'm tweaking:
>sky light intensity
>turning lower hemisphere on/off
>switching between stationary and movable skylight

I can never get the same effect, though, and I'm pretty sure those were the only parameters I might have accidentaly changed.

Btw, I'm interested why at first I was strongly pulled toward the look on the top left. Scrolling through the screenshots, that one seemed the most photorealistic for some reason. What I think would be the reason, is that the details are much less visible, so there are less things that could show that it's a render, but at the same time, the "look" itself CAN be seen in a real world, it's only that it looks like it was lit by some giant reflectors so the colors are burned and everything is flat.

Anyway, not sure what to do now, and in which direction to go.

>> No.591638
File: 745 KB, 1076x786, Fubuki3dwip.png [View same] [iqdb] [saucenao] [google]
591638

>> No.591639

>>591630
>>591629
I think I found the solution! I realized that dynamic GI stopped working, and config file removed the needed line of code to activate it in the editor.

Oh well, if anyone has some general feedback for pics related, I'd still gladly hear it.

>> No.591640
File: 1006 KB, 751x1080, 1510054040853.png [View same] [iqdb] [saucenao] [google]
591640

>>591638
Do Garou next.

>> No.591642

>>591640
Maybe someday, but he's not on my " to do next" list for now.

>> No.591660

>>591611
aren't his arms a bit too thin from the front?
>>591618
how do you expect to get better if you don't even try?

>> No.591661

>>591625
Just cut geometry out for the eyes and flat shade the shapes, as you've done with the hair.

>> No.591669
File: 1.63 MB, 1920x1080, Cube.png [View same] [iqdb] [saucenao] [google]
591669

Here is the textured version

>> No.591671

>>591669
this is supposed to be ceramic type flooring right? seems like it could use less roughness

>> No.591672

>>591669
Looks like some comfy cloth now, something straight out of Etrian Odyssey.

The quality has definitely improved.

>> No.591676

>>591671
>>591672
As they're saying, it looks nice, but it's not clear what material this is just by looking at it.

>> No.591677

Its some sort of Deco Tile material. It might be difficult to read as this is my own design again...
To be complatly honest with you, I am not sure neither were I would use this, maybe in some old grandma kitchen...

>> No.591679

>>591669
Reminds me of the paint textures from SPORE, Spore 2 when????

>> No.591681
File: 3.91 MB, 1838x1080, Closeup.png [View same] [iqdb] [saucenao] [google]
591681

Just fyi, here is another closeup view

>> No.591683

>>591681
the tiny bumps look too strong

>> No.591686
File: 55 KB, 640x480, 1474681538740.jpg [View same] [iqdb] [saucenao] [google]
591686

>>590447
Since no one else did, I'll hit this up.

So if you're not, adjust these proportions using the transpose master (you're in Z, right?). Don't adjust the proportions of each piece individually. You'll lose sight of the end goal that way, having to stop to adjust multiple parts just because you've fixed one.

Second, delete some meshes. You don't need individual low poly spheres for the breasts. Use the ribcage as a singular form here, especially since its a younger character with a less volume in her chest. Similarly, weld your shoulder... ball and your upper arm together. They are one piece in real life and should be one piece in 3D as well. If you want to keep a loop for the definition in the deltoid do that as well.

Thirdly, pose. POSE. POSE ANON. POSE. There is nothing worse than trying to make an unnatural pose feel like a natural character. Drop the arms a little bit into more of an A pose rather than a T pose. Rotate the legs out from the hip and firmly plant her feet on the ground, also making a wider A. Posing varies in what animators/riggers want, but since this is just you fucking around, go with the more natural A pose.Literally almost everyone else in this thread working on characters has this. Don't learn to model like you're working in the early 90's. You're in Zbrush in 2017. Go nuts.

You don't get proportion feedback until your blocking phase A) looks like something, B) you fix your modelling practices first.

>> No.591745
File: 418 KB, 1000x1000, larva.png [View same] [iqdb] [saucenao] [google]
591745

one day, this will be a beautiful butterfly

>> No.591748

>>591745
shit man that's awesome.

>> No.591750

>>591748
:) I really cant wait to animate it!

>> No.591751

>>591681
Now it looks like a grungy bathroom tile. It has lost all of its comfy charm.

You didn't have to slap a weird normal map on it. It honestly looks like a good pattern as-is, even though you don't know what to use it for.

>> No.591775

Who added the 86 to the WIP photo? That thing looks pretty decent. We need more automotive related /3/ content

>> No.591776

>>591775
That's from 3dtuning.

>> No.591777
File: 521 KB, 1426x1809, blueprint.png [View same] [iqdb] [saucenao] [google]
591777

>>591053
Redid a lot of stuff. Also tried freestyle.

>> No.591781

>>591629
Would still appreciate some feedback on textures and lighting.

And grass on these >>591630
But I think I'm gonna redo it once again.

>> No.591788

Its a closeup view of the tile, therefore the normal map details are far more prominent, on further distance like on the cube it is not that visible and just gives some base noise

>> No.591789
File: 3.63 MB, 1896x1080, screenshot000.png [View same] [iqdb] [saucenao] [google]
591789

Anyhow, it still looks way to intense, I reduced the Normal map strenght quite a bit:

>> No.591790

>>591789
if ti's supposed to be linoleum, why is the design bumped?

just use a linoleum normal map.

>> No.591791
File: 112 KB, 1024x683, aomori_20160205_b.jpg [View same] [iqdb] [saucenao] [google]
591791

>>591781
Well, texture and lighting wise it looks pretty good, but the grass looks like seaweed, in winter grass has to be entirely dried up if there's any at all, like the little stick things you have around. When it's cold enough snow doesn't melt when it touches anything, so everything should look very dry, as all the water in the atmosphere is frozen.

Also, you got active snowfall in your scene, and things get increasingly veiled as there is more and more snowfall as per pic related. You could try adding some sort of post effect that grades the image to have reduced contrast, and maybe add a very soft but also very wide bloom that kind of permeates the whole screen, so when there's more snow, your whole view of the screen gets kinda fogged up. It just doesn't look right when you have a crystal clear view of everything when there's snow falling.

>> No.591793

>>591789
>>591790
From what I can tell, the way the current heightmap is set up, the texture is battling between having a distinct surface structure and looking like the type of surface this sort of pattern would be used on.
First, there are no dividing lines nor grout in the pattern, so it doesn't look like tiles, you just can't have one big contiguous tile.
And second, whether this is a tile or surface pattern, it's not supposed to have the different elements be at different heights.
If it were a tile, it would be clear-coated and polished , so everything up to the edges of a tile segment would be largely smooth. If it's a non-coated tile, then it's supposed to be really rough looking, like if it were sanded. Yours currently looks like it has cut stone shading, but it's not laid out like cut stone.
If it's a pattern, then it's definitely not supposed to have a height map and simply follow the shading rules of whatever type of material it's printed on.

>> No.591795

>>591791
Ah, interesting thought about bloom, didn't know that. Yeah, I agree with what you said about grass, I'm planning to do just that today. I tried making it look somewhat like the grass from Horizon Zero Dawn, but actually I don't think I want it like that here, but more like pic related. Thanks.

>> No.591796

>>591795
Although ,I want to mention that I do have a pretty strong fog, it's just that a starting distance is probably too far away, but you can see that around those super tall spruces things get completely invisible. Okay, I'll pull it back a bit.

>> No.591797
File: 1.41 MB, 1920x1080, GC_Kholat_Screen_04.jpg [View same] [iqdb] [saucenao] [google]
591797

>>591796
You don't want a strong fog per se, you want a "thick" fog that feels like it starts obscuring things right out from arm's reach, but gradually fades everything, still allowing you to see things that are somewhat far at the same time, rather than being a brick wall somewhere off in the distance that just disappears geometry when it's too far, that will just make your project seem last-gen and hokey.

What I'd suggest doing is checking out and maybe even playing a game called Kholat, which has arguably the best snow effects in a recent game and it was made in UE4. Gonna dump you some screens to whet your appetite.

>> No.591798
File: 367 KB, 1280x720, 513776473bded5d568fb4c523439ce690b4573d8.jpg [View same] [iqdb] [saucenao] [google]
591798

>>591797

>> No.591799
File: 121 KB, 1024x640, 5912829dae653aefac27e286.jpg [View same] [iqdb] [saucenao] [google]
591799

>>591798

>> No.591800
File: 120 KB, 490x647, head model practice.png [View same] [iqdb] [saucenao] [google]
591800

Send help.

>> No.591801

Thats kinda funny as i originaly wanted ro build a ceiling material. during the process i felt like this is looking more like a tile and switched into that direction, which is likely causing this weird mixup. it looks somehow pleasing but something is off.

>> No.591803

what i could do of course is to place a prominet cut between tiles and rethink how the surface is reflecting the light

>> No.591804

>>591801
Learn to quote, newfaglet.
(Press on the post number of a post you want to reply to).

>> No.591809

reply does not work on my mobile phone :(

>> No.591810

>>591809
Mimi 4chan reader is good if you're using Android. But oh well, if you can't do it, it's ok, just a bit confusing in a conversation.

>> No.591831

>>589727
>being this mad

>> No.591840
File: 171 KB, 1692x1189, fubooty5.jpg [View same] [iqdb] [saucenao] [google]
591840

Was taking a few shots of the body of my 3d Fubuki.
https://imgur.com/a/COTEP

>> No.591841
File: 466 KB, 960x540, 1510855045281.png [View same] [iqdb] [saucenao] [google]
591841

Looking good. Now I'll add some materials for the character, rig and animate

>> No.591844

>>591840
>https://imgur.com/a/COTEP

thicc

>> No.591845

>>589445

The "all roads lead to dick" effect of the front armor panel is a little disconcerting

>> No.591847

>>591840
That shit is hot as FUCK. Will you texture her some nipples or something?

>> No.591848
File: 82 KB, 936x1200, fubuki3front2.jpg [View same] [iqdb] [saucenao] [google]
591848

>>591847
Probably, yeah. As you must have guessed, it's still in progress. Mainly have the underwear to texture, the face to finish, and the hair to do, which are the last big pain in the ass since I don't have much experience with them.

I'll also have the model fully available for viewing on Sketchfab, which is where I also took those screenshots.

>> No.591850
File: 34 KB, 625x749, rock.jpg [View same] [iqdb] [saucenao] [google]
591850

Damn, the appeal of this rock constantly changes after every few brush moves. Am I at least going in the right direction? At least I think I finally "got" the technique, so it can only get better.

>> No.591851

>>591850
Btw I want to achieve a realistic look, not some stylized shit. I know it has to have tons of details, I'll add th

>> No.591852

>>591851
them* later. But I always feel like I'm close to making it look like a turd, lol.

>> No.591853

>>591852
how big is thios rock?
if it is smaller than a car you are overdoying it...

>> No.591854

>>591853
I guess about 2 meters or so, some medium-ish. I don't know, I don't have enough experience with rocks yet. What do you suggest?

>> No.591856
File: 26 KB, 440x657, rock2.jpg [View same] [iqdb] [saucenao] [google]
591856

>>591854
Actually, 2 meters is still too small, I think this would be some large rock as a part of a cliff or something. I'm just going through this video to see his techniques and I started sculpting something similar. He adds a lot of details to it, so ofc, as you said, it would be an overkill for something small.

https://www.cgcookie.com/exercise/exercise-rocky-formations

>> No.591862

>>591856
>https://www.cgcookie.com/exercise/exercise-rocky-formations

use alphas. IDK how it is in blender tho.

>> No.591866
File: 1.97 MB, 627x550, headpats.gif [View same] [iqdb] [saucenao] [google]
591866

i want to move on to the next thing already

>> No.591867

>>590941
>>591840

i hope you modeled her uterus too cuz i want to impregnate that.

>> No.591868

>>591862
There's a place for alphas but you don't have nearly as much control. It can't replace sculpting by hand. It works the same as in zbrush.

>> No.591872
File: 30 KB, 354x585, rock_texture.jpg [View same] [iqdb] [saucenao] [google]
591872

Tried out some texturing with SP since I can't find a decent rock texture that doesn't add tons of additional detail.

>> No.591873

>>591872
But I want to get on this level. I don't see this level of quality even in the best video games. Guess people don't give many fucks about a good rock sculpt.

https://www.artstation.com/artwork/8bq0G

>> No.591874

>>588889
ye

>> No.591877

>>591873
The problem isn't really the sculpting/texturing, it's the dogshit lighting most engines have to use due to realtime limitations. Literally all of the models on Artstation would look half as impressive if anyone used them in a game scenario.

>> No.591878
File: 586 KB, 710x710, 0001-0200.webm [View same] [iqdb] [saucenao] [google]
591878

>>591745
all done. I really hate instagram's video compression but here it is with sound if you want

https://www.instagram.com/p/Bbk4S4JjVmu/?taken-by=ribberart

>> No.591881

>>591877
I actually just experienced it personally. I guess that rock looks okay for the first real try, but I imported it in UE4 and it looks like trash. Unusable. I can get good general lighting, but rocks seem to not respond well to it. Now I have to figure out what to do to get something decent looking in my scene, without having to use some free photogrammetry assets I found on the internet.

>> No.591889

>>591866
You shouldn't make waifushit if your heart's not into it, anon. Take a break and do something else, don't try to push it further for now.

>>591848
Just out of curiosity... are you the guy who did a really sick untextured Boa Hancock sculpt on Sketchfab?

I swear that thing exists, but whenever I search for it now, it doesn't show up. It's as if it mysteriously vanished. It looked like something you'd made previously though.

>> No.591906

>at least half of the thread are anime girls
Just why?

>> No.591909

>>591906
because some people still think anime is ok.
i don't know why they think this, but they do.

>> No.591911

>>591909
Where do you think we are?

>> No.591915

>>590653
Is it Bombernauts? Because it looks very much like Bombernauts.

>> No.591918

>>591889
> are you the guy who did a really sick untextured Boa Hancock sculpt on Sketchfab?

It's not me, no.

>> No.591920

>can't even make good looking trees with SpeedTree
3DCG render for this feel?

>> No.591927
File: 866 KB, 2024x1102, Cube_closeupA3-1.jpg [View same] [iqdb] [saucenao] [google]
591927

Here is another Update. I adjusted the way the shapes tile. I also changes the surface roughness as well as the micro noise.

>> No.591929

>>591911
in 2017.

>> No.591930

>>591929
i want to go back

>> No.591931

>>591929
Wait...this isn't 1994?!

>> No.591932

>>591929
...Nobody tell him.

>> No.591934

>>591930
>>591931
>>591932
Degenerates.

>> No.591935

>>591934
Where do you think we are?

>> No.591936

>>591935
In Hell

>> No.591943

>>591936
i want to go back

>> No.591947

>>591943
You're here forever

>> No.591949
File: 20 KB, 439x420, 59d2e.jpg [View same] [iqdb] [saucenao] [google]
591949

>>591934

>> No.591954
File: 107 KB, 1432x698, 39c9be23a9[1].jpg [View same] [iqdb] [saucenao] [google]
591954

Its Marina from splat2

>> No.591955
File: 320 KB, 960x540, sorrow.png [View same] [iqdb] [saucenao] [google]
591955

Noice

>> No.591957

>>591954
And here we go again.
>>591955
Damn, you're doing this for quite a while. If I remember correctly, you're recreating some old game or animated movie, right? If the shitty lighting is on purpose, than that's ok, but to be honest, you could improve it a lot otherwise. No GI, crushed blacks, low-res textures with unrealistic shading, completely flat white windows. The character looks pretty nice, though. But as I said, if you're doing it on purpose, then great, but I still feel that character looks a bit "too good" compared to his environment. Idk, maybe I'm shitposting, I'd need to see the reference to tell how well done this is. I just know that as a person unfamiliar with the context, I don't like the texturing and the lighting is too weak to see things clearly and it lacks focus.

>> No.591958

>>591848
Fuark bro, I need more. In fact, I think you might be able to make some moneys out of patreon or something

>> No.591959
File: 76 KB, 795x522, som.jpg [View same] [iqdb] [saucenao] [google]
591959

>>591957
Well, 3D is just a hobby for me so I have only spent around a week on this project, the last three days, and four days from around 2 months ago. I'm finishing college very soon, and I'll focus on this project.

Well, I agree the lightning IS shitty. But it's easy to fix, I just have to increase the lights and keep the shadows. The blacks ARE supposed to be black, the character is in some sort of limbo, so there are no walls in the room but there's nothing beyond it. In the original game, the windows are perfectly white, but now that you said it I should add a glass texture.

Here's the pic from the original game. I'm not making the greek bust nor the desk

>> No.591962

>>591959
Ah, see, in this case you did a good job recreating it. But yeah, your scene is definitely a lot darker and you need to boost those lights. Illuminate more of that clock so it gets some depth and you can actually see the inside. Shadows are also not that sharp, so increase the size of your lighting from the outside and add a bit warmth in it. The surrounding can be pitch black, but the environment still lacks some bounce lighting, look at that shadow on the left of the clock (and under the column), it's completely black as well and it fuses with the background, so you should do something about it. And add that fog if you can. Oh, and that floor could use some stronger normal maps and possibly displacement, the tiles are more defined in the original.

>> No.591964

>>589269
Architect/Mechanical Major here.
Do you have anymore views of the building so far? I'd like to see the interior.

>> No.591967

>>591955
>>591959
i remember this from awhile back.
this amount of dedication is awesome.

>> No.591969

>>591967
To be honest, it's okay, but there is not that much going on in the scene, it's still pretty basic. Apart from a character, easily doable in less than a day.

Not trying to criticize the OP, but man, we have to have some standards on this board, and judging by the replies, a lot of people here are very easily satisfied.

>> No.591971
File: 318 KB, 1920x1033, grass_newest.jpg [View same] [iqdb] [saucenao] [google]
591971

Hmm, what do you think about the new version of this grass? I think I finally got something good going on, still needs to be refined, though. I also added some more small dry clumps around.

This time I used fibermesh for generation and then I rendered it to a texture with xNormal and put it on panels.

I could have used numerous simple solutions for it, but I treat this as a practice. I'm somewhat worried if this is still a hack way of doing it, i.e. if I showed it as a portfolio piece, would it be a joke since I didn't sculpt or model it by hand? I mean, of course I can do it, I did it for these small clumps, but the problem is that I can spend hours trying out different things and going back and forth, and it would take forever if I had to model and adjust every single blade by hand every time. After all, fibermesh provides much more organic solution, and now I can sense a bit of similarity with some grass in Uncharted 4, which I'd like to recreate, or at least get close to that level of quality. Still not there, though, I think my grass looks a bit busy, not readable enough.

I honestly don't know what is the most common method in the industry for stuff like this, so I'm trying to do it in any way I can, be it modeling, using photo textures, or generators like this time.

>> No.591975
File: 37 KB, 806x582, 34f42fbdda[1].jpg [View same] [iqdb] [saucenao] [google]
591975

>>591957
Where are we going? Forward!

Progress!

>> No.591976

>>591969
Theres nothing inherently wrong with being satisfied, conisder yourself an art critique but also consider the abilities of the person you speak with, certainly wouldnt be too harsh with a child starting out on their art career, no different from someone at 25. Your keen eye can help take someone a long way, but its a sharp eye and can also cut deeply as well. Its a tool we can use but with good judgement.

>> No.591978

>>591976
I agree. I don't know, I personally am pissed when I don't get a reply harsh enough. Of course it's not easy to hear it, but I got used to it, and I know it's the only way to improvement. My greatest fear is being delusional about my skills and talents, and compared to that, some negative critique or two are nothing.

>> No.591984

>>591978
>Of course it's not easy to hear it, but I got used to it, and I know it's the only way to improvement.

Seems a little dodgy to me. You cannot improve without being told very sternly that your work is garbage? Being praised will always lead to delusion without fail?

Maybe there's a middle ground where a person can think, "Well, I got some feedback for my work, it was generally positive given the context, however it looks like X and X areas could improve.".

>> No.592029

>>591984
Ah I meant criticism in general. Negative feedback is just more useful.

>> No.592043

>>591872
>>591873
Look at your rock. Are you using any references? It looks like random lumpy shapes that have been put together without any rhyme or reason.

Different kinds of rocks follow distinctive shapes and patterns. The rocks of the guy you linked follow a clear direction and have much more cohesive shapes.

Pick a reference and stick to it. Ideally just one, so you don't end up mixing different shapes, scales, or do something that's impossible in nature.

>> No.592048
File: 77 KB, 944x804, TurtleTurrets.jpg [View same] [iqdb] [saucenao] [google]
592048

Started a new game project today, feels good to model a thing.

Still need to bake all this to some reasonable poly model still and will likely tweak the sizes of some of these elements so the read better at the distance the player will see them, but for now I like where it's at.

>> No.592049
File: 77 KB, 805x870, TopdownTurtle.jpg [View same] [iqdb] [saucenao] [google]
592049

>>592048
The game will be played with a top down perspective. This is more or less what they will see.

>> No.592050

>>592048
>>592049
>a weapon to surpass metal gear

>> No.592052

>>591971
It looks really really good now, better than grass in some actual games. Not sure what it’s like in motion though, but looks good from this angle.
Fog still needs work though, did you see my earlier comments? >>591797
Also the snow now looks like a dirt smear over a pane of glass, like you’re viewing things from the inside of an old unwashed car. It needs an overhaul asap.

>> No.592053

>>592049
Nice

>> No.592056

>>592043
Yeah, you're right. I just followed what he did at first and then went on my own. He basically never explained how to make it cohesive, it was just like "jus' carve shit here and there and eventually you'll get something that resembles a rock." I usually use references, but I didn't have a technique and knowledge to recreate those shapes, but I think I'll be able to do it now. It sucks that I can't apparently do it without a reference atm, though. I'm pathetic.

>> No.592058
File: 358 KB, 1920x1036, fog.jpg [View same] [iqdb] [saucenao] [google]
592058

>>592052
Yeah, I saw your replies, here's one more screenshot, I boosted the effect some more. I did reduce a start distance of the exponential fog and added some bloom (but I don't want to overdo it, UE4 is known for that so I'd rather turn it off, but it does complement the scene here). As you said in the last post, yeah I actually meant just a "thick" fog, not a brick wall. Pic related has some more depth so you could see that the falloff isn't drastic, it's just that I don't have much going on far away so it could look more like a wall. If I'm still not getting it right, maybe you could tell me specifically what additional settings to tweak if you're familiar with UE4.

Just to be clear with you and others who might give me some feedback. This is more like a tech showcase than a full game. Almost all the textures are at 4k, and I'm using expensive dynamic lighting with GI turned on, which immediately cuts my fps, like, in half. I'm going for the best visuals possible for me.

If it's not better than Uncharted 4 - it sucks. If it doesn't come close to the new God of War - to trash it goes. I know that my standards may be too high at this point, and I don't expect to actually achieve it, but I want to get as close as possible, so keep that in mind when you give your critique. I don't care nor am I going for the "cheap indie" look. If it sucks, it sucks, just be honest as I would be to myself, it's just that when I look at it for a long time, my view gets somewhat skewed, and I can't see it clearly anymore without some help.

>> No.592059

>>592058
Oh and forgot to mention the ground - I'm not sure why you see it like this, I didn't change the textures and I think it looked pretty good before. It could be the lighting or the angle, since it is not tessellated a lot, so it still looks a bit flat.

>> No.592062

>>592058
Looking better with every shot fampai, now the fog is perfect. Ground is of course lumpy but as you said tessellation is off so can’t judge yet.
You may want to tinker with the blending of the grass a bit so it doesn’t hard clip into the ground like that. Can you try adding a really short fade to the bottom of the sprite? Or maybe just don’t make individual grass clumps so wide, as to make it obvious that they’re projections.

>> No.592071
File: 1.97 MB, 1600x2601, Lighting_test_dfao_lpv.jpg [View same] [iqdb] [saucenao] [google]
592071

>>592058
If you going to best visuals possible, it might be a good idea to setup scene with lightmass. If you not using vxgi, it is unlikely to get a realistic picture. I tryed to use lpv+dfao, and it gave me a decent results, but honestly not what I really wanted to see. Also, if I am not asking for too much, but can you please show your grass material? This is like the best grass I've seen so far.
One more advice if you don't mind. If you going to use characters in your scene, it might be a good idea to turn on capsule shadows, it will help to ground them up.
Here is the picture of my very recent tests.
First, baked static lighting, next one - just dynamic movable skylight/directional light, third - light propagation volume + Distance field AO.

>> No.592075

>>592071
Yeah, dynamic GI isn’t best quality, what it does allow are broad timelapse-style lighting changes and not having to spend time baking, which could be useful on very large scenes. Otherwise baked lighting with like a hundred bounces is always going to look better.

I wonder, does UE4 have an unsharp mask effect built in? I’m just getting into it myself and I see all these UE4 images that do indeed look rather soft in terms of texture, but I think all it would really take is a very fine post sharpen effect to get things looking nicer.

>> No.592076

>>592075
Ue4 does not have built-in solution for image sharpening, but you can easily create your own post-process material and inject in into post process volume array. Just google ue4 "sharpen post process". Or you can go batshit insane and just sharpen your textures, you can also pixelfuck mipmaps inside the engine to make them look sharp.

>> No.592077

>>592075
https://docs.unrealengine.com/latest/INT/Engine/Rendering/PostProcessEffects/PostProcessMaterials/index.html
https://cdn.80.lv/80.lv/uploads/2017/11/Screen-Shot-2017-11-15-at-16.50.14.png - this one should work

>> No.592082

>>592058
warframe/10

>> No.592104

>>592056
Lol what good art is always made by reference you take what you like from many things and create something new. Relax learning is a process it’s silly to be hard on yourself so early on.

>> No.592125

>>592104
It's not like references are bad (especially not in 3D), but I think that if you're a great artist, you should be able to envision things in your mind. It becomes a lot easier when you use references. And what you said about remixing is a completely different thing, and it is true, but here we're talking about literally having a picture (or multiple pictures) in front of you, and copying what you see.

>> No.592166
File: 473 KB, 1114x981, Capture.png [View same] [iqdb] [saucenao] [google]
592166

>> No.592247 [DELETED] 
File: 1.34 MB, 1000x1000, YetiMic test render1.png [View same] [iqdb] [saucenao] [google]
592247

The mic on my desk has been tempting me since I got back into 3D and I figured I'd finally give it a go. Had to follow a tutorial for the wires.

>> No.592287

>>592125
Even if you envision in your mind its being drawn from sources youve seen in real life, we cannot muster up new things out of thin air.

>> No.592355

>>591625
post wireframe I wish to learn

>> No.592541

>>591927
lookin good.

>> No.592623

>>592058
Are the characters supposed to blend into the environment? Otherwise you could just give them some color.