1080x1080

1632600823124.jpg

🧵 Untitled Thread

Anonymous No. 885708

why yes, I enjoy UV unwrapping

Anonymous No. 885709

>>885708
as long as you don't enjoy retopology you're still in the mentally sane camp

Anonymous No. 885729

>>885709
And if you don't enjoy rigging you're still in the human camp

Anonymous No. 885731

>>885708
Try unwrapping Nanite meshes and then come back

Anonymous No. 885738

>>885731
oh, thats simple!
U > Smart UV Unwrap

Anonymous No. 885739

>>885731
Just use the in-house flattener made by the R&D junior

Anonymous No. 885740

UV unwrap deez nutz

Anonymous No. 885742

who shooped the blenderguru face on that sculpt?

Anonymous No. 885752

>>885731
You unwrap them on the lowest subdivision you Muppet

212x237

1631365956794.jpg

Anonymous No. 885753

Yes, I do recreational weight painting DAILY

Anonymous No. 885755

>>885738
Based

Anonymous No. 885763

>>885752
Nope, most of the time they are decimated disasters. I WISH their topo for clean, then obviously it wouldn't matter. Even unwrapping highest subd levels is light years easier than this.

Anonymous No. 885764

Put him in Smash

Anonymous No. 885767

>>885738
The result is usually way too many seams that are also placed in places that are not ideal.

But yeah, these technical things were so much easier when I was just doing this as a somewhat clueless hobbyist beginner. Nobody cared about clean technical results except me - and I didn't care about them at all. Just auto unwrap everything, never sculpt hp details, default bake settings, overuse of smart materials, no worries about texel density... Damn, finishing projects was so fast in comparison.

At this point I feel that Nanite workflow really is a meme. Results are only better in a sense that you can have all the details in geo, but the workflow (after sculpting!) is a disaster IMHO. It's nice to sculpt whatever you want and know you don't have to retopo the mesh and that is how it will look like in engine. But when you start dealing with UVing decimated meshes, shit gets real.

I fail to see where does an artist save time with this workflow.

Anonymous No. 885786

>>885708
You monster.

Anonymous No. 885790

>>885763
Why are you unwrapping a nanite mesh somebody else made again, it will already have UV.

>>885767
I will help you dumbass noobs out.

Blockout sculpt your rock/tree/whatever, zremesh it and projectall, begin working in subdivs for fine detailing, export out subdiv 1 and unwrap, bring it back in, decimate highest subdiv with preserve UV on and export

Anonymous No. 885792

>>885790
>Blockout sculpt your rock/tree/whatever, zremesh it and projectall, begin working in subdivs for fine detailing, export out subdiv 1 and unwrap, bring it back in, decimate highest subdiv with preserve UV on and export
I'll try it, but can it keep all tiny surface details? Everything has to be in the final mesh. I'm not sure I understand how can I work in subd for that kind of fine detailing. If I need to sculpt edge damages of some old hard surface asset, will I zremesh and start with subdiv before or after I'm done with all of that? Because I need all that geo info baked into the final LP/mid poly/however you want to call it, mesh.

Anonymous No. 885832

>>885792
Nvm I think I got it, this is a life saving advice t b h. Now I have no idea why is my studio not using this workflow and everybody is just working with decimated meshes. At first I thought it’s because zremeshed version takes a lot more geo than decimated one, but since I can still decimate it at the end with preserve uvs, I don’t see any negatives with the method you explained.

Anonymous No. 885852

>>885832
>>885792
>>885790
Okay, I've stumbled upon a big problem with this - zremesher refuses to create loops, there are spirals everywhere. I can fix some of them with guide curves but it takes a looong time to do that for complex meshes. At the end it seems like it will still help a bit, but sadly there will probably still remain a ton of manual labor.

Anonymous No. 885867

>>885852
this is normal, when using a decimated version for the low poly in a high->low workflow you needed to just pick shortest path out to make your cuts anyway, it makes it way faster though soon as you have much less to pick from and it will unwrap quick.

Anonymous No. 885961

>>885708
Another tedious task that will be 100% done by machine learning at one point

Anonymous No. 885985

>>885961
Probably going to still be faster to unwrap by hand instead of spending an hour tweaking the settings of some machine learning unwrapper tool that doesn't understand what you actually want to do.

698x658

1564314764202.png

Anonymous No. 885993

>>885985
Lets see how that comment holds up in 5-10 years

Anonymous No. 886085

>>885993
We all just need to collectively lower our standards and auto unwrapping is suddenly already a thing. Every now and then I ponder about the fact we all spend thousands of man hours every day on something so pointless as a perfect stretchless texture map of a random 3d asset. And it doesn't benefit humanity even slightly. Even packing iphones in a chinese factory somehow feels more important in the grand scheme of things.

Anonymous No. 886088

>>885729
I enjoy rigging more than doing UVs

Anonymous No. 886122

Kek
>>>/b/874017411

Anonymous No. 886184

>>885993
how exactly would you communicate intention to the machine learning thing, I can only see it replacing tasks we have already mostly replaced with stuff like wrap for unwrapping humanoids for texturing soon as those are always pretty similar.

If you need to use the UVs to drive anything else, perhaps make one where the UVs are all aligned in order to drive VFX etc. there is no chance.

Anonymous No. 886190

>>885993
That's what they always say. 5 more years, 10 more years, 20 more years. They always keep saying we'll simply think up an idea and the machine will do, but it has yet to happen with anything. 5 years is actually such a long time that I'll be a 30 year old boomer at that point so it's really quite irrelevant if they come out with an intelligent and conscious UV unwrapper that can love you and have sex with you. It's a vague and made up number somewhere far enough in the future that it clearly has no concrete basis in reality. But if it does come out, I'll gladly take it. It'll just make my work easier.

800x679

fq25rzio21021.jpg

Anonymous No. 886194

>>886184
>>886190
Massive cope from these soon-to-be obsolete fleshbags.

Anonymous No. 886197

>>885753
Kek!

346x266

image_2022-03-21_....png

Anonymous No. 887681

>>885709
today is the first time I tried retopology and I can confirm, it made me feel something

Anonymous No. 887730

>>885708
In the order of enjoyability:
rigging>retopo>unwrapping>>>>weightpainting
however, unwrapping is so easy and low effort I don't understand why anyone ever complains about it

Anonymous No. 887744

Anyone who's bitching and moaning about UV mapping or retopo is in the wrong field.
If you view it as a chore fine, but the time you spend on either of those compared to all the other things that goes into making stuff
is just a tiny fraction. If you can't be assed to spend a couple of hours on retopo and UV mapping something
that may take upward to hundreds of hours to build texture and rig you're not cut out for this at all.

While the task can be somewhat menial it is one of those times when you get to turn off most of your brain and just listen to podcasts
or whatever while you go thru the steps of prepping your work for the fun stuff you get to do once it's done.

Anonymous No. 887755

Retopology is satisfying because I see the numbers go down and I have autism.

Anonymous No. 888068

>>887730
I have been doing unwraping on 6 props models , each one have 4 construction models and 4 lods that need little adjust, and was only for one project.
I really like modeling and design, but doing uvmaps just get me tired

Anonymous No. 888078

>>887755
It's opposite for me. I feel like the numbers are never low enough and that activates my impostor syndrome.