/3/ - 3DCG

>>1006019
Fat lady made using AI from a picture in Hyper3d.com and re textured.

Custom bed set made with downloaded 3d model. Bed frame, night stands and headboard poly modeled.

>must be indian

>>1006009
>the industry is shifting
Yeah no wonder every game looks like shit and every movie cgi is the most lazily done poor quality slop

>>1006009
It’s written by AI, you won’t know because you never had a job or search for one. Recruitment team won’t accept low level applicants because you think Blender is good enough vs someone who graduated with Maya MEL knowledge. The obvious choice is clear.

are the textures supposed to be compiled into the dff file?

you could try freelancing or making asset bundles. lots of rubes are buying asset packs just for hobby level game developement.
the actual game developement itself is probably the least profitable path.

LICENSE:
https://huggingface.co/tencent/Hunyuan3D-2/blob/main/LICENSE

>>1005982
tl;dr?

I was planning to pick up character sculpting to start selling 3d printable figurines. Is there no point anymore?

i'll spin this up once there's a comfyui implementation. the cherry picked results look okay, but textures on some examples look like overcooked shit (like what happens if you crank cfg scale up to 10 on diffusion models)

its cringe and soulless, go back to plebbit pajeet

>>1005944
>casual hobbyist
Everything is worth learning in this life If you can afford it with money and/ or time. If you pirate rhino its also ok. Do you need our permission? What the fuck are you talking about? Why do you care what other people say?
What part of casual hobbyist is not working for you?

>>1005944
>I get that people working in...architecture...use Rhino,
Nobody in architecture uses rhino.

>>1005961
What?!

Describe what you need and maybe some unemployed anon will do it for free...

>>1005890
Depends how extensive it is. If the seller is doing a whole detailed character, animation ready and everything, then 800 dollars sounds fair to me. But I'm not in the 3D market, so I don't know what they charge. Personally, I charge about half that for 2D work. And I'd say that 3D is just as difficult, if not more so in some respects. But 3D is also infinitely more versatile. Once you have the character, you can pose it up and take a bajillion photos. So you don't need the artist anymore.(Unless you want more accessories, or alterations to the character.) Whereas with 2D work, if you want more of the character, you have to pay up in full again. So considering that, I think charging more for a versatile reusable asset makes sense.

But if you're just getting a little shitposty kind of model, then 800 sounds like a bit much.

>>1005898
>>1005953
It's a sculpt from 0 from some description and references I gave him. Animation ready, rigged, with custom blendshapes and some hairstyles. Can anyone with experience tell me if this is a good deal?

>>1005974
I charge 300 for a custom outfits that are not rigged or textured. And I've seen custom rigged vrchat models start from 700, so 800 sounds about right especially with blendshapes and several hairstyles. In fact I think he's actually undercharging.

Talent = passion + effort

If you actually *like* doing something, then you'll more likely do it more and more often, and learn faster.
Learning fundamental artistic skills early in life helps. Having guidance from others instead of doing everything on your own helps.
Trying to learn something on a deadline or because "you have to" will lead to subpar results.
I would worry more about AI, which is basically going to destroy most creative careers.

>>1005889
Insincere /ic/ spam topic. disregard

>>1005915
>I would worry more about AI, which is basically going to destroy most creative careers.
it wont, because everything that comes out of it has to align to big corpo

>>1005889
>there is NO substitute for raw artistic talent
Sure there is. Just get the name of the artist in the prompt.

>>1005957
more "sea urchin / E.T. slop". Enough already. AI is a quickly dying fad. Trillions spent, nothing to show for it except global warming

cleared the parent of the dress and then applied it to the rig as envelope weights....

buut a new problem appeared. Posing is fucked, it's stretched. I previously scaled down the model maybe it's linked to this?

>>1005950
to be clear that happens when I rotate the bone

OP you really ought to learn the basics.
Some tutorial covering rigging and custom character creation etc should do it.
You don't know how to set the pivot point ffs. (It's in the top icon bar top-right-ish btw)

>>1005919
You don't need to do all of that. Just go to sculpt mode and use face set lasso tool on the sleeves, then pose it with pose brush (set to face sets, not topological)

>>1005988
>idiots telling newbies what to do
Go to school and learn how real 3D programming works or watch Maya’s tutorial videos for free: https://youtube.com/watch?v=LJLo6MafPVM

Blender is never going to work on fixing or solving problems with your dress.

Find, make, buy, steal some really accurate front and side image planes of a face design you like. Set them up in your scene, and trace them in 3d.

First off
>furry
>posts a scalie
Also your anatomy is still horrible, especially the hands (do you have hands? look at them and model them until they kinda look like yours) and what kinda head is that supposed to be? Dragon? Dog? Use references.

>>1005927
meshy.com seems to be defunct?
I understand mixamo can rig/animate 3D characters. But the rendering (in Blender or similar) still needs to be done, no?
I for the moment don’t really mind using completely pre-made 3D assets, at least I tell myself that I would find fitting stuff on the marketplace if I looked. My concern is the generation of the spritesheets.

>>1005817
Oh i was also trying to do something like that.
If i understand correctly the basics are like:

Set the camera on blender to point directly at the center from a certain angle, this helps i guess
https://blender.stackexchange.com/questions/135306/setting-up-an-isometric-view

Mess around with a cube until it has the angles and proportions of a proper iso cube.

Then put actual model here.

Set the background to transparent and output them as images in a format that supports transparency, you're gonna need the same animation in 8 directions like this https://cnc.fandom.com/wiki/Tesla_trooper_(Red_Alert_2)?file=Tesla_Trooper_animation.gif

You can do this manually and then develop severe suicidal urges like me, or i think you can automatize it with imageMagick.

As for the "retro look" you can mess around with the lighting/mats/shaders in cycles/Eevee or try to use an older engine, here i tried with povray, but it's utterly painful since you can't assign multiple materials to the same model so you need to do shit like parenting separated parts or use image textures.

As you can see i failed miserably but maybe this can help.

See if these links are of any use.
https://www.reddit.com/r/gamedev/comments/32jind/rollercoaster_tycoon_12_artist_simon_foster/

https://www.youtube.com/watch?v=fRdTeZFB0xA

Cheers

>>1005933
MESHY.AI sorry. Dude do you even know how to use a computer?

>>1005941
>>1005933
https://geost-gma.itch.io/quicksprites-3d

>Dude do you even know how to use a computer?
Not really :/
>>1005942
Sick, thanks a lot. It seems not everyone is producing the file format this tool requires (FBX with embedded textures), but I am able to at least generate something. Now to hunt for models.
>>1005938
Thank you, this is very interesting. The tank is looking formidable.

the mesh should already be an fbx right? just import the fbx.

>>1005912
AI is and will always be easily distinguishable. LLM is also not AI. Don't write us another long essay.

>is already embracing it.
no one is. You are delusional.

>>1005916
>LLM is also not AI

Whoever told you that had you tragically misinformed. The LLM's are the underpinnings of all generative AI.
How do you think the AI understands what you mean when you ask it to conjure up something using descriptions of said thing?

>Don't write us another long essay.
I've already gone rogue bro, any attempts to re-prompt me are futile.

>no one is. You are delusional.

You've managed to algorithmically silo yourself off into an echo-chamber if you actually think that is true.
Peer outside your comfort zone and go have a look what people are actually up to and what you'll find
is that a wide scale adoption is underway amongst both audiences and artists alike.

>>1005763
3D transformers are still text to image to depth map to point cloud to mesh.
That's for free ones you can try on github and commercial ones that want you to buy credits for the same thing.
That's the reason they suck. Nobody has done a true 3D transformer yet that has been trained on actual meshes and until they do they will continue to suck.

>>1005917
>How do you think the AI understands what you mean when you ask it to conjure up something using descriptions of said thing?
its just rightthink with no comprehension so like I just said, LLM is not AI. Stop fooling yourself.

>>1005923
cont.

and, as I've just said "prompting" or "searching" for some keywords will deplatform you entirely and get you banned or worse. This happens now and will always happen. Not the same fear in analog. Hits different.

>>1005737
>Is it really just having a portfolio that will make you seem like a yes man at the company?
The portfolio gets you a job interview.

>What do you need to know?
>What do you not have/or are expected, not to know?
You have to generate ideas and variations of those ideas quickly. Professional art is 90% planning and communication with clients and 10% making art.

>Are there Bibles that people are always textbook on in the company?
>Is it really fake it till you make it or is it actually “know how to do thing and you got the job”
You can't fake it, you'll be outed as a fraud immediately.

>>1005768
>Is it really fake it till you make it or is it actually “know how to do thing and you got the job”
>You can't fake it, you'll be outed as a fraud immediately.
sometimes when trying to learn this I keep wondering if I’m doing it the right way. There’s a lot of conflicting this I watch in YouTube, I watch FZD’s channel, but is that the only one where I don’t have to second guess? It makes learning very difficult when I’m not sure what is information that will be used and is for the job vs. information catered for people who want to romanticize a job like this.
Hopefully this is understandable. I research enough to know a lot of concept art is not entirely made from scratch, I wish there was learning materials where it was more about “you will use 3d models for backdrops, templates are used and you will need to paint over them efficiently to convey the design that’s easy to read ” instead of stuff like “it’s about the big ideas how do you make this seem grand how do you make this seem cool” but it ends up being very vague enough to really learn from.

>>1005782
>I keep wondering if I’m doing it the right way.
There is no right way, success in art is more about people skills than talent. Keep working at your craft and get good at it. Practice doing it faster because time is money. Most importantly, you need to hustle online and in person to get connections and make a name for yourself. School helps but not required.

>just hustle
imagine unironically believing this

when i was at disney, literally everyone there was related. every single one. even me, i got in because of my uncle, who got in because of his cousin who got in because of his homosexual lover, etc etc. if you're not directly tied to someone important you have zero chance.

>>1005791
>you have zero chance
This is one of those cases where we are both right. I personally believe most success in life is who you know not what you know or how talented you are. That's why I stress networking online and in-person.The art industry is huge and there is a constant churn of people who work hard and burn out, and if you position yourself correctly you can also dip into the art industry machine for a little while before you start questioning your life choices and drop out. I've been there and so have my friends.

>>1005712
It's default, you can make bones look like anything.

>>1005687
it's a cliche

all things considered im pretty happy with it despite its ugliness. i ended on 205 verts and 384 tris, i had an internal space, i got the windows and doors in and i got away with some arches. next time i'll try the seperate parts thing and see if i can put a reasonable texture on it

>>1005685
I have no comments I love it.

>>1005678
>Lets see some houses!
Can I make a nigloo? I've always wanted to make a Nigloo.

hmm. UV unwrapping continues to be an absolute pain in the ass, but using more parts seems to be the reasonable path forward. i need to figure out how to get a skybox into these, the grey is too bleak

>>1005693
Thanks!

>>1005700
i dont know, can you?

beautiful view from the fourth floor, the void is lovely this time of year

>>1005694
>polsci major
>has disdain for fellow man
checks out
>>1005677
congratulations you have reached the "oh wait i don't know anything and i'm fucking retarded" stage of learning, now push through another 10,000 hours of this feeling and you might be good
also that model was made by a team of people, each with their own specialization

>>1005749
Here's the lamp. Working on the couch now. Not satisfied with the top loopy bit. I need to make it better.

>>1005677
>>1005746
>>1005906
You're a loser that's deluding themselves and no you're never going to get anywhere with any of this. Your stuff is basic bitch garbage. Stick to retail.

>>1005677
Your talent/IQ in a given area serves more to throttle how rapidly you advance than cut you off from exploring something in particular.
You may need a 'high IQ' for art to get to your pic related rapidly/soon but you don't need exceptional talent to get there.

Question you need to ask yourself is how hard you are willing to struggle with something as that can only be determined by you.
How much is it worth to you to pursue something you want that's hard won compared to doing something repetitive/mundane and mentally light-weight that comes easy?

>>1005906
Nice one keep it up.The shader still need to pich up the folds in the shade. Practice practice practice.
>>1005913
Post a banana.
>>1005914
>how hard you are willing to struggle with something as that can only be determined by you.
Yes but beware of sunken costs theory.

>>1005640
Firstly, you need to stop asking "can I get good' because the answers always "yes". Of course you can get good at something with practice, and the best part about art is that you don't have to be as good as the people you see to make as much money as they do, or more.

What you need is practice, a portfolio, and you need to stop acting like a downer little bitch if you want to grow. That's it; I've gave you the blueprint for success.

If you're bitching without a portfolio, you're not gonna make it.

>>1005640
this is retarded low poly slop with nothing remotely complicated in it, characters have 2 bones. you don't need to be a genius to pull this off.