Topology! I'm learning it for the first time and I just

File: 200_s.gif (31KB, 356x200px) Image search: [Google]

31KB, 356x200px

Experiment. Try anything and everything.

>>535212
I would just like to get things right the first time, I would like to experiment but Im trying to become a one man army and do every step of the production process! So sadly time is crucial to me right now I can't afford too much experimenting.

>>535213
impossible. experiment, figure out what works and what doesn't. then, when you're done breaking your back to get that shit working right after your first attempt and doing it, you're going to sit back in your chair and go "i can't use this"

then you do it over again and it will always be better. you will not ever ever do something "right" the first time

better get started now, this is all time consuming.

>>535229
I would example this as tying your shoe, I could experiment until I figure out how its done or I could have someone show me the first time so I dont waste extra time on nothing.

>>535211

http://jetsetradioremixsega.deviantart.com/art/SSB4U-Bowser-553262770

Just copy this.

>>535213
Dude, fuck you. Seriously. Almost everyone here is a one man army on this shit too but that's an excuse. You need to figure out what works and what doesn't. You don't eat sleep and breath this shit and that shows me you don't care to learn these techniques. You have to fail a lot and just because you learn it correct the first time, doesn't mean it's not going to fail or crash on a different model. It's not wasted time if you learned something in the process you fucking faggot.

Experiment you fucking fag cracker

>>535237
I won't bother arguing because I never stated I wasn't working hard on this.

Im not looking for shortcuts Im looking for experience other people have so I can avoid making their mistakes and learn faster.

Working smarter not harder.

>>535233
That's just it though: There is no "right way" or "correct workflow". Sure there are things that are obviously wrong. But, for your example, the difference between modeling bowser with the shell separate or not or making everything one group or not aren't a wrong/right decision.

Some people might just model the shell as part of the model and not care if it flexes when he moves his arms or bends over. Some people might be going for something more realistic and made the shell static and set limiters on the arms so they can't clip into the shell.

Some people will keep every part of the model separate and just combine them when they need to and others will just lump a whole scene together in a single object and somehow it just works for them.

So try both. Try something else. Try it the way some tutorial shows you then redo it yet another way because you didn't like the way that method came out.

Now I will say one thing: Most of the time people model their characters in a neutral (the standard arms out legs kinda spread stand straight pose) pose then rig it because otherwise you might get stretching you don't want if an arm you modeled bent has to be straightened. So on that I would suggest putting him in a neutral post (the one in the image seems fine), then rigging, then reposing however you want.

>>535238
That's why you just have to keep doing it. You have to keep digging. By trying to get it right the first time, you are creating shortcuts. How do you know your roadblock won't lead you to another project or another idea around it? You asked 3 different questions that all require trial and error. If you put it in the t pose, it might clip, it might fuck up, then what? If you make it all in one group it may error in various ways or not play nice once exported. Then what? Every question you have is a collection of trials and errors. There is no easy, straight forward answer. Even from a pro. they all have shit to do. It's up to you to give a shit about these things. E-X-P-E-R-I-M-E-N-T

>>535241
>There is no easy, straight forward answer. Even from a pro. they all have shit to do.

TL;DR: Welcome to the real world. There is no solutions manual. If you ask 100 different designers how to do something you'll 500+ different answers depending on the project, phase of the moon, and whether or no Ralph fucking Wiggum was in tonight's Simpson's rerun.

This has all been really helpful even if its not the answer i was expecting, thank you for your time you guys.

>>535245

Go watch some lengthy tutorials that have you sculpt something then retopo, texture, and animate it

How something is going to animate dictates what kind of topology will be needed, which cuts both ways: Disney is able to automate a fair bit of the process because all their models conform to certain standards (still need to be hand tweaked though)

Anyway, do tutorials, look at other peoples models, try things, build up your own library of assets

>no one has touched on the topic of topology flow
God damnit, /3/.
That's a really fucking important concept, and is something that can actually be learned through tutorials.
No one ever talks about it here, and I think that's a shame because it's important to understand.

There is usually an objectively correct way to model topology when it comes to characters. At least, when we're talking about correct loop flow and pole placement for animation.
A lot of it has to do with anatomy, muscles, how skin deforms, etc, which depends on the subject of the sculpt/retopo.

Humans, for example, generally follow the same facial loop topology, with very little differentiation on the matter.
It almost always follows topology that mimics standard muscle structure and human anatomy.
You will always have loops around the mouth.
You will always have loops around the eyes.
You will always have loops from the eyes to the mouth, or around the nose.
You will always have poles around the corners of the eyes, and mostly always on or around the cheekbones or side of the head.
Etc.

That's not to mention body anatomy, which has it's own list of bullshit you need to understand.

For things like the OP's pic, you'd really benefit a lot from learning human topology/anatomy, and applying the fundamentals you learn from doing those to anything non-human.
You've got to understand why things flow the way they do, why certain poles are where they are, and why things are segmented the way they are, and THEN try more complex stuff.

Look up tutorials on how people do topology, and put in the work by churning out heads until you get it correct.
There is a fuckload of experimentation that goes into this when it's for something that is not easily researched, such as a specialized model, like others here are suggesting.

That's not to say you couldn't just wing it and get it right somehow the first time, but it's highly unlikely due to the absurd amount of things you would have to have knowledge of.

>>535260
^This guy fucks^

>>535211
Is the Bowser from an official game (where presumably he'd only be seen from the side)?

>>535260
Yes this guy knows. As an animator myself that gets fucked by modeling when they don't know what their doing, I can agree.

Google HippyDrome and learn from him. Good body and facial topo is more important than you think. It's one thing to make cool sculpts in zBrush but only the god tier modelers know how to make that sculpt work for an animation pipeline.

forgot what people said here.
you won't learn anything by reading text.

if you really want to know which is what you look to look at examples from pro's.
the more you observe others working, the better you will get.
you can talk about things in theory all day, but 3D is a practical business. you can't learn 3D by reading a book. only by modeling

>>537205 #
>HippyDrome
Also known as Brian Tindall, who animated for Pixar for 12 years, won several awards, and works(worked?) at Laika.

He definitely knows his shit. Would recommend.