Are conventional poly modelers becoming obsolete?

In the past I use to think of them as being vital, while regarding other software as being merely complementary. But increasingly I now find myself questioning this. Sculpting software like Zbrush now have decent modeling tools which when combined with the sculpting features makes it much more user/artist friendly to create stuff. Even hard surface stuff is now easy to create in Zbrush. You can also UV map inside of Zbrush. There is also specialized software for retapo and texturing that are better than what you use in the modeler like Maya or Blender, Modo etc.

But up until recently I still believed that modelers like Maya/Blender/Modo/etc were still needed for at least one thing: Rigging and animation.

And then I saw this:

At this point, is there even a reason to spend money on a modeler at all? It seems to me like almost EVERYTHING you can do in a conventional polymodeling program can now be done better in some other software.

 

I would imagine that "modeler" would also include people skilled at using Zbrush and everything else you listed.

 

You can think other way: everything those programs do you can do in conventional poly modeler, but without added cost and dealing with specific bugs and limitations of each software. For example, idea that "you can sculpt everything in Zbrush" is a trap, as you eventially run into lot of issues if you plan to use your sculpts for anything other than static illustrations. Even the decimated/retopologized sculpts will cost more resources to work with than meshes built by poly modeling; so the few hours you "saved" by sculpting can mean loss of hundreds or thousands hours on rendering, or just by slowing down viewport performance in other software. I believe that the pipelines involving dozens of different software items are mainly pushed by companies that sell said software, to perpetuate the stereotype that it is the "professional" workflow. Almost every character course you see on the web that involves Zbrush+3d Coat+3ds Max+Unfold3d+Substance+whatever can be done with just 3ds Max and some 2d editor, or even Blender only.

 

Cascadeur doesnt replace poly modelling, its an animation tool.

And the 'poly modelling tools' you're talking about aren't actually just poly modellers, they're all-purpose tools. Maya does NURBS modelling too, as well as animation, texturing, simulation, lighting, UVs, rendering, etc.

Also, if you think automated animation simulation tools mean that animator-oriented animation tools are obsoleted, consider why that hasnt already happened because of mocap or why rigging tools arent obsolete despite autoriggers.

I mean, feel absolutely free to set up your pipeline however you want. But the world's Maya animators arent going to be made redundant on the basis of this.

 

This really doesnt look that amazing tbh, and it doesnt beat regular animation workflows and tools that have been around for decades. As righly said above, why would maya animators such as those at disney shy away from what they know for 20 or so years, for this? How could it possibly make them redundant either?

Also, why would some animation tech, replace polygon modellers? An animator != modeller.

Thats like saying that a new version of microsoft excel will replace programmers.

 

I was pointed at this originally by my animator colleague and both of us were impressed by it for what it does. And what it does rather well, IMO, is realism for the physics (gravity-wise, collision-wise etc) and anatomic limitations.
But its got no intrinsic 'character' to those animations, its quite literal. At best it'd be a starting point or shortcut for a real animator.
I see it as having the same value as something like Marvelous Designer; basically, its a specialist, niche tool, but it does what it does rather well.

 

Indeed, such automated software usually is quite limited. Physical realism can quickly become detrimental in game animations. Since breaking the anatomic or physical limits is often beneficial to make animations more expressive or informative. In this case, it will be more straightforward to keyframe everything than try to tweak hundreds of unintuitive parameters such as joint rotation limits or mass of individual limbs.

 

No. Not anywhere close. It's so farfetched it's silly to even discuss TBH.

 

Just to be clear, Cascadeur is working off the equivalent of keyframes on the model, and then is simulating the intermediate motion based on physics and anatomic models. So there is an animator involved.
The better that animator is, the better the results, because the keyframe poses will be better.

But the physics model is realistic. Introducing anything else (classic squash, stretch, bounce?) requires the same animation skills as it ever did, and probably the same tools.

(edit for typo in name)

 

Workflows are developed organically by people doing the work. You just use the best tools to get the job done. Believe me, if I could strip out half the programs I use and do it all in just one or a few, that would be awesome. But when you want to get work done efficiently, right now it's usually best to use a variety programs, each better suited for different task.

The awesome thing is that most of the major software are becoming better bridged together, so workflow becomes easier/simpler with every new update.

 

But why animate in your modeler when you can just use a software designed specifically for animating? I feel like the Maya elitists are just narrow minded and now willing to consider alternatives.

The point that I was making originally is that it seems like there is now a specialized software for everything that you can do in modelers, meaning that unless doing everything in one single software is important to you then there is no real compelling reason to use conventional modelers anymore.

 

Which software beats maya for animation?

 

I don't think anyone necessarily wants to do everything with just one software package. It would be nice because then you wouldn't have to worry about import/export problems and things like that, but any glimpse I've ever seen into professional studios indicates that people usually use a pipeline of various software packages. Anything to save time. I imagine a lot of professional studios may adopt Cascadeur if it's as good as it looks.

The point of software like Maya is not that you use it for everything- It's that you can use it for anything. It looks like Cascadeur is easy to use because it is specialized for character animation and does part of the work for you based on various assumptions that are specific to the field of character animation. You still may need Maya though- because you can animate anything at all in Maya (even if it requires a bit more work) and there's always the possibility that you need to animate something weird that Cascadeur doesn't cover.

Do you really think you'll be able to predict every asset that you'll need for a project? Even so, do think you'll always be able to find specialized software that can author every one of these assets?

 

Conversely, why animate in an entirely new tool, instead of your existing one.

BTW Noone animates in a modeler. They animate in general-purpose 3D packages that include animation tools.

Congratulations on undermining your own argument with conspiracy theory reverse-snobbery nutbar-ism.

BTW, Would that be the Maya which isnt just a modeler and is an industry standard known for the strength of its animation tools, would it?

And the counter-point that I was making is that (a) these arent just modelers in the first place and so already include appropriate tools and (b) not wanting to abandoning existing pipelines, knowledge and skills simply because there's a new shiny is a compelling reason to use one's existing general-purpose package.

And this whole thing is a false dichotomy in the first place.

 

In fact, going back to the thread title, it doesnt sound like this was actually about asking a question at all; seems more to the basis for arguing an agenda based on some sort of beef.

 

Agreed with above post.

Either way, the answer to the thread title is:

NO, not for at least 30 years but likely way longer than that. Everyone in the industry knows about these tools already, theres a reason you are not seeing these things used more in massive production houses - the places that make all the high quality stuff with high talent people.