remain focused on animation and rendering. Richard Culver
Blender had improved a lot on rendering features, but saying that is a very good option today on animation is being too optimistic. Character animation is really poor nowadays on blender, maybe a decade updated. That is very weird, and more on a software that sees itself as esentially an animation package.
The way I look at workflows like texturing and sculpting is that these are very specialized areas. I will make a bet that absolutely zero professional users of any software be that, Blender, Maya, 3D Max, Houdini or C4D even consider using thier main 3D app for a texturing workflow. Pretty much hands down. Zero. That is, aside from baking. Now my finger is not on the pulse of Modo, really. But I think they have done a great job with Modoās painting and sculpting. But still lets be honest. I donāt think the majority of Modo users rely on it in a heavy way. But I am sure that Substance integration is a plus.
I think at some point you have to stop piling things on the wish-list and be realistic. Times have changed in the last 10 years. Just in case no one has noticed.
Blender can no longer consider itself an end to end - that also includes painting sculpting and PBR texture workflows - solution. Not today. When I hear that, it sounds a bit out of touch. That was an old idea that was possible with a spec-based workflow.
I wonāt try to make the argument as to what is useful. That is a slippery slope. But I think it is obvious to anyone paying attention that even the bulk of Blender users have other apps for certain things. That is just how it is.
I think it is just a matter of being aware of what is being used and why. And then, take a look at the limited resources of Blender development and set aside certain things that would be nice and useful (whatever your definition of that is) and focus on the main tools that most users rely on. I say most. And even most hobbyists. Not all, but most.
This is my opinion. And it is definitely also heavily influenced by my own needs and priorities. But, that is why it is my opinion. 
Its s tricky oneā¦ I do use zbrush and substance painter and b2mā¦ Iāll also use after effects, premiere and final cut. But quite often I do those same things in blender. Blender is often good enough and itās easier to stay in one place. There are no open source alternative that id consider for these tasks and it is useful that blender can do so much just fine.
Often Iāve freelanced or been staff at companies and they havenāt had a licence in a timely fashion so blenders"Swiss
Army knife" approach had proven invaluable ā¦
@Richard Culver
I find myself in full agreement. Again. Thanks for putting eloquently into words what i couldnāt, but what i was trying to say more or less.
I might add that even for an hobbyist with NO budget, a lot of professional tools are available. That is as long as you donāt have commercial interests in mind.
You could do anything in Blender, or you could use it in combination with Mari, Nuke, Resolve, Renderman, Octane (soon) whatever else there isā¦
Not everybody wants or needs that, but its truly great that these options are there, and they are extremely helpful for not yet professionals who wants to get there.
Not long ago this stuff would have cost you more than 10000 $.
Now we have easy access to these tools and i am pretty sure i am not the only one enthusiastic about it.
Not to mention that prices in general have gone down (not everywhere), so a limited budget used wisely can get you very, very far.
Then entry barrier for 3D, movies, VFX, VR and games has become much much lower and i think that is great for the industry as a whole.
It also puts more pressure on everybody involved in the market, but i think that can be a good thing.
@Mike Williamson
More often than notāgood enoughā is good enough. There is no point debating that.
But sometimes if you cannot compromise on quality or time or the demands are higher, āgood enoughā doesnāt cut it.
A good Businessman needs to have that in mind. A good artist should too.
I used Bodypaint. Full stop. :o
Am too wishing Blender could be more developed in animation area (with bits of Maya, Houdini, Akeytsu, something newā¦). But I assume thatās also one of the next steps. Since first we need to see clearly WTF is going on (viewport & analytics). Then everyone will have great foundation to start/continue improving on their favorite area. But not before the foundations are laid & some ballast released. Modernizing to keep on the level for the next decade.
Get prepared. Time is running fastā¦
Future looks bright & August is near.
I do think something like smart materials is inevitable. And even a very basic implementation would be useful at some point. I think you could also create some with Eeevee and the principled shader with nodes.
What I was initially reacting to was a smart material painting workflow. And that opens up a can of worms. But that said, with an out-of-the box node set up, using a principled shader and Eevee you could use normal, AO and curve maps to drive various scratch and dirt maps.
Not as intuitive or fast as Substance Painter. But useful.
Yes Richard. Smart materials are really about masking. It is already easy to set up multiple materials and combine them with masks. But maybe not as easy to manage.
You could set up each material definition as a node group and use mix shaders with masksā¦ but this feels clunky compared to substance painter for two reasons:
Node groups need to be appended to new scenes. This could be solved with a better library system (global library available for all blender sessions. Project library saved in blend as now and finally none node group setups in a single material)
Masks in cycles/eevee are not as automatic or flexible as in substance. (Which you mention) it would be great to use so, cavity, curvature, object normal , height, etc. Again some of these are possible now with node groups, some require a bake to access. (Which is what substance painter solves)
So I guess it all comes back to Baking , library management and of course ui!
Library management in blender seems like the biggest problem to solve hereā¦ itās annoying and clunky to set up your own custom brushes as is and then open old scenes and have to append them every time.
I guess nodes will always be slightly slower to setup than layers though.
Interesting that many games now use substance for baking out masks and create their own smart materials in engine. More optimal for texture space that way.
Yeah and I think that was my kind of initial reaction was the can-of-worms to build a painting workflow around it. But maybe with some improvements to Blender as you mentioned would make a sort of hack more feasible, and useful.
I also wonder if anyone has the inclination to build a plugin, post 2.8.
I wrote the old texture paint add-on back in the day. If libraries get global it wouldnāt take muchā¦ and some add-ons do have library like functions but it just feels too kludgey and should really be a native feature.
Global sculpt, paint, materials and node groups would really open up blenders capabilitiesā¦
And workspaces could be interesting too to make a more streamlined ux for sculpt and paint. Too many unknowns in that area for now though.
http://armorpaint.org/
(Armory Painter: PBR texture painting, Node based, GPU acc., Armory Engine)ā¦ anyone with experience to share opinion?
Best examples are Substance Painter and 3D-Coat if Blender makes some sort of material painting feature.
Smart materials is market-speak for what are more or less node groups.
I think what you are asking for is something akin the āedge detection auto-magickeryā stuff that substance,3d coat and all the other stuff comes with, right?
TexTools already has the option to bake all the important maps needed to make the necessary masks for āsmart materialsā. The only thing left to do is scripting an addon to automate the process of dragging and dropping into node editor and assigning them to the different PBR channels.
Well, lets just be careful not to go down the slippery slope of āwell we more or less have that already if you just do XYZ.ā Having a smart material is only a very small component in a much larger and far more complex workflow of painting with smart materials, layering systems, custom masks interactively placed on the fly and so on.
That is different that saying than you could replicate the end result - the final image - using nodes. Sure with maps and node setups and other techniques or addons, you could get .01 of the way there. Or yeah youād have your smart material masking paint over metal with edge wear and so on. Or rock materials. Whatever. But the process of getting there with complex texturing work would be far from intuitive or efficient.
The question that needs to be asked is, how feature rich do you need it to be? And how much effort should go into such a feature that can barely scratch the surface - no pun intended - of the expected workflow for something like Substance Painter or 3D Coat?
I am not sure I have the answer to that. But I think overall, unless someone is going to write a massive plugin, we simply stick to being happy we have a principled shader and Eevee for real time feedback on maps generated externally. And move on to other features we really really need - a lot more!
Or have a proper AO and curvature node like arnold does.
I know thereās the pointiness node but still.
A procedural, endowed with many choices curvature mask in shading nodes system: this is the best solution! And, it is very much needed in Blender, without doubt!
Hello Michael.
I did not know that you had written that addon, obviously you know a lot about this.
Have you discussed this with Blender developers (Campbell Barton, Bastien Montagne, I guess)? That is, to discuss what would be necessary in 2.8 to have a good basement from which in the future to implement all those improvements in Texture Paint or an hypothetical future material PBR painting system.
I think as was said. With Eevee coming in that some form of smart material creation and texture painting seems inevitable.
Also character animation was mentioned here. I personally find Blender superb for character set up, rigging and animation work. And this has already been highlighted as an area for more development. A complete re-write of the rigging system is being planned. But apart from Maya. Which other major 3D app out there is investing as much resources and effort into pushing forward Character animation as Blender is now ? Possibly Houdini ? But Iām currently not as up on that as the others.
I love working with ZBrush, 3D Coat and Substance as much as anyone. They are all fantastic apps. And tightly focused and highly evolved specialist apps like this will always have their place and need. Especially so in these cases you would hope. As all three of these companyās have a very good reputation with their customer base and have always been very open to listening and feedback.
I understand the limited funds and priorities argument obviously. But Blender is not Maya.
I think for Blender to loose itās broader holistic vision would be a mistake. Itās come to be such a fundamental part of what it is. And is a large part of itās appeal and uniqueness. The whole mini digital content production pipeline in one place.
It surely has far more going on right now and also with itās future potential than simply aiming to be some sort of open source alternative version of Maya or Max. And I really feel it should always continue to aim wider than that and continue to follow itās own course and vision. A big part of the power of any tool within Blender is simply having it in one place interacting seamlessly with everything else. Using an already familiar and interlinked interface and methodology.
I think the huge workflow value and advantage there is in having all the tools in one place working together seamlessly is too often overlooked. I think this gets missed a lot in comparing tools and workflows in isolation. Especially against more narrowly focused and highly specialist apps. This can fail to take into account the huge combined value of the whole. And I guess this holistic outlook would go beyond the basic software side too and apply equally to the whole community and user base. The large world wide diversity of needs and uses represented there.
Blender grew up in a different environment to the other big 3D apps. And I make this point from years and years with the big industry apps and also being a long term past industry customer of one of the bigger software companies. I always point out to those Iām introducing it too first time that it shouldnāt be regarded as a typical 3D app. I use the video editor for instance all the time. Why ? Because it does the job. And itās straight forward and stable. And I have a couple of friends now who only use Blender for video editing and grading.
Toka: I too really hope Eevee paves road that will help bring a superior easy to use material painter like 3D-Coat or something similar like Substance Painter as there still have no good opensource material painter.
Someone here tries to make material painter but it has slow developmet and source code filled with Chinese or similar language.
