[numerobis]

+88

[Maximus3D]

Guys, i appreciate that you post "comments" in this thread but please stop with the stupid +X type comments, try to actually type something in using more than just one or two keys on your keyboard. It's not that difficult and it makes threads much more interesting to read for everyone.

/ Magnus

[Bubbaloo]

+1

Couldn't resist.

[Half Life]

I felt that urge too -- but I resisted

Maybe we need a "like" or vote button for these wishes so people who wanted to vote for a wish could do so without needing to create an actual post.

On topic -- I'm using Substance Designer for procedural/vector/raster hybrid material design and it is outstanding (I'm sure I sound like a broken record here) for dirt and aging... there's alot of ground to cover in this arena and it may be best if a company that specializes in that specific toolset developed for Maxwell as a plugin which would allow the Maxwell dev team to focus on what they do best, which is simulate light... it's already somewhat integrated into Max and Maya so it would be a seamless jump for Maxwell users in those softwares.

http://www.allegorithmic.com/products/s ... e/designer

Best,
Jason.

[alexxx_95]

Come on Maximus , relax !!
we can't each time re-build and rephrase the same comment , again and again, so we just up it with a +1000000 because we are just so right with it.

Half Life, could you explain me the difference between Mapzone and Substance designer?

Thanks,

alex

[Half Life]

Well, I didn't use Mapzone for long because I latched onto Substance Designer pretty quickly -- MapZone is a pretty primitive version of Substance Designer and so it is both less powerful and harder to use. Also Substances are unique things which can be used on their own separate from the designer package and this particularly useful in gaming engines via Substance Air.

Here's more information: http://www.turbosquid.com/Substance The player is free and comes with several good sample Substances so you can see how the end user could interact with very few options to achieve huge variations of the same "material".

Best,
Jason.

[Bubbaloo]

Choose the resolution of your substance in real time (from 32x32 to 2048x2048).

Is it limited to 2048x2048?

[Half Life]

When you make them you can specify up to 4096 x 4096 (currently)... they will be limited to whatever upper limit they were created with.

Best,
Jason.

[Polyxo]

I don't think that makes sense at all.
Maybe someone can prove me wrong but I can not imagine any half way decent looking way to Dirt up and Age Models without the
use of Ambient Occlusion/Cavity-Mapping created from the actual Model.

Meaning that Maxwell needed to allow creating such calculations as Pre-Passes. Also Tools to control their intensity and spread
(Brushes/Blendmodes) were required. When content with the effect we finally had to store these maps on disk before applying them to our Mxm's.

While I of course wouldn't mind if someone at NextLimit found a Push-Button solution for that problem I would rather say they shouldn't
waste their time. What's the point in just duplicating functionalities (even to all Plugins if made right) which are available in any High-End
Texture-Creation-Software like Zbrush, Mudbox, Mari, 3DCoat, Bodypaint etc? Even Plugin-Supported 3D-Programs like Modo and others have
that ability natively. I believe that people who feel the need of irregularities, dirt and character giving stuff own such 3D-Painting-Apps anyway
to create their Diffuse Maps, Spec, Subdermal or whatever else is needed.

Why should Maxwell start taking care of just one of these map-types?

[Half Life]

For something like a statue I would agree -- however if you are wanting a brick wall or a concrete sidewalk there is no need to model those details for what is basically a flat surface. A program like Substance Designer is more than capable of making a believable map for such flat(ish) surfaces which can include any amount of distressing you may desire.

The big advantage there is you can texture many buildings at different levels of distressing with only one "Substance" by simply moving a few sliders and outputting different maps... this makes short work of background objects and adds a bit more variation and "imperfectness" to large scenes without the need to hand-tweak lots of textures (either in Photoshop or in a 3D painting app).

Would it be the absolute most realistic thing ever? No, but in a production environment it may very well be the difference between just "OK" versus excellent results in the time given.

Certainly 3D painting apps are deficient as well when it comes to total realism as well -- the best way would be to set up a physical simulation that would take a material(s) + Base Geometry and apply unbiased age though a variety of weather/environmental/usage conditions applied over a user defined period of time... however we would be talking about a huge amount of data that would have to be gathered and processed and the technology just isn't there yet.

The 3D world is still full of compromises that have to be made at the cost of realism and that will likely not change for another decade or so... but clearly simulation is the future of realism, not "hand-painted" maps.

Best,
Jason.

[Polyxo]

Hi Jason,
I wasn't saying anything against standalone Procedural Texture-Creators but wanted to point out that I consider adding means to dirt up Materials to Maxwell itself a bad idea.

but clearly simulation is the future of realism, not "hand-painted" maps

Here - while fascinated of Technology myself - I am not that sure - again talking about Stature type of stuff or CloseUps. I have never ever seen a procedural approach coming
even close to Hand-textured.

[Half Life]

I agree on the Maxwell front -- I think trying to make a one-size-fits-all solution results in a pretty generic result more often than not... instead I prefer software developers to focus on their strengths and let others develop to their own strengths as strategic partners.

Procedurals would not be the same as simulation -- I'm thinking more like RealFlow but for Aging materials/geometries.

Procedurals are simply a useful bridge to that point but I would expect that when "aging simulation" comes into its own procedurals will fall by the wayside. Hand-aging will always be best for aesthetic qualities (much like faux finishes) but they lack the heart of realism because they will not be as heartlessly merciless as an aging algorithm would be

My interest in simulation is very strong -- I chose Maxwell because it is the "Light Simulator"... likewise my vision of materials is to move in a direction where we can "manufacture" materials in exactly the same way they are manufactured in a factory. Ideally I want as much of a 1:1 ratio with reality as I can get from 3D. But at the same time I also accept that we have far to go before that vision is realized and I must work within the current limitations in the meantime.

Best,
Jason.

[numerobis]

Procedurals would not be the same as simulation -- I'm thinking more like RealFlow but for Aging materials/geometries.

yes, sure, this would be nice... but if this means another 4 years without a dirt shader in maxwell i would prefer the ordinary AO/Cavity way first to get something to work with.

[Half Life]

Which is why I posted about Substance Designer/Player -- just supporting the player as a plugin for MXED would open a world of possibilities and would not require anybody who did not want to buy anything to make any purchase(because it is free).

In that scenario you could simply purchase Substances from Turbosquid as needed (or use the free ones I'll be giving away) and if you wanted/needed more authoring ability you would have the option of purchasing Substance Designer as a full-featured texture/normal/displacement/specular/opacity mask generating solution... something Maxwell does not do in any capacity currently, nor (I believe) should it.

The nice thing about that route is it gives a nearly immediate solution with little overhead to bog down the development of much more crucial Maxwell material features (like much faster dielectrics so we can use "real" glass instead of AGS and improved ThinSSS/Coatings and a more robust layer system).

Basically I think it more crucial to put the dev energy into perfecting the core toolset already established than adding totally unrelated new features.

Best,
Jason.

[Polyxo]

The nice thing about that route is it gives a nearly immediate solution with little overhead to bog down the development of much more crucial Maxwell material features (like much faster dielectrics so we can use "real" glass instead of AGS and improved ThinSSS/Coatings and a more robust layer system).
Jason.

Not to forget Ptex - V3 really should support this great Technology natively.