[REQ] Anisotropy

[tungee]

I must say that indigo has the most powerful and physical coorect material system in the Cg world.
The one and only thing i miss is horizontal and vertical anisotropy for Phong material.

Thank you for thids great Renderer!

PS: Im trying to model an vinylplate, but idont looks so good with bumpmaps.

[CTZn]

well I think basically that an anisotropic implementation would be everything but physically correct. Would be a fake. Let's see how much one can push mpd !

[PureSpider]

Yes anisotropic would be a fake... cause i think anisotropic depends on the structure of the object and as CTZn said, let's see where MPD can get us

[StompinTom]

well then the exponent in the phong model is just a fake too. isnt that 'faking' the microscopic roughness of the material?

you have to remember that 3d modeling/rendering is all about models, not the real thing, and that at some point you have to cut corners or else build the real thing.

'physically correct' can only be considered as 'abiding by current scientific models and formulas for physical phenomena'.

the current phong model simulates an even roughness over the material, as far as i know. anisotropy would just be giving this roughness a dominant direction.

[CTZn]

I understand your point, StompinTom. Well I gave my opinion, does not mean I would never use anisotropy if implemented...

[OnoSendai]

Yes, StompinTom is correct.

[Kram1032]

or bump maps actually also are fakes...
They fake shadows by changing normals.... (right?)


That answer of you, Ono... Does that mean, you'll look into anisotropy soon? (I think, for THIS release, it's enough amazing features but anisotropy certainly would also be nice )

I remember that you once already spoke about using the bumpmap gradients to calculate anisotropy...

[StompinTom]

CTZn: nothin wrong with your opinion! just raising my own points

[CTZn]

StompinTom: That's basically what I expect people to do with me

Yep. How further to perception of reality belongs to Ono, understand through implementation.

[Eneen]

Any chances for anisotropy on v4?

[pixie]

Any chances for anisotropy on v4?

I think we already have it since long
https://blenderartists.org/forum/showth ... digo-1-1-9

[Oscar J]

Anisotropy would be great. Bump maps are a little inconvenient. Anisotropy is IMo not really any different from roughness, just directional.

[thesquirell]

I actually really like it the way it is, makes me think about material structure, optical effects, the logic behind it. It is an immersive experience. If it had a slider, or a value to fiddle around with, even though I understand it's pluses, it would kinda ruin it for me. When I achieve anisotropy through surface structure (modeled/bump/normal mapped), it draws me in, and I feel actuall joy. Same thing goes for medium absorption. Writing custom NK data, understanding the nature behind it all, almost beign in a lab of some sort...you just can't compare that type of experience, with, for example Vray. Working with Vray, or Arnold, after that, feels just empty and sad. But, if they choose to add it, I would understand that. As long as they are adding things, not taking them away.

Don't take our lab away!

[Eneen]

Take a look here:
http://www.indigorenderer.com/forum/vie ... 84#p116384