VRay IOR and reflection color correlation?

[noneo275]

Hello.

If IOR (in fresnel equation) defines how quickly reflection changes (from 4% -for glass at normal angle to almost 100% (almost means 99,999%) at 90deg or from 40-60 for example for metals an normal angle to near 100% at 90degrees) AND ALL materials have fresnel reflectance: brick, paper, walls have fresnel - with various VISIBLE reflectance intensity (due to MICROFACETS, "little irregularities" in surface we cannot see "mirror" reflections, rather very glossy for bricks, paper etc & very low IOR: 2-4 = low reflectance) MY QUESION: What is dependence between IOR graph and reflection color slot??? (we know already, that for all materials refl at 90deg =100 (white) - fresnel, of course). REFLECTION COLOR SLOT in Vray mat IS FOR.....???? How gray values in refl color slot changes IOR GRAPH??

Piotr

[noneo275]

01 I think the same about what you wrote, but i need confirmation. And what happens to IOR graph when we are lowering refl. color (from white - 100% to gray - 67% for example). Is IOR graph SQUEEZED?

02 Reflective color for metals = fresnel equation for every wavelength in visible spectrum, so it is colored.

And another big question. Need I "degamma" color in reflection color slot?.

[noneo275]

Example:If material (i dont know,painted plaster for ex) has IOR 4 (refl at normal angle = 36%) & very blurry reflections (glossiness) & 100% (white) in reflection slot And we lower this 100% in refl slot to middle gray (50%, rgb =127). Will IOR graph be squeezed?. If Yes, then REFLECTION AT NORMAL ANGLE WILL BE 36%*0,5 = 18% AND IOR (for 18%) WILL BE 2,48 (every IOR has unique reflection value at normal nagle). Then we are loosing Physical (real) properties of painted plaster material (IOR4 gone to 2,48)

[Scott Schroeder]

Does it look good in the render? That's what I worry about. I think you are getting way too technical for what you are looking to produce. This stuff is art, approach it as an artist and not a technician. I highly doubt the best artists out there in architecture have 100% mathematically and physically correct material settings.

 

Half the beauty of working in 3d is being able to break the bounds of reality and yet produce a convincing realistic image.

[noneo275]

Of course First i looked at A&D graph, long time ago. My question was very specific and i dont think trial and error method is best way to create materials. Sorry but ive seen renderings with extremely good (real) looking metal elements (i couldnt believe, that was rendering) and values of these materials (several kinds of stainless steel), IOR , refl color were very unique AND werent even close to these values, you could see on the internet (Alu or INOX = IOR20-60

[noneo275]

and Fresnel and Reflection arent diff things. They are strictly connected. Every IOR has specific Refl at 0 (normal angle). So if you know IOR value, you know reflection of material. By "reflection" i think you mean Glossiness value, which simulates "microfacets" (with BRDF model - ward for ex). My question was: for what is reflection slot? only to speed up calculations (avoiding extremely low values in glossiness)?

[noneo275]

Well, I would like to create materials like in this rendering I mentioned earlier. Look at http://www.forum.vrayforc4d.com/showthread.php?t=7571&page=2 (LAST POST's attachment). Its stunning (except this bar chair on the left - it looks like "typicaL Vray". And Vray is MATH. and id like to know correlations between things. Is it a shame? Im sorry if my question was too difficult. i just dont know the answer so im asking. And i appreciate "art" things but id like to know "physical" properties to help to "break the bounds" of reality.

[Ismael]

Robert,

 

Maybe you find this useful: http://en.9jcg.com/comm_pages/blog_content-art-143.htm

 

It seems along your thinking lines but any issues with it, try contacting the author.

[noneo275]

Ive seen this tutorial and it inspired me to make my own glass. By the way there are mistakes: no need to place gradients ramp in these slots - simple IOR 1,52 for modern IGU will do. And of course modern glass transmittance is 88 to 91% for 6mm glass (91 for low iron glass). and intrenal reflections should be ON to match values on catalogs (refl 8% for single sheet of glass).

[Scott Schroeder]

Set your phasers to phun!

 

http://bertrand-benoit.com/blog/2013/05/26/materialism-1-5-rough-metal/

 

Mathematically rip him apart in 5,4,3,1, off blast!

[noneo275]

Ismael Orozco - thank you, of course iv seen it and inspired me to make my own glass. Errors in Jan Rybar's glass: no need to use gradient maps , IOR 1,52 will do, second, modern IGU has transmittance 88-91%, last thing: turn ON internal reflections to match values from glass producers catalogs (8% for normal glass)

[Matt McDonald]

V-Ray only uses simple IOR which is why so many people use crazy high values for metals, it's a hack but it's the best you can do with V-Ray or Mental Ray. As I recall, Maxwell uses the Complex IOR.

[John Dollus]

Exactly what Matt posted.

Vray is only fast due to the shortcuts taken with the calculations. It is often referred to and marketed as physically accurate but it is insufficient for scientific analysis. It makes pretty pictures pretty quickly and it does that only because of the sophisticated weighting and optimization of the calculations at render time.

If it were that accurate, then photoshop would not be as prevalent in just about every workflow and there would not be 50 different materials out there for stainless steel.

[noneo275]

Well, there is something like COMPLEX IOR You can calculate it from reflectance. For example : for aluminium (n=1,15, k=7,15) complex ior = 46,45