angle dependent reflection

angle dependent reflection

cikho
Collaborator Collaborator
984 Views
4 Replies
Message 1 of 5

angle dependent reflection

cikho
Collaborator
Collaborator

attached two videos from showcase.

in ref_1 the material is set to basic reflection which works more or less like chrome: the object keeps reflecting no matter the angle at which you're looking at it.

in refl_2 the material is set to angle dependent reflections: the object reflects more when the view angle is very low, and it reflects less when the view point is perpendicular to the surface.

 

this is very useful, for example, with reflective materials with texture where you have to see the texture when looking in front of a surface, and see very clear reflection with skewed angles (reflective wood).

 

In Vred (2018.5) i see no option to make reflections angle dependent.

what am i missing

_________________________________________________________

almost a showcase PRO user 😛

if you find my post useful, please give kudos or mark as solution 😄
0 Likes
Accepted solutions (1)
985 Views
4 Replies
Replies (4)
Message 2 of 5

richardlevene
Collaborator
Collaborator
Accepted solution

By default in all the materials (bar Chromethe reflections are view dependent.

 

You will see there is a "Reflectivity" value or "Use Reflectivity" (in Glass material). This value is the amount it reflects at the facing angle (0 degrees to camera) The reflecting value on the edge (90 degrees to camera) will always be 1.

 

In the Chrome material there is no Reflectivity value as that just determined by Type and with a Chrome is it likely all close to 1 at the facing angle.

 

Best,

 

Richard

 

 

0 Likes
Message 3 of 5

cikho
Collaborator
Collaborator

got it, it works with reflective plastic actually.

my issue was that giving a high reflectivity value (eg .5) to a reflective plastic material it made it look like no dependency was in place.

 

and here comes another counter-intuitive thing in VRED 😐

_________________________________________________________

almost a showcase PRO user 😛

if you find my post useful, please give kudos or mark as solution 😄
0 Likes
Message 4 of 5

richardlevene
Collaborator
Collaborator

Ah ok. Yeah I mean I guess it might not make much sense if you do not know what it means.

 

I guess they do not do the best to explain it within the app or in the help docs.

 

It based on schlicks approximation and for a dielectric material, generally it is 0.04 reflectivity on the facing angle is the right value.

 

The material itself is already setup to give you the right result for that type of material. ie reflective plastic already behaves like a reflective plastic when you apply it. You just change the colour. Chrome already behaves like chrome. So they are meant to be plug and play to an extent. Unlike in other renderers where there generally just one monolithic shader where you have to change all the values to get the result you need. Maybe with just some preset options to get you started.

 

Richard

 

 

0 Likes
Message 5 of 5

michael_nikelsky
Autodesk
Autodesk

As Richard described already all our materials have a fresnel term that causes the angle dependency. The default of 0.04 is representative of a dielectric with an index of refraction of 1.5 which is in the area most plastics or glass has. If you know the index of refraction of a dielectric material you can calculate the proper reflectivity value quite easily:

 

reflectivity = ((1 - ior) / ( 1 + ior))^2

For metals things are a bit more complex since you have a complex index of refraction there with the real part being the index of refraction and the imaginary part being an extinction coefficient. In the chrome and brushed metal materials we provide a couple of presets that encode these value pairs for some materials (they are actually wavelength dependent, so these preset have two rgb values) and use the full fresnel equations for calculation. If you don´t want to use those presets using a reflectivity between 0.8 and 0.9 and a black diffuse color works quite well for an approximation to metals.

 

Kind regards

Michael 

 



Michael Nikelsky
Sr. Principal Engineer