how to interpret Falloff Texture's "falloff index" number?

[itsallgoode9]

How exactly does the Falloff Texture's "falloff index" compare to real world values? The textures defaults at 6 and has a range of 0-15, so the numbers seem pretty arbitrary. Does this setting correspond to real world measurements at all? For example, I want to use this to get an accurate reflection falloff in glass. Can I set this to 1.5 (roughly avg IOR value for glass" and it will work like real world glass reflectivity? If this is not the case, which value in the setting would correspond to a 45 degree angle so I can at least have a point of reference?

[itsallgoode9]

does this node create a linear falloff value or is it more of weighted ramp type of thing? I've just been eyeballing the look when using this node with glass reflection but i'd like to get more accurate with its usage, if these numbers actually correspond with any real world values. I've tried thinking of different ways to test this for myself but haven't really been able to come up with a method.

I checked the manual and it doesn't shed any light on this aspect. a couple tiny thoughts I have though:

1. the numbers in the falloff index seem to work backwards to that of an IOR. even if these numbers don't correspond with real world numbers, it seems inconsistent to have the two values work differently since they are basically doing the same thing. (although thinking about it more, i guess changing it would screw up everybody already using it)

2. The manual says "The fall off map can also be used for other things, like the input to the opacity of glass." Maybe i'm wrong but isn't that incorrect? Isn't the reflectivity of glass the only thing that actually changes based on angle? it may look more opaque because of the higher reflectivity and the angled glass absorbing more light because of it being thicker but that has to do with transmittance and not opacity.

[garytyler]

I'm surprised none of the team has responded. I am also curious about this.

[profbetis]

The numbers are indeed correlated to real life values, although most values are between 1.1 and 1.8

It is the Refractive/Reflective Index of a material as explained here: http://en.wikipedia.org/wiki/Refractive_index

It is defined as n = c/v where c is the speed of light in vacuum and v is the speed of light in the substance. For example, the refractive index of water is 1.33, meaning that light travels 1.33 times faster in a vacuum than it does in water. On the page it states that it is a dimensionless number, which also according to wikipeida "A dimensionless quantity or quantity of dimension one is a quantity without an associated physical dimension" meaning it's a ratio rather than something with units.

This refractive index affects when a photon/wave enters a material (refraction), but also affects the way it bounces off (reflection), and so the same index property changes both.

The wiki page for reflectivity should give you a good understanding of the falloff curve: http://en.wikipedia.org/wiki/Reflectivity
This curve in particular is for water:


Here's a list of various materials and their refractive indices: http://en.wikipedia.org/wiki/List_of_refractive_indices

[itsallgoode9]

are you just talking about index number in general or the settings from the falloff texture node? Seems like you're talking in general about IOR and not the falloff node.

[roeland]

The index value on glossy and specular nodes corresponds to a real world IOR value on dielectric materials, like plastic and glass (metals are not supported yet).

The falloff node works differently, it has a falloff skew factor:
If set to 1 then the value is proportional to the angle between the normal and the camera ray, i.e. if you look from 45° then the value is 0.5.
If larger than 1, then it applies a power curve to the angle.
If smaller than 1, then it inverts the skew factor, and mirrors the power curve.

You can approximate the behaviour of glass with skew factor of 8.0 and a normal value of 0.034.

what the falloff curves look like

--
Roeland

[itsallgoode9]

Thanks roeland, that's exactly what I was looking for! That's not too far off from the numbers I was using when adjusting it visually.

It was said that all the documentation is being rewritten currently; I think it would be useful to put this info in there.

[gordonrobb]

I'm glad that's clear to you - it baffled me totally

I have two questions. Can you use it to set a transition between one thing and another? So if I wanted to sho an irridecent material, where the colour is different when viewed at an angle, I would normally (in LW) have used a gradient, set to angle of incidence, and put in an actual angle. How can I use the falloff number to know what angle '6' is for example.

Also, can I tighten up the terminator between one and the other? (if using it to mix materials)

[itsallgoode9]

I'm glad that's clear to you - it baffled me totally

I have two questions. Can you use it to set a transition between one thing and another? So if I wanted to sho an irridecent material, where the colour is different when viewed at an angle, I would normally (in LW) have used a gradient, set to angle of incidence, and put in an actual angle. How can I use the falloff number to know what angle '6' is for example.

Also, can I tighten up the terminator between one and the other? (if using it to mix materials)

Well, the node in general could be more clear but the reference points at least better let me know how it's acting more control on this definitely would be nice.

To have a little bit more control on this node (although it's not any more precise) I have mapped the output through a color correct node. adjusting the contrast slider can tighten up the transition until the point that it becomes a hard line. So messing with all of those settings cn give you various results.