Complex IOR values
Daz 3D Forums > General > The Commons>Complex IOR values
Complex IOR values
Szark
Posts: 10,634
in The Commons
Has anyone come across a list of Complex IOR vaues for surfaces...I found this https://refractiveindex.info/?shelf=3d&book=liquids&page=water but I have no idea how to translate that in to Iray. I would appreciate if someone could help me out.
Daz 3D is part of 
Connect
DAZ Productions, Inc.
7533 S Center View Ct #4664
West Jordan, UT 84084
Licensing Agreement | Terms of Service | Privacy Policy | EULA
© 2025 Daz Productions Inc. All Rights Reserved.
Comments
I found this one easier to understand.
https://pixelandpoly.com/ior.html
Thanks for that but they aren't Complex IORs but just normal IOR I am looking for Fresnel IOR or Reflectence IOR depending what you read but it seems that most common term is Complex IOR.
It is a good list though. :)
Isn't the Fresnel IOR based on the material in question? The Fresnel IOR should only be different if the Top Coat is a different material, say Skin at 1.47 and a water layer at 1.33.
No take Chrome for example. It isn't a base metal but a metal based coating and has a high Fresnel effect which can't be acheieved with using Metalicity or the IOR and this is where the Top Coat Fresnel IOR values comes in to play. You can see it on Mec4D's metal shaders. After some playing about and getting the MDL Chrome to work in DS I find I can get an exact match to the MDL Chrome Shader with the DS's internal Iray shader using a high Fresenl IOR setting and that I find is the only way to get a realistic result.
However if Iray is like Vray in the sense of using Refraction Weight and IOR then yes I beleive the Fresenl IOR is calculated for those sufraces like Water, Glass etc. I have also read that wood has a higher Fresenl (complex) IOR, around 3-4. Obviously the Gloss (or Top Coat) Roughness and Weights need to be right too.
Chrome is a top coat then. The underlying IOR would be steel and the Fresnel IOR would be for Chrome which as far as I can see varies between 2.3 and 2.9. The thickness of the Chrome layer would also have to be set.
yeah that doesn't work unless Nvidia's MDL Chrome is wrong but it looks right to me and using a vey high Fresenl IOR in the DS shader gives the results like Nvidia's Chrome shader.
I have been trying to find how Iray works in the this regards and how DAZ has inplemented things but man they make it so hard to find information...it really irks the hell out or me. And for Nvidia forums seem to think DAZ3D Iray is a dirty word. :)
I found this passage which explains it better than I can.
"The fresnel effect occurs on the transition between materials with different indices of refraction. Effectively, the larger the IOR difference, the more reflectivity you get. All materials have an index of refraction, but it doesn't work exactly the same way for opaque materials as for transparent ones"
I know rock-all about all this, but surely, opaque materials have by very defintion zero refraction as they do not let the light pass?
It wont because the light penetration of an opaque substance will be different from a transparent one. The underlying surface will only contribute if the thin film matches the penetration distance of the surface material. Ice on water is an example. The thinner the ice layer the more the water layer will affect the Fresnel effect. There are three interfaces that come into play, i.e. Air at 1.0, Ice at 1.309 and the Water at 1.333. The air can be discounted but the thickness of the ice and the depth of water plus the angle of the light and camera will all change the Fresnel effect.
'Refraction Index - This is used anytime the intention is to send light into the surface instead of having it all reflect back. There are several Index of Refraction (IOR) tables online that give the refraction values for different surfaces ranging from air to liquids and to gemstones. This setting can affect either or both Refraction and Transmission.'
The property is also affected by the Thin Walled On/Off setting. (bubble vs marble for example.)
Have you read this?
http://www.nvidia-arc.com/fileadmin/user_upload/iray_for_3dsmax_documentation/pages/pages/dielectric.html
Another thing that came to me after posting was that if there was a layer of air between the ice and the water then that would need to be factored in too. That would mean Air/Ice/Air/Water which again would affect the Fresnel effect as well as Glossy, Backscattering and Diffuse :)
That is the problem with this topic as many think we are talking about the General IOR values of transparent objects but in fact Fresenl IOR is completly different and looks to me to have different values for things like Chrome. I would imagine water fresnel IOR would be the same at 1.33 but since we use Refarction Weight there would be no need to use it. I see what DAZ wrote in the Doc about Fresnel IOR used to Top Coat surfaces and give the example of water but IMO this is needed for things like Chrome to get the right reflections and intensity to be realictic, which for me in my work is important.
I am getting the feeling though it more down to eye balling the effect than having a fixed value like we do for the general IOR values out there.
Light will penetrate even opaque materials depending on their grain size, impurities or, in rock, water quantity.
There has been some scientific studies on this too:)
https://phys.org/news/2016-11-path-opaque-materials.html
Ok so this is all pretty new by the looks hence why I can't find much info on this so it could be possible I am not understanding how Iray does things or how DAZ set up the shader.
But I do find it interesting in rendering Nvidia's MDL Chrome in DS and having to use a Fresnel IOR of 300 to get the same results using the default Iray shader that DAZ3D set up. Plus I also found that 0.010 Roughness in the MDL shader gets a different result in the DS Iray shader. For that to look the same I have to use 0.050.
Seen this? The code means very little to me although I can follow some of it :)
https://docs.chaosgroup.com/display/OSLShaders/Complex+Fresnel+shader
As I understand it, Refraction IOR is _totally unrelated_ to fresnel IOR. Treat them as completely different things.
Fresnel IOR is based on material and there are some big and complicated equations governing them. Also, I'm not sure the common metric of fresnel IOR corresponds directly with the scale Daz uses. IMO, you're better off just... winging it.
Also, as I understand it, metals often have a reflection curve that is unique to the metal and... not in any way managed by Daz' Iray implementation. So it's probably impossible to do it absolutely right anyway.
Yeah I have seen this and sort of get it.
Tim yeah winging sounds to be the option. I just found the need to use Fresnel IOR more in certain curcumstances to get more realistic results and I figured there most some Iray documentation on it that is in English. But alas they don't cater for my low level of math. :)
It's possible, in MDL to use measured values for a lot of things. It's just compilicated to set it up, and requires having enough measured values to properly represent the actual behavior of the material parameter. And it requires actually doing it in MDL, as doing it using the shader mixer would be horribly complicated if not downright impossible.
That is the conclusion I am coming too but I am ok with eye balling it. I have Google images and a Camera. :)
and that I think that is the issue with having enough data which isn't out there yet...well not much
Well, most of the data IS out there....it's finding it, and converting it to be used by Iray. Not much of it is online, and most of the books that have the compiled lists of values are very technical reference works and as such tend to be pricey.
(edit: Note, I'm primarily referring to the "CRC Handbook of Physics and Chemistry", which isn't TOO expensive, but has a LOT of physcial data about a lot of materials. There are more specific tables/books/articles/papers about specific materials with measured values, too.)
good point
Personally, I find that the benefit of realism in a lot of specific details is often lots of work for no noticeable difference.
Not always, and it's worth poking a little to find out. Like, metallicity vs top coat fresnel, the top coat looks reasonably better and it's not much more work.
for the most that is true which is what I normally do but in some case I have found it is a case of doing a lot of referncing and eye balling it.
found this by accident, not sure if its still relevant.
ior for reflection and refraction are absolutely related. its just that for dielectrics/transparent materials k (the complex part) can be ignored.
In MDL up to 1.3 complex ior is not directly supported but a measured curve can be computed based on ior and k
in MDL 1.4 which is what we currently preparing for release, there will be direct support for complex ior.
to use refractiveindex.info, if you use rgb, take ior and k from the 3 reference wavelengths 630 nm, 532 nm and 467 nm
@Szark
In case you missed the above post.
thanks man and thx jjordan.
Have a read at this, or this.
Another resource
Carbon Fibre is just carbon filaments so would have an IOR of 2.4 or so. Carbon fibre made into something would have the carbon IOR mixed with whatever binding agent is used so might have a lower IOR.
@rickwesh
In what way do you think carbon fiber mimics metal?