I am currently doing some design development for a large glazed tower, the architects have a specific glass manufacturer they’d like to use and have selected 3-4 types they are wanting to visually test in different combinations.
I have pretty good data on all types and I’d like to use this data as accurately as possible as I have samples to hand to check the results.
TYPE = Guardian - Sunguard Light Blue 63
Reflectance = 16%
Light Transmittance = 62%
Now, If I use these values in the reflect/refract slots of vray (2.55*percentage=greyscale value) understandably, my glass looks nothing like it is meant to.
Does anyone have any experience with doing projects like this? Like I said, as I have good data I’d like to utilise it. We’re not particularly interested with how the glass refracts, just its reflectance and light transmittance.
I would look at it more with the reflectance value affecting the ior, the higher the reflectance the higher the ior. Not sure if any science or math could be easily derived. Maybe if 0% was 1.5 ior and 100% was 5 ior or something? For light transmittance it might work to directly input the percentage as rgb into the refraction value.
I realise that I can do it by eye… But with the combining different types it would make for many a sleepless night… It would be far better to utilise the information I’ve got especially considering that they will be used for a comparison.
My best idea so far (that I’m now trying) is to create a ‘base’ glass that I then add the necessary values too…
I.E-
Greyscale values = 155 + Reflect/refract percentage (giving values between 155 and 255)
The theory behind this being that for values below 155 for reflectance and refraction it seems to cease looking like glass…
Hi there, cheers for your ideas, I’ll try playing with the IOR for Reflectance see how well that works.
As for the refraction values, that seems to be one of the major issues, taking them below values of 50% (128 greyscale) makes the glass looks muddy and unconvincing.
On a side-note…Is there any plans to create ‘base’ shaders as used in Mental Ray/Brazil? This would aid issues such as this a great deal?
I usually just use the default 1.6 for glass, usually never going above 2.1. 5+ is more of a metal fake me thinks, because that makes it almost like a mirror.
Good points…High rise glazed towers tend to look mirrored though… maybe because the architectural glass used are laminates/coated? The Guardian Sunguard types studied here are anyway…
Having an IOR for reflection is a little confusing…
Remember that (as far as I am aware) vray automatically reduces the transparency of the glass as you increase the reflectivity; ie 100% refletive is 100% opaque.
Id be interested to know why, if correct, Glass does not follow Fresnel too!! Every reflective surface, so every surface, should follow Fresnel?
Kyle, sounds like you might have some long nights in front of you! I agree tho, I have always been confused what IOR is for in reference to Reflection. Maybe the IOR controls both reflections and refractions, and the controls are seperate so you can control each independently. Im sure someone will confirm for you!
About your glass. It is a real shame there is not a method that you can translate your industry figures to our render settings easily. It would make a lot of our lives easier. Keep trying
Cheers for everyone’s advice, it’s an interesting topic.
In the end, I only used my formula (155 + perscentage) as a starting point. And then adjusted by eye from there. Guardian has a good website with reference photography which was utilised. (http://sunguardglass.com). The interesting things is (and as I tried to point out to the architects), Guardian have picked very specific shots to portray their glass in a certain light. For example; for reflective glass images, all reference photography has been shot with an upwards camera angle on a cloudy day etc. For Transparent glass, a close up shot in ambient lighting etc…
As for fresnel reflections… I am not convinced. Most modern architectural glass is some form or composite/laminate, and if you look at it from almost any angle it tends to be very reflective. I think maybe that the type of reflections I’ve dealt with (in this project anyway) are more dependant on light sources/contrast and what is behind the glass, rather than the viewing angle.
Below is an example of one of the glass materials should anyone be interested, this may seem an odd set of setting to most, but they worked a treat. The biggest difference was a Glossy highlight…
