(FINAL)
14 Likes
bella?
Hi Kyle, Yes, it’s Bella.
Though, this type of thing can be done with Maxwell too, or Vray/Corona.
1 Like
non biased renders are sooo goood.
They are a must if you are into Jewelry and need the extra pop.
1 Like
@dsw - what do you think?
Well, nice to tag me on this topic @dale 
Please don’t get me wrong, i know how much work such renderings are, as i do it all day long. So it’s only my opinion. 
I think the technique of bella is outstanding, especially when it comes to reflection, refraction, caustics…
Even though when i look at the image, the refraction seems to be a little bit too much. It would be very interesting which physical parameters for the glass is used.
Also the rendering time would be very interesting. @ThomasAn could you share a few details about the rendering?
Feels like I am in the middle of an ongoing conversation between you two ? 
Swarovski crystals have a lead content that brings their IOR to about 1.7, they also have AB (Aurora Borealis) coatings to emulate Diamond dispersion. Diamonds have a dispersive power of about 0.044, which is the equivalent of an Abbe 22. This is a 60min render.
No, @dale only knows i work for swarovski. 
So i do know a little bit too much about this materials and coatings, especially AB 
Anyhow i don’t think you wanted it to be a realistic rendering for this bull. 
1 Like
Is that either a pragmatic or a practical expectation ?
Wanting a realistic rendering would imply me having the actual object of this shape in hand, to take a photo and try to match the lighting and glass characteristics, which is neither practical nor desirable (I am not running a lab). It would also require me to run a test on the engine itself to determine it’s actual calibration of Abbe against real life objects (not practical either).
Instead, I made some initial assumptions of IOR and dispersive power, dialed them in and run with it to see what happens. The third part of the equation is the cut of the facets which is very important in real life, but didn’t matter to me. I just grabbed a low poly model and run with it.
I take it the actual Swarovski coatings are not as colorful as I remember them, given that you sound jealous of this model’s colorfulness 
(I could have chosen an even more aggressive Abbe such cubic Zirconia or Moissanite).
In my case it’s what it’s required. 
It’s not about the colorfulness, instead it’s again about an realistic AB coating.
Your approach is totally understandable and i appreciate it. 
i was taught a probably ancient myth that swarovski actually uses silver instead of lead. trying to back that up in the web i can not find much evidence besides that it seems to be some myth which keeps floating and some actual products from third party sellers claiming that it is silver. is there anything you could add to resolve any of that?
ok nevermind, i think the myth is not a myth, there is actually far too much evidence looking a bit deeper.
but does that mean there are two kind of materials involved in 2 product lines, lead or silver?
We can think about it logically with a syllogism:
a) I know the render engine follows Snell’s law according to physics theory in literature and does the bending of each wavelength correctly based on that.
b) As partial verification, I know the composition and proportion of it’s spectrum is correct
eg. proportions of green and blue seem to match photo of a rainbow. (loose comparison between Bella rainbow vs photo in the two links bellow)
Given (a) and (b) and given your statement that AB coatings are not as colorful as the OP image, I would conclude that AB coatings are not made to be ABBE-22 … they are probably closer to Abbe 35 or something thereabout. - So my Abbe assumption was wrong about Swarovski coatings.
Due to regulations in many countries it’s not allowed to sell products which contains lead. So no, there is no lead (any more) in the glass. Also there is no silver in the glass, otherwise the products would be much more expensive. It’s a secret mix of non toxic chemicals which results in more weight than normal glass and also an higher IOR.
But there is another product line available which is made out of cubic zirconia Cubic zirconia - Wikipedia which has an IOR more close to diamonds.
Hmm, so if i understand your assumption right, you just changed the IOR and ABBE for a transparent material? I fear it’s not that easy, because as far as i know the AB does not change the ABBE nor IOR. It’s like an optical filter which depends on wave length and view direction and some others.
I made the “hasty” assumption that an AB coating (along with whatever additives might be inside the Swarovski glass itself) would result in an overall Abbe equivalent.
Though it being a coating, I did also add a coating layer to the material, but wasn’t expecting anything of it, because I have no intuition of how that coating combined with the Adde would compound the result. In the render engine the coatings generally behave like a soap bubble causing surface rainbows, but being facetted it causes entire faces to change color. I dialed it down by making it’s IOR about 1.55 (could be 1.6 on that render) close to the main glass so that it’s effect is very subtle. So the rainbows in the render must be from the Abbe and not the coating otherwise entire faces turn green of purple.
So you enabled the “Bella Thin Film” option in the above rendering? Does not look like to me. 
But to say it simple: AB is a thin film coating which looks similar to the examples on the homepage Features | Bella.
Yes, it works like that for rounded things like bubbles. On faceted things, it affects entire facets because of the flatness. The effect can be very intense so I plaid with the settings to minimize it so much that it might as well not have been there.
For example in this iteration it was dialed in so that some faces had a slight yellow tilt
In a previous iteration it was dialed-in for a slight greenish tint
Users like me can play a guessing game trying to dial-in various settings. However, have you guys at Swarovski considered doing a spectroscopy measurement on some of your crystals and making the data available ?
https://refractiveindex.info/?shelf=glass&book=SCHOTT-SF&page=SF1
I would rather select a pre-measured Swarovski complex IOR, than fiddle around in a guessing game 
Sure there were spectroscopic measurements made, but even in our renderings we do not use them. Further, it’s not all about the glass, on such models we us often coating to get at least some colors.
And yes, for the coatings like AB almost the whole facet is colored.
1 Like