[simmsimaging]

I'm testing a scene where I want to have a glassy/liquid shape but need to control the refraction separately from the reflection - I need nice glassy reflections but the IOR required makes the refractions a bit crazy, so this is a cheat.

My glass BSDF is set to an IOR of 1.12 or something low, and then I stacked another layer on top in additive mode to create reflections only. It's set to IOR of 1.5. This is pretty straightforward stuff. The problem is that the top layer does not work as expected - or as it does when stacked on top of a non-transparent base layer.

Rather than get added simple reflections it gets really complex and overbright reflections. It was also crazy bright so I had to lower the layer opacity way down ( ( think it's at 40 percent right now) I tested it by adding the reflections with a coating instead, and while that looked more like normal, it's also hard to get coatings clean and without dispersion effects so I prefer the layered material route.

See the image below for reference. I tested it with a solid base material and it looks as expected - coating is roughly the same as the additive layer, but not at all the same with the refractive base layer.

Is there a known problem with layer stacking like this on transparent materials?

[Half Life]

You've just run afoul of one of my pet peeves about the way Maxwell works -- coatings combined with transparent BSDF's do not work as they should (In my opinion).

Permit me to get on my soap box for a minute

To create anti-reflective glass (the most basic example of the need for this functionality) you need to have the Nd of the coating interact with the Nd of the transparent object to reduce the reflective properties... being able to stack multiple coatings of differing Nd's would be even more accurate for anti-reflective coatings (like you would see on camera lenses).

I don't think anybody else complains about the lack of ability to do this in Maxwell but to me it is a big problem because many many manufacturing processes rely on many stacked microscopically thin coating layers to achieve the finished effect (common plastic ketchup bottles are also a good example of this).

I'm one who believes that reproducing the real-life manufacturing process within MXED is the most logical and realistic method of material making.

I feel better now!

That said, additive layers will not work at all for this type of effect and your results there are completely expected based on how additive layers work.

Best,
Jason.

[simmsimaging]

You've just run afoul of one of my pet peeves about the way Maxwell works -- coatings combined with transparent BSDF's do not work as they should (In my opinion).

Permit me to get on my soap box for a minute

To create anti-reflective glass (the most basic example of the need for this functionality) you need to have the Nd of the coating interact with the Nd of the transparent object to reduce the reflective properties... being able to stack multiple coatings of differing Nd's would be even more accurate for anti-reflective coatings (like you would see on camera lenses).

I don't think anybody else complains about the lack of ability to do this in Maxwell but to me it is a big problem because many many manufacturing processes rely on many stacked microscopically thin coating layers to achieve the finished effect (common plastic ketchup bottles are also a good example of this).

I'm one who believes that reproducing the real-life manufacturing process within MXED is the most logical and realistic method of material making.

I feel better now!

That said, additive layers will not work at all for this type of effect and your results there are completely expected based on how additive layers work.

Best,
Jason.

Thanks once again Jason. I agree that it would be great if the MXED more closely aligned with reality in terms of stackable coatings. While at it - it would be nice to see coatings have a rougnness parameter too - then I would be able to use them far more often.

Guess I'll have to do this one in Vray instead - I need the control of splitting the reflections out.

Thanks again!
b

[Half Life]

For what you are looking to do you can try to boost up the coating to about 10000 nm thickness and the interference (what looks like dispersion) effect will be minimized/eliminated... the thicker you make it the less you will see those effects.

Don't mind me, you can make what you want work in Maxwell(although rendering time will be long) -- I just took the opportunity to gripe

I love coatings -- I just wish they were more powerful in Maxwell... and I'm not a fan of additive layers (except Ghost layers). So I got off on a tangent and forgot to answer your question directly.

Best,
Jason.

[simmsimaging]

For what you are looking to do you can try to boost up the coating to about 10000 nm thickness and the interference (what looks like dispersion) effect will be minimized/eliminated... the thicker you make it the less you will see those effects.

Don't mind me, you can make what you want work in Maxwell(although rendering time will be long) -- I just took the opportunity to gripe

Best,
Jason.

I understand.

I looked at upping the thickness, but even with it set to 15-20K it was still looking a bit off colour, and grainy. It seems an inefficient way to go - unless the coatings can be made to run clearer/cleaner faster somehow....

Thanks /b

[Half Life]

Yeah, uber-long render time... in the several hundreds of hours.

Which brings me to my other pet peeve (it's about caustics), but I'll save that for another time...

BTW, it's worth observing in the "coated" version that the darker areas of the refractions are more pronounced and generally the refraction in general seems a bit off... I've been curious what causes this.

Best,
Jason.

[simmsimaging]

Yeah, uber-long render time... in the several hundreds of hours.

Which brings me to my other pet peeve (it's about caustics), but I'll save that for another time...

Best,
Jason.

If it's about caustics not working with interconnected/penetrating transparent objects then I'm with you on that one!