I’ve placed the above macro that Romullus wrote into my base Corona material in Editor and inactivated it (checkbox off). I did this since day one.
Yes it’s effectively what Pokoy writes could be called ‘IOR map’ but I am ok with this solution which is how FresnelIOR map should have always been (Since no one really uses it because of how cumbersome it is).
Your answer is exactly why I posted the question in the first place. Thanks.
(Report ID=CRMAX-168)
Looking at my answer above I see that I should have put this in a more polite way, sorry for the harsh tone. I should back up my critique with screenshots and examples but I’m not going to be in front of my PC for the next 1-2 weeks and just can’t be more specific right now. Not ideal, but I hope it’s clear why the current approach isn’t very user friendly right now.
What I dislike is that without inspecting maps or map trees you won’t be able to tell what the IOR value is. Also, by using values outside of 0-1 map thumbnails will always be white, making it impossible to quickly check the final map, especially if you stack them you want to know what you’re doing - an all white final map result won’t tell you anything.
By mapping the map/map tree values instead you can immediately see what the final map stack is, since it displays in 0-255 and you specify the max IOR value in a numerical way right within the material’s UI. This way you get the info much quicker with the same result.
I’m eager to see why others don’t see anything wrong in going the more winded route. Maybe I’m missing something and judge too quickly.
Sorry again, Maru.
No problem at all! I am happy to hear any feedback and actually using a more straightforward tone usually makes the message easier to understand.
How do you set a translucency color in the new material? other than the base color.
See:
we are aware that some of the features which are grayed out are not really clear UX/UI-wise on how to enable them. We have a fix ready in our development build, which should arrive to you soon.
In the meantime:
Translucency requires “Thin Shell” in General Tab
Refraction>Thin Absorbtion requires “Thin Shell” in General Tab
Opacity>Clip requires a texture map in the Opacity Texture
Volumetrics require refraction to be enabled
SSS requires Refraction at lower than 100 value.
Source: https://forum.corona-renderer.com/ Why Is the New PBR Better? #msg180472
Seemingly you cannot quite do that anymore. GeorgeK posted this line a couple of pages ago: “In regards to your second question, the translucency fraction is still there labeled as “Translucency” you can enable it and use it without the use of a map simply by enabling the Thin Shell (no inside) option above Base Layer. The color of the translucency depends on your base layer color.”
Maybe I’m wrong 
Ah, damn, it looks like I misunderstood the question, sorry!
For what its worth, you answered a couple of my questions with it, ha!
There is one more thing, due to the connection between roughness of reflections and roughness of refractions, it became virtually impossible to add, for example, fingerprints to the glass surface, because even the stained glass remains with 0 roughness, at first I thought that this could be solved with clearcoat but my attempts did not lead to anything good. It seems a little strange to me that when adding new functions, old ones break) and ofc the same concerns translucency color map, which in some cases just necessary
I hope you wont delete the legacy shader because sometimes some “fakes” lead to good results or maybe it will be a good idea to add “expert” mode in our new physical material which will break those “physicall correct connections”
And I also recently remembered how we discussed the idea of adding a fake thickness function [https://forum.corona-renderer.com/ Real thickness of materials with opacity ](https://forum.corona-renderer.com/ Real thickness of materials with opacity ) for thin meshes that would allow using the opacity map to simulate an effect similar to corona round edges in legacy mode without having to make a separate super-detailed normal map. It would be a great addition to the new thin shell and volume switching function in the new shader
I haven’t tried it myself yet but I suspect that a potential workflow for this could revolve around having the glass be the base in a Corona Layered material while the fingerprints come on top with a mask using it’s own separate material.
I hope the name of CoronaLegacyMtl class would go back to CoronaMtl class which would be really easier to find because of the years of habit and since it completely distinguishes itself from the CoronaPhysicalMtl class I did not grasp the need for changing the class name.
ps: I know it works with both in maxscript because of the compatibility and we have two class for one single thing but I mean the name that shows in the material editor
no doubt this can be solved, but what is the point in this innovation if it does not make working with the render engine easier?
In this case, we may not use maps for roughness or base color at all, but simply mix the necessary materials by masks: D: D: D
Why isn’t this solved by just passing a roughness / glossiness map into the roughness / glossiness channel of a refractive (glass) material? I would presume that would be sufficient for ~90% of cases - maybe even for the remaining edges cases in which a close-up is required(?)
because even a refraction glossiness of 0.99 results in blurry unwanted results for me) ofc it is not “critical” but personal for me, i want to have an option to unlink refl/refr roughness
Working with the coat layer for the fingerprints wouldn’t solve your issue?
logically, this approach should work, but I still cannot get it to produce a similar result. It would be logical to set the required roughness value for the coat layer and apply it using a contrast mask, but as a result, the picture is more noisy and the pattern of smudges does not turn out to be as clear, of course you can always tweak this to death, but it seems to me that this should not work like that.
I think the map should go into clearcoat roughness (inverted), not clearcoat amount. Clearcoat amount needs to be >0 then, of course.
^this 
Nope) you can try it yourself) as I said, a workaround can of course be found, but I cannot understand the ultimate purpose of this simplification if it only complicates the task.
PS and finally, if this is a “physically correct” shader, why should I add a layer of transparent glossy varnish to the glass if it is not there?
