Raytraced Glass that looks like...Glass (4.25)


So 4.25 is released and with ’ production ready ’ raytracing…however! I still struggle to get glass to look and behave like glass. Specifically a car headlight diffuser…

I create a raytrace project, add a PPV (and turn on raytrace refl, transparency etc in details) add an automotive materials glass shader to my mesh and it as per 4.24/Preview 4.25 the result looks nothing like glass blocking light to objects behind it. Only raster translucency let’s light through but does not offer the realism I’m after.

Am I missing something here?:wink: Could someone please share the steps in order to get glass looking glass with raytracing?


I think when I was playing with this I found that the specular channel is actually the IOR - add a parameter that and experiment. Also - change the translucency shading mode to Surface Translucency Volume.

Hey Dannington and thanks for the suggestion.

Yes, Spec is = to Index Of Refraction…
Even if I add a value in the master material of M_Glass_RT_Master from the automotive materials pack [should it not have an IOR value to begin with since it is meant to be a material for glass out of the box!?;)] I still get an unsatisfactory result with RTX.

Here, I have a Chrome headlight surround with a simple sphere with default material on…As you can see with Raytrace translucency the area behind the glass becomes a blanket grey where as with Raster translucency on my PPV I can see the chrome on the inside [like in real life;)…

where as 1.6 is the value given roughly to glass in offline renderers this seems far too big a value in UE4/RTX so by making Spec [IOR] 0.16 instead of 1.6 I can get a better refraction result but Im still left with boring grey behing the glass mesh where as I want to see some of that lovely shiny chrome.

One thing is clear RTX translucency with automotive shaders does not work straight out the box!


just to add…aside from raytraced translucency seemingly not working…adding a sphere behind my glass diffuser mesh and applying both an emissive on one [in image] and standard white or another I get a shimmering effect on the inside of the chrome housing… this same artifact happens on another mesh with M_GlassRT_Master applied [with ior of 0.16] and raytraced translucency. Is this RTX trying to calculate the solution or something else. regardless it never resolved and is always there.

the above shimmering/ brushed artefacts are not raytracing.
i opened the M_GlassRT_Master node network and removed individually each input to the material and only the opaciity fed node network makes this occur. hard to say if its the opacity itself or the other inputs node networks that make it apparent.