Issue with glass

[trixian]

Well the main thing was that on the glass where I used fog, it all rendered close to opaque. (This was what alerted me to this issue at first).
This being because of "insert Vlado's explanation here" as seen up top of this thread.

Without fog it seems to be OK, but I really want that fog in my glass for some reason. As for Fog:yes Bump:no, I haven't tried I think, as the bump\distortion is the main goal here.
Ideally the fog should not interpret the glass as suddenly very thick because of the bump, as the reflections and refractions are being interpreted correctly by the same bump, but who knows

I really hope you figure something out, as I can't seem to get anywhere.

[grasshopper]

Hi, I have struggled with fog and multi-sub maps in the past too:

My issue was that I wanted to show a matted glossy curved edge to some thick saftey glass. I couldn't work out a way to do it and still use the fog multiplier so had to just put up with the reflective edges.
This was before the VrayMultiSubTex but I can't think how it will help this senario. If anyone has resolution to this please let us know.

[RE:FORM_STUDIOS]

I recreated the problem you have; ie variation in the fog colour. I built the 3 ID vrayMultiSub tex and the fog does seem to become uniform again without loss of the bump texture. I only tested a box so the shelled poly might be causing you the problem.

The way I used to create variation in glass is to apply a simple noise map(explicit uv mapped), scale it to the right size, then run a script which would unwrap the object (contain all panes of glass in one file) and randomly move the uv polygons. You would need to do this prior to shelling the object ofcourse. I'll see if I can find the script to randomise UV placement. It was nice as it avoided the problem of having to break the single windows object into seperate objects to randomise either map or IDs.... just a single noise map and single ID did the job. Not perhaps the most realistic result ofcourse. There could be a clever approach possible if you put all your different window pane distortions into one map and then randomly move the UV's by the correct value to jump cleanly to each specific tile in the map. (not the best explanation.. but you get the idea).