Frosted glass

[Peter Milner]

The frosted glass material here is created solely by adding a fine sand texture.

The glass partition is modelled as solid.

The background image shows through the frosting clearly without being broken up and also shows through in areas where it shouldn't be visible.

I shall see what happens if I use blurry transparency instead.

[Chas / Aug Posekany]

How did you model the glass and frosted areas? Is the frosted glass just a thin film over the the full height piece of glass?

[Peter Milner]

No. The frosted section is a separate solid. I've tried using a thin film in the past, but it does odd things with the IOR. Maybe a thin film with IOR of 1.0 and blurry transparency will work. I haven't tried this combination yet.

[Marc Chaumier]

And don't forget, but I think you know it, to let a micro space between the faces of glass and film. In fact, I say that for other nXt users who are not so hardened.

[Peter Milner]

Any development on these bugs yet?

Peter

[Roy Hirshkowitz]

Not yet-- I may be able to play with this a little today. Have you tried just using a mirror for the desk top and seeing if the reflection is accurate?

[Peter Milner]

I did that with the test renders further up in this discussion.

[Roy Hirshkowitz]

Missed all of these too--- thanks.

The second example, showing the "brighter" sky, is probably OK ( think). As the surface becomes less specular it will will reflect more of the sky-- eventually approaching a white piece of paper as your shininess get very low. The area of the sky reflected by the mirror could easily be darker than other areas.

The first example I need to look at more closely-- I should get a chance this morning.

[Roy Hirshkowitz]

Would you mind attaching your test model?-- it will save me a few minutes.

[Peter Milner]

I'll send the model via the upload as I'm having trouble attaching it here.

[Peter Milner]

The point about the brighter sky is that there is a marked difference between 1.0 and 0.99, but there is little difference with values from there down.
This one has sharpness set to 0.2

[Roy Hirshkowitz]

Yep-- I think I'm getting a better handle on this one. It has to do with the code I used for make glossy surfaces converge faster under interior daylighting type scenarios. I may need to remove this optimization in or to solve this. This will certainly have a penalty on many of your scenes (not the current scene since the windows are large)-- for example, glossy plastic at some distance from a window will have more fireflies and take longer to resolve. On the other hand, all of the specular reflections, whether sharp or glossy, will be calculated in a more consistent manner and any intervening windows will be treated exactly the same-- we'll have to see whether it's worth it.

I could use your original office scene as well for testing this....

[Peter Milner]

The samples I've shown with the mirror and pillars all use exterior lighting (sun and sky) and are rendered using Pathtracer.