PLEASE! I really can't achieve a decent frosted glass effect!
In particular I have the need to develop a component (sort of adhesive sheets of frosting) which always goes applied to a glass. Seems really IMPOSSIBLE to get!
Hints accepted!
PLEASE! I really can't achieve a decent frosted glass effect!
In particular I have the need to develop a component (sort of adhesive sheets of frosting) which always goes applied to a glass. Seems really IMPOSSIBLE to get!
Hints accepted!
I was curious if using a bump map would be an effective way to produce a frosted effect. See if this is useful.
Adjusting the bump strength, opacity, and refractive index gives different strengths of the effect.
+120 for me!
I just spent several HOURS trying to get something decent...
please note this should work in both NIGHT and DAY renderings..
I found that while it was "OK" in daytime, the night (DARK BACKGROUND REFLECTION) simply made it look like TINTED GLASS instead.
thanks a lot guys! I think AndyBot solution is great, it has some issues when not looked at just from a frontal view but it's so much better than what I achieved!
Thanks again: I owe you guys a beer!
OK: did some testing and I have to report to the DEVs that this fantastic material has a strange behaviour. It works great if it's "standalone", BUT gets somewhat downsampled if I put it in front of a glass (single or souble-glazed).
I have the need to apply the frosting like sort of a film onto a glass. So I have to keep its frosting abilities even if there's a glass behind... Is it feasible?
I'm curious - Enscape support a frosted glass effect and I used it some times in the past. Why this feature isn't used?
I'd love to, and I really tried (it was my first try) but seems like the amount of glossiness is too low to get the desired result, even if I give the max amount.
Maybe developers could enhance more this behaviour, setting that a 100% glass frostiness nearly equals a 100% diffuse.
I attach an image of what I'm trying to get. Note that I need to have a separate model for the glass window and the film, so basically the material should behave correctly both if standalone (e.g. floating in the air) and if applied NEAR (say 0.25cm) a glass pane.
The test that I did has 3 panes, one is clear glass so you get the reflection like regular glass. With the bump method for the frosted glass, you won't get clear reflections.
I did some more testing, and it seems increasing the opacity makes the material darker and doesn't change the opacity in the rendering. Here are some additional tests, along with the test scene.
minimum opacity, color is maintained. Notice the wine glass is not treated the same through the glass
half-opacity - material gets darker. same concern with wine glass
I really appreciate your efforts Andy!
But at this point I think it's useless for users to struggle to simulate a behaviour when it's clearly hardly reproduceable.
I bump my requesto to developers to increase the control over the "frosted glass" option, or to develop a new shader which renders in something like I showed before.
I'm not quite sure that I understand the request. So just that I get this right - the complaint is that the blurryness you get with frosted glass with maximum settings is not strong enough? Or is the request to have both frosted glass blurring and glossy reflections at the same time? In that case I'd suggest to add a regular transparent material with 0% roughness in front of it:
Not exactly.
In the first place I'm not complaining: I'm just reporting something I feel missing.
What is missing (in my opinion) is a technique to get what you see in the reference image I attached some posts above.
It's not technically a "frosted glass", because it's a frosting film I put ONTO a glass. So I guess some sort of semitransparent translucent very blurry shader is missing.
As an alternative, I could consider to approximate the film with a frosted glass element but in this specific case:
1. the max amount of frosting is too low (in the ref image you can hardly distinguish what's inside the room);
2. when applied in front of another glass pane, the standard frosting material gets reduced (I think it's some sort of issue related to simplification of reflections/refractions of realtime rendering).
I didn't want to sound in any ways harsh with my request, btw.
No worries, I didn't perceive your request harsh in any way
Just want to understand what exactly you feel is missing as I'm still not quite sure what our current material model lacks in order to achieve the result of the reference photo. You can combine semi-transparency with the frosted glass effect like here - and I guess that does look quite similar? There are however certain limitations to those kinds of materials as the technique we use doesn't allow for arbitrary large blur radii.
pibuz I agree there is a limit that is reached when you want a "milky" kind of frosted effect. That does not seem possible with the current material parameters.
Micha It does not appear that the "frosted glass" checkbox does anything as far as I can tell. If others have examples or settings where it's worked, I would love to see it.
Clemens Musterle The result you show COULD be something SIMILAR, and I just tried to reproduce following your kind instructions.
Seems like the good blurriness you get depends mainly on how far the objects are behind the frosty plane. In architecture this can work (in some but not in all cases) but I work in the fair business, and we rarely encounter architectonic dimensions, so basically I have frequently very near objects (about 1m far) so this workflow doesn't lead to the desired effect, sadly.
Is the frosted glass material applied to both front & backside of this plane?
And yes the blur radius is determined by the distance to the object behind as it is in reality, however as I said the intensity is somewhat limited, but a few centimeters should already result in a noticeable blur, as can be seen on the bench in my above screenshot.
I'm also having trouble getting any effect from the "frosted glass" setting. Is there something I should be doing differently in this example file? This is a similar scale as a cubicle enclosed with frosted glass, so it's certainly within the word of architectural scale.
I'm also having trouble getting any effect from the "frosted glass" setting. Is there something I should be doing differently in this example file? This is a similar scale as a cubicle enclosed with frosted glass, so it's certainly within the word of architectural scale
When I downloaded your model I did get the frosted glass look. The strange thing is that where the framed piece and the middle glass overlap you lose all of the forsting and bump from the main view. If you walk around to the other side the bump and frosting show correctly.
pibuz here was my attempt at the glass you showed in your example. It's a 2 layers of glass, or really the front face, and the back face of a piece of glass. The front is glass with the transparency set to a grey-ish material to provide the light transparency. The back face is a a plain glass set to ransparent. If you want the cut out to read properly you need to match the cut out in each plain, but it looks pretty decent.
I'm also having trouble getting any effect from the "frosted glass" setting. Is there something I should be doing differently in this example file? This is a similar scale as a cubicle enclosed with frosted glass, so it's certainly within the word of architectural scale.
Your frosted glass material still had a very low roughness value and therefore the effect was minimal. Roughness basically controls the amount of diffusion you get from the frosted glass. Also I'd highly recommend not to use a white noise texture as bump unless you really want that very noisy, speckled look - I removed that for better demonstration in the screenshot.
The strange thing is that where the framed piece and the middle glass overlap you lose all of the forsting and bump from the main view. If you walk around to the other side the bump and frosting show correctly.
That's a correct observation of a current limitation of the technique. As mentioned before the blurriness of the background depends on it's distance to the frosted glass. However if there's an immediate overlap with another transparent material behind the renderer doesn't know "the actual" depth of the background, but uses the depth of the overlapping transparent object.