Is refraction fog Broken?

[jbbarnette]

I have tried every cobination of values of

fog multiplier ".001-1"

with an almost white blue.
and the
Fog Bias I have tried everything between -5 and 10

I can easily get the swatch to look correct but in render I'm getting absolutely no falloff. What I'm looking for is to do water and for objects to fade off into the blue the deeper they go.

This used to work just fine back in the days of the Beta. and This is described to work this way in the training of Vray on Digital tutors. But in both 2.30 and 2.20 it seems to be broken.

Even a almost white shade of blue in the fog color gets be a very dark result in render. it absolutely no falloff based on thickness.




Now going by this settings and what we see in the swatch as well as the descriptions of how they works that should provide a almost crew that darkens as the distance the tween the surface of the water and the point being refracted increases.


Oh BTW reflection is off and diffuse is pure white

and that fog color is blue-turqoise with the saturation value of .001 so almost white. and the radius of the cylinder is 6 and scene is in CM.

[Dariusz Makowski (Dadal)]

Heya

I think we were doing it yesterday on a test scene. Are your normal's correct and your object is thick ?

[yolov]

message edited

[jbbarnette]

yeah I cant get a gentle falloff to save my life. it seems to go instantly to the ridiculous.

[Dariusz Makowski (Dadal)]

Heya

Can you send me scene maybe or put it here on forum?

Thanks, bye.

[Svetlozar Draganov]

Hello

Here is one example with Refraction Fog and Fog Bias. The values for the Fog Color , Fog Multiplier and Fog Bias are very very low.


Scene File >>fogBias06.zip

[Deflaminis]

I want to revive this thread here because this is very odd in terms of usage. The spinners become hard to use in this case as a very low value is required. I'd like to request that the range of practical usable numbers here be normalized so we don't have to deal with thousandths of a unit. I imagine there is a scientific reason it is the way it is but maybe even a checkbox for a two decimal point move to the right.

Additionally, it seems that the RT engine gets the colors I choose right, but not in the final render.. The fog color in my material editor matches the RT render, but not so for the production render. Here is a screenshot.

Using the 'suggested gamma setup' from 2014 thread vlado posted here: http://www.chaosgroup.com/forums/vbu...788#post579788 Is this problem related to gamma correction on the fog color?

Attached scene file
odd.zip
.

Just not sure how to use this easily.

[jbbarnette]

hmm I have to say that honestly neither of those look right. but I agree that something needs to be adjusted as the numbers required are soo low it is almost unusable. in Maya you almost need a value of .00001with is beyond it level of precision.

[Deflaminis]

Well, I wasn't trying to create something beautiful there, just testing.

[Svetlozar Draganov]

I agree that these parameters are not very comfortable to work with - and the result in most cases is different from what we expect.
I've just run a few tests on Deflaminis scene and I have a way to achieve more predictable results with it.
If Fog Bias parameter is set to 100 - the final color is very close to the one used in Fog Color parameter, furthermore the RT result and Production results are now almost identical and if the Fog-Multiplier parameter is changed both renders follows each other.
I have tried to render different colors with different Hue/Saturation/Value parameters and every change produce the same result between both renders and the end-result-colors matches the input-fog-color.

I could make a proposal to our Developers to change Fog Bias range but I am not sure if this change will work as expected in any situation.
Could you guys run a few tests with Fog Bias set to 100 - and give us a feedback if this works better or it just works better in this scene.

Here are a few results from my tests: