Quote from: pokoy on July 02, 2018, 12:41:13 PM
Quote from: WASasquatch on July 02, 2018, 01:23:18 AM
Did some testing with that note about voxels, and increasing voxels to compensate for radius actually makes the clouds brighter, instead of normalizing them like they are at base radius. I am assuming this is the lighting model that this is coming from?
...
One thing to keep in mind is that clouds are typically much less stretched our over world y-axis, meaning that you could end up with only a few voxels for their vertical span, easily leading to too little illumination information to be accurate. It would be cool if there was a multiplier for voxel count for y-axis only.
That seems to be a struggling problem with clouds and density with lighting. I've been trying to create a cloudy day cloud setup like in WA and so far am not able to get clouds lighting to interact with each other appropriately. Trying some stuff now to see if I can obtain the look with a single cloud layer and a custom density setup but so far doesn't look like it's working either in what is rendering.
Seems clouds don't allow enough light to actually pass through them. For example, even pretty thin and bright clouds end up making your terrain as dark as twilight hours.
Another issue I've been trying to figure out is why there is no ability to adjust, basically, the "specular" roughness of clouds. Any time the sun is in the scene you get a very hard bright spot from the sun, which doesn't cascade over the clouds to create a "bright region" rather than a spot. Even thin clouds blot the sun very well and light the whole cloud area rather than beaming through, while the clouds themselves produce enviro light for the ground to compensate for the sun being blocked as the whole region is glowing from the sun at a direct observable angle
Edit: Hmm, ironically, upping the cloud radius from default stretches this "spot" created by the sun it seems. Doing a test render. Edit 2: More of an illusion in the preview.