Looking ahead at things to come: Voxel-based Volumetric Fog
Looking ahead at things to come: Voxel-based Volumetric Fog

Looking ahead at things to come: Voxel-based Volumetric Fog

Today we want to show you an awesome improvement that will be included in the next release, in addition to a large number of fixes and optimizations: Voxel-based Volumetric Fog.

How it works in the engine

Voxel-based Volumetric Fog uses volume textures as a view-frustum-shaped voxel buffer to store the incoming light and properties of participating media.

Light and FogVolume entities inject their data to the volume textures. After the textures contain all the data to visualize the fog, ray-marching is done through the volume textures.

It accumulates the illumination along the ray from the camera and attenuates the illumination by the fog density.

The illumination is stored to another volume texture and finally it’s composited to the screen by a post-process effect.


How it’s improved from 3.8.1

From 3.8.1 the visual quality was improved. Particularly, flicker and ghosting artifacts were greatly reduced.

To achieve this, downscaled shadow maps and an improved temporal re-projection filter were added to Voxel-based Volumetric Fog.

Furthermore, optimizing and changing the process of the volumetric fog also contributed the improvements.

We captured some high resolution screenshots of this improvement in action, so please let us know what you think in the forums!

All the best,

The CRYENGINE team.