When working on the linear pipeline support for the fog¹, I noticed that, unless I'm wrong, the fog texture is an RGBA image but we only use the alpha channel, setting RGB to {1, 1, 1}.
We may instead store an image with less color components. Also by storing this alpha channel in something like a red channel, we may use sRGB variants and let the hardware do the sRGB conversions.
When working on the linear pipeline support for the fog¹, I noticed that, unless I'm wrong, the fog texture is an RGBA image but we only use the alpha channel, setting RGB to
{1, 1, 1}.We may instead store an image with less color components. Also by storing this alpha channel in something like a red channel, we may use sRGB variants and let the hardware do the sRGB conversions.