Two nodes for working with multi-pass renderer output (V-Ray, Arnold, Redshift, Cycles, Octane, etc.).
Composites a beauty pass with optional auxiliary passes:
out = beauty * (ao_strength · AO + (1 - ao_strength))
+ diffuse_gain · diffuse
+ specular_gain · specular
+ emission_gain · emission
All auxiliary passes are optional; missing passes contribute zero. Inputs are auto-resized (bilinear) to the beauty resolution if needed. This is intentionally simple — for full re-light workflows use a dedicated tool, but for quick AO multiplies and emission boosts this covers 90% of use cases.
Converts a depth pass into a per-pixel circle-of-confusion (CoC) mask:
coc = clamp(|depth - focus_distance| / aperture, 0, 1)
Outputs both the CoC mask (1 = fully defocused, 0 = sharp) and an
in_focus mask (its complement). Use these to drive a separable blur
node for cheap depth-of-field, or as alpha layers for compositing.
The depth_channel selector lets you pick R / G / B / luma when the
depth pass arrives encoded in an unusual channel layout.