Looks like this effect just happens in any case no matter the light source.
This is lit with two sun lamps that approximate the sun and the sky:
Scattering spectrum = Absorption spectrum:
RGB:
Spectral:
Diff:
This second one is quite different mainly due to my own doing. I probably should redo it to be more comparable. But basically, every aspect I could make spectral (in that I’m not relying on an upsampled RGB spectrum but instead I’m using some custom spectrum) I made spectral here. I could just make them 1:1 by always using the RGB value I’d be using in regular Blender instead.
But anyway, in this one the scattering spectrum is the Rayleigh spectrum.
RGB:
Spectral:
Diff:
Seems to me the spectral lighting here is slightly paler than its RGB counterpart. But definitely it is a whole lot less red. That’s the one constant throughout all of these tests. That more spectral version (in that the lighting is also not RGB-based) is clearly even more blue though so yeah, as said, probably not fair.