Yeah, this is a really tricky topic, and I am thinking about / working on a total redo of the rule book
Addendum: Summing Bus Treatment. It's not as to-the-point as it could be.
Looking at the mix description - it also depends on how much was actually processed, which wasn't further commented on. It could have been just "kissing -1,00dB to -2,00dB of gain reduction" per device, or the device could have been mostly bypassed. This wasn't documented by
@Christoph_K (with
MC102 post #111). I have to assume that it was just minimal processing... for flavor.
This is supported by the signal analysis with a the PLR (Peak-to-Loudness) range value of 16,33 LU, and the Loudness Range is about 5,0 LU, compared to a max LUFS ILk of -17,7 LUFS ILk and max dBTP value of -1,37dBTP. Although, it
does feel a bit squashed while taking a quick listen (perceived - due to how strong the drums are processed), but it fits the premise and task. Plus the transients are definitely there.
The general understanding with such setups is "summing glue", which the Fabfilter Pro-MB and later the Studer A800 tape also hint at. Actually a normal process with modern pop/rock music. (as in: running at least a summing bus compression, or a tape simulation, etc - to bring things "more upfront").
I'm not the mix engineer - so I can't further comment on what was done with the mix.
Neither am I the Song Provider - maybe he likes the general idea, but he would still like to see / hear some adjustments in a certain direction before he makes a final decision.
Finding the fine line between
"your mix must be able to stand on it's own and work on all playback mediums" and "
do not finalize (master)/loudness adjust" is not an easy task - I can definitely attest to that.