I've never understood that part tbh. From Wikipedia:
How can you accurately do that on a headphone like the DT1990, which will distort relative volume levels due to its piercing treble and poor bass response?
View attachment 249737
Like, won't a track mixed on the DT1990 sound awful when played back on studio monitors? With dull hi-hats and disgustingly boomy bass, because while mixing you were unintentionally but inadvertently compensating for the headphone's own shortcomings?
First and foremost, headphone mixing is not common, as it is very difficult to do stereo imaging with it, and so it is an afterprocess of standard mixing with studio monitors. Usually these, if not SOTA like Neumann and Genelec or some good JBLs, won't have
good FR perfomance even though they are praised as being the industry standard. The end result,, thus, as I see it, is more relative of the standard we expect of the industry rather than what would be considered neutral and natural, objectively.
The default monitoring gear, like that Yamaha I linked (or even a old version that I forgot the name, but had an atrocious FR) or the infamous classic AKGs, may sound very weird for trained ears but also may amplify incoherences on the original mix. And that is where I see the main use of headphones on the studio environment: either for monitoring on-stage, which requires great isolation and often times benefit from boosted a midbass "bloomy" shelf; or for checking for inconsistencies on the mastering process.
That is of course speculation on my part, but it's what I see when observing friends who work on that area. And I don't think this pans out well with hifi recordings, where we see those SOTA brands more.