So the problem is that there's no metric by which audio devices can be benchmarked?
No, there's just not a quick and easy, yet also relevant to real use, method to do so; and it's not financially interesting enough for hardware sites to spend real money and time on it (that they need to reasonably expect to make back up with the related articles), as opposed to special SSD testing tools, specialized power measurement setups, etc.. I know enough that I could get started with DIY and used equipment, but just with that, which won't be as accurate as something like a brand new AP 2, I would be in $500+, and would then still need time to come up with good,
and repeatable, testing procedures.
Basically, PSRR, CMRR, and related measurements, are what need looking at, along with filtering, more than the possible steady-state performance of the chips. The financial case is there for commercial audio/video product developers, but not so much PC hardware reviewers.