I refuse to believe that these £25 IEMs can perform as well as they undoubtedly do, so I'm going to bin them and go back to relying on expectation bias (AKA my ears) only.
I like the ASR approach: only objective measures can give useful information about the quality of a device.
Subjective impressions often lead to coarse errors of evaluation.
Here the problem is to understand if the measures take into account all the factors at stake.
Perhaps the artificial ear is rigid and not flexible so the measures do not take into account the contractions/deformations of the auditory duct?
From my tests, there are clearly noises/distortions in the 12Hz <-> 30Hz range which are perfectly audible by everyone.
This does not happen with all IEMS (but only with some) and does not happen with headphones.
We must understand how to measure these artifacts because they are audible differences from to the original sound.