Since I’ve got a fair bit of listening time in now, I figured it was time to measure these things (don’t like to do that before I listen for a bit). This is at 72 hours of runtime.
Plots are provided for HEQ (closest to Harmann), HPN, my custom compensation (what is perceptually neutral for me) and the RAW plot (the peak at 4.5 kHz is a common artifact with the EARS tool).
The above plots are smoothed at 1/6th octave, with the same vertical scaling, to match against those @Resolve did in his review.
And I typically do 1/12th octave smoothing, which looks like this (all other things the same):
Fundamentally, these plots look like those done by @Resolve. Which is encouraging, Differences are, I expect, down to variations between miniDSP EARS units, and measuring environment.
Details on measurement rig, approach and calibrations can be found here.