Here is the frequency response measurement of the Sennheiser HD820 versus its open-back siblings (Fig.1):
Fig.1 Sennheiser HD820 frequency response, compared to the Sennheiser HD800 and HD800S
As you can see, this is quite different than
the measurement posted earlier in the thread by someone else. I suspect that the previously posted measurement did not model a proper acoustical load with an ear simulator/coupler -- an apparatus that connects the DUT (device under test) to a microphone in such a way that the working load on the DUT is the same as if used on a real ear [1]. Modeling the input impedance of the human ear becomes increasingly important the higher the acoustic output impedance of the DUT [1]. Also, the need for modeling the correct impedance increases with the proximity of the DUT to the ear [2], which is why we use ear simulators for measuring headphones.
I should also note that we are currently using a newer GRAS High Resolution Ear Simulator (GRAS RA0401). The standard IEC 60318-4 (former IEC 60711) ear simulator was designed in the early 1980's and mimics the input and transfer impedance of a human ear. While the input impedance was based on measurements on human subjects, the transfer impedance was based on the assumption that the ear canal is a simple cylindrical volume with a hard termination. While the GRAS High Resolution Ear Simulator still complies with IEC 60318-4 tolerance band (which is specified up to 10 kHz), it has improved performance above 10 kHz. From 10 Hz to 20 kHz the transfer impedance is within +/- 2.2 dB, resulting in much improved repeatability and more realistic THD measurements [3].
You can find out more about the measurement lab at Head-Fi HQ at the following link:
Head-Fi Audio Measurement Lab
While the HD820 measurement may look rather unusual versus the comparatively very linear frequency responses of its open-back siblings (HD800 and HD800S), a comparison with another premium closed-back headphone (the Sony MDR-Z1R) shows it in a different light (Fig.2):
Fig.2 Sennheiser HD820 frequency response, compared to the Sony MDR-Z1R
The steep drop that starts <200 Hz is there by design. If you want to better understand why it's there, make sure to watch this presentation by Axel Grell of Sennheiser from CanJam NYC 2018 (for discussion of this specifically, start watching at around the 27:00 mark):
Simply put, there was significant consideration given to an internally developed preference curve and psychoacoustics in voicing the HD820 (and Axel Grell does discuss other preference curves when asked about them in the talk above). If you haven't already done so, I strongly suggest you watch the entire presentation.
Given that it's a closed-back headphone, the HD820 will be more sensitive to breaks in the seal than its open-back siblings, but that's to be expected. As I said in an earlier post, I get a good seal with the HD820, even with my eyeglasses on.
As I also said earlier, I definitely prefer the HD820's tonal balance to the HD800S (and, for further reference, I strongly prefer the HD800S to the HD800). That said -- and this shouldn't be at all surprising -- the HD800S is, to my ears, ultimately more resolving. Again, the HD800S is open-back, free-breathing, unencumbered by some of the issues of acoustics/physics that
must be dealt with for this class of closed-back headphone with the HD820. What resolution the HD820 does sacrifice to its sibling, though, is worth the gains in the tonal balance for me. Still, the HD820 is without a doubt one of the most resolving closed-back headphones currently available, and also among the widest imaging of them.
By the way, here are the THD measurements, first versus its HD800 and HD800S siblings (Fig.3), and then versus the Sony MDR-Z1R (Fig.4):
Fig.3 Sennheiser HD820 total harmonic distortion (THD), compared to the Sennheiser HD800 and HD800S
Fig.4 Sennheiser HD820 total harmonic distortion (THD), compared to the Sony MDR-Z1R
The measurements included in this post were made on the same day using:
[1] Brüel, P. V., Frederiksen, E., Mathiasen, H., Rasmussen, G., and Sigh, E. (1976). "Investigations of a new insert earphone coupler," Part I in "Impedance of Real and Artificial Ears," Brüel and Kjær report.
[2] Brüel & Kjær, "Measuring Human Audio Perception," presented at the 2018 ALMA International Symposium & Expo (AISE).
[3] Wille, M. (2017). "High Resolution Ear Simulator," GRAS Sound & Vibration white paper.