Howdy Maya,
Fwiw, I just finished re-reading your post on this in the K371 topic. And found it pretty interesting. FYI, I am not surprised that you found a possible L/R imbalance in the K371's bass response,
In regards to the K371, I can't fully untangle the HPs' own imbalance from how their seal may vary because of my own anatomy's asymmetries and their general susceptibility to seal breach (even performing averages of individual seatings result in different outcomes with the K371 on my head).
The higher-Q notches one can see on one channel vs. another in the bass region are unlikely to be caused by the latter given their shape and the type of headphones (this is not a planar with a fully sealed front volume), and are more likely to be caused by the headphones themselves indeed. Besides, they vary between the samples I've had. So while I'm quite confident that they aren't the last word in manufacturing consistency I'm not sure that I can quantify exactly the degree.
The TLDR anyway is that their L/R consistency, taking all the variables above together, sucks big time at least for the samples I've had and on my own head, and it's very audible even when I move about. It's incredibly annoying to hear a bass rumble suddenly move to the right just because I've leant a little bit on the left.
Re the Apple APM... The differences in measurements in the upper mids and low treble on this model are somewhat confusing. You speculated that it could be due to the pads not interfacing correctly with the dummy rigs, I think. But I also wonder if it could be related to the ANC, or some of the headphone's other DSP functions.. And whether turning those features off might produce a different result.
I am very skeptical that Apple's adaptive EQ works above 800-1000Hz (for the APM, for the APP it might be different). For starters you'd think that if this were the case they'd show fairly low positional variation above that range, but that isn't the case :
https://twitter.com/oratory1990/status/1343323936801644544?lang=en
https://www.rtings.com/headphones/1-5/graph#16092/7913
I've also measured them on my head with in concha mics both with the ear canal open, and the ear canal blocked (cf. image below) - so keeping all else as ceteris paribus as I could - and found that the difference between the two shared characteristics not too dissimilar from other headphones (ie small but noticeable differences between 2-4kHz, which trends match what I'm finding when measuring headphones with both blocked ear canal entrance mics and my DIY probe mic - which is open).
The idea that the APM
could perform some form of DSP magic in that range isn't completely far fetched. If I understand that article well (free access) over-ear HPs still operate in pressure conditions below a few kHz and therefore the inwards facing mic could still provide
some information as to what is happening at the DRP in the ear canal gain region :
https://www.aes.org/e-lib/browse.cfm?elib=20605
But I'm not sure we have any evidence, overall, that the APM is much different from other ANC headphones with a feedback circuit.
BTW I'm not the only person whose in-ear mics measurements for the APM found results that weren't quite superbly well aligned in the ear canal gain region with some rigs measurements (which are all over the place for the APM anyway) :
https://www.reddit.com/r/headphones/comments/os73cl/airpods_max_impressions/
The degree of confidence I can have in my own measurements with the various mics below is a complex, multifaceted subject that depends on various factors such as :
- Seatings to seatings repeatability on my own head for each specific headphones, for each part of the FR spectrum
- Sample variation
- Pads brake in / aging / warming up during use, their impact on which part of the spectrum
- The repeatability of the chosen measurement method, whether in terms of absolute values, or relative differences between headphones
- The relevance of the chosen measurement method vs. known characteristics of the human ear (literature) vs. the range of the FR spectrum we’re evaluating.
- And probably quite a few others
.
All of that can be evaluated in various ways such as plotting results against each others but that would make for a very long and boring post with dozens and dozens of squiggly lines of this kind :
Untangling all these nuisance variables from each others is sometimes a bit difficult but often a fairly rational and demonstrable explanation can be found.
Blankets statements such as “I can EQ headphones within 0.3dB of each others at my DRP” are most likely to be wrong. More specific statements such as “I can EQ
on my own head (and not for you) the right channel of my own pairs of HD560S and HD650 within 0.5dB of each others between 70 and 1500Hz, 1.5dB between 1500 and 4000Hz, 1dB between 4000 and 7000Hz, and unknown above that” are more likely to be accurate, because I’m starting to get a pretty good grasp of the aforementioned factors.
But the TLDR is that for realistic seatings of the APM on my head vs headphones such as the H400SE, HD650, HD560S, Hi-X65 (not the K371, this one's a consistency nightmare across the entire range
), the degree of confidence I can have in my own on-head measurements for relative comparisons between headphones is quite a lot tighter in the ear canal gain region than the variance I've seen from dummy head / test rig measurements between the APM and other headphones or the variance we've seen in general between APM measurements.
I can't really explain these variations - even less the ones above 5kHz where the APM's dummy head measurements aren't particularly informative (and an area of the spectrum where my confidence in my own measurements for them drops as I'm getting increased "noise" from the various nuisance variables above, particularly seatings to seatings variation, albeit nowhere near as bad as the K371). So far fiddling with exactly how they sit around my ears and particularly how even / uneven the pad compression is and its degree makes me float the idea that this is something that might be worth looking into as this does modulate the response in the ear canal gain region quite specifically -
at least for me - relative to the response below 1kHz and in the 4-5kHz range. Also, my own anatomy (and I would suspect most of the population actually) differs quite substantially
around my ears from hammerhead-style fixtures (and even most dummy heads).
Instead of repeatedly attacking me on this subject though, why not listen to what some of the folks from Harman have to say on the subject in the above videos, and then give some your impressions on that?