I still think we're not talking about the same thing here. Of course I agree with you that playing a sine sweep all the way from 0 to 20kHz isn't a good listening test, but I,
@Chromatischism , and
@Dazerdoreal were not speaking about that, we were talking about the resonant peak in a very narrow band. You can't seriously be trying to argue that playing a sine wave up and down this range isn't going to reveal to you where this peak is? Look at the graph above, it's at least a good 15dB, it's really obvious. I agree, you're not going to be able to discover the precise centre of the peak, nor the Q factor, by listening, but you can't tell
@Dazerdoreal they're wrong to have found a resonant peak at 9kHz by using a sine sweep.
If you have a pair of reasonably well setup speakers, maybe even headphones, you can apply a fairly strong, narrow peak via EQ, neither too low or high in the spectrum, in a range that measures well, and sounds smooth when manually sweeping through it, to assess on your own whether or not you can recognise where the peak is with that combination of Q / magnitude. You can even ask someone else to do it and try to see whether or not you can successfully locate the peak.
Given the sort of Q / magnitude that most headphones can exhibit in the treble it should be fairly easy peasy to localise quite a few of the "peaks" in question. Now the question rather becomes "is what I heard a "peak", or a normal elevation surrounded by two dips ?", and which Q / magnitude is it - which I don't think as you said is easy to determine by manually sweeping through a part of the spectrum.
I mean, it should be fairly easy to notice this sort of "peak" (which in the case of the Clear MG rather is a "dip" surrounded by bigger "dips") when manually sweeping through that area :
And you can do the test Oratory suggests to perform and see for yourself which one you think works best for you in terms of
locating "peaks".
Another possibly more complicated question is whether or not you
should be hearing certains peaks / dips in the treble or not (I.e such FR would be preferred to you or not), but in any case it's quite logical to suggest that IEMs' resonant peaks are unlikely to be desirable for most people unless they happen to exactly align with where you should have one, given their origin.
I would like to see the insert microphones you use to measure the resonant frequencies of the IEMs you own in your ears. It is specialist equipment which
@Dazerdoreal and the rest of us are not lucky enough to own, so sine sweeps are a compromise to demonstrate roughly where the resonant peaks are for us.
Oratory was responding to a question regarding measuring a pair of over-ears. For these you can use in-ear mics indeed, but above a certain frequency they can become quite misleading in terms of the notion of "peaks" or "dips" depending on their design. Ideally you'd use a probe tube mic near the eardrum. If you use ear canal entrance mics (either blocked or open), you'll need to apply a transfer fonction to get what the measurement at the eardrum would look like, and above 7-8kHz they may already struggle to accurately describe the relative differences between headphones anyway, let alone the absolute values. And if you use anything that's past the canal entrance, the frequency at which errors start to occur is likely to be lower :
https://audiosciencereview.com/foru...ted-harman-oe-curve-at-home.28130/post-975888
It's the best method to go beyond EQing headphones after ear simulator measurements or by ear, but it isn't completely straightforward either.
For in-ears it's a lot more difficult if the pair of headphones are heavily reliant on seal to perform well as the wire or the probe's tube can compromise it, and anything that isn't near the eardrum will introduce errors quite low in the spectrum.
Best implementations either locate a MEMS mic in the bore of the IEM, but then your measurements will only be valid up to several hundred Hz, perhaps 1kHz or so (here what Harman did for their in-ear articles) :
A series of controlled listening tests were conducted to determine the preferred low frequency response of in-ear (IE) headphones. Using a method of adjustment ten trained listeners adjusted the bass level and frequency of a 2nd order low shelving filter applied to a high quality IE headphone...
www.aes.org
Or pass the probe tube through a dedicated hole (occasionally seen in real ear measurements for hearing aids) :
A cruder approach that I've occasionally tried at home, sacrificing an ear tip :
For in-ears with a decent feedback mechanism any leak with that approach should be minimal enough to be corrected by the feedback mechanism, but I'd still be a little hesitant to use it for passive IEMs.
And there's a very good reason he says to use pink noise instead, and that's because it (along with music tracks that also have a similarly smooth broadband spectrum) have been scientifically determined in blind tests to produce the
most discriminating and reliable listener judgements, as I've mentioned
previously.
The format of the test Oratory suggests to perform isn't similar to the one Harman performed at several occasion, the latter's research is unrelated to assessing whether or not Oratory's method is useful or not to locate peaks. *sigh*