Measure frequency response by ear–the why

Headphone frequency response measurements have been around for ages.

Headphone frequency response measurements suck.

Now, don’t get me wrong, I’m a hard-nosed objectivist when it comes to audio in every other way.  I don’t cringe when DBT experiments show that you can replace coils of speaker wire thick as a snake and worth more than their weight in gold with rusty coat hangers with no difference heard by anybody.  I believe that modern audio performance can be characterized almost in their entirety by frequency response, phase response and nonlinear distortion–and most items in the audio chain perform admirably well on all counts, with the most notable exception being speaker and headphone frequency response.  (Phase and nonlinear distortion are almost more evident in transducers than in any other part of the chain, but have been shown to be controlled to inaudible levels through audio frequencies by many good headphones and speakers, some not even that expensive. (understatement alert!))

Nevertheless, everyone knows that headphone frequency response measurements suck.  Nobody can agree on what coupler, artificial ear or dummy head to measure them with;  after measurement, nobody can agree on the best compensation curve for massaging the graph;  but no matter what choices were made on the above two counts, everybody can agree that EQing the resulting graph to flat does not a good sounding pair of headphones make.  As an experiment, I dare you to try doing so with any one of the headphones in extensive headphone measurement database at innerfidelity that you own.

The problem, as far as I can see, is that we’re still not very good at simulating the mechanics and acoustics of a real human ear using dummy ears or dummy heads.  And probing a real ear is also a non-starter as the acoustics of any probe location other than the eardrum itself is different.  Even if someone were to sacrifice oneself for science and audio and replace one’s eardrums with microphones, said microphones wouldn’t match the acoustic impedance response of one’s eardrums.  Massaging flawed measurement data with compensation curves will not yield accurate results either, as the required compensation curve would change with each model of earphones, its acoustic coupling with the artificial ear being different from that with the real thing, this difference being unique for each model of earphones.

Feel free to correct me on any of the above–I’m not used to “bashing science” myself.

Thus, as an alternative to the above, I hereby propose that we measure frequency response of headphones by ear.  As for how this is even possible or how this can possibly be any good, stay tuned for the next update–“the how”!

As an early spoiler, take a look at the listener feedback on an example of headphone frequency response correction utilizing the result of FR measurements by ear.


About Joe0Bloggs Member of (Joe Bloggs)
This entry was posted in FR measurements by ear and tagged . Bookmark the permalink.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s