The Ultimate Headphone/Audio Thread. Seriously - Page 18

Properly, you should look at many headphones if you want to make any type of sweeping statement like I am, but just for illustration here are the phase responses (I didn't cherry-pick, I swear) for one closed and one (semi-)open headphone:

+ Show Spoiler [graphs] +
Beyerdynamic DT 770 (closed, top), DT 880 (open, bottom) — phase response is the top chart in each, from here

A very sharp change in phase in the treble like that is maybe audibly an issue, could affect sense of imaging. This means that certain sounds containing different frequencies are smeared in time. Note that you see there roughly a phase difference of about 120 degrees (2/3 pi radians) over the span of about 500 Hz (1000pi radians / s), at around the 4 kHz mark. So a group delay around there is maybe about 2/3 pi / 1000pi = 0.00067 seconds = 0.67 ms. Maybe an issue? Granted, wikipedia cites an old paper from '78 saying that group delay under about 1.5 ms in that frequency range is under the threshold of audibility. *shrug* Maybe it's not the phase response that's the issue. Maybe just the frequency magnitude issues, or just the air being sealed feels wrong, or...?

Anecdotally a large number of people who have tried both seem to say they hear positioning better with open headphones. That said, I'd be the first to warn against "common knowledge" and anecdotes with regards to audio, which are often dead wrong.

I'm sleepy so somebody correct me if I goofed somewhere.

Reverb, if it exists, should be evidenced in cumulative spectral decay (CSD) aka waterfall plots. That is, input a brief impulse (which contains all frequencies evenly) and graph the FFTs of what happens next. That shows you the magnitude of each frequency as it changes over time—hopefully to zero, and quickly, unless there are some kind of nonlinear feature like resonance at a particular frequency. In theory these should be shorter with higher electrical damping factor. Closed headphones often have more issues here, but many open cans do as well. That said, generally the decay for most headphones is relatively fast—over in 3-5ms or so down to 30-40 dB below, at least outside the bass. e.g. see here, here, here (note that the latter two sources do more smoothing than the first; some could argue procedural inaccuracies with all their setups)

So maybe that's a bigger deal than the phase issues.

Certainly one of the largest effects should come from elevated bass masking other frequencies.

Clarity mostly comes from an even mids / treble frequency response, mostly, so no unnatural dips or especially spikes that mask other frequencies. Maybe other factors too? It's hard to say for sure what corresponds to what people perceive, especially because perception is such a moving target.