Does Phase Distortion/Shift Matter in Audio? (no*)

[pma]

Examples wanted - OK. This is my favourite example. The two signals below have identical power spectrum (and rms level).



They differ in phase.

However, for an experienced listener it is easy to tell the sound difference:

foo_abx 2.1 report
foobar2000 v2.0
2024-10-31 10:40:30

File A: square_synthesis_inphase.wav
SHA1: cec3a421fcf9bef67e74eef93e8fb81214c8e7d7
File B: square_synthesis_outphase.wav
SHA1: cbe2d8726400680b07414f4533da86ca60f81f67

Output:
WASAPI (shared) : Speaker (USB Sound Blaster HD)
Crossfading: NO

10:40:30 : Test started.
10:41:18 : 01/01
10:41:34 : 02/02
10:41:47 : 03/03
10:41:59 : 04/04
10:42:14 : 05/05
10:42:24 : 06/06
10:42:34 : 07/07
10:42:42 : 08/08
10:42:52 : 09/09
10:43:02 : 10/10
10:43:11 : 11/11
10:43:21 : 12/12
10:43:31 : 13/13
10:43:38 : 14/14
10:43:51 : 15/15
10:44:01 : 16/16
10:44:01 : Test finished.

 ----------
Total: 16/16
p-value: 0 (0%)

 -- signature --
64599d6e0d6df791ef1e90ba7f69d7515cb0f662

The files available for testing are attached below in the in_out_phase.zip file. The 2nd zip file contains both files under test in a row in just one file. The change of perceived pitch can be heard when 1st file transits to the 2nd one, though both signals have the same amplitude spectrum and differ only in phase of the 3rd harmonic component.

 

[Holmz]

Most of the conventional multi-way speakers. The more ways and the lower the crossover frequencies the more prominent are the low-frequency effects of phase distortion (but there is impact at higher frequencies as well).

Extreme example would be a large 3-way with subs, so you'd have XO points at 80Hz, 300Hz, 3kHz ballpark. Listen to these as is vs with compensated excess phase and you'll have no more questions. LF-timbre is affected and sometimes one can hear the difference of the bass delay.

The phase distortion of small 2-way with a single 3kHz XO is quite benign in comparison but it may still be audibly when it is removed, most often it is the rendering of the soundstage that is different, phantom sources and diffuse reverb tails having different dimensions/sharpness, stuff like that.

It is the orders, not the ways.

 

[KSTR]

It is the orders, not the ways.

Well, both. More ways normally means a lower XO point like in the 300Hz region (and another below when using a sub-way). And steeper slopes of course create more phase rotation.

 

[Holmz]

Well, both. More ways normally means a lower XO point like in the 300Hz region (and another below when using a sub-way). And steeper slopes of course create more phase rotation.

Unless we are talking about group delay, we can pretty much just use “step function” response to see things pretty clearly.
But the impulse response is also pretty good at conveying what is happening (or had happened).

 

[gnarly]

Unless we are talking about group delay, we can pretty much just use “step function” response to see things pretty clearly.
But the impulse response is also pretty good at conveying what is happening (or had happened).

It doesn't matter what we look at....phase, impulse, or step.
Every one of those shows that it's both the number of ways, and the orders between the ways, that counts.

All views simply show the effects of phase rotations inherent in IIR / minimum phase crossovers.
The more the ways, the more the total phase rotation.
The higher the order of crossovers, the more the total rotation.

Step just gets more and more screwed with either increasing (in the IIR world)
As does does impulse, and group delay for whatever GD really is.

 

[375HP2482]

The current state of the art in mid fi and hi fi usually sends a pristine signal to the speakers, then the speaker reproduces it with its own coloring, then the room demolishes all that fantastic clean signal and then it hits your eardrums and the brain takes over and does whatever it wants to the sound you perceive (all kinds of biases).

A good post as relates to how domestic spaces corrupt sound purity -- in many cases, more than loudspeakers do. My old boss referred to this arrangement as a "phase chaotic" environment.

The current start of the art in mid fi and hi fi these days can be attained with a $30 earb -- I mean, IEM. Voila! -- no room corruption. Which says a lot for all of those younguns toting earbuds with their phones, blissfully unaware of all the vitriol many audio enthusiasts have been aiming at them.

 

[j_j]

The current start of the art in mid fi and hi fi these days can be attained with a $30 earb -- I mean, IEM. Voila! -- no room corruption. Which says a lot for all of those younguns toting earbuds with their phones, blissfully unaware of all the vitriol many audio enthusiasts have been aiming at them.

Yeah,no, I'm afraid bypassing some atmospheric transmission (no, the space between the diaphram and the eardrum does not suffice), and the entire HRTF of the listener really doesn't replace free-field reproduction. If you have extremely low latency head position correction and a rich enough information about the intended soundfield it can almost be ok. Almost.

But it is cleaner than most rooms.

 

[375HP2482]

No, IEM's don't replace free-field reproduction. I'll take "cleaner than most rooms." And they're better than "almost OK."

Where possible, well-executed reproduction outdoors can yield the best of both worlds.

 

[j_j]

Where possible, well-executed reproduction outdoors can yield the best of both worlds.

It's even possible indoors, but you must treat the room appropriately.