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

[Holmz]

Out of curiosity, what kind of corrections are required on the DSP processing end to get the orderly frequency and impulse response? Seems like it could take quite a bit to hammer out the messiness and all of the room modes in both the frequency and time domain, but I guess if you have capable speakers and plenty of power on tap its an academic discussion.

If we separate the room from the speakers, then ideally one may want the impulse response to be ideal.
Usually that is direct path and no time accounting for wall bounces.

For instance a piano in a dead room and a piano in a bright room both sound like pianos… Just one sounds like it is in a dead room, and the other a bright room.

So that is often FIR based for impulse response and maybe Dirac Live.

Whether you treat the room, or do some room EQ, then that is only amplitude based cutting of peaks.

Ideally the speaker is levelled to be flat in the direct path, and one ignores the bounces.

 

[Cars-N-Cans]

If we separate the room from the speakers, then ideally one may want the impulse response to be ideal.
Usually that is direct path and no time accounting for wall bounces.

For the speakers themselves, yes. I agree. But here the measurements provided are the steady-state in-room responses. Usually room modes can put substantial nulls in the response, so getting the thing to be ruler flat like that may necessitate adding quite a lot of corrections which could come at the expense of things like reduced headroom and increased distortion from stressing the transducers. As can be seen from here, there the phase is essentially near zero degrees throughout the bass region: https://audiosciencereview.com/foru...on-shift-matter-in-audio-no.24026/post-809585

That's at least somewhat eyebrow raising given how much phase shift the room can impose. While countering this with DSP corrections will result in a microphone whos response is simply summed over some measurement period to show no phase shift, I would argue that all this really does is impose the inverse of the phase "error" onto another part of the response, such as the direct sound. Given how messy rooms are acoustically, I would think its somewhat of a moot point to worry about phase outside of how the overall sound sums together. While I don't know if he has any room treatments, generally as we know there are reflections and the room modes take time to die away, so in those terms and in the vein of the thread topic, does it even make sense to think about overall phase response anymore? I would think all you can do is ensure the speakers are well behaved and that there are appropriate corrections to ensure the in-room response is relatively close to what you want it to be. I'm not saying the corrections he shows don't work, but it does run counter to conventional wisdom.

 

[Holmz]

The point (I was intending on making) was:

1. Correct the impulse response as a direct measurement (likely with a FIR).
2. Correct the “in room” part maybe <200Hz or 400Hz on top of that, either also as a FIR, or with IIR/Bi-Quads.

Really, the low stuff, with nulls in the 30-100Hz is likely better with a separate sub… and some trimming (knocking down) of peaks.

 

[Karmacoma]

Hi,
A lot of knowledgeable people in this old thread, and everybody seems to agree phase linearity/nonlinearity is inaudible except in unpractical conditions (if i understand well).
Just bought recently some 2-way monitors, the manufacturer claims they have a phase linearity of 45°, and seems pretty proud about it, so what's his point? Does it makes room correction, subwoofer integration or pairing with other speakers of the brand tuned similarly easier or something? What's the use of such a feature?
Thank you for your answers

 

[Tranquility Bass]

Hi,
A lot of knowledgeable people in this old thread, and everybody seems to agree phase linearity/nonlinearity is inaudible except in unpractical conditions (if i understand well).
Just bought recently some 2-way monitors, the manufacturer claims they have a phase linearity of 45°, and seems pretty proud about it, so what's his point? Does it makes room correction, subwoofer integration or pairing with other speakers of the brand tuned similarly easier or something? What's the use of such a feature?
Thank you for your answers

Would you accept poor phase and impulse response in your amplifiers, DAC's, preamps etc or do you tolerate it in your speakers just because it is hard to get right ?

 

[Mnyb]

Would you accept poor phase and impulse response in your amplifiers, DAC's, preamps etc or do you tolerate it in your speakers just because it is hard to get right ?

What I seen is that the compromises needed when trying to do this in passive speakers are sacrificing other real performance issues and creates additional problems that was not there before ?
Extreme demands of drivers due to only 6db filters come to mind ? Directivity issues due to interference between drivers due to the 6db filters ? And more .
I think the “sound” of these speakers are more those design decisions than the achievement in better impulse response.

And if you have more than one driver or the driver has any size other than a piont compared to the frequency it reproduce you have problems.
You may achieve impulse correctness in one listening place when you are in the rigth position versus the speaker and the drivers , but nowhere else ?

The sound from the speaker in any other angle than the perfect listening spot is not phase correct and neither are othe reflected sound even before the room messes with it .

With the use of active speakers with digital filters and coaxial drivers the cost benefit equation may have changed it might be worthwhile to try , my KEF LSX has phase correction, soe does the ls60 .
I wonder if Genelec are doing this ?

 

[AmadeusMozart]

Would you accept poor phase and impulse response in your amplifiers, DAC's, preamps etc or do you tolerate it in your speakers just because it is hard to get right ?

My view is different from the established view. I have the impression that imaging is definitely better when both drivers are in phase and when there is less phase shift over the audible frequency band. I've designed a two way speaker with a 2.3Khz crossover where they both are in phase and over the audible frequency band there is less than +/- 15 degrees phase shift. It took a long time to find drivers that would seamleslly work together and then find the best crossover point. Crossover is not straight forward either. These speakers have the best imaging I've experienced in the past 50+ years (and I've had some expensive ones).

I know there are many who'll argue the other way and that's fine by me, if they are happy with what they've got that's OK too.

Peace to all.

 

[Karmacoma]

Would you accept poor phase and impulse response in your amplifiers, DAC's, preamps etc or do you tolerate it in your speakers just because it is hard to get right ?

Rhetoric? Only works with people who really know what they're talking about, i don't ahah. To answer your question, as long as I can't hear it and there's no practical consequence i can tolerate anything. If i understand the conclusions of this discussion well, i won't hear it, so why did they put some engineering resources into this? Please. Casually.

 

[Tranquility Bass]

Rhetoric? Only works with people who really know what they're talking about, i don't ahah. To answer your question, as long as I can't hear it and there's no practical consequence i can tolerate anything. If i understand the conclusions of this discussion well, i won't hear it, so why did they put some engineering resources into this? Please. Casually.

Maybe you're not building the speaker right to hear a difference ??

Loudspeaker designer John Dunlavy: By the Numbers...

One of the characteristic traits, I have found, that defines the loudspeaker designer is that they are loners—they seem to avoid one another's company as if on purpose. But if ever you sit down with a designer, all you need to do to open him up is to ask him what he feels to be important in...

www.stereophile.com

Atkinson: To judge from the SC-VI, it seems that you regard getting the outputs of the drive-units in your speakers to arrive at the listener's ears at exactly the same time as being very important. Other than the VI's subwoofers, all its drive-units are recessed into the front baffle. How important is that to the loudspeaker's sonic integrity?

Dunlavy: We participate regularly in the recording of our symphony orchestra here in Colorado Springs. We've also recorded instruments like violins and cellos and timpani in our big anechoic chamber using instrumentation-quality microphones and equipment. And we find that in order to reproduce those sounds with a level of accuracy such that you can not literally hear any difference between the live and the recorded sound, you have to have a speaker that exhibits almost perfect impulse and step responses. The only way to do that is to time-align the drivers very, very accurately, usually within a matter of a few microseconds, then use a minimum-phase, first-order crossover network and get everything right. And you have to have an on-axis response of better—well better—than ±2dB.

 

[Karmacoma]

My view is different from the established view. I have the impression that imaging is definitely better when both drivers are in phase and when there is less phase shift over the audible frequency band. I've designed a two way speaker with a 2.3Khz crossover where they both are in phase and over the audible frequency band there is less than +/- 15 degrees phase shift. It took a long time to find drivers that would seamleslly work together and then find the best crossover point. Crossover is not straight forward either. These speakers have the best imaging I've experienced in the past 50+ years (and I've had some expensive ones).

I know there are many who'll argue the other way and that's fine by me, if they are happy with what they've got that's OK too.

Peace to all.

So, your answer would be ''imaging'' but it doesn't seem to be obvious enough that it's an established/demonstrated fact.
They did it and advertise for it, why do they?

 

[Karmacoma]

Maybe you're not building the speaker right to hear a difference ??

Loudspeaker designer John Dunlavy: By the Numbers...

One of the characteristic traits, I have found, that defines the loudspeaker designer is that they are loners—they seem to avoid one another's company as if on purpose. But if ever you sit down with a designer, all you need to do to open him up is to ask him what he feels to be important in...

www.stereophile.com

What? Oh no please read my first question, I'm not into this, i just bought some Neumann kh 150 monitors, they advertise phase linearity, in this thread people seem to say it's not audible, my question is, is there any practical reason for them to advertise it. Thank you.

 

[Karmacoma]

What I seen is that the compromises needed when trying to do this in passive speakers are sacrificing other real performance issues and creates additional problems that was not there before ?
Extreme demands of drivers due to only 6db filters come to mind ? Directivity issues due to interference between drivers due to the 6db filters ? And more .
I think the “sound” of these speakers are more those design decisions than the achievement in better impulse response.

And if you have more than one driver or the driver has any size other than a piont compared to the frequency it reproduce you have problems.
You may achieve impulse correctness in one listening place when you are in the rigth position versus the speaker and the drivers , but nowhere else ?

The sound from the speaker in any other angle than the perfect listening spot is not phase correct and neither are othe reflected sound even before the room messes with it .

With the use of active speakers with digital filters and coaxial drivers the cost benefit equation may have changed it might be worthwhile to try , my KEF LSX has phase correction, soe does the ls60 .
I wonder if Genelec are doing this ?

You seem to know a lot. Can you answer post #524 please? Even if it's dumb or obvious. Thank you.

 

[KSTR]

Hi,
A lot of knowledgeable people in this old thread, and everybody seems to agree phase linearity/nonlinearity is inaudible except in unpractical conditions (if i understand well).
Just bought recently some 2-way monitors, the manufacturer claims they have a phase linearity of 45°, and seems pretty proud about it, so what's his point? Does it makes room correction, subwoofer integration or pairing with other speakers of the brand tuned similarly easier or something? What's the use of such a feature?
Thank you for your answers

Phase distortion can be audible in very normal practical circumstances for many people but not all people, there a learning process involved... once you know what symptoms to listen for most people can hear it. A typical example is subwoofer integration, the strong phase rotation at low frequency (80Hz or so) is quite audible when compared to a linear-phase variant (without any other changes, of course).

As for your question, it makes room correction a bit easier, but it does not help subwoofer integration much as this depends on low frequency roll-off rather than the crossover characteristics of the main speaker.
Pairing speakers is definitely made easier, a linear phase speaker can be paired with any other linear phase speaker with no phase issues. A typical example would be in HT setup where you must use speaker with equal phase response otherwise the soundstage will have some weird characteristics. A center speaker sure must have same phase response than L and R speakers.

 

[ernestcarl]

Hi,
A lot of knowledgeable people in this old thread, and everybody seems to agree phase linearity/nonlinearity is inaudible except in unpractical conditions (if i understand well).
Just bought recently some 2-way monitors, the manufacturer claims they have a phase linearity of 45°, and seems pretty proud about it, so what's his point? Does it makes room correction, subwoofer integration or pairing with other speakers of the brand tuned similarly easier or something? What's the use of such a feature?
Thank you for your answers

It's just another spec...

Out of the box you want the phase between your speakers to match. Most already do. "Linear phase" design is only applicable with a few certain DSP capable speakers/systems and is entirely optional, plus may only be audible in some conditions. If combining multiple speakers, the phases need to be closely matched where they crossover to avoid cancellation/lobing -- which is why you can't just "willy-nilly" mix different speakers together (e.g. different LCR speaker brands/models). Performance of speakers and subs in-room may skew the frequency magnitude and phase response -- some of that is "correctible" or can be compensated for... Unfortunately, a lot of this will only become more clear when you yourself are designing/integrating speakers or multi-channel setups with different phase profiles manually. There are a lot of design considerations to think about... If you want to play around with the phase response equalization and experience for yourself how (in)audible it is in your own system, go ahead and download rePhase and use a convolver in your playback chain.

 

[Thomas_A]

Maybe you're not building the speaker right to hear a difference ??

Loudspeaker designer John Dunlavy: By the Numbers...

One of the characteristic traits, I have found, that defines the loudspeaker designer is that they are loners—they seem to avoid one another's company as if on purpose. But if ever you sit down with a designer, all you need to do to open him up is to ask him what he feels to be important in...

www.stereophile.com

The Ino Audio speakers are like Dunlavy phase coherent at the crossover. However the constructor Ingvar Öhman clearly says that the audibility vs a traditional good crossover at eg 3 kHz is nil. Phase distortion in the bass region is audible to some however, as demonstrated in this thread.

 

[Mnyb]

You seem to know a lot. Can you answer post #524 please? Even if it's dumb or obvious. Thank you.

Sorry I don’t think I know enough? What are those speakers

 

[Mnyb]

Also in electronics you can achieve better phase without quite as much sacrifices in other areas , in some cases it comes along for the ride like in a hirez DAC .

 

[j_j]

Also in electronics you can achieve better phase without quite as much sacrifices in other areas , in some cases it comes along for the ride like in a hirez DAC .

Why would "hirez" be necessary? Any oversampled DAC that uses a front end symmetric FIR (or even a moderately apodized one) should be fine phase-wise.

 

[Mnyb]

Why would "hirez" be necessary? Any oversampled DAC that uses a front end symmetric FIR (or even a moderately apodized one) should be fine phase-wise.

You are ofcourse correct , i was thinking it would be even easier with very wide bandwidth .

 

[NTK]

Hi,
A lot of knowledgeable people in this old thread, and everybody seems to agree phase linearity/nonlinearity is inaudible except in unpractical conditions (if i understand well).
Just bought recently some 2-way monitors, the manufacturer claims they have a phase linearity of 45°, and seems pretty proud about it, so what's his point? Does it makes room correction, subwoofer integration or pairing with other speakers of the brand tuned similarly easier or something? What's the use of such a feature?
Thank you for your answers

As Amir has said in the title of this thread, our ears are not at all sensitive to "phase distortion". Below is an example of phase distortion I borrowed from miniDSP. The top graph shows the waveform of a 2 kHz square wave reproduced by an uncorrected speaker. The bottom one is the phase corrected version. The phase corrected version definitely looks "more right" to the eyes, but our ears usually can't differentiate between the two.

When a speaker such as the KH-150 already possess the necessary DSP hardware to perform phase correction, then there is really no reason not to do it, as it is "free" (as in it can be done with no extra cost). And Marketing will do what Marketing does.