A Broad Discussion of Speakers with Major Audio Luminaries

[ROOSKIE]

The compression itself looks rather harmless with this floorstander, which I mentioned in reply to the comment that rather very compact bookshelf designs are affected. They are even more so, naturally. Have stumbled across several very compact active vented speakers with pretty astonishing specs in terms of max SPL and lower cutoff frequency, which did not really sound convincing to me in the lower bass region.

There is no replacement for displacement. You want convincing bass and midbass get big woofers or very high xmax drivers or both.

Any small speakers with astonishing specs for high SPL and low bass are lying about things or you need to be filling a very small room like a closet.

Hoffman's Iron Law -
"Hoffman's Iron Law, in the context of speaker design, states that you can only achieve two out of three desirable characteristics simultaneously: low-frequency extension, small enclosure size, and high efficiency/output (loudness). You can't have all three; you must choose two, and the third will be compromised. " ai

That is exactly the problem. If port tuning freq at modest SPL is not hitting a mode, it might sound fine, but with certain sounds like kickdrums at higher levels it excites a mode for a moment.

I don't think you understand what I was trying to say. Oh well. Limitations of no telepathy allowed here.
The very slight shifting of the port tuning is only slightly changing output and at only very high levels.(@least in terms of the amount of change in the f35) There are much much bigger fish to fry in a listening space.
By the way the port/box/system tuning is changing with SPL, heat levels in the voice coil, ambient temperature and pressure, humidity levels and increased resistance in the crossover and voice coil when high power is sent to the system. If the port is close to a boundary it changes the
tuning. It changes by a couple or even a few hrz all the time for many reasons in almost all speakers even if not demonstrated in Erins test.

Then you go into a room where everything is affecting playback.
Heck even opening windows or doors to the room can change in room response measurements.

 

[jhaider]

There is a drop off in off-axis because the microphone is being used in other angles.

That’s an interesting point - most folks use a 1/2” mic with the NFS, right? There have been a few cases where manual turntable measurements show wider treble radiation than NFS does. That’s probably why.

 

[david moran]

You presume so much about me, and obviously without fully digesting what I post, it hurts.
I suggest you consider more out of the box testing, measuring, and listening.
Klippel NFS spins showing only magnitude response, and small room audio science, are not the be all end all to audio.

>> magnitude response, and small room audio science, are not the be all end all to audio.

pls go on

 

[amirm]

I suggest you consider more out of the box testing, measuring, and listening.

I do more of that in a month than you do in your entire life. And I do so based on sound engineering and science, not my lay intuition.

Klippel NFS spins showing only magnitude response, and small room audio science, are not the be all end all to audio.

More myths being created about NFS. NFS fully characterizes the dispersion of the speaker. And that includes both phase and magnitude. There is nothing more to be had. I don't post all that it measures (because they are not audibly significant) but that doesn't mean it is not measuring such. If you don't know such basics of the system, you would do well to stop making such claims.

So far you haven't shown any research or controlled testing to back what you are saying over and over again. In a highly informed thread like this, you better show up with the receipts or we will walk you out of the thread.

 

[ROOSKIE]

Actually Mr Toole clearly states in his document that listeners are able to listen through the room (or simlar wording). So we are capable of somehow "substract" the room from the speaker sound.

In familiar spaces.
Which is probably why listening in new rooms can be hard and judging speakers in newer spaces kind of pointless at least at first.

I don't know how long it takes to adjust to new rooms, when I moved last it seemed to take a couple days.

I don't guess we subtract things, though of course we might, but rather just come to expect them to be there. Almost missing them when they are gone. Like we expect rooms to have their own sounds and character.

 

[Arindal]

There is no replacement for displacement. You want convincing bass and midbass get big woofers or very high xmax drivers or both.

I disagree with that, and see it as one factor why people don't believe the old folks anymore when it comes to sound quality and gear.

The countering evidence is out there, and everyone with $250 bucks can experience it. You don't need much of displacement for convincing bass, as the true demand for max SPL at lower frequencies is astonishingly low. The bluetooth speaker in question has one single active 3" per channel, presumably with effectively a 2.5" diaphragm used in a pretty broad frequency range.

If you don't need enormous SPL and very low frequencies, I am sure the bass will satisfy most of listeners. It seems hard to explain why people should switch to any reasonably sized passive bookshelf speaker or slim floorstander, if it does not give substantial improvement but rather more booming problems, not to mention a big boy.

Any small speakers with astonishing specs for high SPL and low bass are lying about things

I wholeheartedly agree, and see pretty dangerous implications in that: People might turn their back on specs and measurements completely, if they experience a mismatch between what they are told is sciencifically righteous, and what they are actually hearing. A similar transition took place in the 1980s and 1990s, and I see it as a main reason why hi-fi subjectivism is such a persistent belief.

The very slight shifting of the port tuning is only slightly changing output and at only very high levels.(@least in terms of the amount of change in the f35) There are much much bigger fish to fry in a listening space.

I understood you correctly.

By the way the port/box/system tuning is changing with SPL, heat levels in the voice coil, ambient temperature and pressure, humidity levels and increased resistance in the crossover and voice coil

Agreed. That is why I would consider it more helpful to promote compact active speakers equipped with passive radiators and bass limiters as they avoid many of these problems, instead of just staring at cutoff frequency, linear graphs and THD.

 

[amirm]

And whether there is much in teh way of ABX or other studies that say it has even one half of an iota of relevance?

I have already covered one such paper in my video on the topic of phase and a number of other posts here. Here it is again:
On the Audibility of Midrange Phase Distortion in Audio Systems
Authors: Lipshitz, Stanley P.; Pocock, Mark; Vanderkooy, John

 

[respice finem]

Just a thought about "the flesh side" of human hearing: Any transients /impulses will not reach the brain without first passing the "biomechanical apparatus" of the ears. Which will inevitably add its own limitations and imperfection. The brain can process much, but not things that don't reach it.

I am not qualified to state, we are splitting hairs - but it does cross my mind, sometimes...
Why should we care for what we won't hear?

 

[Suono]

And I do so based on sound engineering and science, not my lay intuition

Even 40-30 years ago, speaker manufacturers didn't understand the importance of dispersion and off-axis coherence. Then some people started creating speakers that did. The same goes for phase and time coherence in this discussion.
Since you're a supporter of "you can hear it or you can't hear it," I read a previous post where the author had participated in an ABX test where some listeners found differences between a linear-phase and a minimum-phase xo. If it were only 2 out of 60, we should investigate further. Since no one cares about audio reproduction research and only thinks about profit, this won't happen. Fortunately, there are people with broader knowledge and insight who create phase- and time-coherent speakers. Someone who values direct sound.

 

[Keith_W]

@amirm or @RickS this discussion about impulse response is valuable but (1) does not belong in this thread and (2) should continue. May I suggest that you split it into another thread?

 

[Newman]

@Keith_W it is a broad discussion of speakers, and how to measure (and how to assess the measured result) is pretty on-point...

 

[czt]

Can somebody specify sample music tracks (maybe/preferably with self ABX logs eg. foobar2000) so I can test with them the the audibility of low frequency excess phase correction or M/T min. phase crossover linearization?

 

[Holmz]

I have already covered one such paper in my video on the topic of phase and a number of other posts here. Here it is again:
On the Audibility of Midrange Phase Distortion in Audio Systems

Authors: Lipshitz, Stanley P.; Pocock, Mark; Vanderkooy, John

I’ve highlighted the other part that sit dead center - between the yellow highlighted sections you chucked in.

So the conclusion is… that it is audible and can be heard with high confidence.
And they had no requirement “back in the day” for phase integrity.
... ok … good for them.

I am not sure when “back in the day” is… But they said that this stuff has been known for over 30 years.
So it is past 30 years now.

I am not going to argue that it is the most important, or super important.
But I am also disinclined to see only an FFT where all the phase info is chucked out as being of “no importance”, and just the FR shown.
Especially when the Klippel itself produces the dang plots, which seems to indicate that they must think that it is important enough to do.

Now I can abide “The Luminaries” saying that it is too much work, or that they would not try to overcome it.
But if they could opine as to what is best in theory, then that would be sorta dandy…

 

[Duke]

I have already covered one such paper in my video on the topic of phase and a number of other posts here. Here it is again:
On the Audibility of Midrange Phase Distortion in Audio Systems
Authors: Lipshitz, Stanley P.; Pocock, Mark; Vanderkooy, John

Thanks for posting this!

I had suspected that the loudspeakers were the part of the signal chain where phase response matters the most, and according to this paper, such seems to be the case.

It seems to me that (paraphrasing) "there is no requirement for phase linearity in loudspeakers" is not necessarily equivalent to "there is no audible benefit from phase linearity in loudspeakers."

I think phase linearity may matter more for narrow-pattern loudspeakers, and/or for loudspeakers in either large or well-treated rooms, than for most loudspeakers in most rooms. Let me explain:

The finding that the "audibility [of phase distortion] is far greater on headphones than on loudspeakers" seems to imply that the in-room reflections degrade our ability to hear (and presumably to appreciate) phase linearity. In a live (and unamplified) sound setting, David Griesinger advocates avoiding reflections within the first 10 milliseconds to preserve the phase relationship of the harmonics above 500 Hz and thereby maximize clarity.

So, perhaps midrange phase linearity matters more for loudspeaker/room combinations which avoid significant reflections within the first 10 milliseconds... ?

Or, to put it another way, for most loudspeakers in most rooms, the early reflections may largely "mask" any benefits from loudspeaker phase linearity.

(Just to be clear, I'm not saying that midrange phase linearity matters a great deal, and it's not something I would prioritize at the expense of frequency response or radiation pattern control, both of which imo matter far more.)

 

[Keith_W]

I think phase linearity may matter more for narrow-pattern loudspeakers, and/or for loudspeakers in either large or well-treated rooms, than for most loudspeakers in most rooms.

The problem is, if you believe that phase is audible, the recommendations for speaker/room setup are opposite to the recommendations if you believe that phase is inaudible. Here are a few:

- Reflections: if phase is audible, we want as few reflections as possible since this corrupts phase information. However, if phase is inaudible, attenuating reflections removes the sense of space and leads to an unpleasant dead-sounding room.
- Directivity: if phase is audible, we want narrow radiating speakers to reduce reflections. However, this leads to narrow sweet spot and other compromises.
- DSP: if phase is audible, we want linear-phase DSP with amplitude and phase corrections to give us the ability to make a linear-phase loudspeaker. However, this requires a certain type of system configuration (i.e. a computer in the signal chain), increasing cost and complexity, while reducing flexibility.
- Listening distance: if phase is audible, then we want the direct sound to predominate over reflected sound to reduce phase corruption, which means shorter listening distance. This means point source speakers, not giant multi-driver MTM's.

These opposite recommendations is why there is so much fighting.

Those older studies were done with minimum-phase loudspeakers, and it is unclear if they set up the room to minimize phase corrupting reflections, shorter listening distance, and so on. They may even have been using test recordings that already had phase information that was corrupted by the recording process. The studies are all pre-Griesinger and his thoughts on proximity. Perhaps new studies were done using linear-phase loudspeakers set up specifically to demonstrate the effect, it wouldn't be thought of as being so subtle and only audible with headphones.

My own personal view is that it is audible, but that's from my own experimentation at home making linear-phase filters and comparing them to minimum-phase. It is not evidence and i'm not expecting to convince anyone. And also if JJ says that it's audible, that's good enough for me. So as a good scientist I have to say: until there is formal evidence to prove phase audibility one way or another, at the moment it is unproven, but even sceptics have to acknowledge that there may be a case. Ohm's Acoustic Law prevails for now.

 

[Holmz]

The problem is, if you believe that phase is audible, the recommendations for speaker/room setup are opposite to the recommendations if you believe that phase is inaudible. Here are a few:

- Reflections: if phase is audible, we want as few reflections as possible since this corrupts phase information. However, if phase is inaudible, attenuating reflections removes the sense of space and leads to an unpleasant dead-sounding room.
- Directivity: if phase is audible, we want narrow radiating speakers to reduce reflections. However, this leads to narrow sweet spot and other compromises.
- DSP: if phase is audible, we want linear-phase DSP with amplitude and phase corrections to give us the ability to make a linear-phase loudspeaker. However, this requires a certain type of system configuration (i.e. a computer in the signal chain), increasing cost and complexity, while reducing flexibility.
- Listening distance: if phase is audible, then we want the direct sound to predominate over reflected sound to reduce phase corruption, which means shorter listening distance. This means point source speakers, not giant multi-driver MTM's.

These opposite recommendations is why there is so much fighting.

Those older studies were done with minimum-phase loudspeakers, and it is unclear if they set up the room to minimize phase corrupting reflections, shorter listening distance, and so on. They may even have been using test recordings that already had phase information that was corrupted by the recording process. The studies are all pre-Griesinger and his thoughts on proximity. Perhaps new studies were done using linear-phase loudspeakers set up specifically to demonstrate the effect, it wouldn't be thought of as being so subtle and only audible with headphones.

My own personal view is that it is audible, but that's from my own experimentation at home making linear-phase filters and comparing them to minimum-phase. It is not evidence and i'm not expecting to convince anyone. And also if JJ says that it's audible, that's good enough for me. So as a good scientist I have to say: until there is formal evidence to prove phase audibility one way or another, at the moment it is unproven, but even sceptics have to acknowledge that there may be a case. Ohm's Acoustic Law prevails for now.

Except @Keith_W the evidence shows that people can separate the direct sound from the reflections when the reflections are delayed some number of milliseconds.
As does a Klippel NFS…

 

[Keith_W]

Except @Keith_W the evidence shows that people can separate the direct sound from the reflections when the reflections are delayed some number of milliseconds.
As does a Klippel NFS…

What does a Klippel NFS have to do with it?? We are talking about human perception and psychoacoustics.

 

[Holmz]

What does a Klippel NFS have to do with it?? We are talking about human perception and psychoacoustics.

Mostly just that it is not magic (PFM), and that the brain math is replicated in math and S/W…
so the reflections have little to do with the direct sound’s phase etc.

 

[Arindal]

if phase is audible, we want as few reflections as possible since this corrupts phase information. However, if phase is inaudible, attenuating reflections removes the sense of space and leads to an unpleasant dead-sounding room.

Don´t see much of connection to audibility of phaseshift here as long as it is interaurally the same (inner group delay). Reflections are anyways delayed this or that way compared to direct sound, either helping or deteriorating imaging is rather a matter of their influence on localization and depth-of-field.

if phase is audible, we want linear-phase DSP with amplitude and phase corrections to give us the ability to make a linear-phase loudspeaker. However, this requires a certain type of system configuration (i.e. a computer in the signal chain), increasing cost and complexity, while reducing flexibility.

Active DSP concepts with sufficient computing capacity allow such correction without much of fuzz or extra cost. If it is useful to implement such only for the sake of time alignment, is another topic. Usually there are other reasons why such x-over design is required anyways.

These opposite recommendations is why there is so much fighting.

Don´t see much reason for dispute, as it is never helpful to optimize one parameter (of questionable relevance) such as phase, heavily compromising others. Designing speakers is a matter of compromise, i.e. balancing parameters. Allowing some diffuse reflections in order to help ambience and envelopment, without deteriorating localization stability, is possible.

 

[Duke]

The problem is, if you believe that phase is audible, the recommendations for speaker/room setup are opposite to the recommendations if you believe that phase is inaudible. Here are a few:

- Reflections: if phase is audible, we want as few reflections as possible since this corrupts phase information. However, if phase is inaudible, attenuating reflections removes the sense of space and leads to an unpleasant dead-sounding room.

It is the EARLIEST reflections which (perceptually) corrupt the phase information. So, late reflections = good (for the reasons you mention); and early reflections = bad (Griesinger, Geddes). Minimizing the early reflections while maximizing the late reflections is easier said than done, but not impossible. Is it "worth pursuing"? Opinions obviously differ.

- Directivity: if phase is audible, we want narrow radiating speakers to reduce reflections. However, this leads to narrow sweet spot and other compromises.

Ime suitable narrow-pattern speakers in a time/intensity trading configuration result in a wider sweet spot (and often more uniform spectral balance throughout the room) than is the case with conventional fairly-wide-pattern speakers.

That being said, there IS a trade-off: We miss out on the expansion of apparent source width which comes from strong early same-side-wall reflections.

- DSP: if phase is audible, we want linear-phase DSP with amplitude and phase corrections to give us the ability to make a linear-phase loudspeaker. However, this requires a certain type of system configuration (i.e. a computer in the signal chain), increasing cost and complexity, while reducing flexibility.

This being your area of expertise (and my area of inexperience), I'll take your word for it!

- Listening distance: if phase is audible, then we want the direct sound to predominate over reflected sound to reduce phase corruption, which means shorter listening distance. This means point source speakers, not giant multi-driver MTM's.

Imo it is the time delay between the direct sound and the strong onset of reflections that we want to pay attention to, though not necessarily prioritize above all else. There is a geometric advantage to shorter listening distances; and there is an acoustic advantage to speakers which maintain radiation pattern control down lower in frequency (which typically implies larger speakers). Ime, depending on the specifics, rather large speakers CAN sound coherent at unexpectedly short listening distances.

My own personal view is that it is audible, but that's from my own experimentation at home making linear-phase filters and comparing them to minimum-phase. It is not evidence and i'm not expecting to convince anyone. And also if JJ says that it's audible, that's good enough for me. So as a good scientist I have to say: until there is formal evidence to prove phase audibility one way or another, at the moment it is unproven, but even sceptics have to acknowledge that there may be a case. Ohm's Acoustic Law prevails for now.

My personal view is somewhat informed by feedback from my beta-testers under blind evaluation conditions, though not ideal blind evaluation conditions because the switching time was too long. Their observations were more about spatial quality than sound quality.