A Broad Discussion of Speakers with Major Audio Luminaries

[MaxwellsEq]

That brings me back to ask, if the self-interference of those sound packets flying around is a low frequency phenomenon?

What are sound packets? Is this a quantum characteristic?

 

[Heinrich]

What are sound packets? Is this a quantum characteristic?

A term lend from intuitive imagination, like a burst, which I hope is well defined, only with a broader, less regular spectrum.

 

[Floyd Toole]

Adding more energy to a room that already stores too much is not generally a great way to start your day. So you can't get completely away from the room you've got without some way to get energy out of it as well as into it.

It seems that you haven't read my books or the Welti and Welti & Devantier papers. The multiple subwoofer scheme began with the notion of using destructive acoustical interference to "cancel" - i.e. not provide energy to - specific strong room resonances. Passive combinations of two or four subs in specific arrangements in rectangular rooms work well, but with signal processing, as in Sound Field Management, it works even better. The total energy radiated by the multiple subs in such arrangements can be less than that from a single, or multiple non-optimally arranged subs. At long wavelengths in small rooms pressures add coherently. It is not uncommon to achieve 10 dB higher sound levels (10x power) from 4 subs (4x power). The multiple subs can be smaller. Optimized multi-sub arrangements can increase system efficiency, sound absorption always reduces it. Several examples are shown in the publications mentioned. Seat-to-seat variations are substantially reduced, energetic resonances are attenuated, sound quality is greatly improved as measured in both amplitude and time domains - all with no bass traps. That is a good start from the sound quality perspective.

If that kind of easily audible improvement can be sustained with the added subtlety of "spatial bass", so much the better. However, some evidence and simple instincts say that room reverberation will degrade spatial bass, so JJ is probably right about the need for low-frequency absorbers. Most folks find these to be both expensive and insulting to any sense of room decor, so the audience is likely to be small - custom listening rooms/home theatres. Finally, as spatial bass is a large venue phenomenon, really only classical music is likely to recognizably benefit, and only if the recordings capture the spatial information. With multi-microphone recordings being widespread this is another challenge because the reference is what is heard by a single pair of ears somewhere in a concert hall.

Binaural (dummy head) recordings capture it all automatically, and head-tracked headphone playback is very impressive. Impressive, but antisocial. Achieving it with loudspeakers in small rooms, most of which have significant reflections, is the challenge.

We await the applied science that can make it all happen. Go JJ!

 

[j_j]

It seems that you haven't read my books or the Welti and Welti & Devantier papers.

It seems that I don't AGREE with some of the work. Different.

 

[MKR]

@j_j @Floyd Toole I wonder what your thoughts are on cardioid loudspeaker designs, especially low frequency capable cardioid (which I define as down to ~40Hz, then lower freq covered by monopole bass)? Especially in my experience with cardioid designs, I have had zero need for any room treatment whatsoever, including LF absorbers (bass traps). This experience has been in multiple (different dimensions /shapes) standard US home living spaces (wood framing, sheet rock, carpet, furniture, etc), with same superb results.

 

[j_j]

@j_j @Floyd Toole I wonder what your thoughts are on cardioid loudspeaker designs, especially low frequency capable cardioid (which I define as down to ~40Hz, then lower freq covered by monopole bass)? Especially in my experience with cardioid designs, I have had zero need for any room treatment whatsoever, including LF absorbers (bass traps). This experience has been in multiple (different dimensions /shapes) standard US home living spaces (wood framing, sheet rock, carpet, furniture, etc), with same superb results.

It creates what I can only describe as a complex situation that might work well for some situations, but that might also work very poorly for others. The idea of a controlled dipole is far from new, of course, but how to apply it can sometimes be "interesting".

 

[MKR]

It creates what I can only describe as a complex situation that might work well for some situations, but that might also work very poorly for others. The idea of a controlled dipole is far from new, of course, but how to apply it can sometimes be "interesting".

Thank you for your comments. In what situation would a cardioid be a poor choice? Thus far in every case I have experienced cardioid it only resulted in positive results.

 

[j_j]

Thank you for your comments. In what situation would a cardioid be a poor choice? Thus far in every case I have experienced cardioid it only resulted in positive results.

It's utterly site-specific.

 

[Thomas Lund]

This does not mean that stereo "fails", it means that it is simply not the best format.

At these crossroads between stereo, 3D and binaural formats, it is a good time to take a step back, and consider what to actually require from sound reproduction. The most reasonable would be to begin with perception, rather than e.g. tradition, acoustics or engineering; and to accept the different goals of monitoring and recreational listening.

In humans of any age, Auditory Envelopment (AE) is strong sensory evidence of space. Due to a decades-long research oversight, we largely missed that foundation, and based the sense of space instead on secondary clues, such as reverb time or discrete reflections. As a consequence, inadequate techniques have been developed in recording, distribution and monitoring; e.g. compact microphone arrays, lossy codecs and mono bass "management".

In monitoring, the goal is to convey recordings faithfully, so the temptation for taking shortcuts is limited. At home, it is more about joy, with pragmatic reasons for lowering the bar. However, there is hardly anything audio more enjoyable and comforting than well-tempered AE.

On a forum like this, where readers put up with floorstanders, to get a glimpse of AE latent in recordings, some might also be interested in uncompromised 3D reproduction. Try listening under controlled conditions to fine 3D recordings and full-range channels, rather than pale, electrically summed LF room excitation. Illustration: Technical reasons for AE to vanish, from recording to reproduction.

 

[Floyd Toole]

@j_j @Floyd Toole I wonder what your thoughts are on cardioid loudspeaker designs, especially low frequency capable cardioid (which I define as down to ~40Hz, then lower freq covered by monopole bass)? Especially in my experience with cardioid designs, I have had zero need for any room treatment whatsoever, including LF absorbers (bass traps). This experience has been in multiple (different dimensions /shapes) standard US home living spaces (wood framing, sheet rock, carpet, furniture, etc), with same superb results.

I just noticed that you asked me specifically. Sorry for the delay. JJ is correct that it is not a simple situation.

Obviously, from these discussions there are two separable concerns with sound reproduction in small rooms: (1) corruption of the timbre by resonances, which is different in different seats, and (2) low-frequency perceptual envelopment. I empathize with both concerns, but my research has focused on getting the timbre problem neutralized. In that respect the goal is a flattish, smoothish steady-state frequency response, indicating some control over audible resonances and monotonous bass booms, with all notes present at the appropriate loudness.

Arriving at this goal requires control of the amplitudes of the room modes involved, which begins by understanding how energy is coupled from the loudspeakers to the standing wave structure - and all room setups are different. Put simply, monopole subs are pressure sources and their coupling depends on location relative to the pressure distributions of standing waves, coupling maximally in high pressure regions which is why most end up close to walls. Dipole subs are "velocity" sources, and they couple energy maximally in the nulls, the low pressure regions. More than that, the coupling varies with the angular orientation within the null - rotating the loudspeaker varies the coupling. Cardioid subs are a combination, so they also need to be located with consideration both of location and anglular orientation. Comparing the performance of different designs when located in the same positions in a given room is not helpful in arriving at generalizations about "ideal" woofer/subwoofer designs. For each specific room there will be different optimal locations and arrangements of multiples for each design. Adjustments in level of the individual subs in an array alter the acoustical interference and thereby the level of specific affected modes. If delays are available more modal control can be exercised, and still more with EQ. There will also be differences in system efficiency and cost. It is not simple.

 

[dominikz]

I just noticed that you asked me specifically. Sorry for the delay. JJ is correct that it is not a simple situation.

Obviously, from these discussions there are two separable concerns with sound reproduction in small rooms: (1) corruption of the timbre by resonances, which is different in different seats, and (2) low-frequency perceptual envelopment. I empathize with both concerns, but my research has focused on getting the timbre problem neutralized. In that respect the goal is a flattish, smoothish steady-state frequency response, indicating some control over audible resonances and monotonous bass booms, with all notes present at the appropriate loudness.

Arriving at this goal requires control of the amplitudes of the room modes involved, which begins by understanding how energy is coupled from the loudspeakers to the standing wave structure - and all room setups are different. Put simply, monopole subs are pressure sources and their coupling depends on location relative to the pressure distributions of standing waves, coupling maximally in high pressure regions which is why most end up close to walls. Dipole subs are "velocity" sources, and they couple energy maximally in the nulls, the low pressure regions. More than that, the coupling varies with the angular orientation within the null - rotating the loudspeaker varies the coupling. Cardioid subs are a combination, so they also need to be located with consideration both of location and anglular orientation. Comparing the performance of different designs when located in the same positions in a given room is not helpful in arriving at generalizations about "ideal" woofer/subwoofer designs. For each specific room there will be different optimal locations and arrangements of multiples for each design. Adjustments in level of the individual subs in an array alter the acoustical interference and thereby the level of specific affected modes. If delays are available more modal control can be exercised, and still more with EQ. There will also be differences in system efficiency and cost. It is not simple.

Thanks for sharing your thoughts, very interesting!

To be honest, in general I have a really hard time understanding the purpose of cardioid or dipole subwoofer designs. It seems to me that placement and multiple subwoofer optimization problems are significantly simpler with the typical monopole subwoofers, and at subwoofer frequencies SBIR should not be a concern with close-to-wall placement. Not to mention that monopole subs are more widely available and cheaper...

I do however see a benefit of "cardioid bass" in "main" loudspeakers, i.e. loudspeakers with cardioid radiation down to about 100Hz.
Such loudspeakers can be conveniently placed close to the wall behind them and not get the typical SBIR null in the 150-250Hz range (which otherwise occurs with such placement).

@Floyd Toole Do you have some thoughts on these kinds of designs as one of the possible solutions for SBIR nulls, and any experiences you could share? Thanks!

 

[Floyd Toole]

Thanks for sharing your thoughts, very interesting!

To be honest, in general I have a really hard time understanding the purpose of cardioid or dipole subwoofer designs. It seems to me that placement and multiple subwoofer optimization problems are significantly simpler with the typical monopole subwoofers, and at subwoofer frequencies SBIR should not be a concern with close-to-wall placement. Not to mention that monopole subs are more widely available and cheaper...

I do however see a benefit of "cardioid bass" in "main" loudspeakers, i.e. loudspeakers with cardioid radiation down to about 100Hz.
Such loudspeakers can be conveniently placed close to the wall behind them and not get the typical SBIR null in the 150-250Hz range (which otherwise occurs with such placement).

@Floyd Toole Do you have some thoughts on these kinds of designs as one of the possible solutions for SBIR nulls, and any experiences you could share? Thanks!

I have no experiences with these designs, but reducing the energy radiated towards adjacent boundaries logically is a good idea.

 

[dominikz]

I have no experiences with these designs, but reducing the energy radiated towards adjacent boundaries logically is a good idea.

Thanks for responding!

May I also take the opportunity to say that I find your book remarkable, and that I'm looking forward to the 4th edition.

Thanks a lot for the work you do - many of us appreciate it very much!

 

[Floyd Toole]

Thanks for responding!

May I also take the opportunity to say that I find your book remarkable, and that I'm looking forward to the 4th edition.

Thanks a lot for the work you do - many of us appreciate it very much!

Thank you. But, interestingly, you forum members keep me engaged now that I am retired - and occasionally challenged. In your discussions I come to recognize real world problems and see new ways to explain things to members obviously struggling with limited knowledge. Some of this finds its way into the next edition. Also, the 4th edition will have a better organized companion website, allowing me and my co-authors, Sean Olive and Todd Welti, to expand the content beyond what 500 black and white printed pages allow. Some of it might seem familiar to forum members, as it could be a variation of one of my responses. Informed technical discussions are two way streets. Audio has always had a good share of mythology, wishful thinking, and strongly expressed personal opinions formed under biased circumstances. It fills forums, and many manufacturers and some "experts" keep it supplied with new material. Upcoming generations of audiophiles need to be able to find a rational, science based, alternative. ASR is a positive influence, and the trustworthy loudspeaker measurements are a big part of it. Well done, Amir.

 

[audiofooled]

However, there is hardly anything audio more enjoyable and comforting than well-tempered AE.

Kind of like "seeing the forest for the trees", isn't it? The sensation is not at all subtle.

Problem is that, be it recorded or produced, it's usually latent in VLF interference patterns throughout the signals of wider bandwidth and long duration, on a macro-scale, combined from multitrack mixes. In the midst there would be signals with sharp envelope, mostly mono, that would attract immediate attention.

 

[Thomas Lund]

Kind of like "seeing the forest for the trees", isn't it? The sensation is not at all subtle.

Right. However, AE latent in a production, or AE manifest in reproduction, are not easily quantifiable. Even a systematic vocabulary is missing, after 40 years of academia mostly only considering reproduction timbre.

As much as I appreciate Floyd and Harman’s excellent research on loudspeaker directivity and timbre, it is about time to also consider the hard problem of reproduction: Spatial qualities of the listening room+system, from an AE perspective. AE is even the primary perceptual and emotion-inducing potential 3D sound has to offer over stereo, regardless if AE is naturally recorded or constructed.

 

[Floyd Toole]

Right. However, AE latent in a production, or AE manifest in reproduction, are not easily quantifiable. Even a systematic vocabulary is missing, after 40 years of academia mostly only considering reproduction timbre.

As much as I appreciate Floyd and Harman’s excellent research on loudspeaker directivity and timbre, it is about time to also consider the hard problem of reproduction: Spatial qualities of the listening room+system, from an AE perspective. AE is even the primary perceptual and emotion-inducing potential 3D sound has to offer over stereo, regardless if AE is naturally recorded or constructed.

Hi Thomas,
Everything is important, but each in its own time . . . When I began my research 59 years ago (!), sound quality was a huge problem, and stereo was a novelty. I know you appreciate what I and my research teams have contributed to alleviating the sound quality issue. It is a good start to pleasure when voices and musical instruments sound as they should. Mediocre loudspeakers still exist, but there is no excuse for them any longer. The industry has measurable performance targets that correlate highly with subjective ratings acquired under controlled, blind, conditions and Amir on this website delivers them.

But neutral loudspeakers do not ensure neutral reproduction because of small listening rooms. The early reflections modify directional/spatial perceptions and stereo itself its a fundamental limitation, corrupting both sound and spatial qualities. Low frequency room resonances contribute audible excesses (booms) and gaps (missing or attenuated notes), and these effects are different in different seats. Equalization can improve sound quality for a single listener, if that is the objective. However, movies, video concerts and the like are often social events, to be shared, so seat-to-seat variations become a consideration. It turns out that multiple subwoofers, employed in specific active and passive arrangements can alleviate the seat-to-seat variation problem and global equalization can then address the sound quality issue for multiple listeners. It can work very well from the sound quality perspective.

All of that was achieved with no real consideration of low-frequency spatial perceptions, probably because, over the years, the bulk of what was regarded as important in listener envelopment (LEV) was focused on what was happening in the octave bands from 250 Hz to 2 kHz, and they are still, it seems, an important basis for evaluating concert halls. This, of course, is overly simplistic, as what is measured is typically reverberation time (RT), a time-domain phenomenon. The desirable spatial effects are assumed to follow because of quasi-random arrivals of reflections from many incident angles, and research even identifies "preferred" angles - mainly those associated with side wall reflections. So far so good, because envelopment has been considered to be a prime factor in evaluating concert halls. But to me the generally applied science could be improved.

Bass frequencies have always been a concern, but my observation indicates, in the 125 years of concert hall acoustics, that it is only recently that the attention has included considerations of more than RT and amplitude. From our understanding of the binaural hearing mechanism (my PhD thesis), there has to be more, and David Griesinger was my introduction to anyone paying more than passing attention to directional/spatial/enveloping perceptions at very low frequencies. In large venues it makes absolute sense, and it is one of my continuing pleasures over decades of season tickets to classical concerts. It is real.

However, in small listening rooms there are challenges. The collection of resonances themselves perceptually characterize the size of a room. Very large rooms tend not to exhibit many problematic, distinctively audible, low-frequency resonances - the Schroeder frequency of a concert hall is likely to be around 100 Hz, maybe lower - not my area of expertise. But small rooms have distinct "personalities" - think of singing in the shower. Eliminating clearly audible small-room resonances seems like a fundamental step in the direction of reproducing the sound of large-rooms.

The next challenge is to deliver the necessary poorly correlated low frequencies to the ears of listeners - without losing the desirable attenuation of resonances, and, optionally, reduced seat-to seat variations. Immediately we confront decisions and priorities. Ideally we can identify a method within which multiple objectives can be simultaneously satisfied.

Years ago David Griensinger mounted a test of what might be possible in a "stereo bass" configuration in a relatively dead listening room at Harman. A group of highly experienced and opinionated listeners experienced a protracted experiment of variables, including digitally concocted signals and many musical selections. We heard some spatial effects, but all were in the subtle end of the scale and there was no benefit from bass management crossover frequencies below about 80 Hz. David appeared to be more sensitive than some of the others, and this might suggest a learning curve. He has written that the effect is audible in concert halls down to 60 Hz. You, Thomas, have said 50 Hz, I believe. So it is agreed that there is a low frequency limit, as is likely associated with the spacing of the ears. Obviously, a real test should begin outdoors - most anechoic chambers would not qualify. Once a baseline for audibility is established, add reverberation and vary the crossover frequency to see at what points the desired auditory envelopment is degraded.

That or any other research is no longer my concern; I am retired. But suffice it to say that our understanding of binaural hearing indicates that these effects should be audible, and they are in large halls. Translating the phenomena enjoyed in real halls to multichannel delivery systems in small rooms is the challenge - without degrading sound quality for one or more listeners. As the recording is a key variable, I wish future researchers good luck.

In the meantime, I have a real-life situation to deal with - I have relocated from a comfortably large listening room in California to a significantly smaller (13' x 16'. 4m x 4.9m) room in Ottawa, Canada. It has been convenient that the room is rectangular, making modal control simple, but equipment and furniture are restrictions. It turns out that two small 80 Hz low-passed, bass managed, subs can be located 25% of the distance from the side walls and the rear wall. In the furniture arrangement they are not obvious, and they attenuate the first three width modes, across which the listeners are arranged, and couple poorly into all but the first order length mode. One can walk from side to side across the listening area and little changes, but there remains a first-order mode excess in the rear seats. I am now running the front broadband, floor stander, loudspeakers full range to destructively interfere with the length mode. The front loudspeakers deliver what "stereo" bass there is, but not from the best locations in the room. Playing Davids DropBox files of correlated and decorrelated 100 Hz low-passed noise indicated that I now seem to have retained the desired resonance control while delivering some of auditory envelopment. It is definitely somewhat "out-there", not mono. Interesting, and I may not be done yet. But, when it is good, mono low bass is highly enjoyable with the majority of program material. Cheers.

 

[audiofooled]

Right. However, AE latent in a production, or AE manifest in reproduction, are not easily quantifiable. Even a systematic vocabulary is missing, after 40 years of academia mostly only considering reproduction timbre.

As much as I appreciate Floyd and Harman’s excellent research on loudspeaker directivity and timbre, it is about time to also consider the hard problem of reproduction: Spatial qualities of the listening room+system, from an AE perspective. AE is even the primary perceptual and emotion-inducing potential 3D sound has to offer over stereo, regardless if AE is naturally recorded or constructed.

I agree. It has been my understanding that, on the production side, all the required tools of the trade are there, including monitoring systems which have become increasingly good over the years. Advancements are for real, so is competition. It would be no surprise that investing in proper low end monitoring seems like a step in the right direction. It's an art, but also business.

Good stuff is already out there, if one is so inclined and cares much about spatial quality. This artist, for example, seems to do it very well routinely and in some of his tracks for as low as the systems are capable:

Reproduction in small listening rooms, with regards to spatial quality and AE, is entirely different problem. IMHO, to such an extent that unfortunately, multiple subs summed to mono, seems to be the only strategy that is so far repeatable and common denominator for more than one room/more than one sub of certain specifications. Also, measurement methods are very well established. However, this does not solve for the perception of space, as tempting as it may be to save cost and real estate with small stand mounts and tucking in some subs where suitable. Floorstanders are simply big.

AE requirements for reproduction include quite a lot more than just smooth FR/low distortion in mono bass, be it for multiple seats or not. It is, as you say, room+system, which raises more questions than answers. Small room acoustics must perceptibly disappear for a believable illusion that such long wavelengths are not contained within the walls. The strategy on how to implement both mono and stereo bass in the same room is seemingly polar opposite for practical reasons. It is likely a simulation and measurement method that needs to be well established as a tool that shows when, where and how the two concepts coexist.

It is a matter of multidisciplinary approach to the problem and keeping an open mind about subjects that have been neglected over the years. At this day and age, now that high quality gear and signal processing tools are there, why not take it up a notch or two?

 

[Duke]

Small room acoustics must perceptibly disappear for a believable illusion that such long wavelengths are not contained within the walls.

Agreed. "You are there" is a different - and arguably greater - challenge in the bass region than it is further up the spectrum.

 

[MattHooper]

Even though I know intellectually the challenges of creating the sensation of a large space using stereo speakers in a smaller room (mine is 13’ x 15’ with a large room opening to the hallway), in practice I have found I can get quite a believable-to-me sensation of listening into a very large space if it’s on the recording. This regularly happens with symphonic recordings.
When I close my eyes, if I mentally adjust my “ seating distance” to make the scale of the orchestra make sense, the feeling of listening to an orchestra can be quite realistic. (though, as I say, this is an area where bringing my own imagination, willing to meet the illusion halfway, comes in to play)