A Broad Discussion of Speakers with Major Audio Luminaries

[Thomas Lund]

The field of audio is small. “String of temporary opinions” sadly largely covers it, like palaeoanthropology from where it comes. There are exceptions, e.g. directivity of loudspeakers, the early Nordic standards on FR in reproduction, momentary hearing thresholds etc. However, we rely so much on ABX testing that temporal effects are being systematically overlooked; and it is not only the mono sub fallacy ABX has brought upon us.

As to causality regarding perception, receptors in the skin, mesentery etc., and other physiological elements, may be identified, but that only provides part of the equation, and time influences living beings on a very wide range of scales. In addition, as put by the late, great neuroscientist, Jaak Panksepp, hearing is a special form of touch, where temporal effects have long been found and examined.

Our experiments with AE are not much more than another reason for more temporal studies. It is remarkable, though, how young and old humans appear to recognise abstract LF stimuli in similar ways. Like the other senses, hearing appears to be fundamentally sensitive to stasis vs change, also inter-aural at LF and VLF.

In case you are up for a nerdy weekend with more on science and listening, here is Francis’ talk at the Sidney Shure theatre,

 

[goat76]

Not quite. As I said above, every theory must comprise a causal reason for an effect to occur - basically that *is* the theory. That's missing here (as far as I know). So, we have no theory. Hence we cannot test it. As I said before, it might not be the 'stereo' thing and localisation of phantom sources, but it could simply be decorrelation speaker one and two in conjuction with room modes. I do not dismiss observation, but without reasoning, how to properly recreate it to begin with, just doing things somehow?

When the only parameter that is changed is going from stereo to mono signal in the bass region, and there is audible differences, it can only mean that the audio file contained uncorrelated data in the bass region.

Or if we turn it the other way around. If there wasn't any differences in the actual audio file between the channels to begin with, you wouldn't hear any changes in the sound at all going from stereo making it a mono signal.

The sensation of AE or envelopment doesn't have anything to do with the localization of phantom sound objects or localization of low bass frequency content. It's a way more diffuse sensation than that, the sound is less “trapped” in the space between the two loudspeakers, less congested, and more open and natural.

I suggest you make the comparison listening test yourself, it's not that hard to set it up as I previously described.

 

[Fidji]

When the only parameter that is changed is going from stereo to mono signal in the bass region, and there is audible differences, it can only mean that the audio file contained uncorrelated data in the bass region.

Or if we turn it the other way around. If there wasn't any differences in the actual audio file between the channels to begin with, you wouldn't hear any changes in the sound at all going from stereo making it a mono signal.

The sensation of AE or envelopment doesn't have anything to do with the localization of phantom sound objects or localization of low bass frequency content. It's a way more diffuse sensation than that, the sound is less “trapped” in the space between the two loudspeakers, less congested, and more open and natural.

I suggest you make the comparison listening test yourself, it's not that hard to set it up as I previously described.

I have done this and I can concur that even high quality mono bass is not the best option.

I have 8 sub well controlled DBA/Waveforming setup with good measured performance and full-range speakers capable to go below 20Hz in well treated room, that I am able to control to flat curve via DRC.

My observations:
LR4 crossover at 80Hz places some instruments into dead center e.g. kick drum, making sound stage feel compressed as it competes with e.g vocals.
I get benefits in spatiality and envelopment, while enjoying decay and impulse control of Waveforming up to 55-60HZ LR4 crossover.
No audible benefit in getting lower, from spatiality perspective.

In my specific conditions I get deep cancellation of my L-R speakers at 38Hz, so here I already need to be at Mono signal

As a side note - REAL difference is using heavily adjusted Auro3D upmixer, I have 9.8.7 setup with surround speakers placed at firs reflection points for side and rear wall. 2Ch playback does not have fighting chance against properly setup MCH playback system for both stereo and MCH recordings.

 

[JPA]

For all of you who have doubts about AE, it’s quite easy to set up a listening test to hear the difference between mono and stereo bass. If you want, your can also invite your mother to the listening test if she's interested.

First, set up your subwoofers in a stereo configuration on each side of your main speakers, and obviously connect them in stereo. Open a DAW and put an HPF on the master track, and set it at 80 Hz and with a 24 dB slope. Import all kind of different music tracks you want to test, and just listen to what happens to the bass under 80Hz in isolation when toggling between mono and stereo on the master tracks (most DAWs should have that functionality).

To me, it's easy to hear the effects @Thomas Lund describes as “big”, “open”, “free”, “pleasant”, “light” in favor of stereo bass, in opposite to mono bass, which is described with words like “unpleasant”, “claustrophobic”, “small”, “restricted space”.

How much of a difference it makes is of course highly dependent on the audio production, and if they contain stereo bass information or not.

A 24 dB/oct slope is still going to allow quite a bit of music through between 80Hz and 160Hz. Not an effective test setup in my opinion.

 

[Heinrich]

I suggest you make the comparison listening test yourself, it's not that hard to set it up as I previously described.

And still there‘s no theory behind the claim. So what to test and how? Any effect that may originate in arbitrary phase decorrelation in bass in-room?

 

[NTK]

For those interested in some contrived test signals, I followed Dr Griesinger's method and generated an example using narrow band noise (using 25 multi-tones), with each tone having randomly fluctuating phase. Basically the noise is a sum of tones generated by:

signal = Σ A_n * sin( 2π*f_n*t + ɸ_n(t) )​

The phase is randomly fluctuating in the order of ~0.5 Hz. An example of the ɸ_n(t) function is shown below (horizontal axis is time in seconds, vertical axis is in degrees).

Below are the spectra of the generated noise and their waveforms. The noise is centered at 50 Hz, and the amplitudes of the tones following a bell shaped curve with 50 Hz as its center. The signals have insignificant level of contents above 80 Hz.

With a "bass managed" system with a cross-over of 80 Hz, the left and right channels of noise will be downmixed to mono. The attached WAV file in the ZIP archive starts with the mono downmixed version (5 seconds long) and followed by the stereo version (5 seconds long). The sound clips aren't exactly pleasant to listen to, and I've only listened to them with headphones (I have no access to loudspeakers at the moment). Because of the closely spaced tones and varying phase, there is plenty of beating. The difference between the mono downmix and stereo is clearly audible.

The question, IMHO, is not if stereo bass can be audibly different from its mono downmix — it certainly can. It is how well we can reproduce it in small listening rooms with loudspeakers.

 

[kyuu]

The question, IMHO, is not if stereo bass can be audibly different from its mono downmix — it certainly can. It is how well we can reproduce it in small listening rooms with loudspeakers.

Also whether different is better, and how much so. Most people will have to wildly alter their setups and probably buy new equipment to even attempt to reproduce it. I certainly can't with my current setup.

 

[sceptical1]

Dr. Toole keeps referring to resonances as some all encompassing reason for why speakers are unable to reproduce what is input. Does anyone know what he means by resonance?

 

[audiofooled]

The question, IMHO, is not if stereo bass can be audibly different from its mono downmix — it certainly can. It is how well we can reproduce it in small listening rooms with loudspeakers.

In rooms the most difficult part is energy in vs. energy out. The lower the frequency, the more energy, more control is required. Not easy to do.

Supposedly I invert the polarity on one of the channels in attempt to get close to zero pressure (no smoothing, 15 cycles FDW):

Group delay:

The same, but no FDW:

GD:

A lot could be going on, high and low pressure being equally important. But pressure is only part of the equation.

 

[levimax]

The question, IMHO, is not if stereo bass can be audibly different from its mono downmix — it certainly can. It is how well we can reproduce it in small listening rooms with loudspeakers.

I have 2 large stereo subs co located with the mains in a fairly small room (11 ft X 24 ft) and when playing this clip the second half sounds different than the first half so I don't think reproducing it in a small room is a big issue. I guess "how well" is another question and then "how well" does it need to be reproduced to give a useful "LE" effect.

I have tried quick switching between "mono bass" and "stereo bass" when playing music but when I do so the "mono bass" is 6 dB louder so pretty hard to hear anything beside
louder.

 

[tifune]

. It is how well we can reproduce it in small listening rooms with loudspeakers.

Sorry if I missed it; has "small" been clearly defined within this context?

 

[JRowland]

The question, IMHO, is not if stereo bass can be audibly different from its mono downmix — it certainly can. It is how well we can reproduce it in small listening rooms with loudspeakers.

I switched my mains to full range a few days ago after listening to the Billie Eilish tune @Thomas Lund turned us on to in that other thread, thank you very much by the way. I played with my crossovers and 40 hz worked pretty good and 60 hz was ok at 80 it wasn’t there, subs outside but close to the mains. Mains alone gave me a little of the tingles feel I get in a good church, I very much enjoy it. I’m lucky my mains (R7 meta) get down to 20 hz with room gain and sound great so I’ve been listening to different songs posted having a good time. Green Light by Lorde was another good one, it has hard right panned deep bass a few brief moments.

 

[NTK]

Sorry if I missed it; has "small" been clearly defined within this context?

Small, in this context, is probably everything smaller than a concert hall or church (see Dr Toole's post earlier in this thread). Low frequency envelopment is created by large enclosed space (not necessarily fully enclosed but has plenty of reflection surfaces). The sensation is from our two ears hearing decorrelated bass signals. As J J has pointed out numerous times, to fully characterize "sound" at a point in space, besides the sound pressure level, we also need the 3 components (i.e. x, y, z in 3-D space) of the acoustic particle velocity. The particle velocity tells us in which direction at that instance and at that measurement point the sound pressure wave is propagating. It give us the information on how different the sound will arrive at our two ears. If the sound wave is propagating (or bouncing back and forth) in the left to right direction, it can give a different sensation than if it is in the front to back direction, when the sound pressure magnitude is the same.

...
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.
...

 

[Newman]

Sorry if I missed it; has "small" been clearly defined within this context?

Domestic = small
Mixing and mastering studio = small

If you are wondering whether your listening room at home is small, it is.

Cheers

 

[livinon2wheels]

A small transducer is able to reproduce higher frequencies and these are "faster" than lower frequencies - that is all there is to it. Within a given bandwidth, all loudspeakers are equally "fast". There are no "fast" woofers because the "speed" is determined by the highest frequency it reproduces. If it is bass managed, that is typically 80 Hz. The room resonances are the "slow" part, and unless these are controlled, bass will be "slow", as MaxwellsEq said. Science is different from subjective reviewing - facts are involved.

Dr. Toole, i have a specific question that I have been pondering as I attempt to redesign systems for my home that is being rebuilt. For an overall sytem size is it reasonable to assume there is a minimum amount of cone area and xmax cone travel necessary to reproduce bass frequencies down to below 20 Hz given a listening room's volume? For example lets say you have a 4000 cubic foot room and you want to have bass extension flat to 20 hz @ a sound pressure level of 115 db. Is there a formula for calculating such a thing or a way that you are familiar with to arrive at a reasonable approximation of achieving this. I want to make sure I get there but I don't want to spend more than I need to to achieve the desired results. Does this make sense? Thanks in advance for addressing my question.

 

[Newman]

is it reasonable to assume there is a minimum amount of cone area and xmax cone travel necessary to reproduce bass frequencies down to below 20 Hz given a listening room's volume?

While we wait for the reply you seek, I will propose that the minimum will apply to a bass reflex cabinet tuned to 20 Hz (not a sealed cabinet), and the sub placed in the corner of the room.

But what about a 20 Hz bass horn?

If you seek the answer for a sub type and placement that you have in mind, it might be worth mentioning it.

Cheers

 

[Keith_W]

Dr. Toole keeps referring to resonances as some all encompassing reason for why speakers are unable to reproduce what is input. Does anyone know what he means by resonance?

I would normally let Dr. Toole answer this for himself, but he appears to have left the thread. A resonance is an oscillation when an object is excited at, or close to its natural frequency. In your listening room, things like glass windows, doors, speaker cabinets, etc. can all resonate. He has said elsewhere (can't find it right now) that by far the largest source of resonance is the speaker driver. He decried high end audio where "heroic" measures are taken to reduce cabinet resonance but speaker driver resonance is relatively ignored.

 

[bmc0]

The question, IMHO, is not if stereo bass can be audibly different from its mono downmix — it certainly can. It is how well we can reproduce it in small listening rooms with loudspeakers.

Thanks for the test signal. For whatever my personal anecdote is worth, the difference in spatial quality is clearly and easily audible with my loudspeaker setup.

 

[Sokel]

Thanks for the test signal. For whatever my personal anecdote is worth, the difference in spatial quality is clearly and easily audible with my loudspeaker setup.

The difference is audible, yes.

If I had to describe it, first part is more coherent but confined between the speakers.
Second part sounds like it's fluctuating a little but gives the impression of different things going around at R and L.

Also, for some strange reason second part sounds a little louder.

My set-up is stereo all the way down to (F3) 31Hz, EQ'd for room up to 100Hz, etc.

 

[Vuki]

Dr. Toole, i have a specific question that I have been pondering as I attempt to redesign systems for my home that is being rebuilt. For an overall sytem size is it reasonable to assume there is a minimum amount of cone area and xmax cone travel necessary to reproduce bass frequencies down to below 20 Hz given a listening room's volume? For example lets say you have a 4000 cubic foot room and you want to have bass extension flat to 20 hz @ a sound pressure level of 115 db. Is there a formula for calculating such a thing or a way that you are familiar with to arrive at a reasonable approximation of achieving this. I want to make sure I get there but I don't want to spend more than I need to to achieve the desired results. Does this make sense? Thanks in advance for addressing my question.