What contributes most to 3-Dimensionality of sound systems deliver?

[KlausR.]

The first post of this thread mentioned a 360° mechanical/electrical sound reproducer; the MBL 101 X-Treme loudspeaker system. And the OP's experience was the most 3-dimensional sound he has ever experienced in his life. That 360° dispersion is the main contributor for that 3D holographic sound illusion, plus the integrated room acoustics with it.

I would assume that with such designs the level of the early reflections is higher than with forward firing speakers and that the spectrum of those reflections is closer to the spectrum of the direct sound, provided that the room boundaries are not too absorbing. On the other hand sound field measurements in small rooms with omnidirectional speakers as source have shown that the field is highly directional. So with omnis we still have a directional sound field with a stronger participation of early reflections.

Now, if say we take a classical music recording … on CD, with the ambiance of the concert hall recorded on his master, and play it in our acoustically treated room; the clues (ambiance hall) coming from the front soundstage will be perceived by our brain, and distributed in our room around our ears.

And that hall ambiance recorded on the recording will transmit well into our brain as if we were in that hall.

You certainly will hear that the recording has been made in a large room, but you will not be able to feel as if you were in that room simply because for having the impression to be in a large room you need the early reflections to come late and you need the reverberation time to be high, but your small listening room doesn’t give you either and the cues from the recording are not the correct ones. There is psychoacoustic literature on what cues are important for room size perception, but that’s a whole topic on its own.

The off-axis dispersion if close to the on-axis one (say up to 75-90°) will contribute to the 3D imaging, holographic sound.

I wonder how the off-axis behaviour is able to have an effect when the reflection points are treated?

Also, a full classical orchestral music recording is impossible to recreate @ realistic live levels...classical chamber music, small acoustic jazz ensemble are more feasible for our limited speaker's systems in our rooms...even from larger ones...full range.

My speakers put out a healthy 123 dB, so realistic levels are not really a problem?

There are many types of 3-dimensionality from the sounds of music in our homes.

The same sound event will be perceived differently by different listeners, soundstage included. The reason is the difference between the individual in-ear responses, and there is no “hearing better” or “hearing worse”, all there is, is hearing differently. And this is exactly why I skip the subjective/listening part of reviews, because the reviewer does not have my ears with my personal and unique head related transfer function.

Back to my question: which components of the sound field are responsible for soundstage and 3-dimensionality? What component creates width, what component creates depth?

Klaus

 

[fas42]

Back to my question: which components of the sound field are responsible for soundstage and 3-dimensionality? What component creates width, what component creates depth?

Klaus

IME, as a simple answer, it's the ease with which one can register the low level acoustic cues that were picked up by the microphones - it's a quality issue, not a "Is there something there, or not?" That information is always picked up, whether the recording engineer intended it to happen or not - a poor quality playback so muddles this detail that it doesn't make sense, and subjectively soundstage and 3D don't exist; improve the rendition to a very high level, and then you can pick the cracks in the wall of the recording studio , .

 

[16hz lover]

Big stiff non resonant woofers with a lot off extension without much roll off , increases the soundstage, off course the housing must be able to cope with the energy .

+1

 

[KlausR.]

Back to my question: which components of the sound field are responsible for soundstage and 3-dimensionality? What component creates width, what component creates depth?

IME, as a simple answer, it's the ease with which one can register the low level acoustic cues that were picked up by the microphones - it's a quality issue, not a "Is there something there, or not?"

With components of the sound field I meant stuff like frequency response, phase response, direction of direct sound and reflections etc., and parameters such as angle between listening axis and loudspeakers, distance, level, distance between loudspeakers, etc. For instance, interaural cross-correlation seems to be a very important parameter, the lower, the better the 3D impression.

Klaus

Edit: Here's some info re: interaural cross-correlation I've found in one of my write-ups.

A high interaural cross-correlation is equivalent to high binaural similarity which results in low subjective preference (Schroeder 1979). Conversely, for more dissimilar ear signals (high binaural dissimilarity) the preference is high. Acoustic asymmetry lowers interaural cross-correlation which results in an increased sense of spaciousness and generally in increased preference (Ando 1977, Toole 2006). In order to obtain zero cross-correlation (maximum dissimilaritiy) in two-channel stereo systems the angles of the loudspeakers with respect to the listening axis should be 23, 67, 126, 158º (Damaske et al. 1972), according to Ando (1998) the optimum angle for all sound sources is 26º. Interaural cross-correlation decreases (with subsequent improved spatial impression) when the reverberation time of the room increases (Kurozumi 1983).

Ando (1977), “Subjective preference in relation to objective parameters of music sound fields with a single echo, Journal of the Acoustical Society of America, vol. 62, no.6, p.1463

Ando (1998), “Architectural Acoustics”, Chapter “Design of electroacoustic systems”, Springer

Damaske et al. (1972), “Interaural crosscorrelation for multichannel loudspeaker reproduction” (Interaurale Kreuzkorrelation für mehrkanalige Lautsprecherwiedergabe), Acustica, vol. 27, p.232

Kurozumi et al. (1983), „The relationship between the cross-correlation coefficient of two-channel acoustic signals and sound image quality”, Journal of the Acoustical Society of America, vol. 74, no. 6, p.1726

Schroeder (1979), “Binaural dissimilarity and optimum ceilings for concert halls: More lateral sound diffusion”, Journal of the Acoustical Society of America, vol. 65, no. 4, p.958

Toole (2006), “Loudspeakers and rooms for sound reproduction – a scientific review”, Journal of the Audio Engineering Society, p.451

Klaus

 

[NorthStar]

1. I would assume that with such designs the level of the early reflections is higher than with forward firing speakers and that the spectrum of those reflections is closer to the spectrum of the direct sound, provided that the room boundaries are not too absorbing. On the other hand sound field measurements in small rooms with omnidirectional speakers as source have shown that the field is highly directional. So with omnis we still have a directional sound field with a stronger participation of early reflections.

2. You certainly will hear that the recording has been made in a large room, but you will not be able to feel as if you were in that room simply because for having the impression to be in a large room you need the early reflections to come late and you need the reverberation time to be high, but your small listening room doesn’t give you either and the cues from the recording are not the correct ones. There is psychoacoustic literature on what cues are important for room size perception, but that’s a whole topic on its own.

3. I wonder how the off-axis behaviour is able to have an effect when the reflection points are treated?

4. My speakers put out a healthy 123 dB, so realistic levels are not really a problem?

5. The same sound event will be perceived differently by different listeners, soundstage included. The reason is the difference between the individual in-ear responses, and there is no “hearing better” or “hearing worse”, all there is, is hearing differently. And this is exactly why I skip the subjective/listening part of reviews, because the reviewer does not have my ears with my personal and unique head related transfer function.

6. Back to my question: which components of the sound field are responsible for soundstage and 3-dimensionality? What component creates width, what component creates depth?

Klaus

I would love to know if anyone with a large room and system can recreate a full orchestra symphony as live in a classical concert hall?
Kids can get close to two hundred decibels inside their cars.
Some theaters (public or private) can get 150dBs. Would that be the right recipe to recreate a live volcano erupting within say 3-5 miles from where we stand/sit?
Is loud means closer to the 3D real event?
Are two speakers best for 3D-dimensionality or is more better?
Where are the limits on the MBL 101 X-Treme loudspeakers system? Are they limitless in their 3-dimensionality? Are they the best @ that 3D holography than say horn speakers?

Multi-speakers; are they a gimmick in the year 2016 from ultra hi-end audiophilia forums, or is there any science behind it that can be counted as valid enough to blow the winds in the four corners of the globe and boil the oceans surrounding them?

Is DSP a consideration to emulate 3D space from music reproduction? Say Yamaha for instance with their real venue measurements from judicious microphone positioning and with their data encapsulated/transcribed inside DSP chips working @ high speed?

Computers with multi-core processors (8 to 16 cores) and trillions of parameters information communicated in less than 0.001 second; can they compete with room treatment acoustics?

Where are we today in the reproduction of the very best 3D sound illusion in our homes?
_________

Sometimes the best answers are the questions?
_____

1. An acoustic music concert is projecting sounds in 360°. Could be a small jazz ensemble in a small jazz club or an auditorium, etc., could be a classical chamber orchestra with twelve musicians playing in an art space or small concert hall, could be blues band playing in a long narrow bar club or private live house...tango, flamenco, samba, salsa, ...
There is no amplification, everything, all the instruments and singers are acoustic, strictly.
The room will dictate what we hear, and the speakers in our own rooms would need to match those live events from their dispersion pattern in tandem with our own room acoustics.
Can two speakers do that? Omnipole, horn, direct radiator, dipole, bipole, ...?

An electric rock, or blues, or jazz, or electronica, or punk, or metal live music concert...in a stadium or cultural centre or arena or amphitheater...; use amplification and speakers onstage with direct front dispersion plus multiple subwoofers with 360° envelopment (several techniques are employed, depending). When recording those live events, the various methods are almost limitless and same with the results all over the place. This is artificial music created in artificial venues and recorded and mixed artificially.
We need artificial loudspeakers to recreate the experience in our own rooms...with DSP and room EQ to adjust the sound to our own cohesive preference if not with the taste of the recording/mixing sound engineer.

Brief, the two worlds between un-amplified acoustic and amplified electric music. They are completely different; they project and reflect totally differently.

2. I agree. What is your take on DSP recreated acoustic spaces?

3. I believe, that for good 3D imaging a loudspeaker system with the attribute of excelling @ both on and off axis has a superior advantage.
...In the frequency response, with angles from 0° to 90° having close proximity in their measured responses, in particular from 500Hz to 15kHz.

4. 123dB; is that good enough for a full symphony orchestra, with 100 musicians, to be faithfully recreated in your own room's dimensions @ home?
Can a two-thousand seats hall space be transposed live in a room of say 40' L by 23' W by 15' H?

5. Assuming that two listeners have similar sets of ears with the same geometry design and size of orifice and level of overall balance; yes, that would be our best reference.
But we simply don't have this luxury in our audio/music passion/hobby/business. So even the replacement of fuses inside our audio electronics will sound different from one listener's set of ears to the next.

6. a) IMO...the recording, its source, and the source to reproduce it. Then the speakers chosen to match the room where they'll be reproducing, and balancing their acoustics.
6. b) IMO...great imaging from two speakers having the exact same frequency responses, and with close proximity in their on and off axis responses, plus air around them...meaning fair distance from the front and side walls...say no less than five-six feet. ...The larger the room the better. ...So, space. 3D spaciousness with width, depth and height is best reproduced in larger spaces...I truly believe. And only from the best stereo music recordings.

 

[NorthStar]

? http://www.ambiophonics.org/Tutorials/HowGood_1.html
? http://www.ambiophonics.org/TechnicalPapers.html
? http://www.ambiophonics.org/Tutorials/RGRM-RACE_rev.html

(((•))) http://www.ambiophonics.org/papers/AES123NYC0710.html

 

[fas42]

With components of the sound field I meant stuff like frequency response, phase response, direction of direct sound and reflections etc., and parameters such as angle between listening axis and loudspeakers, distance, level, distance between loudspeakers, etc. For instance, interaural cross-correlation seems to be a very important parameter, the lower, the better the 3D impression.

As I implied, IME all the parameters you mention are irrelevant - with a particular system I can shift from terrible, non-existent 3D, to a superb sense of space and positioning of sound elements; and then back again to mediocre. At no time in this process is the FR, phase, reflection makeup, angles, distance, etc, etc, altered in the slightest - the only variable is the quality with which low level information in the recording is presented.

 

[KlausR.]

1. I would assume that with such designs the level of the early reflections is higher than with forward firing speakers and that the spectrum of those reflections is closer to the spectrum of the direct sound, provided that the room boundaries are not too absorbing. On the other hand sound field measurements in small rooms with omnidirectional speakers as source have shown that the field is highly directional. So with omnis we still have a directional sound field with a stronger participation of early reflections.

1. An acoustic music concert is projecting sounds in 360°. Could be a small jazz ensemble in a small jazz club or an auditorium, etc. … There is no amplification, everything, all the instruments and singers are acoustic, strictly.
The room will dictate what we hear, and the speakers in our own rooms would need to match those live events from their dispersion pattern in tandem with our own room acoustics.
Can two speakers do that? Omnipole, horn, direct radiator, dipole, bipole, ...?

Traditional speakers cannot, because of their radiation pattern. What one would need for instance is a multifacet speaker such as a dodecahedron with 12 individual units and heavy DSP. Further you would need your room to replicate the performance venue. None of this is feasible at home so we have to live with what we can get from our 2-channel, i.e. some sort of illusion. Maybe that wave field synthesis does a better job, maybe systems like ODEON?

2. You certainly will hear that the recording has been made in a large room, but you will not be able to feel as if you were in that room simply because for having the impression to be in a large room you need the early reflections to come late and you need the reverberation time to be high, but your small listening room doesn’t give you either and the cues from the recording are not the correct ones. There is psychoacoustic literature on what cues are important for room size perception, but that’s a whole topic on its own.

2. I agree. What is your take on DSP recreated acoustic spaces?

If I look at stuff like ODEON, it looks as if it’s possible (www.odeon.dk).

3. I wonder how the off-axis behaviour is able to have an effect when the reflection points are treated?

3. I believe, that for good 3D imaging a loudspeaker system with the attribute of excelling @ both on and off axis has a superior advantage.

Agreed, so why treat reflection points with absorbers? Doing that you no longer enjoy the benefits of good off-axis behaviour!

4. My speakers put out a healthy 123 dB, so realistic levels are not really a problem.

4. 123dB; is that good enough for a full symphony orchestra, with 100 musicians, to be faithfully recreated in your own room's dimensions @ home?

Peak values of 120 dB have been measured in orchestra pits, so the answer is yes.

Can a two-thousand seats hall space be transposed live in a room of say 40' L by 23' W by 15' H?

Not with old fashioned 2-channel stereo, I’m afraid.

5. The same sound event will be perceived differently by different listeners, soundstage included. The reason is the difference between the individual in-ear responses, and there is no “hearing better” or “hearing worse”, all there is, is hearing differently. And this is exactly why I skip the subjective/listening part of reviews, because the reviewer does not have my ears with my personal and unique head related transfer function.

5. Assuming that two listeners have similar sets of ears with the same geometry design and size of orifice and level of overall balance; yes, that would be our best reference.
But we simply don't have this luxury in our audio/music passion/hobby/business. So even the replacement of fuses inside our audio electronics will sound different from one listener's set of ears to the next.

The problem is that you simply don’t know what exactly the in-ear frequency response of another person looks like and how it compares to yours, unless you measure both. The responses of 10 individuals have been measured in Shaw, “Earcanal pressure generated by a free sound field”, J. of the Acoustical Society of America 1965, p.465, and the differences were substantial (e.g. 20 dB at 3.3 kHz), so yes, you might hear a difference when using an audiophile fuse while I might not.

6. Back to my question: which components of the sound field are responsible for soundstage and 3-dimensionality? What component creates width, what component creates depth?

6. a) IMO...the recording, its source, and the source to reproduce it. Then the speakers chosen to match the room where they'll be reproducing, and balancing their acoustics.
6. b) IMO...great imaging from two speakers having the exact same frequency responses, and with close proximity in their on and off axis responses, plus air around them...meaning fair distance from the front and side walls...say no less than five-six feet. ...The larger the room the better. ...So, space. 3D spaciousness with width, depth and height is best reproduced in larger spaces...I truly believe. And only from the best stereo music recordings.

The recording and all what is before you don’t have any grip on, you have to take it as it comes. If the recording is bad, your system won’t make it any better. If the recording is good and your system is bad or not well placed in the room, the system will make the 3D impression of the recording worse.

@ fas42
By the same token I am able to imply that the recording is irrelevant. Take a good recording with excellent imaging and soundstage: this particular recording I can shift from terrible, non-existent 3D, to a superb sense of space and positioning of sound elements; and then back again to mediocre, by playing it through different systems, with altered FR, phase, reflection makeup, angles, distance, etc, etc, .

“two speakers having the exact same frequency responses”: which manufacturer measures and indicates on his website how well the two speakers of a pair are matched? So far I have seen none?

“close proximity in their on and off axis responses”: how many manufacturers do actually measure frequency response on and off-axis and show the graphs on their website? A handful perhaps?

The question is still not answered: which components of the sound field are responsible for soundstage and 3-dimensionality? What component creates width, what component creates depth?

Klaus

 

[jkeny]

As I implied, IME all the parameters you mention are irrelevant - with a particular system I can shift from terrible, non-existent 3D, to a superb sense of space and positioning of sound elements; and then back again to mediocre. At no time in this process is the FR, phase, reflection makeup, angles, distance, etc, etc, altered in the slightest - the only variable is the quality with which low level information in the recording is presented.

Absolutely agree, Frank
An example of this was in an audiophile friends house that I brought Richard (Opus112) to when he visited recently. The guy had setup up two systems for us to hear - a PC based front end & a SD card player front end. Both these were playing into the same DAC, same amp & same speakers. Room was far from optimal with various pieces of furniture & piano around.

Sound from the PC based system was between the speakers - 2D. Sound from the SD card player presented a deep soundstage - what I am referring to here. Could this have enveloped us & put us in the middle of the presentation if the room was attended to - probably?

PC system became much more 3D when a transformer was used in the signal path but only if this trafo was isolating the signal - when trafo input to output 0V were common, the sound stage depth was greatly reduced.

My belief is that the 2D sound was the result of masking of the low level detail by common mode noise - when this masking was removed we had full depth of soundstage.

As we both suspect, the low level details are what is being used by the brain to create the illusion of believability & we have found this is supported in our reading of the research in Audio Scene Analysis.
Like you, I believe that " FR, phase, reflection makeup, angles, distance, etc, etc" are not the important factors in soundstage low level information retrieval & presentation is the determining factor.

 

[NorthStar]

The question is still not answered: which components of the sound field are responsible for soundstage and 3-dimensionality?
What component creates width, what component creates depth?

Klaus

On the recording source, the microphones. ...And their positioning. ?
On the reproduction end, the multichannel setup; the speakers, and the Digital Signal Surround Sound Processor. ?

? Dispersion, Imaging (Toe-in +/-), Reverberation (reflections, echoes), Panning and Phase. ...Spaciousness. ?
_______

Just some good 3D sound articles:

? https://www.google.com/patents/US9154897
? http://www.lsi.usp.br/~regis/FariaRRA_ImprovingSpatialPerceptionInVR_VECIMS2005.pdf
_______

• http://www.bbc.co.uk/rd/projects/surround-sound-with-height
• http://www.soundonsound.com/techniques/inside-track-bjorks-vulnicura
_______

? http://www.ambiophonics.org/papers/AES136IRIS.pdf

 

[fas42]

My belief is that the 2D sound was the result of masking of the low level detail by common mode noise - when this masking was removed we had full depth of soundstage.

As we both suspect, the low level details are what is being used by the brain to create the illusion of believability & we have found this is supported in our reading of the research in Audio Scene Analysis.
Like you, I believe that " FR, phase, reflection makeup, angles, distance, etc, etc" are not the important factors in soundstage low level information retrieval & presentation is the determining factor.

Nice to see you around, John, you've been quiet here lately! Glad to hear you bumped into Richard, hope there was plenty of good talkin' and listening ...

Interesting get together with the audio friend just up the road, last night - his system goes up and down depending on how much he's played with things, and disturbed parts that actually were working well - from which he has to recover !! There were problems last night, and it was not really sorted out by the end of the evening - he needs to do some significant modding of a connection, which he's not 100% keen to do, understandably so; it's bite the bullet time! However, the 3D aspect was pretty decent, quite reasonable sense of space; and the relevant part is that he has an another audio friend, whom I've never met, who has gone the conventional route, just keeps throwing money at it, to "solve" things. Mega, premium brand speakers, the works -and on the "right" recordings it apparently can be quite spectacular; my friend visited him recently with a USB of tracks to try out, and afterwards felt pretty good about things, ... lots of that music came across very tiny, and unworthy, a small image barely as big as the speakers themselves, with no life. He played the "worst" track from that session last night, and it worked very well, in spite of his rig needing work - I could see why there was an issue with his friend's very expensive setup: a very 'produced' pop recording, lots of depth, and different reverb and effects around all the elements - the sort of thing that ambitious systems often completely muck up ...

 

[jkeny]

Nice to see you around, John, you've been quiet here lately! Glad to hear you bumped into Richard, hope there was plenty of good talkin' and listening ....

Yea, lots of good chat & audio was had over two days.
To be honest, I haven't read much on audio forums that has interested me recently - it's all the same ol' stuff over & over. A couple of times I have written a post & then decided not to post it as I didn't want the bother.

 

[microstrip]

(...) As we both suspect, the low level details are what is being used by the brain to create the illusion of believability & we have found this is supported in our reading of the research in Audio Scene Analysis.
Like you, I believe that " FR, phase, reflection makeup, angles, distance, etc, etc" are not the important factors in soundstage low level information retrieval & presentation is the determining factor.

I have written it before - even if the classical factors such as FR, phase, reflection makeup, angles, distance are well set, it is some kind of low level information that needs to be properly handled to reach a 3D image illusion that takes a lion part. My system is a good proof - cold the conrad johnson GAT and LP275Ms sound flat and small, after two hours warmup such as now , it has a real feeling of 3D in space, with a big sound. The same thing for the relative height of instruments in a few recordings - and please do not tell me it is due to tweeter and medium height, my Soundlabs have a single full range panel and the Quad ESL63 are point like speakers!

 

[NorthStar]

Francisco, can you transport us (teleportation) into your room so that we can also experience your 3D illusion? :b
How? Through the power of the Internet. I know it's less than perfect but a youtube music video is our best technological tour-de-force @ the moment.

* The day that we'll be able to share exactly what we're listening to in our rooms with our gear and our acoustics that'll be the day!
** John, glad to see you, and I miss Richard.