Envelopment: Toole, Multichannel, Binaural

[Fitzcaraldo215]

Seems to me that this...

“The answer, of course, is that multichannel specifications are designed to recreate movie sound effects,”

This, not technical limitations, is the real roadblock. If the ultimate goal is to create concert hall sound, the most fundamental problem is that so few recordings even attempt to capture it in the first place. There are some audiophile and classical recordings that do, using the current multi channel systems, but multi channel is a small fraction of a shrinking home audio market, music, as opposed to movies, is a small fraction of the multichannel content market, and content that aims to reproduce natural hall ambiance (as opposed to, say, five Eagles singing harmony from five different speakers) is a small fraction of the multichannel music market.

There is just not enough money in it to change the way hardware is made, much less recordings. It will have to be synthesized from stereo, or your listening options will be so limited that...well, who was it that said audiophiles use music to listen to their equipment, not vice versa? This is that. In the extreme.

The bad news is that, if you do not like classical music, you may be right.

The good news is that if you do like classical music, and if you know and love the sound of live classical music, there are thousands of recordings out there to please you. But, unfortunately classical is a niche. It looks pretty big to those inside the niche, but it looks insignificant to those outside it.

 

[Fitzcaraldo215]

Wow! Great write-up! The link to that AES book seems broken though?

I agree with most of your points. EDIT: even though I have heard stereo setups which felt real to me, but only when seated exactly in the sweetspot.

I would add that well-designed and well-placed omni speakers also are able to recreate some of that which is often lost with conventional stereo.

Listening to a well-implemented and music oriented m.ch. setup is very high on my todo list this year.

Omni speakers, like Bose's approach decades ago, relies on using reflections in a small listening room to try to recreate what happens in a large concert hall. Guess what? Ain't gonna happen. But, enjoy the belief that it does happen.

 

[iridium]

Actually it is very simple to achieves those goals with:
A logarithmic spaced, articulated spiral array in a horizontally asymmetric configuration employing frequency tapering and divergence shading, which includes isophasic high-frequency and mid-frequency apertures, hyper-cardioid low-frequency transducer sections, is controlled by finite-impulse response filtering digital signal processing, and works well with a psychoacoustic infector.

iridium.

 

[Fitzcaraldo215]

Movie application is very different to music reproduction. Movies use effects to simulate off-screen 'occurrences'.

Surround sound for domestic music entertainment is trying to recreate the recorded ambience by modifying the domestic ambience. Music performances generally don't have direct sounds coming from the sides or rear. Is it really so good or is it just another illusion/preference?

What you ignore is what most people ignore. Direct sound is only a part of what is heard live. There is actually much more unseen but heard reflected energy perceived in most seats in the hall than direct energy. Our eyes deceive us. Our hearing follows different rules, though our mind prioritizes what we see over what we hear. We see the performers. We therefore understand sound as direct sound. We do not see all the infinite number of reflection paths of that direct sound on its way to our ears in the hall, but that is what our ears hear.

Here, for starters. I cited it before, but read this:

http://www.regonaudio.com/Records and Reality.html

There is also the commonly held "Bose Fallacy", that listening room reflections are somehow going to recreate and "fix" the hall reflections that the stereo paradigm takes way. Bose made millions from that. But, good luck with it. It don't work.

 

[Scott Borduin]

Absolutely awesome, Scott. Beautiful, insightful writing about one of my favorite subjects.

@Fitzcaraldo215 , thank you. There is much to chew on in your reply, and I promise to get to it tomorrow when I'm caught up on sleep and can organize thoughts.

Scott

 

[Scott Borduin]

Actually it is very simple to achieves those goals with:
A logarithmic spaced, articulated spiral array in a horizontally asymmetric configuration employing frequency tapering and divergence shading, which includes isophasic high-frequency and mid-frequency apertures, hyper-cardioid low-frequency transducer sections, is controlled by finite-impulse response filtering digital signal processing, and works well with a psychoacoustic infector.

iridium.

Ha! Well done.

Almost certainly intentionally, you raise an interesting point - the theory and technology of advanced audio reproduction is increasingly inaccessible for the consumer of that technology. That is true for me, even though I have a strong technical background. To make an analogy, my other major hobby is in the automotive space, where I own part of a repair/performance shop and do a lot of tuning on modern powertrain control modules (PCMs). A modern PCM will contain literally hundreds of models of engine and transmission behavior, almost all of which will make no sense on the surface to our customers who primarily have background tuning carburetors, distributors, transmission shift cables, and the like. The thing I try to do is explain complex technical subjects that I might understand at some level of technical depth, but in ways that are accessible and applicable to people from different perspectives. You all will be the judge of how well I do that!

Scott

 

[iridium]

What you ignore is what most people ignore. Direct sound is only a part of what is heard live. There is actually much more unseen but heard reflected energy perceived in most seats in the hall than direct energy. Our eyes deceive us. Our hearing follows different rules, though our mind prioritizes what we see over what we hear. We see the performers. We therefore understand sound as direct sound. We do not see all the infinite number of reflection paths of that direct sound on its way to our ears in the hall, but that is what our ears hear.

Here, for starters. I cited it before, but read this:

http://www.regonaudio.com/Records and Reality.html

There is also the commonly held "Bose Fallacy", that listening room reflections are somehow going to recreate and "fix" the hall reflections that the stereo paradigm takes way. Bose made millions from that. But, good luck with it. It don't work.

Remember that old true saying: Bose, no high, no lows.
Makes you wonder what music halls they used to model their speakers!
Oh, I know, it was the closet in the Marketing Department.

iridium.

 

[iridium]

Ha! Well done.

Almost certainly intentionally, you raise an interesting point - the theory and technology of advanced audio reproduction is increasingly inaccessible for the consumer of that technology. That is true for me, even though I have a strong technical background. To make an analogy, my other major hobby is in the automotive space, where I own part of a repair/performance shop and do a lot of tuning on modern powertrain control modules (PCMs). A modern PCM will contain literally hundreds of models of engine and transmission behavior, almost all of which will make no sense on the surface to our customers who primarily have background tuning carburetors, distributors, transmission shift cables, and the like. The thing I try to do is explain complex technical subjects that I might understand at some level of technical depth, but in ways that are accessible and applicable to people from different perspectives. You all will be the judge of how well I do that!

Scott

Just as long as it has at least one Vertex Magneto; then I am very happy.

iridium.

 

[Cosmik]

Our eyes deceive us. Our hearing follows different rules, though our mind prioritizes what we see over what we hear. We see the performers. We therefore understand sound as direct sound. We do not see all the infinite number of reflection paths of that direct sound on its way to our ears in the hall, but that is what our ears hear.

You seem to be agreeing with my contention that we "hear through the room", but rationalising it as an illusion caused by vision.
When you are at one of your concerts do you close your eyes and suddenly find yourself afloat in a swirling sea of paths and reflections? And then open them again and everything is back to normal? Or does it continue to sound pretty much the same either way? For me, it's the latter.

 

[oivavoi]

Omni speakers, like Bose's approach decades ago, relies on using reflections in a small listening room to try to recreate what happens in a large concert hall. Guess what? Ain't gonna happen. But, enjoy the belief that it does happen.

Nah, not quite. That's a simplistic argument actually. Bose relied on the dominance of ambient over direct sound, with the drivers playing into the front wall and not towards the listener. Good omni speakers can be listened to in the near-field, with point-source dispersion, and the direct sound thus becomes dominant. Linkwitz' recommendation is that omnis need to be listened to near-field for critical listening, and I agree with him. What ideally happens then is that the direct sound determines the spatial and distance cues that the ear perceives. This necessitates stereo recordings where these cues have been preserved. The ambient sound - which is almost an exact replica of the direct sound - then softly reinforces these cues from the direct sound, while also providing a pleasant sense of envelopment.

There's a lot of prejudice out there on the omni idea, a fact I find slightly annoying. That said, good omni designs are also hard to find, with most people immediately thinking of things like MBL, which can't be listened to near-field.

EDIT: Like I said, I still haven't auditioned a good multichannel rig for music. So I'm not saying that omnis work better for recreating the real thing than multichannel. I'm just saying that good point source omnis (which excludes most commercial offerings, the best I've heard so far were a modified pair of Linkwitz Plutos, and I'm still awaiting delivery on my Morrisons), well placed and with the listener placed near-field, can do a pretty good job of making stereo sound like a real acoustic sound field.

 

[oivavoi]

I had MBL speakers at one time. I found them to be addicting at first, not so much later on. The problem was that the sense of space was too similar recording to recording - a factor of speaker/room interaction, probably.

See comment above. Problem with the MBLs is that they can't be properly listened to near-field, which makes the listening room the dominant perceived space.

 

[RayDunzl]

Problem with the MBLs is that they can't be properly listened to near-field

Why not?

 

[oivavoi]

Why not?

The drivers are placed too far from each other (and the mid driver is very large), so the signal from the different drive units needs some distance to properly integrate and fuse.

 

[RayDunzl]

Can't that be "corrected" for a nearfield listening position with a dose of DSP?

 

[oivavoi]

Can't that be "corrected" for a nearfield listening position with a dose of DSP?

Perhaps! Haven't tried. But intuitively, I would assume that kind of thing to be a hard law of physical sound waves that DSP can't change, just like DSP can't make small woofers play deep bass without distortion...

 

[svart-hvitt]

multi-
ˈməltē/
combining form

1. more than one; many, especially variegated.

Stereo, 2 channels, is, by definition, multichannel.

(sorry)

Smartass!

I will strike back at you in mischievous ways for pointing that out!

 

[Kal Rubinson]

Was the demo Auro or Atmos?

There were 2 events I attended although many other speakers (people) used the system for demonstrations. This is what Morten Lindberg tells me about the system:

1. Source was our original production files at 24/352.8 from a Pyramix workstation thru a HAPI converter over Ravenna AoIP.
2. For this second session with Dan we also used Pure Audio Blu-ray thru a StormAudio ISP 3D.16 ELITE unfolding 9.1 at 24/96: www.aes.org/events/143/spatialaudio/?ID=5814
3. Loudspeakers were PMC IB2S XBD-A: https://pmc-speakers.com/products/professional/active/ib2s-xbd
+ for the heights: PMC twotwo.6 https://pmc-speakers.com/products/professional/active/twotwo6​

 

[Fitzcaraldo215]

Remember that old true saying: Bose, no high, no lows.
Makes you wonder what music halls they used to model their speakers!
Oh, I know, it was the closet in the Marketing Department.

iridium.

Actually, Dr. Bose published a number of academic research papers, possibly at AES also, while still an MIT Professor. One of those carefully measured the levels of direct and reflected sound energy from an audience perspective. I do not recall the hall/halls he used or how many different seating locations. I believe Symphony Hall Boston was used, possibly others, as well. He came up with the 8/1 reflected to direct energy ratio, enshrined in his famous 901 with 9 identical drivers, one facing front, 8 facing the rear on 2 angled faces.. Also, it was possibly the first successful commercial speaker to employ active EQ for the bass.

So, there was a fair bit of actual science involved, imperfect though it may have been in his research and the implementation. Yes, there was also a huge amount of marketing hype, as well, once the speaker was commercialized.

He likely was not the first, but I think he helped put the importance of powerful hall reflections on the map. Subsequent research went into much more detail with much greater refinement.

I think the 8/1 ratio was too crude and oversimplified, myself. He also did not seem to mind the common use of his speaker in discos and PA systems, but turned around so that the more powerful 8 speaker side faced the listeners.

 

[Scott Borduin]

Absolutely awesome, Scott. Beautiful, insightful writing about one of my favorite subjects.
I also don't see much point to Mch for studio recordings, since studios are deadened spaces to facilitate multitrack stereo recordings, including techniques like extensive panning and mixdown.

Well, if you've ever heard Dark Side of the Moon in 5.1, you've heard how surround effects can create interesting artistic impressions in pop music independent of enhancing any sense of performance space.

I really do not like the terms "envelopment" and "surround sound". I think they are easy to misconstrue. Also, the multidimensional space, ambiance, etc. feels totally natural to me live, not specifically "enveloping" or "surrounding", though they are that. The sense of space, ambiance, etc. are just an inseparable part of the total, singular impression of live sound, somewhat vaguely and diffusely. My sense is that good Mch recordings convey this in a similarly natural way, without undue emphasis of the presence of additional speaker channels. That should be totally seamless in a properly set up Mch system. And, I have countless recorded examples of where it is just that.

I used the word envelopment because Toole uses it, and because it is apparently a term commonly used in describing concert hall quality. But I get what you're saying - I don't sit in Schnitzer hall and think I'm being enveloped.

For whoever reasons, I am not a fan of artificial synthesis of additional spatial channels. I believe that the ITU standard for the 5.0/.1 Mch scheme adopted by Sony/Philips for their Scarlet Book for SACD (the SACD equivalent of CD's stereo-only Red Book) clarified and became an important standard for Mch music. That was even beyond SACD, not only in home speaker setup, but also in determining what music recording engineers actually used and aspired to in miking, mixing and production for music playback with most realistic reproduction. At least one Mch label, Channel Classics, uses only a 5 channel ITU mike array. Others may use more mikes, but the same ITU array is the backbone of the setup, and final mixing and mastering is done via an ITU speaker array.

So, I am not enthusiastic about needing to add extra speaker channels at +- 60 degrees in order to possibly create additional "envelopment" beyond what is already there, as captured by mike arrays and engineering according to ITU standards. A pleasurable sonic effect? Yes, possibly. But, is it more "real"? Questionable. I prefer to stay with what the engineers mixed and mastered from mikes in the hall. I have been satisfied with that.

I found the link to Toole's AVS forum post on this: https://tinyurl.com/y7gflkjl . Toole is pretty knowledgable about theater sound, and is not impressed that the ITU standards are based on much in particular. And he seems to think that placing sides at +-60 is best for music, regardless of whether there are additional speakers in the system. His thoughts about 9.1 are for systems which need to serve up HT effects as well as music. If you're happy with the results you get from your setup, of course no need to change. For those of us who never got 5.1/7.1 to sound exactly natural, simple experiments with relocating side speakers are easy and free!

 

[Fitzcaraldo215]

...

For a time, I experimented with true discrete multichannel, but was never fully satisfied with the results. To my ears, the surround channels tended to pull the stereo image forward, and add some spaciousness, but real envelopment seemed elusive. Perhaps it was just my particular setup, or I just never found the right recordings, but I wonder what would have happened had I just known to experiment with surround channels placed further forward. Toole himself suggests a "9.1" configuration with the two extra speakers at +-60, decorrelated from the side channels via simple delay of 10 ms or so (in a post on AVSforum, the link to which I have lost).

...

Scott - yes, as you point out, Mch modifies the frontal soundstage vs. stereo as well as providing surround ambience or envelopment. Elsewhere you cover the value of the center channel, which does enhance the frontal image tonally and spatially, as well as "anchoring" the frontal image against head movements and uncentered seating positions. This seems grounded in good science.

In the above quote, you refer to pulling the frontal image forward into the room toward the listener. I agree completely that this happens, and I find it a good thing. That is part of why I find that discretely recorded Mch recordings offer greater depth of image consistently. It is not a loudness thing, but the depth dimension spreads from in front of the speaker plane to, as stereo does, behind the plane. I also find discrete Mch recordings render instruments or voices with more dimension, like the sound of the body of a cello more "fleshed out" in almost 3D on good recordings. And, the apparent soundstage is wider in Mch, which should not be surprising.

Forgive my subjectivism in this, particularly in a science forum. However, others and I believe we have heard these imaging properties on numerous Mch systems in comparison to stereo. I do not now if there are any scientific listening tests which would verify these specific properties, nor any which refute them.

However, if these properties are desirable and real, they are achieved by complex interactions between the different speaker channels, including phantom imaging between front, center and surround channels.

To achieve these imaging properties in accord with and with fidelity to the master recording, I believe in maintaining reasonably close adherence to the standard ITU speaker layout at 0, +-30, +-110 degrees, with near identical speaker voicing and calibration for equal channel volumes and speaker distances. Deviations from this, if not too great, are not necessarily terrible, and the imaging differences vs. the mastering studio may not be noticeable or identifiable. But, I do prefer, after some experimentation and difficult room constraints, to stick with the standard used in (most) discrete Mch recording and mastering. It may be just a mental thing, but I find the results better with consistency.