Speakers

Are Integrators Deceiving Clients with Their Immersive Audio Setups?

Adding speakers for Dolby Atmos, DTS:X or Auro 3D to create great immersive audio only works if extra speakers’ signals are processed individually, but most surround sound processors only offer 11 channels. It’s a problem.

Are Integrators Deceiving Clients with Their Immersive Audio Setups?
Are your clients getting what they paid for in their home theater immersive audio system?

·

There is extreme “channel confusion” taking place in the custom installation industry, but it has nothing to do with Internet reselling or distribution. It’s the channel confusion related to “discrete” versus “processing” channels in surround-sound or new immersive audio setups.

Integrators are proactively deploying object-based surround sound systems such as Dolby Atmos, DTS:X and Auro without surround sound processors (SSPs) or receivers (AVRs) that can accommodate more than 11 discrete channels. The result is clients are possibly not getting what they paid for in their home theater immersive audio system.

Arnaud Laborie, founder, CEO and chairman of the board for Trinnov Audio, makers of SSPs, notes that the problem is that integrators are confused between a “discrete” channel and a “processing” channel.

Laborie deciphers the situation for CE Pro:

“Dolby Atmos and DTS:X are both object-oriented formats, and Auro has announced that it is moving in this direction as well. ’Object-oriented' in this context means that once you go beyond the basic 7.1 speaker setup that we all know well, additional sounds are not assigned to specific ‘channels.’

"Instead, ‘audio objects’ are defined composed of a combination of the sound itself and metadata to describe the location of the sound over time relative to the main listening position (MLP). It is the responsibility of the SSP or AVR to ‘render’ these objects appropriately to the available speakers.

“Currently, we know of no DSP-based solution for more than 11.1 (12 total) uniquely-rendered channels."
— Arnaud Laborie, Trinnov Audio

“Almost all current SSPs employ a DSP platform from one of a couple DSP chip companies to do their 3D Audio decoding and rendering of the audio objects. These companies depend on large volumes in order to recoup the significant investment in R&D and tooling required for designing new chips. This need for high volume leads to them making decisions about their architectures that are driven by the mass market.

“Currently, we have not been able to verify any DSP-based solutions for more than 11 uniquely-rendered main channels (excluding subs). As recently as ISE, a couple companies announced the ability to move this number up to 13 from 11, which is certainly a step in the right direction. This means that, regardless of the number of speakers used, the effective spatial resolution is 7.1.4, or perhaps 7.1.6, regardless of how many output connectors reside on the rear panel.

"For example, a large theater might have several speakers on the left sidewall; however, they would all be receiving the Left Surround (Ls) signal, albeit perhaps with different delay and level adjustments.

"We expect that a DSP-based solution for perhaps as many as 16 discretely-rendered channels will become available toward the end of 2017, which will be an improvement, to be sure. However, even then, manufacturers will have to develop (or update) their products using the newly-introduced DSP engines, which will take some time.

Conflicting Immersive Audio Format Specs

“There remain differences based on the architectural decisions made by the three immersive audio companies. For example: The Dolby Atmos specification allows for a maximum of 34 unique speaker locations (plus LFE), and provides for up to 24 separate speakers with unique signals at listener level, plus another 10 channels above the listener.

"Auro 3D provides for as many as 13 unique speaker locations (plus LFE). Hence, if you have a high-channel-count system, you will usually end up arraying some of the channels by sending them to more than a single speaker location. 


Related: Home Theater Demo Guide for Showing Differences Between Immersive Audio Formats


"DTS:X currently allows for 11 speakers out of a total of 30 possible locations (plus LFE), or a 7.1.4 arrangement in the commonly-used parlance. Once again, in a higher-channel-count system, you would normally end up setting up arrays of surround and height speakers to ensure that all the speakers are used.

“Clearly, Dolby Atmos currently provides the greatest potential for high spatial resolution. Trinnov recently announced a coming, optional software update that will allow it to render all 34 discrete channels provided for in the Dolby Atmos specification, plus another 14 output channels (48 total) to support complex bass management systems, or for speakers employing the Trinnov's Altitude32’s active crossovers.

“However, the differences do not stop there. All three immersive audio formats listed above have different recommended locations for the additional loudspeakers (beyond the basic 7.1 layout we have used for years). This fact has led to some people advocating for multiple sets of somewhat redundant loudspeakers, and using only the ones that are pertinent to the format being played at the moment.

"We feel this is a significantly inefficient waste of available resources. Moreover, it seems that it must be a difficult proposal for the client to accept: paying for perhaps 30 channels of amplification and speakers and then using only a dozen or so at any one time.

What Is the Solution?

All is not lost for integrators. Laborie explains that Trinnov has a 3D Microphone that can accurately measure the actual location of each speaker in space (to within 2 degrees of azimuth and elevation, and 1cm in distance), relative to the reference position.

That information, combined with the company’s software called Remapping technology, can create phantom images of these sound objects that will be perceived as being located accurately, even if there is no speaker at the indicated location. 

Laborie explains:

“In principle, this process acts like stereo imaging, except that it operates all around you rather than only in front of you. This combination of unique technologies allows us to reproduce all three immersive audio formats from a single, sensible array of loudspeakers. 

"Adding speakers to create great spatial resolution only works if you can process all of these extra speakers’ signals individually, and with full knowledge of where they exist in the actual listening space."
— Arnaud Laborie, Trinnov Audio

“As with stereo, these phantom images work best when you are in the ‘sweet spot.’ When multiple listening positions are considered, the rationale for high speaker counts becomes even stronger. This is so because the location of the sound will be less dependent on phantom images and instead be determined more by the location of actual loudspeakers.

"This follows the same reasoning as having a Center channel speaker rather than relying on the phantom image from Left and Right. It locks the sound to a known, physical location, regardless of where you sit in the audience.

“However, adding additional speakers to create great spatial resolution only works if you can process all of these extra speakers’ signals individually, and with full knowledge of where they exist in the actual listening space. 

“Imagine entering an old-fashioned hi-fi demo room with a wall full of left/right pairs of speakers. Now imagine that all of them were playing the same stereo signal, at the same time. You would not expect the imaging to be precise (or even good), regardless of where you were in the room. This situation is analogous to reproducing multichannel soundtracks on more speakers than there are discrete signals to be used. At best, you have an extremely blurred, unfocused sound image.

“There is no question that maximizing the immersive audio experience available is both complex, and also extraordinarily rewarding. There is no substitute for a truly high-resolution spatial audio experience to complement the latest offerings in high-resolution video. Optimizing the design and implementation of these systems requires experience and sound judgment as well as technical skills. The extra effort is worth it.”



  About the Author

Jason has covered low-voltage electronics as an editor since 1990. He joined EH Publishing in 2000, and before that served as publisher and editor of Security Sales, a leading magazine for the security industry. He served as chairman of the Security Industry Association’s Education Committee from 2000-2004 and sat on the board of that association from 1998-2002. He is also a former board member of the Alarm Industry Research and Educational Foundation. He is currently a member of the CEDIA Education Action Team for Electronic Systems Business. Jason graduated from the University of Southern California. Have a suggestion or a topic you want to read more about? Email Jason at [email protected]

Follow Jason on social media:

Jason also participates in these groups:
LinkedIn · Google+

View Jason Knott's complete profile.



CE Pro Magazine

Not a Magazine Subscriber?
Subscribe Today...It's FREE!!

Comments

Posted by scolburn on March 27, 2017

Deception means intentional misleading. That’s not what is going on here. It’s more like ignorance of what’s needed to optimize object-based audio compared with traditional multichannel home theater design.

Sound objects have size. For immersive audio to pull listeners into the action on the screen rather than distract them away from it, sound objects need to keep their size constant. With chip-based renderers if you put multiple speakers on a single output channel, that size changes depending on how the sound object pans through the array and where listeners are relative to the array. E.G. Imagine a human sized voice growing and shrinking as it moves around the room. Distracting, no?

Having more native discrete channels certainly helps. When not possible, using speakers with wide broadband dispersion and high power response for side and back surrounds has proven to work quite well and helps immersive audio become engaging like it should.

Steve Colburn
Triad Speakers

Posted by highfigh on March 25, 2017

Why is it OUR fault? We’re just selling the BS formats the manufacturers trowel out that are half-baked when they hit the market. If they would bother to wait until the products/formats are fully-formed and mature enough that we won’t be the ones who have to deal with PO’d end users who expect this stuff to work after a firmware update, we wouldn’t have to do so much damadge control. Do we get anything for this? If anyone thinks they’ll be paid for the time we take fixing manufacturer’s screw ups, they’re naive. We used to be paid for warranty service, but they killed that off because it cost THEM too much. What about OUR costs that aren’t recovered? HDMI boards don’t grow on trees!

Since we’re on the topic, who needs surround modes to make a system do a bad impersonation of a Large Hall, Small Hall, Jazz Club or any of the others? 5 channel, sure. Party mode, to eliminate the delay when using Zone2 when the problem could have been handled internally? Uh-huh. Get rid of the others- nobody needs them!

Design the surround system to create a TRULY realistic sound field and it may become exciting. Until then, I’d rather to listen to two channels.

Posted by highfigh on March 25, 2017

Why is it OUR fault? We’re just selling the BS formats the manufacturers trowel out that are half-baked when they hit the market. If they would bother to wait until the products/formats are fully-formed and mature enough that we won’t be the ones who have to deal with PO’d end users who expect this stuff to work after a firmware update, we wouldn’t have to do so much damadge control. Do we get anything for this? If anyone thinks they’ll be paid for the time we take fixing manufacturer’s screw ups, they’re naive. We used to be paid for warranty service, but they killed that off because it cost THEM too much. What about OUR costs that aren’t recovered? HDMI boards don’t grow on trees!

Since we’re on the topic, who needs surround modes to make a system do a bad impersonation of a Large Hall, Small Hall, Jazz Club or any of the others? 5 channel, sure. Party mode, to eliminate the delay when using Zone2 when the problem could have been handled internally? Uh-huh. Get rid of the others- nobody needs them!

Design the surround system to create a TRULY realistic sound field and it may become exciting. Until then, I’d rather to listen to two channels.

Posted by scolburn on March 27, 2017

Deception means intentional misleading. That’s not what is going on here. It’s more like ignorance of what’s needed to optimize object-based audio compared with traditional multichannel home theater design.

Sound objects have size. For immersive audio to pull listeners into the action on the screen rather than distract them away from it, sound objects need to keep their size constant. With chip-based renderers if you put multiple speakers on a single output channel, that size changes depending on how the sound object pans through the array and where listeners are relative to the array. E.G. Imagine a human sized voice growing and shrinking as it moves around the room. Distracting, no?

Having more native discrete channels certainly helps. When not possible, using speakers with wide broadband dispersion and high power response for side and back surrounds has proven to work quite well and helps immersive audio become engaging like it should.

Steve Colburn
Triad Speakers