Brief Summary

A common criticism of Dolby Atmos is that it is impractical to present immersive audio in the home and the only way most listeners will experience it is over headphones. This is often presented as an advantage that stereo has over Atmos, as correctly placing Atmos speakers in a domestic environment is too demanding for all but the most committed enthusiast. However in too many cases stereo has exactly the same issue. What is the problem and what does it mean? 

Going Deeper

With all the discussion around the practicalities of presenting immersive audio content in the home it’s easy to think that stereo doesn’t suffer the same drawbacks. Stereo has limitations in the sense that it presents a front facing soundstage which extends from one speaker to the other and no further. But because it is so well established it’s easy to overlook a basic flaw with the stereo experience: Most of the time when people hear stereo over loudspeakers in the home, they don’t actually experience stereo. Why is this?

How Stereo Works

Stereo recording and reproduction was invented by Alan Blumlein in the 1930s. His early work with EMI covered many aspects of stereo but he is probably best remembered for the mic array which bears his name. A Blumlein Array uses a crossed pair of directional microphones. These would have been fig 8 ribbons, though the first cardioid microphones were being developed at this time. The Blumlein Array consists of two fig 8 mics, coincident in space but pointing at 45 degrees either side of the centre line. More general XY coincident pair miking can use any directional pattern but the underlying principle is the same - it relies on level differences between the channels to create the stereo illusion.

At the same time in the USA Harvey Fletcher was developing an alternative to Blumlein’s coincident pair approach to capturing performances in stereo. His approach used spaced omnidirectional microphones to capture the performance and these differing approaches are often used to illustrate the roles the level differences captured by Blumlein’s technique and the time-based differences captured by Fletcher’s spaced microphones both play in creating believable stereo. The pan pot as we recognise it from mixers uses Blumlein’s level differences to place a sound in the stereo soundfield and while less common, delay based panning is a widely used technique

The Problem With Stereo

However, stereo reproduction over loudspeakers presents an issue which affects stereo playback regardless of which method is creating the illusion of stereo. You need to hear an equal contribution from each speaker and to achieve this you need to be in the so-called ‘sweet spot’ between the speakers. A small movement out of this central area where both speakers are heard equally results in both level and time-of-arrival differences which result in one speaker being heard so much more clearly over the other that the stereo image collapses and the perception of a ‘phantom centre’ element between the speakers disappears.

Exactly how wide this sweet spot is depends on the room’s acoustics as much as it does on speaker placement, but it is narrow. Even at large outdoor events it only takes a movement of a couple of metres either side of the centre line for stereo to collapse to ‘dual mono’. A phantom centre image is a fragile thing.

The Wall of Sound

One of Harvey Fletcher’s early experiments was using what was called the “Wall Of Sound” This had nothing to do with Phil Spector’s recording techniques or the infamous PA system owned by the Grateful Dead. It involved using a large number of microphones hung in a line across an orchestra and replaying the results over speakers placed in the same locations in a separate listening room. Unfortunately there was no way of recording that many channels at the time so this system was live-only. The reason I mention this experiment is that it illustrates the well known principle that the only way to ensure consistent placement of a sound in space is to place a speaker at that location. Virtual (or ‘phantom’) placements can be achieved by playing back from two (or more) speakers but the larger the angle between the speakers the more location-dependent the listener’s experience becomes.

Consistently Locating Sounds For Everyone

This issue is particularly well known in film and TV sound where the presence of a visual cue for where a character is on screen makes the effect of an off-centre listener hearing dialogue, coming from a character speaking on-screen, heard from a speaker located to the side of the screen unacceptable. The solution, a dedicated centre channel speaker, has been a feature of film sound for many years. It exists to anchor dialogue and on screen sounds where they belong - on-screen.

Live sound reinforcement suffers the same issue, and in the old days of point source PA systems some events used centre clusters above the stage to help with this ‘one side or the other’ dual mono issue. But since the 90s the line array, those characteristic long J-shaped hangs of PA speakers either side of the stage, have transformed the live experience in terms of fidelity, but have done little to alleviate the lack of stereo for concert-goers. Most live shows use stereo mixing sparingly because mixers understand that, unless you are standing in front of the Front of House position, you’re unlikely to hear both channels.

L'Acoustics L-isa

Changes are happening in the live sector. Immersive is as much of a buzzword in the live sector as it is in studios but it often means something slightly different. An excellent example of how top end PA manufacturers are embracing the possibilities presented by modern equipment is L’Acoustics’ L-isa system which can be used to create immersive systems, many of which can do all of the things we are used to in Atmos mixing, and for live music performance are often being used to fix the familiar ‘one side or the other’ stereo problem.

Very briefly the way it works is to create a system of overlapping speaker arrays which cover the width of the stage and uses clever processing to send audio to the relevant speakers to anchor audio in the soundfield at a given position rather than relying on a large LR speaker system to cover a large audience. Thereby giving a consistent soundfield experience over a large audience and presenting the sound of a performer from the same position as their position on stage from any listener’s perspective. Although L-isa can do far more than that.

Domestic Music Playback Systems

So back to domestic stereo. Car playback is a special case, a car interior is a fully designed environment, more so than a studio control room, and if the manufacturer prioritises audio system performance they can get very good results. But in the home, unless you are one of the very few people who has a dedicated listening room or home theatre (or studio, but that’s not strictly ‘domestic’), your listening experience is probably some kind of ‘dual mono’ most of the time. We all have horror stories of people who have one speaker in the living room and the other in the kitchen, or behind the sofa but correct setup of speakers is pretty rare these days. Even if you have a dedicated Hi Fi in your house, there is probably only one place from which good stereo can be enjoyed. Instead what I see is either ‘narrow stereo’ from a single appliance or smart speakers which bounce sound off surrounding walls to give an increased sense of width, but neither can be classed as a convincing stereo experience comparable to sitting between a properly placed pair of monitors.

Good smart speakers can sound very good for casual listening, but my recent experience of the Sonos Era 300 convinced me that the same can be said of domestic playback of Atmos content via a smart speaker. Which brings us back to headphones.

Headphones For Stereo And For Atmos?

It’s a common comment that the only version of an Atmos mix which really matters is the binaural version (coincidentally binaural techniques were pioneered by both Harvey Fletcher and Alan Blumlein back in the 30s). So few of the listening public have access to any other suitable playback system. The thing is, the same can be said of stereo. Although many of the stereo systems which exist in homes deliver a sense of generalised spaciousness, to get a convincing stereo soundfield, for most people that only happens in their headphones.

This is a positive conclusion for me. Much of the point of domestic playback over loudspeakers is to be able to enjoy music communally and the quality of smart speakers has risen and risen. The best sound very pleasant and are convenient and easy to live with. The fact that smart speaker solutions which work as well as they do with immersive content are available is a huge positive, as good immersive mixes sound great and it’s never been an either/or choice between stereo and immersive. I’d expect immersive playback systems to become standard in the home in the same way that stereo replaced mono - People replace their hardware over time and the new hardware supports immersive.

Check out this video from the early days of Stereo. Companies tried to sell stereo to early adopters but the world eventually switched to stereo because over time it became the default. Why wouldn’t the same happen with immersive?

I’d say that the argument that Atmos isn’t practical in the home is, in most real-world cases, a flawed argument. Not because of the requirements for Atmos but because of the requirements for stereo! Would you agree?