What is Dolby Atmos?

Dolby Atmos is having a moment. Although the immersive audio format dates back to the first term of the Obama administration ¹, for most of its life so far it’s only really been known to the niche audiences of movie/home theater enthusiasts and Blu-ray Disc collectors.

That changed with its adoption by Apple Music. Anyone subscribed to the service can stream any Atmos-encoded songs in its catalog, at no extra charge. Along with lossless quality ², Atmos is a key differentiator for Apple Music from Spotify, which right now supports neither of those enhancements.

The Atmos basics

Now here’s the tough part: What is Atmos?

Even as someone who’s been immersed (pun intended) in its world since the early 2010s, I still struggle to explain it to most people. Fundamentally, it’s metadata embedded in a Dolby TrueHD or Dolby Digital Plus stream, with information about where a sound (codified as an “object”) should be played relative to other sounds within a 3-D map of a theater or home. The New York Times (gift link below) offered a pretty straightforward explanation this month:

Dolby Atmos, introduced in 2012, was initially developed for movie theaters and the home theater market. Because it offers a wider palette than stereo, and differs from traditional 5.1 and 7.1 channel setups, Atmos allows engineers—typically mixing across a dozen or more speakers—to put sound sources in front, to the side, behind and even above the listener.

So:

• More, more, more: It’s more complicated than stereo (i.e., two channels), with the ability to assign over 100 total objects in any direction. Accordingly, to get Atmos playback, everything in the chain—the master recording, the delivery service (such as Apple Music), and the playback equipment—must support its proprietary technology. The simplest setup for most people will be Apple Music plus AirPods; a more complex setup would be something like a 4K Ultra HD movie with Atmos soundtrack, a compatible disc player, and an Atmos/capable home theater.
• Spatiality: It emphasizes spatial audio that comes from multiple directions at once and/or moves around in a seemingly “3-D” way. For example, with AirPods on, if you set your phone flat on a table and then walk around it, it’ll feel like the vocals in the song (if it has them) are hitting your ears from different directions as you move—a good demonstration of how Atmos always maintains relative spacing between its objects.

Still, when I first heard about Atmos in 2014, I wasn’t clear how it differed from the (also-immersive) surround sound setups that movie and home theaters had used for years. But there are both subtle and overt differences:

Subtle

Atmos consists of audio “objects” that are specified as rectangular coordinates within a specific three-dimensional space, such as a theater or a sound system.

For example, Object A can be mixed to be “above” Object B, and so when it comes out of the speaker system it’ll seem like it’s coming from a very different source location.

The key is the relative positioning of such sounds as coordinates in 3-D space. This design means that Atmos content isn’t locked to or defined by presence in specific discrete channels, which can seem like a benefit or a drawback depending on your perspective:

• On the “benefit” side, you don’t need speakers that are in a specific configuration, e.g. center, rear-left, surround/right, and so on, because Atmos is adaptive and can keep sounds in the right relative spacing even when coming from a single Atmos-enabled speaker, which may use drivers to send sounds in distinct directions. In this way, Atmos allows for simplified setups.
• On the “drawback” side, it’s an open question if the simulated surround sound (basically, the sensation of sounds moving around in coordinated space despite only coming from a somewhat simple set of speakers or even headphones) they get from Atmos is a good or even close approximation of the 6 or 8 discrete channels on a traditional setup.

Overt

If someone knows anything about Atmos, they know that it has vertical channels. Many Atmos soundbars have speakers on the top, to fling audio toward the ceiling.

Helicopter sounds are a famous Atmos use case, because the verticality that Atmos allows can make it seem like the sounds of the rotating blades are coming from directly above you.

There is a rival technology called DTS:X that works virtually the same way.

But is it worth it?

Whenever a new commercial sound format like Atmos comes along, I feel like it’s healthy to be initially skeptical of it. That’s because these innovations are often presented as offering, at great cost, richer sound that meets the standards of self-proclaimed “audiophiles,” whose claims—about how you can hear sounds with noticeable improvements in clarity solely via better encoding and playback technology—unfortunately usually lack strong evidence.

It’s fine to be an audiophile about something like noticeable background noise on a podcast ³ or heavy distortion on a song that’s been mastered so “hot” that even listening to it on a low volume can hurt your ears. But audiophile arguments about basically any of the following strike me as shaky:

• Lossless audio: It’s a lot of extra data for virtually no change in perceived quality.
• “Hi-res” anything: Movies like Akira pioneered really high sample rates (the so-called “hypersonic effect”) that like the above don’t seem to sound any better than more modestly produced audio.
• Super Audio CD/DSD: A failed format/technology that offers much higher bitrates than CD as well as 5.1 surround. Blind listening tests couldn’t identify SACD vs CD at any rate above chance, i.e., of flipping a coin. The 5.1 piece is different, because that is distinct from stereo-only CDs, but playing back a 5.1 SACD mix is almost impossible without a Blu-ray player with 6+ RCA jacks.
• “3-D” sound: This is the marketing lingo around Atmos and it seems strange if only because our world is 3-D already and we hear sounds from different points constantly.

The truth is that the human ear has a very limited frequency range, especially compared to animals such as cats, and this range deteriorates with age. Stereo CD audio, which is 16-bit with a 44,100 Hz ⁴ sampling rate and a 20 Hz to 20 KHz frequency response, is already beyond the range of what a human adult can perceive

Side note: Vinyl and tape are far worse in this respect, so yes the CD really was a big breakthrough and an exception to the audiophile skepticism outlook I proposed above. It allowed for more dynamic range—there are bass frequencies I can hear on the CD version of the Captain Hollywood Project’s Animals Or Human that I can’t hear on the vinyl—and it also doesn’t degrade with each listen like analog media do.

We also only have two ears. Stereo sound is usually superior to mono, at least for music (podcasts are a different story) for this reason. I’m not clear that a similar jump can be made with spatial audio, and one of the audio engineers interviewed by the NYT had similar thoughts:

She noted that there are evolutionary and biological reasons that sound sources coming from behind and above listeners can be unsettling or anxiety inducing. She also observed that music is a potent form of communication in large part because the consummatory phase happens entirely in the listener’s head. Having clearer and more sound sources can actually make it harder to know what to pay attention to.

My own experience listening to Atmos has been that it seems cool sometimes—I liked how the voices and sound effects were spaced in the mix on George Miller’s Three Thousand Years of Longing—but overall it seems a lot like old stereo (on headphones) or surround sound (in home theaters) to me.

On headphones

AirPods are still relying on just two speakers to deliver their 3-D sound. The upcoming Apple Vision Pro may go further, but I mostly agree with the quoted interviewee that we may be getting a lot more audio information that we won’t know what to do with and that won’t enrich our listening experience.

In home theaters

Setting up a proper home theater is difficult. I had to place the speakers in my 5.1.2 (that’s five channels plus one subwoofer and two Atmos height channels) system strategically, ensure that I was using the HDMI eARC port correctly, and learn about how Dolby TrueHD differed from Dolby Digital Plus ⁵.

“Wait, timeout” you’re probably saying. “I thought Atmos was supposed to replace 5.1 and 7.1?”

It is, sort of—Atmos offer the potential for an immersive surround sound-like experience with fewer speakers and configurations, but it can also be supported on newer iterations of the same types of complex home theater systems it’s positioned as an alternative to.

On such systems, Atmos to my ears doesn’t differ much from the old 5.1 or 7.1, with the exception of occasional vertical sounds. People in the market for simple Atmos home sound systems (e.g., a 2.1 soundbar with Atmos) were probably never going to set up a 5.1 or 7.1 system, so Atmos really is something new and exciting.

The proprietary march goes on

Dolby Atmos is proprietary technology from a company whose business model Knives Out cinematographer Steve Yedlin has called “charging for extra signal processing.”

A cynical take on Atmos would hold that it’s all part of a scheme to make even the act of listening to music on headphones more expensive and exclusive. It’s off a piece with the removals of standard headphone jacks from phones. The implication will be that you can’t get the “full” experience without a full Atmos sound chain.

I’m not sure I’m that cynical yet. However, I do think Atmos is akin to other diminishing-return innovations such as 4K HDR and the Apple Vision Pro, in that it seems to oversell itself because its predecessors (stereo audio, 1080p video, and the smartphone, respectively) are still more than good enough for the use cases it purports to solve. It’s a fascinating concept but its biggest effects seem to be not on how much people enjoy what they’re hearing and seeing, but on the creation of long meta-takes and explainers like this one and the NYT piece.