All About Atmos – a guide for MEs. Part 1

Seemingly out of nowhere, Atmos has become one of the most talked about formats. But if you haven’t worked in film audio, or have been burnt previously by investing in surround for audio, it can seem all a bit bewildering. 

We’ve been making Atmos monitoring gear since 2014, and have a lot of experience and clientele in this area. To help the conversation, we’ve written a comprehensive set of notes to explain what Atmos is and isn’t, what it requires to make it work, and how to make it work with a traditional stereo mastering suite.

This is part 1. Click on the links to go straight to the other parts, or keep on reading :)

About Atmos -Part 1 : Atmos basics

  • What is it and how does it work?
  • Beds and objects
  • Atmos, 7.1, 5.1, Stereo and binaural
  • How do you master an Atmos mix?

About Atmos -Part 2 : What is the role of a ME in an Atmos mix?

  • Music or Video?
  • Stem mastering
  • ITB or Analogue chains
  • Back to school

About Atmos -Part 3 : Setting up an Atmos room

  • Rooms & Speakers
  • DAWs and other kit
  • Monitoring

About Atmos -Part 4 : Integrating a stereo room with Atmos

  • What to think about
  • Crookwood solutions

Part 1 : Atmos Basics

What is Atmos?

Atmos is a proprietary surround sound format, developed by Dolby.  It was released in 2012 primarily as a film (movie theatre) sound format, but has evolved since then.  It also continues to evolve, so bear this in mind, given this article was written in early 2022.

There are a few key features to Atmos:

  • On top of a traditional 2D surround system (speakers arranged at ear height spaced around you), Dolby has added additional speakers above the listener’s ears (in the ceiling), which can add a 3D sense to surround audio
  • As well as normal surround mixes (called “beds” in Atmos speak), you can individually position or move a huge number of audio sources (“objects”) anywhere within the 3D Atmos audio space.  In total you can have up to 118 channels of mono or stereo (stereo = 2 channels) audio objects within one Atmos mix
  • In common with most Dolby formats, the Atmos mix will automatically downmix (“render”) itself to suit the capabilities of the user’s monitor system.  It does this using mixing metadata about where an object should be, which means that it will translate well to another Atmos system, a 7.1 or 5.1 surround setup, or even a stereo, binaural headphones setup or Ambient (B format) setup.
  • On the subject of headphones, there is an important feature.  Atmos provides for a separate loudspeaker experience, and a separate headphone experience – a mix can sound different across the two methods of listening.  Atmos does this by rendering a binaural mix instead of a stereo mix for headphone users. Done right, binaural mixes can create a much better stereo, or even surround effect than a normal stereo mix can provide when played on headphones

How is an Atmos mix made?

In an ideal world, an Atmos mix is made in a properly calibrated Atmos room.  The actual mix is ITB, and uses a Dolby plug in to record and pan the audio objects & beds, which are imported.  This plug in available for many DAWs, although the film people tend to use a ProTools / Avid setup.

You can’t master the whole Atmos mix, but you can however process (master) the individual objects or beds of an Atmos mix, and then re-render them into the Atmos mix. This is sort of like stem mastering.

The output of an Atmos mix is a set of files – a master file, a delivery file, plus various other rendered files for discrete formats checking/QC or production on different listening environments (stereo, 7.1, 5.1, binaural etc).

What do I need to make/ monitor an Atmos mix?

  • At a minimum, Dolby say you need a 5.1.4 speaker setup.  However, they recommend a 7.1.4 setup. This is 7 speakers on a horizontal plane, 1 subwoofer (LFE) and 4 overhead speakers.
  • You will need a treated room that can support these speakers, with possibly some speaker delays and DSP correction.
  • A DAW that supports the Dolby plug ins, with sufficient power to be able to process the audio objects etc.
  • The Dolby plug ins.
  • A monitor controller/ audio interface that can drive/control the speakers.
  • A Dolby Atmos licence.

This is just an overview.  Part 3 of the article deals with this in more detail.

How “fi” is an Atmos mix?

Atmos supports a maximum of 96KHz 24 bit files for beds and objects.  However, the output file has to be down-sampled to 48K. Atmos itself is not lossy, but in order to stream an Atmos file to consumers, it goes through another Dolby process called “Spatial Coding”.  This is a lossy format, and reduces the bandwidth from a potential 100Mbps for cinemas to 384kbps for streaming.

How do I master an Atmos mix?

Good question. The answer depends on where you are placed in the production chain.

If you are, say, doing classical work, the chances are that you are in control of the whole process from recording to delivery file.  In this case you’ll effectively create the mix using your recorded sources, processing them prior to mixing them in Atmos, then QC’ing the final outputs for delivery.

If, however you are doing a traditional ME job, the Atmos mix will have been decided for you, and now you have a few choices:

  1. You work with the Artist at an earlier point to essentially mix and master the stems used for the Atmos mix.
  2. You leave the Atmos mix alone, and instead just provide a separate stereo Hi-res mix and a binaural mix for headphones.
  3. You look to tweak the provided mix and objects to produce the “final” Atmos delivery files (including binaural), as well as perhaps doing a separate stereo mix

Note, you cannot process an Atmos file – it must be reloaded into a DAW and edits made to the underlying beds and object audio files.

Stereo and Atmos

Because Streamed Atmos is a lossy format with a 48K maximum delivery sample rate, it arguably cannot compete with a basic CD for audio quality.  This means that for most delivery programs (Spotify Hi-Res, Amazon Hi-Res, Tidal, Apple Lossless etcas well as CD & Vinyl), you will still be asked to produce a separate stereo master, as per your normal ME work. The Atmos renderer can produce a downmixed stereo file, which you can stereo master as per normal, or you can re-pan all the beds and objects into stereo. But this rendered file is not ideal as a starting point for stereo mastering. A stereo mix is quite different from a downmixed Atmos stereo mix.  Really you will need a separate stereo file, as per your normal work.

Binaural Mixes

Within the Atmos environment, you can make a separate binaural mix.  This uses another Dolby plug in, and lets you re-place the objects and beds, to make them work better in a headphone environment.  You don’t re-process the audio files however, just re-position them. Experienced MEs have noted however that while a Atmos speaker render is forgiving, a binaural mix is not, and you may have to tweak the speaker mix to ensure the binaural mix sounds good too.


This is another area that ME’s should be involved in.  Because the master Atmos file is rendered differently on different setups, a ME should prove that these downmixes work well.  If an issue is found, then the master Atmos file will need to be tweaked to correct the issue. See binaural mixes above.


Surprise, surprise, Apple have their own streaming implementation of Atmos, called “Spatial Audio”.  While it’s very similar, there are differences, so again the files need to be checked on an Apple decoder as well as a Dolby one.


Atmos provides many new opportunities for ME’s, albeit requiring significant investment.  It’s not certain that Atmos will end up being the default music delivery system, but Apple’s involvement may push it this way.

However, there is still the strong case that Stereo is not dead, especially the Higher res versions of Stereo.

The plot thickens…

Part 2 deals with the more ethereal issues of what an ME can add in value to the Atmos process, Part 3 looks at the technical requirements, and Part 4 looks at how an ME can setup an Atmos room.