How do you handle music in Dolby Atmos?

I’ve had a lot of Atmos mixes over the last few days for final quality control before we use them for (UHD) Blu-ray authoring. These mixes are all from other studios. I noticed that most of them mix music above “ear level” exclusively to the left and right Top Middle channels (Ltm/Rtm). One reason for this may be that Atmos uses these two ceiling channels for the 7.1.2 bed.

To be honest, when we mix Atmos in our own studio, we usually send the music to all six ceiling speakers (with a few exceptions). (Most of the time just room reverb.) Personally, I find it illogical that the music only comes out of the two middle speakers. On “ear level”, all the speakers are (usually) in use for playing music.

So I would be interested to know how you deal with this? Why do you do it this way and not that way?

What I’ve read somewhere is that the best monitors for the end listener are often the front left and right monitors. But this explanation doesn’t work for me. It’s secondary. Personally, I deploy music everywhere (except on the front center if there’s dialogue, and on the LFE, with a few exceptions). I also raise the treble instruments a little towards the ceiling and keep the bass well down. It’s an artificial logic, but one that has been well coded over time, with the bass rising from the earth and the treble from the sky (without exaggerating the positions). It’s best not to break the codes, except creatively and for the right reasons. In sound effects, where I’m just a (passionate) beginner, this is also my logic, but there are exceptions, such as a thunderclap (part of the bass in the sky) or a cat meowing at your feet (treble below), or a thousand other such things.

On the other hand, I’ve noticed that in many cases (Amazon, Netflix, Apple), at least in my Atmos 5.1.2 room, which I haven’t had time to configure perfectly (unlike my studio, where I put all my energy), the music is often too loud in the rear. I’m not quite sure why. Maybe it’s the fear of getting too close to the dialogue in the center and drowning it out. However, there are dynamic EQ treatments that can protect their frequencies. That’s what I’m doing here with Fabfilter. Music, ambiences and effects are slowed down in the center as soon as there is dialogue, with a very rapid release. I say center, but my dialogues are always slightly widened on the left and right fronts, as are their reverbs.

But your question is really very wide-ranging and I’ve only brought out the first points that come to mind, without being too methodical (sorry).

1 Like

Totally. Anything less than x.x.4 is utterly useless for more than the occasional effect (… the proverbial helicopter passing above the listeners’ heads). x.x.2 is especially bad for the proper spatial impression of reverbs or real top-layer mics from actual recordings.

1 Like