I think there’s a lot of potential here. I’ve thought about this for years and been trying to mix music in atmos since way before it was available on streaming services.

Now that Apple has implemented it on Apple Music I want to talk about the huge problem I have with the way they’re marketing it and how most people are consuming it…

Spatial Audio on headphones and smart speakers needs to stop. I like soundbars like the Sonos arc, but seriously, stereo headphones? I feel like this is worse than lowering/compressing the quality of music files for streaming. If a mixing engineer mixed a song in atmos on a true atmos system then they’re down mixing with some sort of software into stereo format, do we really think this will ever work well?

What are you opinions on this? Do you any of you actually like these stereo binaural reproductions of atmos mixes? Is this the future? I personally do not like any of the atmos mixes I’ve heard on stereo heapdhones.

I understand we have to start somewhere. I hope this evolves quickly and consumers get better options for cheaper to consume the atmos format. But are they getting too far ahead of themselves here?

Yeah, I've been thinking about this a lot as I watch mixers spend thousands and thousands on Atmos setups. There's obviously a lot of interest and hype, and I can understand how much fun and inspiring mixing in the immersive format can be. I'm still struggling to see how this becomes the "new stereo" for the exact reasons you point out.

I'm not sure of the actual percentages, but it seems like most people will continue to consume music in headphones. I perused Apple Music's Atmos playlists last night with my AirPod Pros, and while there are definitely some very interesting mixes, I didn't necessarily feel immersed in the music in the way you would sitting in a multi speaker room. The mixes did seem more spacious, and the bass seemed more defined, but there's only so much a pair of earbuds can do.

Granted I have not experienced Atmos through a soundbar, but I'm still just not seeing the masses going out and spending money on another device. I hope I'm wrong and the format survives.

atmos (and other immersive formats) may have its charme for use in large cinemas but for music only releases, imo anything beyond 5.1 is absolutely pointless - no love for any downmixing here either...

I am on beta iOS iPad 15 and beta Monterey( both using the M1 chip), and also using the airpod max headphones. I think it’s hit or miss getting the full experience on anything else. Overall I notice a general quality increase, i think because of less stream compression. On the surround side, my gut feel is that a lot of the music mixing sound guys don’t really get 5.1 mixing let alone Dolby Atmos. What i am hoping is at least a couple of great film soundtracks are moved over(like “Interstellar”), so atleast there’s a quality reference to correctly mixed atmos music.

What do you feel is missing, or is 'wrong'?

Feels like I am more "on" the stage than in front of the stage, so my positional queues are off. I can't see the actual channels in my current setup, but heres what my ear is telling me given a 5.1 primary mix:
x.1 - rather than having the drops and effects going through the sub it seems like its always on
center - seems like voice is hard panned LR than bleeds into center, i prefer the opposite where voice is primarly center with only reverb sends going to other channels
LR - its either hard panned in music or the listeners position is way to forward (ie way too much sides)
LR sides - they seem the same volume as the forwards (which i think is throwing my listening position off)

Overall i really doubt these were mixed on a post dubstage, i dont have the expertice though to really know what is off.

Since I haven't worked with binaural spatial at all I'm curious if these things are the result of translating a surround mix to that format or if it's the actual mixing aesthetic that you're experiencing. What would your guess be?

They are using a dolby decoder that is supposedly happening on the listeners end to convert the true atmo's mix to stereo using the binaural thing that i dont really understand. All I know is more often than not it sounds much worse than the original stereo mixes. The mixes sound great on true atmos set ups!

Hi All,
My business partner and I have been posting in the New Product and Mastering forums about this. The thread in New Products is really long now. John Hanes has been posting there as well. His insights are excellent. I didn't want to simply duplicate those posts. I'm happy to offer some opinions here too if you wish.

We have been working in the format for about 1.5 years, and built an Atmos Music Enabled studio in Nashville last year.

I believe the format has so much potential. Unfortunately, the rollout with Apple Music has been difficult.
We both spent the past week trying to answer questions, and explain the format to people unfamiliar with it.

-mike
Axis Audio
Nashville, TN

Part of the issue is that Apple is using a different algorithm for the binaural processing than is provided by the Dolby Renderer. This makes it difficult as mixers are not able to accurately monitor their binaural spatial audio mixes in real time.

@ Mike@Axis is right that the Apple version of it has been a problem as it is a different presentation than what we are creating in the studio.

My view, which i've had some pushback on, is that I'm mixing for the full speaker setup and the binaural downmix is a convenience for users to hear "something". I strive to make the full Atmos mix magical and immersive and to have the binaural render be as close to the stereo mix as possible.

One pushback is that some producers / artists want the binaural render to be EXACTLY the stereo mix and this is not possible. Especially not possible on current Apple "headphone" render (not binaural because it is seemingly ignoring binaural settings from Dolby) and especially not possible with the future Apple "head tracking render" which seems to give everything a "mid" effect setting.

The other pushback is that some have the opinion that since most people are listening on headphones in binaural that I should make the binaural mix magical and immersive at the expense of the full speaker Atmos mix. (which ends up sounding like crap). Also not possible if different streaming services are going to apply different functions than what I set.

This is the polar opposite of pushback 1 , so can't please everybody. Also I'm hoping that as more speaker arrays, full Atmos setups, and better binaural decoding with proper HRTF and personalized HRTF come around that the mix for full speakers will sound better and better on headphones.

One factor to consider is that the spatial information is encoded in the sound file but the decoding is where the illusion comes from. That means as the technology gets better the mixes that you already made will sound better and better over time as the tech evolves

Hi Patrick, the surround sides being too loud with and present is something I notice on a lot of the Spatial Audio mixes that are on the Apple Music platform. And I'm talking about 7.1.4 speaker playback here. A lot of the mixes seem to have gone through some sort of template to get a lot of upmixed Atmos content go out on the platform te have enough material available. The other guess is that some of this Spatial Audio music is indeed mixed on and for headphones, compromising the speaker playback. But headphones are the majority of music consumption nowadays, so I can see why they would do this (though I don't like the results)... Greetings, Thierry

I think (and hope) this could be of interest to many of you.
Last week, we released Sienna Sphere, a plug-in suite (and System-Wide native application) for immersive sound that can convert immersive Bed tracks into binaural ones using them for headphone monitoring without the need for a complete multi-speaker setup.
You can install a 30-day free trial to test.



More info on this GS thread: https://gearspace.com/board/new-product-alert-2-older-threads/1367648-acustica-audio-presents-sienna-sphere-turn-your-headphones-into-atmos-studio-post15757859.html

Hope it helps!

I'm starting to think of binaural renders of atmos mixes as mere "backwards compatibility". I understand that isn't definitively correct, but from a functional view, it suits a similiar problem: "How do we make this new tech work with old equipment (HRTF aside)?". The ability to still work across such a wide range of environments is a big strength, and with my fingers crossed, one big reason we may see this protocol take root.

This article on pro tools expert I think is a must read for anyone doing immersive music on apple products:
https://www.pro-tools-expert.com/pro...on-apple-music

You want to stop other people from listening to spacial Atmos on headphones? All listeners of Apple Music using Apple / Beats headphones are receiving the spacial mix by default. There's a lot of money invested in this, so it's not going to stop.

The "binaural" render for headphones is a work-in-progress. It will keep improving as time goes on. Mix engineers are still learning. Hardware designers are still learning. Expect future headphones to do a better job on this than today's cans.

Just got my copy of "Dolby Atmos #1 How it Works" and am stoked to dig in!

Having never mixed on anything more robust than 2.1 - yet - I have to wonder how many mixing techniques get thrown out the window. One reason we need to compress and EQ is because we are trying to fit an entire rock band (with eight chainsaw-like guitar tracks, 16 drum mics, BG Vox, keys, reverbs, delays, etc.) into a little L+R image.

For those who are/have mixed in a multi-channel environment like a 7.1.4, do you find there is less effort to carve out EQ and compress signal when you have all these extra channels?

Is there any kind of standard format for placement? Or is it still kind of wild west - like the early days of stereo?

It is super exciting, I'm not doing Atmos music mixing professionally but here is what I do:
1) put my objects into a 7.1.4 bed for listening
2) have the reverbs "staged", so don't try to do all the reverbs in the 7.1.4 but I stage the reverbs doing some on the mono/stereo level, then a bit every time i mix a track stage together.
3) after I get a 7.1.4 bed sounding good, i listen to it only on the quad speakers with a simple mix down - i then go back to step 1 to fix things
4) after I correct things I see if the 7.1.4 to mix down stereo works - then back to step 1
5) I do a final mix down to mono to see if it's muddled (kind of like what your saying), which it almost always is - I'd say 80% of the time it wasn't the mix creating the muddle but the arrangement ... which requires a bunch of rethinking.

This is a kind of old school mentality for checking out beds, I have no idea if the current music mixers check their beds this way anymore.

Very cool approach and it makes good sense to me. Having always checked my mixes in mono - even when told "We don't care about the phase meter anymore because everyone is listening in stereo these days." - I always felt like there would be no harm done if the stereo mix sounded good in mono. Ten years ago, it did feel like my approach may have been dated, but now that we see how much tablets and cell phone speakers are used today, I'm glad I stuck to my gut instincts. I guess you never really know where your mixes will be played at, and considering today's digital works may be around as long as there is electricity, we should mix in a way that sets us up for success across anything, which may include some holographic-futuristic mono source 100 years from now.

I'd like ask about how you stage your reverbs. Are you staging several stereo reverbs that are panned in the beds to create a 3D room simulation? Rather than using - and I don't even know what reverbs are available for an "immersive" mix - a reverb designed for >5.* ?

Something a lot of people don't realize is that many people's first exposure to a new artist is their songs being played in mono at a concert. Mono remains a big big deal for artists.

For sure less EQ and compression. For the reason that you can just pan stuff around to discrete speakers and not have it fight with other competing elements (sonically).

There is more dynamic range available too as there as the same total energy gets spread over a larger array of speakers (the wording is off but you get the point).

Try Steven Wilson's King's Ghost. Compare the Atmos version to the Stereo one and note how much sidechain compression is going on in the Stereo mix to get that low bass and kick in the same space.


There is none, the song and arrangement will dictate your choices.
However, in practice:
You probably want to ground everything important to the groove in the front channels (bass, drums, percussion elements that make up the groove, guitars that provide groove, etc). Vocals usually go to the front speakers. Either to the center or to a mix of L C R (where it is louder in the center speaker). Labels sometimes can request to not stick the vocals in the center channel so that the naked vocals cannot be lifted easily.

And then there are more technically informed choices and principles:
If you pan multiple objects around all the time, the effect gets lost. Movement draws attention when it contrasts against a steady backdrop. Doesn't matter where you are panning to.

Use LFE sparse. Most systems have bass management which means that the regular speakers don't cover the full spectrum and rely on the subwoofer to fill in the gap below 120 Hz. So in practice, if you have a kick, everything below 120 will already go to the sub on a bass-managed system. If you now add additional sub to it, it can become pretty boomy quickly.

Not all speakers need to be used at all times. It is actually a very nice effect (again, contrast) to just rely on the front speakers and sometimes add something very distinct in the back. Or to "open and close" the soundfield by going from a mix utilizing all speakers suddenly to just fronts or even just center.

Difference between phantom center and discrete center-like channels. The latter position a sound explicitly on a spatial dimension regardless of your listening position, the former only are perceived as between two speakers if you are dead in the middle.

Very true, Dolby Atmos Music mixing is now going through a phase of technical novelty, where mixes are experimenting the same way mono went to stereo: drums left and vocals right, showing off that 2 channels is more than 1 :-) Eventually esthetics will become more important again...

I feel like there are three groups of people creating Atmos mixes:

Engineers who have done Surround (5.1 mostly) and now just simply upgrade to Atmos given that the requests they get are for Atmos. These have plenty of experience with the concepts of multi-channel immersive mixing. There is only an additional curve specific to the Atmos format (beds/objects mostly) but they have figured out a lot of the things already from working in 5.1 when it comes to placement etc.

Engineers who get into Atmos by using headphones. These don't really have experience and might even have never heard a 5.1 album in their life… It's the cheapest way to get in there but arguably gives the worst results too as it might not translate to discrete speaker setups that well.

Engineers who get into Atmos because they are requested to do so, work with discrete setups (ie: spent a lot of dough on getting their room specifically setup for Atmos), but have not done the 5.1 thing either and therefor have the same learning curve, but will generate mixes which probably translate much better.

Regarding mono: I can't remember the last time when I heard a band in mono in a concert… Any concert I've been to in the last 20 years was always in stereo (meaning L and R don't have exactly the same content) or even quad in the case of King Crimson and Steven Wilson.
Sure, the imaging is not the same for everybody, depending on where you are in the venue, but skewed imaging does not equal a mono mix for me.

Same here.

And even if it was mono I don't really understand what experiencing live in mono has to do with mixing in Atmos or other formats.

@ santibanks : those 3 types Atmos mixe(r)s are probably not far from the truth.

Regarding live mono: you'd be surprised how many live sound engineers actually keep a lot of the mix mono for obvious reasons.
I'm talking regular rock concerts with L and R main speaker setups.
Of course there are artists/mixers using the stereo/quad/larger systems. Kraftwerk and Pink Floyd come to mind.
And there certainly are new developments like d&b Soundscape and L-Acoustics L-ISA, with live sound discovering what cinema had been doing for a long time...

Regarding Mono and Atmos mixes: for me Apple gave a great example with one of their first demo's when Spatial Audio was released on the iTunes platform: they proved with a Marvin Gaye song that the mono mix was the best, better than the stereo mix, which in turn was better than the Atmos version (or did they try to tell something else ? :-)
The advantage of the mono mix in Atmos, was that it actually came out of one (center) speaker, just the way mono should be listened to! ;-)
This is not to say Dolby Atmos isn't good (I like it a lot), but sometimes things (like older music releases) should better be left untouched...

Cinema has been far more mono than most people realize. Everything important to the story, i.e. dialogue and Foley effects go front and center so every location within the theater can hear them clearly. The score goes front L&R. Surrounds are used for ambience and sound effects in a balance that can be 10dB. off in any direction depending on where a listener is seated.

Hi everyone, I'm gonna buy 7.1.4 Atmos system and starting from scratch a new studio setup but I need your pro advises please :

If Atmos is the new standard with less EQ and Dynamaics needed, do you think I still need to get an analog console I've ordered with 500 slots to plug Api Module or go fully digital like composiing/mixing directly my track in Atmos using only plugins (apart from the analog chain to record mic & instruments) ? And fold down the Atmos mix to Stereo and master this stereo file with an Analog mastering chain ?

If keeping the console with all 500 modules, that would mean mixing first Stereo, then bouncing all stems & import it on an Atmos Session.

Appart from prices, which workflow will be the best ?

As you see, according to the better workflow, I won't need the same Hardware setup for my new studio. I will be so happy to have your thoughts, pro experience. Thanks a lot

While almost all dialog goes center Foley goes where it needs to go. If it follows a character that's talking then it'll often go with the dialog, but if it's a character that isn't talking it may very well pan.

Score doesn't just go L&R, it goes into surrounds as well.

There are a bunch of modern directors that pan the hell out of dialogue too!