Dolby Atmos for the rest of us...

[mgw38]

In recent years people have chosen to consume video from inferior devices (phones, tablets) just for the convenience. Installing 12 speakers is the opposite, so niche. But the compatibility with other setups is surely an appeal in Dolby Atmos.

The fact that Atmos combines channel based audio (i.e. the Atmos "bed") with object based audio (i.e. the Atmos "objects") ensures full backward compatibility without loss of quality. This makes it exceptionally interesting. It honestly took me a while to realize that.

People kept asking me about the mp4 encoded version of Atmos (i.e. Dolby Digital Plus with Joint Object Coding, or DD+ JOC for short). For the longest time I did not get what the advantage of that format is. But what it does is encode Atmos as regular channel based 5.1 audio (which plays back perfectly on any consumer system or device, including phones) with additional meta data that allows Atmos systems to reconstruct the object based Atmos experience.

That way the audio will play back perfectly regardless of what the system is capable of. However, very few producers or mixing engineers have yet understood how to take full advantage of the possibilities this offers. At this point most Atmos mixes are just fancier versions of panned stereo.

[AudioBabble]

It's certainly a very "smart" technology... and I should probably mention - lest Dolby get all the credit - that the main drivers behind the research and implementation of the standard for object based audio were these guys in France around 2015-2018: https://www.orpheus-audio.eu/
[and apparently they received further funding in 2020]

What I find intriguing is the fact that the individual tracks (or objects to use the correct parlance) can be extracted from an ADM WAV file - up to 128 of them... plus the discrete output channels.
This would logically seem to suggest that an ADM WAV file could be HUGE in terms of data size, depending on how may "objects" are defined... if I understand what I've read so far.

Unfortunately I don't have any ADM encoded WAV files to play with just yet to find out if this is true.

[EdgarRothermich]

May I chime in?

About the question whether Dolby Atmos is a gimmick or following the same fatal footsteps as Quadraphony or DVD Audio? Absolutely not! It is a total different playing field and the original obstacles about where put all the speakers (and how to pay for it) is irrelevant for the end consumer. The de-facto listening environment is headphones, and binaural rendering of Dolby Atmos (or any immersive format) will do the trick. It will get better with personalized HRTF (where a lot of big companies pumping a lot of R&D $$$ into it at the moment).
Soundbars and Smart Speakers will do the rest for the living room.
There is quite some momentum in the professional studio scene at the moment to get on the Dolby Atmos bandwagon since Apple Music joint the party in June 2021 (and Logic 10.7). With Sony's own immersive contribution "360 Reality Audio", it is getting already crowded and the headache increases even more for engineers to figure out how to handle all that.

I can provide a few links with valuable information about Dolby Atmos.

Here is the video "Mixing in Dolby Atmos - #1 How it Works" with lots of information about Dolby Atmos from general concepts to more details about the technology


It is the companion video to my book, the first Dolby Atmos book "Mixing in Dolby Atmos #1 How it Works"
http://dingdingmusic.com/-titles-/dolby-atmos-1.html

In my other YouTube video "Logic Pro - What's New in 10.7" I provide more details about the Dolby Atmos integration in Logic.


There are also two articles I have written for Logic Pro Experts where I discuss topics about Dolby Atmos, especially Apple's Spatial Audio aspect with Dolby Atmos.
https://www.pro-tools-expert.com/produc ... pple-music
https://www.pro-tools-expert.com/produc ... headphones

And here is the Podcast Episode "Dolby Atmos for Music" talking about that topic
https://podcasts.apple.com/us/podcast/p ... 0543612967

Enjoy

[EdgarRothermich]

What I find intriguing is the fact that the individual tracks (or objects to use the correct parlance) can be extracted from an ADM WAV file - up to 128 of them... plus the discrete output channels.
This would logically seem to suggest that an ADM WAV file could be HUGE in terms of data size, depending on how may "objects" are defined... if I understand what I've read so far.

Unfortunately I don't have any ADM encoded WAV files to play with just yet to find out if this is true.

Yes this is true.
ADM BWF is technically a wav file, that's why some people are confused and try to play it back on their audio player.
However, the ADM (Audio Definition Model) is kind of an extension, also referred to as NGA (Next Generation Audio) because it is an audio file with much more data (metadata) for various purposes. ADVM BWF is not proprietary to Dolby but the "Dolby Profile" of their ADM BWF version is. Different manufacturer can create different flavor to include their set of metadata. In this case the ADM BWF file representing the Dolby Atmos Master File, it includes all the metatdata for the Dolby Atmos mix and the description what to do with all the audio data embedded inside the wav file.
And yes, the ADM BWF file is an interleaved audio file that can include up to 128 audio channels and that can get quite big (GB or even TB if you talk about full 2h movie mixes).With the exception of the Bed channels, all the other Object channels don't contain any pan information, which is stored separately as metatdata and the Renderer during the playback puts everything together based on the available speaker setup.

[jdnz]

Lfe speaker is less critical. In an Atmos setup the lfe speaker is not used as a subwoofer and thus has less relevance for Atmos music.

I'm trying to get my head around this aspect... I had a look at the dolby website. From what I understand then: the LFE is for EXTRA LF info (<120hz) that may or may not be present in the overall mix. Further, there is no crossover in a conventional sense for the rest of the speakers - i.e. they all receive full frequency, not just >120hz?

Do I have that right?

that’s true in terms of Dolby’s spec ( which was originally aimed at theatres where obviously you’re going to have large, full range speakers all around), but in the consumer space you need to watch out, I’ve got a Yamaha atmos ht setup, and the receiver’s bass redirection DEFINITELY applies even when it’s decoding atmos ( i.e. everything under the cutoff freq I’ve set gets redirected to my sub)

makes sense in the consumer space as a lot of people will have speakers so small they’ll struggled to get down to 50hz, let alone 25hz

[jdnz]

But audio streaming services are starting to accept Atmos Music and systems such as Apple’s spatial audio can take advantage of the object oriented music format with fairly standard consumer gear.

Which, if you think about it, kind of reduces the whole thing to a bit of gimmick in some ways...

I checked out some dolby atmos mixes myself on Amazon HD with normal headphones. All I noticed was a nice stereo "spread".

That is because producers have not yet tapped into its full potential.

those of us who are old, grey haired and cynical will remember when quadrophonic was the ‘next big thing’. Just like atmos there was very little quadf music releases that actually made any meaningful/sensible use of it, the only thing I can remember that was a good quad mix was ‘Dark Side of the Moon’ ( which was trully impressive )

the majority of music is now consumed via headphones - I just can’t see immersive/atmos music releases being a big draw card

[AudioBabble]

Thanks @Edgar for chiming in! Lots more info to soak up, which is more than welcome

From a purely consumer perspective (personally speaking), as I said earlier I haven't exactly been blown away by what I've heard so far on normal headphones... however, the head-tracking addition could make all the difference.

From what I've read, this is a way of "re-enforcing" the illusion of 3D space so that an object can appear to remain fixed in space relative to head movements. So actually if one were to remain absolutely still there would be no difference to normal headphones.. but of course that's unlikely.. well.. unless it was meditation music!

I don't know if this is the case, but I can imagine consumer earphones like apple's including a miniaturized headtracker.. if they don't already.. bringing the 3D experience to the every-day level.

One thing I have been mulling over though is when 3D sound is useful and when it's not.

One could argue that all sound is 3D in reality and therefore this is just a step up in terms of realism. This is true for certain applications, for example classical performance recordings. But realism isn't the only goal in recorded music by any means.

Personally, I think a 3D mix could be distracting for some kinds of music, for example straight-ahead rock'n'roll or blues. I've seen that many current "pop" acts seem to have 3D mixes out now... I baulk at actually listening to them, but can imagine this is more of a bandwagon thing than anything else.

Obviously I'm talking about music here... film, gaming theatre and installation art are a different matter... but I feel like the kind of music that can benefit the most for both creators and consumers is that which has an extra-dimensional quality in the first place... "dark side of the moon" or "the wall" would be great examples. Any kind of "head" music I suppose, EDM for sure...

That said, in much the same way as Hendrix, The Beatles, etc made creative use of stereo, I could imagine certain techniques in the 3D field becoming commonplace, for example moving reverbs around in 3D space, moving a mix from a focused position to a "surround" position for dynamic purposes during a break or intro section, etc, etc

Is 3D to stereo what stereo was to mono? Time will tell... but I definitely agree that it is up to producers to make the most of the creative opportunities it offers.

[AudioBabble]

I’ve got a Yamaha atmos ht setup,

My ears pricked up there... tell me more - was it expensive?

[jdnz]

I’ve got a Yamaha atmos ht setup,

My ears pricked up there... tell me more - was it expensive?

depends on what you call expensive - compared to my previous 5.1 setup not a lot - it was just a case of adding the height speakers ( in-ceilings ) and reconfiging the amp so it knew it had an atmos speaker setup to drive.

In absolute terms - the amp itself wasn't bad ( a basic yamaha like the rx-a2a will do - nz$2K - the lower end/cheaper Yamaha rx-v series don't have enough outputs to do dolby ). The real cost was the speakers - which is one of those 'how much $$$ do you want to throw at it' questions ( centre speaker plus 2 matched pairs for front/rear, plus a powered sub, scouring the used market you could do it quite cheaply, going high-end and new you could blow a fortune )

[AudioBabble]

A nice, relatively affordable setup. Interesting how the room is intentionally reflective (you can hear it in the presenter's voice). I wonder if the room could be said to be genuinely dual purpose however as I wouldn't want that much reflection in a stereo mixing environment.

You might want to skip through the entirety of "rocketman" or not - depending on your preference!

Still fairly drool-worthy however

[AudioBabble]

mind you I just realized it;s just a demo room

[ThomasHelzle]

I was involved in two stereoscopic movie productions when the last hype wave started ~2009/10.
So I learned a lot about how S3D works. Other than good surround sound, the visuals are a fake, since our eyes rely on much more than stereoscopy for depth perception (and a significant percentage of humans actually doesn't have it at all) - and the images still are all on the same screen plane, so our eyes can't for instance zoom onto a detail, since the distance is the same for all of them. Many people just get a headache.
I saw some good (for instance "Up") and some very bad ("Avatar") S3D movies and by working on two myself came to the conclusion, that for me personally, it doesn't really add anything important to a good movie, although it can be nice to have, and it can be quite annoying in a badly made one like Avatar.
In the end it's the story, the acting etc. and depth is like a 2% bonus maybe if done well.

Similar things apply to surround sound in music IMO, especially in the way it' usually done ATM.

I have a 5.1 setup since I love to sit "inside the music" which already works with a quad setup.
I let things wander around me, love to move effects as well as instruments, but with most music, I don't see it as much as a real necessity as in movies, games or installations, simply since an emotional song works in the cheapest mono radio almost as well as in the most perfect surround setup.

The question that is missing for me in most discussions: What's the actual point?
Is it just to add something new because we can and it sells units or is there an actual use and benefit to it?

When we listen to music, the musicians usually aren't jumping around us, we don't have sounds zip left and right.
Music is mostly something that happens in relatively relaxed secure environments.
The position of instruments isn't really that significant.
It's the melody, the rhythm, the sounds...

Where spatial hearing really shines is when you are walking through a forest and can instantly hear and locate the breaking twigs a predator treads on, or the rustle of leaves where some dinner may hide.
So that's where movies, games and some installations can really benefit, since you are in a location aware situation (and many movies and games have topics where those ancient senses are triggered).

But with music - again, in my personal view - it's more a "nice to have" if some sounds are moving around, but in most genres the best you will gain is probably better spatial resolution and some feel of roominess, which already works rather nicely with a quad or 5.1 setup and even things like open baffle speakers etc. can provide some of that impression with stereo.
So that's why I'm rather sceptical if it will make a big impact in music.

And binaural sound with headphones - although I use it myself - isn't anywhere close to a really full surround experience. Front/back perception as well as up/down are rather weak in all the systems I tried so far (with non-customised solutions).

So other than the technical excitement and interest:
What do you guys see as the deeper benefit or possibilities for higher-channel-count music?

Cheers,

Tom

[fairlyclose]

Following ThomasHelzle I worked as an external advisor/researcher with a spatial audio research centre at a university in Australia for many years and there are limitations to what is possible, some of which are perceptual and psychological - of course those limitations can be exploited for creative benefit too.

But for example when we turn our heads whilst listening to sound cues and looking at the sound generators there is a temporal mismatch in how the brain integrates that information and how it is delivered through the tech. The sound and vision come from outside the body but the movement information is generated within the body - and there is a lag that cannot be overcome. Hence motion sickness (severity varies between people).

Also having prominent sound appearing from behind elicits an orienting response - you turn to see what it is. That is something to exploit - I did for a piece in 2006 using a 20 channel system at this venue (fantastic architecture) https://fedsquare.com/venue-hire/the-edge Some people really did not like that I had done that, others really did.

With all of this it is worth thinking of sound as an adaptive sense developed through evolutionary history - for example, what is it that we are wanting to hear above us, what are the frequency ranges. And then in terms of integration of all the sensory systems, with movement vision and sound.
And then keep in mind that displacing and reassembling sensory cues in foreign contexts is also interesting eg ASMR So we can play with our evolutionary histories as well.

[mgw38]

The lag is no longer an issue these days. This is one of the reasons why spatial media tech has more promise today than it had only a few years ago.

[fairlyclose]

The lag is no longer an issue these days. This is one of the reasons why spatial media tech has more promise today than it had only a few years ago.

That is good news. How did they solve the lag between movement and sound/vision?
EDIT a quick scan reveals cybersickness is still a thing. I struggle to see how it can ever be completely overcome given the sensory mismatch is hardwired into our bodies - you cannot get the processed sound and vision perceived at the same time as the vestibular cues of motion. The media content can be designed so it doesnt intrude as much or as often ie mismatch occurs less often and less strongly