That software can simply convert everything is clear, but what do the inventors and really convincing sound engineers say to should solutions?
If the music industry is satisfied with it, it will only be the Frauenhofer story part 2. Actually nothing where the nerds will be satisfied with. That’s where the HiRes market is.
I’m not talking about music “converted” to atmos. That’s no better than the labels mass converting old rebook releases to MQA or HiRes and marketing them as such. Creative artists/recording engineers are beginning to create music with atmos in mind and will mix to it. That is where the real interesting spatial releases will come from.
Atmos is no more a fluke than when 4.0 came out, then 5.1…
There are serious mixing engineers mixing Atmos audio. The next wave is Atmos audio in High Resolution. Who knows 5.1.4 High Res? 192? 96 would be awesome. of course with all good engineers/mixers.
The Process Behind a Great Dolby Atmos Mix
In today’s livestream (8pm EDT), we sit down with David Frangioni to discuss the processes that go behind making a great spatial audio Dolby Atmos surround sound mix. We talk about why some Atmos mixes aren’t as immersive as others and how to best configure your system to take full advantage of spatial audio for music and movies.
David Frangioni is an award-winning veteran of the music and audio-visual technology industry, with expertise ranging from being a drummer and producer, to an audio consultant, technologist, integrator and recording engineer. He’s worked with such artists as Aerosmith, The Stones, Ringo Starr, Elton John, Ricky Martin, Steely Dan, Sting, Bryan Adams, Journey, Styx, Phil Collins, Shakira, Rascal Flatts, Cher and Chick Corea. In the A/V world, he led Audio One to more CEDIA awards in a given time period than any other audio-visual system integrator in the USA and in the music business, he has worked on dozens of Gold & Platinum albums.
Watch the interview on Audioholics Patreon page.
Pretty convincing evidence that Spacial Audio is here to stay, just like 7.1.4
I’m glad we survived the Mono to Stereo wars in the last century. LOL
The reason your thinking is to conservative is that you need capital to make all this work, and the commercial side will drive the 95% that can’t tell the difference, but the rest will benefit from this.
I don’t understand what’s necessary here. Is it sufficient to have Roon send the 7.1 multichannel stream to an Atmos-capable AVR? Isn’t that something it can already do?
Probably worth noting that ATMOS is not hard coded to any specific speaker layout as previous formats like quad or non-atoms 5.1 mixes were. ATMOS is a object oriented format that specifies where in space sound will come and at present It can support up to 64 speaker configurations (as an example the trinnov altitude a/v processor can currently support 32 speakers and is designed to be expanded in the future to 64) and when you play an ATMOS encoded recording your a/v processor will – in real time – do whatever is necessary to play optimally regardless of whatever speaker setup you have without loss of any sounds. Additionally, if you listen to ATMOS mix with headphones from an apple device that supports apple’s spatial audio, it will process the ATMOS mix into an optimal binaural spatial format (or alternatively there is a dolby binaural format that some devices support).
It’s my understanding that the only way ATMOS recordings can be delivered with full functionality today is via an HDMI connection. If you try sending an ATMOS recording via optical it will be converted to 5.1.
So to get back to the original question in the previous post, in order for ATMOS to work optimally with Roon - at minimum, you’d need an ATMOS capable A/V processor, a Roon core machine with an HDMI output, the music services Roon accesses needs to have full ATMOS recordings available and Roon would need to be able to tell the music service to stream the ATMOS recording not the normal HD or SD mix. I suppose it would be possible to have an HDMI connection on a Roon enabled endpoint that could receive the ATMOS stream from the Roon core and send the ATMOS stream via HDMI to an A/V processor but I don’t know of any endpoint configured with HDMI. But none of this works today. I’m not sure exactly where the roadblocks now but I am sure that ATMOS spatial recordings are here to stay and Roon has to be able to support ATMOS playback if it is to remain relevant in the high-end audio world.
Thanks, Craig.
The latest Archimago post reviews one:
I was wondering if it would work for this application.
Maybe this is the real issue.
Probably, but the roadblock at this point is Roon being able to access and stream full ATMOS files as you suggest. That needs to get solved first. When that does, then devices like this would likely work. In my case I have a nucleus that already sends non atmos multichannel music that Roon can access to my AV processor via its HDMI output so I wouldn’t have the need for a hdmi enabled endpoint but many might find it useful in their setup.
So the format isn’t simple multichannel, eh? Is the format documented somewhere, or is it proprietary, do you know?
I noticed that one of the images in step 10 of that Archimago post is labelled “Notice BTW that my output channel layout is 5.1.2. The receiver is re-mixing to the Atmos format with Dolby Surround Upmixer from 7.1.” So I figured it was something the receiver could do.
Proprietary dolby. You basically use the format to put a sound somewhere in virtual space. A simple example, would be your specifying a singers voice comes from the front center. Then when you play the recording, if you have a center channel, your ATMOS enabled A/V processor puts the voice in that speaker. If you don’t have a center channel the voice gets mixed into the left and right channel in an appropriate fashion. Don’t know where the documentation for the format is but of course it exists.
And yes, most A/V processors can upmix almost anything to full playback from all speakers to crudely simulate full ATMOS, and sometimes these sound surprising nice IMHO, but it is a rather poor approximation to what you would get from a properly mixed ATMOS recording decoded properly.
Why couldn’t roon just hand off the stream to the endpoint ala Tidal connect for Atmos?
Given how easy it is to playback Atmos using Apple Music and Tidal, I find it hard to believe Roon are either going to require an HDMI connection or not offer Atmos streaming.
Its my understanding (I could be wrong though) that the ATMOS stream Tidal will deliver is a binaural mix for playback on headphones. That would be crap in ATMOS home setups. I believe you can get a non binaural ATMOS mix for a home ATMOS setup via the Apple 4K via Apple music but I haven’t tested that yet. And ATMOS currently can not be delivered to a A/V processor via any other connection than HDMI. This is not something Roon can change.
Huh, that’s pretty awful on Tidals part.
Atmos is going to have a tough time gaining adoption if they put arbitrary hurdles in the way of streaming.
Yes, I can see that it would be. I was wondering if the object metadata Atmos uses could be carried by metadata records in, for instance, a FLAC stream.
I do believe I was wrong in my earlier post. If you are running the native Tidal or Apple Music with the right subscription, and have the right device (Apple TV 4k for both, or devices like fire tv for Tidal) connected to to your A/V processor you will get ATMOS delivered properly for your home setup. What doesn’t exist today is for something like Roon to be in the middle of that flow like it is for non ATMOS music and still have the proper ATMOS flow work. That’s the key thing that needs to be solved.
OK in that case Room can just pass the endpoint the Tidal URL presumably I.e. like Tidal connect.
That makes more sense, thought Dolby were shooting themselves in the foot 
Roon never just passes a url. RAAT passes pcm, dsd or pcm with mqa signaling info to Roon ready devices. It would take a major update of RAAT to handle ATMOS like there was for MQA IMHO or for another class of non RAAT devices to be supported like Airplay for example…
Atmos is very much documented. It is fixed channels (up to 7.1 aka bed) with additional audio objects placed in 3D space. HDMI 2.0 support up to 32 channels in 1536 kHz (1536 / 32 = 48kHz, or alternatively 1536 / 192 = 8 channels. The audio objects are in lossless and can also be of different size and position, also they can move in realtime. Atmos is not basically a channel based system, it plays back using all speakers you have in your setup to reflect the audio objects position and size. Netflix, Disney+, HBOMax etc all support Atmos streaming. Atmos can be based on lossless Dolby TrueHD 7.1, or lossy Dolby Digital+ 5.1 - all alernatives have x number of dynamically moving audio objects with different size (size can also change over time, not only position).
For Roon we want Atmos over HDMI and for the Roon Ready endpoints (for example Trinnov Altitude which support multichannel audio over ethernet).
Think of Atmos as “drawing” or rendering audio in 3D space with all alternative speakers you have in your system.
Edit:
DTS got their own audio object based alternative called DTS:X, it is very much the same but with in-ground speakers and center ceiling speakers in addition (including Voice of God, directly above - also called 0h or T0). Hence DTS:X is technically better, but very few mixes uses dynamically moving objects - they are very often fixed to specific position to reflect 7.1.2 or 7.1.4 (Dolby Atmos mixes did same concept before, but now they have evolved so much more mixes are using the full potential of the platform - this is most likely connected to software used for mixing).
I wasn’t saying it wouldn’t require an update, just saying that it is possible 