ATMOS - Differences between Beds and Objects

Hi all,

While there is a lot on either one, and some posts have some explanation as to one difference or two between Atmos Objects and Beds, I thought it would be nice for you experts to post on here in more detail the differences in these two things used in Atmos mixing. And possibly where the use of one over the other is better for best results. I see some confusion still on here on this, so hopefully this post can serve to explain / clear up any confusing things.

I have been mixing 99% in Stereo, and have limited experience in surround formats, due mostly to my clients simply being uninterested in these, I do want to learn this new Atmos format, especially since Apple has decided to put their weight behind it (for now), so I might be able to actually mix and release both remixes of older albums, and also experiment with Ambisonic recordings of new material, alongside traditional recording methods for Stereo releases in the near future.


Fundamentally, objects/beds work the same across DAWS when mixing for Atmos, however, it makes a great difference on how quickly you arrive at mix nirvana - what type of music are you planning and what’s your setup like?

Not to rant, Dolby and Avid/ProTools make it ridiculously complicated for the average musician/producer to work between the two products.

Over the next couple of week, I’m putting together a post on my tried and tested predictable Atmos workflow that includes:

  1. PC spec and configuration
  2. Nuendo 11 DAW configuration
  3. Audio A/D ASIO routing
  4. Exclusive monitoring in headphones
  5. Ambisonic setup for headphones
  6. Track layout/template
  7. Best practice use of objects/beds for types of audio, instruments, and returns
  8. Mixdown to stereo, binaural and other surround formats
  9. Verification testing in Mac based Dolby Production Suite
  10. Verification testing through consumer devices
  11. Package for AvidPlay distribution
  12. Provide complete example Nuendo project

Thanks! I am sure that will help figure out some setup thing.

The already released material is Country / Bluegrass. Taking 24 tracks or less and remixing / reimagining each song into an Atmos experience. We are still thinking about doing this for certain. As far as new material, it would be electronic music, experimental stuff, with more movement in the space than traditional music. More creative leeway mix wise.

My setup eventually will be a full set of 7.1.4 speakers (JBL), plus a few different sets of headphones for mix checking. Currently on Nuendo 11 using Focusrite Dante I/O on Macs.

We would be doing separate Stereo mixes, and then move onto the Atmos mixing after that. I have not ever found an “automatic” foldown to Stereo that is acceptable to myself nor my clients.

Personally I think that Atmos is actually the least ideal immersive format when we talk strictly about modern music production. (… some kind of "Beta >< VHS situation). The “Object” approach is great for media work, but more or less incompatible with the workflow we are used to in music production nowadays, where mixing and mastering are integral parts of the artistic process, and not so much about purely “technical delivery”. The minuscule details we usually take care for when it comes to positioning, balance, compression and/or room are now quite arbitrary “meta data”-based decisions of a rendering device.
… what’s more - how is a mastering engineer supposed to deal with dozens of moving “Objects”, when there is no actual “mix bus” anymore? The methods we tried-end-tested for more than half a century now are not applicable.
My personal decision: For music I try to avoid any kind of moving “Objects”, but I mix for a channel-based delivery format instead that gets encoded into “Beds” (or fixed “Objects” with size set to 0) as a last step, after mastering has taken place.

… would be interested to hear any contradicting opinions, though. :slight_smile:

Well that… does not answer my question. But interesting nonetheless. Thanks.

Well, I apologize — I thought I was responding to this part of the question:

1 Like

No need to. Any info can help, although it would be nice to get technical info on the title of my post.

Traditional mastering screws up many of the benefits of immersive Atmos mixing. It really is a new sonic world. To get the most out of it, you do have to let go of many traditional mixing and mastering techniques.

Thx but that in no way answers my question. Cheers

Beds are 7.1.2 and objects can be panned in 7.1.4. The rear high speakers really make a difference.
And another benefit of objects over beds is that you have additional settings for near, mid, and far for binaural. But remember, for music most people by far are listening to the binaural version on headphones and not an Atmos surround speaker system. Dolby didn’t license that functionality to Steinberg or any other company. I got tired of trying to overcome that severe lack of functionality, I just purchased the Dolby Atmos Mastering Suite (DAMS) to work along side Nuendo. Plus the hardware to accomplish porting 128 channels of audio from Nuendo to DAMS. Gets expensive very quickly.

1 Like

Thanks Paul, that makes more sense. Yes I am hearing from other pros that there is a raging competition between Dolby and a large label… concerning the format. There is a lot of weirdness flying about. I just want to figure out the technical stuff so I can offer it to my clients as an option, see one of my above replies.

I would suggest doing the Atmos mix first, folding that down and tweaking your stereo mix from there. Certain very good spatial qualities can make it down into the stereo downmix. So much so there are times when I think I left the surround speakers on! A friend of mine experiences the same thing. If you are going to move into Atmos mixing, make the most of it, it really is special. It has its issues such as downmixing from 7.1.4 to 5.1. The sides of the 7.1.4 get lost in 5.1, I hate that. That is why you need to mix back and forth between different downmixes, you can strike some balances that are special too, BUT IT IS WORK!!!

1 Like

Well said Paul!

Hi Paulwr,

Not too long ago, we mixed to Mono first, for radio and PA systems such as Muzak (still in use in some form at public spaces near you), and then moved that final, approved Mono mix up to stereo for Vinyl records, cassettes, then compact discs, anything with the Stereo format. And once Quad and then 5.1 came around, we upmixed to those, again, from final approved Stereo mixes. the results were better.

I prefer that method, over having to begin from a 7.1.4 (12 speakers) format and then have to fight my way down back to a simpler format. Seems as entirely too much work, and too much of it uphill.

But, you do as you wish. This Atmos thing is, once again, The Wild West…

I have spoken with a few peers about Atmos, and their approaches, and nearly all of them work as I wrote on here, up from a simpler final Stereo mix. One of their main issues is the fight between Binaural and Atmos, and moving between the two formats. Atmos is really for a speaker situation, Binaural is for headphones. I think I will end up making different mixes for each format, instead of trying to “fit a square peg in a round hole” type of situation. Even though it may take longer, I will not be second guessing anything, and not have to backtrack to adjust a different format just so foldown algorithm is happy.

Ps, yes, I still begin mixing in Mono, to this day. :wink:

Yes, we are certainly in the Wild West with Atmos and really the immersive movement in audio in general. Standards outside of Theatrical are not really in existence yet.

And yes, Atmos speaker vs Binaural (BIN) foldown can be a bitch. But if your are doing primarily music delivery, you don’t want to leave your BIN mix untweaked from a downfold anyway. I forgot if it is Tidal or Apple, but one of them you can submit a separate BIN file from the Atmos mix under the same release name. Make very sure they are the exact same length. They all should adopt that. Meanwhile, it is still good to keep in mind that over 90% of listeners only use headphones and even stereo speakers to listen to immersive music. It’s a good reason the strike a balance between stereo and BIN mixes.

1 Like

For me, the difference of Atmos beds vs. objects in a nutshell is:

The 7.1.2 bed has fixed positions. The (remaining) 118 objects can be placed, panned or circulated freely in the 3D space around the listener.

Beyond that:

From my listening experience, the Sp0tifies, Amaz0ns use a kind of pimp-it-up -to-3D approach and lable it “3D”, “binaural”, “360” or even “Atmos”. Kind of a proof for that is, that they offer it only on smart phones (-> headphones or stereo speakers only). On a PC or Macbook with an Atmos-AVR connected, this sound option is not visible.

Tidal offers true Atmos, e.g. via their app on a FireTV 4k. Listening experience is so la la.

As many people above kind of mentioned, the real value on Atmos comes - and here it is better to say “immersive audio” - if you utilize the 3D capabilities right from the beginning. It gives an additional grade of freedom and creativity.

Sure, it does very likely not provide this additional value to all kind of music. Like a painting artist, that does not use all colors and brushes for a new painting. But it is good to know, to have it if needed.

LG, Juergi

1 Like

Thanks juergi,

Closer to what info I was looking for.

I am going to be experimenting with all this stuff until I have a good handle on it, so the more tech I know about the format, the better. Cheers!

I’ve been looking into that same question. My understanding:

The obvious difference is that the channels within a bed are mapped directly to speaker channels and can’t be panned (a stereo bed maps to the front left and right speakers etc). Objects can be panned anywhere in the 3D space.

From Dolby’s documentation:
“Depending upon the position and size metadata applied to an object, objects and bed channels can be sonically identical. For instance, an object placed in the left front with size set to zero will be identical to placing the audio in the Left channel bed.”

That’s a rich statement. “Depending upon the position” means that an object panned between speaker positions will not sound the same as a mono source panned as a phantom image between channels in a bed. “[depending on] size metadata” means that an object allows you to change the apparent size of the sound source within the 3D space, by cross-propagating the signal into neighboring speaker channels and doing other acoustic tricks.

If an object is panned to a speaker position and If the Size is set to zero, it will sound like a channel in a Bed.

I wrote more about this on my last video Update: Setting up surround reverb in Nuendo Atmos with Cinematic Rooms - YouTube

1 Like

Thanks leobro.

My takeaway is that I have to test as well, to determine the best solution for a given instrument or sound.
Bed = No panning possible. Object= Can be panned freely.

Good day @noeqplease

Going back to my original response, and summarizing a bit - Specifically for Atmos, 7.1.2 beds are best for either, existing surround audio or surround capable returns that are processing 7.1.2 audio and mapping appropriately to the speakers in the bed. So, technically good panning is possible in the bed, if you are driving it with a 7.1.2 capable plugin. I do it on every project using Cinematic Rooms Professional, and Slapper (delay).
My basic setup is this in songs with less than 50 tracks:
1 - Every instrument is an object
2 - The bed only contains existing/mixdown surround audio and FX returns that are processing audio with 7.1.2 out.
3 - In net new projects, I start from 7.1.4 and mix down. Once you start delivering to the various music streaming services, and FireTV/Roku apps, you will find your workflow will demand you work in this manner in order to manage through the fixing binaural vs Atmos differences.

Anyone starting net new in stereo/mono are still adjusting to the “Wild West”.