Dolby Atmos - Speaker vs. Headphone playback

[JimofMaine]

More so than any point in audio history...it deserves to be recognized. It clearly depends on subjective listener criteria. And improvements are coming soon to headsets mixes unlike speakers. What device/DAW and what headphones are you using to hear Atmos mixes. The point is is it's not the headphones creating spatially but the technology.

 

[zdjh22]

With a personalized HRTF and a measured headphone, what the Smyth does in hardware should be possible in any software binaural renderer.

I use both an A16 and a software binaural renderer (Virtuoso, including a head tracker). Getting a personalized HRTF for Virtuoso is not easy, and I haven’t done it yet (working on it). There’s no comparison, but perhaps with an HRTF Virtuoso (and other software renderers) will come closer. They aren’t there yet, except for perhaps Sony 360 RA with an in-studio measured HRTF.

The A16 measures not only your HRTF and your headphones, but your sound room as well. It is uncanny how closely the sound out of my A16 matches my physical speakers.

 

[DrKlahn]

I use both an A16 and a software binaural renderer (Virtuoso, including a head tracker). Getting a personalized HRTF for Virtuoso is not easy, and I haven’t done it yet (working on it). There’s no comparison, but perhaps with an HRTF Virtuoso (and other software renderers) will come closer. They aren’t there yet, except for perhaps Sony 360 RA with an in-studio measured HRTF.

The A16 measures not only your HRTF and your headphones, but your sound room as well. It is uncanny how closely the sound out of my A16 matches my physical speakers.

I'm open the idea of headphones being able to create a comparable experience, but I'll point out again that I would bet the majority of Atmos headphone users being mentioned in this thread are not anywhere close to this level.

Myself I regularly have some friends over for some drinks and listening to the latest acquisitions, so to me even if I could invest in the right equipment to properly experience Atmos on headphones there's just not much appeal for me. But if there is for others, then I'm all for them enjoying immersive music how they want to.

I do feel that the mix engineers should be evaluating their work on both to make sure the experience is optimized for everyone.

 

[jimfisheye]

This might sound a little shallow but even half-assed mixes tend to sound just fine on headphones. Only fully finished mixes work across systems with speakers in rooms. There are many shades of grey as well as the opinion that not working on speakers in rooms is a moot point! But this is a thing. Mixing on headphones is usually a huge mistake because your mix sucks on everything else!

 

[zdjh22]

I'm open the idea of headphones being able to create a comparable experience, but I'll point out again that I would bet the majority of Atmos headphone users being mentioned in this thread are not anywhere close to this level.

Myself I regularly have some friends over for some drinks and listening to the latest acquisitions, so to me even if I could invest in the right equipment to properly experience Atmos on headphones there's just not much appeal for me. But if there is for others, then I'm all for them enjoying immersive music how they want to.

I do feel that the mix engineers should be evaluating their work on both to make sure the experience is optimized for everyone.

My listening room is limited to 7.1.2 because of my AVR and where it's possible to place speakers. I definitely enjoy that physical speaker Atmos experience, and like you I've had friends over to listen to my latest acquisitions. More often it's trying to convince friends that multichannel music is worth the investment, and that most of us missed out when quad was around and readily available.

The A16 experience for me is exactly the same as the speaker experience, down to the sound moving as I turn my head, and that's the common experience of other users. The magic for me comes from the fact that I can measure my 7.1.2 layout in different ways, for example, sitting backwards or even on the floor (or on the floor backwards). The result is a greater number of virtual Atmos speakers - my speakers, but just in new physical positions - and I typically listen to 10.1.6 virtual setups. I also now have a true quad setup, where my large L/R speakers are used for my virtual Ls/Rs. Further, commercial A16 PRIRs (in the A16 world, a PRIR is a combination of an HRTF and an impulse response measurement) for speakers far more expensive than I would buy are cheaply available and let me experience Atmos with different physical (but virtual) setups.

An A16 is expensive, but it lets you use Atmos, PCM, dts:X, and Auro3D virtualized physical layouts that would be far more expensive to actually build. A number of other A16 users and potential users have told me they can't listen to a full blown physical Atmos system because of various limitations (in an apartment building where they can't play loudly, in a dwelling where speaker layouts aren't easy to accomplish, etc.). For them the A16 is a godsend. We all hope that cheaper software binaural approaches will eventually match the experience.

As to the question of mixing on speakers vs headphones, the target audience for the original A8 and for the A16 is audio engineers who want the option to mix on headphones (maybe not exclusively, but on occassion when the engineer is away from the studio and needs to work on something). Given that both can provide virtual sound from any particular set of measured speakers and speaker layout, a portable replication of a studio is possible. Particularly for sound mixing for movies I believe the A8 has been popular. Both are professional bits of gear that happen to be great for multichannel audio enthusiasts. Look up the Sony 360 RA environment - it seems that Sony also believes in this application.

 

[DuncanS]

My listening room is limited to 7.1.2 because of my AVR and where it's possible to place speakers. I definitely enjoy that physical speaker Atmos experience, and like you I've had friends over to listen to my latest acquisitions. More often it's trying to convince friends that multichannel music is worth the investment, and that most of us missed out when quad was around and readily available.

The A16 experience for me is exactly the same as the speaker experience, down to the sound moving as I turn my head, and that's the common experience of other users. The magic for me comes from the fact that I can measure my 7.1.2 layout in different ways, for example, sitting backwards or even on the floor (or on the floor backwards). The result is a greater number of virtual Atmos speakers - my speakers, but just in new physical positions - and I typically listen to 10.1.6 virtual setups. I also now have a true quad setup, where my large L/R speakers are used for my virtual Ls/Rs. Further, commercial A16 PRIRs (in the A16 world, a PRIR is a combination of an HRTF and an impulse response measurement) for speakers far more expensive than I would buy are cheaply available and let me experience Atmos with different physical (but virtual) setups.

An A16 is expensive, but it lets you use Atmos, PCM, dts:X, and Auro3D virtualized physical layouts that would be far more expensive to actually build. A number of other A16 users and potential users have told me they can't listen to a full blown physical Atmos system because of various limitations (in an apartment building where they can't play loudly, in a dwelling where speaker layouts aren't easy to accomplish, etc.). For them the A16 is a godsend. We all hope that cheaper software binaural approaches will eventually match the experience.

As to the question of mixing on speakers vs headphones, the target audience for the original A8 and for the A16 is audio engineers who want the option to mix on headphones (maybe not exclusively, but on occassion when the engineer is away from the studio and needs to work on something). Given that both can provide virtual sound from any particular set of measured speakers and speaker layout, a portable replication of a studio is possible. Particularly for sound mixing for movies I believe the A8 has been popular. Both are professional bits of gear that happen to be great for multichannel audio enthusiasts. Look up the Sony 360 RA environment - it seems that Sony also believes in this application.

I must use my A16/HD800S more!

 

[DrKlahn]

My listening room is limited to 7.1.2 because of my AVR and where it's possible to place speakers. I definitely enjoy that physical speaker Atmos experience, and like you I've had friends over to listen to my latest acquisitions. More often it's trying to convince friends that multichannel music is worth the investment, and that most of us missed out when quad was around and readily available.

The A16 experience for me is exactly the same as the speaker experience, down to the sound moving as I turn my head, and that's the common experience of other users. The magic for me comes from the fact that I can measure my 7.1.2 layout in different ways, for example, sitting backwards or even on the floor (or on the floor backwards). The result is a greater number of virtual Atmos speakers - my speakers, but just in new physical positions - and I typically listen to 10.1.6 virtual setups. I also now have a true quad setup, where my large L/R speakers are used for my virtual Ls/Rs. Further, commercial A16 PRIRs (in the A16 world, a PRIR is a combination of an HRTF and an impulse response measurement) for speakers far more expensive than I would buy are cheaply available and let me experience Atmos with different physical (but virtual) setups.

An A16 is expensive, but it lets you use Atmos, PCM, dts:X, and Auro3D virtualized physical layouts that would be far more expensive to actually build. A number of other A16 users and potential users have told me they can't listen to a full blown physical Atmos system because of various limitations (in an apartment building where they can't play loudly, in a dwelling where speaker layouts aren't easy to accomplish, etc.). For them the A16 is a godsend. We all hope that cheaper software binaural approaches will eventually match the experience.

As to the question of mixing on speakers vs headphones, the target audience for the original A8 and for the A16 is audio engineers who want the option to mix on headphones (maybe not exclusively, but on occassion when the engineer is away from the studio and needs to work on something). Given that both can provide virtual sound from any particular set of measured speakers and speaker layout, a portable replication of a studio is possible. Particularly for sound mixing for movies I believe the A8 has been popular. Both are professional bits of gear that happen to be great for multichannel audio enthusiasts. Look up the Sony 360 RA environment - it seems that Sony also believes in this application.

Again, I don't doubt your findings. What I was trying to say is that I doubt that the majority of Atmos headphone experiences are anywhere close to approaching what setups like this provide. And likely aren't approaching what a modest speaker setup can provide.

I'm not disparaging any engineer for their preferred approach. I'm simply saying that I think as part of the QC of your work you should be evaluating on speakers and headphones if that is what your audience is using for playback. Just to ensure there are no anomalies on either. But if the headphone tech is that indistinguishable for you, that may not be necessary.

But the main point of the thread is playback/consumption and I'm still of the opinion that the average Atmos headphone is probably not on par with a modest speaker setup. I could certainly be wrong though.

 

[zdjh22]

But the main point of the thread is playback/consumption and I'm still of the opinion that the average Atmos headphone is probably not on par with a modest speaker setup. I could certainly be wrong though.

I completely agree. Consumer headphones intended for spatial audio are not going to match the experience of a modest consumer AVR/speaker setup. Virtuoso software on a Mac playing Spatial Audio from Apple Music, or from a video streaming service, using a quality pair of headphones and a physical head tracker comes close. But that software approach has some compromises, is a bit picky needing some expertise, and is nowhere near as consumer friendly as an AVR plus speakers plus a UDP or a streamer. A Mac plus multiple DACs plus speakers is maybe the cheapest way into Atmos, but also a very fiddly option.

At a more audiophile grade, so say $5K+ rather than a consumer setup at maybe $1.5K+ (lots of compromise on speaker quality), spatial headphone experience can fully match physical speaker experience. I came to multichannel audio through a couple of decades of simpler surround sound for TV and movies, so already had a 5.1 speaker setup. Getting to 7.1.2 was just incremental, and I suspect many folks can go that way and be extremely happy.

 

[minimumtumbleweed]

May I ask what headphones you've listened to Atmos music mixes? Now that this thread has been moved to speakers vs headphones, maybe we can get to the bottom of your bias.

I have not used any officially licensed Dolby Atmos or Spatial Audio headphones, because, quite frankly, they all fail fundamentally in delivering high audio quality. You're not using Airpods or Beats or TWS headphones because you care about audio quality, you're using them because they are convenient and wireless. But, apparently you can use any headphone if you really want. I tried Sennheiser HD 650, HiFiMan Anandas, and my current pair, the Fiio FT5 which are all very good midrange audiophile headphones. I preferred the stereo mix in all cases. You can go on about the software, but in the end the hardware is going to do the heavy lifting. I say, if a person wants to listen to atmos music on their headphones, go for it, more power to them. But to say that it is the ideal way to listen to it, or to suggest that Atmos music should be mixed for headphones, is misguided.

 

[JimofMaine]

They cannot. And I say this as a headphone lover and collector.

You're clearly talking about stereo mixes in headphones...I am not limiting it to stereo. Ambisonics' spatial accuracy in headphones is better than consumer speaker setups. Again, not talking about which spatial presentation you may prefer...maybe your opinion is that speakers are more immersive, offer more tactile connection, more "Atmos" feel, etc. Fine, but not the point I was making. Accuracy of location belongs to Ambisonic, binaural listening.

 

[JimofMaine]

I have not used any officially licensed Dolby Atmos or Spatial Audio headphones, because, quite frankly, they all fail fundamentally in delivering high audio quality. You're not using Airpods or Beats or TWS headphones because you care about audio quality, you're using them because they are convenient and wireless. But, apparently you can use any headphone if you really want. I tried Sennheiser HD 650, HiFiMan Anandas, and my current pair, the Fiio FT5 which are all very good midrange audiophile headphones. I preferred the stereo mix in all cases. You can go on about the software, but in the end the hardware is going to do the heavy lifting. I say, if a person wants to listen to atmos music on their headphones, go for it, more power to them. But to say that it is the ideal way to listen to it, or to suggest that Atmos music should be mixed for headphones, is misguided.

Wow, claiming something is misguided while also admitting you're not listening to the format as intended seems misguided.

Don't take my word for it: look at Dolby's sample music on their website. They clearly state, "Best Heard on Headphones"

Plus, look at the virtual reality world, all headset based Binaural, HRTF encode and decode. Apple's new Pro Vision uses headsets with head tracking binaural audio. In Gaming, HRTFs are the biggest and newest tech, and yep over headsets only. These examples of spatial over headsets are everywhere, if you look. Headset are king.

Personal preferences aside, headphones are spatial when playing binaurally rendered spatial content. So these enumerated things must exists:
* binaural signal-either recorded w binaural microphones or Ambisonic microphones or mathematical translation of an audio signal w applied HRTFs, head related transfer functions
* So be encoded with HRTFs
* And be decoded HRTFs.

If all these are not in use, it is not a Dolby Atmos Binaural mix you are listening to. May I suggest you try not to predict its sound, if you have not experienced it.

It's dogmatic to say speakers are more spatial, I get that. The surround community has had good reason to think headphones are a step down from speakers--spatially speaking. After all, when talking 5.1 and 7.1 and converting them to stereo for headsets, these conversions definitely lose spatiality.

Dolby Atmos is designed differently. The format is ESPECIALLY well suited for headphone, binaural listening.

Personally I like speaker listening better, however I recognize that headphones have large, distinct spatial advantages.

 

[DrKlahn]

Don't take my word for it: look at Dolby's sample music on their website. They clearly state, "Best Heard on Headphones"

I'm not sure that is much of a smoking gun when the vast majority of people are not consuming a web page from a device hooked up to a full spatial audio system in their media room. Most people are looking at the site on their phones where, yes, headphones would be an improvement.

Again, I'm open to the idea of headphones and quality Atmos reproduction. But I remain skeptical it would match my speaker setup based on experiences with headphones trying to replicate immersive audio in the past. And I know very few people that will invest in high end headphones, meaning most of the Atmos headphones are probably just Atmos in name only. But to be fair, very few people will move beyond a soundbar to an actual speaker setup.

All things being equal, I would still have a hard time steering someone towards headphones vs speaker setup (provided the space and funds aren't an issue). There's just so much more you can do from an entertainment standpoint with the latter.

 

[minimumtumbleweed]

Wow, claiming something is misguided while also admitting you're not listening to the format as intended seems misguided.

Don't take my word for it: look at Dolby's sample music on their website. They clearly state, "Best Heard on Headphones"

Dolby is trying to sell a product. If they told you that it only really works when you buy at least 7 speakers, and several need to be installed in your ceiling, as well as a receiver, and probably a subwoofer, they would lose sales to those who are not willing to make that investment (i.e., most people). From Dolby's website:

“The Binaural Renderer is used for monitoring Binaural audio on headphones, creating Binaural re-renders, and for writing Binaural metadata to .atmos Master files.

The Binaural Renderer reproduces the experience of a multi-channel loudspeaker listening experience as closely as possible over headphones.”

So even in their own words they are trying to replicate the experience of a home theatre, not vice-versa.

Plus, look at the virtual reality world, all headset based Binaural, HRTF encode and decode. Apple's new Pro Vision uses headsets with head tracking binaural audio. In Gaming, HRTFs are the biggest and newest tech, and yep over headsets only. These examples of spatial over headsets are everywhere, if you look. Headset are king.

Personal preferences aside, headphones are spatial when playing binaurally rendered spatial content. So these enumerated things must exists:
* binaural signal-either recorded w binaural microphones or Ambisonic microphones or mathematical translation of an audio signal w applied HRTFs, head related transfer functions
* So be encoded with HRTFs
* And be decoded HRTFs.

If all these are not in use, it is not a Dolby Atmos Binaural mix you are listening to. May I suggest you try not to predict its sound, if you have not experienced it.

You don't actually need "Dolby-licensed" headphones to "properly" listen to it, I already brought this point up. Any headphones will work. I find it hard to believe that significantly worse headphones are going to magically sound better just because they are "licensed". The processing is still happening.

It's dogmatic to say speakers are more spatial, I get that. The surround community has had good reason to think headphones are a step down from speakers--spatially speaking. After all, when talking 5.1 and 7.1 and converting them to stereo for headsets, these conversions definitely lose spatiality.

Dolby Atmos is designed differently. The format is ESPECIALLY well suited for headphone, binaural listening.

Personally I like speaker listening better, however I recognize that headphones have large, distinct spatial advantages.

Can't imagine a bigger soundstage than a movie theatre, which is in reality what Atmos was purpose-designed for.

 

[JimofMaine]

I'm not sure that is much of a smoking gun when the vast majority of people are not consuming a web page from a device hooked up to a full spatial audio system in their media room. Most people are looking at the site on their phones where, yes, headphones would be an improvement.

Again, I'm open to the idea of headphones and quality Atmos reproduction. But I remain skeptical it would match my speaker setup based on experiences with headphones trying to replicate immersive audio in the past. And I know very few people that will invest in high end headphones, meaning most of the Atmos headphones are probably just Atmos in name only. But to be fair, very few people will move beyond a soundbar to an actual speaker setup.

All things being equal, I would still have a hard time steering someone towards headphones vs speaker setup (provided the space and funds aren't an issue). There's just so much more you can do from an entertainment standpoint with the latter.

Very thoughtful response and I agree technical reasons are to blame for Dolby itself recommending headsets to hear Atmos. But isn't that just another way to say the same thing? Dolby says headphones are better at SHOWCASING a billion dollar product. Believe them.

Totally agree with you that many (including me) prefer a Dolby Atmos speaker presentation for movies verses a headset. Doesn't change the fact that the headset version has better spatial representation. There's more to sound quality and character than spatialization. Now for Atmos music it's a toss up for me. Depends on my mood. Sometimes I like the headphone presentation much more than the speaker. Almost without fail I like the original stereo speaker presentation the best.

However, many Atmos Binaural headphone presentations, to my ears, are better than and original stereo, headset presentation! This is pretty amazing. (And to beat a dead horse, a Dolby Atmos mixdown to 2.0 headphone stereo is unlistenable. Especially compared to the original stereo mix or the Dolby binaural.)

 

[JimofMaine]

Dolby is trying to sell a product. If they told you that it only really works when you buy at least 7 speakers, and several need to be installed in your ceiling, as well as a receiver, and probably a subwoofer, they would lose sales to those who are not willing to make that investment (i.e., most people). From Dolby's website:


So even in their own words they are trying to replicate the experience of a home theatre, not vice-versa.

You don't actually need "Dolby-licensed" headphones to "properly" listen to it, I already brought this point up. Any headphones will work. I find it hard to believe that significantly worse headphones are going to magically sound better just because they are "licensed". The processing is still happening.


Can't imagine a bigger soundstage than a movie theatre, which is in reality what Atmos was purpose-designed

Don't disagree with you on these things. The Dolby text about rendering is about in Studio mastering...which maybe was done on a headset (more common) or on 7.1.4 or better. It's about creating the ADM file, not a statement about translation.

Not all headsets get the Binaural file. On Apple Music, for instance, you must have Apple Spatial Audio enabled or you actually will be listening to a stereo mixdown of the Atmos mix. Only some Apple headphones are capable of this.

Along these same lines, Gamers will tell you there are specific binaural spatial cues only available with Dolby Certified headsets and xBox/etc. enable Dolby audio. There are paid subscriptions for such access. Obviously, regular headsets can be used, but no payment, no access to binaural Atmos content.

So the processing isn't just happening. It's more complicated than that.

 

[zdjh22]

Don't disagree with you on these things. The Dolby text about rendering is about in Studio mastering...which maybe was done on a headset (more common) or on 7.1.4 or better. It's about creating the ADM file, not a statement about translation.

Not all headsets get the Binaural file. On Apple Music, for instance, you must have Apple Spatial Audio enabled or you actually will be listening to a stereo mixdown of the Atmos mix. Only some Apple headphones are capable of this.

Along these same lines, Gamers will tell you there are specific binaural spatial cues only available with Dolby Certified headsets and xBox/etc. enable Dolby audio. There are paid subscriptions for such access. Obviously, regular headsets can be used, but no payment, no access to binaural Atmos content.

So the processing isn't just happening. It's more complicated than that.

On Apple Music on a Mac, you can use Loopback (or Blackhole) with OSX’s built-in Atmos renderer to pipe the multiple channels you’ve configured (say 5.1, or 7.1.4) into a software binaural headphone converter like Virtuoso, and then use whatever headphones you desire to listen. Virtuoso has equalization files for many dozens of makes and models of headphones, and they are willing to create additional files on demand. Virtuoso also supports several different head trackers so spatial audio like with Apple headphones can be accomplished (i.e. move your head and the spatial sound pattern moves appropriately).

The problem is that you are stuck with the three (IIRC) HRTFs that ship with VIrtuoso. If your head/ear/body shape match one of those, great. But if not, you won’t get the full desired effect. You can configure all sorts of listening rooms with different speaker numbers and distances, and sometimes that will help compensate. But the lack of a personal HRTF is really the issue with all binaural audio. Virtuoso does support SOFA-format HRTF files, and you can for example purchase a measurement kit from Earfish (the University of Antwerp research program) and send your data to them for conversion to your personal HRTF.

A realtime binaural renderer like the A16 relies on a prior direct measurement, using tiny in-ear microphones, of your HRTF convolved with the particular speakers you’ve used for the measurement and the environment those speakers are in. So with a big room, you get more reverb. A smaller room with sound dampening, like a mixing room, yields a very different result. The in-ear mics can be used while wearing your over-ear headphones to generate a personal headphone equalization file, which compensates for both headphone response and your particular ear shapes and resonances. For IEMs there is a procedure to generate an equalization file by progressing across the frequency spectrum and setting L/R gains to achieve perceived equal sound levels (much harder than it sounds). The standard headphones that most A16 users employ are the Sennheiser model HD800s. In my experience the much cheaper Sony MDR-MV1 headphones are sonically similar - nearly identical - to my HD800s phones, but ergonomically they are not as nice.

I know that there are multiple sources for generating HRTFs from head and ear scans/photographs.

If you are interested in trying binaural samples of Atmos or other multichannel setups that came off an A16, you can go to the 3D Sound Shop website and download (or play in Chrome) samples for about a dozen different brands of speakers. These were produced with the owner’s head/ears (his HRTF) and a pair of HD800s headphones. There are instructions for configuring an A16 to demo the files with different headphones with your personal equalization file, as after all these commercial PRIRs are for the A16. But without an A16 you can nevertheless use whatever headphones you might have, and simply play the demos via your PC. You’ll still experience binaural (multichannel) audio, it just won’t be as good as what you could get with a setup matched to your body/ears.

 

[zdjh22]

(Deleted - somehow this got posted twice)

 

[austinsignal]

Interesting thread. I use Audez MM-100 phones to check binaural fold downs, and Apple Airpod Max for checking Apple Spatial fold downs. Neither are other-worldly expensive, but both sound very, very good. And with the Max's, I'm comfortable knowing that the people that curate Apple's various playlists are likely listening to them (so important to artists/managers/labels). Many (most?) of the indie artists I mix for evaluate their mixes on headphones of some kind. Lots of airpods out there for sure. So - I spend a fair amount of time listening my ATMOS mixes on phones and toggling to the original stereo master for comparison. It can be challenging, as artists will have listened to their stereo masters many, many, many times prior to hearing the binaural/spatial mixes they get from me...so they're very keen to hearing differences. In particular, if the original stereo masters have mid-side/wideness things going on, the stereo version can sound very wide on headphones...and an ATMOS mix that is spread beautifully across the whole immersive sound stage on speakers can actually sound narrower (less immsersive?) than the wideness-enhanced stereo master. So there are panning tricks and others, including using the object distance parameters in the Dolby downmix settings that can really enhance the spatial presentation over headphones. Things really get interesting when the stereo master was mastered from stems, as opposed to from a stereo mix. Then, all kinds of processing can happen on individual stems (drums, vox, etc) as opposed to the whole mix. When done well the resulting stero mix can be very immersive and truly great-sounding on headphones. Ferreting that out and developing a surround mix that can compete with that on headphones is challenging, tbh kinda fun. 'ah shit...so that's what they're doing' moments... My thoughts on headphone fold downs: 1. Do no harm (they can't sound worse than the stereo mix on phones) 2. They should be distinguishable from the stereo mix on headphones 3. Hopefully the headphone mixes will sound better. 4. Don't try to compare headphone mixes to the ATMOS speaker version, as they aren't remotely the same thing for the consumer. Lastly, lately I've been doing whatever I can to get artists in an ATMOS room to hear their material in all its glory. Just last week I sent Richard Houghten to Todd Burke's The Listening Room LA in LA to hear his new album that I mixed recently. He also got to hear another album I did for him a while back on Apple Music. And, just the other day he bought a pair of Airpod Max - and he's pretty excited about the quality of Spatial mixes (some, anyway!) on those phones. I think he's a convert at this point - I may lose him as a mix client as he's now conjuring how to record/produce in ATMOS as opposed to the remix route....ok with me as I'm all for moving the format forward!

 

[austinsignal]

On Apple Music on a Mac, you can use Loopback (or Blackhole) with OSX’s built-in Atmos renderer to pipe the multiple channels you’ve configured (say 5.1, or 7.1.4) into a software binaural headphone converter like Virtuoso, and then use whatever headphones you desire to listen. Virtuoso has equalization files for many dozens of makes and models of headphones, and they are willing to create additional files on demand. Virtuoso also supports several different head trackers so spatial audio like with Apple headphones can be accomplished (i.e. move your head and the spatial sound pattern moves appropriately).

The problem is that you are stuck with the three (IIRC) HRTFs that ship with VIrtuoso. If your head/ear/body shape match one of those, great. But if not, you won’t get the full desired effect. You can configure all sorts of listening rooms with different speaker numbers and distances, and sometimes that will help compensate. But the lack of a personal HRTF is really the issue with all binaural audio. Virtuoso does support SOFA-format HRTF files, and you can for example purchase a measurement kit from Earfish (the University of Antwerp research program) and send your data to them for conversion to your personal HRTF.

A realtime binaural renderer like the A16 relies on a prior direct measurement, using tiny in-ear microphones, of your HRTF convolved with the particular speakers you’ve used for the measurement and the environment those speakers are in. So with a big room, you get more reverb. A smaller room with sound dampening, like a mixing room, yields a very different result. The in-ear mics can be used while wearing your over-ear headphones to generate a personal headphone equalization file, which compensates for both headphone response and your particular ear shapes and resonances. For IEMs there is a procedure to generate an equalization file by progressing across the frequency spectrum and setting L/R gains to achieve perceived equal sound levels (much harder than it sounds). The standard headphones that most A16 users employ are the Sennheiser model HD800s. In my experience the much cheaper Sony MDR-MV1 headphones are sonically similar - nearly identical - to my HD800s phones, but ergonomically they are not as nice.

I know that there are multiple sources for generating HRTFs from head and ear scans/photographs.

If you are interested in trying binaural samples of Atmos or other multichannel setups that came off an A16, you can go to the 3D Sound Shop website and download (or play in Chrome) samples for about a dozen different brands of speakers. These were produced with the owner’s head/ears (his HRTF) and a pair of HD800s headphones. There are instructions for configuring an A16 to demo the files with different headphones with your personal equalization file, as after all these commercial PRIRs are for the A16. But without an A16 you can nevertheless use whatever headphones you might have, and simply play the demos via your PC. You’ll still experience binaural (multichannel) audio, it just won’t be as good as what you could get with a setup matched to your body/ears.

wow. those samples are incredible on my Airpod Max phone....are there any music samples anywhere?!

 

[JimofMaine]

On Apple Music on a Mac, you can use Loopback (or Blackhole) with OSX’s built-in Atmos renderer to pipe the multiple channels you’ve configured (say 5.1, or 7.1.4) into a software binaural headphone converter like Virtuoso, and then use whatever headphones you desire to listen. Virtuoso has equalization files for many dozens of makes and models of headphones, and they are willing to create additional files on demand. Virtuoso also supports several different head trackers so spatial audio like with Apple headphones can be accomplished (i.e. move your head and the spatial sound pattern moves appropriately).

The problem is that you are stuck with the three (IIRC) HRTFs that ship with VIrtuoso. If your head/ear/body shape match one of those, great. But if not, you won’t get the full desired effect. You can configure all sorts of listening rooms with different speaker numbers and distances, and sometimes that will help compensate. But the lack of a personal HRTF is really the issue with all binaural audio. Virtuoso does support SOFA-format HRTF files, and you can for example purchase a measurement kit from Earfish (the University of Antwerp research program) and send your data to them for conversion to your personal HRTF.

A realtime binaural renderer like the A16 relies on a prior direct measurement, using tiny in-ear microphones, of your HRTF convolved with the particular speakers you’ve used for the measurement and the environment those speakers are in. So with a big room, you get more reverb. A smaller room with sound dampening, like a mixing room, yields a very different result. The in-ear mics can be used while wearing your over-ear headphones to generate a personal headphone equalization file, which compensates for both headphone response and your particular ear shapes and resonances. For IEMs there is a procedure to generate an equalization file by progressing across the frequency spectrum and setting L/R gains to achieve perceived equal sound levels (much harder than it sounds). The standard headphones that most A16 users employ are the Sennheiser model HD800s. In my experience the much cheaper Sony MDR-MV1 headphones are sonically similar - nearly identical - to my HD800s phones, but ergonomically they are not as nice.

I know that there are multiple sources for generating HRTFs from head and ear scans/photographs.

If you are interested in trying binaural samples of Atmos or other multichannel setups that came off an A16, you can go to the 3D Sound Shop website and download (or play in Chrome) samples for about a dozen different brands of speakers. These were produced with the owner’s head/ears (his HRTF) and a pair of HD800s headphones. There are instructions for configuring an A16 to demo the files with different headphones with your personal equalization file, as after all these commercial PRIRs are for the A16. But without an A16 you can nevertheless use whatever headphones you might have, and simply play the demos via your PC. You’ll still experience binaural (multichannel) audio, it just won’t be as good as what you could get with a setup matched to your body/ears.

Yes binaural and Ambisonic playback over regular headphones has been around awhile, but it hasn't been without its challenges. Results are sort of all over the board as mentioned here regarding personalized HRTFs.

Dolby Atmos is a valid attempt to fix many of the complaints. Apple also has been instrumental. And as you may have guessse its not without controversy between these two. One area of contention is the binaural renderer: Apple is insisting on using their own HRTF render. DolthAtmos ADM file

Interesting thread. I use Audez MM-100 phones to check binaural fold downs, and Apple Airpod Max for checking Apple Spatial fold downs. Neither are other-worldly expensive, but both sound very, very good. And with the Max's, I'm comfortable knowing that the people that curate Apple's various playlists are likely listening to them (so important to artists/managers/labels). Many (most?) of the indie artists I mix for evaluate their mixes on headphones of some kind. Lots of airpods out there for sure. So - I spend a fair amount of time listening my ATMOS mixes on phones and toggling to the original stereo master for comparison. It can be challenging, as artists will have listened to their stereo masters many, many, many times prior to hearing the binaural/spatial mixes they get from me...so they're very keen to hearing differences. In particular, if the original stereo masters have mid-side/wideness things going on, the stereo version can sound very wide on headphones...and an ATMOS mix that is spread beautifully across the whole immersive sound stage on speakers can actually sound narrower (less immsersive?) than the wideness-enhanced stereo master. So there are panning tricks and others, including using the object distance parameters in the Dolby downmix settings that can really enhance the spatial presentation over headphones. Things really get interesting when the stereo master was mastered from stems, as opposed to from a stereo mix. Then, all kinds of processing can happen on individual stems (drums, vox, etc) as opposed to the whole mix. When done well the resulting stero mix can be very immersive and truly great-sounding on headphones. Ferreting that out and developing a surround mix that can compete with that on headphones is challenging, tbh kinda fun. 'ah shit...so that's what they're doing' moments... My thoughts on headphone fold downs: 1. Do no harm (they can't sound worse than the stereo mix on phones) 2. They should be distinguishable from the stereo mix on headphones 3. Hopefully the headphone mixes will sound better. 4. Don't try to compare headphone mixes to the ATMOS speaker version, as they aren't remotely the same thing for the consumer. Lastly, lately I've been doing whatever I can to get artists in an ATMOS room to hear their material in all its glory. Just last week I sent Richard Houghten to Todd Burke's The Listening Room LA in LA to hear his new album that I mixed recently. He also got to hear another album I did for him a while back on Apple Music. And, just the other day he bought a pair of Airpod Max - and he's pretty excited about the quality of Spatial mixes (some, anyway!) on those phones. I think he's a convert at this point - I may lose him as a mix client as he's now conjuring how to record/produce in ATMOS as opposed to the remix route....ok with me as I'm all for moving the format forward!

Very cool to hear about your experience! Especially with Houghten. Are you using Audiomovers Apple Binaural Renderer to evaluate the final?

For those who don't mix in Atmos, there is a controversial move being made Apple--they are applying their own binaural rendering HRTFs to all Atmos music. (supposedly because they are better and they claim to be able to personalize and literally measure every listener's unique head.) So you do not hear the Dolby Atmos binaural render through Apple Music. As an engineer using Atmos in a DAW, you listen and make decisions over headphones to the Dolby binaural monitor. This is quite astonishing. Audiomovers is the only software I know that let's you preview how it will actually sound on the Apple platform.