Anyone tried Atmos music upmixing?

[radio3]

I am thinking of going to multichannel for audio in order to have more immersive sound when entertaining guests and that kind of thing. The vast majority of content available, of course, is stereo rather than native multichannel, so most of the time will need to be upmixed.

Could anyone here share their experiences with listening to music using Atmos? I would love to hear what people think of it. Do you like it and find the results are satisfying and immersive?

 

[PaulD]

I have tried it (in a studio) and I greatly prefer high quality stereo reproduction. There may be some Atmos music material that works well, but for me music is usually an experience with a definite front where the performers are, and anything from the sides or rear is ambience or room reflections.

I have even worked in ambisonics as well as cinema formats. For me, stereo is fine for music. It is in cinema and VR where more immersive surround formats have an advantage.

 

[Pdxwayne]

My friend have 2 Atmos setups in his home. He always upmix stereo music to Atmos. Both systems sounds great to me.

 

[Sancus]

I listen to just about everything upmixed using the Auro3D upmixer. It is a lot better than the Dolby or DTS upmixers. I am not alone in this opinion.

The upmixers use various tricks to add ambience(ie: stuff-that-sounds-like-reflections) and spread the sound over the front 3 channels. It is rare that I dislike this effect, and it's also a fairly subtle effect.

For those who think it is artificial, well, it's really no different than what you get from a dipole or a set of speakers with unusually wide directivity, just moreso. In those cases you are also adding artificial ambience to your playback system, as most modern recordings were made with narrow directivity studio monitors.

I strongly recommend the multi-channel chapter of Floyd Toole's book if you're interested in understanding multi-channel audio.

 

[Archaea]

I listen to just about everything upmixed using the Auro3D upmixer. It is a lot better than the Dolby or DTS upmixers. I am not alone in this opinion.

The upmixers use various tricks to add ambience(ie: stuff-that-sounds-like-reflections) and spread the sound over the front 3 channels. It is rare that I dislike this effect, and it's also a fairly subtle effect.

For those who think it is artificial, well, it's really no different than what you get from a dipole or a set of speakers with unusually wide directivity, just moreso. In those cases you are also adding artificial ambience to your playback system, as most modern recordings were made with narrow directivity studio monitors.

I strongly recommend the multi-channel chapter of Floyd Toole's book if you're interested in understanding multi-channel audio.

Auro 2D adds harmonic distortion and increases the bass trims and decreases the treble trims in a very measurable way. (At least on my Denon X6700H implementation) It’s not just adding ambiance, it’s changing EQ and adding 2nd order harmonic distortion for fuller sounding bass. It’s tailoring the sound. That’s not a bad thing, but it is worthy of mention and some transparency.

https://www.avsforum.com/threads/wh...g-what-is-going-on-behind-the-scenes.3169754/


Upmixing stereo to Atmos is kind of a misconception. You can upmix stereo with Dolby Surround and DTS Neural X but not with Atmos nor with DTS X. Atmos and DTS X are object based audio. There are no objects in stereo music. So the terminology used this far in this thread is a bit off.

I upmix my stereo listening nearly all the time. My personal favorite on Denon receivers is Dolby Surround. It gives a a strong centered sound and front sound stage presence, and fills in the room nicely through the surround and ceiling channels. It seems to be more subtle than some of the other DSPs.

 

[Chromatischism]

Auro 2D adds harmonic distortion and increases the bass trims and decreases the treble trims in a very measurable way. It’s not just adding ambiance, it’s changing EQ and adding 2nd order harmonic distortion for fuller sounding bass. It’s tailoring the sound. That’s not a bad thing, but it is worthy of mention and some transparency.

https://www.avsforum.com/threads/wh...g-what-is-going-on-behind-the-scenes.3169754/


Upmixing stereo to Atmos is kind of a misconception. You can upmix stereo with Dolby Surroung and DTS Neural X but not with Atmos and not with DTS X. Atmos and DTS X are object based audio. There are no objects in stereo music.

Are you sure the distortion isn't just increasing due to the bass boost?

 

[richard12511]

As it is out of the box, I prefer strict stereo over upmixed atmos.

However, with a few easy tweaks, Atmos can be made to surpass stereo.

Default Atmos upmixer has 2 main issues(imo):

1. It worsens the stereo center image. You lose that holographic sound to center image.
2. You hear sounds coming from behind you, which(at least for me) breaks the illusion that there is a performance happening in front of you.

Problem 1 can be mostly corrected by enabling "Center Spread" in the atmos upmixer settings via your AVR.

Problem 2 can be fixed by manually turning down the level of the rear surrounds(though you'll want to turn them back up for movies/games, which gets tiresome). How much you need to turn them down depends on your speakers and listening distance. For me it was around -4.5dB, but you'll just have to play with it. Find a good piece of music where your brain is perceiving that there are speakers playing music behind you, then slowly starting turning them down, little by little. At some point, it's almost like a switch flips in your brain. Your brain stops perceiving the sound as coming from speakers behind you, and starts perceiving the sound behind you as echoes coming from your front speakers.

Done right, I find it superior to stereo. Stereo imaging in the sweet spot is identical, but now is perceivable over several seats, and the sense of envelopment is far greater. You kinda get the benefits of both wide dispersion(stable image, sense of envelopment) and narrow dispersion(tight imaging) at the same time.

 

[Archaea]

Are you sure the distortion isn't just increasing due to the bass boost?

Possible, but the ratio seems off.
7dB more bass for 15db dB more second order harmonic distortion. Also I think you can hear the harmonic distortion. Auro 2D on my AVR sounds different than just cranking the bass 7dB more on DTS Neural X or Dolby Surround or Stereo. The bass sounds fuller, has harmonics — it’s subtle, but I suspected there was harmonic distortion added when I first heard it this summer (almost like a restorer function or something) and subsequently on multiple occasions noticed it sometimes made bass on music I was VERY familiar with sound a bit “weird” or just “different” and not only just louder. I’m constantly toying with my sub trims and playing my favorite tracks moving the subwoofer trims around and I could tell there seemed to be something there more than just the 7dB trim increase.
My subs are loafing in the test I linked, so it’s not that they are being overloaded or being driven into distortion. (I have eight 18” sealed subwoofers powered by four INuke DSP 6000 amps) and as you can see on the FR and Distortion charts I linked - I was playing my test tracks back at very moderate SPL levels.

 

[Archaea]

As it is out of the box, I prefer strict stereo over upmixed atmos.

However, with a few easy tweaks, Atmos can be made to surpass stereo.

Default Atmos upmixer has 2 main issues(imo):

1. It worsens the stereo center image. You lose that holographic sound to center image.
2. You hear sounds coming from behind you, which(at least for me) breaks the illusion that there is a performance happening in front of you.

Problem 1 can be mostly corrected by enabling "Center Spread" in the atmos upmixer settings via your AVR.

Problem 2 can be fixed by manually turning down the level of the rear surrounds(though you'll want to turn them back up for movies/games, which gets tiresome). How much you need to turn them down depends on your speakers and listening distance. For me it was around -4.5dB, but you'll just have to play with it. Find a good piece of music where your brain is perceiving that there are speakers playing music behind you, then slowly starting turning them down, little by little. At some point, it's almost like a switch flips in your brain. Your brain stops perceiving the sound as coming from speakers behind you, and starts perceiving the sound behind you as echoes coming from your front speakers.

Done right, I find it superior to stereo. Stereo imaging in the sweet spot is identical, but now is perceivable over several seats, and the sense of envelopment is far greater. You kinda get the benefits of both wide dispersion(stable image, sense of envelopment) and narrow dispersion(tight imaging) at the same time.

two things
1)Your center channel speaker is not the same as you left and right speaker I’ll wager. If it were, I predict you would not need to use “center spread”. I’ve had matching LCR my last three speaker upgrades and last two rooms, and I have always disliked “center spread” (but I do like stereo) People I’ve continually heard advise using “center spread” have physically different center channel speakers than their L/R.
2) Turn off Dynamic EQ. With Dynamic EQ off Dolby Surround no longer makes the surround too loud, and you don’t have to mess with channel trims. I like Dynamic EQ on for movies at lower volumes and and off for movies at reference. For music at lower volumes in just stereo Dynamic EQ is great, (extra bass). But in a upmixing format as you said, the surround speakers overwhelm unless you crank down the trims.

 

[richard12511]

two things
1)Your center channel speaker is not the same as you left and right speaker I’ll wager. If it were, I predict you would not need to use “center spread”. I’ve had matching LCR my last three speaker upgrades and last two rooms, and I have always disliked “center spread” (but I do like stereo) People I’ve continually heard advise using “center spread” have physically different center channel speakers than their L/R.
2) Turn off Dynamic EQ. With Dynamic EQ off Dolby Surround no longer makes the surround too loud, and you don’t have to mess with channel trims. I like Dynamic EQ on for movies and off for music.

Center channel is identical to left right in my case. Perhaps it's different in my AVR, but without center spread on, my Left and Right speakers get relegated to mostly playing echo and ambiance noises(similar to movies). It makes the center image sound like you're listening to a mono source through one speaker, which makes sense, as the center is now playing all of the dialog and 90% of the instruments. Occasionally on old Stones album or Jazz album will have some hard panned images that get played through the left/right, but for the most part(with modern music) they just play echo noises.

I'm not sure about how I feel about Dynamic EQ. I've got it off right now, but I've played with it from time to time.

***Edit: I should clarify that I've never had Dynamic EQ on in the system I was talking about with the Atmos upmixer(Yamaha AVR). The times I've used Dynamic EQ have been with my Denon AVR in my office, where I'm listening very quietly while working.

 

[Archaea]

Center channel is identical to left right in my case. Perhaps it's different in my AVR, but without center spread on, my Left and Right speakers get relegated to mostly playing echo and ambiance noises(similar to movies). It makes the center image sound like you're listening to a mono source through one speaker, which makes sense, as the center is now playing all of the dialog and 90% of the instruments. Occasionally on old Stones album or Jazz album will have some hard panned images that get played through the left/right, but for the most part(with modern music) they just play echo noises.

I'm not sure about how I feel about Dynamic EQ. I've got it off right now, but I've played with it from time to time.

Cool. Well there goes that theory anyway. I thought I was onto something. I guess sometimes there’s no accounting for personal tastes and preferences. If I were to hear your room or you to hear mine - I wonder if we’d hold our expected preferences. I too have like/dislike relationship with dynamic EQ. I mainly wish you could just lower the quantity of the effect. Yes you can set the Reference Level Offset to -5 , -10 , 15 but I sure don’t like how after you increase the main volume beyond those Reference Level Offset levels that the effects Dynamic EQ apply actually reverse. (So you get LESS bass and LESS boost to the rear speakers if you use that setting. Since I like having a reference capable system and demoing things at reference for guests or for spirited listening sessions that pretty much eliminates anything but -5 out of principle and even that sits ill in the mind.

 

[richard12511]

Cool. Well there goes that theory anyway. I thought I was into something. I guess sometimes there’s no accounting for personal tastes and preferences. If I were to hear your room or you to hear mine - I wonder if we’d hold our expected preferences.

I could definitely see where a feature like that is a matter of taste. IMO, it makes the center image too good . My brain likes to at least have the illusion that I'm still listening to stereo.

 

[Sancus]

Auro 2D adds harmonic distortion and increases the bass trims and decreases the treble trims in a very measurable way. (At least on my Denon X6700H implementation) It’s not just adding ambiance, it’s changing EQ and adding 2nd order harmonic distortion for fuller sounding bass. It’s tailoring the sound. That’s not a bad thing, but it is worthy of mention and some transparency.

https://www.avsforum.com/threads/wh...g-what-is-going-on-behind-the-scenes.3169754/


Upmixing stereo to Atmos is kind of a misconception. You can upmix stereo with Dolby Surround and DTS Neural X but not with Atmos nor with DTS X. Atmos and DTS X are object based audio. There are no objects in stereo music. So the terminology used this far in this thread is a bit off.

I upmix my stereo listening nearly all the time. My personal favorite on Denon receivers is Dolby Surround. It gives a a strong centered sound and front sound stage presence, and fills in the room nicely through the surround and ceiling channels. It seems to be note subtle than some of the other DSPs.

That's interesting. It is worth noting that Auro2D and Auro3D sound pretty different, with Auro3D being more configurable since you can adjust the amount of effect and the size of the replicated venue. Auro3D requires 4 height channels to activate. You can enable unused channels to force this and it is still significantly different from Auro2D even with 0 or 2 height channels. (E:you already know this )

I suppose I can make these sorts of measurements with REW so maybe I'll do that out of curiosity at some point.

I really don't like the DTS: Neural-X upmixer for stereo because it hard pans vocals to the center channel(some people seem to like this, I really hate it) and activates the surrounds too much for my taste. Interestingly it does NOT do this with actual multi-channel music, and I like it for that. I should also note that many true 5.1 albums have significantly more bass mixed into them than stereo recordings of the exact same music.

I haven't really experimented much with Dolby Surround, but I may do that. FWIW, I tend to like a pretty bass-emphasized curve in general so the changes Auro2D makes are fine with me, but it is good to be aware that they exist!

E: I should have read further down that thread. I like Auro3D at a strength of around 8, so I guess I maybe don't like as much bass as the 7dB default for Auro2D. Of course, Denon AVRs have basic bass/treble tone controls that you can use IF you're not using Audyssey Dynamic EQ, so you can probably adjust that if you want as well. Using Dynamic EQ AND Auro2D would probably produce a LOT more bass.

I also think it would be interesting to measure the surround and center channels for each of the upmixing algorithms to see what's going on there, as I think the spatial differences are more significant than the timbre changes. And it is worth noting that the surround channel content affects the timbre of the final experience as well. But that is obviously a lot of work!

 

[Sancus]

two things
1)Your center channel speaker is not the same as you left and right speaker I’ll wager. If it were, I predict you would not need to use “center spread”. I’ve had matching LCR my last three speaker upgrades and last two rooms, and I have always disliked “center spread” (but I do like stereo) People I’ve continually heard advise using “center spread” have physically different center channel speakers than their L/R.
2) Turn off Dynamic EQ. With Dynamic EQ off Dolby Surround no longer makes the surround too loud, and you don’t have to mess with channel trims. I like Dynamic EQ on for movies at lower volumes and and off for movies at reference. For music at lower volumes in just stereo Dynamic EQ is great, (extra bass). But in a upmixing format as you said, the surround speakers overwhelm unless you crank down the trims.

FWIW, I did some experimentation just now with Dolby Surround and I can see why someone might prefer it to Auro2D. I do still prefer Auro3D though. I have identical L/C/R(all Genelec 8351Bs) and while I did note the difference between center spread on/off, I thought it was fine with center spread off, at least in the handful of recordings I listened to(some classical and pop).

 

[Kal Rubinson]

Auro 2D adds harmonic distortion and increases the bass trims and decreases the treble trims in a very measurable way. (At least on my Denon X6700H implementation) It’s not just adding ambiance, it’s changing EQ and adding 2nd order harmonic distortion for fuller sounding bass. It’s tailoring the sound. That’s not a bad thing, but it is worthy of mention and some transparency.

Agreed. I use it for TV and for the cable music streaming in my CT setup and I find its application "fleshes out the sound" without disturbing redistributions. OTOH, I do not use it for serious music playback from my file library, even with the stereo tracks because it makes the imaging cloudy.

I have tried both dts and Dolby upmixing in place of (and before I tried) Auro2D and rejected them because of disturbing source relocations. I have not yet tried Auro3D but the additional necessary speaker channels will be installed within a couple of weeks.

 

[Abe_W]

As it is out of the box, I prefer strict stereo over upmixed atmos.

However, with a few easy tweaks, Atmos can be made to surpass stereo.

Default Atmos upmixer has 2 main issues(imo):

1. It worsens the stereo center image. You lose that holographic sound to center image.
2. You hear sounds coming from behind you, which(at least for me) breaks the illusion that there is a performance happening in front of you.

Problem 1 can be mostly corrected by enabling "Center Spread" in the atmos upmixer settings via your AVR.

Problem 2 can be fixed by manually turning down the level of the rear surrounds(though you'll want to turn them back up for movies/games, which gets tiresome). How much you need to turn them down depends on your speakers and listening distance. For me it was around -4.5dB, but you'll just have to play with it. Find a good piece of music where your brain is perceiving that there are speakers playing music behind you, then slowly starting turning them down, little by little. At some point, it's almost like a switch flips in your brain. Your brain stops perceiving the sound as coming from speakers behind you, and starts perceiving the sound behind you as echoes coming from your front speakers.

Done right, I find it superior to stereo. Stereo imaging in the sweet spot is identical, but now is perceivable over several seats, and the sense of envelopment is far greater. You kinda get the benefits of both wide dispersion(stable image, sense of envelopment) and narrow dispersion(tight imaging) at the same time.

For instance, on my Yamaha processor, a batch of different tweaks/settings/levels etc can be saved under one of 2 "setting patterns". This prevents me from tweaking each time for upmixed music VS movies/games. I can select each pattern with the click of a button. If your processor has such features, it would be worth exploring.

 

[Kal Rubinson]

That's interesting. It is worth noting that Auro2D and Auro3D sound pretty different, with Auro3D being more configurable since you can adjust the amount of effect and the size of the replicated venue. Auro3D requires 4 height channels to activate.

I should have read further down that thread. I like Auro3D at a strength of around 8, so I guess I maybe don't like as much bass as the 7dB default for Auro2D.

I have not yet tried Auro3D but the additional necessary speaker channels will be installed within a couple of weeks.

Well, I am happy to say that I prefer Auro3D, even without the tweaking, simply to avoid the annoying bass boost in Auro2D. Useful for some stereo sources but not necessary with the better ones.

 

[Abe_W]

I have tried it (in a studio) and I greatly prefer high quality stereo reproduction. There may be some Atmos music material that works well, but for me music is usually an experience with a definite front where the performers are, and anything from the sides or rear is ambience or room reflections.

I have even worked in ambisonics as well as cinema formats. For me, stereo is fine for music. It is in cinema and VR where more immersive surround formats have an advantage.

So...i bought a nightclub/bar last year (screwed up time to get into this business because of covid, but, i had my reasons). I have been making some upgrades to the venue's audio and sat through a couple of band setups+rehearsals. Auditing from a few different spots, it sounds very little like a stereo experience. It sounds a whole lot like my atmos setup at home.

I am still learning a lot about venues. But, i don't buy this front stage stereo only crap argument for music anymore man. It is a vestige left over from an archaic unsophisticated time in sound reproduction that needs to be snipped off.

 

[tifune]

I really don't like the DTS: Neural-X upmixer for stereo because it hard pans vocals to the center channel(some people seem to like this, I really hate it) and activates the surrounds too much for my taste. Interestingly it does NOT do this with actual multi-channel music, and I like it for that. I should also note that many true 5.1 albums have significantly more bass mixed into them than stereo recordings of the exact same music.

Just wanted to echo this sentiment - I played around with it quite a bit because my room benefits from Front Wides, which Auro does not use. I wanted to like it, but in addition to above it's as if the sound phases in and out of existence across different speakers. It's not at all subtle and my largely untrained ears couldn't find any pattern to it for troubleshooting.

If you have front Wides but no heights, dolby with center spread is my preferred choice for 2ch upmixing. But, that's maybe only 20 people on the whole planet...

ultimately Auro3D is best and since the time this thread was started you can enable it with only Front Heights on Denon 6700 and 8500, presumably the rest of the current line but those are the only 2 I've tested.

 

[Chromatischism]

Well, I am happy to say that I prefer Auro3D, even without the tweaking, simply to avoid the annoying bass boost in Auro2D. Useful for some stereo sources but not necessary with the better ones.

Bass boost is still applied to Auro 3D. I have a Quick Select on my Denon that switches the sound mode and reduces the sub channel by -4 to -5 OTOH.