Skip to main content

The 13.4.1 S2 update added hi-res (Ultra HD) and Dolby Atmos audio support from Amazon Music Unlimited. With this update, Sonos released this great article about hi-res audio and how you can listen to it on Sonos. It’s a very detailed and well-written article:

https://blog.sonos.com/en-us/hi-res-audio-guide

In the US we have had some regulation of the audio amplifier power rating since 1974. This is a look at the rule and the problems.

Many manufacturers have taken advantage of this vacuum by publishing a confusing array of unrealistic power claims. Some go so far as to slap a sticker on the front panel with an inflated power figure that's based on only one-channel driven at 6-ohms and 10% THD.

https://www.audioholics.com/amplifier-reviews/ftc-amplifier-rule-help-protect-home-audio-consumers-today

 

If the industry won’t establish and enforce standards then we are reduced to getting government involved which is rarely the best solution.


Cite what??

 

Actual evidence for your claim.  

 

If you are refering to my last statement about the audible difference between currently available HD and UltraHD tracks, that is more a conclusion, based on those fundamentals of digital signal porcessing which I tried to briefly summarize before, rather than a claim.

 

You’ve clearly done enough research on the topic to know that what you’ve stated is far from accepted fact. You had to know that if you post a theory like this on a forum with others knowledgeable on the subject, it’s not going to be blindly accepted.

Shame on me :)

If you are refering to my last statement about the audible difference between currently available HD and UltraHD tracks, that is more a conclusion, based on those fundamentals of digital signal porcessing which I tried to briefly summarize before, rather than a claim.

 

Your "fundamentals of digital signal processing" are gibberish.  You made some claims that are not backed up by any mathematical or practical evidence.  All evidence points to there being no benefit of either 24 bits or sample rates over 48 kHz in the playback of digital audio.

I basically agree with the statement above. And while it is fairly obvious (at least to me) that there is no audible benefit by increasing the bit/sample resolution form 16 to 24, I don’t think there is a similar conclsuion on the increase in sampling rate, yet. For 2D stereo recordings and stationary sound sources there are some studies out there suggesting that there is no audible benefit from increasing the sampling rate above 48kHz. This is where we might just stop the discussion with a simple 16bit/44.1kHz is good enough, period.

OTOH the psychoacoustic propeties around the detection of minimum audible angle and depth of a sound source from low to high frequencies are still very much under investiagtion and not entirely understood. All I was stating is, that when the CD was developed the focus was primarily on accurate reconstruction of the amplitude spectrum while the phase spectrum was of lower importance.  


If you are refering to my last statement about the audible difference between currently available HD and UltraHD tracks, that is more a conclusion, based on those fundamentals of digital signal porcessing which I tried to briefly summarize before, rather than a claim.

 

Your "fundamentals of digital signal processing" are gibberish.  You made some claims that are not backed up by any mathematical or practical evidence.  All evidence points to there being no benefit of either 24 bits or sample rates over 48 kHz in the playback of digital audio.


Cite what??

 

Actual evidence for your claim.  

 

If you are refering to my last statement about the audible difference between currently available HD and UltraHD tracks, that is more a conclusion, based on those fundamentals of digital signal porcessing which I tried to briefly summarize before, rather than a claim.


He really made the point why 192kHz sampling frequency is required if you want to accuately reproduce the gun fire of a laser blaster flying across a cinema theatre.   

But is it necessary to render an improvement, audible in a controlled blind listening test, to a 2 channel music recording? 

 


This is why digital audio with increased sampling rates of 96kHz or even 192kHz would indeed provide a very noticeable benefit as it allows for more precise positioning and depth of the sound sources.

On that basis the difference between Red Book and Hi Res in any blind test should be like ‘night and day’, and yet: https://www.realhd-audio.com/?p=6993

The author nails it in this paragraph: “Is I’ve often stated in these articles, it is the production path that establishes the fidelity of the final master. Things like how a track was recorded, what processing was applied during recording and mixing, and how the tracks were ultimately mastered. If all of these things are done with maximizing fidelity as the primary goal, a great track will result.”

Again, 16bit/44.1kHz is completely sufficient in terms of fidelity, as it keeps the quantization noise low enough (-96dB) and reproduces all the audible frequencies (up 22.05kHz).

I should have said, that most current music productions do not take full advantage of the higher spatial resolution you get when using 96kHz sampling frequency and it’s debatable whether a rock/pop production would ever exploit it. With classic music, when done properly, you can definetly hear it.

You should read the methodology in detail. The test samples auditioned by several hundred individuals compared the full resolution originals with RedBook equivalents. The conclusion speaks for itself: “Hi-Res Audio or HD-Audio provides no perceptible fidelity improvement over a standard-resolution CD or file. “

The author of that study originally set up a recording label specifically for the production of high-resolution audio, from recording to delivery. He of all people would surely have wished that Hi Res was detectable. According to his findings it wasn’t.


My quick summary:

Regarding 16 vs 24bit/sample resolution: As a streaming/transport format there will be no audible gain as long as studios are finally compressing the dynamic range of their master recording to fit into the 96dB provided by a 16bit representation of a signal. Also, as has been said before, I doubt anyone (even with golden ears) can hear the difference as 96dB SNR sounds "fantastic" while 144dB (which is what you theoretically get from 24bit) is just overkill. However, as an internal format in studio- as well as in listetning equipment 24bit/sample resolution makes total sense for doing proper volume control and eq in the digital domain. But this is happening anyways even if your transport format is "just" 16bit/sample.

Regarding sampling rate, the discussion is a litte different though:

There seems to be common consensus that a sampling frequency of 44.1kHz is sufficient to accurately reproduce the audible frequency spectrum. In fact, according to the Nyquist Theorem this allows for reproducing frequencies up to 22.05kHz and only young children can hear frequencies above 20kHz while the hearing of an average adult Joe is capped at 18 or even just 16kHz. So, all good here? Well not quite...

Ever since the CD appeared in the 80's many audiophiles keep claiming that a good analog record still offers more accurate reproduction of the sound stage and more precise positioning and depth of the instruments. They are right!

This is because there is a (incorrect) notion that equates the frequency spectrum with only the amplitude spectrum but it neglects the corresponding phase spectrum. While it's true that the human ear (and brain) cannot hear the amplitude of frequencies above, say, 18kHz our two ears can extremely well detect phase differences between frequencies that are much higher! So while we cannot hear those frequencies as tones, we can detect the tiny differences in runtime which it takes those inaudible frequencies to arrive at the left and right ear respectively. In other words, our spatial location capabilities are of much higher resolution than our frequency hearing capabilities. Btw, this effect is heavily used by 3D sound systems like Dolby Atmos or THX.

This is why digital audio with increased sampling rates of 96kHz or even 192kHz would indeed provide a very noticeable benefit as it allows for more precise positioning and depth of the sound sources.

I say "would" because every track from Amazon labeled "Ultra HD" which I have seen (or been listeing to) so far is just 24bit/44.1kHz. So it gives me the "useless" 24bit/sample resolution but falls short of providing higher sampling rates which could really make an audible difference.

 

As long as you get HD it’s “fantastic”, there is currently no audible difference to “Ultra HD”. My hope is they provide more and more content higher sampling frequencies in the future. Then it will make a difference!
    

 

 

Cite?

Cite what??


This is why digital audio with increased sampling rates of 96kHz or even 192kHz would indeed provide a very noticeable benefit as it allows for more precise positioning and depth of the sound sources.

On that basis the difference between Red Book and Hi Res in any blind test should be like ‘night and day’, and yet: https://www.realhd-audio.com/?p=6993

The author nails it in this paragraph: “Is I’ve often stated in these articles, it is the production path that establishes the fidelity of the final master. Things like how a track was recorded, what processing was applied during recording and mixing, and how the tracks were ultimately mastered. If all of these things are done with maximizing fidelity as the primary goal, a great track will result.”

Again, 16bit/44.1kHz is completely sufficient in terms of fidelity, as it keeps the quantization noise low enough (-96dB) and reproduces all the audible frequencies (up 22.05kHz).

I should have said, that most current music productions do not take full advantage of the higher spatial resolution you get when using 96kHz sampling frequency and it’s debatable whether a rock/pop production would ever exploit it. With classic music, when done properly, you can definetly hear it.

Anecdotically, I remember when hearing a keynote from one of the invertors of THX at an IEEE signal processing conference back in 2000. He really made the point why 192kHz sampling frequency is required if you want to accuately reproduce the gun fire of a laser blaster flying across a cinema theatre.   


My quick summary:

Regarding 16 vs 24bit/sample resolution: As a streaming/transport format there will be no audible gain as long as studios are finally compressing the dynamic range of their master recording to fit into the 96dB provided by a 16bit representation of a signal. Also, as has been said before, I doubt anyone (even with golden ears) can hear the difference as 96dB SNR sounds "fantastic" while 144dB (which is what you theoretically get from 24bit) is just overkill. However, as an internal format in studio- as well as in listetning equipment 24bit/sample resolution makes total sense for doing proper volume control and eq in the digital domain. But this is happening anyways even if your transport format is "just" 16bit/sample.

Regarding sampling rate, the discussion is a litte different though:

There seems to be common consensus that a sampling frequency of 44.1kHz is sufficient to accurately reproduce the audible frequency spectrum. In fact, according to the Nyquist Theorem this allows for reproducing frequencies up to 22.05kHz and only young children can hear frequencies above 20kHz while the hearing of an average adult Joe is capped at 18 or even just 16kHz. So, all good here? Well not quite...

Ever since the CD appeared in the 80's many audiophiles keep claiming that a good analog record still offers more accurate reproduction of the sound stage and more precise positioning and depth of the instruments. They are right!

This is because there is a (incorrect) notion that equates the frequency spectrum with only the amplitude spectrum but it neglects the corresponding phase spectrum. While it's true that the human ear (and brain) cannot hear the amplitude of frequencies above, say, 18kHz our two ears can extremely well detect phase differences between frequencies that are much higher! So while we cannot hear those frequencies as tones, we can detect the tiny differences in runtime which it takes those inaudible frequencies to arrive at the left and right ear respectively. In other words, our spatial location capabilities are of much higher resolution than our frequency hearing capabilities. Btw, this effect is heavily used by 3D sound systems like Dolby Atmos or THX.

This is why digital audio with increased sampling rates of 96kHz or even 192kHz would indeed provide a very noticeable benefit as it allows for more precise positioning and depth of the sound sources.

I say "would" because every track from Amazon labeled "Ultra HD" which I have seen (or been listeing to) so far is just 24bit/44.1kHz. So it gives me the "useless" 24bit/sample resolution but falls short of providing higher sampling rates which could really make an audible difference.

 

As long as you get HD it’s “fantastic”, there is currently no audible difference to “Ultra HD”. My hope is they provide more and more content higher sampling frequencies in the future. Then it will make a difference!
    

 


Shame on me :)

I basically agree with the statement above. And while it is fairly obvious (at least to me) that there is no audible benefit by increasing the bit/sample resolution form 16 to 24, I don’t think there is a similar conclsuion on the increase in sampling rate, yet. For 2D stereo recordings and stationary sound sources there are some studies out there suggesting that there is no audible benefit from increasing the sampling rate above 48kHz. This is where we might just stop the discussion with a simple 16bit/44.1kHz is good enough, period.

OTOH the psychoacoustic propeties around the detection of minimum audible angle and depth of a sound source from low to high frequencies are still very much under investiagtion and not entirely understood. All I was stating is, that when the CD was developed the focus was primarily on accurate reconstruction of the amplitude spectrum while the phase spectrum was of lower importance.  

 

CIte for the following definitive statement in your original post, please:

While it's true that the human ear (and brain) cannot hear the amplitude of frequencies above, say, 18kHz our two ears can extremely well detect phase differences between frequencies that are much higher! So while we cannot hear those frequencies as tones, we can detect the tiny differences in runtime which it takes those inaudible frequencies to arrive at the left and right ear respectively. In other words, our spatial location capabilities are of much higher resolution than our frequency hearing capabilities. Btw, this effect is heavily used by 3D sound systems like Dolby Atmos or THX.

 

Because in your above quoted post, you seem to say the claim is “still very much under investiagtion (sic) and not entirely understood.”  So which is it, a definitive fact or something to be investigated?

BTW, THX is a quality standard, not a codec like Atmos or DTS.


In a broader context home audio is a mess anyway where definitions are concerned - HiFi and Audiophile quality as two classic examples. So in that tradition even someone like Amazon plays fast and loose with the definition of HD, applying it to the CD format. And if I am not mistaken, Sonos has also jumped on that HD bandwagon thereafter.

PS: And of course, the total mess over definition of output power in watts ranging from rms to PMPO...that one is a doozy.


Agreed, Kumar, but even an agreement on the nature of the snake oil nostrum might benefit folks who look for these false “key words” in marketing statement might ease some of the posts we deal with. 


 

I get where you’re coming from, and support it, but getting various companies to pay attention to a “standard” as opposed to their own bottom line, and marketing strategies seems like a false hope. 

Won’t work probably because it is snake oil to start with, in terms of what is audible about it. As opposed to the HD and beyond video side of things where there is a universally adopted standard definition for things like HD Ready, HD, 4K etc,


Because it worked so well with the CEC consortium, and the various issues many manufacturers have had implementing that?

I get where you’re coming from, and support it, but getting various companies to pay attention to a “standard” as opposed to their own bottom line, and marketing strategies seems like a false hope. 


Why can’t the music industry standardize these designations and require compliance. Not sure who or how it would be done, just want transparency and honesty not marketing speak. 


I guess it’s a case that people want to know that they are getting the quality of audio they are paying for. Whether it’s actually needed, or not, becomes a separate question. 

Some say in some circumstances, that they can hear a difference between HD (16-bit) audio and UltraHD (24-bit) audio.. I’m not one of those people. I have been quite happy in the past with the lesser 320 AAC lossy audio, so I’m more than happy with the Amazon HD standard.

It’s not actually costing me any more to have the higher quality Amazon audio anyway on Sonos… (24/48 Atmos/UltraHD audio in some instances) …and as long as it all plays without interruption and sounds good (to me), then I’m okay with that, especially as a Prime Movie/Music Unlimited annual subscriber, this ‘merged’ Amazon music service is actually costing me £50 (per year) less than before.


Further to the preceding - I believe that this Hi Res thing is being peddled just to sow dissatisfaction among vulnerable users of technology which has peaked in terms of what it can objectively deliver, such that they spend more money to pad the pockets of these peddlers. 

Instead of doing what none of the peddlers of hardware and services are doing to address the big issue which causes much more aggravation - distracting sound level changes from one track to the next when playing playlists while using the random shuffle feature. Having to constantly move the volume control around to deal with this is very aggravating and damages the listening experience.

Atmos is a different case; that does sound very different from HD/SD, but there too personal preference may mean that it isn't an improvement. But it at least does not need labels to announce itself; the sound does that job.