There is no point to distributing music in 24-bit/192kHz format

Article from : Xiph.Org ( Apr 4, 2013 )

Articles last month revealed that musician Neil Young and Apple's Steve Jobs discussed offering digital music downloads of 'uncompromised studio quality'. Much of the press and user commentary was particularly enthusiastic about the prospect of uncompressed 24 bit 192kHz downloads. 24/192 featured prominently in my own conversations with Mr. Young's group several months ago.
Unfortunately, there is no point to distributing music in 24-bit/192kHz format. Its playback fidelity is slightly inferior to 16/44.1 or 16/48, and it takes up 6 times the space.
There are a few real problems with the audio quality and 'experience' of digitally distributed music today. 24/192 solves none of them. While everyone fixates on 24/192 as a magic bullet, we're not going to see any actual improvement.
 

First, the bad news

In the past few weeks, I've had conversations with intelligent, scientifically minded individuals who believe in 24/192 downloads and want to know how anyone could possibly disagree. They asked good questions that deserve detailed answers.
I was also interested in what motivated high-rate digital audio advocacy. Responses indicate that few people understand basic signal theory or the sampling theorem, which is hardly surprising. Misunderstandings of the mathematics, technology, and physiology arose in most of the conversations, often asserted by professionals who otherwise possessed significant audio expertise. Some even argued that the sampling theorem doesn't really explain how digital audio actually works [1].
Misinformation and superstition only serve charlatans. So, let's cover some of the basics of why 24/192 distribution makes no sense before suggesting some improvements that actually do.
 

Gentlemen, meet your ears

The ear hears via hair cells that sit on the resonant basilar membrane in the cochlea. Each hair cell is effectively tuned to a narrow frequency band determined by its position on the membrane. Sensitivity peaks in the middle of the band and falls off to either side in a lopsided cone shape overlapping the bands of other nearby hair cells. A sound is inaudible if there are no hair cells tuned to hear it.

Above left: anatomical cutaway drawing of a human cochlea with the basilar membrane colored in beige. The membrane is tuned to resonate at different frequencies along its length, with higher frequencies near the base and lower frequencies at the apex. Approximate locations of several frequencies are marked.
Above right: schematic diagram representing hair cell response along the basilar membrane as a bank of overlapping filters.
 
This is similar to an analog radio that picks up the frequency of a strong station near where the tuner is actually set. The farther off the station's frequency is, the weaker and more distorted it gets until it disappears completely, no matter how strong. There is an upper (and lower) audible frequency limit, past which the sensitivity of the last hair cells drops to zero, and hearing ends.

Some Videos

Read more at : Xiph.Org

Nice article, thanks for sharing.

there's a lot going on beyond what the human ear can hear....
and it's not all about audio,
some of it is about resolution and math.

the math turns into audible payoff,
especially in the 32 bit floating point math,
but truth be told,
the better target would be 60kHz

The great audio myth : Why you dont need that 32bit DAC

batsbrew
there's a lot going on beyond what the human ear can hear....
and it's not all about audio,
some of it is about resolution and math.

 
And resolution beyond what humans can hear is irrelevant for audio intended for human listening.


And the bit depth and sample rate used for DSP processing can be higher than the delivery format when appropriate. 

batsbrew
the math turns into audible payoff,
especially in the 32 bit floating point math,
but truth be told,
the better target would be 60kHz

 
So, assuming you understand the math you're referring to, can you tell us roughly how many calculations might become audible at a given bit depth?

batsbrew
NO.
 
but i don't really care about that.
 
i do believe, that you can't do any harm by having the highest resolution available for recording...
 
playback, though, is another matter.
 

I would say that one can do harm if one convinces others that they need to do or buy (!!!) things that offer no benefit in the real world.
 
And sometimes people (without realizing it) make themselves look silly by claiming to hear stuff that couldn't possibly be audible. I don't think it's fair to either let people embarrass themselves or buy stuff that doesn't really offer any benefit simply because they don't understand how the math involved in digital audio relates to the real world.
 
OTOH there is of course nothing wrong with going overkill if there is little or no downside involved. 24bit recording is a good example of this, as it allows one to leave lots of headroom to avoid clipping without having to worry about quantization errors.

fret_man
Higher samples rates are useful during non-linear processing. Think limiting, compression, saturation. These all cause distortion which can cause aliasing. For example, the 5th harmonic of 10kHz is 50kHz, which is inaudible. But if this is processed at 48kHz, then the alias frequency will be 50k-48k = 2kHz, which IS audible. In my opinion that's the main benefit of higher sample rates. But now-a-days, those plugins use built-in sample rate conversion to do their non-linear processing at the higher rates needed to prevent aliasing.

 
Sure, but the the article being discussed is about delivery/playback formats.
 

In fact, higher rates are more susceptible to clock jitter effects, so things may sound better at lower rates anyway.

No, I think you mean higher frequencies (meaning the audio itself, not the sampling rate) are more susceptible to jitter. But jitter, at least in terms of audio sampling, has basically been irrelevant for quite some time now anyway.

Factors for determining bit depth needed in the real world:
 
1. Quantization error + dither (i.e. SNR of the medium).
2. Peak to average ratio of the audio plus any headroom (i.e. how far below 0 dBFS the average level is).
3. Average playback level in the listening environment (dB SPL).
4. Noise in playback equipment and listening environment.
[EDIT] 4a. Source noise in the recording.
5. Absolute threshold of hearing (ATH).
 
Once the level of the QE + dither is below #4 or #5 (with a reasonable margin for error), you're done. Increasing bit depth can't add resolution because the resolution is in fact limited by #4 and/or #5 and not the bit depth.

Is it the 24/96 or the "_mastered_ for vinyl" that makes it sound better than cd (or mastered for cd)?

Here's a honest plugin developer :

For many years, processing speed, sample rates and copy protection were limiting.  Things have changed.  Digital Technology can now come very close to having all the wonderful non-linearity and “warts” of the best analog gear.  We say “close” because most of *******’ analog gear will pass 150KHz easily, and that is something that is impossible with current digital technology.  But hey, who hears above 20KHz?  ::Spock Eyebrow Raise::  But it is difficult getting everything right up to that 20KHz, especially at sample rates of 48KHz and below.

Want to know who it is ?, just copy the text and google it ;-)

Ok it just music peeps.