• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Is high-resolution audio audible or not audible and a waste of data?

markanini

Major Contributor
Joined
Feb 15, 2019
Messages
1,819
Likes
1,915
Location
Scania
To some extent I disagree, but one has to be someone who built a perceptual codec to do much. :) But yes, testing a perceptual codec requires perception.
You're right, let me retract and say looking at graphs doesn't tell enough, in normal circumstances.

Because an efficient perceptual codec takes advantage of auditory masking. It will appear like holes in the spectrogram, but will be hard to ABX. The strategy has left enough bandwidth to allocate for harmonically complex and transient signals when needed.

As a contrasting example, a badly implemented, or configured, codec may have smearing artefacts that are easy to ABX. Yet the spectrogram will appear intact to Nyquist on a spectrogram when zoomed out, as most view them.
 
Last edited:

ZolaIII

Major Contributor
Joined
Jul 28, 2019
Messages
4,233
Likes
2,496
This sound capture would be better in 24-bit than the 16-bit attached.
Whereas the 24 bit? Let me shawe (it down to about 20 bits) and pack it for you (WavPack lossy) will be smaller than 16 bit Flac (tho there are bunch of zeros there anyway as it's only two fuss impulses). Anyway you won't be able to recognise difference both audio or visual (to point it on graph).
 

pma

Major Contributor
Joined
Feb 23, 2019
Messages
4,661
Likes
10,870
Location
Prague
Whereas the 24 bit? Let me shawe (it down to about 20 bits) and pack it for you (WavPack lossy) will be smaller than 16 bit Flac (tho there are bunch of zeros there anyway as it's only two fuss impulses). Anyway you won't be able to recognise difference both audio or visual (to point it on graph).
The posted file is 16 bit, not 24 bit. With similar signals, the ABX tests 16/24 and also 44.1/96 were positive. With usual pop music, they are not, of course.
 

ZolaIII

Major Contributor
Joined
Jul 28, 2019
Messages
4,233
Likes
2,496
The posted file is 16 bit, not 24 bit. With similar signals, the ABX tests 16/24 and also 44.1/96 were positive. With usual pop music, they are not, of course.
I did a bit more elaborate test with a native 96/24 track (Peanut Vendor) about two years ago and I wanted to assure you in my findings (audio visual not being able to pin the difference) how (limited) compression can be applied smartly, that's all.
 

j_j

Major Contributor
Audio Luminary
Technical Expert
Joined
Oct 10, 2017
Messages
2,312
Likes
4,844
Location
My kitchen or my listening room.
Because an efficient perceptual codec takes advantage of auditory masking. It will appear like holes in the spectrogram, but will be hard to ABX. The strategy has left enough bandwidth to allocate for harmonically complex and transient signals when needed.

As a contrasting example, a badly implemented, or configured, codec may have smearing artefacts that are easy to ABX. Yet the spectrogram will appear intact to Nyquist on a spectrogram when zoomed out, as most view them.

All of that is pretty much true, although a good codec will usually not leave holes, rather it will add noise in places it's not going to be audible. Your comment about a BAD codec is spot on.

I can't pass it along, but if you know somebody who's heard the "13 dB miracle" that rather graphically makes the point.
 

markanini

Major Contributor
Joined
Feb 15, 2019
Messages
1,819
Likes
1,915
Location
Scania
although a good codec will usually not leave holes, rather it will add noise in places it's not going to be audible.
Technically yes. Viewed at a glance, the output on a spectrogram will look like some ranges are chiseled out to a black abyss, and still everything is working as it should.
At least I have failed to ABX samples that had that appearance, along with others that participated in public listening tests.
 

j_j

Major Contributor
Audio Luminary
Technical Expert
Joined
Oct 10, 2017
Messages
2,312
Likes
4,844
Location
My kitchen or my listening room.
Technically yes. Viewed casually, the output on a spectrogram will look like some ranges are chiseled out to a black abyss, and still everything is working as it should.
At least I have failed to ABX samples that had that appearance, along with others that participated in public listening tests.

That's reasonably unlikely, but it can happen. More often you'll see more noise below the peaks in the spectrum, but it can happen.
 

lowgain

Member
Joined
May 26, 2019
Messages
31
Likes
69
Comparing lossy audio by graphs is flawed. It says nothing about detectable differences in a DBT with a PCM reference.
To be fair, I would say that drawing conclusions is flawed, but you can compare anything. English is not my mother tongue, maybe the word comparing includes a notion of judging?

Anyway, at the time, I thought that modern codecs wouldn't cut frequencies, and that's why I initially did the spectrograms. Since Tidal didn't say what they used, I wanted to show that at least both Spotify and Tidal were using some advanced lossy encoding.

I was surprised to see that the results for Tidal, but also noted that I didn't properly blind test them. Now, Tidal does say which lossy encoding they use.
 

lowgain

Member
Joined
May 26, 2019
Messages
31
Likes
69
So it was tested Spotify at its highest setting and Tidal at it's highest lossy setting "high" in the desktop application.
CD // Spotify "Very high" // Tidal "High"

Maybe you should rename your uploaded pictures so there can be no misunderstandings what is compared.
I can't seem to edit that post any longer?
 

Rufus T. Firefly

Active Member
Forum Donor
Joined
Sep 8, 2021
Messages
155
Likes
107
Location
St. Louis
I was tested at the University of Ghent (Belgium) because i could hear "exceptional" high they said, but my first test at the age of 10 was just above 19kHz. They started to test me because i complaint about high pitched sounds of old tv's and other electronic devices that nobody hears. Never heared about people who hear above 20kHz. Do you have links to those studies?

Btw, now at the age of 43 i still can hear until 16854Hz (test done a few months ago). And that is also exceptional they say at the university, especially for a former dj and sound engineer who worked often in very loud envirroments. But i don't see myself as a golden ear, it's more a problem because i hear all kind of stuff i don't want to hear. I can also hear very silent sounds, that normal humans can't, and that is a big problem. I don't want to hear the insects that live in my house (in every house) when i'm in my bed, i want silence.
I had a similar experience decades ago when I asked a salesperson about the high pitched noise in their store. She said it was their security system and it was very rare that anyone could hear it.

I didn't have a hard time like you do with unwanted noises and quite enjoyed have a what I considered a great sense of hearing...until I developed a mild case of tinnitus from taking a Z-pak. Now I have a constant ringing which sounds horrible but isn't really annoying except when there aren't other sounds to cover it up.
 

Tangband

Major Contributor
Joined
Sep 3, 2019
Messages
2,994
Likes
2,811
Location
Sweden
Perhaps it's time to remind people (not necessarily you) that the actual noise due to air molecules hitting your ear drum (which is what makes "pressure" exist, of course) is somewhere in the neighborhood of 6 to 8 dB SPL (flat weighted). No, you can't hear that, barely, because of the white spectrum of that particular noise (the noise floor here has some other components, which is common everywhere, it's hard to block low frequencies) has little energy in any given ERB.

But for 20-20K that's a solid minimum noise level. The only way to get rid of that noise is to remove the air from both sides of the eardrum, and I suspect the listener would be upset, at the very least. It's not an experiment to actually attempt.
True. Me and many more having a slight tinnitus has much worse ”ear sinad ” than that. The microphones and the recording rooms are analog and has a certain noise level.
But, as the technology is available , there is no reason to not use 24 bit or more in the studio.
 

HarmonicTHD

Major Contributor
Joined
Mar 18, 2022
Messages
3,326
Likes
4,846
True. Me and many more having a slight tinnitus has much worse ”ear sinad ” than that. The microphones and the recording rooms are analog and has a certain noise level.
But, as the technology is available , there is no reason to not use 24 bit or more in the studio.
Studios use 24 bit or now even 32 bit recording / processing for convenience, efficiency and having all the flexibility in subsequent digital mastering / mixing. Sound quality is not the driving factor.
 

j_j

Major Contributor
Audio Luminary
Technical Expert
Joined
Oct 10, 2017
Messages
2,312
Likes
4,844
Location
My kitchen or my listening room.
Studios use 24 bit or now even 32 bit recording / processing for convenience, efficiency and having all the flexibility in subsequent digital mastering / mixing. Sound quality is not the driving factor.

Using double float is the way to go. It's not that hard, and you know that keeps the calculation noise out of your final 16 bits.

Doing processing in 16 bit fix, or even 32 fix, is just bad practice.
 

HarmonicTHD

Major Contributor
Joined
Mar 18, 2022
Messages
3,326
Likes
4,846
Using double float is the way to go. It's not that hard, and you know that keeps the calculation noise out of your final 16 bits.

Doing processing in 16 bit fix, or even 32 fix, is just bad practice.
Thanks for adding the scientific explanation, not something I am deeply familiar with.

My comment was more from a practical point (studios). Setting up recordings quicker, dialing in levels and being more forgiving for mistakes and running as many digital effects in the DAW (mostly 32 or 64Bit) to one’s heart content, saves time and money.
 

sweetsounds

Active Member
Joined
Apr 24, 2019
Messages
146
Likes
294
Yes, I had already seen this interesting video some time ago. But the theme here is slightly different: we are talking about the time spread of transients when switching from signals with a high sampling frequency to a lower one and the fact that our auditory system could detect this effect in the form of an alteration of the localization of the sources in the space.
Hi,

I know, it is confusing, but people are having a wrong picture how digitalization works and forget the step of bandwidth limiting the signal first.

So the sequence is
- bandwidth limiting / low-pass of 22kHz
- analog digital conversion
- digital-analog conversion
- reconstruction filter

The last step is also rightfully looked at by Armin in the DAC tests.

The band width limiter modifies the original signal, it smoothens it out. That's why timing accuracy is much better than 5 us.

All the rest is math:
Fourier transformation (100s of years old: the time and the frequency domains are identical descriptions of signals).
Nyquist-Shannon theorem: a band-width limited signal can be perfectly represented with twice its bandwidth. This means digital and analog are the same.

So what is the learning nugget: the only limitation of 16/44.1kHz is the bandwidth limitation and the dynamic range of 96dB to 100dB, not the fact that there is discrete time sampling/digital.
The reconstructed signal is 100% the same as the bandwidth limited input signal, if you have a steep reconstruction filter.


By the way, vinyl does the same, it is dynamic-range-limited to approx. 70dB (resolution of perfect vinyl) and bandwidth limited by RIAA encoding and the medium. Vinyl could reproduce 50kHz. However, because of the cutting process significant errors/dynamic jumps are introduced. With Vinyl you hear more artefacts, but some people might prefer these artefacts.

Notice: vinyl is a mechanical process, the stylus will not be able to follow the groove exactly and jumps. This can add fake dynamic range to records. Don't know if it's audible, but vinyl definitely adds artefacts during wriitng and playing due to mechanical interactions.
 

Germinal

Member
Joined
Jan 5, 2022
Messages
24
Likes
15
I had a hard time telling apart 320kb/s mp3 files from WAV files. I'm sure distinguishing 16 bit 44kHz wav files from higher resolution files would be even harder for me.
I'm streaming TIDAL/Spotify/play CD and LP.. and I wouldn't be able to tell the difference in a blind test really.. all sounds good.. and my gear ain't cheap :s

On Qobuz I listened to some 192khz.. nah.. I don't hear it...

must be my ears
 
Top Bottom