24bit vs 16bit, the myth exploded!
Aug 15, 2021 at 3:48 PM Post #6,271 of 6,480

71 dB

Headphoneus Supremus
Joined
Sep 17, 2017
Posts
1,690
Likes
557
Location
Helsinki, Finland
What you are showing here works with a simple sinewave. However, it is not relevant because an actual musical waveform is much more complicated than a sinewave.
Fourier demonstrated that it can be represented with sinewaves, but this series of sinewaves must generally be infinite.
Sinewaves are not magically simpler in this sense. Complex musical signals are just as combination of sinewaves so if something works for one sinewave, it works for a million sinewaves. To proof this I did the same for music, now in STEREO! Look and be amazed!

Band-limited signals have a finite/limited series of sinewaves.

samples2.png
 
Last edited:
Aug 16, 2021 at 4:08 AM Post #6,272 of 6,480

gregorio

Headphoneus Supremus
Joined
Feb 14, 2008
Posts
4,125
Likes
2,244
The process is much more random than you think.
What I think is irrelevant, it's what the science proves/demonstrates that's important here and what it proves is that it is NOT random.
The dispatching of the elementary sinewaves in the time domain in totally random. This is the reason why you can always have information between two samples, even though the elementary components are band limited.
I have no idea what you mean by the "dispatching of the elementary sinewaves in the time-domain". There are no "elementary sinewaves in the time-domain" unless you're reproducing nothing but a single sine wave.
What we sample is not sinewaves but the surface of the waveform.
No, what we sample is the amplitude of the waveform.
If you don't sample this surface with a very high precision, then the reproduction of the sinewaves that are inside is distorted.
Now you are contradicting yourself again. In another thread you repeatedly state how good DSD is but DSD has just one bit resolution, the very lowest precision that's possible in digital audio! So which is it, do you need very high precision, in which case DSD must be absolutely terrible or is your statement above false?
We need to be aware that the AD converter doesnt have any access to the content of the waveform.
Of course it has access to the content of the waveform. If it has no access to the content of the waveform then what do you think a converter is converting? Even by your own incorrect definition of "what we sample", isn't this "surface of the waveform" part of the content of the waveform?
What is under the waveform is band limited, but the surface is not.
There is no under or surface, there's just the waveform and a waveform is either band limited or it's not, it cannot be both at the same time.
We generally admit that people older than 30 cannot hear anyting above 16 KHz. Then we could sample a good analogue master on tape at a sample rate of 32 Khz and see if it sounds as good as 44.1 KHz.
Not sure what you mean by "we could" because science and engineers have already done that. In fact it was standard practice in some parts of the audio industry at one time, so it's probably been done millions of times! Although under controlled test conditions probably only a few thousand times or so.
We can also try to interprete the sonic results achieved with DSD.
Why do we need to interpret the sonic results achieved with DSD when we can precisely measure the sonic results objectively.
This technique is very different from PCM and the sound depends a lot on the analogue filtering after the bitstream
No, it's very similar to PCM and one of the main points of DSD was to reduce the quality requirements of the analogue filtering. So again, pretty much the exact opposite of what you're falsely stating!
We need to understand that the waveform is the product of a mathematical function.
Why do we need to understand a statement that is false?
Auditions do confirm my theory.
No they don't. Controlled tests demonstrate no one can tell the difference, in some sighted tests the subjects can't tell the difference and in some they can but even these cases do NOT confirm your theory! What subsquent controlled testing confirms is the theory of placebo effect.
People who have learnt the basics of traditional digital audio generally consider for certain that a sample rate above 48 KHz cannot benefit to sound reproduction and they are submited to a nocebo effect because hearing the difference would lead them to reconsider what they learnt and this often makes them uncomfortable.
Which is why we haven't only tested scientists and audio engineers but also the public, audiophiles and students who have no knowledge of the basics of digital audio. The results are the same!
High resolutions audio formats are becoming more and more successful, with offerings from Apple, Qobuz, Amazon and Tidal. The people who are ready to pay a premium price for them are expected to hear the benefit they bring to the sound. Why would they pay more for they same sound quality ?
Isn't that obvious? If marketing can convince them that the sound quality is better (even though it isn't) people will pay more. A very large part of the audiophile world is entirely based on this fact! Did you not know that?

G
 
Aug 16, 2021 at 5:53 AM Post #6,273 of 6,480

71 dB

Headphoneus Supremus
Joined
Sep 17, 2017
Posts
1,690
Likes
557
Location
Helsinki, Finland
Isn't that obvious? If marketing can convince them that the sound quality is better (even though it isn't) people will pay more. A very large part of the audiophile world is entirely based on this fact! Did you not know that?

G
Some people are predominantly sensing -type meaning they absorb information "as it is" while some other people are intuitive -type meaning they analyse information and seek for hidden meanings within it. Sensing- types are typically better at detail while Intuitive -types are often better at critical evaluation of the information. Sensing -types, especially those who are also feeling -types rather than thinking -types are easier to convince of unwarranted claims and are therefor potential clients for snake oil sellers.

People experience the World differently and process information differently. What is self-evident for some may not be self-evident for some others. Myself being an INTJ explains why my way of thinking about audio is "Is this good enough?", "Why pay more than X euros for headphones when X euros is about the price category giving the most bang for the buck?" etc. Given my personality type and my knowledge of digital audio, I am one of the last people on Earth to be convinced of the "benefits" of hi-rez audio. This gives me the insight that xSFx -personality types without knowledge of digital audio are probably massively more easily convinced of the "benefits" of hi-rez audio.

It would be interesting to see studies of the correlations between personality types and the way people do audiophile hobby. I did Googling, but didn't find anything good.
 
Last edited:
Aug 16, 2021 at 7:36 AM Post #6,274 of 6,480

audiokangaroo

Head-Fier
Joined
Jul 26, 2021
Posts
70
Likes
2
Location
France
Gregorio, I'm not sure that you have a very good understanding of Fourier analysis.
The signal is made of elementary sinewaves. The waveform is the surface of this phenomenon. It is the output of a mathematical fonction.
The input of the fonction consist in sinewaves from the audible band and the position of these elementary signals in the time domain.
If you don't agree, what is you explanation of the waveform ?
 
Last edited:
Aug 16, 2021 at 11:54 AM Post #6,275 of 6,480

castleofargh

Sound Science Forum Moderator
Joined
Jul 2, 2011
Posts
9,483
Likes
4,875
Gregorio, I'm not sure that you have a very good understanding of Fourier analysis.
The signal is made of elementary sinewaves. The waveform is the surface of this phenomenon. It is the output of a mathematical fonction.
The input of the fonction consist in sinewaves from the audible band and the position of these elementary signals in the time domain.
If you don't agree, what is you explanation of the waveform ?
I wish you could join us back into the real world at some point.

You replied to my last comment that you didn't think you were overconfident. This might just be your biggest mistake. And I’m saying that right after quoting complete nonsense about the surface of an audio waveform. Alluding to what, audio fishes swimming below?
Goggle how a microphone works before posting even more science fiction. You have mistaken almost everything that can be so far, and that while people were explaining it to you in details. Forget about Nyquist, Fourier, and sampling for now. start at the beginning. What's a mic? What can it do? What signal can it deliver? Spoiler: no fish.
 
Aug 16, 2021 at 1:35 PM Post #6,277 of 6,480

castleofargh

Sound Science Forum Moderator
Joined
Jul 2, 2011
Posts
9,483
Likes
4,875
Darn I was hoping to combine two of my favorite hobbies, fishing and audio. :slight_smile:
I was only talking about the electrical signal coming from a mic. But with a little Pono imagination, there is still hope.

UCpS8VM_d.webp

Apparently, hornitophiles need 384kHz to grow wings and fly with the birdies.
 
Aug 16, 2021 at 2:27 PM Post #6,278 of 6,480

audiokangaroo

Head-Fier
Joined
Jul 26, 2021
Posts
70
Likes
2
Location
France
I wish you could join us back into the real world at some point.

You replied to my last comment that you didn't think you were overconfident. This might just be your biggest mistake. And I’m saying that right after quoting complete nonsense about the surface of an audio waveform. Alluding to what, audio fishes swimming below?
Goggle how a microphone works before posting even more science fiction. You have mistaken almost everything that can be so far, and that while people were explaining it to you in details. Forget about Nyquist, Fourier, and sampling for now. start at the beginning. What's a mic? What can it do? What signal can it deliver? Spoiler: no fish.
What do you think I didn't understand about the way a mic works ?
I'm open to learn new things if you have something to explain. However, the waveform problem is exactly the same at the voltage level and at the air pressure level.
The waveform is the result of a process. I don't want to be overconfident, but this process seems to have a combinatory dimension and a random dimension.
I don't think that explaining the production of the waveform with a mathematical function is a wrong idea.
You have the right to compare my theory with science fiction, but it would be more useful if you could explain precisely what is wrong among the things I tried to explain.
In my opinion, considering that the waveform has the same frequency content as the audio band is simplistic.
I can give you again the example of multiplication, which is a rather simple mathematical function. If you take as input the integer numbers from 0 to 9, you can see that the
output range goes from 0 to 81. The output range is wider than the input range. I dont see any reason it should be any different with the building of a waveform from elementary audio frequencies.
We must be aware that what makes audio frequences that are present in the waveform audible is the decoding work done by the inner ear. If the waveform is not reproduced with a very high level of accuracy, the decoding will be more difficult and as a result the audio message will sound less natural.
 
Last edited:
Aug 16, 2021 at 3:39 PM Post #6,279 of 6,480

71 dB

Headphoneus Supremus
Joined
Sep 17, 2017
Posts
1,690
Likes
557
Location
Helsinki, Finland
I'm too tired of decoding audiokangaroo's posts and trying to make heads or tails out of the bizarre claims. It is like a perfect example of backward thinking from the conclusion: Analog is best, then comes hi-rez digital audio and then comes CD audio. How can this belief be justified? Well invent some crazy theories about wavefrom surfaces and how band-limited signals aren't actually band-limited. Its so insane I feel bad for audiokangaroo...
 
Aug 16, 2021 at 3:45 PM Post #6,280 of 6,480

Nickhasarrived

New Head-Fier
Joined
Feb 24, 2020
Posts
30
Likes
24
Location
Toledo, Ohio
What do you think I didn't understand about the way a mic works ?
I'm open to learn new things if you have something to explain. However, the waveform problem is exactly the same at the voltage level and at the air pressure level.
The waveform is the result of a process. I don't want to be overconfident, but this process seems to have a combinatory dimension and a random dimension.
I don't think that explaining the production of the waveform with a mathematical function is a wrong idea.
You have the right to compare my theory with science fiction, but it would be more useful if you could explain precisely what is wrong among the things I tried to explain.
In my opinion, considering that the waveform has the same frequency content as the audio band is simplistic.
I can give you again the example of multiplication, which is a rather simple mathematical function. If you take as input the integer numbers from 0 to 9, you can see that the
output range goes from 0 to 81. The output range is wider than the input range. I dont see any reason it should be any different with the building of a waveform from elementary audio frequencies.
We must be aware that what makes audio frequences that are present in the waveform audible is the decoding work done by the inner ear. If the waveform is not reproduced with a very high level of accuracy, the decoding will be more difficult and as a result the audio message will sound less natural.
I think what you are talking about matters more to the way a DAC decodes the information. The number of samples will matter less to the way a DAC actually works.
 
Aug 16, 2021 at 5:43 PM Post #6,281 of 6,480

audiokangaroo

Head-Fier
Joined
Jul 26, 2021
Posts
70
Likes
2
Location
France
I think what you are talking about matters more to the way a DAC decodes the information. The number of samples will matter less to the way a DAC actually works.
DAC technology is interesting, of course, but this is not what we are discussing here. The number of samples is directly related to the frequency band we want to capture from
the waveform.
Basically, we want to capture perfectly every audible frequency in order to be able to reproduce them. These frequencies are sinewaves and we have to make sure that they are reproduced as perfect sinewaves and not as distorted sinewaves, because the difference is audible. What is not easy to understand here is that the perfect reproduction of
these sinewaves whose frequency is limited to 22 KHz, depends on the capture of other sinewaves whose frequency is much higher. These ultrasonic frequencies are inaudible as such, but they are for mathematical reasons, necessary to achieve a perfect capture of the waveform. As the perfect capture of the waveform is necessary to the perfect capture and reproduction of the audible sinewaves, these very high frequencies have to be captured in order to achieve high fidelity. If they are not present in the sampled content, frequencies under 22 KHz will be distorted and the output of the DAC will not sound natural.
 
Last edited:
Aug 16, 2021 at 6:19 PM Post #6,282 of 6,480

71 dB

Headphoneus Supremus
Joined
Sep 17, 2017
Posts
1,690
Likes
557
Location
Helsinki, Finland
DAC technology is interesting, of course, but this is not what we are discussing here. The number of samples is directly related to the frequency band we want to capture from
the waveform.
Basically, we want to capture perfectly every audible frequency in order to be able to reproduce them. These frequencies are sinewaves and we have to make sure that they are reproduced as perfect sinewaves and not as distorted sinewaves, because the difference is audible. What is not easy to understand here is that the perfect reproduction of
these sinewaves whose frequency is limited to 22 KHz, depends on the capture of other sinewaves whose frequency is much higher. These ultrasonic frequencies are inaudible as such, but they are for mathematical reasons, necessary to achieve a perfect capture of the waveform. As the perfect capture of the waveform is necessary to the perfect capture and reproduction of the audible sinewaves, these very high frequencies have to be captured in order to achieve high fidelity. If they are not present in the sampled content, frequencies under 22 KHz will be distorted and the output of the DAC will not sound natural.
Are you by any chance talking about Gibbs phenomenon?
 
Aug 16, 2021 at 6:40 PM Post #6,283 of 6,480

audiokangaroo

Head-Fier
Joined
Jul 26, 2021
Posts
70
Likes
2
Location
France
Are you by any chance talking about Gibbs phenomenon?
Gibbs phenomenon is a mathematical phenomenon. I guess it is due to the fact that a sinewave is a sinewave, which sets a limit to the achievement of a perfect square wave.
This is not what I was talking about here, however. My point about sampling, which has to be confirmed, of course, is that the reproduction that the form of the signals that we can hear depends on the perfect capture and reproduction of the global waveform (what I called the surface). I think that this waveform is made of audible and inaudible frequencies and that each of those signals has to be captured. If we don't capture them all, the reproduced waveform will be different from the original, and the sinewaves produced by the decoding job in the inner ear will not be perfect, which is potentially audible.
As you can see, I don't pretend that we can hear frequencies much above 20 KHz, like bats. All we have to do is to make sure that audibles sinewaves are perfectly reproduced and decoded.
 
Last edited:
Aug 16, 2021 at 6:55 PM Post #6,284 of 6,480

71 dB

Headphoneus Supremus
Joined
Sep 17, 2017
Posts
1,690
Likes
557
Location
Helsinki, Finland
Gibbs phenomenon is a mathematical phenomenon. I guess it is due to the fact that a sinewave is a sinewave, which sets a limit to the achievement of a perfect square wave.
This is not what I was talking about here, however. My point about sampling, which has to be confirmed, of course, is that the reproduction that the form of the signals that we can hear depends on the perfect capture and reproduction of the global waveform (what I called the surface). I think that this waveform is made of audible and inaudible frequencies and that each of those signals has to be captured. If we don't capture them all, the reproduced waveform will be different from the original, and the sinewaves produced by the decoding job in the inner ear will not be perfect, which is potentially audible.
As you can see, I don't pretend that we can hear frequencies much above 20 KHz, like bats. All we have to do is to make sure that audibles sinewaves are perfectly reproduced and decoded.
Well, if you don't believe 44.1 kHz can reproduced audible sinewaves perfectly that is your problem. Sampling theorem proves it. My resampling examples show strong evidence for it too not to mention all the listening tests. You keep ignoring the evidence that debunks your claims. That is why people get frustrated with you.
 
Last edited:
Aug 16, 2021 at 7:16 PM Post #6,285 of 6,480

Dogmatrix

Headphoneus Supremus
Joined
Apr 1, 2011
Posts
2,076
Likes
495
@audiokangaroo I see the point you are trying to make and I also think there may be something there
Such exploration is true science whether it proves fruitless or genius is for the future but the endeavour is science
Unfortunately this thread in spite of its name has little to do with science it would be better named sound engineering
You will not find any help in your quest here , I see the personal attacks are already beginning
I wish you all the best in your search for the truth
 

Users who are viewing this thread

Top