Back to basics. Definition: Analogue audio

[MCH]

Are we talking audio signals only? If not, there are plenty of natual or man made signals that are not continuous by definition and are measured as such, and i would not call them digital either. Reasons being, for instance but not only, that not all values are allowed. For instance in mass spectrometry (hint: spectrometry not spectroscopy precisely as a consequence of being a result of counts, not a continuous wave). Another example everyone knows is the ionizing radiation, commonly measured in counts per unit of time.

I have no clue of electronics, but i have the feeling that digital as used in audio might imply some sort of coding? (It is a question)

In any case, i think that the definition as chosen by Amir in the video is spot on for the purpose of explaining that the signal can get noise or whatever, all this discussion is just.... a forum discussion

 

[Ken Tajalli]

Are we talking audio signals only?

Good point, lets say:
Limited to electrical signals in general, but ultimately Audio.

 

[DonH56]

So would you call DC an analogue signal, it is a continuous voltage and in time does not change.I have to ask to get a clearer answer.
After all, I am after bullet-proof Def. Think of it as not coming up with a good description, but something like a legal contract! That leaves no room for misinterpretations or abuse!

So Analogue is not restricted in time or amplitude, unless one sets a range.? do you agree

I intend to cover Analogue first - though it is proving difficult!
It is like saying, what is white, and the answer comes coloured is this and that, so what is not coloured is white!

Thank you, I am aware that there are set-in-stone definitions at IEE or IEEE, but there are some members here who are not engineers and they have loosely collected understanding on analogue signal.

DC is just another signal that happens to be constant (ideally). You can continuously vary an analog supply's output voltage, while a digital power supply quantizes the voltage level you can set. After start-up, time is irrelevant, so you could consider it a sample of one or whatever. I see no particular reason to treat DC differently than any other signal in this regard.

If you clip the output of an amplifier, is that still analog? You can spend the rest of your life defining boundary conditions. At some point I say "good enough" and move on.

I am not really interesting in getting into some big debate. The IEEE Standards are good enough for me, lowly engineer I may be. I am an analog engineer so used to dealing in shades of grey and nothing is ever completely absolute. To misquote (whomever said it), expecting black and white answers to everything is the path to madness.

In the audio world, AFAIK if there is a data converter (ADC or DAC) in the signal chain, then it is digital from the output of the ADC until the output of the DAC. I think that is all most audiophiles care about. They will consider any chain having a digital (conversion) step anywhere in it to be "digital".

Too pedantic for me - Don

 

[Ken Tajalli]

I am not really interesting in getting into some big debate. The IEEE Standards are good enough for me, lowly engineer I may be. I am an analog engineer so used to dealing in shades of grey and nothing is ever completely absolute. To misquote (whomever said it), expecting black and white answers to everything is the path to madness.

I like your posts! There are some gems in them.
So, Analogue is dealing in shades of gray, & that there is nothing completely absolute.
Black & white answers only relate to logic circuits, where shades of gray has no place.
Are we in agreement?
Are we getting closer to the answer?

 

[wwasilev]

Ok, I didn't read all the responses, so I could be repeating other posts. But, in electrical engineering we were taught that linear systems must obey two rules;

1) Superposition: f(x)+f(y)=f(x+y)

2) Homogeneity: for f(x)=y then f(Cx)=Cy where C is a constant

Digital systems don't obey those rules.

Please correct me if I'm wrong.

 

[MRC01]

I'll take a stab at it. I will not try to pedantic but this may be more complex (pun intended) than one would think.
...
In an analogue signal, a and b can take any value in the real number space ... even if you limit the values of a and b to be between -2 and +2 ... i-e in maths terms -2=<a<=2 and ... "a" and/or " b can take ANY, value in between... 1.094765974 or 1.985546548468484884 or -1.93584848484894848452487... etc...

In a digital signal. You limit the value that can be taken by "a" and "b". They only take what is called "discrete" steps. usually 2 to the power of something... that is 2 x 2 x 2 ..etc thus for 4 bits, "a" or "b" can only take strict values in 16 steps..., equally spaced steps.... of 0.25... thus you would have: The only value that can be taken by "a" and "b" would be:

-2, -1.75, -1.50, -1.25, -1.00, -.75, -.50, -.25, 0, 0.25, 0.50, 0.75, 1.00, 1.25, 1.50, 1.75, 2... "a" and "b" shall only take one of those values.. Nothing in between ...
Digital signal aka discrete signal....

Peace.

Exactly. A better term for what we call "digital audio" would be "discrete audio".

 

[Blumlein 88]

Where does FM radio transmission of audio fit in? I'd assume analog.

 

[MRC01]

Where does FM radio transmission of audio fit in? I'd assume analog.

I'd call conventional FM radio analog, but they also transmit HD Radio over FM, which I would call "discrete" or "digital".

 

[DonH56]

Where does FM radio transmission of audio fit in? I'd assume analog.

AM, FM, PM are all "analog" modulation schemes in my book.

 

[antcollinet]

Recent replies on "Is Digital Audio Transmission Analog?" has shone a light on the (mis)definition of Analog audio.
Many have sort of asserted that since there are conductors and an electrical current flowing through it, then it is analog!
See these examples:
- “Digital” is a mathematical representation. Signal transfer is through voltage and voltage is an analog quantity. Digital signal representation is transferred by means of analog quantity - voltage.
**********************
- Digital is nothing more than a useful technique, or like *** said, a mathematical representation. Even the data storage on our computers is just different levels of stored electrons on memory cells. Theoretically everything in life is analog. Isn't it beautiful?
*********************
- No. The first thing you learn in Digital Design 101 (actually 2-something) is that digital is just a special case of analog.
*********************
Without a solid, unified definition, we can not go forward in a scientific discussion.
I think it is time to have a bullet-proof definition for Analogue signal in general and Analogue audio in particular.
Ofcourse we could do a little Google and come up with some, but what do you think?
Afterwards, we could have a bullet-proof definition of Digital signal and digital audio in particular.
But for now, let us talk about Analogue Audio only.
It may sound mundane, but it seems we need to define certain phrases before we use them.

Any takers?

Really not sure what you are trying to achieve here. Based on your statements in the the thread about Amir's "Is digtal audio transmission analogue" info video,you seem to want to prove Amir wrong in his statement that digital signals are transmitted in the analogue domain.

You are on a road to nowhere with that.

 

[Ken Tajalli]

Really not sure what you are trying to achieve here. Based on your statements in the the thread about Amir's "Is digtal audio transmission analogue" info video,you seem to want to prove Amir wrong in his statement that digital signals are transmitted in the analogue domain.
You are on a road to nowhere with that.

Yes, that is my understanding. But I am not cocky nor self-assured enough to argue my own to death.
Instead, I am open to discussion.
Or should I do as you do, and come up with comments such as "you are on a road to nowhere!".
Is it sacrilege to question Amir?

 

[MRC01]

Claude Shannon and others in the field of information theory explored how to define information in terms of math and its relation to physics (energy / entropy), and how to quantify with mathematical precision the amount of information in any signal or encoding scheme, whether analog or digital/discrete.
James Gleick wrote a history & summary of this called "The Information". It's a great read for those interested in the theoretical side of what this is all about.

 

[Thomas savage]

DC is just another signal that happens to be constant (ideally). You can continuously vary an analog supply's output voltage, while a digital power supply quantizes the voltage level you can set. After start-up, time is irrelevant, so you could consider it a sample of one or whatever. I see no particular reason to treat DC differently than any other signal in this regard.

If you clip the output of an amplifier, is that still analog? You can spend the rest of your life defining boundary conditions. At some point I say "good enough" and move on.

I am not really interesting in getting into some big debate. The IEEE Standards are good enough for me, lowly engineer I may be. I am an analog engineer so used to dealing in shades of grey and nothing is ever completely absolute. To misquote (whomever said it), expecting black and white answers to everything is the path to madness.

In the audio world, AFAIK if there is a data converter (ADC or DAC) in the signal chain, then it is digital from the output of the ADC until the output of the DAC. I think that is all most audiophiles care about. They will consider any chain having a digital (conversion) step anywhere in it to be "digital".

Too pedantic for me - Don

Yea Don , like 2 beers in and you've of just thought ' to hell with this nonsense im putting on some music and getting the Mrs to run to the store for another chilled 6 pack , I dont work 70 hours a week for this shit ' ... .. .

Doctor Savage prescribes more beer for you .

 

[LightninBoy]

Digital audio is transmitted via analogue. That is a fact and there is absolutely no way to transmit it otherwise. There is no need alter definitions to avoid this reality. Simply concede it and move on.

But people are conflating the audio encoding mechanism with the transmission mechanism. The digital encoding enables data to be transferred with *zero* data loss using analogue (ie anything physical) means. This includes wires spanning the oceans, satellite transmissions over radio waves, etc. This is the point that needs to be hammered home again and again.

Digital encoding enables data to be transmitted over physical (analogue) means with zero data loss.

Digital encoding enables data to be transmitted over physical (analogue) means with zero data loss.

One more time ...

Digital encoding enables data to be transmitted over physical (analogue) means with zero data loss.

 

[Ken Tajalli]

Interesting reading.

Analog transmission - Wikipedia

en.m.wikipedia.org

Also:

 

[FrantzM]

So would you call DC an analogue signal, it is a continuous voltage and in time does not change.I have to ask to get a clearer answer.
After all, I am after bullet-proof Def. Think of it as not coming up with a good description, but something like a legal contract! That leaves no room for misinterpretations or abuse!

So Analogue is not restricted in time or amplitude, unless one sets a range.? do you agree

I intend to cover Analogue first - though it is proving difficult!
It is like saying, what is white, and the answer comes coloured is this and that, so what is not coloured is white!

Thank you, I am aware that there are set-in-stone definitions at IEE or IEEE, but there are some members here who are not engineers and they have loosely collected understanding on analogue signal.

A signal that can take any value between the limits of a set of numbers is NOT digital. It is analogue, time invariant i-e not varying with time.

 

[MRC01]

A signal that can take any value between the limits of a set of numbers is NOT digital. It is analogue, time invariant i-e not varying with time.

To clarify, DC is time invariant (ideally), but many other analog signals are not - they do indeed vary with time. The time varying aspect of the signal is where all the fun and music is.

 

[Ken Tajalli]

A signal that can take any value between the limits of a set of numbers is NOT digital. It is analogue, time invariant i-e not varying with time.

Question: if it is not digital, it must be analogue?
Is there just two options?

 

[MRC01]

I think so, yes. I see Digital vs. Analog as mutually exclusive.
The reason is: the encoding of information is either in discrete states, or it is not.
If it's discrete, then we call it digital.
Otherwise, anything else, not being discrete implies it uses a continuously varying encoding or set of states, in which case we call it analog.
Perhaps the words "discrete" and "continuous" would be more concise than "digital" and "analog", because the latter words have other meanings & connotations.
---
Either way, digital or analog, we can encode any information we want. And we can compute information content/density equivalent conversions from one to the other (how much bandwidth you need, etc.). One advantage of digital, is that it facilitates perfect (error-free) storage and transmission and error identification and recovery methods.

 

[antcollinet]

Yes, that is my understanding. But I am not cocky nor self-assured enough to argue my own to death.
Instead, I am open to discussion.
Or should I do as you do, and come up with comments such as "you are on a road to nowhere!".
Is it sacrilege to question Amir?

Of course not.

It is just somewhat obvious that a digital signal on an interconnect (wire/cable/fibre/pcb trace etc) is transmitted with analogue signals (values of voltage/current). There is no such thing as a square wave, they have rise and fall times. The rise and fall are not straight lines, the "top" and "bottom" of the "square" wave are not flat. The transitions from rising/falling edge to top/bottom are not sharp corners.

In fact digital waveforms can (especially at higher speed) look surprisingly close to sine waves.

In other words, they are continuously varying voltages - different from one moment to the next. Engineers must pay attention to all the things that can impact analogue signal transmission - response (analogue) of the driving and receiving devices, complex cable impedance, capacitive and inductive coupling of noise, transmission line effects, reflections etc etc - in order to make sure they get from one end to the other without errors.

The only difference from an analogue encoded signal is how we interpret the signal once the receiving device has received it.