Talk:Analog signal

These three very similar articles need a proper merge and copyedit, but I haven't got time at the moment.

---

Sources: Some of an earlier version of this article was originally taken from Federal Standard 1037C in support of MIL-STD-188.

Now refactored a bit to flow better. Still needs more copyediting. -- The Anome 18:36, 29 Aug 2004 (UTC)

Thoughts

I don't have citations for these ideas, and I don't quite know how to work them into the relevant articles, but I'd like to throw them out there:

1) "Analog" and "Digital" describe two different kinds of computing machinery. The word, "analog" did not come into common use until after digital computers were invented.

2) "Signal Processing" machinery is computing machinery. A physical signal is a measurable quantity that contains information about some other quantity of interest, and signal processing machinery is machinery that transforms a signal in some useful way that typically can be described by a mathematical law.

3) "Analog" originally was a noun. An analog used to be a thing that somehow represented or stood for some other thing. When the "information" contained in a signal was simply the value of some other measured quantity, then it was acceptable to say, "the signal is an analog of the original quantity."

4) The word "electronic" describes electric circuits that process signals.

5) A "logical signal" can be imposed on a "physical signal" by means of modulation. In fact, there can be multiple levels of modulation. (e.g., subcarriers on a FM broadcast radio station.)

6) All "physical signals" are analog signals. All "digital signals" are logical signals.

7) "Noise" is the difference between the signal that you receive, and the signal that you wished to receive.

8) One reason for modulation is to provide immunity from noise. Digital signalling is often used for this purpose, but all-analog solutions can also provide significant noise immunity (e.g., FM broadcasting vs. AM broadcasting.)

Drondent 01:01, 18 July 2007 (UTC)

Link suggestions

An automated Wikipedia link suggester has some possible wiki link suggestions for the Analog article, and they have been placed on this page for your convenience.
Tip: Some people find it helpful if these suggestions are shown on this talk page, rather than on another page. To do this, just add {{User:LinkBot/suggestions/Analog}} to this page. — LinkBot 00:52, 18 Dec 2004 (UTC)

Do we want articles that aspire to sounding erudite? Articles that are comprehensive to the point of being chopped hash? These are the properties found in standard encyclopedias. We can be different by having plain, clear explanations.

Page move: Analog signal

To conform with digital signal and discrete signal (not digital (signal) and discrete (signal)). The fundamental difference between "analog signal" and "analog (signal)" is that the latter is a disambiguation of "analog" but "analog signal" is a perfectly acceptable usage.

Support

1. Cburnett 16:29, Jun 12, 2005 (UTC)
2. James F. (talk) 20:26, 13 Jun 2005 (UTC)
3. —Michael Z. 2005-06-17 04:47 Z
4. taestell 01:32, Jun 19, 2005 (UTC)
5. I've been doing link disambiguation for Analog, and the two topics are virtually identical. Nihiltres 22:44, 28 October 2006 (UTC)

Discussion

Please put discussion here instead of in the voting section.

"Analog (signal)" swapped with "Analog signal" (preserve history of "Analog signal") after uncontested support of 4 to 0. Cburnett 05:02, Jun 19, 2005 (UTC)

False assertion?

Currently, the text claims that a digital may be distiguished from an analog signal by the meaningfulness of small fluctuations in the signal amplitude the latter and meaninglessness of them in the former. That doesn't sound right: I'm not aware of a natural law that says one can't make use of small differences in the signal amplitude in encoding and decoding digital data. Can anyone defend the current text's claim? Terry Oldberg 18:31, 15 December 2005 (UTC)

I think what's being referred to here is an effect called quantization, which limits the minimum possible capturable per cent difference between two volumes in a digital recording to the inverse of the number of bits in its sample resolution times 100. In a 16-bit audio recording, for instance, the minimum recordable difference in volume will be a 100/16, or 6.25, per cent difference in volume. The greater the sample resolution, the smaller the difference in volume can be recorded in a digital capture. This is an inherent physical limitation of digital recording, whereas in analog recording, which is produced by a continuous rather than a discrete process, this doesn't happen (although you get noise related to the medium — for example, line hum or tape hiss — creeping in).

A similar effect known as aliasing happens in the sampling rate: digital audio files cannot record a frequency higher than the Nyquist frequency, defined as half the sampling rate of the recording: e.g., a digital recording with a sampling rate of 44 KHz cannot record a frequency greater than 22 KHz.

Having said all this, the article's description is vague, almost to the point of inaccuracy. It should say that analog recording is a continuous wave function resulting from the impression of a fluctuation in the voltage of a current, caused by sound pressure, on a medium, whereas digital recording is a discrete (i.e. stepwise) series of snapshots of amplitude, with a maximum sampling frequency and sample resolution, and that these characteristics impose predictable limits on the maximum frequency and minimum difference in volume a digital recording can capture. HarmonicSphere 03:11, 2 February 2006 (UTC)

Correction

It was formerly asserted that a microphone's operating mechanism is a variance of current caused by sound pressure. This is technically incorrect (compare Microphone), since it is not current that varies with the sound pressure, but voltage. I've made this correction in the article. HarmonicSphere 03:03, 2 February 2006 (UTC)

...Unless you are talking about a carbon microphone—still found in older telephone sets—A carbon microphone effectively is a resistor whose value varies with sound pressure. When a carbon microphone is placed in a circuit that is supplied by a constant-voltage source (e.g., a battery), then the current in that circuit that will fluctuate in response to sound pressure fluctuations. Drondent 00:35, 18 July 2007 (UTC)

Added wikilinks

I made the anon changes, forgot to log in

Incorrect definition?

Although I haven't found any other (clear) definition of an analogue signal, I think the very first line of this article is not correct. Unless I'm very much mistaken, a signal is analogue as long as it is continuous in amplitude, and is not related to continuity in time. For instance, a Compact Cassette is considered an analogue medium. However, a compact cassette has a finite number of magnets, and is thus not continuous in time. Cassandra B 18:06, 22 April 2007 (UTC)

But we pretend that it is continuous. Any difference between the signal we expected to read off the tape and the signal that we actually read off of the tape is called "noise." If we want to analyze the noise, then it may be useful to know that part of it is due to the granularity of the medium. Drondent 00:30, 18 July 2007 (UTC)

The introductory definition was very poor. It completely missed what an analogue signal really is. It suggested that any time continuous signal of constant amplitude is an analogue signal. But it isn't. To be analogue it has to be an analogue of some other quantity. I still don't think the definition is perfect (in that it isn't quite vague enough). I B Wright 10:21, 8 August 2007 (UTC)

Unrefenced Material

Some kind Administrator has added an {{Unreferenced}} tag to the article. The subject matter of the article is such that any reasonably competant engineer could write such an article without reference to any other material. In practice such an article is likely to peer reviewed and edited by (hopefully) equally competant engineers, and as such, the article really doesn't require much in the way of references or citations. I B Wright 10:21, 8 August 2007 (UTC)

Infinite Resolution Myth

This article makes the false assertion that an analog signal has theoretically infinite resolution, then contrasts this with limitations of digital systems:

Since an analogue signal has a theoretically infinite resolution, it will always have a higher resolution than any digital system where the resolution is in discrete steps.

First, the wording is imprecise. I think what is meant is "[...] it will always have a higher resolution than can be represented by a digital system [...]"

More importantly, the statement itself is a persistent myth that is the opposite to truth. Any analog signal will theoretically always have noise, and any analog transducer will have a finite response to changes in the signal. This places limitations on the resolution and bandwidth of analog systems. This is a consequence of the laws of physics and not just the imperfection of real-world implementation. Since any analog system will have a finite resolution and bandwidth, a digital system with a finite bit width and sampling rate can be built with the same fidelity as any analog system. There is therefore no point in specifically pointing out the limitations of digital systems.

63.81.178.62 (talk) 22:28, 17 July 2008 (UTC)

Thanks! Good work. Binksternet (talk) 18:44, 18 July 2008 (UTC)

Face clocks are not digital

Just because a clock uses a digital signal at some point doesn't make it a "digital clock." All this does is confuse elementary school kids. What designates a clock as "digital" is changing digits, not a digital signal. See the English dictionary [1]. This is in contrast to the dictionary definition of analog clock[2]. Oicumayberight (talk) 06:03, 14 April 2009 (UTC)

From the opening paragraph of the article: To be an analog device, 3 criteria have to be met.

1. The time varying signal has to be continuous.

Correct. I B Wright (talk) 14:15, 14 April 2009 (UTC)

2. It has to represent some other time varying quantity.

Correct. I B Wright (talk) 14:15, 14 April 2009 (UTC)

3. That other time varying quantity also has to be continuous.

Not Correct. It can be, but it doesn't have to be. I B Wright (talk) 14:15, 14 April 2009 (UTC)

Most types of clock fail on the first criterion in that they do not move in a continuous movement but move in discrete steps. This makes the movement digital.

Technically correct. Though there are plenty of examples of true analog clocks. Examples: the water clock; the candle clock etc. However all clocks that depend on some type of escapement or electronic frequency source are not technically analog. I B Wright (talk) 14:15, 14 April 2009 (UTC)

A note was added that for almost all practical purposes such clock can be considered analogue even though they are not in reality. The term 'analog' is commonly used to describe the presentation of the time in that thge clock has hands, but the use is technically inorrect. 86.133.11.175 (talk) 13:46, 14 April 2009 (UTC)

It is probably fair to say that the term 'analog' in the clock context is used to describe the type of display rather than the modus operandi (which is resolutely digital) in order to distinguish it from the type of display that most people would recognise as being 'digital'. As has been noted: this is for all practical purposes, true and a dictionary of language would pick up on the common usage. But this article is a technical article not an article on language and word usage.

Personally, I don't believe the concept of the analog display is out of place, but as a technical encyclopeadic article, the technical innacuracy must be pointed out.

As a footnote: the term 'digital' does not necessarily mean that the output consists of displayed digits, though can include it. Many years ago a constructional project appeared in an electronic magazine that presented the opposite connundrum. It was a clock with a digital display but the modus operandi was in fact analog. I B Wright (talk) 14:15, 14 April 2009 (UTC)

This article is also an article about signals, not equipment nomenclature. It's inappropriate for this article to define the entire piece of equipment as "digital" just because it contains digital signals. Almost every piece of modern electronic equipment contains digital signals, but we don't call them digital, especially without sources. See WP:DICT Oicumayberight (talk) 16:18, 14 April 2009 (UTC)

I didn't introduce the subject into the article. Perhaps you should read WP:DICT. It make prescisely the point that this isn't a discussion on language useage. And, yes, a piece of equipment that solely uses digital signals is described as digital device regardless of any display presentation, just as a device that uses purely analog signals is described as an analog device. Any device using both is a hybrid device. Most mechanical clocks contains no analog signal whatsoever whether you like it or not. 86.133.11.175 (talk) 16:33, 14 April 2009 (UTC)

It's not about whether or not I like it. It's about whether or not it's sourced. Where's your sources for these claims? Oicumayberight (talk) 16:47, 14 April 2009 (UTC)

I can't believe you are replacing sourced material with unsourced material. Oicumayberight (talk) 15:45, 14 April 2009 (UTC)

Virtually any reference work on the subject of analog signals makes the point that they must be continuously varying. Why do you have so much dificulty grasping the fundamental principle that something that moves in discrete steps does not meet this criterion? Your references were not sources as they referenced the language usage of the term 'analog' in connection with the subject and did not address any technical issue whatsoever. 86.133.11.175 (talk) 16:25, 14 April 2009 (UTC)

At least I had references. Oicumayberight (talk) 16:28, 14 April 2009 (UTC)

Irrelevant references (for the issue in point). 86.133.11.175 (talk) 16:34, 14 April 2009 (UTC)

It's very relevant, especially when you are using a article to define equipment nomenclature. It's in violation of WP:DICT. Oicumayberight (talk) 16:37, 14 April 2009 (UTC)

Then I suggest you read what has been written in the light of the comments on both this and your own talk page. You just want to describe a handed clock face as analog (a language derived use of the term not a technical one - in violation of the very policy you cite to try and bolster your failed argument). As has been pointed out, this is a technical article not an article on language use and to describe a mechanical clock as analog without any qualification that it really isn't would be to provide a misleading article. The article describes the technical aspect, but notes the language usage as it would be incomplete without it, bearing in mind, that I don't believe that you are by any means alone in misunderstanding the usage. 86.133.11.175 (talk) 16:49, 14 April 2009 (UTC)