Discussion in 'Pro Audio Equipment' started by audiokid, Jan 12, 2011.
What does analog mean? In audio, how do you explain analog or analog?
Hmmm... 38 views and a day later and still no responses, so I'll give it a try, just for the sake of discussion.
I think the existing definition in recording.org's Audio Terms is already very well written, and reads:
An "analog" is a close but not perfectly accurate copy of (in this case) a signal or sound. For instance, a signal recorded on tape is an analog of the original sound, and although it may be a nearly perfect duplicate, it will not have exactly the same characteristics as the original.
Perhaps it might also help someone who is still wrestling with the term to know that it is derived from the word analogy, which was first used in 1536. Its roots are far from the world of audio. An understanding of the word analogy helps to understand the word analog. If we were to describe the human heart as a pump, that would be an analogy of thought or idea, but the pump itself -which is an object- would be the analog of that analogy.
It also might help to know that analogs can vary in quality, and that more than one analog of the same thing can exist. An Edison phonograph cylinder, vinyl record, 8 track cartridge, cassette tape and compact disc of Bach's Fugue in D minor are all analogs, and of varying quality.
Now I can almost hear the muttering of those who are saying surely a CD is not an analog, but in the strictest sense of the term, a CD is indeed an analog of the original live performance, in the same way a record or tape is.
Sometimes words have more than one meaning or application, as is the case with the word analog. In modern audio parlance, analog has come to mean any sort of audio equipment or process that is not digital. Analog equipment is used to manipulate analog signals, which are theoretically continuous in frequency and amplitude -just like any real audio performance- whereas any digital equipment must first convert analog signals into digital signals, which are instead comprised of discontinuous quantities of bits or steps that are usually fine and numerous enough to fool us into thinking we are listening to continuous analog signals.
I'm surprised one of the EE degrees hasn't helped you out here Mr. Admin.
I'll take a stab at it to be corrected or shot full of holes by our esteemed EE friends.
An analog audio device manipulates electrical voltage throughout the process of creating, transferring, amplifying, modifying, or recording an electrical waveform that is representative (analogous) of the changes in amplitude (volume) and frequency (pitch) of the source.
Well, I'm gonna go out on a limb and add to the fantastic posts above with another perspsective...
In audio it is the conversion of sound vibrations through mechanical energy and then through electrical energy and then back to mechanical and sound vibrations to represent the soul of sound in a wholistic, flowing, organic manner. It contains continuous energy without reducing sound vibrations to cut up, dualistic, on/off, yes/no, (right/wrong?) pieces of the whole, regardless of how many samples are taken in how short a time, which does have it's place as well.
So lemme have it, I'm here to grow, not be right
In mind my analog is defined simply as the lack of a binary.
Digital circuit are circuits meant to convey information using binary as the main form of communications. The binary alphabet has 2 letters; 1 or 0, on of off, true or false. Its amazing what you can do with just 2 letters.
In order to achieve binary communication you need to create those letters, you need analog circuitry to do this. All digital components, weather they are complex FPGA's or simple logic gates have analog circuits inside. These circuit are achieved with transistors typically (you could uses vacuum tubes )that are driven to a point of saturation. When saturated these transistors act as switches either off or on. Each transistor can be used to represent a single bit, using many transistors you can make binary words....
Some define this is digital circuitry due to saturation and switch like behavior of the circuitry. I disagree with that, fundamentally this type of circuity is governed by the same natural laws of physics that an analog circuit is. We are just using differently. So what makes something digital in my mind is use of the binary language to covey a message.
Analog would be defined as conveying information using voltage and/or current levels.
Here's my attempt: An analog audio signal is one whose output varies continuously as a function of time. Analog devices are those that modify or produce analog signals. Analog signals are distinguished from digital or discrete signals where both the input and the output of the function can take on only a finite collection of values.
Digital means ten (digits=fingers) and the key is not binary, but the fact that we use the natural or counting numbers or another discrete (discontinuous) quantity to represent the input and output. In practice, we use a finite collection of values. The base that we use to represent the values in irrelevant. The Russians experimented with a base three digital computer back in the day.
The key isn't what you are measuring. Digital signals represent voltages and currents too. The distinction is whether you represent the voltage as any real number (continuous) as a function of time (continuous) or as one of, say, 2^24 numbers (discrete) sampled 44.1K times per second (discrete).
I thought I'd pull this bit into a separate post because this is pretty esoteric and sort of off topic. I think the last sentence is dead on - digital vs. binary is a matter of the way humans interpret the circuits. But remember the question of whether the "real" circuit is analog (Maxwell's equations) or discrete (quantum mechanics) is a pretty deep philosophy of science question as well. My position is that we can't know, we won't know. (Apologies to Hilbert. I might have that carved on my tombstone.) This allows me to bring up a quote that I saw on Twin Peaks last night (got the boxed set for Xmas). "What we observe is not nature itself, but nature exposed to our method of questioning." - Heisenberg
It's funny, I didn't think the OP was remotely struggling with the term, but was merely interested in the intellectual excercise of a discourse on an alternative elaboration. :smile:
I think dvdhawk is assuming that there are standard technical definitions that the EE crowd will chime in with. It's actually quite hard to give a technically accurate definition of analog. Both of my definitions of analog and discrete use terms like continuous in an intuitive sense and I could have my mathematicians stripes ripped from my shirt for going so. Not that anyone cares, but from a mathematicians point of view a technically better definition of a digital signal would be: a function whose domain and codomain are both countable sets with discrete topologies.
Wow Bob thanks for the comments I really enjoyed that! I agree with you, and your words are much clearer than anything I could have rattled off.
I live down on the hardware level, once and while jumping up to the assembly or even C code, but mostly I play with physical components and try to get them to behave the way I want them to. When you step into the world of computing you are in different space. Your in a place we have created out of our math. Everything should work as we defined it there, unless we made an error in logic. However in the component level our math is a approximation at best of whats really going on.
Analog in my mind lives in the space that we understand only through approximation, whereas digital is more the world we created.
Sure Bob, but there's nothing remotely incorrect about your usage of the word continuous, and you are too strict in thinking that words like continuous cannot be used in an intuitive sense or non-mathematical vernacular without risk of appearing un-mathematician like. In fact, I would argue that without the complement of ordinary language, mathematics, as a disclipline, could not even exist.
Couldn't we surmise from this that there is a theoretical application of the word as well as a practical one? Or perhaps one is a function of the other in real time usage?
Anyway, I'm awake now..........thanks......
The only assumption I made (maybe incorrectly) was that he wanted a succinct elementary definition such as one would find in the RO glossary of terms to help those who aren't engineers have a better understanding.
My mild surprise was that one of those EE's who are exceedingly qualified to do so, hadn't regurgitated a simplified technical definition they had learned on day 1. Or until they had time to clearly answer the question, copy/pasted an appropriate wikipedia or google definition. We're very lucky to have some brilliant minds willing to share their extensive knowledge as it relates to these things. So in their absence, I took up the challenge of trying to define it (as I understand it) without any outside reference.
I wouldn't be surprised if the admin's working knowledge of analog equipment is as good as, or better than, mine.
But that may "depend on what your definition of 'is' is." - Slick Willie
LOL Actually digital is real time too.
It think the problem here is the question is essentially philosophical. That is more subjective in the end, and hard to nail down definitively.
What I consider incorrect in my usage is that I really would rather not rule out things like (perfect) square waves and sawtooth waves in my definition of an analog signal. But of course, they are not continuous in either a precise or intuitive sense. When a couple of the most common examples of audio signals don't fit your definition of audio signal...well I'd like to think that was un-mathematician like...but I know some engineers who would disagree. I agree with your point about ordinary language, which is why I did not define an analog audio signal to be a function from a real time interval to a real or complex Banach space.
I'm just an idiot technician not an EE but I've always realized that our analog audio travels at the speed of light. While our high quality 24-bit, 192 kHz sample-based recording is really slow in comparison. Probably why everybody in digital land keeps trying to emulate analog land. So I really want to recorder that's even better than DSD which is sampled at nearly 6,000,000 times per second. That's too slow. I want DSD sampled at 669,600,000 miles per hour, 186,000 mi./s, what is that in "hurts"? That might make digital a good analog to analog?
I don't know any analog of myself
Mx. Remy Ann David
Hey DVDHawk, I actually misread your post that I quoted the first time around, as I thought you meant that Admin possesses more than one Electrical Engineer’s degree, and that you were being playfully sarcastic by saying that at least one of them should have been sufficient for him to furnish a definition. In this context, perhaps my tongue in cheek reply is more understandable. :tongue:
Make that out of cheek.
Analog in audio makes reference to the signal, weather it is from an electric waveform(analog) or a digital wave.
Digital Audio at some point may become analog, which subjects the wave to conversion into the electric realm and must be carried by a pair (or more) of wires.
In the case where Audio is traveling through a wire, it must see ground. Ground is a term for a point at which all electricity is disrupted to a zero reading in current, meaning no signal is present. Ground is also a term for a copper rod that is pounded in the earth next to your electric service in your building.
The Analog Audio on a wire is referred here as being unbalanced and is subject to interference. Interference occurs when other magnetic fields or frequencies interrupt the transfer from the source and introduce noise.
Analog Audio that is then transferred by cable or wire and can also be balanced and therefore uses a transformer or an electronic circuit to null any interference over the course of a cable run. This is usually done by shielding, which is braided wire which overlaps two or more conductors that carry the signal.
Analog Audio excels at complex details and seems to be the gold standard by which most professionals still go by in regards to the recording and mixing processes.
Circuits that carry an analog signal are more subjected to noise and artifacts thereby making them easy targets for criticism in the DAW world.
Analog Audio has a long history, there has been many advances in technology over the years, but the biggest drawback in with this can be the cost.
Many recording artists, including myself, have significant investments in Analog Audio.
Within a world of Digital upgrades and outdated technology, the Analog Audio components are still the most critical pieces of gear in the chain.
"Ground is also a term for a copper rod that is pounded in the earth next to your electric service in your building."
Ummmm.....Not quite. The rod is the Grounding Electrode. The use of it is called Grounding. A Ground Rod is considered a Supplemental Method of Grounding. The NEC defines 'Ground" as: A conducting connection, whether intentional or accidental, between an electrical circuit or equipment and the earth, or to some conducting body that serves in place of the earth.
When the rod, plate, ring, water pipe, or other approved means of grounding is connected to a circuit, it becomes Grounded.
In practical use the Grounded Conductor is the Neutral whereas the Grounding Conductor is the earth or bond.
Never get the two of them mixed up or confused as a Grounded Conductor carries the load through a halfcycle in every AC circuit. Getting between it and the load is a shocking experience.....
We now return you to your regularly scheduled scientific analogies. This Electrical Instructional Video Was brought to you by BigDog Electric.
Separate names with a comma.