Is a purely-analog chip possible without sampling?

or

--
========================================================================
          Michael Kesti            |  "And like, one and one don\'t make
                                   |   two, one and one make one."
    mrkesti at hotmail dot com     |          - The Who, Bargain

Oh, don't get technical with me. I hate it when people get technical.

(But, technically, one could easily quantize an audio waveform to less than e-)

John

Maybe if you were more familiar with 'this person's posting history, you'd know why.

That history would show you that Mr. Radium is either ill or a troll. Occasionally interesting discussion arises during the course of correcting his silly preconceptions, but that's not enough reason to encourage him.

___

-S maybe they wanna rock. maybe they need to rock. Maybe it's for the money? But That's none of our business..our business as fans is to rock with them.

She could have been using her dad's car, which had that sticker on it.

Cigarette in one hand, cell phone in the other? How did she manage to steer?

Michael

Oh, please. I've participated in Usenet for almost 25 years and I assure you that I recognize what is going on here.

If he is ill then he deserves better treatment. If he is a troll then he deserves to be ignored. The fact is that the likes of you and Mr. T enjoy feeding trolls.

He is encouraged either way.

--
========================================================================
          Michael Kesti            |  "And like, one and one don\'t make
                                   |   two, one and one make one."
    mrkesti at hotmail dot com     |          - The Who, Bargain

At the basic level, an audio cassette recorder is sampling. There are grains of magnetic material that are magnetized in the tape. The levels are remembered by the strength of the field, I believe.

So, this is similar to the scheme used by those intersil voice recording chips, in which analog levels are stored in tiny capacitors. It is still 'digital', in the sense that discrete samples are obtained, but those samples can take on an 'analog' value.

Regards, Bob Monsen

Really? Analog cassette uses sampling? Then, is there any electronic audio storage device [analog or digital] that does *not* use sampling? Or is sampling an inescapable monster of everything?

Basically, yes. See the page at

for a description of how ferromagnetic materials (such as are used in cassette tape) work.

Quantization and uncertainty/noise are pretty much inescapable in this universe.

Even your beloved variable-density optical sound recording technology is sampled/quantized, in a way somewhat analogous to the way in which cassette recordings are.

The film consists of a clear carrier layer, with an impregnated data layer consisting of discrete particles of metallic silver. The original film stock was made using particles of silver iodide, which is photosensitive. When the film is exposed to light and then developed, those particles of silver iodide which were exposed to light turn into metallic silver; those which were not exposed, wash away.

The darkness or optical density of the film strip depends on the number and size of the silver particles. The apparently-continuous voltage that creates the sound you hear, is actually related to the total number of silver particles blocking the light flowing through the readout mechanism at any given moment of time.

--
Dave Platt                                    AE6EO
Friends of Jade Warrior home page:  http://www.radagast.org/jade-warrior
  I do _not_ wish to receive unsolicited commercial email, and I will
     boycott any company which has the gall to send me such ads!

You *could* sample sufficient points along a analog tape recording to reconstruct the audio digitally, but that is *not* how audio tape recorders work.

The recorder creates a varying magnetic field on the moving tape, that is passing over the recording head. Considering the pre-equalization of the audio signal, the field is analogous to that audio signal at any moment.

In playback, the magnetic pickup responds to the varying field on the moving tape by outputting a voltage that is analogous to the changing field. It is then equalized by a circuit complementing the pre-equalizer.

No "sampling" takes place and nothing digital occurred.

Well, it's "sampled" to the extent that the varying magnetic field, sensed by the pickup head, consists of the sum of the individual magnetic fields (vector and magnitude) of the large number of individual magnetic domains within the ferromagnetic recording layer.

It's not digital, and it's not sampled at regular intervals the way that a digital-audio or bucket-brigade-chip signal is.

--
Dave Platt                                    AE6EO
Friends of Jade Warrior home page:  http://www.radagast.org/jade-warrior
  I do _not_ wish to receive unsolicited commercial email, and I will
     boycott any company which has the gall to send me such ads!

Using the same logic, you would argue that a microphone samples the sound that reaches it. That's a stretch beyond common usage.

If, by "sampling" you mean "quantized, then, yes indeed, that's precisely what happens in a microphone, and yes, it is a stretch beyond common usage because it's common usage that's wrong.

The total force on any surface, inluding the diaphragm of a microphone and your ear drum is the net result of individual discrete collisions of air molucules with that surface. Each collision is most assuredly discrete. The net force looks continuous only because we whoose to integrate it over sufficiently long averaging time, but it is discrete whether your common usage embraces it or not.

And even disregarding that level of quantization, the fact that one might make the statement that a medium is "continuous," it does not then follow that the medium has the properties of infinite resolution as if there were no quantization going on.

No, I did not mean quantize, and that is not what happens in a microphone.

A microphone "converts" sound pressure to a voltage, or to a change in capacitance, or into a change in inductance, etc.

The pickup head in an audio tape recorder converts a moving magnetic field to a voltage.

The microphone responds to the combined effects of all the input forces.

I suspect your set of lexicons includes a belief there is no such thing as a Direct Current Voltage.

My ear may quantize the result, but the microphone doesn't.

Well, my point was that you can think of everything as digital or analog. Certainly digitized sound seems analog when played. One doesn't hear the individual samples.

However, assume that you could record some data in a truly analog way. That would mean that at every point, your recording medium contains exactly the same information as was present at every instant. Now, consider the following question. How much information is contained in a particular region of the media? Suppose you could somehow speed up somebody reading the library of congress out loud to an incredibly fast rate, so that the time required to play it would fit within your tape. Could you record that information, and reliably get it back from that interval of media?

Of course, the answer is no. It is not possible to do that, because the speeding up would require very high frequencies. In terms of information theory, those frequencies would require a sampling rate that is twice the rate of the highest frequency, so the information density would be very great. For the theoretically perfect analog recording, that information density is infinite.

What this all implies is that any piece of physical media is going to have a limited information density, and as a consequence, will not be able to store ANY analog interval precisely. You always lose something. You can represent this limit as a sampling rate, or as the rate of iron oxide dots flowing past on a tape, or as the maximum number of flip-flops you can squeeze onto a bit of silicon, or as the maximum number of capacitors you can put on an intersil chip times the ability of the output device to discriminate different levels of output...

So, sadly, you can't get away from the monster of information theory.

Regards, Bob Monsen

Platt"

sound

Of course it does.

Of course, but do you believe electricity is linear below the single electron level, assuming we could even record or measure to that level? Do you believe thermal and other noise does not exist in a microphone signal? Do you not believe there are finite limits to the linearity of any electrical device?

a

Your right, only a voltage at any given instant, which may be fairly constant for a given period of time.

It's obvious you have no idea what "quantise" means in a technical sense. In this universe everything has finite limits. Analog recordings are no different, nor the signal from a microphone, or even the variations in air pressure we call sound.

MrT.

It's obvious you are an argumentative idiot who would like to count angels on the head of a pin.

Regardless of the topic of discussion or the correctness of your position, Don, you just lost the debate.

-- ======================================================================== Michael Kesti | "And like, one and one don't make | two, one and one make one." mrkesti at hotmail dot com | - The Who, Bargain

There isn't -- unless you measure it for an infinite time.

And a related question: What is the low-frequency response limit of a DC amplifier?

Isaac

Yes! Isn't it neat?