Is the player piano roll a digital or analogue medium?

Actually, they're analog storage, not digital.

the above is from that link. but what do i know?

but of course you mean digital to be 'expressed in numerical form.' But what are numbers? can ones and zeros be expressed by holes and absence-of-holes? That's the way it works for the aluminum in a CD, why not paper?

Were the cassette tapes used to load 80s computer games an analogue or digital medium?

this guy agrees with eyeball kicks, i think.

the above is what he says(jesus,how bored am i?)

apparently no piano-roll recording devices were capable of recording dynamics; someone else sat that and made notes about it and added that "control" information to the roll afterwards.

are there chopin rolls? amazing to think that there are "recordings" by him that we can hear

At least that's how I think a player piano works...

So a music box would also be digital, if a player piano is. (unless that's not how a player piano works.)

I think you're right - the abacus is digital.

Cripes!!!!!!!

And someone creating a MIDI file on a keyboard without quantising/looping the results is (before altering pitch/attack/decay/etc) creating something analogous to a player piano roll, yes?

The relationship between the pits and lands on the data surface of a CD (which can be various lengths within various tolerances) and the actual datastream is not all that simple; I tend think of CD as an emphatically digital medium because of its optical reading method.

I don't deny for a second that there are anomalies, tolerances, imperfections, etc., on the physical surface of a CD and in the physical process of reading it; however, the idealized representation of the data is still digital. This is unlike a piano roll.

And a related question: if I make an exact print of one piano roll from another piano roll, is the copy a digital one?

A third question: are the early computers which used punch hole technology digital or analogue?

Simply, an analog computer is a computing device that has two distinguishing characteristics:

1. Performs operations in a truly parallel manner. Meaning it can perform many calculations all at the same time.
2. And operates using continuous variables. Meaning it uses numbers that that change not in steps, but change in a smooth continuous manner.

By constrast, a digital computer can only perform sequential (one at a time) operations, and operates on discrete (noncontinuous) numbers.

no.

digital. (their holes were either on or off and also at a specific position -- the player piano roll could have holes at any interval at all)

i'll take a stab at this one: on a cd you have a certain series of pits and lands. they vary somewhat in position (though within certain tolerances, of course). what is important is that groups of them form binary numbers have values from 0 to 65535 (i think that's right, the values not really important though). the important part is that there are a SET NUMBER of these values FOR EACH SECOND of music. you can vary the location of some note position all you want, but in a 44 khz mp3 (for example) it's going to end up SOMEWHERE in one of 44,000 locations for that second of music. compare this to a player piano or a music box - you can literally make the mark anywhere on the roll, there are (in theory but not practice, of course) an infinite number of possible positions to put a hole in the area of paper that represents one second of music.

not according to doug coward's analog computer museum!

The analog-ness of a paper-roll's representation of time is not its resolution, it's that it isn't internally calibrated.

One simple distinction is that digital data can be read and written at different, and varying, speeds, whereas analog information can only be reproduced accurately if it is read at the same speed it was written. (This isn't completely general, but it's true for audio at least.)

If you want an early mechanical medium very similar to the piano roll that actually did store digital information, look up the Jacquard Loom. It too represents data with physical holes, but the key difference is that in weaving the unit of progress is the shuttle throw, not time, and the Loom encoded its data by throw.

------------------------------------

DIGITAL - DEFINITIONS

Webster's 1913 Dictionary

Definition: \Dig"i*tal\, a. [L. digitals.]
Of or performance to the fingers, or to digits; done with the
fingers; as, digital compression; digital examination.

Computing Dictionary

Definition: A description of data which is stored or transmitted as a sequence of discrete symbols from a finite set, most commonly this means binary data represented using electronic or electromagnetic signals.

ANALOGUE - DEFINITIONS

Webster's 1913 Dictionary

Definition: \An"a*logue\ (?; 115), n. [F. ?, fr. Gr. ?.]
1. That which is analogous to, or corresponds with, some
other thing.

The vexatious tyranny of the individual despot meets
its analogue in the insolent tyranny of the many.
--I. Taylor.

2. (Philol.) A word in one language corresponding with one in
another; an analogous term; as, the Latin ``pater'' is the
analogue of the English ``father.''

3. (Nat. Hist.)
(a) An organ which is equivalent in its functions to a
different organ in another species or group, or even
in the same group; as, the gill of a fish is the
analogue of a lung in a quadruped, although the two
are not of like structural relations.
(b) A species in one genus or group having its characters
parallel, one by one, with those of another group.
(c) A species or genus in one country closely related to a
species of the same genus, or a genus of the same
group, in another: such species are often called
representative species, and such genera,
representative genera. --Dana.

Computing Dictionary

Definition: (US: "analog") A description of a continuously variable signal or a circuit or device designed to handle such signals. The opposite is "discrete" or "digital".

Analogue circuits are much harder to design and analyse than digital ones because the designer must take into account effects such as the gain, linearity and power handling of components, the resistance, capacitance and inductance of PCB tracks, wires and connectors, interference between signals, power supply stability and more. A digital circuit design, especially for high switching speeds, must also take these factors into account if it is to work reliably, but they are usually less critical because most digital components will function correctly within a range of parameters whereas such variations will corrupt the outputs of an analogue circuit.

So a piano roll is digital!

This about sums it up I guess. (Due to part 2 of the above, the piano would then in itself be digital in pitch, at least given that the volume is digital (thus enabling us to ignore shifts in pitch due to hitting velocity), as opposed to e.g. a violin.)

Red herrings include:
i) "stored as number" – this can also be done in an analog device, the distinction is surely rather that of discrete vs continuous.
ii) "binary" – wouldn't be any less digital if computers worked in e.g. base three (with low-level voltage states positive-negative-zero).

I guess so, I was pondering that as well. This is perhaps why there can be a player piano but not for example a player violin. A player violin would have to either divide a single string into steps (discrete values) and only allow certain notes to be played or it would have to be able to continuously slide up and down a string which would make it analog.

OTM.

Some of the stuff upthread is so wrong and dumb.

xpost:

One could envision encoding methods that could do a good, but not perfect job of making a player violin by recording string position at a very fine time/space slicing interval.

But regardless of how fine the slicing intervals were, wouldn't the slicing of a length of string into discrete steps essentially make it digital? I guess I'm basically arguing that a fretted guitar neck is digital!

big xpost

There was another sentence after that. It seems like if you're arguing that piano roll technology was capable of recording musical pitches in an analog manner you would have to demonstrate how it would be possible to use a punched paper system to record a continuously variable tone from an instrument like a violin.

http://sonic-arts.org/darreg/dar12.htm

--------------------------------------------------------------------------------

Why? Electronic devices can be analog. It's not an electronic/mechanical distinction as already mentioned upthread.

Well yes, that is why it's interesting. It seems to teeter on the edge of saying that almost any process has a digital component. For example, a chemical reaction that depends on a threshold is digital. It may be that any effect that depends on a threshold being reached is digital in nature. This would include electrochemical reactions like synapse firing patterns.

Did the word 'digital' originally refer to fingers? Ie, it's digital if you can count it with your fingers, ie, discretely. So: a bunch of grapes can be said to be digital, insofar as it is a cluster of discrete events which may be counted.

No, if you counted the number of grapes you would have a digit but the grapes themselves are not digital. That's highlights the major problem with people in this thread trying to define things as "being digital" or analog when the "thing" in question is actually a complex system that may be a mixture of digital and analog processes.

For example "if a robot could play a violin, would it be analogue??" is meaningless on its own. Is the robot controlled through analog control voltages or a digital computer? If it's a digital robot playing the violin we wouldn't say that the whole thing is "digital." But if the control data (the piano roll) was a MIDI signal then we could say that portion of the violin playing robot was digital.

The same applies to the question "Were the cassette tapes used to load 80s computer games an analogue or digital medium?" The cassette tape is obviously an analog medium being used to store digital data that has been encoded into an analog signal. Likewise a player piano takes an instrument that uses a digital series of pitches and stores those pitches digitally while storing the player's timing in an analog manner. So it's a hybrid device and can't be accurately said to "be analog" or "be digital."

http://img.stern.de/_content/51/06/510692/robot8_311.284046692607.jpg

i saw a real pianola this weekend and got to touch it! it was broken, tho :/