79 lines
3.2 KiB
Markdown
79 lines
3.2 KiB
Markdown
---
|
|
tags: [analogue]
|
|
---
|
|
|
|
# Analogue and digital
|
|
|
|
Analogue and digital are paradigms for recording information, specifically
|
|
information about an object or state that obtains in the world.
|
|
|
|
|
|
|
|
## Analogue
|
|
|
|
Here are two examples of analogue measurement/ capture:
|
|
|
|
- Measuring the temperature of a room with a mercury barometer
|
|
- The mercury in the barometer expands or contracts in response to the heat of
|
|
the ambient environment.
|
|
- We say that the position of the mercury relative to the container represents
|
|
the current temperature
|
|
- Measuring the weight of an object by placing it on a weighing scale
|
|
- We say the position of the needle on the scale represents the weight of the
|
|
object
|
|
|
|
> In each case, the position of the needle or the mercury isn't actually the
|
|
> temperature or the weight, it is a representation of this property. Thus in
|
|
> both cases, the analogue device works by analogy hence why we call it
|
|
> analogue.
|
|
|
|
This isn't the only sense in which "analogue" is used however. It doesn't always
|
|
mean an analagous representation of a physical property. We talk about
|
|
**analogue signals** as signals that vary continuously and which do not align to
|
|
digital values.
|
|
|
|
## Digital
|
|
|
|
### Limitations of analogue for computation
|
|
|
|
From a computing point of view, the data that can be captured and recorded by
|
|
analogue devices is simply too diverse and non-uniform. Computers are universal
|
|
machines whereas analogue devices typically only measure one property well.
|
|
Whereas a computer can record and playback music and take the temperature, to do
|
|
this in an analogue way would require a complex machine that can represent two
|
|
unrelated data sources. Imagine if a computer was able to achieve universality
|
|
through analogue means, it would most likely be very large physically and very
|
|
complex computationally since the computer would need to be able process data
|
|
from, e.g. for grooves on vinyl, needles on weighing sensors and mercury in
|
|
vacuum tubes.
|
|
|
|
> The types of analogue systems used are so different and variable that creating
|
|
> a common computing device that can understand all of them is nearly
|
|
> impossible.
|
|
|
|
Furthermore:
|
|
|
|
> Computers require highly reliable and accurate representations of certain
|
|
> types of data. Analogue representations of data can be difficult to measure
|
|
> precisely (and are potentially infinite), tend to decay over time, and lose
|
|
> fidelity when copied.
|
|
|
|
**Computers need a way to represent all types of data in a format that can be
|
|
accurately processed, stored and copied.**
|
|
|
|
### Why digital is better suited to computational requirements
|
|
|
|
A digital system represents data as a sequence of symbols where each symbol is
|
|
one of a limited set of values.
|
|
|
|
Computers only use two symbols for each value: 0 and 1.
|
|
|
|
Although a digital system could use more than two symbols, adding more would
|
|
[increase the complexity](Why_computers_use_binary.md#from-circuits-to-programs)
|
|
and cost of the system. A set of only two symbols allows for simplified hardware
|
|
and improved reliability.
|
|
|
|
All data within a computer from images to sounds are sequences of 0s and 1s. In
|
|
practice, the sequences are stored in various ways. On a CD, the binary digits
|
|
are stored as bumps (0) or flat spaces (1). On a flash drive, the 1s and 0s are
|
|
electrical charges etc.
|