Operator`s manual

Version 1.20 271
/ APPENDIX G
WHAT IS SAMPLING?
Sampling is a process where we record sound digitally. All natural sound comes in the form of
variations in sound pressure. Using a microphone, we can convert those changes in air pressure
into rising and falling voltages. Once they exist in that format, we can process them through
ANALOGUE TO DIGITAL CONVERTERS (ADCs) to turn those voltages into streams of digital
data. Once they exist as digital data, we can edit them with alarming precision.
SOUND MIC
VOLTAGE
WAVEFORM
ADC
DIGITISED
WAVEFORM
In the ADC, the sound is ‘sampled’ at a rate of 44,100 times per second. You can liken this to film.
If we take a lot of photographs in very quick succession, when we play them back (also in quick
succession), we have the illusion of movement. The same is true of sampling. If we take enough
samples, we get an accurate reproduction of the sound. To carry this analogy even further, if you
think of the very early days of film where they didn’t take so many frames in a second, the results
were jerky and distorted. The same could be said about old samplers - because they sampled less
(that is, the sampling rate was lower), the sound quality was not so good. In order to reproduce
sound accurately, you need to sample at a frequency that is at least twice the upper reaches of the
sound’s frequency range. In other words, if a sound contains frequencies that extend to, say,
15kHz, you need to sample at 30kHz at least. Instruments such as cymbals which are very bright
and contain many overtones need to be sampled at 40kHz. A bass drum, however, which has very
few upper harmonics, could feasibly be sampled at 20kHz.
The S6000 samples at 44.1kHz (the same as compact disc) and at 48kHz (the same as DAT) so
you can be sure you are getting the highest quality sound from your sampler.
The digitised waveform is held in RANDOM ACCESS MEMORY (RAM) where it exists as numbers.
As you know, computers and microprocessors are very good at dealing with numbers and so we
are able to rearrange those numbers and so alter the sound.
At the end of the process, we need to be able to convert those numbers back into an electrical
analogue waveform in order to hear them and so the numbers are reconstituted into an analogue
signal via DIGITAL TO ANALOGUE CONVERTERS (DACs) and output to your mixer or amplifier.
However, one of the inherent problems with sampling is the RAM and because of cost, it is not
possible to have an endless supply of it installed in the sampler (although the S6000 does allow up
to 256Mbytes of RAM to be installed offering nearly fifty minutes of mono samples to be held in
memory!!). As a result, recordings (or samples as they are more commonly known) are often fairly
short.