Making sounds with analogue electronics - Part 9: Sampling in an analogue environment

Loops of tape cannot be used because the start of the sound would not be synchronized with the pressing of the key; that is, by arranging for the tape to be pulled back into the bin each time the key is released, it automatically goes back to the start point of the sound, ready for the next press of the key. There have been several other variants on the same idea from other manufacturers, but the Mellotron is the best known (Figure 3.8.2).

Because the capstan is the same size for each key, the tape for each key needs to be recorded separately, with each tape producing just one note (although several tracks are available on each tape, with a different sound on each track). The tapes are thus multi-sampled at 1-note intervals. Recording user samples for such a machine requires time, patience and attention to detail: the levels of the sounds must be consistent across all the tapes, for example.

The 'frames' that contain complete key-sets of the tape bins can be changed, but this is not a quick operation. Because of the difficulty of recording your own sounds onto a tape, these tape samplers can almost be regarded as being sample-replay instruments rather than true samplers.

Tape loops
By looping a piece of tape around and joining the end to the beginning with splicing tape, it is possible to create a continuous loop of tape which will play the same piece or recorded material repeatedly. The only limitation on the size of the tape is physical: short loops may not fit around the tape recorder head and capstan, whilst long loops can be difficult to handle as they can easily become tangled.

The repetition of a sequence of sounds produces a characteristic rhythmic sound, which can be used as the basis of a composition. As with the Mellotron tape player, synchronizing the playback of the start of a loop is difficult, and synchronizing two loops requires them to be exactly the same length, or to have very accurate capstan motor speed control. Tape loops are thus usually used for asynchronous sound generation purposes.

Pitch changes
Analogue tape recorders have one fundamental 'built-in' method of modifying the sound: speed control. Changing the speed at which the tape passes through the machine alters the pitch of the sound when it is played back. This can be either during the record or the replay process.

For example, if a sound is recorded using 15 inches per second (ips), and replayed at 7.5 ips, then it will be played back at half the speed, and thus will be shifted down in pitch by one octave. Conversely, sounds that are recorded at 7.5 ips and replayed at 15 ips will be played at twice the normal speed and will thus be shifted up in pitch by one octave.

Note that the pitch and time are linked: as the pitch goes up, the time shortens, whilst lower pitch means longer time. The 'length' of a sound is exactly the length of the piece of tape on which it is recorded. If the tape is played back faster, then the tape passes over the replay head faster, and so the sound lasts for a shorter time. (The same is not necessarily true for digital samplers…)

This 'pitch halving and time doubling' was used to a great effect by guitarist Les Paul in the 1950s. Using the technique of recording low-pitched notes at a slow tape speed, and then replaying at a faster tape speed, he was able to achieve astonishingly fast and complex performances on guitar. The same technique is still a powerful way of changing the pitch of sounds, or for enabling virtuoso performances at slow tempos.