User manual

The AMS Pitch Changer employs a very fast microprocessor which examines the digitised
musical signal within the memory and performs a number of tests to isolate two points in
time which have the highest correlation with each other in terms of waveshape. The pitch
shift algorithm contained in the program memory uses this information in conjunction
with the required pitch ratio to minimise the ‘glitch’ problem normally associated with
real time pitch changing. With suitable programmme material almost perfect pitch
shifting can be obtained without the characteristic ‘fluttering’ of other units on the
market. For very different programme material and very wide pitch ratio settings, the
performance can be further enhanced by fitting the de-glitch option. The wide
bandwidth and excellent noise performance of the basic delay function is also retained
when changing pitch, as is the vco facility. Combining all these features, and the ability
to key in a pitch ratio of crystal-locked stability means the engineer may create automatic
double tracking of hitherto unobtainable quality and effectiveness. In certain cases,
feedback enhances the ‘fullness’ of the sound.
Feedback, used with care, can be used to great effect; for instance, at a pitch ratio of 2.0
feed-back generates harmonics of the input, and single notes on a bass guitar for
instance can be made to sound as if they were played on an organ. Chords, or beat
frequencies between two notes on the bass guitar will make it impossible for the
microprocessor to identify the frequency of the incoming signal, and will cause the effect
to be marred. Nevertheless, the lack of the usual ‘wobble’ on the output means that it is
now possible to use a pitch changer musically at pitch ratios other than 1.01.
1. Interesting arpeggio effects can be obtained by programming delays longer than
200-300mS and applying feedback to just short of instability. This effect is especially
useful if musical intervals are programmed as the pitch shift, such as 1.26. When
applying feedback whilst pitch changing, care should be taken when shifting the
frequency down by significant amounts, since if instability is reached successive cycles of
feedback will cause any input frequency to tend towards zero. On the way, the output will
pass the resonant frequency of most things in the control room, including the engineers
and the monitors - so BEWARE!
2. The unit performs at its best on programed material with recognisable pitch and
amplitude; material with high elements of attack does not allow perfect splicing, since
two compatible points in the memory do not exist. However, as mentioned above, the
de-glitch option will significantly improve the performance under these circumstances.
3. In an attempt to ensure that all S-DMX users are familiar with ‘pitch ratio’, not only as
an effect but also as a musical tool, we have included the following data. We have done
this knowing full well that most of the people who are reading this manual will already be
familiar with the theories expounded below. However for those who are uninitiated the
following may help:
Each note in the chromatic scale is separated from the next by a semitone increase in
pitch; the lowest common musical interval. This semitone increase is not a fixed
frequency. If for example we move from A’ to A#’ we will need to increase the frequency
by 26Hz to effect a semitone increase in pitch. If we now add 26Hz to G#’ we will not
arrive at A”. To move from G#’ to A” we will need to add 49Hz to effect the same
semitone increase. This is because the semitone is a frequency ratio or pitch ratio and is
therefore dependent on the frequency of the note being played.