It has been said that the more complex the envelope, the better the creative possibilities. The history of 'the envelope' is one of the continuous evolution. The beginning lies with organ technology, where RC networks were used to try and damp out the clicks caused by keying sine waves on and off, and then the clicks ended up being generated deliberately so that they could be added back in as 'key click'.
Trapezoidal waveform generators followed, which provided control over the start and finish of the envelope. ADSR-type envelopes, and their many variants, were used for the majority of the analogue synthesizers of the 1970s and 1980s. The advent of digital synthesizers with complex multi-segment EGs has made the ADSR appear unsophisticated, and analogue synthesizers designed in the 1990s have tended to emulate the multi-segment envelopes by adding additional break-points to ADSR envelopes.
The suitability of an envelope has very little to do with the number of segments, rates, times or levels. Instead, it is connected with the way that things happen in the real world. There are two things to consider:
- Many instruments have envelopes with exponential attacks rather than the much easier to produce linear slopes which many analogue synthesizers use. One solution to this is to add in two or more attack segments and so produce a rough approximation of an exponential envelope. This is much easier to achieve in a digital instrument than in analogue circuitry.
- Envelopes often change their shape and their timing in ways that are related to the note's pitch and the velocity with which it was played. Most modular and monophonic analogue synthesizers are not velocity sensitive, and so instruments that depend on this sort of performance technique tend to suffer (e.g. pianos). Changing the attack times with pitch can be quite complex in an analogue synthesizer – you need an EG with voltage-controlled time parameters, and this can require a large number of additional patch-cords and control knobs (Figure 3.7.5).
FIGURE 3.7.5 Envelope scaling using voltage control. (i) An ADSR envelope. (ii) The same envelope with the attack, decay and release times reduced proportionally. (iii) The same envelope with just the attack time reduced. (iv) The same envelope with just the decay time reduced. In order to produce each of these envelopes, a voltage-controlled EG would need both ganged (all time altered equally) and individual controls.
Sophisticated multi-segment envelopes suffer from being harder for the user to visualize the shape of the envelope being produced. Probably the best compromise is an ADSR with a couple of attack, decay and release segments, and control over the slopes: 'function' generators meeting this sort of design criteria are beginning to appear. EG design research is still ongoing.
3.7.5 Discrete versus integration
Early analogue synthesizers used individual transistors to build up their circuits. This 'discrete' method of construction was gradually replaced by ICs, usually op-amps for the majority of the analogue processing. Custom chips began to integrate large blocks of circuitry into single chips: a VCO or VCF, for example. Finally, by the mid-1980s, complete VCO, VCF, VCA, LFO and EG circuits could be placed on a single 'voice' chip intended for use in polyphonic analogue synthesizers.
In the 1990s, the VCO would probably be replaced by digital generation techniques, with analogue filtering and enveloping from VCF and VCA chips. The specialist chips that are used can become collectors' items, particularly some of the older and rarer designs.