Shopping For A Synth
Sometimes The Little Things Count
When shopping for a synthesizer, most people know to look for major features like the amount of patch storage, whether split and layered keyboard configurations are possible, and so on. But paying attention to features that make it easier to programme sounds is very important if you want to attain your maximum level of creativity. Craig Anderton offers guidance.
Craig Anderton explains what to look for.
When shopping for a synthesizer, most people know to look for major features like the amount of patch storage, whether split and layered keyboard configurations are possible, and so on. But paying attention to features that make it easier to programme sounds is very important if you want to attain your maximum level of creativity. I've found the following features to be very useful when programming synths; of course, no one synth has all the features listed below, so pick the ones with the greatest relevance to your needs and desires.
Lower-priced synths use one envelope generator to feed both the VCF and VCA (Voltage-Controlled Filter and Amplifier). While this is adequate in most cases, having a separate envelope generator for the VCF and VCA gives more flexibility. It's even better if there's a third envelope generator for the VCO (Voltage-Controlled Oscillator), or some means of allocating one of the other envelopes to the VCO when desired.
Being able to bend up or down to pitch via a programmable amount helps greatly when synthesizing wind, plucked string, fretless, and percussive instruments. Wind instruments tend to start off slightly flat and bend up to pitch; plucked strings and drums start off slightly sharp and bend down to pitch. With fretless instruments, few instrumentalists can hit the exact right pitch - like vocalists, they'll start a little flat and slide up to the right pitch. These are subtle effects, but programming them in can aid greatly in the realism of a patch. If an autobend function isn't available, check whether an envelope generator can be pressed into service to perform the same function.
A multitimbral instrument lets you assign different voices in the synth to different sound programs and different MIDI channels; thus, you can have more than one sound available from a single synthesizer when driven by a MIDI sequencer. For example, with an eight-voice multitimbral instrument, you might assign six voices to play a piano part on MIDI channel 1, and two single-note percussive effects to play on channels 2 and 3. This kind of feature helps stretch your synth's value.
MIDI continuous controller 7 can carry master level data that determines the overall volume output of a synth. Before implementation of this MIDI parameter became commonplace, programming level changes meant programming several patches at varying volume levels, and initiating a patch change when appropriate. This technique used up memory locations, though, and changing patches might produce a short glitch. Using controller 7 data (typically from a sequencer) to control volume lets you get more mileage out of a single patch.
Some synths have built-in chorus, delay, and so on. This offers two big advantages; you don't need to invest money in an external processor, and frequently, the signal processing parameters can be stored as part of a sound patch. It's like getting a MIDI signal processor for free.
Most people like dual oscillators because they can be detuned slightly to create a 'fatter' sound, but you can accomplish the same effect with a single-oscillator synth and an external chorus unit. The reason I like dual oscillators is because you can tune them to different intervals (octaves and very high harmonics are my favourites); also, synths with two oscillators often allow 'hard sync', a nasty-sounding synth effect that's really good for rock leads and bass parts.
Velocity responds to the dynamics of your playing, aftertouch to pressure exerted on a key after it's down. Both allow you to build more expressiveness into your patches.
Preset machines, where the sounds are etched in ROM and cannot be altered (except temporarily), are great for getting up and running with a minimum of effort, but they're a drag if you like to come up with your own custom sets of patches. Fortunately, the patch storage space problem can be alleviated with voice editing software but this is not extremely practical for quick patch changes in a live performance situation.
Parameter-control synths, where you can only adjust one parameter at a time, make for tedious programming. A good voice editor will, in conjunction with a personal computer, display all parameters simultaneously so that you can get an instant overview of your patch. Some editors let you print out hard copies of patch data, set up libraries of particular patches, randomise parameters, and other goodies.
It's very handy to compare an edited version of a patch with the unedited version to see which one you like best. Most newer synths have this feature; with older synths, one trick is to dedicate a specific patch location as an 'edit buffer'. Copy the patch to be edited over to the buffer, and do all your editing in that location. As needed, you can compare the edited version to the unedited version. When you like the edited one more than the original, overwrite the original patch.
The more waveforms, the better. Synths that only produce square and sawtooth waves will always tend to sound a bit 'buzzy'; having access to triangle and pulse waves makes a big difference. Even better are the synths that offer multiple digitally-generated waveforms (Ensoniq ESQ-1, Kawai K3, Roland D50, Korg DW8000, etc) as these allow for more options.
Most synths have pitch bend and modulation wheels, but sometimes you want to be able to change some other parameter, such as envelope attack, filter cut-off, etc. If your synth uses a knob/slider for parameter value selection, you can call up the parameter you want to vary in real time, and use the knob to make the desired adjustments. This technique is not convenient with synths where parameter values are entered with a numeric keypad (although these kinds of synths generally cost less).
Although this feature is less relevant these days since most synths can receive MIDI program change commands, sometimes it's easier to just set up a preset series of patch changes for a song and step through them with a single press of the footswitch during live performance.
I'm often asked why some synths have, say, 16 voices when you only have ten fingers anyway. The answer is that if a patch has a significant amount of envelope release time, a note can continue to sound even after you've taken your finger off the key. If all the available voices are sounding, then trying to play a new note will 'steal' one of the voices, thus cutting short the envelope's release phase and producing an unpleasant truncation effect. Although you'll seldom run into this problem with short, percussive patches, string washes and other long-release sounds often need all the voices they can get. One alternative to putting a lot of voices onboard is the MIDI 'overflow' mode offered by some synths; if all the voices in the synth are tied up, extra notes are re-directed out of the MIDI port to a second synth module.
It's useful to have lots of third-party patch support available for your synth. Even if you create all your own original programs, you can learn a lot from other people's programming techniques, and these patches often serve as an excellent point of departure for your own experiments.
As mentioned at the beginning, no one synthesizer has every desirable feature imaginable. But if some of the ones above sound important to you, take them into account when looking for a good synth.
© 1987 Electronic Musician magazine ((Contact Details)) and used with the kind permission of the Publishers.