Most synths now boast an infinite variety of sounds, but what the manuals don't tell you is that you need an infinite amount of time to explore them properly. Paul White suggests some short cuts to synth heaven.
I'm a great fan of technology, particularly when it makes my life easier, or allows me to achieve something that I couldn't before. Where I do lose patience is when technology builds barriers between me and my ultimate goal, and sadly, that's where many synth editing systems fall down — they frequently make it so awkward to change anything at all, that very often you're discouraged from even trying.
This only becomes a real problem when the synth's preset sounds are not quite suitable for the use you had in mind. The same goes for sound cards — if you do find a selection of off‑the‑shelf sounds that are ideal for your needs, then great, but more often than not, they'll need some tweaking.
Fortunately, a compromising approach can be taken to sound editing, turning an off‑the‑shelf patch into a custom one with the minimum amount of effort. In many cases, all you need to do is to alter the attack and/or release times, but most modern synths build their patches from two or more sounds or voices, each of which may have its own level envelope settings. Unless your particular synth has an overall envelope control function, you'll need to find out how to access these separate envelope parameters. If you have a poor memory for button sequences, write the procedure on a card and stick it to the top of your synth. On most occasions, you'll only need to change the A (Attack) and R (Release) values. Decay and Sustain can usually be left as they are.
Another approach to customising patches is to layer existing sounds together — there's nothing new about this, as all additive‑type synths employ layering within their own patch creation systems. If, for example, you need a string patch, but find analogue strings sound too soft and sampled strings too edgy, you can combine the two and achieve different results by varying the balance. If you don't mind a little low‑key editing, you can do this by choosing an existing synth patch on an additive‑type instrument, then trying out a few different raw samples or waveforms in place of the ones provided in the preset. Just run through the various string and pad sounds, then see which two work best together. You can still keep all the envelope and modulation parameters of the original patch, although with strings, it's often worth tweaking the attack and release times.
Most synths allow this kind of low‑level editing to be done quite easily, but you don't have to edit at all if you don't want to; you could simply layer two presets from two entirely different synths, or layer two or more parts of a multitimbral synth. If you're working with a sequencer, this is achieved by copying the original part to another sequencer track, and then playing the copied part back via the new instrument, as well as the original one. If you're playing live, all you have to do is link two MIDI instruments together and arrange your patch mapping table so that the matching sounds are called up by the same Program Change number.
Staying on the subject of strings, I've found that you can make a simple string patch far more interesting by using two sounds with different attack times, so that the sound timbre changes as the second instrument comes in. This trick also works well with choir patches, where it can be effective to choose a second sound that's an octave higher than the first.
Layering isn't restricted to sounds from the same family, nor do they have to be complementary instruments, as with the endless piano and string layers available. To create a new‑sounding picked instrument, for example, you can combine virtually any sounds, providing they have suitably fast attacks. Harpsichords, guitars, hammer dulcimers, bass guitars tuned up by an octave or two, pianos and the like all work well together, resulting in an instrument that has some of the characteristics of its component parts, but also has its own distinctive sound. One example of this type of layering that works exceptionally well is where a short, percussive sound, such as a stick click or noise burst, is added to the start of a conventional bass synth sound. This really adds attack and definition to the note, without robbing it of any of its depth.
If you want to break away from recognised sounds altogether, but the instruments you have are full of instantly‑recognisable GM‑style patches, you can still create radically new sounds by remembering just two simple facts. The first is that the brain relies heavily on the attack characteristics of an instrument in order to recognise it, and the second is that familiar sounds can seem quite different when transposed up or down from their normal range.
Take a GM piano for example — if you slow the attack, the identity of the piano is lost immediately, and an abstract, bowed sound results. Similarly, notice how much more pronounced the attack part of the sound becomes when a mandolin is dropped by a couple of octaves. High flutes can be lowered in pitch to provide deep pipe sounds, while bass instruments can be raised a couple of octaves to provide a new lead instrument. Slow brass can be given a fast percussive attack, church bells can be given a slow attack to create a haunting quality — in fact there's no limit to what can be achieved, simply by messing about with the level envelope and voice transpose parameters.
The next phase in creating your own patches through layering is to choose one recognisable and one abstract sound. Simple filter sweep pads sit alongside string samples quite happily, but you might also find that layering the attack part of a distorted guitar with a piano provides you with a less predictable alternative to your existing piano or clav patches. More exotic results can be achieved by layering unpitched sounds (such as water bells), with tuned percussive noises (such as wood blocks or marimbas).
Guitar players have known for decades that the guitar itself is only a small part of the whole sound. The effects and the amplifier contribute at least as much to the end result, and it's the same with synths and samplers. We tend to think of synthesizers as instant gratification in a box, but perhaps it would be more constructive to think of them only as a starting point.
One of my favourite tricks is to find two recognisable sounds or instruments that have some tonal element in common, and then crossfade from one to the other as the note evolves. Human voices are similar in many ways to reed instruments, flutes and violins, and if you choose the right crossfade rate (simply by using the normal level envelope parameters for each sound), you can create quite an eerie, morphing effect that sounds far more subtle than the crossfade it really is. Brass pads merge nicely into strings, different types of wind instrument can be merged one into the other, and choirs can be merged into (or out of) just about any smooth pad sound.
My current favourite is a patch that starts out as a clarinet, and then moves into a slightly electronic‑sounding voice created from a voice waveform, rather than from a full vocal sample. I've managed to make this effect work on both my Proteus 1 and my old K1, where it has become a mainstay of my library music and new age (actually, middle‑age) endeavours. There's just something very evocative about these crossfades, but you never know if the combination has the necessary magic until you try it. The secret is to match the fade‑out rate of the first sound to the fade‑in rate of the second, so that the patch level remains nominally constant during the changeover.
Crossfading can also be used to graft the attack of one sound onto the sustain of another, which can yield some musically interesting hybrids. Elsewhere in this article, I mentioned that the attack of an instrument provides strong clues as to what that instrument is, so if you take something like a plucked attack and then quickly merge into a wind or string sustain sound, you surprise the hearing system and make it pay attention.
Many budget synths don't provide resonant filters, but if you do have a machine that includes them, you might find it easier to choose a patch with the right filter sound, and then change its component waveforms or samples, rather than choosing a preset that sounds right and trying to set up a suitable filter effect. If you don't have a filter section but would like to create some of the analogue sounds you hear on records, you might find that one of the currently available, MIDI‑triggered external filter boxes does the trick, such as Peavey's Spectrum Analog Filter or Waldorf's new 4‑pole filter (to be reviewed in SOS shortly). Most of these devices use a single filter section, so they don't work in the same polyphonic manner as internal synth filters, but for solo lines, bass lines or pad chords, they can add a huge range of depth and expression to an otherwise bland digital sound.
Both internal and external effects can be used to add character to a sound, and chorus is an obvious example to use on string pads, as it helps create an ensemble effect. On the other hand, chorus also tends to 'de‑focus' sounds, so if you want something to stand out, it might be better to err on the side of fewer effects. Reverb and delay effects also have their creative uses, but if you're working with a sound that already has a long release time, then you might find that delay and reverb only clutter it up. As a general rule, delay and reverb work best on percussive sounds or sounds that finish abruptly.
Finally, although you've probably got a pretty good idea of the effects that can be created using conventional multi‑effects boxes and internal synth processing, don't neglect the less obvious things, such as patching your synth through guitar pedals or speaker simulators. Speaker simulators (as found in most recording guitar preamps) are good for taking the rough edge off digital sounds and for adding mid‑range warmth, while overdrive units can be used to 'vintage‑ise' organ sounds, or add bite to solo lead lines. Even putting the synth through a guitar cab and re‑miking it can be worthwhile.