Dave Stewart's Music Seminar (Part 3)
Simple But Not Obvious
Part 3: Famed keyboardist Dave Stewart continues his monthly column designed to get you thinking more about your music.
What is this thing we call 'feel'? We say 'it doesn't feel right — the snare feels late. Try bringing it forward 15 milliseconds'. Orchestral musicians don't know what we are talking about. I once overdubbed keyboards onto a tape of some film music performed by 40 orchestral players. At the start of one piece, the conductor had counted everyone in. His voice was quiet, but audible — '1, 2, 3, 4...' — I played the downbeat. About a second later the strings crept in, as if playing the second or third beat of the bar. I looked at the composer in alarm. 'Don't worry,' he said, 'it's normal. They're supposed to be playing the downbeat. Try starting a little later'.
This seemed mighty strange to me, but clearly orchestral musicians' sense of rhythm is far more fluid than that of their rock counterparts. Hence the old story (stop me if you know it) of Clem Cattini, veteran British session drummer, who on one session was getting more and more frustrated at trying to play with a string section. 'What's up, Clem?' asked the producer over the talkback, having noticed the drummer frowning and muttering to himself. 'Anything wrong with the balance?' 'Yes,' Clem replied, 'Can I have the strings a little sooner in the headphones, please?'
Ha! Apt rejoinder, Clem! However, these classical Charlies must know something we don't, because when the guy at the front waves his stick around, they all play together. For the life of me, I can't discern a beat in that continuous, languid movement... can you?
Perhaps the gap is widening — at band rehearsals, young musicians, well versed in click tracks and EBU timecode, snap at each other for coming in four milliseconds early, while the symphonic players spend so long delaying the downbeat that the concerto is over and the audience have gone home before the first note has sounded. At any rate, many of us older, non-classical musicians have been forced by the emergence of the machine as a dominant instrument in popular music to completely re-think our ideas on timing and rhythm — and a painful process it has been, too.
I take it that many of you reading this will have some experience of working with drum machines and sequencers. If so, you'll know that trustworthy gadgets can sometimes make rhythms feel appallingly stiff. Some of you may have tried to rectify the problem by shifting the rhythmic placement of the sounds around, say by delaying the bass drum 10 milliseconds or advancing the snare by one 96th of a quarter note. This can work, especially with a badly processed drum sample which contains a small amount of silence at the front of the sound, or a rhythmically ambiguous event such as a flam or a grace note (the sort of things shown in Figure 1).
Another approach is to move the whole drum part ahead of or behind the other parts so as to create more rhythmic tension or relaxation. But even after doing all this rather tedious stuff, you can still end up with a rhythm track that feels stiff or plodding — instead of an easy, athletic lope, it becomes a dismal trudge through the mud, at any moment threatening to come to a complete standstill.
The reason is that 'feel' and swing are not always determined by rhythmic placement. There is another, equally important consideration, and that is the question of arrangement — most especially the choice of sounds and their dynamics. In order to illustrate what I mean, I want to show you the simple parts that play in the chorus of the song Heatwave, which Barbara Gaskin and myself have recorded. Under the chirping crickets, mystic keyboard textures and mysterious bells and gongs, the bass and drum parts shown in Figure 2 are pumping away.
The critical thing about these parts is not the choice of notes, or the way that the snare drum is 306 nanoseconds ahead of the rest of the track (it's not) — it is the dynamics. The hi-hat part (played by an Emu SP12 sampling drum machine) uses three samples: quiet, loud, and open. If the quiet sample is set at too high a volume in relation to the loud sample, the part feels like Figure 3 — ie. too stiff. Too quiet, and it sounds like Figure 4. Only when the volume balance between the loud and quiet hi-hat is right does the rhythm begin to feel good. Nothing to do with rhythmic placement, but everything to do with dynamics. Similarly, the main bass part (although sequenced and quantised) has to be played with the right accents in order to pump along in the right manner. (For those who find numbers fascinating, I've printed out the MIDI velocity values for the first bar.)
This means that if a keyboard is to play the bass line (which it always does in my current music), a dynamically responsive sound must be chosen. Are your bass sounds dynamically responsive enough? If not, try increasing the 'velocity sensitivity' of the appropriate oscillators/operators/partials/sound generators (whatever you want to call them — the things that make the noise) till the difference between a loud note and a quiet note is clearly audible. You may have to increase the overall volume level of the program, as it will tend to go down as the velocity sensitivity is increased — experiment till it feels right to your particular touch. If your sounds are already too dynamic (ie. too much difference between a quiet note and a loud note), then reduce the velocity sensitivity and overall volume.
A cautionary note: do not be tempted to go through your sequencer parts editing all the velocities to be the same from section to section — life's too short for that kind of thing. If it doesn't sound right, just play it again until it does!
One other important factor that affects the feel of the main bass part is the length of the notes. Most of the time, the notes (marked staccato on the score) are extremely short — something like one sixth the length of an eighth note, my sequencer tells me. This means that the longer, accented notes at the end of certain bars really stand out: dup, dup, dup, dup, dup, dup, dup, DAAAAH!!
These rhythm parts are, of course, only the bare bones of one part of the song arrangement, but even in this the choice of sounds was of paramount importance. The drum samples were all mine, a mixture of dry hi-hats, ambient snares, distant echoing toms, and close thudding bass drum that could never coexist in any real life acoustic environment but nevertheless somehow magically work together.
The main bass sound was a blend of two DX7 programs — one a warm, deep fretless bass sound, the other a hard slap bass. The latter was mixed in at a fairly low level to add definition to the front/attack of the fretless sound. To add size and aggression, I added the Wagnerian 'Big Bass' sound, a thunderous mix of three DX7 metallic bass programs and a sampled grand piano playing bass octaves. Played live, it helps to widen the beat somewhat and offset the rhythmic precision of the sequenced parts.
On top of this pulsating rhythm racket, other keyboards picked out some lines that defined the chord sequence without actually playing pads or long notes. Figure 5 shows two of the parts. Note the left hand of the chordal part playing its own, happy little independent bass rhythm — the sort of thing my left hand does when I'm not looking. Interesting to note how it fills in the gaps in the acoustic guitar part!
All music © Broken Records 1990. The song 'Heatwave' is published by Flag 22 Publishing. Title used by kind permission. Import copies of Dave Stewart & Babara Gaskin's CD The Big Idea, featuring 'Heatwave' and nine other tracks, can be obtained via mail order from: Broken Records, (Contact Details).
Broken Records will pass on any correspondence to Dave & Babara.
Feature by Dave Stewart
mu:zines is the result of thousands of hours of effort, and will require many thousands more going forward to reach our goals of getting all this content online.
If you value this resource, you can support this project - it really helps!