Magazine Archive

Home -> Magazines -> Issues -> Articles in this issue -> View

Using Sequencers (Part 1)

An Introduction to the Use of Sequencers

An introduction to the use of the perennial tools of the electronic composers trade. With some explanation of programming from music and notation.

Some form of automatic playing facility has been available to the Electro-musician for many years now. As the Synthesizer has always worked on trigger, gate and pitch codes of an electronic nature, be they Control Voltages or Digital Information, it was always a logical step to try different methods of controlling the Synthesiser other than with a keyboard. The first generation of sequencers, those working on analogue triggers, were very limited. They had something between 8 and 32 notes (all of which had to be individually tuned using a different potentiometer for the Pitch Control Voltage of each note - which made them relatively expensive in terms of the necessary hardware). Note-length was usually determined by a single Gate knob or simply by how fast the sequencer was run. This meant all notes had the same time value and no rests were possible. This sort of sequencer gave the characteristic feel to the early music of Tangerine Dream, Kraftwork, Faust and other early pioneers of electronic music. 'On the Run' from 'The Dark Side of the Moon' by Pink Floyd is a classic example of the sort of thing which could be done with such devices (an early EMS sequencer being used on this track). Perhaps the greatest exponent of the analogue sequencer is Georgio Moroder, who in the late Seventies used the more expanded models to invent a new type of music 'Electronic Disco' as exemplified in his production of Donna Summer's 'I Feel Love' or his Soundtrack for the film 'Midnight Express'.

The new generation of sequencers which worked on Digital Technology arrived with the Eighties and offered a vastly increased number of notes with an ease of programming unheard of before. One of this new breed, a British product called the 'Spider' offered 256 notes (or spaces) and programming from the keyboard of its companion, the 'Wasp' at a price which even the Home Electro-musician could afford. Indeed, digital synthesizers like the 'Wasp', could even be run by home computers. These days, the wide range of sequencers on the market are programmable in the same way as computers, with full edit functions as well as being programmable from the musical keyboard. They also allow musical subtleties such as 'staccato' or 'legato' phrasing and tempo variation to enable 'authentic' musical performances.


Trigger. A Pulse, either digital (an electronic code) or analogue (a voltage) which causes the synthesizers envelopes to open, allowing a note to begin sounding.

Gate. The continuance of the trigger pulse which holds open the envelopes in the sustain phase (i.e. after attack and interrupting decay and preventing release), simulating the amount of time the finger remains on the keyboard.

Real Time. A sequence which plays back exactly as it was recorded. Useful for fast programming and solo performers' accompaniment but can cause synchronisation problems for group or multi-tracked performance.

Pulse Time. A slower but more controlled method of programming either from a musical or computer keyboard, allowing composition, edition and synchronisation (often referred to as 'Step-by-Step').

Clock Signal. An internal or external series of regular pulses which form the basis of all trigger pulses and allow synchronisation between sequencers drum machines, etc. Often referred to as 'tempo' or occasionally 'frequency'.

Space. The name of the facility on most sequencers of the digital type which allows you to insert rests into the sequences (occasionally this is actually labelled 'rests').

Delete. The name of the facility which allows you to erase wrong notes and spaces from sequences.

Insert.The name of the facility which allows you to add extra notes or spaces in the middle of an existing sequence.

Step (Forwards or Backwards). The facility which allows you to move about within a sequence without changing anything until you find the point where a change is required. Step number tells you the point at which you are on more sophisticated sequencers.

Loading Sequencers

Sequencer controls on the Roland JX-3P.

The most immediately obvious way to do this if you have a real-time sequencer is to simply press record and play the music in. But there may be several reasons why this is not the best idea. Most obviously you may not trust your technique or timing. Don't forget that if you want to 'loop' (repeat) the sequence you will have to 'play' the stop button at the appropriate point as well. You may wish to synchronize the sequence to another or to a drum machine (very difficult with a real-time sequence and only possible if the internal clock which was used to analyse your playing can be synchronised fast enough or indeed, is externally accessible at all). Many real-time sequencers have no edit facilities available.

Sometimes it may be possible with polyphonic sequencers to put one of the parts (preferably the most difficult) down in pulse time and then add the rest in real time. Whether you do this or put everything in step by step, you will still need to take care with the clock/trigger value.

To decide this, look for the smallest note or rest value in the music you wish to program. This will determine the value of one trigger in pulse time or the clock speed in real time. Even if this note-value occurs only once in the piece, if it is not allowed for, you will not be able to program either this note or the one that follows it properly.

If in real time, set the Clock ('Tempo' or 'Frequency') so that the fastest note(s) you are going to play are correctly recorded (you will have to experiment till you get this right). On some real-time sequencers (the DSX for example), the clock is permanently set at a suitably fast tempo. If you find notes missed out or the wrong value, then you must set the clock faster, until this no longer happens. However, this will use up the memory faster, making your potential sequence shorter. If you cannot find a compromise, you will be forced to resort to pulse time loading (even this may not be possible, although it does tend to record evenly-noted sequences more efficiently - minims and demi-semiquavers together will probably be more efficiently recorded (i.e. longer sequences will be made possible) in real time). If there is a 'Ready' feature on your sequencer which will start recording as soon as you touch the keyboard this will help to save space and prevent timing problems at the beginning of the sequence (fatal if you want to 'loop' and/or synchronize your sequence).

The Elka Synthex sequencer panel.

If you are going to program in pulse time, each pulse (that is each potential trigger) has to have the value of the shortest note in the piece of music and everything must be worked out on this basis, (There is one exception to this rule - where the notes are not all multiples of the shortest note-value e.g. where you have triplets or fifth notes, etc., which we will come onto later).

Now you must program in everything in multiples of this time-value. For example, if your shortest note is a crotchet, you will have four pulses to the bar in common time (4/4) and a minim will last for two pulses and a breve for 4. Or if your shortest note is a quaver, there will now be 8 pulses to the bar, a crotchet 2 pulses, a minim 4, a breve 8, etc.

Similarly in different time signatures, you will have different number pulses per bar. In 6/8 with quavers 6 pulses, in 2/4 with hemi-demi-semi-quavers 16, 9/16 gives 18 demi-semiquaver pulses per bar. Hopefully you will not have to work with this memory-wasting hemi-demi-semi business to often, but I have tabled most of the possibilities below.

OSCar's sequencer programming buttons.

However 'hemi-demi' problems are easy compared to triplet and fifth-notes in terms of pulse assignment and sequencer space wastage problems. If you have triplets, fifth-notes, etc., and the notes they are divided down from (i.e. 3 or 5 times their value) carry on as above. The complications really set in when you have notes only slightly different in length from each other e.g. quaver (1/8 note) and quaver triplet (1/12). To solve this we must go back to the fractions we did in school. What we need here is the lowest common denominator (LCD); a pulse length which allows us to program both the eighth note and the twelfth note. The LCD of 1/8 and 1/12 is 1/24.

So we need 24 pulses per bar to cope with quavers and quaver triplets. To program a quaver, we need three pulses (3/24 = 1/8) and for a quaver-triplet (2/24 = 1/12) we need two. Again all the ramifications of this are tabled below.

Chart for EMT
(Click image for higher resolution version)

Now take the number of pulses per bar and divide the total memory space by this. Say we have 16 per bar (semi-quavers in 4/4) and memory is 1288 notes (memory is usually close to a power of two as digital memory is stored as binary or hexadecimal numbers - 128 = 27; 256 = 162, then we have room for 8 bars of music. With 24 per bar (quavers and quaver-triplets in 4/4 we only have room for 5 and a bit bars. You see now how triplets and 'hemi-demis' swallow up storage space!

This table shows even more clearly how the high resolution necessary for triplets and 'hemi-demi-semis' uses up memory space rapidly; without them, 128 notes gives you more than 10 bars, with them to get five bars is doing well. So we must be careful what music we choose to load into our sequencer; in pulse time, music with lots of notes of the same time value tends to be much more efficient. But beware the piece which is entirely crotchets except for a couple of quavers! You will still have to program the entire piece in quavers pulses.

Now we have sorted out the mathematics, with our pulse assigned a time value and we know how many bars we can store, how do we program in the music? For time value notes equal to our pulse-length (i.e. the shortest) nothing could be simpler. Simply play the notes on the keyboard in your own good time and the sequencer will play them back in perfect time. If your sequencer has a legato facility, don't forget to make the most of this by holding down the last note while you play the next one for notes which are phrased in the music. To program a space simply press the key marked 'rests' or 'space'.

To enter notes of longer length, we must first know how many pulses they will take up, i.e. how many times bigger are they than our shortest time-value. For instance a breve is worth 4 crotchets, a minim 16 hemi-demi-semi-quavers, a dotted crotchet three quavers and so on. Now we must make sure that this many pulses are taken up in our sequence. If we are inserting a longer space, there is no problem. We simply press the space/rests button the appropriate number of times, 4 for a minim in quaver pulses, 6 for a dotted crotchet rest in semi-quavers and so on.

How we do this for notes is dependent on the comprehensiveness of our sequencer. On the ideal machine, we could simply press a key corresponding to the the note value (indeed on the Elka Micropiano, this is how it is done, allowing maximum efficient use of the memory space). However, most sequencers don't have this feature and we have to program longer notes "long-hand". On some, this is done by playing the note - which takes up one pulse - and then pressing the "rests/space" button to fill in the remaining number of pulses(essentially legato-phrasing that note to a number of spaces, which, has precisely the required effect). On others we must press the "step forward" button the appropriate number of times (don't forget the playing of the note always takes up one pulse), while we hold down the note in question.

On sequencers which make no provision for longer length notes than the pulse, it is still possible to get this effect by sacrificing either repeated notes, or spaces. To do this however we need to make changes to the envelope on the synthesizer. If we are going to manage without spaces, we must turn up the release on both filter and amplifier envelopes so that it lasts as long as the note-length required. Then we simply fill in the remaining necessary pulses for our note-length with spaces. The release controls will then hold the envelopes open after the trigger has been and gone, making the note the required length, i.e. until a new note is triggered.

If it is repeated notes we are going to sacrifice, then we must turn release up a bit and sustain to full on both envelopes. This will prevent retriggers of the same note sounding. Now we simply fill the number of pulses required with the same note repeated the appropriate number of times. Of course, both these methods entail some loss of musical freedom, as changing the envelopes always modifies the sound somewhat, but this still gives us greater flexibility in accurate programming of note length.

Let us now look at some music to see how we would go about putting it in a sequencer.

Let us begin with a typical synth pattern which is mainly in semi-quavers (1/16th notes) in 4/4. For those of you who don't read music I am putting a programming notation below the music lines. A retriggered note is shown by its name, whereas a note tied over more than one pulse is represented by t (don't forget you can use a space if you have no facility for tying notes) and means a legato tie (if available). There are 16 pulses to the bars in this example, although the pattern repeats itself within the bar.

(Click image for higher resolution version)

You will notice that the second phrase is merely the first one transposed up a major 3rd and that the third one is the same up a tone from the original. This means that if your sequencer allows transposition, you can loop a single bar (or even half-bar) and transpose up 2 tones after 8 bars (16 half-bars) and up 1 tone after another 4 bars (8 half-bars).

Now here is a drum pattern to accompany the bass-line. This is written out in our usual format - bass drum on the bottom line, then snare, with hi-hat crosses on the top line with o and c indicating when it is to be opened and closed. We also include our usual block chart for ease of programming onto drum machines.

(Click image for higher resolution version)

If you are synchronising your bass pattern with a drum machine, you will need to program a trigger on every 16th note. This can be done using the toms on the Korg KPR-77 or the Roland TR-606. However, if you cannot put 'spaces' in your trigger, try programming the rhythm below on the Lo Tom to trigger the sequence and only load 14 notes per bar (missing out the two t beats). This trigger pattern will then play the notes in the correct rhythm.

(Click image for higher resolution version)

Now let us move on to add chordal accompaniment to the rhythm section. Here is the music part with the three lines written out as they should be programmed in (r means a rest should be programmed):

(Click image for higher resolution version)

Again a 16-pulses-to-the-bar trigger will run this sequence in time with the rhythm parts. However, if your sequencer doesn't have a 'space' facility or enough memory, then the same effect can be obtained by loading these notes; and using this triggering pattern from the Hi-Toms on your drum machine:

(Click image for higher resolution version)

Now we have the accompaniment, we can add the melody line, either manually or on the sequencer, if you have a fifth line. By looping the lines sequenced so far, you can play the melody and then improvise over eight bars E6 (+ 2), four bars G = (Double Harmonic) and four bars F = (Pentatonic Major) cycle. Anyway here is the tune:

(Click image for higher resolution version)

(Click image for higher resolution version)

As for the choice of synth sound to play each of these lines, this is to some extent dictated by the nature of the different roles they play within the piece and the length and speed of the notes. Try a full sound (PWM or sawtooth) with a fast filter envelope on the bass-line. The chords of the polysequence sound good with sharp attack and a fairly quick decay, whereas the lead line needs a good 'sustain' sound to carry it over the long notes. A woodwind-type sound, perhaps with some delay vibrato, adds a nice lyrical feel to the piece, which is, incidentally, called 'The Pulse'. Anyway, have fun experimenting with different sounds once you have loaded your sequence notes. This is of course the joy of using sequencers: you can program your music and then make the 'producer's' decision of how it should sound.

The first of the new series of E&MM demo cassettes features this piece on the Oberheim system (reviewed elsewhere in this issue) and on more modestly-priced equipment from Roland.

'The Pulse' Copyright 1983 Paul Wiffen.

Series - "Using Sequencers"

Read the next part in this series:

All parts in this series:

Part 1 (Viewing) | Part 2 | Part 3

More with this topic

Browse by Topic:


Previous Article in this issue

Oberheim OB8

Next article in this issue

Stop Press - MemoryMoog Plus

Electronics & Music Maker - Copyright: Music Maker Publications (UK), Future Publishing.


Electronics & Music Maker - Jan 1984

Donated & scanned by: Stewart Lawler




Using Sequencers

Part 1 (Viewing) | Part 2 | Part 3

Feature by Paul Wiffen

Previous article in this issue:

> Oberheim OB8

Next article in this issue:

> Stop Press - MemoryMoog Plus...

Help Support The Things You Love

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!

Donations for December 2021
Issues donated this month: 0

New issues that have been donated or scanned for us this month.

Funds donated this month: £4.00

All donations and support are gratefully appreciated - thank you.

If you're enjoying the site, please consider supporting me to help build this archive...

...with a one time Donation, or a recurring Donation of just £2 a month. It really helps - thank you!

Small Print

Terms of usePrivacy