Site hosted by Build your free website today!

Multidimentional Polyphonic Sequencing


As everyone knows, an analog sequencer is a DC step voltage generator device which produces a series of independently
adjustable control voltages over time.

This kind of device, which was first developped by Don Buchla in the 60's, can be used in various ways : as a DC voltage
source, as an audio source, as a waveform generator source or as a logic programming device. In basic sequencing,
it is often used as a DC voltage source to control the frequency of a VCO (1 Volt/octave input) as well as the
frequency cut-off point (Fc.) and amount of resonance (Q factor) of a VCF.

Sequences Within Sequences

As you might have already discovered by yourselves, sequences often behave in unexpected ways due to strange streaming

Researchers Steve McAdams (Stanford School of Medecine) and Albert Bergman (Dept.of Psychology at McGill University)
have done extensive research on the subject of "sequence streaming effects" (i.e a sub-liminal auditory perception
that a given sequence is suddenly splitting into two or more sub-sequences streams).

In their work entitled "Hearing Musical Streams", they claim that the perceptual effects of a sequenced sound is dependent upon
the musical context in which that sound is embedded : i.e a sound's perceived pitch, timbre and loudness are
influenced by the sounds that precede it, coincide with it and even follow it in time! (1)

This is what Steve McAdams has to say on the subject of "streaming": "The faster the tempo of the sequence, the greater the degree
of breakdown of the tones into separate and narrow streams: when the tempo is very high, the sequence disintegrates and
every tone's individual frequency is beating along in its own stream.
Thus, in a tonal pattern, the particular relationship between frequencies - and not just the frequency separation
between adjacent tones - plays a vital role in the formation of streams".

Paradigm Shift

Alas, music generated by analog sequencers is often monotonously repetitive and lifeless : this is due mainly
to a "cultural myth" which suggest that the timebase of a sequencer should be a metronomical clock with a rhythm in 4/4. The end result, is often
too machinistic: each step in the melody having the same length!

The following illustration shows a small 4 x 8 steps sequencer, where the usual step pushbuttons assign switches have been
replaced by Banana Jacks. Why?: To be able to activate individual steps by Pulse control! Indeed, by doing so, each step can now have its own tempo and rhythm on a vertical plane (see Figure 1).

Two CD4526BC counters, both driven by an external clock, are used for complex clocking purposes : they can be configured either in
parrallel (2x1 stage) or in series (2 stages).

In this case, the first counter (A) is programmed - either by means of binary switches or by Pulse control - to deliver a rhythm in 3/16,
while the second counter (B) is set for a rhythm in 8/16. So, sequencer steps 1 to 4 will have various frequency divisions
of a rhythm in 3/16, while steps 5 to 8 will have frequency divisions of a rhythm in 8/16.

Note, that these counters are also programmable frequency dividers: each output Q1 to Q4 is half the tempo of the
preceding output (see Figure 2).

For example, when output Q1 of Counter A is patched to Sequencer step 1 Pulse control input, only the control voltages settings of
potemtiometers A1 B1 C1 D1 will be available at the respective A B C D outputs (Blue banana jacks).

Now, that we can address each step in the sequence with a different clock tempo and rhythm, let's patch four
independent MONO synthesizers - or a multitimbral polyphonic synthesizer -with some interesting percussive sounds.

Then, tune all VCO's of Synth 1 to a known initial frequency, while in full VCA mode. Next, connect Sequencer CV output row A to the 1 Volt/oct.inputs of all your VCO/VCF.
That done, apply a *5V Gate to Sequencer step 1 Pulse input and program potemtiometers A1 B1 C1 D1 for a suitable tonal melody
in a given key. Proceed to program all the other steps in the same manner.

Recheck the coherence of all melodic lines, especially steps 1 and 5, 2 and 6, 3 and 7 and 4 and 8, until you are satisfied with the results. Finally, patch counter A pulse outputs Q1 and Q2 to trigger the two envelope generators with VCA1 and VCA2 set in normal mode.
Refine your percussive patch if needed.

Proceed to connect the other three synths in a similar way, using the remaining Control voltages and the other Counters Qn Pulse outputs.
That done, recheck if the melodic lines of all 8 steps on the vertical plane, are coherent for all your four Synths
and make the necessary corrections if necessary.

Play with your "clockless sequencer" for a while...

Now, that we have seen how the sequence behaves on the Y plane, let's experiment with clocking trajectories on the X and Z planes.

Obviously, you can "step" the vertical sequences on the horizontal plane. To do that, use a slow clock (like output Q4 of Counter B) and patch it to the sequencer horizontal clock input (gates n steps ---> on the X plane).

Also, if you have an Up/Down counter input, connect it to output Q4 of counter A (gates n steps <---> on the X plane).

Similarly, if you have a vertical clock input, connect it to the sequencer Pulse output 8 (gates n steps Top/Bottom on the Y plane).

Finally, you can use the RESET,HOLD and RND functions to make your polyphonic sequence even more livelier and exciting.

Enjoy ...

Andre Stordeur

(This article is an excerpt of "Ye Olde Timer's Analogue Cookbook"by the author).