The Waiting Room
by Paul Batchelor
A sine wave table is used for the many sinusoid LFOs used in the patch. A table of notes is created using gen_vals, which will be used for the main melody.
_sine 8192 gen_sine _seq "62 69 72 71 74" gen_vals
The pads provide the canvas for the patch. For a sound generator, a buzz generator is used, called gbuzz. gbuzz produces a set of harmonically related sinusoids, which approximately sounds like a sawtooth. The strength of each successive overtone is modulated with a sine LFO via osc*. This modulates the overall timbre of the sound.
62 mtof 0.3 8 3 15 inv 1 0.2 _sine osc 0.1 0.6 biscale gbuzz
This sound gets repeated 2 more times to form a triad with notes C, D, and A, which outlines some sort of D7 chord. The phase and frequency of each LFO is tweaked to make the voices more individual.
60 mtof 0.3 8 3 10 inv 1 0 _sine osc 0.1 0.6 biscale gbuzz + 12 inv 1 0.75 _sine osc 0.1 0.6 biscale gbuzz +
The summed signal is attenuated by 6dB and put through a butterworth highpass filter.
-8 ampdb * 400 buthp
The chorused pluck sound is the lead instrument sound that plays the melody. It begins with a clock signal generated via clock, and then put through a maytrig. This signal is duplicated. One will feed the sequencer, the other will be the trigger signal for pluck.
0 77 1 clock 0.4 maytrig dup
The melody generated comes from the seq table generated above. tseq is set to shuffle mode, picking notes randomly. A small bit of portamento is added as a stylistic choice, but it is not necessary.
1 _seq tseq mtof 0.003 port
The rest of the arguments for pluck, then a lowpass filter butlp. The last argument to pluck is the lowest intended frequency to be used with pluck. This is a static value that sets the buffer size. This parameter is worth experimenting with a bit, as it drastically changes the sound and decay of the sound.
0.5 50 pluck 3000 butlp
The puck signal is fed through a string resonator, which is the filtered version of pluck. This string resonator is tuned to F#, a major third above "D". The idea is to roughly simulate nodes and antinodes of a string body. When the plucked string plays an F# it will ring louder. streson and pluck are known to build up a lot of DC, so a dc blocker is placed at the end.
66 mtof 0.91 streson dcblk
This string sound is duplicated, and fed into a chorus effect. Sporth does not have a chorus effect, but a makeshift one can be made easily using a vdelay and slow, small modulations of the delay time. For good measure, a dc blocker is used here as well (I must have been getting a lot of DC when making this patch!)
dup 0.4 0.1 1 0 _sine osc 0.002 0.008 biscale 0.2 vdelay + dcblk
The chorused instrument is then fed into a delay line whose delay time is set to a dotted-eigth note duration. This delay is fed into a butterworth lowpass filter to keep it out of the way, then scaled to make it quieter.
dup 0.7 77 bpm2dur 0.75 * delay 2000 butlp 0.5 * +
This signal attenuated by 3dB, then added to the rest of the mix.
-3 ampdb * +
This instrument provides the bass sound, a constant pulsating drone. It is comprised of two detuned bandlimited sawtooth oscillators tuned to a D.
26 mtof 0.2 saw 38.1 mtof 0.3 saw +
As stated in the title, this is a subtractive patch, and the lowpass filter used here is diode, a filter design based of the one used by the TB303. The filter cutoff is modulated by a slow-moving LFO whose period is 8 seconds.
8 inv 1 0 _sine osc 500 2000 biscale 0.1 diode
This sound is added the rest of the patch. The entire patch is attenuated by 3 dB.
+ -3 ampdb *