The GS Multiplex module

GS Multiplex (short for Granular Synth Multiplexer) is a companion module for the Granular Synth. It was specifically designed to act as a bridge connecting the N-Step sequencer to the Granular Synth but has several other uses, in particular the production of complex layered sounds.

It uses the same time-division multiplexing mechanism as the GS MIDI X module. Multiple channels of information are transmitted down single cables in order to make it appear that the Granular Synth is doing multiple things at once.

In very rough terms GS Multiplex turns the Granular Synth into four semi-independent monosynths. Each of these monosynths is called a timbral part or just a part.

Timbral Parts

The term “timbral part” is used to mean a grain stream whose parameters can be controlled with significant independence from other grain streams. By significant I mean beyond simple factors like pitch and amplitude.

On its own the Granular Synth is monophonic and has a single timbral part.

With GS MIDI X we have up to 128 simultaneous notes available (controlled by a MIDI channel) but still only a single timbral part. Although with the keyboard split mechanism one could argue that the GS MIDI X setup does provides two timbral parts. But as this works with parameters just having an offset applied or not depending on which side of the split a note is on it’s not really a significant level of control.

With GS Multiplex we have up to 4 simultaneous notes available (controlled by four sets of pitch, gate and velocity inputs) and four timbral parts.

GS Multiplex turning the Granular Synth into four monosynths

The level of independent control over these timbral parts is not complete though. Not every single control on the Granular Synth is duplicated in each part. But one does have control over pitch, amplitude, seeding and four other configurable parameters. This gives us a good deal of creative control without it becoming too complicated or taking up too much screen space.

GS Multiplex’s control of amplitude is also quite versatile as it’s utilized to provide velocity response, manual and voltage controlled volume adjustment, manual and voltage controlled muting/fading and a handy solo feature.

Interfacing with N-Step

GS Multiplex was designed with the Adroit N-Step sequencer in mind so it serves as an ideal bridge between N-Step and the Granular Synth.

GS Multiplex linking an N-Step sequencer to a Granular Synth

The four CV channels of N-Step pair with the four parts of GS Multiplex and the same color-coding is used for the knobs as one will often use the red channel of N-Step to control the red part of GS Multiplex and so on.

This arrangement makes it straighforward to produce four-part harmony but this is just one of several possible configurations.

You would normally position a GS Multiplex module immediately to the right of an N-Step Main module. The various outputs and inputs are then nicely aligned so that it’s easy to connect the modules with short cables and relatively neat wiring.

Example of connections between N-Step and GS Multiplex

Color-coding the cables as shown in the example above is a good idea especially when sample and hold and perhaps the use of N-Step’s Bernoulli gates add further complication.

Note that you don’t have to pair N-Step’s CV channels with GS Multiplex parts, you could use some CV channels to control parameters other than pitch (velocity for instance). But it’s common that you will want to pair up at least one or two. Sometimes you may have a single CV channel wired to multiple parts as you can achieve some super layered sounds with this approach.

It’s also possible to drive one Granular Synth with two or more N-Step sequencers. Indeed this can be particularly useful when working on a computer with limited resources as with some careful management of offset (perhaps using ATQ) just one Granular Synth can be pressed into producing output that sounds like several separate instruments playing simultaneously.

Linking N-Step to the Granular Synth via GS Multiplex is such a large subject that it’s covered separately HERE.

Quick start guide

This quick start guide doesn’t show off any exciting features, it just covers the groundwork required to do basic interfacing. We’ll look at more interesting patches later (and in the separate page on using GS Multiplex with N-Step).

If you are really impatient then you could skip to the “Drums for beginners” section but it’s recommended that you work your way through the material here in order to get a full grasp of how GS Mulitplex works.

Let’s load up what could be considered the minimum patch…

GS Multiplex Init Patch

Downloading presets

This patch doesn’t do anything particularly interesting or novel (it just lets you play the Granular Synth monphonically in gated seeding mode with MIDI input from your DAW, a MIDI keyboard or Voltage Modular’s little virtual keyboard) and like any INIT patch it sounds pretty awful but it does introduce most of the wiring and control settings needed to get going.

Saving this cabinet in your Library as something like “GS Multiplex Init” will help speed up future patch building.

The patch provides conventional monophonic control of GS Multiplex’s red part (the upper of four rows) with these connections…

I/O Panel PITCH is connected to red V/OCT.

I/O Panel GATE is connected to red GATE.

I/O Panel VEL is connected to red VEL.

The connections below are where things start to get interesting.

GS Multiplex’s V/OCT is connected to the Granular Synth’s V/OCT.

AMP is connected to the Granular Synth’s AMP external CV input.

GATE is connected to the Granular Synth’s GATE.

SEED is connected to the Granular Synth’s SEED IN.

The Granular Synth’s SEED OUT is connected to the GS Multiplex’s SEED IN.

Outputs that are multiplexed have labels that are underlined.

A few changes have been made to the Granular Synth’s default settings…

1) The built-in Pad test sample has been loaded into the buffer (by clicking RECORD and selecting Pad from the menu). This just gives us a fairly neutral place to start exploring from.

2) The AMP knob has been turned fully CCW and the AMP parameter’s attenuvertor knob has been turned fully CW. This allows the AMP output of the GS Multiplex module to fully control grain amplitude in the Granular Synth.

AMP Settings

So 5 V from the AMP output causes a grain to have maximum amplitude while lower voltages produce quieter grains. There’s no point in producing silent grains so 0 V actually stops grains from being seeded. It’s important that the Granular Synth’s AMP knob is turned fully CCW in order to benefit from this efficiency gain (as well as to stop sounds that should be muted from leaking through).

3) SIZE has been turned up to about 70% so that grains last a reasonable amount of time.

4) The Envelope Generator’s LOOP button has been disengaged so that the envelope gate corresponds to the GATE output of the GS Multiplex module without gated looping occurring. Although the Envelope Generator is not actually used in this patch.

Envelope LOOP disengaged

5) The OMDU switch has been set to Monophonic. This disables the arpeggiator but leaves quantization active. To disable quantization too set the switch to Off.

A few initial things to observe…

You should be able to hear sound when you play the keyboard.

It’s monophonic and the sound currently sustains when you hold a key down.

The LED rings around all of the GATE input sockets light up when you press a key. This is because the inputs are normalled so that if you plug a cable into a GATE input socket the sockets beneath also receive the same signal unless this is overriden by another cable being plugged in. This reduces the number of cables required when creating layered sounds.

Normalled gate inputs

Only the red part’s ON button is engaged so the other parts ignore the shared gate signals they are receiving.

The red VOLUME knob controls the volume of the sound.

If you are playing on a MIDI keyboard that is velocity sensitive then if you tweak the VEL IMPACT knob you should notice that you can adjust how much impact varying how “hard” you play the keys has on the loudness of the result. Providing that the VEL IMPACT knob is turned up you should be able to see the red meter to the right of the volume control change in response to how hard (or more accurately how fast) keys are being struck.

The red OCTAVE knob controls the octave of the sound.

MS is short for multi-seeding. Because the red part’s MS button is engaged, holding down a key causes the note to be sustained by grains being repeatedly generated. The stream of grains is produced at the rate of the Granular Synth’s internal clock (that’s why there’s a connection from the Granular Synth’s SEED OUT to the GS Multiplex’s SEED IN).

If you disenagage the red part’s MS button then only a single grain is produced when a key is first pressed. This is called one-shot mode.

This one-shot/multi-seeding distinction is the same as with the GS MIDI X module but instead of it being determined by whether or not a cable connects the SEED OUT and SEED IN sockets the connection is there all the time and you can switch between modes using the MS button of each part. Therefore you can use a mix of one-shot and multi-seeding in the same module.

A tour of the controls and I/O

The bulk of the interface consists of four virtually identical rows of sockets and controls. These are called the red, orange, green and blue parts.

The blue part

A part has (from left to right)…

A V/OCT socket that accepts a standard one volt per octave pitch input.

A GATE socket that accepts a regular gate/trigger input.

How the next few elements work together is explained later.

A VEL socket that accepts a 0 V to 5 V velocity signal. If nothing is connected a default velocity of 75% is used (as if 3.75 V was being fed to the socket).

An ENABLE input. If nothing is connected then the part is always enabled.

An ON button. When disengaged a part is muted (unless it’s soloed).

A FADE switch that gives a choice of instantaneous switching (the center OFF position) or fading using one of two different envelope settings.

A SOLO button that isolates the selected part and mutes the others. Clicking on an engaged SOLO button returns things to normal.

A VOLUME input socket that accepts a 0 V to 5 V signal. If nothing is connected this has no impact on volume.

A VOLUME knob.

A meter showing the combined effects of the last few elements.

An OCTAVE knob for octave shifting. The octave can be shifted up or down by up to three octaves. The default 12 o’clock position results in no shift.

An MS button. When enaged the part is in multi-seeding mode otherwise it’s in one-shot mode.

A P1 knob that sets the value of parameter 1 (or the rate of an LFO as discussed shortly).

A P2 knob that sets the value of parameter 2 (or the rate of an LFO as discussed shortly).

A P3 knob that sets the value of parameter 3 (or the rate of an LFO as discussed shortly).

A P4 knob that sets the value of parameter 4 if nothing is connected to the socket on its right.

A P4 input socket that if connected accepts a CV signal for parameter 4 that is used instead of the P4 knob setting.

At the top of the module we have…

Top of the module

A VEL IMPACT knob that allows adjustment of how much impact a velocity CV sent to a part’s VEL socket has on the amplitude of the part. With VEL IMPACT fully CCW the VEL CV has no impact and the result is the default 75% scaling (the same as when nothing is connected to the VEL socket). When set fully CW the VEL CV has full impact with 0 V resulting in silence and 5 V resulting in 100% scaling.

An ATTACK A knob that sets the attack time of the A envelopes.

A RELEASE A knob that sets the release time of the A envelopes.

An ATTACK B knob that sets the attack time of the B envelopes.

A RELEASE B knob that sets the release time of the B envelopes.

The attack and release knobs are all calibrated the same. The minimum period (fully CCW) is 3 ms, the maximum (full CW) is 25 seconds and the default (12 o’clock setting) is 276 ms.

A + button. When engaged this affects the seeding of any part that’s in multi-seeding mode by adding an extra seed on the rising edge of the part’s GATE input. Most of the time the effect of this button is very subtle but it becomes obvious in certain circumstances (especially when using low seeding rates).

A SEED IN socket. This must be connected to the Granular Synth’s SEED OUT socket in order for multi-seeding to work.

LFO RATES / VALUES switches for P1, P2 and P3. When a switch is down (VALUES mode) the column of knobs below set values between 0 V and 5 V for each part. These values are multiplexed and sent to the Granular Synth via the corresponding output socket at the bottom of the module.

When a switch is up (LFO RATES mode) the knobs below instead set the rates of four sinewave LFOs. The outputs of these LFOs are then multiplexed and sent to the Granular Synth via the corresponding output socket at the bottom of the module.

Parameter 2 is in LFO RATES mode

When in LFO mode LEDs appear to the top right of the knobs to give some indication of the LFOs state. In total there are 12 independent LFOs available. When GS Multiplex starts up it randomizes the phase of all 12 oscillators to prevent them all beginning in sync with each other – this is particularly important when the LFOs are running slowly as the idea is that they change grain parameters fluidly and organically and we don’t want to have to wait for the phases to diverge to get this effect.

At the bottom of the module we have…

Bottom of the module

An independent submodule that’s a simple logic sequencer dscussed later.

A handy latching button that simply sends 5 V to the socket below when engaged and 0 V when it’s disengaged.

A multiplexed V/OCT output socket that should be connected to the Granular Synth’s V/OCT.

A multiplexed AMP output socket that should be connected to the Granular Synth’s AMP external CV input.

A GATE output socket normally connected to the Granular Synth’s GATE. This output is not multiplexed instead its a conglomerated gate that is high if any part’s gate is high and that part’s amplitude is greater than zero.

A SEED output socket that should be connected to the Granular Synth’s SEED IN.

Multiplexed P1, P2, P3 and P4 output sockets. These can be connected to any of the Granular Synth’s grain parameters’ external CV inputs (except AMP which is already in use). The available grain parameters are therefore OFFSET, PITCH, SIZE, SHAPE, PAN and REVERB (FX).

To help you see what each parameter is assigned to in a patch you can customize the P1 through P4 labels by clicking on them and selecting a label from a pop-up menu.

Label menu

Remember that these are just labels and there’s no way for the module to know if you are using them accurately. You might prefer to leave the labels as P1 through P4. when experimenting with wiring and only use the custom labelling facility once a patch is fairly stable. Also sometimes the choice of labels provided might not be appropriate, for instance you might be using P1 to control both size and shape at the same time.

The multiplexed control of grain parameters is a very important aspect of using GS Multiplex and is discussed in depth later.

How amplitude control works

A sub-section of each part is dedicated to amplitude control – rather like a simple mixer channel but with the added benefit of an automated fade in/out feature. Long fade times of up to 25 seconds can be set and this can have considerable impact when used creatively.

Note the amplitude affects the amplitude of a grain when it is seeded and doesn’t have any impact on the grain after that, so changes are only noticeable when there are a series of grains being produced.

It’s all pretty straightforward in practice but there’s sufficient going on here that it’s worth discussing the detail for a full understanding of how amplitude control works…

The sockets and controls involved in a part’s amplitude control

The diagram below shows how the amplitude control might be structured if it was constructed from individual modules.

Amplitude control mechanism

The dotted lines in the diagram show default connections when a socket has no cable connected. So if nothing is connected to the ENABLE input socket and the ON button is engaged then 5 V is fed to the gate input of the part’s envelope generator.

The operation of the envelope generator depends on the setting of the FADE switch. In the image above the switch is set in its left position therefore the attack and release of the generator are controlled by the ATTACK A and RELEASE A knobs. So if the generator’s gate is high then its output will climb at the rate set by the ATTACK A knob and then stay at 5 V.

If the SOLO button is disengaged then the output of the envelope generator passes on to the first in a series of VCAs.

If nothing is connected to the VEL input socket then it defaults to a standard 75% velocity. No matter what the VEL IMPACT knob is set at, 75% is the base level so the velocity impact scaling just passes on the 75% value to the control the first VCA. So the output of this VCA is 75% (or 3.75 V in this scenario) .

If nothing is connected to the Volume CV input socket then it defaults to 5 V. So the second VCA passes on 100% of the signal (i.e. the 3.75 V).

Finally the VOLUME knob controls the final VCA’s gain. So if the knob was set at 100% then the final voltage would be 3.75 V (75% strength).

In the image the VOLUME knob has the its default setting of 75% so the final amplitude is 75% of 75% of 5 V which equates to about 2.8 V or 56%. The meter shows this 56% reading.

To get to 100% volume then the VEL socket needs to be fed a 5 V input, the VEL IMPACT knob needs to be fully CW and the VOLUME knob needs to be fully CW.

The above sounds way more complicated that it really is. In practice the 75% defaults are chosen simply to allow some latitude in adjusting levels up and down.

As the Granular Synth’s response to amplitude CV is linear the final voltage after multiplexing has an exponentially law applied. Paradoxically the main benefit of this is that it makes the automatic fading sound more “linear”.

How the seeding logic works

The seeding logic is just what determines whether a grain is created. The diagram below will make it easier to understand for some people but don’t worry if it doesn’t make any sense to you.

Seeding logic

If the amplitude from the amplitude control section is zero then no seeding occurs. This has the enormous advantage that CPU isn’t wasted on computing things that can’t be heard.

If the amplitude is greater than zero and a part’s MS button is disengaged (as shown in the diagram) then when the GATE input receives a gate signal it converts it into a trigger and a single grain is seeded. This is called one-shot mode as only one grain is produced for each gate signal.

One-shot mode

In the CV Watcher image shown above the yellow trace shows the GATE signal and the green trace shows the seed output from the seeding logic mechanism in one-shot mode. The red trace shows the SEED IN signal and you can see that it has no effect on the timing.

If the amplitude is greater than zero and a part’s MS button is engaged then when the GATE input is high a grain is seeded every time a trigger is received from the Granular Synth’s internal clock via the SEED IN socket. This is called multi-seeding mode as multiple grains might be produced for each gate signal (depending on how long the gate is high and how fast the clock is running). It’s also called gated seeding as seeding occurs while the gate is open.

There are two slightly different types of multi-seeding behaviour depending on whether the little + button at the top of the module is engaged.

The + Button

First let’s look at what happens when the + button is disengaged (as it is shown in the seeding logic diagram)…

Multi-seeding with + button disengaged

The green trace here shows that when the gate is high the SEED IN triggers are let through. This is often exactly what we want to happen. However, sometimes we want to preserve the rhythm of the gate – so that sound occurs as soon as the gate goes high. This is where the + button comes into play as it adds an extra seeding trigger when the gate rises, as shown below.

Multi-seeding with + button engaged

Neither approach is inherently better, it depends on the application. So you have the option to try out both and use which sounds best in a particular situation.

The logic sequencer

In the bottom left of the module we have a sub-module called the logic sequencer. It’s a very simple yet surprisingly useful little thing.

The logic sequencer

It’s independent of the rest of the module so you can use it for tasks that are completely unrelated to GS Multiplex if required.

The four outputs Q1 through Q4 are gate outputs. One and only one of these outputs is active at a time. When active an output is 5 V, when inactive it’s 0 V. An illuminated LED ring shows which output is active.

Clicking on the STEP button or sending a trigger to the socket above the button causes the currently active output to become inactive and the next one to become active. The whole thing cycles so the next step after Q4 is Q1.

Clicking on the RESET button or sending a trigger to the socket above the button forces Q1 to become active.

Here’s what the outputs look like when driven by a clock…

Red is clock, yellow is Q1, green is Q2 etc

So all dead simple stuff.

The only other thing of note is that you can shorten the cycle by connecting either Q4 or Q3 to the RESET input.

Reset by Q4
Reset by Q4

The magenta trace above shows the brief moment when Q4 goes high. This only lasts for one sample so for all intents and purposes one can consider the cycle to be Q1, Q2, Q3.

Reset by Q3
Reset by Q3

The cyan trace above shows the brief moment when Q3 goes high. This only lasts for one sample so for all intents and purposes one can consider the cycle to be Q1, Q2.

The main application for the logic sequencer in the context of GS Mulitplex is using it to switch parts on and off (or fade them in/out) by making connections between the Q outputs and ENABLE inputs.

When used with N-Step one could connect one of the EOC outputs from N-Step to the STEP input and set quite long attack/release times so that the “orchestration” morphs from one sound to another as a sequence repeats.

OK that’s all the technical reference stuff covered, I hope it’s not been too much of a slog. Let’s now look at putting the theory into practice with some demo patches…

Drums for beginners

This tutorial uses a GS Multiplex module and a Granular Synth along with a Misfit Audio Drum Trigger Sequencer (an 808 style drum machine that ships with Voltage Modular Core) to implement a simple drum kit consisting of a kick, snare, tom and closed hi-hat. This is just one way to use the GS Multiplex module and possibly not all that typical either but having such well-defined sounds to play with will help you quickly find your way around the controls.

GS Multiplex Drums Patch

You should hear a pretty conventional one bar drum loop playing. The Granular Synth is using ATQ to automatically slice up a sample that contains the drum sounds. But the focus here is on the GS Multiplex module.

When you read below something like “click on the red SOLO button” this is just shorthand for click on the SOLO button of the red part (the SOLO button isn’t red).

SOLO buttons

Click on the red SOLO button. This isolates the kick drum. Notice that the meters on the other parts have dropped to zero, this is because SOLO is implemented via amplitude control and the meters show the final amplitude value for each part.

Click on the orange SOLO button. This isolates the snare.

Click on the green SOLO button. This isolates the tom.

Click on the blue SOLO button. This isolates the hi-hat.

Click on the blue SOLO button again. This cancels the solo and all four parts become audible again.

ON buttons

Click on the red ON button to disengage it. This mutes the kick drum.

Click on the orange, green and blue ON buttons. When all of the ON buttons are disengaged there should be no sound.

Now click on the red SOLO button. Note that the kick drum comes back in. This is because the SOLO buttons override the ON buttons. The SOLO buttons also override the ENABLE state and bypass any fade that is in operation. If you look at the diagram in the amplitude control section above you’ll see that this is because a SOLO button switches to 5 V instead of the output of the part’s envelope generator when engaged.

Click on the red SOLO button again to cancel the solo and return to silence.

Then click on all the ON buttons so that they are all engaged and all four parts can be heard again. As you do this notice that the meters don’t immediately jump up. They rise at the rate set by the ATTACK A knob which defaults to about a quarter of a second which is just slow enough to be noticable.

VOLUME knobs

Adjust the volume knobs. As you do this the meter next to the knob will reflect the change but it’s temporarily obscured by the tooltip that pops up to tell you exactly what value you are setting.

If you turn a volume knob full CCW (and then release the mouse button to get rid of the tooltip) you’ll see that the meter will show zero volume and the part will be muted. Note that clicking on a part’s SOLO button will not restore the volume. If you check the amplitude control diagram again you’ll see that the volume knob is applied after the SOLO switch.

If you turn a volume knob fully CW you’ll see that the meter doesn’t go all the way to the top, instead stopping at 75%. This is because nothing is connected to the part’s VEL socket and the default velocity in this situation is 75%. If you feed a 5 V signal into the VEL socket and the VEL IMPACT knob is fully CW you’ll see the meter go all the way to the top.

Adjust the volume knobs to get a mix that you like (making sure that all four parts are still audible).

OCTAVE knobs

Try adjusting the OCTAVE knobs. This mangles the drums sounds in a slightly over the top way. You’d normally make finer tuning adjustments than this with drum sounds, but the OCTAVE knobs are really useful when working with regular pitched sounds.

The range is plus or minus three octaves. As the Granular Synth PITCH knob can shift plus or minus two octaves as well the combination of the two gives more than enough scope.

Parameter knobs

The column of knobs above the P1 output are controlling the OFFSET parameter of the Granular Synth. See the red cable from P1 to the Granular Synth’s OFFSET external CV input.

OFFSET settings

The OFFSET knob is set fully CCW and the attenuvertor fully CW. This means that a CV input of 0 V maps to the beginning of the buffer and 5 V maps to the end of the buffer.

Because the offset ATQ button is engaged offsets will be snapped to the nearest transient in the buffer. These transients are automatically set at the beginning of the individual drum sounds. ATQ doesn’t do a perfect job of this but it’s good enough for our purposes and it saves us having to manually set the offsets and size parameters.

If you adjust the red P1 knob you’ll see that you can change the sound from the kick drum to any of the other sounds. So it’s the P1 knobs that are mapping each part to a particular drum sound.

Notice that the red cable is just a regular cable carrying one signal. Yet in this case it’s managing to carry four signals – the offset of each part. This is the magic of time-division multiplexing. What’s happening is that if say all four drum parts are triggered at exactly the same time the P1 signal rapidly switches between the red, orange, green and blue knob settings in coordination with a rapid burst of triggers from the SEED output socket. Then at the other end the Granular Synth rapidly creates a grain for each of the four seed triggers, each with the correct offset value. This all happens so quickly that it appears to be both instantaneous and simultaneous.

OK so in this instance we want to leave the P1 knobs where they are, but the P3 knobs are being used to control panning via the cyan colored cable.

PAN settings

Again the parameter knob (this time PAN) and the attenuvertor knob are set fully CCW and fully CW. So 0 V is hard left and 5 V is hard right.

Tweak the P3 knobs to rearrange the stereo positions of the drums.

Finally P4 is being used to control the amount of reverb for each part. Again the parameter knob is set fully CCW and the attenuvertor fully CW.

As nothing is connected to any of the sockets to the right of the P4 knobs the knobs work just the same as the others. Try adding some reverb to the snare sound by turning up the orange P4 knob.

One problem you will eventually come across when repatching the P1 through P4 outputs to different destinations is that when you unplug a cable from one of the Granular Synth’s parameter’s external CV inputs you forget to reset the knobs.

This means that you’ll have accidentally programmed the parameter to have maximum modulation from whatever internal modulation source is assigned the + side of the attenuvertor. Often this will be the default uniform random source. So you end up with a mysterious “bug” where the Granular Synth produces strange sounds or even silence. So be aware of this! The fix is simply to set the attenuvertor to its default 12 o’clock position when you unplug the cable. If you have the Voltage Modular settings set the way I do then just double clicking on a knob resets it to its default position.

So far we’ve left the P1 through P4 labels as their defaults in order to emphasize that they are generic parameters. But you have the option of labelling them with something more appropriate.

Click on the P1 label and select OFFSET from the menu. Then instead of the output being labellled P1 it’s labelled OFFSET.

Label menu

Use the same method to change the labels P3 and P4 to PAN and REVERB then save the patch.

The changed labels

Cinematic Strings

This patch shows how to use GS Multiplex to produced layered monophonic sounds with the Granular Synth. It produces a rather dramatic string sound. A couple of the stock Voltage Modular Delay modules are used to add an extra sense of space.

Cine Strings patch

The patch is controlled via input from your DAW, a MIDI keyboard or Voltage Modular’s little virtual keyboard but can easily be adapted to be driven by a sequencer. Although as this is a lush pad sound all it really needs is some very slow one finger keyboard work.

Cables for pitch, gate and velocity connect Voltage Modular’s I/O Panel sockets to the V/OCT, GATE and VEL inputs sockets of the red part…

Control connections

The vertical white lines on the GS Multiplex’s panel serve as a reminder that signals plugged into the sockets are normalled to the ones below. So in this instance all four parts receive the same pitch, gate and velocity signals. This saves us having to laboriously patch 9 extra cables.

Although this is helpful most of the time it’s worth noting that you might sometimes accidentally layer sounds via these normalled connections. If a part’s ON button is engaged then it will play along even if that wasn’t your intention.

The ON button for the blue channel is disengaged at the moment so only the upper three parts are currently audible. We’ll fade in the blue part later.

The OCTAVE knobs are set so that the red and orange parts use the input pitch, the green part is one octave lower and the blue part is one octave higher.

All of the MS buttons are engaged so all parts produce a stream of grains while a key is held down.

The next thing to look at is what’s happening with the parameter 1 column of knobs (those above the P1 output socket). The label P1 has been changed to PITCH using the left-click mechanism discussed earlier in order to make things a little clearer.

One source of confusion is that the Granular Synth has both V/OCT and external PITCH CV inputs. The V/OCT CV is always fixed at one volt per octave but the PITCH input goes through an attenuvertor. In this patch the V/OCT signal tells the Granular Synth what the base pitch is and the PITCH signal is used to offset individual parts in order to form a chord.

Note that one doesn’t have to tune parts to different intervals in order to create layered sounds. We are just doing it here as it’s a useful trick to master and it’s not entirely obvious how to implement it.

The cable from the P1/PITCH multiplexed output feeds the external CV input of the PITCH parameter of the Granular Synth so the knobs set pitch offsets for each part. They are set up to form a minor chord.

This isn’t entirely straightforward to do as the knob range is 0 V to 5 V which is great for general purpose parameters but too wide a range for easily selecting pitches with semitone precision. So how is the chord tuning done?

An alternative and slightly easier method for setting pitch offsets using the LSSP Chromatic Scale module is described later.

The PITCH parameter knob on the Granular Synth has a -24 to +24 semitones range so if were to set it fully CCW and the attenuvertor fully CW we’d get a four octave range when we vary the CV from 0 V to 5 V which would make setting the parameter knobs on the GS Multiplex module very tricky.

However if we set the attenuvertor to 25% this reduces the range to just one octave. We can do this with precision by right-clicking on the attenuvertor, selecting Edit Value and typing in 25.

Attenuvertor set to 25%

Now if we leave the PITCH knob at the default 12 o’clock position then a 0 V CV input causes no pitch shift and a 5 V CV input causes a shift up of just one octave. This means that the parameter 1 knobs on the GS Multiplex module give no pitch offset when set fully CCW and raise the pitch by one octave when set fully CW.

To get a minor chord we need a root note which means zero offset, a minor third which means an offset of four semitones, and a perfect fifth which means an offset of seven semitones.

The red part is used for the root note so its knob is set fully CCW.

The orange part is used for the minor third so its knob needs to be set four semitones up from fully CCW. As the Granular Synth’s OMDU switch is set to M it snaps pitches to the nearest semitone so it’s reasonably easy to do this by ear.

The orange SOLO button was clicked and then the orange P1 knob was set fully CCW. Then slowly turning it up I counted each semitone jump until it reached four.

Then the same process was used with the green P1 knob although this time counting to seven.

This is a bit fiddly but not that difficult once you’ve done it a couple of times. If you have a good musical ear you could just tune up a chord without any counting. Another method is to do it numerically looking at the tooltips. 1.2 produces a minor third, 2.84 produces a perfect fifth. If you want a major chord use a major third instead by setting the orange P1 knob to 1.6. These numbers are just what I happened to end up with but because of the quantization there’s a range of settings that will all produce the right pitch.

The image below shows the final P1/PITCH knob settings.

P1/PITCH knobs set for minor chord

As mentioned earlier If you have the LSSP Chromatic Scale module (or some other convenient source of note voltages) then you could achieve the same result by using the P4 input sockets instead.

Using the P4 inputs for V/OCT offsets

In the image above the red cables are supplying the required 1 V per octave offsets via the P4 inputs. (A P4 knob is ignored if its socket is used). Then the P4 output is patched to the Granular Synth’s V/OCT input alongside the existing V/OCT cable. This adds the voltages together.

Now PITCH modulation on the Granular Synth by P1 is no longer required.

No pitch modulation

Returning to the earlier patch, Parameter 2 is used for modulating OFFSET and Parameter 3 for modulating PAN. Both use LFOs for this so the switches above the knobs are set in the up position so that the knobs control the rates of 8 sinewave LFOs instead of setting direct values.

P2 and P3 modulating OFFSET and PAN

The offset modulation by P2 makes the start position of grains in the buffer vary a little in order to add a bit of variety between parts. In this instance we don’t want the offset to vary across the entire buffer so the Granular Synth’s OFFSET and attenuvertor knobs are set to limit the modulation to a smaller region.

Knobs set for limited modulation range

The pan modulation pans the notes in the chord across the entire stereo field. But as each part has its own LFO, the modulation is done at low speeds and the reverb smears the imaging it sounds a bit more subtle than a typical auto-pan. Also although the pan CV changes continuously once a grain is seeded it doesn’t move.

As mentioned earlier the blue part starts off muted. This is because we’ve been saving it to demonstrate an automatic fade. While holding a note down click on the blue ON button. A fourth high note will fade in over a four second period, adding more drama. Disengage the ON button and the extra note will fade away.

The FADE switch for the blue part is in the right-hand position so the ATTACK B and RELEASE B knobs control the speed of the fading.

To demonstrate how the logic sequencer can be utilized, we’ll make a simple modification to the patch to make the extra high-pitched note fade in and out now and then.

To achieve this the blue ON button is left permanently engaged, the logic sequencer is clocked by the keyboard gate signal and the Q4 output is patched to the blue part’s ENABLE socket. Now every fourth note played brings the high note into the mix for a while. Providing that the pad is being played relatively slowly this works quite nicely – adding a mysterious element to the sound every now and then.

Extra wiring for spooky fade (red cables)


GS Multiplex is part of the Granular Synth bundle, available for download at the Cherry Audio store.