Note Randomization

Have you ever tried to write a song picking notes at random? Of course, most of the time, it's gonna sound like crap as is but if you keep at it and add some delay when all is said and almost done, it may end up sounding decent and worth saving (for youtube at least).

Practically, one would do the random thing on one pattern and then copy/paste/modify that pattern in sequencer mode to build up a full song. Trying to build up a song just on the KAOSS pad and tapping at random for all the patterns might be a little too much though. As with everything, it should be done in moderation (I think).

It's quite easy to experiment with note randomization on the KORG DS-10 thanks to the KAOSS pad. Since the pattern is looping when you record, it's quite a no-brainer to figure out that after a while of doing random taps, you might just end up with something that works. All the notes you randomly pick are automatically in key (that's the beauty of the KAOSS pad). This explains why, no matter what the notes end up being, there ain't gonna be any dissonance in the final product.

It's too bad the Korg DS-10 doesn't have a randomizer like this piano roll randomize tool from fl studio. I think one could waste a lot of time playing with it.



Writing a song pattern on the DS-10 using a whole bunch of random notes ... and a good dose of delay.

In the video, I am playing mostly with the pitch as I couldn't care less about the gate. Recall that KAOSS pad 1 has the pitch on the X axis and the gate on the Y axis. If you are really ambitious, you may to want to play with KAOSS pad 2 which controls the cutoff and the peak and even KAOSS pad 3 which is pretty much a free for all.

In KAOSS pad mode, click on the SET button and you have the ability to choose the key (C is the default) and the scale (Ionian is the default). When you play with the stylus on the pad and then go into the sequencer mode (to check things out), you will see that the notes are all of that scale and key.

Now, if somebody could explain the difference between the SMT (smooth) and HLD (hold) settings in the KAOSS pad, I'd be real grateful.

Making a song

Songwriting is partially based upon repetition. You have the basic construct, verse leading to the chorus, repeated throughout the song with maybe a bridge along the way. A song will not "catch" if there's no melodic hook to hang on to. So, it makes sense to start the song creation process by creating one single pattern containing the melody, in our case, just a motive since you only have 16 steps maximum to work with. Then, it is customary to create similar patterns that are variations of the main pattern, in other words, develop the initial motive. There is a fine line between creating a song that's repetitive enough so that the listener can make sense of it and one that's boring because too repetitive. So, you may need to have several motives (melodic phrases) to be able to build a full song. At some point, you may venture into creating melodic phrases that span several patterns and creating songs with unrelated melodic phrases connected by bridges.

For starters, try to limit yourself to diatonic tones, that is, tones that are part of a given scale. For example, you could limit yourself to the C major scale, which are the seven white keys on the keyboard starting at, say, middle C (you probably already know that there are natural half-steps between E-F and B-C and therefore no black keys between those notes). Considering solely the diatonic tones makes song writing less intimidating and will most likely make your tracks more pleasing to the ears of others.

In the next posts, we are gonna look into music theory in order to have a better understanding of tones and how they relate to each other.


In this video, we create a short song using 5 patterns. The first pattern contains the main motive (SYNTH1), repeated in patterns 2 and 3. Patterns 4 and 5 are slight variations of pattern 1 to break the monotony of the melody.

Delay, Flanger and Chorus

After the signal is mixed, effects (FX) can be applied, if desired. Delay, flanger and chorus are available for all input voices, that is, SYNTH1, SYNTH2, SYNTH1+SYNTH2, DRUMS and ALL. Effects can of course be turned OFF.

An effect processor takes an input signal and transforms it into a processed signal. Typically, what you get at the other end (of the effect processor) is a mix of the original unaltered (DRY) signal and the processed modified (WET) signal. When looking at an effect processor control panel, you will always see a DRY/WET knob. If turned all the way to the left (0%), only the input signal comes out. If turned all the way to the right (100%), only the processed signal comes out. Anything in between is a mix of the two signals.

Delay

A delay line, as its name applies, takes an input signal and generates a processed signal that's delayed by a certain number of milliseconds, the delay time. A delay effect is not the same as reverberation. Indeed, reverb produces a kind of blurry echo that gets washed out, while a delay line produces a clear echo, that's a priori as clear as the input signal.


Here's what you see when you go into the FX control panel and switch on the DELAY effect.

The SYNC slider switch enables you (when on) to sync the delay time with the beat-per-minute main clock. It means that the delay time will correspond to 1/16 notes, 1/8 notes, 1/4 notes, etc.

The L/R RATIO knob enables you to adjust the left and right delay times.

The FEEDBACK knob controls the amount of feedback. When the processed signal is "fed back" into the input of the effect processor, it is called feedback. With feedback, sounds get repeated forever but the repeating sounds get quieter and quieter (gain less than 1).

Flanger and Chorus

Those two effects are actually similar. They are both delay lines with a modulated delay time. Modulated by what you may ask. By a Low Frequency Oscillator (LFO), of course. When the delay time is modulated, interesting things happen, but let's delve a bit into the effects of adding a delayed signal, for now.

When a delayed signal is superimposed onto a source signal, the frequencies (overtones) are kinda filtered by the effect processor, in other words, the overtones you hear depend on the delay time. A flanger or chorus filters the frequencies in the manner of a "comb filter": some frequencies (peaks) go through amplified while others are attenuated and even canceled (notches). When the delay time is large, the pitch tends to go down. When the delay time is small, the pitch tends to go up. So, when you modulate the delay time, you create pitch modulation. The "depth" on a flanger or chorus processor controls how pronounced the effect is, in essence, the depth of the notches (see the DEPTH knob on both flanger and chorus control panels).

What's the difference between flanger and chorus then?

The flanger has a very small delay time range and produces a "whooshing" sound. The chorus doesn't make the sound sweep of a flanger and produces a "thicker" or richer sound, result of very slightly detuned and out-of-sync voices.


This is the control panel for the FLANGER effect.

The LFO FREQ knob is for adjusting the frequency of the Low Frequency Oscillator (LFO).


This is the control panel for the CHORUS effect.

The chorus effect has the same controls as the flanger, without the feedback (used on a flanger but not on a chorus).

Modulation using Envelopes

Following again the post about patching, here is a little video about modulating sound oscillators with the Envelope Generator (EG).


In the above video, I have set the pattern to just 2 steps, one being filled with a sound, the other empty, just to make it easier to hear the effect. Initially, the envelope is just pure sustain, it's changed to something more refined with some attack and release. This envelope is then used to modulate the pitch of both oscillators (since oscillator 2 is off the mix, we are actually just acting on oscillator 1). As you can hear, the pitch of the sound follows the envelope: it goes up during the attack and goes down during the release.

Note that the envelope is always modulating the amplitude or volume: it's an internal or hard-wired modulation. Indeed, the volume goes up in the attack phase and down in the release. So you might have been modulating with the envelope all along without really knowing.

Modulation using Oscillators

Following the post about patching, here is a little video about modulating oscillator 1 with the Low Frequency Oscillator (LFO) and then with oscillator 2 (VCO2).


We put all 16 steps of oscillator 1 at the same pitch in order to better hear the effects of modulating the pitch. The LFO "envelops" the oscillator with a low frequency waveform resulting in the pitch going up and down as the notes (that are all at the same pitch to begin with) are being played in the sequence. The speed at which the pitch goes up and down is regulated by the LFO frequency knob. When switching to modulation by the 2nd oscillator (VCO2), the frequency of the modulation is adjusted in the SYNTH EDIT control panel using the VCO2 PITCH knob. As you can hear, the frequency of the modulation is much higher than with the LFO.

Patching

Patching here refers to external patching where you use patch cords (wires with male jacks at each end) to connect sources to destinations. Internal patching refers to hard-wired routing.


Above is what you see when you go into SYNTH PATCH (SYNTH1 and SYNTH2) mode. Focus is given on signal routing with "virtual" patch cords that you plug into the female jacks.

What patching does is to give you control over how the audio signal (from the oscillators VCO1 and VCO2) is modulated. The arrows indicate the signal flow.

You have the destination signals all arranged in a row right below the knobs. The knob for each destination signal controls the amount of modulation. It is bi-directional, meaning that the center position turns the effect off, turning the knob to the right increases the effect of the modulation signal and turning the knob to the left increases the effect of the inverted modulation signal. Well, that's until you use VCO2 to modulate in which case the knob you plug the signal into becomes uni-directional, meaning that it has to be turned all the way to the left to turn the effect off. Turning it to the right increases the effect of the modulation signal. Yeah, it's a bit confusing.

Plugging a control signal into the PITCH IN jack modulates the pitch for both Voltage Control Oscillators (VCOs). To control just one VCO, plug the cord into VCO1 PITCH or VCO2 PITCH.

Plugging a control signal into CUTOFF IN modulates the frequency cutoff of the Voltage Controlled Filter (VCF).

Plugging a control signal into VCA IN modulates the control voltage of the Voltage Controlled Amplifier (VCA), in other words, the output volume.

The modulation signals are all aligned at the bottom, gathered in the Modulation Generator (MG) sub-panel: Low-Frequency Oscillator (LFO) with its 4 types of waves, Envelope Generator (EG) and VCO2. As you can see, the 2nd Voltage Controlled Oscillator has dual functionality: it can be used to produce sound and to provide a modulation control signal. Note that the real KORG MS-10 does not have a 2nd oscillator, there's just one.

Quite simply, you may use a patch cord from any jack at the bottom (modulation signals) to any jack just above (destination signals).

Low-Frequency Oscillator

Synthesizers rely on Low-Frequency Oscillators (LFOs) to modulate signals, like pitch, amplitude or cutoff (in filters). Waveforms available for LFOs are generally sine, triangle, sawtooth, pulse (rectangle), noise (random) and sample-and-hold waves.

Modulating the pitch of a sound with a sine wave LFO produces a sound that goes up and down cyclically (following the ups and downs of the sine wave) from the base pitch (of that sound without modulation). This modulation of the pitch is referred to as VIBRATO. If you modulate the amplitude (loudness) or the filter cutoff of the sound instead of the pitch, the effect is often called TREMOLO. When the LFO is a square or rectangle wave modulating the pitch, you obtain an effect called TRILL (goes back and fourth between 2 pitch values).

What is a sample-and-hold (S-H) wave? It is a stepped signal coming from a source signal (usually some random noise coming from an internal noise generator) sampled by a clock source. When the clock sends its trigger to the LFO module, it samples its source signal and holds the sampled value until the next trigger. In other words, the source signal is sampled at regular intervals forming a stepped signal, each step corresponding to the value of a sample.

It is called Low-Frequency because it operates usually under the 20 Hz threshold (of things that you can't really hear by themselves).


Above is what you see when you go into SYNTH PATCH for synth1 or synth2. In this post, we focus on the MG (Modulation Generator) module.

As you can see, KORG calls an LFO a Modulation Generator (MG), which is not a bad name since it's exactly that. The supported waveforms are: triangle, sawtooth, pulse (square) and sample-and-hold. The output wave of the Modulation Generator can serve as input to whatever module accepts modulation. The frequency knob controls the frequency of the signal, in other words, its speed. The BPM (Beat-Per-Minute) slider switch enables you to force the LFO's frequency to be in sync with the main beat (per minute).


The two jacks on the right of the MG panel (as seen above) are for modulation with the Envelope Generator (EG) and the second oscillator (VCO2). Those two have nothing to do with the Low-Frequency Oscillator (LFO), they are just other sources of modulation.

So, to be perfectly precise, KORG's MG (Modulation Generator) is not exactly equivalent to an LFO (Low-Frequency Oscillator) as it includes the LFO and other sources of modulation (EG and VCO2 on the Nintendo DS version of the KORG MS-10).

Envelope Generator (EG)

The envelope is what shapes a sound. It gives the sound a beginning, a middle and an end. The most common of envelopes is the ADSR which stands for Attack, Decay, Sustain and Release. The envelope begins with the attack phase during which the level goes from 0 to the peak (maximum level for the sound). Then comes the decay which is a drop from the peak to the sustain level. During sustain, the level of the envelope remains flat until release which is a drop from sustain level to 0.

The idea of an envelope to shape sounds comes from Helmholtz, a great mind famous for his equations. The ADSR, as we know it today, comes from Moog, a pioneer in synthesizer design. The envelope Generator is probably the most important module in a synthesizer as it enables to make sounds feel like they are coming from real instruments.

When the keyboard detects that a key is being pressed, it turns on the GATE signal, which tells the synth to start the envelope (attack) for that key (and associated sound). When the key is depressed, the keyboard turns off the GATE signal, telling the synth to start the release part of the envelope.


Above is what you see when you go into SYNTH EDIT for synth1 or synth2. In this post, we focus on the EG (Envelope Generator) controls.

The ATTACK knob controls the time it takes to go from 0 to peak level. If set all the way to the left, there is no attack and the sound goes straight to the decay phase.

The DECAY knob controls the time it takes to go from peak level to sustain level (always lower than or equal to the peak). If turned all the way to the left, there is no decay and the sound goes straight to the sustain phase.

The SUSTAIN knob controls the level of the sustain phase. If all the way to the left, the sustain level is 0 and therefore silent. If all the way to the right, the sustain is at 100%, which means its level is at the peak. Notice that this knob does not control time (as the other knobs) but level.

The RELEASE knob controls the time it takes for the sound to go from the sustain level to 0.

The signal from the Envelope Generator can be used to modulate modules that accept input control voltages (in a voltage controlled analog synth), for instance, the volume (obviously otherwise we would not be able to hear the sound envelope), the pitch, the filters, etc.

Pitch

The pitch of a sound is directly related to its fundamental frequency. The higher the frequency, the higher the sound and vice-versa.


Above is what you see when you go into SYNTH EDIT for synth1 or synth2. In this post, we focus on the PITCH controls.

The OCTAVE knob enables you to shift the pitch one or two octaves up and down.

The PORTA knob controls the portamento effect which is a pitch bend from one note to the next (or one pitch value to the next). The more you turn the knob, the more time it takes the oscillators to go from note to note and therefore the smoother the transition. When turning the knob to the right, you introduce a lag which limits the speed at which oscillator signals are allowed to change and smooths out pitch changes. Depending on the instrument, it may sometimes be referred as glissando. On a synthesizer, it is usually called glide.

EG INT stands for Envelope Generator (internal as opposed to external, in other words, hard-wired). The envelope's signal can be used to modulate the pitch. If you turn the knob to the right, the pitch rises and falls according to the envelope's signal. If you turn the knob to the left, the envelope's signal is inverted, which makes the pitch falls when the signal goes up (attack) and rises when the signal goes down (decay and release). Modulation of a signal (here, the pitch) by another (here, the envelope) is a fundamental concept in analog sound synthesis.

So far, we have seen that the envelope can modulate VOLUME, FILTER and PITCH. There are other parameters that can be modulated by the envelope but those three are the most commonly found on synthesizers.

Voltage Controlled Amplifier (VCA)

The Voltage Controlled Amplifier (VCA) lets you modulate the volume of a sound (one oscillator wave or the combination of two oscillator waves). How? Simply by feeding the VCA the signal (voltage) coming from the Envelope Generator (EG) or, in general, whatever signal carries a control voltage.

If you look at the Envelope Generator, the voltage goes up (attack) as a key is being pressed on the synthesizer keyboard or the sound starts, goes down a little (decay), stays level (sustain) and goes down again (release) as the key is being released or the sound ends. So it's quite natural to have the envelope modulate the amplifier (VCA).


Above is what you see when you go into SYNTH EDIT for synth1 or synth2. In this post, we focus on the VCA (Voltage Controlled Amplifier).

The slider switch enables you to choose the modulation source signal: Envelope Generator (EG) or GATE. As you probably know, the gate lets the sound pass through, between the pressing of a key and its release. Voltage is 0 when the gate is closed and at the maximum when the gate is open. So clearly, it can be used to modulate the Voltage Controlled Amplifier (VCA) although it kinda defeats the point of having an envelope generator at your disposal. Having the envelope modulate the VCA makes a lot more sense.

The LEVEL knob adjusts the level of amplification to be given to the signal. It's more or less a volume knob if you will.

The DRIVE knob on a synthesizer is usually there to boost a low signal prior to distorting (overdriving) it. On the Nintendo KORG DS-10, it's not very clear what it does especially since the actual KORG MS-10 doesn't have one. Best guess is that it is simply an OVERDRIVE knob to produce distortion.

Filters

In a synthesizer, a filter as its name implies removes (more exactly, attenuates) some frequencies from a signal, for instance, an oscillator's signal. On the Nintendo KORG DS-10 (and the KORG MS-10 for that matter), it is referred to as a VCF (Voltage Controlled Filter) because it is an analog synthesizer (as opposed to digital).

A low-pass filter (LPF) lets the low frequencies go through and removes (attenuates) the high frequencies.

A high-pass filter lets the high frequencies go through and removes the low frequencies.

A band-pass filter lets the frequencies within the band go through and removes both the low and high frequencies.

As you probably know, a sound is a combination of sine waves of various frequencies and amplitudes. For a given sine wave, amplitude relates to how high and low the sine wave goes. When the sine wave frequencies are all whole-number multiples of the lowest frequency sine wave, the waves are called harmonics and the lowest frequency wave is known as the fundamental. The harmonics above the fundamental are commonly called overtones. Filters basically remove some of the harmonics of a sound.


Above is what you see when you go into SYNTH EDIT for synth1 or synth2. In this post, we focus on the VCF (Voltage Controlled Filter).

The CUTOFF is the threshold frequency for the various types of filters that are available (see the slider switch at the bottom): LPF (lowpass), HPF (highpass) and BPF (bandpass). In the case of the bandpass filter, the cutoff frequency is a misnomer as it should really be called the center frequency. Note that the cutoff frequency can be modulated with an LFO (Low Frequency Oscillator) or in Korg's jargon, a MG (Modulation Generator) for interesting effects.

The PEAK refers to a phenomenon called resonance. When resonance is applied to a filter, there occurs a signal gain (as opposed to a cut) near the cutoff frequency. The peak directly relates to the amount of resonance applied, in other words, the gain.

EG INT refers to filter modulation by the internal envelope generator (the controls to the right). An envelope generator shapes a note played on the keyboard into attack (voltage rises), decay (voltage drops), sustain (voltage stays put) and release (voltage drops) phases, commonly referred to as ADSR. The envelope generator output (voltage) modulates the cutoff frequency as a note is being played, from the time the key is pressed (attack) to the time it is depressed (release). As the voltage rises in the EG signal, the cutoff frequency rises to the maximum level set by the CUTOFF knob and as the voltage drops, the cutoff frequency drops. This is one way of modulating the filter.

On most analog synths, there are two envelope generators: one for the filter and one for the amplifier. On the Korg DS-10, there is only one for the amplifier but you have the capability to use that single envelope signal to modulate the cutoff frequency. So it's not too bad.

Oscillators

An oscillator is the basic source of sound in synthesizers. It is the signal of an oscillator that you shape (for example, with modulation) to obtain the sound you desire. The Nintendo KORG DS-10 has two oscillators, named VCO or Voltage Controlled Oscillators, which you can superimpose for more sound richness.


Above is what you see when you go into SYNTH EDIT for synth1 or synth2. In this post, we focus on the VCO controls.

For both oscillators, you can choose the type of wave signal: triangle, sawtooth, square (aka pulse), noise. The default is sawtooth which is kind of a whole-purpose base signal. The balance knob enables you to mix the two oscillators. The VCO 2 pitch knob lets you control the pitch of the second oscillator relative to the first one. You usually want the second oscillator to be at a higher pitch if you intend to sync them.

Oscillator sync (VCO SYNC) is an important tool since it will sync (synchronize) the second oscillator (usually called the slave oscillator) to the first oscillator (the master oscillator). If the second oscillator is modulated and its pitch changes (for example, with a low frequency filter or LFO, called a modulation generator or MG on KORG synths), oscillator sync forces the second oscillator to start over every time the first oscillator starts a new cycle, locking its base frequency (to be always the same as the master's). Syncing also prevents a phenomenon called beating, which usually happens when superimposing two oscillators operating at different frequencies. It's most noticeable when the second oscillator is at a slightly higher pitch than the first one (the difference between the two frequencies gives you the frequency of the beat). Practical note: when you use oscillator sync, you may want to mix down the master to actually hear the oscillator sync effect.

KAOSS pad

The KAOSS (kaossilator, in full) pad is a neat little tool to compose music or "jam" on an already composed piece of music.

There are actually 3 KAOSS pads at your disposal per channel, each with a X and Y axis. The idea is that by putting the stylus on the pad, you are actually inputting an X and Y coordinates to the software which in turn determines the corresponding values for whatever effects were assigned to the X and Y axis. In the first 2 KAOSS pads, you have the ability to control NOTE and GATE for pad 1 and PAN and VOLUME for pad 2. These have corresponding grids in sequencer (SEQ) mode.

The 3rd pad is quite free-form as you can select (from a list) a bunch of effects on the X and Y axis. When you record, whatever is stored will also appear in the KAOSS X and KAOSS Y grids of the corresponding sequencer (SEQ).

What's great about these KAOSS pads is that you can play around with the stylus on the pad, record something and then go in the sequencer to see what you have done and edit it !

With the KAOSS pad, you can pretty much do all the music creation as mentioned above but you can also "jam" with it. If you don't record, you can apply all kinds of effects to your music as it is being played using the stylus on the pad. It is like free-styling your own tunes.


In the video above, we make use of the KAOSS pad (number 3) to control the CUTOFF (KAOSS X) and the PEAK (KAOSS Y) of an existing random melody. Once we are happy with what we hear, we start recording. Then, we go into SEQ mode and see that the recording shows up in the KAOSS X and KAOSS Y grids of the sequencer. Neat!


In this second video, we change what the KAOSS X and Y axis control on pad 3. We choose (for whatever reason) OCTAVE for X and DRIVE for Y, record something and then go back to the SEQ to see that the KAOSS X and Y screens contain what we have just doodled on the KAOSS pad.

What's in a gate ?

When you cycle through the control screens in SYN SEQ mode, one of them is called GATE. What is it and what can be done with it?

The gate in a synthesizer controls the duration of the note you are playing. In the Nintendo KORG DS-10 title, the gate enables to shorten the duration of the sound being played at each beat of the loop (pattern).

The term LEGATO means that there is no gap between the notes being played, that is, the keyboard virtual keys are never actually depressed, it's one long continuous note modulating. The lower the gate (in terms of percentage), the shorter the duration of the sound. Playing around with the gate obviously enables to put rhythm into loops.


The (very short) video above shows the effects of the gate settings on the notes being played.

Introduction to the KORG DS-10

The KORG DS-10 is a synthesizer/sequencer for the very popular Nintendo DS portable gaming system. It is meant to be a simulation of the old KORG MS-10, a quite popular analog synthesizer.

What you can do with this little marvel is create beautiful music. To make a song, you have to create patterns first, from just 1 to 16. A pattern on the KORG DS-10 may have two keyboard/synthesizer tracks and four drum tracks. Each track is made up of steps, typically, a note. So, if you think about it, each pattern can have up to six tracks or channels. In other words, at any moment the DS can play up to six channels/notes at the same time.


This above is the main control board for the KORG DS-10. At the top, it says that we are currently in pattern 1 which is set to have 16 steps (default value). Remember that each step corresponds to a note (or a silence of course). For each pattern, you have a bunch of tracks to work with : 2 synthesizers (SYN1 and SYN2) and 4 drums.

When you work on a synthesizer track, there are many different ways to actually enter notes or sounds : with the keyboard (KBD), with the KAOSS pad (we'll get into that in a later post) or directly in the sequencer (SEQ). The drum tracks are edited just like in the synthesizer sequencer but it's more streamlined.

The notes that you enter in the sequencer can be modified in many ways with the GATE, VOLUME, PAN, KAOSS X and the KAOSS Y controls. Even the plainest of melodies may sound incredible with KAOSS.

Once you are satisfied with the pattern tracks, it is time to mix the 2 synthesizer and 4 drum tracks to your liking. After that, special effects (DELAY, FLANGER and CHORUS) can be applied to the tracks separately or as a whole.

The last step in making music is the actual song creation. It is nothing more than a collection of patterns. Just like with any sequencer, you may place any pattern anywhere along the song timeline. Couldn't be easier.

Needless to say, this is an amazing piece of music software.