Inspired by the classic DX7 style FM synthesis, the reface DX offers some different wrinkles. Stereo, dual Insertion Effect processing, capacitive touch data entry, and more. The original DX FM required, at minimum, a short stack (2 Operators) to generate a complex waveform. In reface DX each Operator can be made to generate a complex waveform. There are four Operators, each is initially a sine wave generator, but each can be fedback on itself to create more complex waveforms from a single Operator.

Frequency Modulation is basically using the output level of one waveform to modulate or change the output of another. This is accomplished by feeding the output of one oscillator into another. The Modulator is not heard directly, but you do hear its influence on the result from the other oscillator which is called the Carrier. FM, yes like the radio signal FM… Those radio signals are transmitted at frequencies far above what we can hear, DX-style FM takes place with frequencies well within our range of hearing. But it is the same FM.

As musicians, we can better understand this Modulator:Carrier relationship by thinking musically. Imagine a violinist bowing a single string with their right (bow) hand and adding vibrato (pitch modulation) with their left hand. The bow puts the string in motion at a pitch of A440 (it carries the audio signal), while the left hand modulates the pitch by shortening and lengthening the vibrating string several times a second. The bowed string is the Carrier, the vibrato is the Modulator.

The frequency of the vibrato might be 4 cycles per second. Below our audible perception as a steady pitch… But you do hear the influence of this variation on the resulting sound. Dr. John Chowning, the Father of FM, discovered as you sped up the vibrato (modulator) you could effect a profound timbre change in the oscillator acting as the Carrier. Doing this electronically, he was able to speed the vibrato up to any frequency. As the Modulator frequency crossed into the audible frequency range, this caused a change in the overall tone of the resulting sound. And as the output level (called “modulator index”) increased so did the harmonic complexity of the results. Sidebands that included harmonic and even non-linear harmonic levels were generated and the complexity of timbre opened the door to a virtually unlimited range of tonalities.

Why this is musically important is sounds are made up of complex harmonic relationship, FM is a synthesis method that lends itself well to recreating both musical and non-musical timbres in an extremely wide variety of categories. Sound design with this FM engine can be mastered via this hands-on interface.

The Operator (oscillator) includes the sine wave generator, an Amplifier, an Amplitude Envelope Generator, and inputs from other Operators and the keyboard. This is different from the typical analog Oscillator – because it includes the EG as apart of this single block – each tone source has its own Amplitude Envelope. In your typical analog synth, the Oscillators share the same AEG.
The Amplitude Envelope Generator is responsible for changes in output level of this Operator over time. You can control how it comes in, what it does while it is in, and how it disappears when you let go of the key. In the analog world this was referred to as ADSR or Attack-Decay-Sustain-Release – four parameters that defined the “shape” of Loudness in the case of the AEG. On the reface DX, you have 8 parameters to define this shape. The parameters are four LEVELs and four RATEs.

Basically you control how fast (Rate) the envelope travels from one Level to the next. In future articles we will show you how to take command of these Rates and Levels to create the results you require. Rates have values between 0 and 127, where the higher the number the faster the travel… think of the rates as miles per hour… higher is faster. This is not a strict TIME parameter, because if the distance it has to travel is longer the same value for RATE may not result in the same amount of elapsed time. So Rates and Levels work in combination. A Rate of 100 does not denote a specific value of time, if the distance it needs to travel is greater, it simply will take longer to travel that distance. Think real world again: Even if you are moving at 100mph, if you have to travel a greater distance it does take more time.

Follow the Synth Basics articles, here on YamahaSynth.com for tips and tricks on programming your reface DX. If you have questions or comments on this article, please go to this forum thread to disucss.

According to Dr. Chowning, the Father of FM, one of the major innovations that Yamaha engineers brought to the table was the Feedback loop. Basically, routing the output back to the input to generate more complex tones. While this is similar to having additional modulators, it is not exactly the same thing – but will get you similar sonic results. In the reface DX each of the four Operators can be fedback on itself; this means a single operator can be a complex wave source. In the original DX7, for example, you had some ability to use Feedback. When all Operators are simply sine wave generators, if you wished to modulate a Carrier with a Sawtooth or Square Wave you would need to create a two Operator Modulator “stack” to generate that modulation wave to feed into the Carrier. Now any single Operator can generate the Sawtooth Wave or a Square Wave.

What this will translate into is you can build more complex waveshapes with fewer Operators; you can use your Operators to add fine details to your sound because it take less of them to create complex tones. You can simply build really phat sounds, quickly and easily. It’s something new to explore. Let’s begin with an Initialized Voice.

Press [FUNCTION]
Navigate to screen #4
Select “VOICE INIT”
Select “YES”
Press [ALGO]

This time lets select Algorithm #12… where each Operator is a Carrier (which means it is capable of audio output). Again, only Operator 1 has its Output Level set to 100, in an initialized Voice.

Playing the keyboard, you will hear the pure (and lonely) sine wave – no harmonics. Let’s apply some Feedback: Much like a microphone in front of a speaker column or an electric guitarist standing up close to the amplifier, Feedback is the process of signal traveling in a loop, output back to input, over and over again. What the engineers designed this time is a way to channel that energy specifically toward creating Sawtooth type Waves (those that contain every whole integer multiple of the Fundamental – ie, all harmonics) or toward creating Pulse type Waves (those missing certain odd numbered harmonics according to the width of the pulse). A Pulse Wave that is 1:2 (square) will be missing every other harmonic; it contains only the odd numbered harmonics. A pulse that is 1:3 will be missing every third harmonic, a pulse that is 1:4 will be missing every fourth harmonic and so on.

Press [FB] Feedback
You arrive at a screen showing the four Operators set to 0 (0 here means sine wave). As you swipe up in touch column 1 you will be sending the output of Operator 1 back into itself and thereby generating a Sawtooth Wave output. The higher the value toward 127, the brighter the result. The wave will contain all harmonics.

Notice that there is a “law of diminishing returns” in effect here: as you increase the value, the sound gets richer and brighter, however, once you exceed a certain value the sound starts to over modulate and thin out. Hint: when setting Output Level – use your ears, they will serve you better than the mathematics!

Try the opposite direction, again, as you increase the value by swiping down from the 0 centerpoint, the sound gets richer and rounder, as it tends toward the Square/Pulse Wave output. And again as you near the extreme the sound begins to thin out again. 

Extra Credit: We mentioned that only Carrier Operators can be heard – Modulators are only heard by how they impact the timbre of the Carrier. You will discover when working with Operators and creating basic wave shapes, it is useful to use the ability to switch Algorithms. When you want to hear your Operator you can switch to an Algorithm, like #12, where all Operators are Carriers and will output audio. The integrity of your programming within the Operator is maintained even though you switch it from Modulator to Carrier, for example. In general, the more complex the Modulator’s waveform the more radically it will influence the Carrier you route it through. If your Modulator is very complex, it may quickly result in generating so many sidebands (all frequencies) it will begin to approach non-musical sound (Noise). Noise contains all frequencies simultaneously. Sometimes the best sound is obtained through subtle use of this modulating “influence”… like applying vibrato to a violin or to a synth lead, sometimes a little is all you need. Applying too much can lead to a bit of chaos. The “Modulation Index” is a fancy way of saying the “amount” of modulation you are applying to the Carrier (Modulation Index = the Output LEVEL of the Modular). As we go forward, we’ll show you how to isolate the contribution of each of your Operators. Next we’ll look at the Amplitude Envelope Generator – how loudness or output is controlled over time. The old “ADSR” (Attack-Decay-Sustain-Release).

If you have any questions or comments about this article, please let us know in this forum thread.