Chapter 7: Instruments that Create Sound
Using Technology to Pinpoint What Creates Sound
In February 2020, Yamaha released the YC61, a stage keyboard featuring the newly developed Virtual Circuitry Modeling (VCM) Organ tone generator. The following year, we released the YC73 and YC88 to round out the series at three models. Although these are stage keyboards rather than synthesizers, they are internally indistinct from the latter in that they contain AWM2 and FM tone generators in addition to the VCM Organ tone generator. We chose “YC” for the model name in a nod to the YC-10 combo organ released in 1969, a series that would go on to fuel the “band boom” of the 1970s. Although intended to be based on a different concept to Yamaha Electone electronic organs, the series shared the same audio output system as the Electone. Fifty years later, we have revamped the YC Series with the new VCM tone generator.
Interestingly, the underlying VCM technology originated as a research project of K’s Lab, an in-house research team headed by Toshifumi Kunimoto, a Doctor of Information Science also known as “Dr. K.” The team was formed in 1987 to develop the next generation of tone generator systems after the commercialization of Yamaha’s FM and AWM tone generators. The lab was engaged in a groundbreaking effort to mathematically determine the principles by which acoustic instruments produce sound, laying the groundwork for the Virtual Acoustic (VA) tone generator that would appear on the VL1 synthesizer in 1993. However, calculating the acoustic characteristics of conical wind instruments such as saxophones required such a vast number of formulae that the VA tone generator would overwhelm the digital signal processor (DSP) technology available at the time. In search of a solution, K’s Lab discovered a paper on branching theory published in Japanese in 1977 (Note 1). Branching theory holds that the acoustic characteristics of a conical pipe can be approximated by two cylindrical pipes. Using this theory, the lab used cylindrical pipes of a fixed diameter to simplify the formulae, making the new tone generator system more practical and leading to the release of the VA tone generator. Incidentally, we applied branching theory beyond the field of electronic simulations, using it to design the Venova, a casual wind instrument. The fully acoustic instrument debuted in 2017 and won the Good Design Award in Japan.
Note 1:
Ensui horn no hankyoshin shuhasu no kyumenha riron ni yoru kaiseki - ensui-gata kangakki no naru shuhasu (Analysis by the spherical wave theory of the anti-resonant frequency of a conical horn - Sounding frequency of cone wind instrument), Junichi SANEYOSHI, 1977.
K’s Lab was involved in the development of many more tone generators after the VA, including the AN1x (1997), a DSP-based analog modeling synthesizer, and the EX5 (1998), which featured the Extended Synthesis hybrid tone generator system. Around 2001, the lab began developing effects using physical models of analog circuits.
Analog circuits are a combination of capacitors, transistors, and other components; even with components of equal capacitance and amplification factor, different manufacturers and models result in slight variance. Especially in the case of sound signals and other electrical signals with varying frequency, the frequency characteristics of the components have a substantial effect on the sound. In other words, a pair of effectors with the same function sound different if their analog circuits contain different components or were assembled differently. This is one reason musicians and engineers prefer one product over another, and why they choose so carefully.
K’s Lab went on to study the behavior of analog circuits and develop digital effects using physical models for several mixing consoles, including the Yamaha DM2000VCM professional digital mixing console (2000), the first product with “VCM” in the name. VCM would appear in many more products as it continued to evolve. It is the underlying technology for Open Deck, a technology that simulates rotation speed fluctuation and other aspects of sound quality experienced when recording and playing music with a tape recorder, and also featured in RND Portico plugins, a collaborative effort with Rupert Neve Designs.
Along the way, the developers of the YC61 decided to apply this technology for simulating analog circuits and other physical phenomena with DSP to the development of a new organ tone generator, and focused on using VCM technology to simulate a tonewheel-type organ tone generator.
Unlike electronic oscillator-based organs, tonewheel-type organs produce sound using the electromagnetic induction generated when the metal gears rotate near the pickup. Picture an electric guitar. In the case of an electric guitar, the strings vibrate from side to side above the pickups to create electromagnetic induction, causing electricity to flow. The higher the vibration frequency of the strings, the higher the pitch. Tone wheels feature gears with jagged teeth. When the gears rotate, the distance between the metal and the pickup changes in the same way as strings vibrate over pickups, ultimately creating the same effect. If two gears are rotating at the same speed, the one with more teeth produces the higher vibration frequency. If two gears have the same number of teeth, the one rotating faster produces the higher frequency. Therefore, the pitch can be adjusted by changing the rotation speed or number of teeth.
Actual organs have a tone wheel of the requisite frequency for each pitch, and the sound is created by combining them with tone wheels that produce the harmonics for each note, with volume adjusted using a lever called a drawbar. Tone wheel organs involve the mechanical element of gears rotating; thus, fluctuations in rotation speed, variance in the physical shape of the gears, and other factors impact the resulting sound. The pickups, which consist of magnets and coils that pick up the sound, add even more factors, including the type of magnet used, the number of times the coils are wrapped, and the thickness of the wire. The VCM Organ tone generator utilized VCM technology to allow DSP to compute these parameters in pursuit of more realistic organ sound.
The YC61 features VCM technology other than the VCM Organ tone generator. While most organs are generally made with an amplifier equipped with a rotating speaker, such as the Leslie, the YC61 models this rotating speaker sound with an effect called the VCM rotary speaker. VCM Effects supply phaser, flanger, and other vintage analog effects that are popular among professional musicians.
The YC61 also features an organ sound supplied by an FM tone generator. The oscillators on FM tone generators (digital components known as operators) produce exceptionally clean sine waves. The YC61 used this fundamental characteristic to simulate an organ tone generator with the transistor oscillator used for the Electone tone generator, and featured a total of six types of FM tone generator organs, with eight drawbars that act as carriers, as well as modulators and feedback. The conventional drawbars can be used to enjoy a completely new FM organ sound. Since both the VCM and FM organs run on digital circuits, the YC61 features fully digital replicas of the sounds of electric and electronic organs that have been played for over 50 years.
Efforts on the YC61 went beyond the tone generators, delving into the role of the stage keyboard and examining every aspect of the user interface. While many synthesizers and keyboards to date have only one user interface in which buttons and dials are used to select and edit voices, the YC61 has three: an Organ section, a KEY-A section, and a KEY-B section, each of which can be programmed with voices and activated or deactivated as needed. This visual clarity and pre-programming ability make it easier for keyboard players to switch between piano, electric piano, and organ voices, as they often need to do when playing with bands, and is better for live performances thanks to easier switching between synth voices with piano voices momentarily layered with an electric piano or split with an organ sound while playing. This is the same concept as the CP73 and CP88, which were developed around the same time, and was carried over to the CK61 and CK88 stage keyboards released later.
MONTAGE M Released Ahead of 50th Anniversary of Yamaha Synthesizers
In 2016, Yamaha released the MONTAGE, a flagship hybrid synthesizer with AWM2 and FM-X sound engines. Seven years later, in October 2023, we released the MONTAGE M featuring the new AN-X sound engine. In fact, the “M” in MONTAGE M signifies the fact that, with three sound engines, the instrument has evolved from a hybrid to a multi-synthesizer.
The new AN-X sound engine is a virtual analog modeling type that uses digital technology to simulate an analog synthesizer. Compared to the AN sound engine (also virtual analog modeling) on the AN1x released in 1997, its specifications are significantly advanced to the point that, when combined with the array of real-time control of the MONTAGE, players can now enjoy more complex voice editing.
First, the basic structure has three oscillators (compared to two with the AN1x) as well as a separate noise oscillator. There are seven LFOs (up from two), and the filters have been upgraded to dual filters for more complex sound creation, for example using a high-pass filter and a low-pass filter at the same time, or creating two low-pass filter peaks and manipulating each one independently. Each oscillator can be routed directly behind the filters, making it possible to create unfiltered sound from certain oscillators. We also added parameters for more analog-like simulation, namely Ageing (equipment age), and Voltage Drift (voltage fluctuation) which simulates variation in oscillator pitch and filter cutoff variations and other attributes. Additionally, realistic analog synthesizer behavior is simulated down to the details, for example setting whether the oscillator phases are aligned or shifted by keystroke, or controlling the variation in pitch of each oscillator. The result is a warmer sound than expected of a digital synthesizer.
The AWM2 sound engine has also evolved substantially from the original MONTAGE, with the number of elements increased from 8 to 128. Normally, there are only eight elements, but opting to add elements allows more approaches and greater freedom in sound creation, for example layering eight or more waveforms simultaneously without using multiple parts, or creating voices with completely different instrument sounds or chords assigned to each key.
Elements are another key to broadening the range of expression the instrument is capable of. The XA Control parameter for controlling how an element emits sound, for example an element that sounds only when the keys are played legato or are released, makes it possible to add noise-like sounds more like those of an acoustic instrument. It is also possible to simulate variations in sound through random playback of slightly different waveforms of the same instrument sound.
Another difference between the MONTAGE M and previous models is the method of saving voices. Each voice is built on 16 parts selectable from one of the AN-X, FM-X, or AWM2 sound engines, which are saved as a unit called PERFORMANCE. Through the MOTIF series, there was a clear distinction between VOICE mode, where voices were switched one by one between piano, organ, and synth voices, and PERFORMANCE mode, which gathered multiple parts to render VOICE mode voices for a layered sound. As for the relationship between VOICE and PERFORMANCE, the format of saving voices as voice and multi-parts was introduced around the time the sound engine section became multi-channel (multi-part); from the EX5 onward, the names VOICE and PERFORMANCE were used more explicitly and became synonymous with the voice saving format of Yamaha synthesizers. However, from the MONTAGE onward, the VOICE method of saving single parts was abandoned; everything is now saved as a PERFORMANCE.
The MONTAGE M allows the user to choose between AN-X, FM-X, and AWM2 for each part of a PERFORMANCE; for example, setting Part 1 to AMW2 Piano, Part 2 to FM-X Electric Piano, and Part 3 to AN-X Pad Sound results in a layered sound consisting of acoustic piano, electric piano, and synth pad. This makes it possible to combine all sound engine systems in an oscillator-like manner to create sounds, and using the Motion Sequencer and SuperKnob to simultaneously control parameters across parts provides an approach that transcends the bounds of sound engine systems even for real-time voice editing.
Smart Morph for Creating Voices from Voices
The Smart Morph function (Note 2) allows MONTAGE M to machine-learn two or more voices and use the constituent elements of each to create a new voice that can be operated graphically.
Suppose A is a synth pad voice, B is an electric piano voice, and C is a synth brass voice (either FM-X only or AN-X only). When this is entered onto the Smart Morph Edit screen and the Learn button is pressed, the screen takes on the appearance of an XY pad displaying the machine learning results in color. The display shows a white square with a connecting line from each voice, and a single blue square in the upper left corner. The user can touch the screen to move the blue square. If they move the blue square to the white square corresponding to A (the synth pad), the synth pad voice sounds. If they move the blue square to the white square corresponding to B (the electric piano), the electric piano voice sounds. Moving the blue square anywhere between A, B, and C causes the MONTAGE M to automatically generate a voice that blends the respective elements proportionately. The colors on the screen show the relative strength of the voices; for example, the closer the blue square to B (the electric piano), the stronger the electric piano element.
Thus, users can create new voices from existing voices. They can also use the SuperKnob to move the blue square on the map (the XY pad-like screen) in real time for more complex voice creation.
Note 2:
Smart Morph is available on the MONTAGE 3.5 OS or higher and the MONTAGE M (FM-X only), and can be used with AN-X on the MONTAGE M 2.0 OS and higher.
Evolution of the User Interface
As sound engine systems become more sophisticated and gain new functions for sound creation that can be combined to create new voices, it becomes more difficult to master the techniques required to use them toward that end. Of course, a color touch screen can vastly improve operability and visibility, but MONTAGE M goes beyond and incorporates even more ideas.
First, the main display in the center of the main unit allows the user to switch voices with a simple touch. With the Category Search function, the user can search through voice categories by touch rather than using a drop-down menu. These display switching and touch buttons are linked to the LIVE SET and CATEGORY buttons and the 16 buttons below them on the right side of the unit so that either can be used to make changes, and the LED lighting of the buttons and the display highlights are always linked.
The MONTAGE M is also equipped with helpful solutions for editing voices. For example, when editing AN-X voices, there are three oscillator sections and settings for pitch, LFO, pulse width, modulation, and other detailed parameters that cannot be displayed on a single screen. Therefore, there is generally a screen for each oscillator with buttons or drop-down menus for moving between oscillators. The hierarchy became more complicated due to filter and amp sections, envelope settings, and other added features, making it more difficult for the user to reach the desired parameter screen. To solve this problem, the MONTAGE M has a NAVIGATION button on the right side of the main unit. One push, and the main display shows a block diagram of the voice being edited. Returning to the block diagram provides an overview of the sections each parameter belongs to, putting the desired parameter in reach with a simple touch of the corresponding section on the main display.
The MONTAGE M also has a sub-display in the upper left corner that complements the main display. The sub-display can be operated with the eight buttons at the top, the PAGE button on the left side and the eight knobs and QUICK EDIT button at the bottom, and can display the parameters of the selected voice on the main display independently of the main display. The sub-display also features a distinct display layout showing the basic signal flow of Oscillator, Filter, and Amp layouts in the same way for AWM2, FM-X, and AN-X, making voice editing more intuitive there than on the main display, which shows a unique layout for each sound engine system.
Additional features include color-changing LED lights on the button switches on the panel that indicate when the function of a button has been switched. This and other thoughtful developments have made this multifunctional, parameter-rich synthesizer more straightforward for users.
MONTAGE M 2.0 Debuts on 50th Anniversary of Yamaha Synthesizers
In this milestone year marking the 50th anniversary of Yamaha synthesizers, the MONTAGE M operating system was updated to version 2.0. The update includes many additions and improvements, starting with waveforms from the CFX, Yamaha’s flagship concert grand piano, which were released in 2022. Also included are the recently popular Shimmer Reverb effect and the Smart Morph function, which was previously only compatible with FM-X and can now be used with AN-X. Here, we will introduce another upgrade to 2.0: support for the MIDI 2.0 protocol released in 2019.
The MONTAGE M is Yamaha’s first MIDI 2.0-compatible synthesizer. By virtue of transferring MIDI data in 32- to 128-bit packets as opposed to the conventional 8-bit, MIDI 2.0 offers substantially higher data resolution than MIDI 1.0. Using control change data for transferring data on volume and other parameters as an example, with MIDI 1.0, only seven bits are available in the data area, resulting in maximum resolution of 128 steps (27). The data for parameters like pitch bend comprises pairs of seven-bit bytes combined to create 14-bit bytes, yielding a maximum resolution of 16,384 steps (214). MIDI 2.0 data areas feature 32 usable bits for a staggering maximum resolution of nearly 4.3 billion steps (232 = 4,294,967,296). Velocity is also dramatically improved, with 16 bits for a maximum of 65,536 values.
It should be noted that these figures only represent what is possible with MIDI 2.0; actual equipment may not be capable of such high resolution. Nonetheless, the MONTAGE M SuperKnob has 1,024 values from 0 to 1,023, an extremely important feature for compatibility with MIDI 2.0 (and much more than can be transferred within the 128 steps of MIDI 1.0). MIDI 1.0 was created in 1982, and Yamaha released its first compatible instrument, the DX7, the following year. In the four-plus decades since then, seven- and 14-bit bytes have been enough to make electronic instruments sufficiently expressive, but now, we have entered a new age where more resolution is needed.
The MONTAGE M 2.0 OS supports resolution of 10 bits (1,024 steps) for parameters such as velocity, aftertouch (polyphonic and channel), SuperKnob, pitch bend wheel, modulation wheel, 8-knob, 8-fader, foot controller, and sustain. In other words, a MIDI 2.0 connection is required to take full advantage of this feature in combination with a DAW or other software. Of course, it is possible to work without any loss of quality when using the MONTAGE M by itself.
The MONTAGE M8x also features a new keyboard to allow users to take full advantage of this high-resolution MIDI capability. For the FSX keyboards on previous models, velocity values are calculated by detecting the speed at which each key passes between several detection points throughout the stroke. In contrast, each key on the GEX keyboard on the MONTAGE M8x has coils attached to it and the keybed beneath it, and velocity is detected using the electromagnetic induction generated by the distance between the two coils. Pressing a key decreases the distance between the coils and changes the voltage generated at the ends of the coils (as the coils get closer to each other, the voltage decreases and a stronger sound is produced). Using the distance between coils means the vertical position of each key can be detected at many more points, increasing resolution and making it possible to capture velocity on a finer scale. The GEX keyboard also supports polyphonic aftertouch by virtue of detecting when a key is pressed deeper than normal based on the distance between the coils, making it suitable for the dramatically increased resolution of the MIDI 2.0 era.
With synthesizers, we tend to focus on the structure of the sound engine system—analog or digital, AWM2 or FM—but the fact that after 50 years Yamaha continues to improve user interfaces, keyboards, and other components to give players what they want is proof that the company considers synthesizers to be musical instruments worthy of the very best effort in development.
Proposing a New Relationship between Software Synths and Hardware Synths
Since it became mainstream to use DAW to produce music, the way we think about the relationship between software synthesizers and hardware synthesizers has changed in many ways.
Once the DAW-based production style became the default, music producers began to see hardware synthesizers as master keyboards, and composers and arrangers viewed them as items through which to experiment with motifs. These new roles were a major factor in Yamaha’s design of the MOTIF synthesizer.
We attempted to provide editing software to make hardware synthesizer sounds available as their software counterparts with audio interface functionality, but ultimately failed to change the paradigm of using hardware synthesizers for live performances and software synthesizers for production. As computer specifications continue to improve, a growing number of musicians are using software synthesizers in their live performances, to the point where it feels as though hardware synthesizers are falling out of favor completely.
Amid this trend, a program called Expanded Softsynth Plugin for MONTAGE M, or ESP, emerged to shape a new relationship between software synthesizers and hardware synthesizers.
Although ESP could correctly be described as a software synthesizer version of the MONTAGE M, the key is in its strategy. ESP is available free of charge to registered users of MONTAGE M; it is not sold separately. Also, since the synthesizer is not used like a dongle, it will work even in the absence of the MONTAGE M. This is very important for professional musicians, making it useful in two cases:
1. Musicians who produce at home and often travel to the studio for vocals and mixdown sessions
Many musicians appear to be working this way these days. ESP gives them access to the same sounds in a DAW environment as with their MONTAGE M, while leaving the hardware at home. In the past, musicians often created audio files to take to the studio instead of their hardware synthesizers, but ran into trouble because they could not alter the voices during mixdown sessions. Although other plugin synths are an option for making major changes, ESP is optimal for making minor edits.
2. Musicians who record while on tour
Many musicians encounter problems when they want to record with an artist while they are on tour and their equipment is in transit and therefore unavailable. Some get around this problem by purchasing multiple sets of identical equipment for touring and recording, something that ESP eliminates the need for. Another advantage of ESP is the ability to load voices used for new recordings directly into a MONTAGE M to perform the songs on tour.
This demonstrates how ESP combines the advantages of hardware synthesizers and software synthesizers so that users can benefit from both. The fact that ESP is available free of charge to registered users of MONTAGE M may seem like a bonus feature, but the idea is that the MONTAGE M synthesizer comprises both the hardware and ESP. This is Yamaha’s answer to the current hardware vs software debate, as the company celebrates the 50th anniversary of its synthesizers.
The Next 50 Years
Fifty years have passed since Yamaha released the SY-1 in 1974. Starting with a project to increase the expressiveness of the Electone, we have made numerous technological innovations in sound engine systems, keyboards, user interfaces, acoustic technologies, and more, culminating with our 2024 release of MONTAGE M 2.0.
The past half-century is filled with synthesizers that exist only because Yamaha is persistent and earnest in its quest to turn its ideals into reality and share its distinctive technologies with the world. There are many ways to describe our approach of using the array of technologies cultivated by our forebears to create innovative products that satisfy the needs and exceed the expectations of the times, based on our belief that synthesizers are instruments for creating sounds. One may be the fervent desire to use Yamaha technology to create instruments that create sound. Sustaining this aspiration will drive us to discover technologies to pen new chapters in the history of synthesizers for the next 50 years and beyond.
