The technology of sound
Recreating vintage classics using cutting edge techniques
In our never-ending pursuit of perfect tone, we are driven to seek out the best, most accurate methods of recreating legendary synths and keyboards. Whether by mapping out the DNA of every unique electronic component, or by analyzing the precise sonic properties of acoustic instruments, we create the most flawless, detailed emulations possible.
What is TAE®?
TAE® is "True Analog Emulation" - Arturia's exclusive technology that accurately reproduces the tone, waveshape, tuning and other detailed characteristics of an analog synthesizer.
Better reproduction of analog oscillators
TAE® recreates the characteristics of analog oscillators in amazing detail. The transparency and clarity of our analog synthesizer emulations give musicians the freedom to be inspired by classic sounds or explore new textures knowing that the previous limitations of digital oscillators will not interrupt the creative flow.
Better reproduction of analog filters
Filters are a major element in subtractive sound synthesis. It is important to reproduce the exact characteristics of analog filters in the digital domain, as well as the individual characteristics of classic instruments that give each one its unique sound.
Implementation of soft clipping
Soft clipping can give the sound more punch and presence, not to mention limiting the amplitude of the sound. It’s another important aspect of the circuitry that gives analog synthesis its character.
How do we do it?
A great attention to detail
TAE® instruments are created using the most advanced DSP techniques by today's standards. Depending on the context, our instruments are modeled starting from mathematical models, physical models, simulations of their circuitry. From then, we sit next to the instrument we are modeling and listen to every little nuance in the sound it produces. It could be the response of the filter cutoff knob, a small artifact in the waveform of an oscillator, even the noise that is generated by the circuit. Then we find ways to reproduce those behaviors in the digital domain. It takes a lot of effort and a good knowledge of the analog instrument we emulate, because those subtle variations to the "perfect" sound are never the same from one synthesizer to the next, but it is what makes our emulations so true to the original.
Better reproduction of analog oscillators
TAE® oscillators are very similar to their analog counterparts for several important reasons.
One of the main reasons is that they are "free-running"; i.e., they are not sampled, wavetable-based or generated from a 0-point when a note is played. Each waveform is generated dynamically, which is more consistent with the way the original devices worked and with the laws of physics. Thus our oscillators avoid one of the major tell-tale signs of digital waveform generation and provide that level of "liveness" found in the classic analog synthesizers of the past.
Standard digital synthesizers produce aliasing in high frequencies, and also when using Pulse Width Modulation or Frequency Modulation. TAE® allows the production of oscillators that are completely free of aliasing in all contexts (PWM, FM, hard sync, etc.), at no extra CPU cost.
Temporal representation of the “sawtooth” waveform
of a hardware synthesizer
TAE® oscillators are highly optimized for each digital environment, so you’ll be able to play our instruments exactly as you would if you were playing the original analog synthesizer. Again, the waveforms are not pre-sampled and regenerated with digital perfection; each note has a life of its own.
Another factor is that the original analog oscillators were basically unstable. And we’re not necessarily referring to the pitch; actually, their waveforms were always slightly different from one period to the next. And due to analog hardware sensitivity, new period trigger times varied with the temperature and other environmental conditions.
So we factored that in, too: TAE® simulates the oscillators’ instability, which helps to create a warmer and fatter sound.
Temporal representation of a “sawtooth” waveform
reproduced by TAE®
Original analog oscillators used the condensers' unloading to produce common wave shapes (sawtooth, triangle and square). This means that waveforms were slightly curved or distorted in other ways that are considered highly desirable in a musical context. TAE® allows the reproduction of a condenser's unload in order to give you the original analog sound.
And since TAE® oscillators behave in all of these “lifelike” ways, when you play a simple chord with the raw oscillators you won't get the impression of digital "fog" or a static sound. Instead you’ll hear a refreshing transparency and clarity.
With TAE® you can even produce complex PWM sounds with unprecedented quality and free of any aliasing. This is true for all of our TAE® waveforms (square, triangle, saw, etc.). Our instruments produce sounds which are unique on the market, because PWM is very difficult to achieve on certain waveforms like triangle or ramp waves. TAE® brings new levels of reality in a virtual analog synthesizer, so that you can achieve new levels of creativity in your music.
Better reproduction of analog sounds with direct filter circuit modeling
TAE® reproduces the characteristics of analog filters in the digital domain.
Let’s take our Prophet-V for example. Due to advances in computer processing power, Prophet-V can employ direct filter modeling techniques to achieve unprecedented accuracy in the emulation of the original 4-pole low-pass analog filter. By modeling the operation of the individual hardware components of the filter circuit, the warm nuances synonymous with the original analog instrument are recreated.
This graph is a frequency domain plot that serves as just one example of direct circuit modeling in action; it shows the generation of harmonics at multiples of the resonant frequency when the filter is in self-oscillation mode, for both the virtual (blue) and original (red) instruments.
These harmonics are characteristic of the Prophet 5's filter and are due to the non-linear behavior inherent in its analog circuitry. The generated harmonics add to the richness and warmth of the sound produced by the filter. As a result of the direct modeling of the analog circuitry the same sonic characteristics are present in Prophet-V, thus giving the user a truly analog sound.
Harmonics generated by the filter circuit when in self oscillation prophet 5 original: blue, prophet virtual: red
Clipping me softly…
The importance of soft clipping
In analog synthesizers the resonant filter uses a current-limiting function, which prevents the signal from being too loud (soft clipping).
TAE® emulates this current-limiting function, making the sound more natural. It also allows filters to enter self-oscillation like original hardware synthesizers do.
Also, soft clipping can be a kind of saturation effect, but with a very particular shape. Common shapes can't give good results compared to analog soft clipping; it is a peculiarity of the output amplification stages of analog synthesizers, and is a critical component to the analog sound.
What is Physical Modeling?
Recreating acoustic & electric instruments using advanced physical modelling.
Arturia are renowned for their analog component modelling and giving legendary synths a new lease of life as software instruments, but how do we recreate physical instruments with strings, mechanical parts, and made of resonant material?
Physical modelling is a process of recreating the physical reactions and unique interplay of both physical and electronic components using mathematical models and algorithms to create a realistic reproduction. This scientific process is used in many fields, from seismology to the automotive industry. Interestingly, the same principles that create accurate earthquake predictions and more efficient engines and vehicle design can also be used to give vintage instruments a new lease of life in software.
When it comes to enjoying the sound of each classic electric and acoustic instrument featured in our V Collection, there are only 3 real options to consider.
Option 1 would be to own every instrument, maintain them, tune them, and keep them in a humidity and temperature-controlled, warehouse-sized studio. Nice in theory, but totally impractical and impossibly expensive in reality.
ThreeWaves Music Store (NJ)
Option 2 would be to use audio sampling technology. In the past, this would have been the obvious solution to recreate these keyboard legends. If something sounds great, record it playing individual notes, and it’ll sound just like the real thing when you play it back, right? Not exactly.
To sample, or not to sample?
Sampling in various forms certainly had a massive effect on the course of popular music. From the early ‘60s, the Mellotron let musicians perform using tape recordings of real instruments and voices, while in the late ‘70s the Fairlight CMI kick-started the digital sampling revolution. However, sampling has its limitations.
A sound snapshot
Not matter how well recorded, any sample is just that: a sample. Single notes sound just like the original, but when you begin to perform and add that human touch, the sound feels static and lifeless. Your ears are incredibly sensitive, and they can easily detect the lack of nuance and interplay that comes with sampled sound.
To create expressive, dynamic sounds with sampling, you need to capture several velocity layers, from the softest touch to the hardest hit on the keys, and on as many keys as possible. To save on disk space, hardware and software instruments often restrict this sampling to just 4 or 5 layers, and blend them together to create the areas in between.
To accurately capture every layer, note combination, and parameter option on even a simple keyboard would require ludicrous amounts of data. While this is certainly possible using today’s technology, it’s massively impractical. Sampled instruments by definition use huge amounts of RAM to store their samples for quick recall as you play them.
Chasing your tail
A common practice for sampling is to accurately capture the initial “attack” and“decay” elements of the sound, but use a process called “looping” to simulate the “sustain” and “release” of the sound to save space.
Where a real instrument’s sound will continue to evolve and change over time, often you’ll find that sampled instruments simply repeat themselves until you release the key.
Modelling: the way forward
While there’s no doubt that many of the hardware and software instruments that use samples sound great, we weren’t satisfied.
We wanted more. We wanted to create dynamic instruments that responded just as the originals did. We wanted to use the latest, cutting edge technology to recreate the iconic physical keyboards of the past.
And so, we come to
Option 3 Physical Modeling. Arturia’s passionate, skilled developers and engineers were already well versed in the reproduction of classic synths; analyzing their analog electronics, modelling their components, and matching the original sound. This process, Arturia’s exclusive True Analog Emulation® technology, saw the rebirth of some of the greatest analog synths of all time as affordable, accurate software instruments.
As Arturia’s software offer grew, so did our ambition. Our users wanted to be able to play instruments with physical, vibrating components, from classics like the Fender Rhodes to the iconic Clavinet and Wurlitzer; the mighty VOX Continental; as well as all manner of acoustic pianos.
We are Arturia, and we dare to be different.
For our team, it felt like a natural progression to use the same methodology and painstaking attention to detail used to create our award-winning software synths to model some of history’s most iconic acoustic and electro-acoustic instruments.
This process gives musicians an incredible level of control and sensation of realism. Let’s take a look at some of the great features of physical modelling:
Feel the power
Physical modelling makes full use of the impressive processing power found in today’s personal computers. Convincingly recreating an organic, acoustic sound in real time that’s modelled on a physical instrument requires a lot of CPU power. That power would have been science fiction even a few short years ago. Now you can embrace this technology and shake off the drawbacks of sampling.
No layers, just sound
Instead of transitioning between a handful of sampled sounds for different velocity levels, physical modelling creates a seamless, perfectly fluid experience just as you’d expect from the physical instrument. No sudden jumps in volume or tone that put you off your stride, just beautiful recreated sound that you won’t want to stop playing.
Our deepest sympathies
The delicate interaction of each note the instrument plays has a subtle but mesmerizing effect on the sound as it evolves. A phenomenon known as “sympathetic resonance” creates harmonic interplay that simply can’t be recreated using samples. Physical modelling brings this beautiful quirk to life, as virtual strings, tines, and speakers carry vibrations to virtual cabinets and soundboards, giving extra depth and realism.
Make a change
Because the behaviour of every single component has been meticulously analyzed and recreated, you can ignite your creativity by personalizing the instrument to suit your sound. Want to try a crazy drawbar combination on your organ? No problem. Want to try something a bit more adventurous with your piano? Swap out the wood cabinetry for glass or metal. Want to change the pickup position on your Rhodes? Just turn a knob. Physical modelling makes seemingly impossible things easy, fun, and inspiring.
Making the model
Arturia’s physical modelling is at the cutting edge of acoustic technology, the most recent chapter in a story that began with pioneering scientists in the 1970s.
From there, major breakthroughs in the modelling process from Stanford University in the USA and IRCAM in France opened the doors to exciting new methods and concepts.
Stanford University (US)
Audio engineers were now able to make detailed, accurate models of instruments, with their acoustic properties and nuances beautifully replicated. No small task, as every element that produced the original sound needed its own dedicated algorithm, from the transient detail of the attack to the harmonic structure as the note releases, from the type of material the instrument is made of to the size and tonal properties of the virtual room it was played in.
The benefit of giving this attention to detail is that it provides a huge level of expression and finesse to the performer, far beyond what is possible with sampling alternatives. While it’s true that no digital recreation could top the tactile, vital feel of playing a classic acoustic instrument or vintage keyboard, physical modelling can now bring us closer than ever, without the pitfalls of sampling, and at an affordable price.
Pushing the boundaries
Over the years, Arturia has been at the forefront of developing and progressing the technology of physical modelling, staying true to our company aim of bringing amazing sounds and instruments to a new audience
The power of Arturia’s physical modelling technology can be found in the following software instruments:
In partnership with IRCAM we pioneering acoustic instrument reproduction in 2006 with Arturia Brass, recreating the unique timbre, rasp, and zingy tonal properties of a brass ensemble, including trumpet, trombone, and saxophone.
A leap into the world of drums and percussion, the Spark incorporates physically modelled drums alongside sampled and synthesized instruments to give musicians an extra level of depth and realism.
The first Arturia classic keyboard emulation to use physical modelling, it reproduces the acoustic properties of the reeds, hammers, mechanical noise, amplification, and even the resonant sound of the plastic body, creating and virtual instrument with outstanding realism and flexibility.
A powerful piano software suite, making full use of physical modelling technology to accurately map out and reproduce the whole process of piano recording. From the delicate nuances of components and hammers, to microphone placement and the sound of the ambient environment, Piano V became the go-to resource to create the perfect piano sound.
Perfectly recreating the legendary Fender Rhodes, the iconic sound behind countless hits, would always take more than samples. The growl, percussive power, and soulful tone of this classic has a special place in many hearts, and by the power of physical modelling, every nuance is recreated, from its tines and cabinetry, right down to the awesome effects pedals and amp.
The latest step in Arturia's physical modelling journey, Clavinet V represents the culmination of the Phi story so far. The sonic backbone of funk, R&B, soul, and jazz since its release, to rely on samples to recreate this incredible instrument would be a huge injustice. Thanks to Arturia's modelling know-how, Clavinet V has the same rasp and vibrancy of the legendary original.