The numerous „new“ releases in the field of analog synthesizers often seem to obscure the view of those developers, who not only try to develop „alternative“ sound generation systems and instruments, but also endeavor to bring them to the market. Especially the latter does not always seem to be easy…
One who has succeeded in both, has kindly agreed to give us an interview – Stephan Schmitt, founder of Native Instruments and Nonlinear Labs, and mastermind of the C15.
You’re probably wondering why „us“? This interview is also published – kind of back/back – on Amazona.de.
If you are interested in C15 sound samples you can listen to some while reading this interview. Please scroll right down to the end and press the orange Play button. The sound examples from my SoundCloud will start to play. In total about 23 minutes of audio demos.
I hope you enjoy reading it!
Hello Stephan! First of all, thank you for being available for this interview. Let’s start with Nonlinear Labs right away. What was your motivation after Native Instruments (NI) to found a new company with Nonlinear Labs (NL)?
Interview Stephan Schmitt – Nonlinear Labs
Native Instruments products are primarily used in digital music production and DJ applications. They integrate well with common sequencers and loop-oriented tools. But my personal passion has always been live music and non-electronic music styles.
NI was not the ideal environment to translate the experience gained in the development of digital sound generators into instruments for performing musicians. That’s why I decided to found Nonlinear Labs. With a small company you can also implement product concepts for smaller target markets.
The C15 is the first instrument to follow this idea and is a further development of the NI Kontour Software Synthesizer. Without going into the differences in detail at this point[note – please see separate information at the end of the interview], why did you further develop the synthesis engine in this direction?
The synthesis engine of the C15 has a longer history. I had already developed the reactor instruments Spark, Cha-Osc, Prism and Skanner for NI. When we started with the C15 at NL, I also developed its sound generation in Reaktor under the working title „Phase 22“. Colleagues at NI asked if this could be turned into a software instrument. I got involved in a license deal and the result was given the name „Kontur“.
In order to meet the requirements of typical NI customers, Kontour received four flexible modulation generators. The C15 doesn’t have this, because we leave the influence of sound to the musician. To achieve this, we have further developed many aspects of synthesis, effects and control. The experience with contour was very useful, and we could also use the results of the sound design commissioned by NI further.
Although the MIDI CC 15 is „undefined“, I assume „C15“ doesn’t stand for it?
There are two derivations for the name. First one: We wanted a plain product name like „DX-7“. The „C“ initially stood for a keyboard instrument as in „Clavichord“ or „Clavinet“. The „15“ exemplyfies 2015 the year of completion of the first near-production prototypes.
But now I like the following interpretation better: „C14“ (actually the number is written in superscript form before the letter) is a radioactive isotope of the element carbon. It decays with a half-life of 5730 years and is therefore used in archaeology to determine the age of organic materials. „C15“ is a carbon isotope that disintegrates to half its size in about 2.5 seconds. It’s like a piano envelope. (https://en.wikipedia.org/wiki/Isotopes_of_carbon)
Carbon is the basis for all organic substances and of course also wood. Since our C15 can sound very organic and wood plays a major role, I think the name is well suited.
Let me come back to MIDI, as it seems to be discussed most intensively in connection with the C15. Specifically, the lack of a MIDI interface. In addition to your personal decision, which you have already explained in other interviews, are there also technical reasons?
The C15 captures and processes the touch and release velocities of the keys and the movements of pedals, ribbons, bender and aftertouch in high resolution (with more than 4000 steps instead of 127 steps in MIDI). With an external MIDI connection we would have had to translate it into a coarser resolution. More exotic formats like NRPNs and CC #88 could have been used as High Resolution Velocity Prefix, but then it would have been questionable if the connected MIDI hardware or software could do something with it. By the way, the way we interpret the continuous sustain pedal is not compatible with the sustain function provided in MIDI.
It didn’t seem attractive to me to put development effort into bad MIDI compromises and to open a „Pandora’s box“ at the same time, because then adaptations to various environments would certainly be required.
How does the C15 generally look in terms of speed and resolution, e.g. the controller and keyboard data processing?
As mentioned above, the internal resolution is 32x higher than in MIDI. As sampling rate for the analog controllers we have chosen 80 Hz, for the keyboard it is 8000 Hz.
Although I sometimes miss the possibilities MIDI offers, I think that a manufacturer’s decision should be respected, especially since there are not only the technical reasons, but also the personal reasons already mentioned. Let me ask you though – did you ever regret the decision not to implement MIDI in the C15?
No, it still seems to me a consistent and necessary decision, which is more and more understood also in the circle of potential customers. Many producers of electronic music also have little desire for MIDI. You prefer to sample a phrase, continue editing the audio clip and assemble it into the arrangement. This way they „create facts“ and get on with the work, while you may never finish the fiddling in the MIDI editor.
I find it interesting that other areas of the C15 do not receive nearly as much attention. So one reads little about the flexibility of sound generation and the topic of sound shaping is generally not mentioned…..
Well, for example, when someone at a trade fair sits down at the instrument who can play a bit, he often stands up with shining eyes after getting a small impression of the tonal potential and playability.
The sound is actually generated while playing the C15 and this brings me to your recommendation to buy three additional pedals when buying the C15. I noticed, for example, that my Yamaha pedal is clearly superior to the Roland EV-5s in workmanship, but inferior in control. Does NL recommend pedals from other manufacturers?
At the moment we recommend and often deliver the Roland EV-5 and DP-10. However, we need to do a more thorough examination of pedals from different manufacturers in order to make better recommendations. It is also possible that one day we will offer our own pedals.
Do you think about increasing the number of macro controls, of which there are currently four? Four is not enough for me, especially since with the four connection options for the mentioned pedals, two ribbon controllers, aftertouch and the pitch stick, there are currently eight possible interventions.
Meanwhile, the four macro controls are too few for me. Therefore, we plan to increase their number to six in a major future software update. That would be an update, where we also change the foils on the front panels.
In order to stick to the topic of future developments, are you planning to integrate other controllers or to offer the C15 with other keyboards in the future?
There are still no very concrete plans for further controller elements. However, I can imagine e.g. something joystick-like.
First of all, we want to get even more out of the existing hardware. This is especially important to me when it comes to the keyboard. We’ve already put a lot of research into it to equip it with sensors that continuously measure how deep each key is pressed. This would open up new possibilities for expression, such as the soft insertion of notes or modulations similar to polyphonic aftertouch.
Colleague psv-ddv rightly pointed out the technical challenge and the possibility of an implementation of Last Note Aftertoch during a discussion on Amazona regarding polyphonic aftertouch. Is NL thinking about this implementation, or are you considering other solutions?
We are modifying the audio engine so that note-selective modulations for aftertouch, pitch bending or pedals are possible. Besides „Last note“, „highest note“, „lowest note“, „pressed keys“ or split ranges are also interesting.
Do you think about upgrading the hardware in the future?
Every customer will have the opportunity to update the hardware. For the customers of the first series, there will also be no costs associated with this. It is important to us to build a sustainable product with the C15 that can not only be repaired but also upgraded.
Very nice to see that the now almost inflationary concept of sustainability is actually put into practice here. Let me return to the meaning of wood as you mentioned at the beginning. What about the wood types for the case, will you offer alternatives?
With the decision to produce a large part of the instrument from solid wood, the question of the choice of wood has already arisen. So far we have beech and pear in use. Beech is available in three finishes: transparent, slightly tinted and with black glaze. The latter has been in great demand since we showed it at the Musikmesse. We’ll have to reorder first.
So there is a certain individualisation about the type and varnishing of the wood. Maybe someday we can offer a wider selection.I think the C15’s synth engine is extremely flexible and productive. What has surprised me for a long time is that the complexity remains well hidden to you, because you get good results very quickly and easily. The possibilities of intervention play their part, but this may also be due to the fact that not every parameter that can be called up can be modulated and thus a certain clarity is maintained on a small scale. In this respect, however, I still have one or the other wish, such as the possibility of directly influencing parameters using velocity, or e.g. negative keytracking in order to be able to modulate accordingly in the case of the Chorus parameter in the Reverb Engine. In the lower registers, for example, a played chord could be „widened“ in the upper registers in the melody or solo line to maintain a clean and assertive sound.
The synth engine is still undergoing an evolution process. Therefore I am always grateful for suggestions. We will try to expand the tonal possibilities without sacrificing clarity.
To get from the details here perhaps to a superordinate whole, can you give us insight into further future developments of the C15 software?
I have already mentioned some upcoming extensions. Important topics are currently dual mode and morphing. This allows the splitting and layering of groups of voices and their assignment with two presets, by the way also the routing of signals between the groups. Morphing allows seamless transitions between two presets, controlled by controller, velocity or key position. However, many software topics also concern a further development of the user interface.
You provided a good keyword with the user interface. As already mentioned, the complexity is not obvious over long distances, which is sometimes due to the playful approach, but it still exists. In my opinion, it becomes obvious when, for example, one wants to capture the feedback possibilities, even the volume ratios. Are you planning an alternative display using WiFi to give you an overview?
Many possibilities are still unused in the graphical user interface. We think of the graphical representation of the signal flow, the envelopes, the filter frequency response, the waveforms, shaper curves or spectra. Real-time level meters also make a lot of sense.
What role will the operation of the C15 via WiFi generally play in the future?
The haptic hardware user interface will always have a high priority. Therefore, we will not shift too much to the graphical interface. But of course it is especially predestined for visualizations and administrative tasks.
What about other synth engines? Do you already have plans you can share with us?
There are some drafts in the drawer. We will focus more on physical modeling, but there are also additive and modal approaches, an independent method for generating oscillator waveforms or even a Farfisa-like sound generation.
What exactly do you have in mind and if it should go towards the organ, what does it look like with the more famous organ classic?
I’m not that interested in the classics. There are enough companies to take care of it. To rebuild a Sinus drawbar organ or to develop a Farfisa clone would be boring.
Inspired by old videos with Can and her keyboardist Irmin Schmidt, who sent his farfisas through a lot of effects, I thought instead about what one could do nowadays with a bank of rectangular oscillators and octave dividers, for example in connection with velocity and with envelopes and filters, which in principle work not per voice but globally. This creates an instrument with a very individual character, which also requires a special way of playing.
But I am particularly interested in the topic of physical modeling, since it was one of the last big developments which also relied especially on the interaction of the musician with the instrument. Probably therefore it was replaced by Virtual Analog, which in turn had to give way to the „original“. From my point of view, it is nice to see that there is a manufacturer who is at least reconsidering this issue. In which direction will Nonlinear Labs go?
The renaissance of analog synthesis surprises me again and again. With digital components so many interesting possibilities arise, which are still used far too little.
A preliminary remark on the term „Physical Modeling“: It describes a group of synthesis methods inspired by the behaviour of acoustic instruments. The aim is to digitally reproduce the physical processes of excitation, the development of vibrations and complex interactions.
However, I do not think it makes much sense to try to simulate acoustic instruments as completely as possible. Especially when you are dealing with a keyboard instrument as a playing surface, you will always be disappointed, because a large part of the character of an acoustic instrument is created by how the musician interacts with it. For example, touching the strings of a guitar is such a complex physics and such a direct physical influence that cannot be transferred to other interfaces.
So I won’t take part in the race for the highest realism that also exists for example with sample-based sound generators. However, it is worth experimenting with components that are also used in physical modeling in order to give digital instruments the most expressive behavior possible. We are talking about exciters, resonators, nonlinear components and signal paths, as they can already be found in the C15. If acoustic realism is not a dogma, it can also be combined well with typical „electronic“ components such as oscillators and filters.
As preparation for a future C15 sound engine, a structure with several coupled (delay-based) comb filters was created some time ago. It seems promising to me, but requires further development.
Anyone who knows Native Instruments‘ „Prism“ knows that I have also dealt with modal synthesis. A larger number of narrowband 2-pole filters are used as resonators. We at Nonlinear Labs will continue this experience in the future.
A combination of both approaches, i.e. delays and band filters, is also exciting. A colleague is experimenting with a matrix as a possible core for a percussion synthesizer.
As we are a small team that develops a complex product in many aspects with the C15, new sound engines are not to be expected at very short notice. But perhaps we can offer something in this area next year.
We can’t get to the end of this interview without talking about your customer service and sales model. I would like to comment on the first and describe it as examplary. Short reaction times, short distances… many other manufacturers, including dealers, could take a leaf out of your book. Your sales concept – to do everything directly – is coherent and testifies to being close to the customer. For very different reasons. But now to your sales model, which is not commonplace and which I kindly ask you to explain…
We want customers with very different budgets to be able to play the C15 and to keep their financial risk low. That is why we have developed a flexible hire-purchase model. You can use an instrument with partial payments starting at 2% per month and then become the owner at your own pace. No interest is charged.
You can also return the instrument at any time. We will then refund that part of the payment which exceeds the sum of the minimum instalments of 2% per month. This also applies to the return of fully paid instruments. Returns are processed and offered at reduced prices, in which we also see a contribution to sustainability.
I know I’ve been taking up too much of your time. I could think of other questions, like whether there are plans for developments independent of the C15… but that would probably go too far. I would like to take this opportunity to thank you very much and wish you and Nonlinear Labs all the best and a lot of success for the future.
C15 vs Kontour:
- High resolution (thousands of steps) for velocity, hardware controls and parameters.
- Each of the Macro Controls can address more than 80 parameters.
- Flexible mapping of the eight physical modulation sources (bender, aftertouch, 4 pedals, 2 ribbons) to the four Macro Controls.
- The Oscillators got the Fluctuation feature providing a large variety of noise and random signals.
- Control over the Phase offsets of the Oscillators.
- The new Chirp filter allows to control the bandwidth of the phase modulation.
- The Feedback Mixer got its own shaper for improved control of the feedback behaviour.
- The Feedback paths can be controlled by Envelope C.
- The Envelopes got adjustable Attack Curves.
- The Decay 1 segment of the Envelopes can work like a static „hold“ segment.
- Envelope C gets more flexible, can have a bipolar shape and has more destinations.
- The delay of the Comb Filter can be modulated by the Oscillators (similar to phase modulation).
- The State Variable Filter has been re-designed to be more effective and flexible in band-pass and band-reject modes.
- Stereo Pan for the Oscillators and Filters in the Output Mixer.
- Adjustable key tracking for Stereo Pan of voices.
- Adjustable Stereo spreading of Unison voices.
- The Cabinet now supports stereo processing.
- The Gap Filter now also works as an 8-pole bandpass filter and offers adjustable resonances.
- The Flanger extended by a 4-pole Allpass for phaser-like response.
- The Flanger can be controlled by a key-triggered envelope.
- More control over the stereo processing of Flanger and Echo.
- The Reverb is now based on a new algorithm with a high quality and smooth transitions between different Sizes.
I would like to thank „psv-ddv“ for brainstorming and Peter Grandl of Amazona.de.