Italian synthesizer brand Frap Tools has been a force in the modular synthesizer world for over a decade, earning a reputation for uncompromisingly high-quality, functionally dense modules with a design sensibility often leaning toward Buchla and Serge. Under the offshoot brand Frap Audio, they also ventured into pro audio with the Dynamics 2806 500-series compressor. Most recently, however, they've completed their first polyphonic keyboard synthesizer: Magnolia.

We caught up with the company's Simone Fabbri and Giovanni Grandi to talk about the journey it took to realize Magnolia, the design decisions that went into it, and the overall trajectory of Frap Tools as a synthesizer company.

Origins Of Magnolia

Eldar Tagi: Hi folks—great to have you back for another interview! The spotlight this time, of course, is on Magnolia, your new boundary-blurring bi-timbral analog through-zero FM polysynth. Given your strong roots in the modular world, we’d love to hear how this project came to life. How long has it been in the works? What sparked its development, and what ideas or needs inspired its creation?

Frap Tools: Over the past ten years of designing modules we simultaneously developed all the building blocks required to manufacture a complete instrument, and the cool thing about those “blocks” is that these have been engineered by us from scratch: I think that this is the crucial point. At Superbooth, many people asked on which chip the oscillators or the filters were based, and well, they’re based not on a chip but on our own schematics. This small difference allowed us to have something to say in the crowded world of keyboard synths, because the resulting sound for sure is by nature different from any of them – you may like it or not, that’s for sure, but at least it can’t sound as another instrument on the market.

On top of that, after designing and playing open-architecture instruments like Eurorack modules, we began to appreciate the beauty of closed-architecture designs, which require more care into what to make explicit, what to hide under the hood, and what to exclude at all. I would say that the “closed synthesizer” that we’ve left behind when we approached the modular synths has in fact lied in front of us all the time.

Magnolia Polyphonic Analog Thru-Zero FM Keyboard Synthesizer

$4,829.00

In stock

ET: Can you talk about the name Magnolia—what does it represent for you, and how does it connect to the instrument's identity?

FT: We wanted our instrument to be perceived as organic and lively on one side, and timeless on the other. We discussed many names from the natural realm, like mountains or minerals, but Magnolia won because it embodies two key elements: a living being, and an evergreen tree. Plus, it sounds so good!

ET: This shift from modular to integrated format must have brought about some design recalibrations. When building a self-contained instrument—where architecture, interface, and user experience are tightly fused—did you find yourself making different decisions than you would in a modular context? Were there new freedoms or unexpected constraints?

FT: Oh yes. Every challenge we faced can be boiled down to two main factors: different architectures, and different traditions.

The first point is that the modular synthesizer is an open system that is up to the performer to close, and there is an implicit agreement between the player and their instrument on the fact that some connections may produce "bad" sounds (or no sounds at all). On a keyboard synth there is no such agreement: only a selected number of possibilities must be made available, whereas others must be kept inaccessible (for example, sending gates into the audio path). Many of such conventional connections are the results of years of "natural selection," which brings us to the second point.

Analog keyboard synths have a very strong conservative tradition in terms of interface and user experience. Many professional and amateur keyboard players own multiple synthesizers and expect to transfer their skills from one to another almost seamlessly. (There are of course some cases of experimental interfaces, but everything considered, they are a small percentage of the whole set of analog synths ever produced.) We wanted Magnolia to fit within the "orthodox" synthesizers, so we had to accept the fact that our Eurorack interface language was not appropriate and elaborate a new, consistent one.

Architecture and Sound Design

×

ET: Magnolia is polyphonic, but not exactly in a traditional subtractive sense. Each of the instrument's voices is built around a Brenso-inspired dual-oscillator core—a complex architecture rarely seen in polyphonic synths. What do you find most compelling about this architecture in a polyphonic context, and what kinds of sonic or musical possibilities does it open up?

FT: We think that the complex oscillator design in the style of the Buchla 259 offers an excellent balance of features and accessibility for sound designers. It has enough ways to manipulate a sound's timbre without being distracting to use. However, it makes sense only if it is supported by an adequate modulation matrix, just like in the modular world. For this reason, we think that a huge benefit of a keyboard-driven complex oscillator are polyphonic modulation sources like velocity and aftertouch: being able to control its parameters independently per voice is an incredibly expressive tool, and the response is very natural and organic.

ET: An important feature of Magnolia's sound is the thru-zero frequency modulation. I think it is fair to say that analog TZFM has gained a particular rise with the resurgence of modular synthesizers, and in the context of standalone keyboard synths—Magnolia is effectively the first of a kind. What makes this synthesis approach particularly compelling for you to work with, and why did it feel important to bring it into a polyphonic keyboard instrument domain now?

FT: We think that one of the reasons FM synthesis is generally considered "harder" to learn than subtractive synthesis is that it is always first presented as a mathematical model rather than a sonic landscape, which is not the case for subtractive synthesis, for example. Many musicians learn subtractive synthesis without having a clue of its mathematical foundations, which they can delve into at a later stage. And that's where we think that "modular-style" analog FM can be interesting for musicians: it bypasses the concepts of operators, index, and precise ratios of digital linear FM, leaving just two oscillators, one modulating the other. It is a more primitive, almost feral way to do FM: it is possible to explore beatings, in-between values, and imperfections that are impossible to achieve on a precise digital synthesizer. However, we should all agree that from a mathematical standpoint TZFM is nothing new in the world of FM: it is just a linear FM correctly implemented in the analog domain.

ET: Another big part of Magnolia's sound is in its analog wavefolding and shaping circuits. If I understand correctly, this also comes from Brenso. Are there any specific differences in the application or performance of these circuits here?

FT: The core is the same and so its sonic character, but there are some substantial differences in how we implemented the controls. Once you are no longer able to modulate every parameter with the CV you like, as in a Eurorack system, many of Brenso's iconic controls become slightly redundant, and so we chose to simplify its design by assigning to each of Magnolia's oscillators a limited set of "defining attributes." One became the "West Coast" one, with TZFM and wave folder, and the other became the "East Coast" one, with fine tune and PWM.

More specifically, some differences are minimal (for example we reduced the wave folder's range and removed its symmetry control); some others are more relevant, for example the wave shape selection: instead of having a complex shaping section as on the Brenso, we preferred to equip each oscillator with a one-knob continuous shape control.

We were in doubt as for where to put the Sync control: logically, the sync sound is often associated with the classic Moog-like subtractive synthesis; however, we found out that it sounds gorgeous when FM'd or folded, so we put it in the "West Coast" oscillator.

×

[Above: The insides of the Frap Tools Magnolia.]

ET: The filter section on Magnolia also derives inspiration from your Eurorack filter module—Cunsa. Can you talk about how filtering is implemented in Magnolia?

FT: This is the perfect example of what we were talking about before. Our Cunsa module is a very deep filter with plenty of colors available. Its Character control adds a crucial variable and opens up new combinations together with the input gain and the Q controls, which need the kind of experimental attitude that modular synth players have. However, on a keyboard synth there is no room for redundant features, especially because the filter's input is always the same—as opposed to the infinite possibilities of the modular world. We thus chose to design a streamlined version of Cunsa for Magnolia, which picks only the best of its features. In the case of the high-pass filters, we altered the original design quite substantially, because we had to find a filter topology that could work well with FM patches as well as subtractive ones. It was quite a time consuming task!

No Menus — No Problem

×

[Above: detail of the first Magnolia prototype.]

ET: One particularly impressive aspect of Magnolia is how elegantly you've handled modulation assignments—offering deep flexibility without relying on screens or menu diving. What guided that decision, and how did you balance immediacy with complexity in the modulation architecture?

FT: The absence of large displays was non-negotiable from the beginning, mostly to preserve the product's longevity and to encourage a hands-on patching style. The push-and-rotate combination has been around since the early days of virtual analog synthesizers, and we liked it quite a lot because it reminded us of the modular world: it can easily become an instinctive gesture for the player, almost like physically drawing a virtual cable from source to destination. Adding the Value button to monitor the destination amounts and the Toggle button to turn them on and off allowed us to show all the details without having to use a massive display.

ET: When you showed the instrument at Superbooth 2025, it was still in development with a few features missing. You've mentioned an effects section, and a distortion circuit. Can you talk about the final arrangement you settled on? And as an extension to this question, what other significant changes to the instrument have you made since?

FT: We wanted our effects to be the icing on the cake and complement the analog dry sound without being too much of an "ear candy." (Sorry for the abundance of dessert-related metaphors.) We chose to feature a distortion section that is based on our CGM analog Eurorack mixer with a controlled feedback circuit that can be anything from subtle enhancement to complete sound annihilation. After that, we added just two choruses and two delays tailored to the sonic character of Magnolia: such effects are quite useful and can enrich a patch without being too distracting, unlike complex shimmering reverbs or phasers. The most important thing is that the dry signal is always analog.

Other things that have changed from the prototype are the oscillator's range, some printed graphics, and the preset management system, which now uses three-digit decimal numbers instead of the former 8 groups / 8 banks / 8 presets. Other news are actual additions to that prototype, like a Macro modulation knob that replaced the Master Tune knob, and a sophisticated keyboard-based polyphonic random value generator that we named "Polymove."

×

ET: Your designs often draw from design philosophies typically associated with the so-called West Coast synthesis paradigm, which tend to favor non-keyboard control methods. In the description of Magnolia you explicitly imply that it falls closer to the more recent takes on creating a sort of hybrid instrument merging elements of East and West Coast approaches. So my question is this: What led you to choose a traditional keyboard interface for Magnolia over something else like a touchplate interface? Did you consider any alternative forms of interaction during the development process?

FT: No, we never considered other controllers, even though it is a topic that we bring up on a regular basis. As a matter of fact, the whole design process was the opposite: we started from the keyboard and then incorporated selected sonic elements of the West Coast modular tradition of the 1970s. Magnolia was meant to play "regular" music with chords and melodies from the start.

Magnolia: A New Chapter

ET: One aspect that's often overlooked when discussing instruments is the role of time—specifically, how extended engagement with an instrument reveals its subtleties, quirks, and deeper layers. As someone who has likely spent more time with Magnolia than anyone else so far, are there particular features or behaviors you hope users will attune to as they spend more time with it?

FT: We tried to make Magnolia as self-explanatory as possible, but there might be a couple of things that need a little extra care: the modulation matrix and the gain staging.

We already talked a bit about the modulations previously, so let us add that we had to make some design choices that sacrificed immediacy in favor of consistency: for example, there is no pre-programmed vibrato-over-modwheel feature (one needs to assign the LFO to the pitch and then the modwheel to the LFO amplitude), or a button for filter tracking (one needs to assign the keyboard track modulation to the filter like any other destination).

The gain staging is important because there are a few points where loud and quiet signals bear significant sonic differences, and we recommend musicians to treat amplifiers and mixers as nuanced parameters rather than just on-off circuits.

×

[Above: Behind the scenes at Frap Tools during Magnolia development.]

ET: Magnolia clearly represents a new chapter for Frap Tools. Do you see it as a singular project, or the beginning of a broader exploration into standalone instruments?

FT: We would love to explore the world of standalone instruments and we have plenty of ideas for sure!

More from Frap Tools

12hp

Bagai Analog Sample + Hold

$345.00

In stock

42hp

Cunsa Quad Multi-Mode Filter

$1,168.00

In stock

30hp

Brenso Complex Oscillator

$765.00

In stock