The day has come: the most hotly anticipated Eurorack module of the year is here...and it brought a friend with it! Of course, I'm talking about the Make Noise PoliMaths and QXG, which, as I write this are (rhetorically) flying off of Perfect Circuit's shelves.

20hp STAFF PICK

PoliMATHS Eight-Channel CV + Audio Event Generator

$459.00

In stock

QXG is an all-new quad low pass gate/stereo mixer module, and PoliMaths is a new take on Make Noise's iconic function generator Maths. PoliMaths drops the original Maths's voltage processing/mixing/analog logic capability in favor of deeply expanding its capacity for signal generation. Indeed, PoliMaths is eight function generators in a single module, easily capable of complex arrays of control voltage and audio generation.

These modules, along with the recent MultiMod and Jumbler, are among the first entries into Make Noise's proposed New Universal Synthesizer System (or NUSS, for short). The full details of NUSS aren't yet entirely clear, however, some specific features of PoliMaths and QXG clarify a lot about the system's general direction and technical execution. In the bulk of this article, I'm going to describe PoliMaths's core functionality, along with a brief explanation of how it relates to QXG. I'll also share a couple of thoughts/ideas about NUSS's general concepts, so far—but, perhaps, we'll come back for a more extensive discussion of what NUSS is and how it works in the not-too-distant future.

PoliMaths is a Function Generator

You can think of PoliMaths like a supercharged version of the original Maths's function generator channels. Where the original Maths could generate two simultaneous functions, PoliMaths can generate a staggering eight. Like the MultiMod and Jumbler before it, PoliMaths is focused on harnessing multiplicity through a unified user interface. That is to say, while it is, technically, composed of eight distinct function generators, they share a set of macro-level parameters, and cannot be utilized fully independently from one another. As we go, though, you'll see that it offers several interesting methods for interrelating channels, jumping between them, and applying incremental changes that affect each channel with slightly different levels of intensity. So, first, we're going to talk about the basic parameters themselves and how they impact a single channel. Then, we'll expand to discuss the various ways that multiple channels can be addressed/manipulated relative to one another.

The basic parameters on PoliMaths's left side should be familiar to those who have used Maths. Individual Rise and Fall knobs determine the timing of the function's attack and decay stages; the Curve parameter takes you from logarithmic to linear and extreme exponential envelope curve shapes. Strength, as found on other Make Noise releases such as 0-Ctrl, impacts the total output amplitude range of a function generator. Cycle, as you might guess, causes the function generator to retrigger upon completing its cycle, turning it into a looping source of continuous modulation.

×

So, what's up with the right side? Well, in short, it allows you to add continuous fluctuations to the function's output, essentially imbuing your function generator with more oscillator or LFO-like characteristics. The Osc control determines the amount of this "oscillation" that is added to the function generators' outputs; the Shape control determines the shape of the oscillations, offering a continuous blend of saw, triangle, and ramp shapes (with noise/random shapes available by applying an additional offset to the Shape CV input). Rate, fittingly, determines the rate of these oscillations, ranging from slow sub-audio frequencies up into quite high audio frequencies. In low-frequency range, you could use this feature to, for instance, add a sort of "tremolo" or "fluttering" by applying the output to the control input of a VCA or lowpass gate. Depending on the selected shape for the oscillations, this could also result in delay-like effects. At audio rates, that same patch could be used for dynamic amplitude modulation. Applied to other parameters, of course, all sorts of interesting timbral changes could start to unfold.

The function generators can be made to cycle by using the Cycle tactile switch and corresponding gate input. There are, presently, two cycle modes: Cycle All, which turns all of the channels into free-running cyclical function generators, and Follow the Leader, which chains the channels in cyclical round-robin fashion, with each channel triggering its nearest neighbor upon completing its own cycle. This even allows for multiple "chains" of "follow the leader" to occur simultaneously—enabling multiple cascading modulation chains to play out at once.

Additionally, all of the parameters we've discussed have dedicated CV inputs, imbuing PoliMaths with a significant degree of dynamic potential.

It's important to note here that PoliMaths is not exactly an "upgraded" or "new" version of the original Maths. PoliMaths is, in essence, a signal generator—and has no "standard" utility as a signal processor. So, you cannot use PoliMaths to mix signals; you cannot use it as a slew for external signals; and you cannot use it to perform the same types of waveshaping, filtering, or logic processes that give the original Maths its Swiss-army-knife appeal. However, PoliMaths is—as its name implies—still quite adept at a large number of common modular synthesizer tasks/utilities. I'd recommend not comparing it too closely to its similarly-named ancestors: while there is some overlap between these devices, they have different specialties and can even play quite nicely together. So, don't go throwing out your "old" Maths; it still has a useful and interesting role to play in a post-NUSS world.

In order to dive more deeply into what makes PoliMaths especially interesting and unique, we talk about how it handles macro-level control.

PoliMaths on a Macro Scale

We'll start by stating that PoliMaths has, presently, three primary "Span Modes" in which it can operate, selected using the Mode tactile switch. These modes are referred to as Channel Index, Round, and Parallel—and, depending on which you have selected, the behavior of the Activate input and Span parameter will change. Let's describe them each.

In Channel Index mode, the Span control can be thought of as a means of addressing the array of function generators in order to "target" the one you want to trigger. With nothing patched to the Activate gate input, the targeted channel will trigger instantly upon the Span value reaching them. In this way, you can use the Span control to "strum" across the channels, activating them one at a time. Likewise, you could use an incoming control voltage to modulate Span, thereby activating channels in accordance with the shape of the incoming voltage. You could use a sequencer to create repeating patterns of activations; you could use a random voltage generator for unpredictable activations; you could use an LFO to sweep through channels, activating them in predictable, repeating patterns.

×

When something is patched to the Activate input, the targeted channel is not triggered unless a trigger is received at the Activate input. This can be useful in several situations—for instance, it enables repeated activation of a single channel. Likewise, if modulating Span with a continuous function, the Activate input can act like a sort of "sample and hold" for the Span parameter, ensuring that channels only activate exactly when you want them to—perhaps because you want to align activations to a metric grid, or perhaps for some other, more unusual purpose.

In Round mode, channels activate in round-robin fashion when a trigger is received at the Activate input. With Span set just clockwise from center, the activated channels will increment by 1 with each successive activation, in typical round robin fashion. Turning Span farther clockwise, the next channel to be activated will increment by higher and higher numbers, creating activation patterns where individual channels are triggered in changing order. With Span counterclockwise from center, the same behavior is observed, but with channel number decrementing rather than incrementing—essentially employing a round-robin algorithm that counts "backwards." Round mode finds special utility when using PoliMaths as a sound source...a special case that we'll touch upon specifically later in the article.

In Parellel mode, each channel is triggered by an internal "clock divider," which uses the Activate input to set the base clock rate. As Span is turned to the left or right, the module consults an internal process that maps Span to different variable clock divisions for each output. A logic table is included in the manual which describes how divisions are mapped to the various outputs for different Span values, but in short, it can be thought of as a way to place progressively slower divisions toward the left or right side of the module. All in all, it's a powerful way to do eight-channel triggered function generation, and one could go far using it as a core means of rhythm generation within their system.

Spread & Per-Channel Behaviors

×

Perhaps one of PoliMaths's most interesting feats has to do with the implementation of its Spread parameter, as well as the module's CV inputs as a whole. First, we'll say that some control voltage inputs on PoliMaths are global, affecting each channel instantaneously and equally to one another. The Curve and Shape inputs, for instance, always immediately impact the behavior of every channel. The other inputs, though, are a different story. The five other CV inputs (Rise, Fall, Strength, Rate, Osc) each feature apparent attenuverters with gold polarity indication legends. These parameters each have a special relationship with the Spread parameter, and operate somewhat differently from typical Eurorack control voltage input + attenuverter pairs.

The idea is this: if you leave it such that nothing is patched into these inputs, the attenuverters serve to determine how strongly the Spread knob + CV input impact the core parameter in question, and in what direction that impact is applied. You can broadly think of Spread as being a means of applying a channel-number-based weighting to each of these core parameters. The "attenuverter" allows you to dial in the intensity of this channel-based weighting, with the Spread knob itself affecting all such parameters in concert with one another. For instance, you could make it such that the fall time is quite fast toward the left half of the module, but gets slower toward the right; you could make it such that the Oscillations are faster toward the right and slower toward the left; and, you could use the Spread knob and CV input to alter the strength of these weightings dynamically. Interaction with the Spread parameter is continuous: if you alter the Spread amount or the attenuverter-based weighting values, you will hear the impact of these changes instantaneously.

×

Something even more interesting and unusual happens when you patch an external control voltage into any these five Spread-enabled parameters' CV inputs. The parameter in question is disconnected from this internal Spread-based control, and the effect of the attenuverter no longer applies a channel-number-based weighting to the parameter. Additionally, the impact of the attenuverter and corresponding CV input will not be heard immediately: instead, control signals are, in a sense, "sampled" and distributed to a channel (or channels) only when the next Activation event occurs. In this way, each channel of PoliMaths could have a drastically different set of parameter values from one another, each arbitrarily defined—and not necessarily applied as a channel-number-based gradient, as is the case using the Spread functionality alone. Make Noise refers to this concept as Modulation Dissemination in the PoliMaths manual.

The astute among you might note that this type of "fanning out" of an essentially monophonic control source to a series of distinct, autonomous, similarly-structured entities is a common part of several other instruments and general electronic music processes. This type of data distribution is not dissimilar to approaches seen in Harald Bode's Warbo Formant Orgel, or in Buchla's early approaches to "polyphonic adapting," as seen in 1970s 200 Series devices such as the 264 Sample & Hold / Polyphonic Adaptor, the 266 Source of Uncertainty, as well as the 237, 238, and 219 keyboard controllers. These instruments have inspired other modern devices, such as Intellijel's Shifty—esssentially, using a series of sample and holds to sample & distribute an incoming control voltage to several destinations in order, allowing for a modular approach to polyphonic voice allocation/data distribution. Similar approaches are found in other common polyphonic contexts. For instance, many "cloud-based" granular synthesis implementations use a similar data distribution scheme. Seldom are the parameters of individual overlapping grains updated simultaneously; it's perhaps more common for control values for each of the grains' parameters to be "sampled" when a new grain is instantiated, and to remain unchanging until the grain in question has ended.

So what can this mean in the context of Polimaths? Well, one obvious (and, to this writer, somewhat mind-blowing) possibility is to send a 1V/Octave control signal from a controller into the Rate CV input with the Rate attenuverter turned all the way up. Send your controller's gate/trigger output to the Activate input. Set Mode to Round, and set Span just barely clockwise from 12:00. Turn up Oscillations. When you play a note on your controller, it sends a new control voltage value to the Rate input; that value is immediately sampled by virtue of the trigger sent to the Activate input. Likewise, Round mode with the described settings ensures that the next channel in line is triggered, with the sampled Rate CV input used to define the instantaneous Rate of the oscillations. It so happens that, when its associated attenuverter is turned up to 100%, the Rate CV input tracks at 1V/Octave. Congratulations: you've turned Polimaths into an honest-to-goodness polyphonic synthesizer, with the possibility for per-voice audio output (or the option of Submixed groups of outputs—look up Submixing in the product manual, as we will not cover it in this article).

As part of this experiment, you could apply other information to different parameters. Perhaps you would want to apply a keyboard's velocity value to the Strength parameter; perhaps you would want to trigger a random voltage generator with each keypress and apply that to the functions' stage duration controls. Perhaps you want to use something other than a keyboard to control your peculiar modular polyphonic synthesizer: use random generators, cycling functions, sequencers, any of the typical CV and rhythm generation tools you have lying around. Of course, the possibilities are vast.

A New Direction

There are a couple of other features I have not directly addressed. The Accumulate input, for instance, is a means of selectively holding activation signals until a specific desired time. So, triggers to Activate are not immediately deployed: they are stored until such point that a trigger is received at the Accumulate input, activating all previously-triggered inputs simultaneously (and "disseminating" any control voltages present at these prior activations to the appropriate channels, as well). We also haven't discussed the Reset intput: its behavior changes depending on the Span mode, either resetting a specific function generator or resetting the selected Span algorithm's behavior to an initial state. Likewise, we haven't touched on PoliMaths's several behind-the-scenes setting controls, including options for selecting between unipolar or bipolar oscillation and output submixing. We also haven't deeply discussed Cycling behavior, nor details about the Channel Index output's behavior. No doubt we'll return to these ideas in the future, when more details about the NUSS series as a whole are understood.

While we don't quite yet know the nature of NUSS as a whole, we do know that a multi-channel sound generator is coming soon. Indeed, even as I write this, Make Noise's founder Tony Rolando is on a tour around the country showing the NUSS system complete with this new sound source, purportedly called MultiWave—an eight-voice oscillator that utilizes many of the "NUSS" concepts we've discussed. We also know that the "Channel Index" signal produced by Multimod and PoliMaths can be used with the PoliMaths Span input in Channel Index mode, allowing you to chain multiple PoliMaths and ensure that the same channels are activated at the same time on each. It seems that MultiWave will behave similarly, either "following" a MultiMod or PoliMaths or, perhaps, "leading" a PoliMaths via its own incoming-data-to-channel-index-control mode(s).

So, we don't know all of the details—but we can start to see some ideas emerging that will likely be found throughout upcoming modules. The idea of Spread as an interesting macro control method; the implementation of these multifunction, context-based CV inputs; the idea of different Modes for channel activation; the Channel Index signal as a means of synchronizing multiple devices; etc. One can imagine how concepts and structures like these might play out in other devices—both the soon-to-be-released oscillator, or even other types of devices altogether.

×

I should also note that we haven't yet addressed the behind-the-scenes patching possible between the NUSS-oriented modules that we do know about already. If you look at the back of your PoliMaths or QXG, you'll see small ribbon headers. These can be used to "pre-patch" groups of signals from one device to another. So far, the most obvious use cases are to connect PoliMaths's eight outputs to the eight audio or control voltage inputs on a pair of QXGs, allowing inter-module connection without the need for any front-panel patching (quite the boon in contexts where macro groupings of eight channels seem to be the norm). One might think of this as somewhat similar to the "Polytip" connections in recent Tiptop Audio modules, which send multiple channels of signals over front-panel USB-C-style connections. But, unlike these Tiptop devices, Make Noise have tucked away these types of connections behind the panel, leading users toward more long-term normalled configurations of modules. We suspect that this is a wise move, given the apparent direction that NUSS is heading. We anticipate, naturally, that the NUSS sound generator will also feature such an output header for direct connection to QXG—but beyond that, time will tell.

All of that said, I hope that my quick, humble description of PoliMaths has made it clear that it isn't just meant to be part of a larger NUSS ecosystem. It's a densely functional module that could serve Eurorack systems of any style. It could be a source of many channels of evolving, inter-linked modulation; it could be a sound generator; it could be a rhythm generator; it could be a way of selectively deploying targeted modulation to a set of destinations. Part of its brilliance is that, while it appears to be an integral part of a newly-evolving ecosystem, it fits equally well into existing systems. And, in either case, we suspect it will lead to some new and unique ways of interacting with sound.

You Might Also Like

10hp

Jumbler Analog Signal Combiner + Distributor

$259.00

In stock

10hp

MultiMod Complex CV Generator + Voltage Processor

$349.00

In stock

0-CTRL Desktop Touch Keyboard / Sequencer

$399.00

In stock