Long awaited and much anticipated, Novation's digitally‑modelled analogue synth module is the company's most ambitious offering yet.
Although they established themselves as providers of budget analogue bass monosynths with the BassStation series, Novation have, to date, lacked an instrument to put them in the synthesizer Big League alongside heavyweights Korg, Roland and Yamaha and fellow European upstarts Access and Clavia. While all these manufacturers have brought out polyphonic digital synths utilising physically modelled analogue synthesis, Novation's only foray into sound modelling has been their DrumStation rackmount drum module. Now, however, Novation are launching the Supernova, and for all those people who've been waiting and waiting, yes, it's finally out — first models started shipping around the middle of June.
Based on the custom Analogue Sound Modelling digital technology Novation first developed for the DrumStation, this synth module represents a significant leap in terms of pricing, technology and functionality. As well as being Novation's first synth to break through the £1000 price barrier, it's also their first polyphonic, multitimbral digital synthesizer. Yet at heart it embodies the same enthusiasm for analogue synthesis technology that has always characterised Novation. So will the Supernova light up your sky at night, or is it destined to be eclipsed by bigger stars?
The Supernova is a 3U 19‑inch rackmount module which packs in a traditional waveform‑based subtractive analogue synthesis architecture, modelled digitally in software, 8‑part multitimbrality, an 8‑part arpeggiator, and seven effects for each part in multitimbral mode! It comes in two versions, offering 16‑ and 32‑voice polyphony respectively, with the 16‑voice model being upgradable to 32 voices. The 32‑voice model also has twice the 512 Program and 256 Performance memories of its 16‑voice cousin, though the polyphony upgrade for the latter model also includes the additional memories.
Novation's new module has two play/edit modes: Program and Performance.
With the module in Program mode you can play a single Program at a time, on the user‑selectable Global MIDI channel. Each Program has seven effects, including reverb, chorus and distortion, and can be assigned a preset or user‑programmed arpeggio drawn from a common memory pool. To use the Supernova MIDI‑multitimbrally, you must select Performance mode; this gives you eight Parts, each of which can play a single Program complete with all its effects. You can assign each Part to either the Global MIDI channel, Omni (ie. all), or one of channels 1‑16. As you can also set a key range for each Part, this means you can create anything from a single‑channel split/layer texture of up to eight Programs through to an 8‑channel MIDI multitimbral configuration with one Program per channel.
In Performance mode you can also use up to eight arpeggios at once — one assigned to each of the eight Parts, triggering a single Program with associated Part‑specific effects processing. Individual Parts, complete with any effects processing, can be routed to any one of four pairs of audio outs on the module's rear panel; you can use any of these eight outputs as individual outs by panning one or more Parts hard left or right.
Memories of the BassStation's less than sturdy construction and its rather toy‑like appearance haven't done Novation any favours in subsequent years. However, the Supernova is set to lay those memories to rest. The new module is solidly professional in both appearance and construction, and at the same time sleekly and distinctively stylish, thanks to its metallic blue casing, deep‑blue fluorescent LCD (á la Lexicon) and plethora of red pinpoint LEDs and custom‑designed knobs and buttons. Both the response and contours of the knobs and buttons are reassuringly firm, yet smooth and comfortable, helping to give the Supernova a plush feel.
Novation have also managed to pull off their usual trick of balancing a generous controller‑rich front panel on the one hand, and a clearly organised and accessible layout on the other, with knob and button positioning that's just on the right side of cramped. The user interface has been thoughtfully designed to make parameters as accessible as possible, via a combination of knobs and buttons and LCD menus; LCD‑based parameter editing is straightfoward and fast, thanks to a combination of page up/down buttons and a Fast Data knob plus a pair of increment/decrement Nudge buttons for each of the two LCD rows.
The new module is solidly professional in both appearance and construction, and at the same time sleekly and distinctively stylish.
A Supernova Program's sound sources consist of three main oscillators and a pink noise source. In addition, each of the main oscillators has its own 'virtual' slave oscillator for creating sync effects, and there are also two ring modulators which take their inputs from oscillators 1 + 3 and 2 + 3 respectively. For each of the three main oscillators you can select a square or sawtooth waveform, and set pulse width for the former; Novation plan to add further waveforms in a future update, making them accessible via the Special button in the Oscillators section. You can adjust the octave (+/‑2), semitone (+/‑12) and fine pitch (0‑127) amounts for each of the three main oscillators, but not for the noise source, which is a shame. Incidentally, OS updates are uploadable into the Supernova via MIDI from a sequencer as MIDI files; an uploaded file will be loaded into temporary RAM, checksummed for load errors, then copied into Flash ROM if OK. Internet users will be able to download updates for free from Novation's web site.
The outputs from the (sync'ed) main oscillators, the noise source and the two ring modulators are routed through a mixer section, where each source's level can be set independently. Each level can also be modulated by any combination of five sources, with a separate modulation amount being specifiable for each source. The five sources — LFO1, LFO2, envelope 2, envelope 3, and mod wheel — form one side of a 5x5 modulation matrix, with level, pitch, pulse width, sync, and 'soften' (more on this in a moment) forming the destination parameters. Only the level parameter applies to all four sound sources and the two ring modulators.
The standard oscillator sync effect is available for each of the three main oscillators via the sync level parameter in the modulation matrix, with a mod level parameter also available for each of the five mod sources. However, the Supernova also goes beyond traditional oscillator syncing capabilities with three additional parameters, not part of the modulation matrix and so not dynamically controllable: sync skew amount (‑64/+63), formant width (0‑127), and sync key follow amount (0‑127). Together these three parameters — again, settable individually for each of the main oscillators — further increase the timbral modification possibilities available on the Supernova prior to the filter stage. In essence, the soften parameter rounds off the 'sharp edges' of a waveform, which has the effect of reducing the waveform's harmonic content and creating a 'softer' sound; for instance, a square wave can be reduced all the way down to a sine wave. Soften can be applied independently not only to each of the three main oscillators but also to the noise source. In addition, for each one of the three main oscillators you can set the pitch wheel range and direction (+/‑ 12) and the degree of mod wheel and aftertouch control of LFO 1 modulation amount (‑ 64/+63 in each case).
The output of the mixer section is, of course passed on to a filter section, which consists of a single resonant filter, capable of being driven into self‑oscillation. You can choose between low‑pass, band‑pass and high‑pass filter types and 12, 18 or 24db/octave roll‑off. Currently the filter doesn't have a notch filter type, but Novation plan to add this in a future OS update. Other filter section parameters are resonance and overdrive, the latter being a saturation effect which can be used to create a fuller, warmer sound, and there's also a 2x5 modulation matrix, with filter cutoff point and resonance amount as the destinations, and LFOs 1 and 2, envelopes 2 and 3, and the mod wheel as sources (as in the oscillator section). In addition, you can specify separate aftertouch modulation amounts for frequency and resonance, and the degree to which aftertouch and mod wheel individually will affect LFO2 modulation of frequency and resonance. You can also set a filter tracking amount and a Q norm amount; the latter determines to what extent, if any, higher levels of resonance will produce a louder output signal from the filter.
The Supernova has three ADSR envelopes — one assigned to amplitude, the other two assignable. The attack, decay and release stages each have maximum durations of 20 seconds. You can specify the degree of control that attack velocity has over each envelope's modulation depth (controlling the volume of the Program in the case of envelope 1), the amount of delay for envelopes 2 and 3, and the amounts of mod wheel and aftertouch control over envelopes (again, controlling Program volume).
Each envelope can be set to single or multi triggering (single prevents legato notes from retriggering the envelopes). There are also several parameters, as follows, common to all three envelopes:
- Portamento type (porta or glissando — continuous or semitonal gliding).
- Portamento mode (linear or exponential glide — exponential is used for TB303‑type portamento).
- Poly mode (determines whether or not a repeated note will use the same voice).
- Oscillator mode (percussive or ensemble — determines whether or not the oscillators are reset on note on).
- Glide type (normal, autoglide or eight pre‑glide types).
Regarding the latter parameter, autoglide is used for TB303‑type glides, with no portamento on non‑legato notes, while the pre‑glides provide initial up or down pitch glides, with a choice of two‑,five‑, seven‑ or 12‑semitone ranges, and duration determined by a portamento time parameter.
The LFO section consists of two LFOs, each with speed and delay settings and a choice of square, saw, triangle and sample‑and‑hold waveforms. The Supernova's LFOs can operate at audio frequency rates, a feature which lets you create sounds not possible with the standard LFO speed range. For convenience, an LFO Range parameter lets you set whether the Speed knob will operate over Slow (moderate), Normal (standard synth) or Fast (full) ranges. You can alternatively sync one or other or both LFOs to Internal or MIDI clock, depending on the Global sync setting, and select a 'note duration' for each LFO, ranging from a 32nd‑note triplet up to 12 bars; for internal sync, the rate is set by the Supernova's arpeggio speed, and for MIDI sync by the external sequencer's tempo. In this way you can use the LFOs to create rhythmic effects sync'ed to a MIDI sequence.
Other parameters available per LFO are offset amount, envelope 3 LFO speed mod amount, aftertouch and mod wheel LFO speed mod amounts, single or multi delay trigger, and keysync or freewheel LFO triggering (ie. whether or not the LFOs are in phase and reset with each new note played). Many of the modulation amount parameters mentioned above have a +/‑ value range, so you can determine not only the amount but also the direction of modulation — so, for example, more aftertouch decreases the amount of modulation.
Each Program can draw on up to seven effects modules: distortion, stereo reverb, stereo chorus/flanger/phaser, stereo delay, comb filter, EQ, and stereo panner. At present they're organised in a single configuration, with distortion, EQ, comb filter and panner organised in series, while reverb, chorus/flanger/phaser and delay effects operate as individual send/return effects (with send amounts settable individually) after the comb filter and before/after the panner. However, Novation are planning to add multiple configurations in a future OS upgrade.
For the distortion effect you can simply set drive amount (0‑127) and mod wheel amount (‑64/+63), while for the EQ you can set treble and bass boost amounts (‑64/+63), and for the comb filter, the filter frequency (0‑127) and comb boost (‑64/+63). The send/return effects offer more programmability. In the case of the delay, for example, you can set delay time, feedback amount, stereo width, left/right delay ratio, delay signal high‑frequency damp, mod wheel control amount, and sync. Like the LFO rate, the delay time can alternatively be synced to internal or MIDI clock, with a note length ranging from 32nd‑note triplet to 12 bars. With the panner you can, of course, set a fixed pan position (L63‑R63), but alternatively you can specify a pan type (autopan, tremolo, L‑R, R‑L), pan speed, pan depth, and dry level (controlling the amount of source signal that gets through).
Overall, the effects are capable enough, and definitely a worthwhile addition to the instrument — all the more so for being Part‑specific in Performance mode. While they're not always the richest or smoothest effects you might hope for (the reverb, for instance, has a ringy, warbling quality when it's laid on thick), they integrate with and augment the analogue quality of the instrument to good effect.
As I said earlier, a Performance consists of up to eight Programs, which are assigned to eight Parts. MIDI channel reception and (for the arpeggios) transmission for all eight Parts defaults to the Global channel, as defined in Global mode, but you can assign each Part to a different MIDI channel if you want, for MIDI multitimbral use. There are six settings buttons in Performance mode:
- Tune: allows you to set semitone (+/‑24) and fine‑tune (+/‑63 cents) transpositions for individual Parts.
- Range: assigns key ranges to Parts (within C0‑C8).
- MIDI: sets the send/receive channel for the selected Part.
- Velocity: lets you set a response curve (normal, inverse, crossfade high, crossfade low) or a 'from or to' numeric value (from various preset values — a freely settable value range would have been better), so there's scope for velocity crossfades and switches.
- Output: allows you to route each Part with its effects, as described earlier.
- Polyphony: sets each Part to Poly, Mono or Off.
When you select a Part, using the relevant Part Edit button, all the synthesis and effects settings for the Part's Program are made available via the relevant front‑panel knobs and sliders. You can fully edit the Program in Performance mode within the context of the other Programs assigned to the Parts. Then, when you come to save the Performance, after you've chosen a destination memory and entered a Performance name you're asked whether or not you want to update the Programs. The default value is No, but alternatively you can select Yes or Each. If you select Yes, any edited Programs are automatically written to the relevant Program memories. However, if you select Each you can go through each Program one by one and decide whether or not you want to update it in memory. At this point you can select a different Program memory to Write into; in addition, the Compare button lets you listen to the Program stored in the destination memory — an extremely useful feature.
As with the Program Banks, all Performance Banks are writable, and Novation have left Bank B empty in the 16‑voice model, and Banks B‑D empty in the 32‑voice model, so there's plenty of free memories for storing your own patches and edits.
Novation have trounced the competition by allowing Programs to keep all their effects and associated settings when assigned to Parts in Performance mode. For each Performance Part you can use the assigned Program's effects settings or create your own specifically for the Performance. Unfortunately, however, you can't copy effects settings from a Program into the Performance memory, so 'tweaking' becomes a laborious process. This shortcoming is something Novation plan to address in a future OS update.
You can mute and unmute individual Parts live using the eight Part Edit buttons (a feature which really comes into its own with multiple arpeggios). Just press the relevant button 1‑8 to select the Part, then each subsequent press switches the muted/unmuted status. What isn't possible is muting or unmuting of more than one Part at the same time, which is disappointing for live mixing. However, the Supernova is apparently able to read multiple Part buttons at once, and Novation are thinking of introducing different button‑press options to make multiple Part muting possible.
The Supernova provides 128 preset monophonic and 128 preset polyphonic arpeggios, and allows you to create a further 128 arpeggios of your own (64 mono and 64 poly). In Program mode, each Program can be assigned an arpeggio from this common pool; you can also program the tempo (64‑191bpm), whether or not the arpeggiator is automatically enabled when the Program is selected, and whether or not it's latched (continues to play when you take your fingers off the keyboard). In Performance mode, these settings are per Part within a Performance (except for tempo, which is the same for all Parts), either adopted from the Part's assigned Program or else programmed within the Performance itself.
Other Program‑ and Part‑specific arpeggiator parameters include number of playback octaves (1‑4), gate time (25, 50, 75 or 99%), arpeggio direction (up, down, or as played), key reset on/off, mute on/off, latch type (repeating or one‑shot), note length (32nd‑note triplet to 12 bars), and velocity response to trigger notes. For monophonic arpeggios you can also set how the Supernova handles situtations where the number of trigger notes is greater than the number of different notes in the arpeggio sequence. Another feature of the Supernova lets you transpose any part of a played arpeggio in real time, using notes sent on a user‑programmable Arpeggio Transpose channel; this feature can be turned on or off for individual Parts in a Performance. You can also choose, for each Program and Part, whether arpeggio notes play internally, via MIDI, or both; in Program mode, arpeggio notes are transmitted on the Global MIDI channel, in Performance mode on the channel of the relevant Part. Individual arpeggios can be dropped in and out in Performance mode by muting and unmuting the relevant Parts — though, as mentioned earlier, disappointingly only one at a time.
If the arpeggiator is to be used in conjunction with a MIDI sequencer, the Supernova must be set to External sync so that it will lock to the sequencer. With External sync enabled, the Supernova reads MIDI Start, Stop and Continue commands, along with MIDI clock data. The arpeggiator tempo is clocked to the sequencer's tempo (and therefore to any sequenced tempo changes). However, you can use the Arp Sync function (available per Part in Performance mode) to change the base note length of an arpeggio, so that it's possible to have different arpeggios playing at different rates and looping at different times. In practice, the Supernova's arpeggiator starts, stops and syncs to a sequencer accurately, both in an extended linear play situation and when you have a sequence loop running repeatedly.
Last, but by no means least, user arpeggio programming is done in Global mode. Here it's possible to select one of the 128 available user arpeggio memories, choose a sequence length of 1‑64 steps, and assign a note number, velocity amount and gate time to each step. Providing that the arpeggio you're creating is assigned to the Program in use, you can get instant feedback on your step‑time editing efforts simply by playing trigger chords on your keyboard or looping an arpeggio trigger note or chord in your sequencer as you work.
The Supernova marks Novation's arrival in the big league.
The Supernova marks Novation's arrival in the big league. It's a versatile and feature‑packed analogue‑modelled synth that convinces sonically, functionally and visually. Eight‑part multitimbrality, an 8‑part arpeggiator, seven effects per part, and a maximum 32‑voice polyphony help to put the Supernova on the map as a serious challenger in the 'new model analogue' market. But what really mark it out as a classy synth are its seductively rich, warm, smooth and natural analogue‑modelled sound, and the character, versatility and sheer playability of its Programs and Performances.
Combine all this with an accessible, controller‑laden front panel for easy editing, and a sophisticated MIDI implementation for effective integration with today's sequencer‑based recording setups, and you have a traditionally‑styled synth module for the late '90s and beyond, that will appeal to a wide range of hi‑tech musicians and quite possibly become a workhorse instrument in project and professional studios alike. A cosmic supernova is a short‑lived phenomenon. This Supernova, on the other hand, looks destined be a bright star in the synthesizer firmament for some while to come.
The Supernova is one of the most realistic and satisfying analogue‑modelled synths to come onto the market so far. A bass end with both punch and fullness is one notable aspect of the Supernova's sound, and the module's many and varied bass sounds are one of its strengths. Smooth, mellow, analogue‑style pads and strings are another forté, as are more ethereal and metallic pads, and the fullness of sound that the Supernova can achieve just in Program mode is impressive. The synth's sonic range goes from the charmingly delicate to the raucously brash, taking in everything from cheesy little analogue 'filler' sounds to warm, rounded analogue brass sounds. Analogue‑style electric pianos are also handled well, as are synthesized kicks, snares and hi‑hats.
Overall, 'smooth' and 'mellow' are the descriptions which seem most characteristic of the synth's sound, though it's by no means limited to these characteristics. And while it reminds me to an extent of Roland's JP8000 in character, the Supernova has its own sound.
Although this synth doesn't need to rely on its effects, they make a welcome addition, and 'fuse' well with the synthesized sounds. Novation have hit just about the right balance of accessibility and flexibility with the seven effects, while the module's ability to retain all Program‑specific effects processing in multitimbral mode is a boon — even if it needs to be handled with caution to avoid the effects swamping the sound.
Twiddling Over MIDI
Novation first championed MIDI sound edits with the BassStation, albeit in a limited way. With the Supernova they've really gone to town, making extensive use of both MIDI controllers and Non‑Registered Parameter Numbers to transmit data from all of the module's 28 knobs and 98 buttons, plus all but one of its LCD menu‑based parameters (the Restore parameter in Global mode). NRPNs were devised as a way of extending the MIDI controller map of 128 controllers. MIDI controller numbers 98 and 99 are the NRPN LSB and MSB respectively, allowing for a total of 65,536 controllers! MIDI controller number 6 (the Data Entry controller) is then used to edit the value of the parameter indicated by the NRPN LSB/MSB number. Novation use only the LSB, giving 128 controllers, but they use the data values of some of these NRPN controllers to provide value ranges for multiple Supernova parameters. The entire list of parameters and values takes up seven A4 pages in the manual.
Of course, all this parameter data can be recorded into a sequencer via MIDI, for subsequent playback to the module, or it can be used to edit another Supernova via MIDI, while anyone who wants to develop a synth editor module or a sequencer mixer map will have an absolute field day with the possibilities on offer (Novation themselves will be offering some mixer maps for free download from their website).
Analogue Sound Modelling
16 voices (expandable to 32) or 32 voices
OSCS PER VOICE:
3 oscillators with virtual sync, plus noise generator
12, 18, 24db/octave, resonant; low‑/high‑/band‑pass options; overdrive control
2 assignable LFOs capable of operating at audio rates
Distortion, Comb Filter, EQ, Reverb, Chorus/Flanger/Phaser, Delay, Panning
56 available simultaneously in Performance mode (8 Parts x 7 effects each)
16‑voice model: 512 (expandable to 1024); 32‑voice model: 1024
16‑voice model: 256 (expandable to 512); 32‑voice model: 512
384 patterns (128 preset poly, 128 preset mono, 128 user); each Part has its own arpeggiator in Performance mode
8 assignable polyphonic outs, 24‑bit 128x oversampling delta‑sigma DAC each
In, Out, Thru
All knobs, switches and parameters transmit MIDI controllers and NRPNs; arpeggiator can transmit via MIDI; arpeggiator rate, LFO speed, and delay time can sync to MIDI clock
Built‑in, multi‑voltage, automatically switches for different voltages and currents
3U 19‑inch rackmount; 133mm x 222mm (including knobs) x 483mm (HxDxW)
- Rich, warm smooth sound, with a solid bass end.
- Sonically versatile.
- Controller‑rich, well laid‑out front panel for easy editing.
- Knobs and buttons send MIDI data.
- Crisp, visually pleasing display.
- Multiple effects for each Performance Part
- Multi‑part arpeggiator in Performance mode.
- Eight audio outputs.
- Only one Performance Part can be muted at a time.
- Part edits and select/mute on the eight buttons can be confusing.
- You can't copy Program effect settings into Parts for Performance‑specific tweaking.
A very desirable analogue‑modelled synth module combining stylish appearance and accessible editing with a versatile but essentially analogue sound.
16‑voice model £1299; 32‑voice model £1699; 16‑voice expansion card (for 16‑voice model) £449. Prices include VAT.