Arturia emulate both the classic Prophet 5 and the lesser-known Prophet VS synthesizers in their latest software instrument.
Arturia's previous soft synths — the Minimoog V, Moog Modular V, ARP 2600V, and CS80V — have been generally well-received by reviewers and players alike, so Prophet V joins a respected family of recreations of vintage instruments. However, it differs from Arturia's earlier products, because it emulates not one, but two vintage classics: the Prophet 5 from 1978, and the hybrid analogue/digital Prophet VS from 1986.
Prophet 5 Emulation: Oscillators & Filters
Arturia's literature states that they modelled their Prophet 5 emulation on the Rev 2 and Rev 3 models of the original synth, so the best instrument for comparison would be a Rev 3 Prophet 5 or a Prophet 10, which, if you ignore a handful of enhancements, is just a pair of Rev 3s in wolf's clothing. As luck would have it, I just happen to have a Prophet 10 to hand!
Opening the filters fully on both synths (using both the cut-off-frequency knob and the filter envelope), defeating any modulation, and selecting the sawtooth wave on Oscillator A on both the soft synth and the Upper synth of my Prophet 10, I played middle 'C' on both, and expected to hear an all-but-identical sound. To my surprise, I didn't. The original synth was noticeably brighter, the soft synth sounding just a little 'softer' than the original, and the difference was also clearly visible on an oscilloscope. The square waves and triangle waves were closer to the originals, but all three of the soft synth's waves sounded less bright than those generated by my Prophet 10.
The noise generator on my Prophet 10, while suffering from the cyclic 'thump' that bedevils early Prophets, sounds 'whiter' than that of the soft synth, and a spectrum analyser confirms this, showing that the soft synth's white noise spectrum exhibits a marked fall-off above 5kHz, ending up almost 20dB down at 20kHz. This is equivalent to placing a 10dB/octave low-pass filter across the output! If these results are a consequence of the filter emulation, rather than any significant differences in the spectra of the noise generators themselves, they might explain the cyclic waveforms' lack of brightness when compared with the originals.
Analysis of the range of filter cut-off frequencies is more encouraging. The soft synth has a lowest cut-off frequency of 10Hz, a maximum when controlled by the cut-off frequency knob alone of 8.47kHz, and (when used at 44.1kHz) a maximum when pushed to the limit using CVs of 20.5kHz. This compares well with the Prophet 10, which has a minimum of 10Hz, a maximum when controlled by the cut-off frequency knob of 8.55kHz, and a maximum when pushed to the limit using CVs of... well, I don't know. The last I saw, it had crashed through the Nyquist limit of my analyser and was heading upward toward 30kHz without pausing for breath.
Also good news is the way in which you can 'play' the self-oscillating filter. The filters in both synths produce the same, haunting sound, and track the keyboard so accurately that you can play them just as you would the oscillators. By the way, instead of offering a Keyboard tracking on/off switch, the soft synth offers a knob to vary it from zero to 100 percent. No, it's not original, but it's a well-chosen enhancement.
Envelope Generators & Modulation
Moving on, I measured the fastest transients that the two synths can generate, which revealed some stunning results. The original synth was more than reasonable by the standards of the time, with a fastest VCA Attack of around 1ms, and a fastest VCA Decay of about 9ms. Nevertheless, Prophet V blows these figures away, with a complete AD cycle of just a millisecond or so. This makes it very snappy indeed.
On the other hand, my measurements of the maximum stage times revealed that, in contrast to the Prophet 10's times of approximately 25s, 50s, and 40s (Attack, Decay and Release respectively), the Prophet V's times of 8s, 11s, and 12s preclude many of the long sweeps and gentle effects that you might otherwise hope to coax from it. This is particularly disappointing, because Arturia have in the past (for example, on 2600V) extended the maximum envelope times to make such effects possible.
Next, I tested the transient response of the filter envelope. The Attack is extremely quick (again, approximately 1ms), but I was surprised to find that the tail extends for almost a quarter of a second. What's more, the bump toward the end of the tail is not unique to this particular click; it appears on all of them. As a consequence, sounds that begin with a filter click have a slight squelch. It's not a huge problem, but it's worth flagging.
As has been noted many times before, the combination of a flexible LFO, lots of destinations in the Mono-Mod section, and the memorably named Poly-Mod made the Prophet 5 particularly flexible. Prophet V recreates all of these, but adds MIDI synchronisation of the LFO, and replaces the LFO amount knob with an LFO/Noise control that determines the ratio of LFO and/or pink noise that comprises the modulation signal in the Mono-Mod section. Again, these are well-chosen enhancements that do nothing to detract from the character of the original.
It's notoriously difficult to recreate analogue FM on a digital synth, because the vagaries of analogue oscillators are amplified when they are used in this way. The differences between my Prophet 10 and Prophet V are very noticeable. For example, when increasing the amplitude of Oscillator B's triangle wave when used as the Poly-Mod source for Oscillator A, the overriding effect in Prophet V is of the pitch dropping, whereas on the original it's of the pitch rising. When using the sawtooth or square waves as the modulator, the two synths are similar, but fully clockwise on the soft synth is equivalent to around the 11:30 position (sawtooth) or 1:30 (square) on the original, so the genuine Prophet offers a wider range of effects. There's also a bug here. Switch off all the waveforms on Oscillator B, and the FM effect should disappear. It doesn't, and a seemingly random amount is applied to Oscillator A.
Before leaving the voice structure of the Prophet 5 emulation, I should note that the soft synth retains important facilities such as oscillator sync, glide, the release on/off switch, and unison. To this, it adds new ones such as single triggering and legato-only glide in unison mode. It also allows you to determine the pitch-bend range (which the original synths could not) and to specify an amount of detune in Unison mode (likewise). These are all sensible additions, the combination of which turns Prophet V into a stonkingly good monosynth. What's more, whereas the original synths use last-note priority when played polyphonically and high-note priority in Unison mode, Prophet V offers Reset, Circular, Low, High, and Last note-priority options in both modes, which allows you to tailor it for your favourite playing style.
So How Does It Sound?
I compared some of the original Prophet 5 patches set up on one manual of my Prophet 10 with the factory sound bank supplied with Prophet V. I started with the classic Brass and Strings settings from patch locations 11 and 12. To start with, these sounded a million miles away from the respective patches on the soft synth, but by adjusting the parameters on the Prophet 10 to match Prophet V, or vice versa, I could get the two instruments to the point that — although they were not identical — it would have been churlish to deny that the soft synth was emulating the original.
Greater differences arose when creating brighter sounds, and those with bright, percussive transients. This is when the 'softer' sound of Prophet V became more evident. So, for example, the hammer attacks of electric pianos were slightly softer on Prophet V, as were heavily modulated and sync'ed sounds.
I also tested for things such as zipper noise, and found that the filter quantisation was very similar to that of the original but, when swept quickly, it offered a smoother response for filter blips and splats. In other tests, experimenting with fine control of the pulse width (using the Shift key to adjust the soft synth in smaller steps) demonstrated that the quantisation of the soft synth can be far finer than on the original, which is definitely in Prophet V's favour.
One problem, however, was very apparent. Many of the 'big' sounds in the factory patches distorted nastily. Curing this entailed no more than reducing the level in the Mixer and/or the Master Volume control, so I would love to know whether this was a consequence of my setup or careless programming.
Prophet VS Emulation: Vector Synthesis
There's a lot of mystique surrounding the Prophet VS, but it's a simple synthesizer by modern standards, because the concept of placing four digitally generated waveforms at the cardinal points of a compass and mixing them using a joystick or envelope generator has now become basic fare. Other than this, the VS's signal path is analogue, with each of its eight voices based on Curtis filters and VCAs, articulated by two dedicated envelope generators and modulated by two LFOs.
So, what made the Prophet VS special? As always, the answer is 'the sound'. The VS has a depth that has rarely been equalled and (according to some) never surpassed. However, the build quality, expense, and unintuitive user interface prevented it becoming a world-beater.
Prophet V's Prophet VS emulation includes the 96 ROM waveforms contained in locations 32 to 127 in the original. (The VS also allowed you to create and store 32 user waveforms in locations zero to 31, but the soft synth does not offer this.) Setting up my own Prophet VS and using this as both a sound source and the MIDI controller for Prophet V, I compared the waveforms, both audibly and using the signal analyser. Starting with wave 32, the sine wave, I set up a simple patch with no filtering, modulation, or effects, and listened to both instruments. The sounds were very different. The real Prophet VS demonstrated a lot of harmonic and enharmonic contamination, plus aliasing at high frequencies. Prophet V did not do so, and its output was much closer to a mathematically pure sine wave. Whether you view the soft synth as closer to the ideal, or an inaccurate reproduction of the original is, of course, a matter of taste.
The sounds of the other simple waves were somewhat closer to one another, although the differences were always clear. For example, the sawtooth wave on the original VS is bolder and brassier than the slightly muted version on the soft synth, and the square wave is also brighter. The softness in Prophet V continued to reveal itself as I tested all manner of waveforms, and it didn't appear to be a question of EQ; the soft synth's waves seemed to contain a different balance of harmonics when compared with the original's. Another interesting quirk revealed itself during these tests — wave 75 (Vocal 1) is inverted on the Prophet V's emulation.
Unlike the original, which merely offered a button to select oscillator A, B, C, or D, the soft synth shows all four simultaneously on screen, with the wave number and coarse and fine frequencies permanently visible for each. This is much clearer than on the original, as is the method for balancing the oscillators and creating vector envelopes: just click on Envelope in the Modulation section and then select Mixer, whereupon you can 'pull' the vector points into position to create dynamic effects much more quickly and easily than was possible on the original.
Filters, Amplifiers & Envelope Generators
As before, the reason for the consistent lack of top end in the soft synth again revealed itself when I tested the filter section. As usual, I set the resonance to maximum and swept the cut-off frequency to obtain the filter's range. The Prophet VS's filter has a cut-off frequency well above 20kHz, with numerous artifacts in the pass band. No wonder the original synth has a deep and complex sound! In marked contrast, the maximum frequency of self-oscillation obtainable from Prophet V in VS mode is just a little over 5kHz. This is at a setting of 71; increase the cut-off frequency to 72 and the self-oscillation disappears. Strangely, this doesn't mean that the maximum cut-off frequency is just 5kHz, because the filter actually opens to nearly 20kHz. Clearly, the filter software is buggy, and needs to visit the doctor.
Unlike the original, Prophet V's VS mode also offers high-pass, band-pass, and band-reject modes. These don't always operate as you might expect. For example, past a cut-off setting of 80, the high-pass filter begins re-introducing some of the low frequencies it's meant to be removing. The efficacy of the band-reject filter also deteriorates markedly between settings of 74 and 78, with the rejection band becoming wider and shallower as you increase the value. Most noticeable of all, however, are the problems with the band-pass filter, revealed by passing white noise through the filter in band-pass mode with the resonance set to maximum. (Remember, the filter won't self-oscillate at high frequencies, so the noise allows you to hear what's happening.) Increasing the cut-off frequency beyond a value of 81 introduces all manner of strange, low-frequency artifacts. Surely Arturia couldn't have got the filter so wrong?
As it happens, they didn't. When I first used Prophet V, I conducted all my tests with the Volume control set between half and full. But when checking the results, I performed the tests at a range of Volumes, and at low Volumes the majority of the inappropriate filter responses disappeared. Clearly, that visit to the doctor is urgent.
Unusually, the Prophet VS had no amplifier section as such, with the amplitude, pan, and envelope controls spread across the Voice Control and Envelope Group sections. In an improvement on the original, Arturia have placed these in a single section with a helpful graphic display of the envelope itself.
Prophet VS envelopes (there's one each for the filter and the amplifier) are five-stage affairs with six looping modes and eight options for the number of times the loop occurs. The Prophet V recreates these almost precisely, accurately emulating the fastest Point 0 to Point 1 transient of 11ms and slowest stage times of 40s. Almost. Arturia have missed a trick: a fastest transient of 11ms would make the Prophet VS a very sluggish synth, and experienced programmers know a trick that creates a much faster transient lasting just 2.5ms, but this setting gives a transient of 22ms on Prophet V.
Modulation & Fairy Dust
The modulation matrix on the Prophet VS was another area in which it exceeded the bounds of traditional analogue synthesis. Prophet V takes this a step further, offering an additional modulation source (the amplitude envelope) and no fewer than nine additional destinations that include individual pitch modulations for Oscillators A, B, C, and D. Strangely, it also loses one destination, the chorus depth, but if you count the number of routes available, the VS offers 28, while Prophet V offers a whopping 82, so there's ample potential here for creating expressive sounds or off-the-wall effects, as you choose.
Nonetheless, there are areas in which Prophet V falls short of the original VS specification. Two of these are significant. Firstly, there's no arpeggiator. Secondly, Prophet V has lost the VS's Double mode, which provided split and layer capabilities, with user-controlled detune and programmable delay of the onset of the linked patch. These facilities enabled the VS to make huge, complex noises that rivalled much larger, heavier polysynths, so their absence from Prophet V is a great shame.
Sonic Comparisons
Most soft-synths look similar to the originals, and while the Prophet 5 emulation in Prophet V looks like a Prophet 5, the emulation of the VS does not copy the original design exactly, which is a good thing in my opinion. Too many companies think that they can stick a pretty picture onto some standard digital building blocks, and that people's eyes will fool them into hearing what they see. It must have taken guts to update the Prophet VS's control panel and, while the change of design may be a little disappointing for the purist, it's clearer and easier to program that the original.
So, how does the sound of the soft synth compare to the original? To test this, I again started by comparing the factory patches with Arturia's factory patch bank. The nice chaps who contribute to the Analogue Heaven synth forum told me where to obtain the SysEx file I needed to re-program my VS, and after a few moments downloading and uploading, I was ready to go.
The results were unequivocal. Prophet V is not as bright and — in a way that will appeal to some players and not others — it does not exhibit the aliasing that characterised the VS's sound. This may make it a 'better' synth in an objective sense, but it is not true to the original, so you'll have to decide if this is important to you. The soft synth also lacks some of the punch of the original VS. This may be (in part), because the fastest transient is too slow, but I think that it goes deeper than this. No matter how you program the soft synth, it lacks something when compared with the original. Don't get me wrong — it still sounds very good, and I'm sure that you'll be happy with it. But it's not as good as the original.
Patch Library, Effects & Hybrid Mode
If you have used any other Arturia soft synths, you'll know how the company organises patch libraries and goes about offering chorus and delay, basic MIDI facilities, and MIDI synchronisation of timed parameters. You'll also know how to determine the number of voices, keying priorities, and so on, so I'll say no more about these. Furthermore, Prophet V shares its siblings' ability to learn MIDI controllers in a quick and intuitive manner so, if you have a suitable hardware controller, it can spring to life for real-time programming and tweaking in a way that will forever be impossible on the original.
The effects are basic. The chorus offers controls for rate, depth, wet/dry mix, and three options of complexity. The delay comprises independent delay lines for the left and right channels, with delay time and feedback control for each, plus a global wet/dry mix. But the depth of the chorus is determined not only by the Depth control, but also by the Rate, which is wrong. Likewise, the Delay does not work correctly if the Time is set to its maximum. (If the feedback is anything other than maximum, there's no delayed signal. If it's at maximum, the delayed signal isn't delayed, and doesn't pass through the feedback loop.) This is basic stuff that should have been caught in Quality Control.
Arturia are justifiably proud of Prophet V's Hybrid mode, and at first sight it seems to offer a great deal, combining the Prophet 5 and the Prophet VS to create a mega-synth of great capability. You only have to look at the enhanced modulation matrix — which now includes sources and destinations from both synths — to get a feel for the range of possibilities that this offers. In particular, this allows you to route velocity and aftertouch to nine destinations in the Prophet 5 architecture, turning the soft synth into a bastard child of a Prophet 5 and a Prophet T8. For some players, this may alone justify the cost of Prophet V.
But when you delve deeper, all is not quite as it seems. Rather than let you combine the two synthesizers freely, or even, as in the Korg Legacy bundle, assign both synthesizers in a 'Combi' mode, Arturia provide a patching diagram that allows you to select the nature of the oscillators used in slots A, B, C, and D, and then to determine how you route them along either or both of the 5's signal path and the VS's signal path. This means that you can't create a software Prophet 10, or emulate the Double (split and layer) mode on the original VS, which is a great shame. But despite its limitations, Hybrid mode is a powerful architecture. Let's face it — the ability to pass the Prophet 5's oscillators along two separate signal paths, one with a multi-mode filter, is not to be dismissed lightly.
It would have been easy to describe Prophet V, throw in a handful of eulogies about how it sounds identical to the original synths, and walk away. But Prophet V is not identical to its inspirations; it's a good imitation that offers a number of well-chosen enhancements and an equal number of errors that should have been sorted out. So, what about those differences in sound? Do they matter? I suspect that they don't. Even if they do, history suggests that there will be an update in a few months that will sort out many of the problems. Therefore, a better question is: 'Is it a good synthesizer?'
The answer to this depends, to some extent, on the power of the computer on which you're running it. Prophet V is hungrier than a NYPD cop who hasn't had a doughnut for the whole morning. With just a half dozen notes played, it fully taxes a minimum specification machine, but if you have processing power to spare, it becomes an excellent source of warm polyphonic pads, brittle PPG-esque timbres, complex hybrid sounds, and huge monosynth leads and basses.
To conclude, there have already been numerous soft synths purporting to emulate the Prophet 5, and I have not been impressed by any of them. Arturia have created something closer to the original than any of these, and I am confident that Prophet V will continue to improve. Furthermore, by creating this strange marriage of the Prophet 5 with one of my all-time favourite synthesizers, the Prophet VS, the company have shown the imagination needed to take their offering way beyond those of their competition.
Original Source - Sound on Sound
