Cubase 11’s new SuperVision plug‑in makes it easier for Pro and Artist users to trust their ears.
Gavin Herlihy’s ‘Why Your Ears Are Lying To You’ article in SOS May 2021 (http://sosm.ag/why_ears_lie) was a sobering reminder that you can’t always trust what your ears are telling you. This is why it’s helpful when recording, mixing and mastering audio to have good metering options to support your listening‑based decision making, and the Pro and Artist editions of Cubase 11 added a powerful new audio analysis and metering plug‑in: SuperVision. I can’t hope to cover its 18 different analysis modules in detail here, but by way of an introduction, I’ll consider how it might help with two common practical mixing tasks: setting your master bus processing and checking for lead vocal masking.
First, though, I need to go over some basics. SuperVision is an insert plug‑in, so you can use it to visualise your audio at any point in a signal chain. In addition, you can also use multiple instances in a project. By default, a new instance opens with a single Level module. However, the tool‑strip at the top includes a drop‑down menu to select from any of the 18 module types. The two top‑right buttons allow you to split the display of the selected module vertically or horizontally, to add a further module, and you can repeat this process to display up to nine modules in a single instance. You can resize the overall plug‑in window and customised layouts of modules can be saved as presets for instant recall.
The ‘cog’ button opens a further options page for the currently selected module (I’ll discuss some examples below), and other useful buttons in this tool strip allow you to pause, hold and reset the values in the current module. Holding Alt/Option while clicking on the Pause or Reset buttons applies the action to all modules in the plug‑in. Usefully, you also have the option of resetting a module’s display each time playback commences.
Master Bus Processing
While not everybody is a fan of mix‑bus processing, it’s common practice either to mix into a processing signal chain on your master bus or to add some master‑bus processing in the final stages of a mix; EQ, dynamics processing and saturation are all commonly applied. Precisely how much of this processing (all, some or none) you actually ‘print’ to your final mix is a broader question, but if you’re going to export your mix with the processing in place, it’s helpful to have as much information as possible about the changes that result from this processing. A number of SuperVision’s modules can help here, but I’ll focus on a combination of two: Loudness and Wavescope (shown above).
The Loudness module provides a pretty comprehensive numerical and visual summary of the loudness of an audio signal. The exact display can be customised in the Settings window and in the screenshot I’ve selected LUFS and the EBU +18dB display scale (I’ve left all other settings at their defaults). As well as a True Peak value, you get the three different standard time‑based averages of loudness: Momentary Max, Short‑Term and Integrated, measured over 100ms, three seconds, and the whole playback period, respectively. Typical targets for streaming services are ‑1.0dB for True Peak and ‑14 LUFS for Integrated loudness but, whatever loudness targets you’re aiming for, SuperVision lets you see if you are hitting them. The Range values and visual display show the dynamic range, with higher values indicating greater differences between the loudest and quietest sections of your audio. This can be a useful additional indicator of the impact of your dynamics processing.
The Wavescope module provides a real‑time waveform display for the audio signal. In the Settings, you can adjust the meter’s integration time (how long, in seconds, the ‘window’ is that you are seeing at any one time) and scale the display in various ways. This module does a pretty simple job but when combined with the Loudness module, and with identical instances of SuperVision placed both directly before your master bus processing and directly after, it can be really helpful in informing your processing decisions.
The first screenshot (above) shows an example. Top‑right in the MixConsole, you can see instances of SuperVision placed before and after my master bus processing chain. Comparing the two SuperVision instances, the numerical loudness data and visual waveform changes introduced by the processing are obvious. Such visual feedback is undoubtedly useful in terms of identifying any over‑ambitious loudness gains or potential destruction of transients. Of course, this approach isn’t just useful on your master bus; the feedback can be just as helpful on a drum or lead vocal bus.
Spectrum Curve and Multipanorama can provide visual assistance if you suspect other instruments are conflicting with the all‑important lead vocal.
Detecting Vocal Masking
Another two‑module combination, Spectrum Curve and Multipanorama, can provide visual assistance if you suspect other instruments are conflicting with the all‑important lead vocal. To make the visualisation easier, I’ve set up a stereo bus for my lead vocal that receives the vocal plus any of its send effects (reverb, delay, etc). In this mix, the main instruments ‘competing’ for frequency space with my vocal are some guitar and piano parts, and I have routed these to a further bus.
The second screenshot shows how you might configure an instance of SuperVision on the vocal bus described above. In this case, I’ve activated the external side‑chain input, and specified the guitar/piano bus as its source. The upper panel shows the Spectrum Curve module, and in the right drop‑down menu you can choose what frequency curves are displayed. In this example, I’ve selected both the main channel (in this case, the vocal, shown in blue) and the side‑chain input (the guitar/piano bus, in white). Within the Settings panel for the Spectrum Curve module, I’ve activated the Masking option. With this engaged, SuperVision highlights (using blue vertical bars) the frequencies in the main signal that are most likely to be masked by the side‑chain signal — for my example, this is the 300‑600 Hz range, making this an obvious target for some modest EQ cuts on the guitar/piano bus, if I find that I need to give my vocal a little extra space in the mix.
The lower panel shows the Multipanorama module. This X‑Y display maps the audio intensity of a signal across the stereo image (horizontal axis) and frequency range (vertical axis). You can flip the axis and change the colour of the display in the module’s Settings panel. At present, for this module, you can only visualise one audio signal at a time in the display. Still with the module selected, you can use the top‑right drop‑down menu to decide whether the display shows the main audio channel or the side‑chain input, so you can easily flip between the two signals. Not only can you see the frequency ranges where they might clash, but you can also see whether those potential clashes apply to their stereo placement. This might inform your panning decisions for the instruments or, if you are feeling particularly brave, help with any Mid‑Sides EQ you might like to apply to the instrument bus, perhaps making frequency space for your vocal just in the centre of the stereo image, while being able to leave the instrument’s EQ intact towards the sides.
Of course, it would be great if the Multipanorama module could display both the main and side‑chain signals at the same time and better still if they could be colour coded. Perhaps that’s something the boffins at Steinberg can add at some point. It’s also worth noting that this module is very useful for judging the impact of any multiband stereo image processing you might apply using Cubase 11’s new (for Pro and Artist) Imager plug‑in... but that’s a topic for another day.