The FabFilter-MB plugin is complex at first, but the tools it offers are incredibly useful and versatile. With ti we can perform 4 types of dynamics processing, as well as introduce equalization, mid-side stereo imaging, and advanced frequency-specific and image-specific external side-chain processing.
The FF-MB is a multi-band dynamics processor - with it you can create up to 6 bands, each can be downward compression, upward compression, downward expansion, or upward expansion. This may seem convoluted by I’ll explain each processing type in greater detail in future chapters.
The crossovers can range from 6dB to 48dB, and we can change this value by clicking the band icon, and dragging up or down. Additionally, these bands double as an EQ, letting us amplify or attenuate the ranges prior to any dynamics processing.
Let’s take a listen to a combination of each processing type before we delve deeper into them in the next few chapters.
Downward compression is the default processing type - it’s created by combining the compression function with a negative range. As a result, whenever the signal goes above the threshold, the signal is attenuated, based on typical parameters like your ratio, range, attack and release times, and soft or hard knee setting.
One thing to keep in mind is that the higher the frequency that you’re compressing, the faster you can set your attack and release without worrying about unwanted distortion. That said, avoid super-fast attack and release settings on the lowest frequencies.
Lastly for this chapter, I want to cover the lookahead function which measures the signal ahead of time, resulting in more compression and the capturing of fast transients.
Let’s listen to downward compression, and vary the attack and release, as well as the ratio.
To create upward compression, use the compression mode and combine it with a positive range - notice that whenever the signal falls below the threshold, the band is amplified. This is great for amplifying quieter details of a signal, and Lessing the dynamic range, but from the noise floor up.
The same concepts from the last chapter apply here, so let’s take a listen to upward compression, and notice how we can hear a lot more details.
Downward expansion is more commonly referred to as gating - it’s created by combining the expansion mode with a negative range. As a result, whenever the signal above the threshold, it’s amplified, resulting in a greater dynamic range - this is a great setting for reducing background noise.
It’s also useful if you’re working on a track that was aggressively compressed, and you want to reverse some of that processing by adding dynamics.
Let’s listen to it on some compressed drums, and notice how we bring dynamics back into the track.
Last up we have upward expansion, which is created by combining the expansion mode with a positive range - whenever signal goes above the threshold, the signal is amplified. This is helpful when you’re trying to increase the dynamic range by amplifying transients, in turn making the loudest aspects of a signal louder.
Let’s take a listen to this processing on the same drum track, and notice how it increases the dynamic range, but in a different way than last chapter.
Let’s move on to some of the more advanced functions. If we click the expert tab we can change the detection settings, if we’re measuring the internal or external signal, hear the isolation of the detected range, and blend the detection between stereo and either mid or side.
In band mode, the full band is being measured and detected by the threshold - in free mode, we can cause a different frequency range to be measured. For example, I could compress 200-500Hz, whenever 2-5kHz goes above the threshold.
So let’s listen to this example, and have 2-5kHz level cause compression to 200-500Hz, but, use the mid image as the detection.
If we want to cause dynamics processing to a very distinct set of frequencies, here’s how we can achieve that. Create and center the band over the desired frequency range, then drag the left and right crossovers as close as possible, before increasing the slope to 48dB per octave.
Next, open the expert tab, and select ‘Free.’ Lastly, isolate the detection with the free band, down to a single frequency. Now the processing will only occur when that exact frequency or small frequency range crosses or falls below the threshold, depending on the type of processing you’re doing.
If you do this, be sure to use either linear phase or dynamic phase processing at the bottom.
Let’s take a listen to it and notice how particular the processing becomes.
Since the FF-MB operates in Mid-Side, we can use some of the parameters to either widen or narrow our stereo image, often in a frequency-specific way. The easiest way to do this is to create a band and pan it to either the mid or side image.
A more complex way would be to use the expert mode, and with ‘Band’ selected, cause either the mid or side image to be compressed, expanded, and so on.
So let’s listen to a combination of filters that affect the stereo width, including some side image upward expansion on the highs, and mid-image upward compression on the lows - notice how the highs dynamically widen, and the lows become more detailed and centered.
Earlier, I briefly mentioned the various phase modes at the bottom - they include linear phase, dynamic phase, and minimum phase. Linear phase will result in the most accurate filters, and the flattest frequency response, but, it may negatively impact transients due to your DAW compensating for the latency.
Minimum phase doesn’t introduce this issue; however, the frequencies around the crossover points will interfere with one another, causing amplification or attenuation to those frequencies. This becomes more apparent the higher the slope value at which these crossovers are set.
Lastly, Dynamic phase is a combination of the 2 - it can generally avoid the problems associated with the previous 2 phase settings, so it’s typically the recommended setting.
Let’s listen to 3 tracks, each processed with the same dynamics processing, but with varying phase settings, and let me know if you notice a difference.
Let’s go back to the Expert tab, and use the external setting to cause a track other than the one we’re processing to be used as the detection. For example, let’s say I want to affect my vocal, and use the instrumental for my detection - here’s how I’d do that.
I’d set the instrumental as the external side-chain or the key. Then I’d select external in the expert mode. When the Band setting is selected, the frequency range of the external signal will trigger the processing.
That said, if I want say 200-400Hz of my instrumental to cause expansion to 2-5kHz on my vocal, I’d use the ‘Free’ setting and adjust the range accordingly.
Lastly, let’s say I want the mid image of the instrumental’s 200-400Hz to cause this expansion, I’d change the stereo link setting to mid.
This particular example is great for increasing vocal clarity, so let’s take a listen to it.