Here are three hypothetical audio issues. Because they only present themselves some of the time, regular compression or EQ often don’t fix them satisfactorily. Do you use a multi-band compressor or a dynamic EQ to fix them?
You have a vocalist who gets harsh in the midrange but only when they sing in their high register
A stereo bounce of a track has a bassline that gets too loud on certain notes
An acoustic guitar has a boxy resonance but only when a particular chord is played
Which of you choose depends largely on personal preference and experience. Multi-band compression and dynamic EQ are different but there’s also an awful lot of overlap between them. We know there is a difference but may not exactly what is it?
Multiband Compression Vs Dynamic EQ
Arguably multiband compression and dynamic EQ are for different things but there is so much common ground between these two processes that they often end up getting used for similar things. Dynamic EQs are very suitable for surgical corrective work but they are so flexible that they can be used for the kinds of tasks traditionally associated with multi-band compressors. The converse is also true, while designed for the kind of dynamic range compression tasks which go beyond what can be achieved with a broadband compressor, a multi-band compressor can easily be used as a static or dynamic equaliser and so the difference between them isn’t necessarily clear. They both change the level of an area of the audio spectrum in response to the level in that area while leaving the rest of the spectrum unaffected.
As a process, multi-band compression is older and more familiar, the only analogue dynamic EQ I’ve ever come across is the BSS DPR902II whereas multi-band compressors have been around for decades in both software and hardware forms.
The problem multi-band compressors were created to solve is the issue we’ve all experienced when the level in one area of the spectrum. With broadband compression, often the bottom end, triggers gain reduction, which affects the whole signal, often resulting in us not being able to use as much compression as we’d like. A well known example would be the kick drum causing the cymbals to pump - occasionally a good thing but usually not. For issues where the bass is affecting the top end, just filtering the low end out of the side chain of a single band compressor can help a lot but if we want greater control then passing multi-band compression can stop signals in different areas of the spectrum from affecting each other.
A dynamic equaliser achieves similar results but in a different way. As well as being a regular static EQ, they offer dynamic control of the gain of each EQ filter. In the same way as with a compressor, when the key signal exceeds the threshold the gain is reduced, or increased as most dynamic equalisers can act as compressors or expanders.
They could be seen as each process coming from a different end of the production process as multi-band compression is often associated with mastering and broadcast processing. That is processing that happen at the end of the production process to a mixed track. Whereas dynamic EQ is arguably more of a bottom up process in that it is a development of an equaliser, which is a tool used at every stage of the production process. It’s by no means a hard and fast rule but it makes sense, though with plug-ins we can use as many instances of any process as our systems will support. The limitation of only having one hardware unit no longer applies.
What Is The Difference Between A Multiband Compressor And A Dynamic Equaliser?
To start, lets look at how a multi-band compressor works. In the diagram below, with the audio running from top to bottom, the audio is split into discrete frequency bands using crossovers. Dynamics processing is applied to each band separately and the bands are summed to produce full bandwidth audio after this processing has been applied.
Contrast this with dynamic EQ, the process appears similar but the dynamic gain control element of the process is via each filter’s gain control. The effect can be very similar but there is no use of a crossover involved.
Pros And Cons Of Multi-band Compression Vs Dynamic EQ
|Favourable Q/Gain dependency||Phase shift caused by crossovers|
|Covers entire spectrum by default||Less flexible control of bandwidth|
|Bands can't overlap|
|Not usually capable of boosts|
A table showing the pros and cons of dynamic EQ vs multi-band compression would be the converse of this table, so what does this mean? Let’s look at each of the points in more detail, starting with the Cons…
Phase Shift Caused By Crossovers - This is the biggest sonic difference. Crossovers aren’t perfect. Linear phase crossovers exist but incur latency, if you want lower latency then the crossover will colour the sound more. the biggest sonic difference between a multi-band compressor and a dynamic EQ is that when set to a null setting with no processing happening a multi-band compressor will still change the sound due to the presence of these crossover filters. A dynamic EQ doesn’t.
See the video below for a demonstration of this.
Less Flexible Control Of Bandwidth - A dynamic EQ is capable of very specific targeting of frequencies at very narrow Q settings. Because of the fixed shape of the crossover filters this isn’t the case with a multi-band compressor.
Bands Can’t Overlap - This is also related to the crossover. It is possible to place bands of a dynamic EQ very close to each other or even have them overlap. This isn’t the case with a multi-band compressor.
Not Usually Capable Of Boosts - Most multi-band compressors only offer downward compression and so can’t boost frequencies dynamically.
Covers Entire Spectrum By Default - While it is possible to set up a dynamic equaliser to affect the entire frequency spectrum, if you want to apply some processing across the entire spectrum then this is exactly when a multi-band compressor is designed to do and it really would be missing the point to try to use a dynamic EQ for this.
Favourable Q/Gain Dependency - if you’re not familiar with Q/Gain dependency then check out this piece about the Sonnox Oxford EQ, which is one of the few static EQs which offers a choice of Q/Gain dependency settings.
Q/Gain dependency affects the way an EQ responds when you use it because of how the gain changes the width of the filter. On a static EQ this has no effect on the sound once a final setting is chosen but does affect the way the EQ feels in use. With a dynamic EQ the gain changes during the mix so it absolutely has an effect.
Many dynamic EQs don’t take this into account and can sound less transparent as a result. A consequence of the adjacent filters in a crossover in multi-band compressors is that they can sound more transparent than some dynamic EQs, which don’t take Q/Gain dependency into account.
The Sonnox Dynamic EQ does take Q gain dependency into account and uses constant Q filters for a more consistent sound.
What Do You Use?
Which do you use, multi-band compression or dynamic EQ? Do share your thought and experiences in the comments below.