Everybody knows what is a Multiband Compressor so I’m not going to get too deep on what it is, but I am going to help you use it like a pro.
The thing about Multiband Compressor (MBC) is that it’s like a ninja sword. It’s not the smartest move to start using it until you’ve mastered the wooden sword first.
In this post, you will learn everything there is to know about the MBC- how to use it on vocals, different instruments, and more. It is the perfect tool for controlling and even somewhat automatically mastering a lot of different sources. That’s why it has taken its first steps at radio stations.
Yeah, I know I said I’m not going to teach you what it is, but this is only a short section for the new guys. Basically, a Multiband Compressor is a 3 or 4 (or more) compressors in one plugin, spread over the whole frequency spectrum.
Each compressor is working at a different frequency range, giving you the ultimate dynamic control over the channel. You can also think about it as a type of dynamic EQ that allows you to separately compress each and every frequency range.
The Multiband Compressor is a great tool for controlling and shaping a simple or complicated source with one tool. For example, if you want to shape and control a vocal channel and you want to surgically compress every frequency band while getting a very specific result, the MBC is the best tool to do it with.
Think about compressing a single vocal channel; You can compress the low range in 5 db to get a very controlled low end while leaving the mid and high bands a little bit more loose.
It’s an express lane straight to dynamic heaven.
I always suggest starting with leveling the dynamics for the whole channel. Sometimes it takes a little pre-fader automation work or changing the levels for each event on your channel, which is my preferred method.
The goal is to create a stable RMS level for the whole channel and work your way up from there.
After that, I add a regular digital and transparent compressor for a little more control over the dynamic range. This compressor is shaving the top 2 or 3 db- that’s it. Only do that if you feel like the raw vocal is all over the place dynamically.
This is the part where you throw in your favorite MBC on the chain. Keep in mind, I don’t suggest going with “The best sounding Multiband Compressor” but with the one that you are most familiar with. This is important.
First, try looking for problems. If you need to, you can open an EQ plugin before the MBC just so you can find problems by easily swiping across the frequency range.
For example, if you find a harsh high mid frequency, you can immediately set up one of the high-frequency bands on the MBC to shave 2 or 3 db’s off of it.
Remember, everything we do with a Multiband Compressor should be done in a corrective manner and very subtly, unless we’re looking for something else other than control.
After that, you can put a regular LA2A or a 1176 style compressor and get an overall processed and controlled vocal. I think I did that on almost every vocal I’ve ever mixed in the last few years.
For me, the trick with multiband compressing is not to overdo it.
It’s best if you use it only to dynamically shape your source and, only then, maybe do some heavy lifting with a regular compressor and a limiter after that if you feel the need for it. When you’re done processing your vocal channel, bypass only the MBC, does it sound like you lost some of its magic? Great, you did a good job.
In some cases, it’ll sound better without the MBC, even after hours of fiddling with it. In this case, turn off your ego and the MBC and move on to the next channel in your mix.
In most cases, a regular good compressor would be just fine.
Now that you have control over the different frequency bands across the whole range, you can use it to shape and create a certain color for your channel. It’s very similar to an EQ shaping of the source but it is more dynamic and alive.
You can create a bottom-heavy guitar sound without crowding the entire low range, or a vocal sound with a nice bright character that cuts through the mix without it being too harsh. You can add in the pretty stuff and cut out the ugly and unnecessary garbage.
Of course not. As a matter of fact, in most cases, you won’t have to use all the bands because you’d only need to shape and control one or two areas. Let’s say you want the mid-range section to breathe and be free while you do want the low and high ends to be tighter. Easy- go for it.
In short, an expander will expand the dynamic range of a source once it goes above the threshold.
Let’s say you want the high band to be compressed and controlled, the midrange to bite and be more aggressive, while the low mid is compressed and the low end is pumping and kicking you in the stomach- You can do that. It works beautifully with bass guitars, different synth sounds, and with almost any other source, really.
The real secret in expanding is finding the sweet spot for the attack and release.
Every source needs its own settings and, once you find it, you can make an electric guitar or bass much punchier and help them jump to the front of the mix more easily. On vocals, you can really affect the performance and give it more bite and aggression with the right settings.
It’s important to know that there are no specific settings for each source, you just need to tweak and find it for yourself.
This is a cool one- I call this “compressing without compressing”. Meaning, the processed channel will not get affected unless something else covers it. Let’s say you have a cool electric guitar riff that holds the song and you want it in the middle of the song together with the vocal but they are both sitting on the same frequency range and you don’t want them to clash for the whole song, this is what you do.
Send the vocal to a parallel bus and the bus to trigger the sidechain in the MBC on the guitar channel, it’s easy to set it up. It’s a little different on each digital audio workstation application but the idea is pretty much the same.
Look for the frequency range where most of the vocal sits and set the sidechain on the guitar’s channel right on that same range. Now, every time the vocal plays, it triggers the compressor for that same frequency range on the guitar channel and compresses the guitar without losing it in the parts where there’s no vocal. Lovely.
I can only tell you how I personally do it, Here it is. First, you have to get the overall dynamic of the vocal settled. You don’t want the RMS to get too crazy because you need a steady level going into your Multiband Compressor. So you do a basic compression on the vocal before it goes to the MBC.
Make sure to keep it loose, meaning slow Attack and fast Release. After that, you need to set the bands to expand instead of compress. Usually it’s only changing the Ratio parameter to a positive value. This tells the band compressors to create a boost every time the signal’s going above the threshold.
You don’t have to engage all the bands. You only need one or two, sometimes three.
If you use more than that, the source might break and not be as powerful, if that makes any sense. Most of the “punch” power is located in two ranges: it’s the low mid and high mid. I don’t usually touch the middle of the range, because that is where the natural fundamental is living and I don’t like changing it.
So if it’s a deep male voice, the “low punch” is between 80hz and 300hz and the “high punch” is around 2k to 5k.
Try these ranges as a starting point for every male voice. If it’s a girl, the highs are pretty much the same and the lows are a bit higher- let’s say from 150hz to 350hz. Turn off all the other bands. Now, solo the low band and set the ratio to an exaggerated value like +6db, the Attack on 30 milliseconds, and the Release on 70. This is just to put you in the ballpark.
These parameters will probably change along the way as you’re searching for the sweet spot. Let the vocal play and start lowering the threshold. At some point you will start to hit the point and the meter will show a few db’s of gain each time the threshold is being crossed. It will sound like the peaks are jumping out of the monitors- this is what we’re looking for.
When you finally find the exact timing for the Attack and Release, set the threshold to be a bit higher, and set the Ratio to 2-3 db max, depending on how aggressive you want it to be. Do the same for the High section.
Usually the Attack and Release will have to be shorter because the higher frequencies are faster. This also works great on synth channels, basses, and guitars. I don’t recommend doing it on acoustic guitars, although I’m learning new things every once in a while so never say never.
When you record drums, the overhead microphones will pick up more highs than lows. To me, the idea is to get the overhead channel to sound almost like a full range drum set. Yes you can do it with an EQ and leave it there, and you can also use the MBC for a little more dynamic result that sounds a bit more “Alive”. This is how I do it.
I don’t compress before the compressor. I just do a high cut and low cut with an EQ before I send it to the MBC. Now you start with “EQing” with the Multiband Compressor. Yes, you heard right. You can shape the channel’s EQ curve with the gain feature on every band. After you do that, you start compressing a few db’s on every band. Play with the Attack and Release until you like how it sounds. Remember, these are fast sounds, so shorter Attack and Release will get you there more easily.
Make sure to work in solo mode on each band and take your time with it. Remember to be gentle, we are not trying to break the source, only to make it sound a little bit more alive. After you finish processing, try to switch the MBC in and out and look for what it does against just doing the same with an EQ. If you like what you hear better with the MBC, we did good 🙂 Next.
You’d be surprised to know that acoustic guitar is one of the hardest instruments to mix, let alone record. Of course, you can use an EQ and compressor to process an acoustic guitar, but how about having both of them in the same plugin and work their magic in a much more intelligent and dynamic way? Ok, the truth is that a rhythm acoustic guitar channel is made up of few different processors so the MBC is just a small part of it but still, it really helps the guitar to find its place in the mix.
This is how I do it- I start with an EQ to give it a basic shape and high and low cuts. Then, throw in the Multiband Compressor and start working on the “dynamic shape” of the guitar. You can make an acoustic guitar breathe and move naturally on the spectrum with no sudden peaks or crazy holes in the frequency range.
Let’s take a rhythm acoustic guitar for example. With the right process, you can make it sound like a steady warm acoustic wall of chords that wraps the whole song in sweet harmony. So we use all the bands and we start with the ratio for all the bands on 5db. From there, we start lowering the threshold until we see some compression. At this stage, we can already hear what the MBC is doing to the guitar.
Now, this is a big one for me- although we compress each band separately, I still find it very important to keep the same Attack and Release values on all the bands. If we don’t do that, in some cases the guitar might break and lose it’s energy and drive because we messed with its internal dynamic properties, which we don’t want.
We only want the whole frequency range to be consistent and balanced. With a rhythm acoustic guitar, you’d generally want to have a fast attack and a slow release, Play with it until you find the right timing for your acoustic guitar part. You don’t have to work on each band in solo mode. Remember, the goal is just to have balanced rhythm acoustic guitar across the whole song. When done the right way, it can make a world of difference in the final result.
In cases like this, where the piano takes a really big place inside the song, it’s very important to control its dynamics and use it as an emotional ground for the whole song. This is the perfect scenario for using a Multiband Compressor. Again, we are using all the bands here and we start with the main section for the piano which is the low mid to mid frequencies. Most of the energy in pianos is living between 150hz to 2Khz, so we can start there.
As a starting point, we set the Ratio to 5db on all the bands and lower the threshold to see some compression. If the part is consisting of long chords, you want the Attack to be around 50ms+- and the Release to be around 200ms+-.
You should always play with these parameters and not take these numbers as gospel. The reason for that is that every piano has it’s character and unique properties. The main goal here is to find the right inner groove for the instrument and the way it plays. It’s very important to set the Release to the right value. You don’t want the piano to “Pump” out of rhythm or be too lose. You have to look for the right timing.
When working with Multiband Compressor on a piano, you don’t want to stay in solo mode all the time, because you’re EQing and compressing at the same time, so you want to know exactly what you’re getting out of your speakers. Also, make sure that you keep the same Attack value for all bands so you won’t “break” the energy of the piano. As for the Release, here, you have much more freedom because of the nature of the piano.
The lower notes tend to sustain longer and have more energy, so you can try and set the release for higher values.
After you find the right timing for the Attack and Release, make sure to find the right Ratio. I, personally, don’t like to squash the piano too much, so, for me, it’s between 2 to 3db of gain reduction on each band. It is very subtle, but inside the mix it makes all the difference.
As I said in the last acoustic guitar tip, it is one of the hardest instruments to mix, and even the most experienced audio engineers and producers will sometimes struggle with that. Sometimes acoustic guitars can have unpleasant and shrieking high content frequencies, especially when the strings are new. You can fix that with an EQ, but when you just create a hole in the guitar’s natural frequency response you might just lose good information. You want to fix that in an automatic and intelligent way. Of course, the MBC to the rescue.
Usually, these shrieking frequencies are between 2Khz to 8Khz.
So, it’s going to be very simple. First, yes, this is a mastering tip, but you shouldn’t be afraid of it if you’re not yet at that level. Assuming that you have healthy levels and good balance across your mix: no distortions, no clipping and you’re entering the final stage of the mixing process, throw in your favorite MBC on the mix bus and start here. It might change from genre to genre, but it’ll give you a good starting point. E
ngage only 2 bands, the lowest frequency range and the highest frequency range. Bypass all the others.
Set the Cross-over point to 150hz. Set the Ratio to 3db, Attack around 50ms, and the Release at around 120ms. Now, start lowering the threshold in solo mode. Now you see and hear a little compression. If you feel that it’s not enough for you, try increasing the Ratio. Remember, don’t get the threshold too low or compress more than 2 or 3db’s, because you might break the energy for the whole mix. After you get the required compression, increase the gain for the low band in 2db. That should make your lows more collected and controlled.
This band should range from 2k to 10k. Set the Ratio to 3db. Attack – 3ms, Release 6ms, and start lowering the threshold to see some compression. Once you have reduction, increase the gain in 2 to 3db to compensate. Again, try to play a little bit with the parameters to get the timing right. Remember, it should be very subtle. Bypass the MBC on and off to hear the changes. If you like it, we did good.
Techniques for Mixing with Multiband Compression
Yeah, it’s not a music production tip, but we all do everything these days, and it’s a pretty amazing tip, so I’m not going to leave it out. It’s simple, assuming that you have healthy levels, no distortions, no clipping, and you’re entering the final stage of the mixing process, throw your favorite Multiband Compressor on the channel and engage all the bands. Now, look for the busiest part in your signal and set the ratios on all the bands to 8db. Next, look for the right threshold level for your signal until you start seeing some compression. Do it on all the bands and let it go.
Try to look for unpleasant peaks in the signal and work from there. This is not a steady and repetitive audio source, like music, so it’s all over the place. But it’s enough to have 8db of gain reduction on all the bands, and you already have good dynamic control over the signal. Some MBC’s have an automatic Attack and Release control. This will be the perfect application for that.
It’s a little funny, but some female singers have some kind of an annoying high frequency that makes them unpleasant to listen to, and it might ruin the whole performance. I don’t like to name names, but I think one very good singer that has an annoying high-frequency thing in her voice is Jorja Smith. She is a great singer, but if you listen to her natural voice on her show in Tiny Desk Concert, I couldn’t take more than 2 minutes. It’s just painful.
So what do we do?? Easy! We drop an MBC on that joint 🙂 Usually, the annoying range is right between 3k to 7k.
How do we find it? We first open an EQ on the channel and create a narrow band with a boost of 10db. Then, we sweep across that range to find the painful spot. Once we’ve found it, we set one of the high-frequency bands to live right on this area with a narrow band if possible. This is a good starting point:
Again, start lowering the threshold to see some compression and go from there. Make sure you’re not overdoing it, because you don’t wanna kill the natural character of the vocalist. It is important to understand that every MBC is working a little differently, so you might get different results on different MBC’s. That is why you should play with the parameters around those starting points to find the sweet spot for your source and your Multiband Compressor.
Don’t forget to compare your work with the work of others from the same genre. It makes a world of difference if you’re aiming to the level of other professionals.
Remember! Always switch the MBC on and off when you finish editing, it’ll give you an overview and let you know if you like the result or not. There will be some cases where it’ll not work, don’t fight it. Switch it off and go on with your mix.
Play with it, have fun with it, tweak for hours until you’re able to use the MBC like playing an instrument. As a matter of fact, this should be your way of thinking throughout your whole musical journey.