More of Less: The Art of Stacking Compressors

Sometimes, the tweak of a single decibel of gain reduction can be the deciding factor in whether or not a mix pops out of the speakers just right.

Dan Gluszak makes the case for compressing in stages, a.k.a. “serial compression.” [Image by Flickr user Andrew Buttita.]

As mixers, we often find ourselves feeling as if we’re attempting to decode hieroglyphics when an artist uses vague terms like “smoother” or “squishier” to convey what they’d like to hear differently. “Are they talking about EQ? The way the dynamics glue? That could mean literally  anything!”

Whatever they mean, it’s our job as mixers to get it there. And for the times that it is a matter of compression, the smallest adjustments can sometimes seem to make or break a mix. A single move made to a compressor’s ratio or threshold can trigger the “Eureka!” moment for an artist sitting in on a mixing session.

Too Much of a Good Thing

For many mixers still cutting their teeth, one of the most blatant red flags indicative of inexperience is over-compression.

Much like when a novice guitarist spends his or her first week with a delay pedal and every new idea sounds like a snippet off one of The Edge’s demo tapes, compression tends to be approached with a very heavy hand for those who have just opened their first BF76 in ProTools.

But it just makes everything sound so exciting, doesn’t it?

Yes!… and exciting is good!

However, shaving down decibels in small increments over multiple stages can achieve a far different end result than doing it all at once. Oftentimes, “stacking” compressors with the right approach can yield even better results.

Squeezing Drums

When looking at the waveform of a printed mix, the most extreme transients usually come from the kick and the snare.

The subject of mix bus compression is an ongoing debate in its own right: Some mixers prefer to keep the mix bus untouched, others mix into it from the start, and others slap it on at the very end. Assuming you are utilizing some kind of mix bus compression, these kick and snare transients are likely to be the main elements to which the bus compressor will react.

Although the mix bus compressor tends to get a lot of credit for gluing an entire arrangement together, that bangin’ drum sound punctuating the mix likely began taking shape a few steps earlier. All together, the kicks and snares on a great-sounding mix may end up getting as much as 6dB of cumulative gain reduction by the time the final mix gets printed.

That sounds like a lot of compression if you apply it all at once. Indeed, taking 6dB off in one fell swoop can sound amateur, and is very different than taking off a couple dB here and a couple dB there.

For many mixers, maybe 2dB are coming off at the level of the individual tracks, perhaps another 1dB is getting taken off at the drum bus, and another 2dB on the mix bus.

The key is that if you don’t apply it all in one place, you’ll be working towards a noticeably more cohesive-sounding and more musical blend of all the voices of the kit.

Get Low

When it comes to bass, we’ll often have more than one printed track to work with. There’s almost always a DI, and on some recordings, maybe one or two amp tracks to choose from.

It’s important to be careful of blending too many low frequency sources, as phase relationships can get particularly tricky down there. Nonetheless, in terms of dynamics we may sometimes want to treat an amp track differently than a DI.

In most cases, an amp track will print a little smoother than a DI. Depending on how the amplifier circuit and the speaker are pushed, you’ll often get some inherent compression in the signal before it even hits the mic. Due to the smoother nature of an amp, and the more percussive nature of a DI, it can be wise to treat them differently in our mix, and to look for different things from each source.

An amp track is a natural place to look for saturation and sustain, and a DI track will often contain more transient “punch” and articulation. Whether you want to enhance these built-in characteristics or counteract them is up to your tastes, and the needs of the mix.

Once each track has been addressed on its own, an additional layer of glue over both of them on a dedicated bass bus can help to ensure the combined sources are playing nice together as one. Give the LA-2A (or Waves’ faithful digital emulation) a shot to work its time-tested intangible wizardry for this particular task.

Nice Pipes

Sometimes, it’s best to get the job done a little bit at a time. [Image by Flickr user simon_music]

The human voice is an extremely dynamic instrument that often needs to be addressed in multiple stages of delicate and tasteful compression.

Michelangelo’s statue of David likely bore a closer resemblance to the Michelin Man than the eventual end result between its excavation as a blocky slab of marble and its final meticulous chisel. He chipped away in stages. The same can work for compressing the human voice.

A vocal can often benefit from a gentle pre-compressing before we call in the big dogs for the heavy lifting. You may have even been bold enough to have taken some off the top on the way into your DAW. (The Empirical Labs Distressor does a masterful job at this).

This pre-compressing could be a gentle 1.5:1 ratio, inside which the signal lives comfortably, even at its softer moments. (As opposed to just taming the peaks). The heavy lifting could come in the form of a ruthless 10-15dB smashing at a not-so-subtle 3:1 or 4:1 ratio. This type of extreme crushing can prove necessary when trying to get a vocal to cut clearly through a loud and punishing rock mix.

Another approach to serial compression is to start with a fast attack compressor, just taming the transient peaks, so that a slower attack compressor further down the line behaves in a more consistent fashion.

For vocal arrangements that are stacked with layers and harmonies it can also be helpful to use a smooth optical compressor over the entire vocal bus to help glue it all together. The Sonnox Oxford Inflator also does something somewhat magical in this application

Parallel Universe

Parallel compression can prove to be an extremely helpful tool as well. For those not familiar with the technique, it’s the process of duplicating a signal and reducing its dynamic range tremendously — generally with a very fast attack setting, to chop off the transients and bring out the sustain in between them.

This over-compressed signal is then blended back in with the original source to produce what usually ends up being a much fuller and richer overall tone. This technique proves to be extremely useful when polishing off a drum blend, as well as even tucking an entire parallel mix behind the main mix bus.

(Note: In the software world, we must be very mindful of phase relationships here, as plugin compressors can introduce latency unless we compensate for their delay.)

With its wide range of attack times and patented THRUST circuit, the API 2500 is an extremely flexible tool capable of producing a large tonal spectrum in parallel applications. The SoundToys Decapitator can also add a kaleidoscope of color underneath a drum bus with its vibrant palette of harmonic distortion styles.

There is No “I” in “Team”

All in all, we must call upon a wide array of tools to achieve an optimal end result. A coach building a winning team won’t stack the bench with 20 some-odd centers; he must assemble a network of supporting players to elevate his star players to their ultimate potential. Thus, all of the different flavors and styles of compression we call upon must work together to achieve what no single compressor can do on its own.

“EUREKA!”

Sometimes a touch of multi band compression across the entire mix can be the Gorilla Glue that seals the deal on a mix.

Producer/Mixer extraordinaire Greg Wells attests to this in telling the story of how he tied the knot on mixing MIKA’s hit song “Grace Kelly“.

Though it’s easy to go in and second guess any of the thousand-plus EQ moves you made along the way (and we must be careful not to do the mastering engineer’s job for him!) sometimes a gentle move on a multiband comp can be the answer to a track’s missing mojo.

Summing it Up

No matter what kind of compressor you use, the idea is to work the gain reduction in small, careful increments to assemble a team of compressors each pitching in for the greater dynamic good.

A great mix isn’t the result of a single compressor doing all the work. Eventually, you will have all of your meticulously chosen attack and release times working together to create a mix that breathes just perfectly for the energy and tempo of the song.

And when you hear the mix jumping out of the speakers juuuust right, trust your gut and know when to walk away.

Dan Gluszak is a freelance producer/mixer/engineer who is based in Los Angeles, California. He also doubles as a touring and session drummer.

For more on Serial Compression, try Beyond The Basics: Serial Skills, Simplified.

Please note: When you buy products through links on this page, we may earn an affiliate commission.