The topic of mix bus compression – mixing through a compressor on the master output – can stir up a lot of conversation among engineers across many genres (live, recording, broadcast, etc.).
For example, ask studio engineers about bus compression, and five bucks says their responses will include the word “glue.” It’s probably because bus compression acts on the mix as a whole, as opposed to an individual element, so it can create an “everything stuck together” kind of sound.
Unfortunately, this is all too often overdone, squashing all the life out of a perfectly good mix. To my ear, more than 2 or 3 dB of gain or reduction starts to sound…dull. So I never use a bus compressor to try to control the dynamic range of a mix; rather, I compress individual inputs or subgroups as needed.
That said, there’s often value in using bus compression artistically. A compressor can have very unique “sounds” due to its design, and there are times when I’ll use a bus compressor for “color” or “flavor” rather than strictly gain reduction. In these cases, I’m not seeking transparent compression as I would be for dynamic control, but rather using a certain compressor because of its non-linearities or distortion characteristics as a part of the sound. (This is somewhat like how we expect absolute transparency from power amplifiers while a guitarist looks to his or her amp for intentional coloration as part of his or her tone.)
In the studio, my go-to unit is the Shadow Hills Industries Mastering Compressor. Okay, I can’t afford the hardware unit, but the Universal Audio plugin version is killer. I dial in a touch of optical compression, maybe 2 dB of makeup gain, and let the (emulated) output transformers add a bit of, well, “glue.” This creates a pleasing effect that’s quite subtle and appropriate for many (but not all) genres.
My other favorites are Universal Audio’s LA-2A and 1176 LN, although I’m far more likely to use them on drum groups or vocals. Many studio engineers also swear by the Fairchild 670 (likely a plugin, unless they’re lucky enough to have access to the original hardware) and the Manley Variable-Mu.
There’s a noticeable exception to my “only 2 or 3 dB of compression” policy – “New York style” compression. More properly called parallel compression, and most commonly used on drums, this technique involves mixing a compressed signal with its uncompressed “original.”
For an aggressive drum sound, I bus kick and snare to an 1176 in the infamous “all buttons in” mode, crank it up until it distorts, and mix this back into the drum mix at maybe -20 dB. It creates a crunchy, raggedy sound while still retaining the transient impact of the uncompressed signal. (Watch out for latency-induced comb filter bugaboos if you try this live with a digital desk.)
Then there’s the question of whether to stereo-link the bus compressor. Linking advocates say that linking left and right prevents gain reduction from being applied to only one side of the mix, which would cause instruments to wander in the stereo field. If you use only minimal compression, of course, this isn’t likely to be an issue, especially in a large venue live situation, where the idea of a stereo field only applies to a small portion of the seats anyway.
The argument against linking comes courtesy of Dave Rat, and makes a lot of sense to me: In the linked mode, if the mix becomes unbalanced towards one side, gain reduction is applied to both sides, which does nothing to combat the imbalance. In dual mono mode, the wayward side is brought back in line.
As noted at the outset of this discussion, bus compression is a topic of interest among many in professional audio, so with that in mind, we reached out to a couple of additional engineers to get their take. Thanks very much to Karrie Keyes and SoundGirls for their help in making this happen.
Front of house engineer Fela Davis: “Early in my front of house career, I used stereo compression on my main outs when mixing on small analog boards to make the overall mix sit better and add more depth. Now, with so many compression options in even the most inexpensive digital console, I’ve found myself not needing to add it to my main outputs.
“I like compression on certain individual channels and groups but that’s when I’m mixing rock or electronic music. Adding stereo compression to my main outs after group compression muddied panning and spacing of my mix. Currently I mix a lot of acoustic music so heavy compression isn’t needed. I’ll compress individual channels when mixing small three- or four-piece acoustic bands, and I tend to not use groups on small bands because I like to tweak the channels more with a smaller band.
Front of house engineer Michelle Sabolchick Pettinato: “In mixing live sound, I find some compression or limiting on the master stereo bus output of the console to be a very useful tool. I’m particularly fond of the Waves Maxx BCL (analog unit) for this purpose. The limiter on the Maxx BCL is the function I use most frequently because it catches the transients without squashing the entire mix.
“A specific example: The drummer for Styx plays a lot of ghost notes, and in order for those to be heard, I really have to push the snare in the mix. It allows me to be able to drive the snare channel hard enough without hitting the limiters on the PA.
“Compressing the snare drum channel hard enough to control the difference in level between the ghost notes and hard hits can create some unpleasant tonal characteristics and really squash the dynamics of the snare. So I use some light compression on the snare channel and let the Maxx BCL keep it from jumping out in the mix but still allowing the ghost notes to be heard.
“I also like to send all of my drums to a stereo group and add some light compression. This gives the drums a cohesive sound and prevents things like a random tom jumping out in the mix. With proper attack and release you can also add a little punch to the entire drum mix.”
For even more on the topic from the recording and TV/film side of this issue, go to here.