Getting a great drum sound is a critical part of mixing. Even well-recorded drums can be tweaked to greater sonic heights with judicious use of processing. In this excerpt from "Jacquire King Mixing Lifeboats," you'll get a feel for some of that as Jacquire applies EQ and parallel compression to the drum bus.
SIDE BY SIDE
He creates a parallel setup by sending the multitrack drum mix to two separate aux channels that are fed by the drum bus. One has no compression, and the other is compressed heavily and brought into the mix in parallel. He does this by creating two aux tracks that are both fed by the drum bus, which includes all the individual drum tracks.
Jacquire’s parallel compression setting features two identical drum channels, one compressed, one not.
He points out that he could also create parallel compression by using the compressor's mix knob. That's because a mix knob controls the ratio of processed to unprocessed signal, which creates the same condition as if you have two channels outputting the same signal, one compressed and one uncompressed, and you use faders to adjust their relative levels.
That said, Jacquire says he doesn't think it sounds as good to use a mix knob for parallel compression, although he admits it's a subjective call. He points out that another advantage of his two-channel parallel setup is that it's easy to swap out compressors if you don't like the first one you try.
If you have a compressor and you like how the parallel compression sounds using its mix knob, he says, then you won't need to set up a secondary dry channel. The compressor he's using for this application is the UAD emulation of the Empirical Labs EL8 Distressor.
He finds it easier to leave the Pro Tools fader for the parallel channel at 0dB (unity gain), and use the Output fader of the Distressor plug-in to raise and lower the compressed drum sound.
TONE IT UP
He explains that he also is going to make some EQ adjustments on the drums. He'll start by applying EQ across the drum bus and then might later do some individual channel tweaking. He temporarily turns off the wet (compressed) aux channel and opens up a UAD Neve 1081 EQ plug-in. He will make his EQ moves on the dry channel, and then copy the plug-in and its settings to the compressed channel, so both are getting the identical treatment.
His objectives for this EQ application are to carve out the boxy midrange, give the drums a low-end boost and open up the top end. He explains that any subsequent moves he makes on individual drum tracks will mostly be about managing the midrange and getting the drums to fit together more tightly.
He starts by trying to decide whether to set a low-end boost at 50Hz or 100Hz. After listening, he decides that the kick and the other drums have a nice low end already, and boosting at 50Hz makes the kick drum compete too much with the bass.
He thinks that for this song, the low-end focus should be the bass with the drums sitting above it in terms of the frequencies that are emphasized. Sometimes it's the other way around, he points out, where the kick is the lowest element, and the bass sits above it. So he ends up using a bell filter on the drum low-end, boosted at 100 HZ.
The EQ settings used for the wet and dry drum bus channels.
He EQs the top of the drums at 15kHz, with a moderate boost. He says 10kHz is too close to the snare and can overly affect it and can also throw the midrange out of whack. He likes 15kHz, which opens up the articulation of the snare and overheads in a pleasing way.
Next up, he sets the plug-in's Low-Midrange band. After listening and trying different ones, he settles on a cut at 690Hz. He also sets a broader Q (bandwidth) for the filter and likes it. He decides against using the High-Mid band of the equalizer, for the moment.
GETTING WET
Next, he turns on the "wet" (compressed) drum bus channel. He wants it to be more aggressive. He sets the Distressor's ratio to its highest setting: Nuke, which is equivalent to a brickwall limiter. He experiments with the High-Pass Filter on the Distressor's Detector circuit to get more upper-midrange accentuation of the compression. With the High-Pass Filter in on the Detector, the compressor doesn't respond as much to the bottom end. (Bass frequencies tend to trigger a compressor more heavily than mid and high frequencies.)
He says he often turns on the Distressor's Link button, which is designed to link it to another Distressor, without actually linking it. He does so because it changes the distortion characteristics in a way he likes. He tries it with and without the Link button in and decides that without sounds better in this situation.
TIME IS CONSTANT
Next, he experiments with settings of the Attack, Release, Input and Output knobs. He uses a slow attack to let transients through. He shortens the release. He says he wants it to pump and be exciting. Faster releases cause more pumping because the compressor is clamping down and releasing more often. When he takes the compression out for a second by turning the output knob to 0, you can hear the difference.
He sets the Attack knob reasonably slow (about 7.5 out of 10), which means it's letting a lot of the transients through. With the Nuke setting on, he's getting a lot of gain reduction, which has the effect of bringing up the room ambience, because it's narrowing the dynamic range.
Nevertheless, if he were to set a slow release, he says it would flatten the tails of the sounds, which isn't what he wants. He says he's trying to get the drums to "jump" a little more. If the release is fast, the compressor will clamp down on the tail end of the transient, but let the bright initial "articulate" aspects of the sound come through.
The drum bus compression setting on the UAD Empirical Labs Distressor plug-in.
He says the wet-to-dry ratio for parallel compression is typically 60-40. He says he's getting is closer to 50-50. The Output knob is a little under 3, Release at 2.1, Attack at 7.7 and Input at 7.7.
He doesn't use the Distressor's Distortion buttons, and he doesn't set the High-Pass Filter on the audio circuit (which is separate from the Detector circuit), because that would thin out the output.
TRY THIS
Did you notice how much the attack and release parameters impacted the sound when Jacquire was compressing the drums? Although threshold and ratio are critical for setting a compressor, the attack and release time, which are called "time constants," also have a major impact on the way a compressor operates.
As Jacquire mentioned, the attack time governs how much the transients are going to come through. When you're compressing drums, that's crucial. With a fast attack time, the compressor clamps down on the transient, reducing its impact and punch. If you slow down the attack time, the transient gets through, and the compression starts on the post-transient part of the sound. In-between settings will split the difference.
The first set of examples are of a multitrack drum kit mix, being compressed by the UAD Distressor plug-in, and setup for parallel compression with the same dual drum-bus routing that King used in the video. So that you can better hear the impact of the compression settings, the non-compressed fader was turned off.
Example 1a: The attack is slow. The transients are coming through nicely.
Example 1b: Now, the attack is quite fast. Hear how squashed the transients are.
The release governs how long the compressor holds before it lets go and can be retriggered. With slow releases, the compressor holds the attenuation longer, giving everything a very contained sound, which lessens pumping and room tone.
With a fast release, you can create "pumping" effects, because it’s compressing and releasing more quickly. You hear more of the room tone, as well.
Example 2a: Here's that same drum example but this time the Distressor's release parameter is slow, giving it a more contained sound.
Example 2a: The release is quite fast, causing more pumping. Notice the room tone that this setting brings out.