How to Use Compression Effectively when Mixing

Alternatives to Downward Compression

Instead of applying downward compression, which can add both desirable and undesirable timbre alterations to the signal, we have alternatives.

For example, wave shaping could be used to bring forward quieter details. This is a good work around for lower peaks and then increasing the output gain.

I always recommend this free MWaveShaper which can introduce nearly constant amplification to quieter levels. By creating a shape in which lower amplitudes are expanded without any alteration to the peak level, we can introduce more detail without the need for peak down attenuation.

Conversely, we can reduce the dynamic range via saturation - which will introduce gradual or soft-knee peak down attenuation while reducing the range from the noise floor up by introducing harmonics.

These harmonics will typically fill the low mid to high mid ranges, resulting in reduced dynamic ranges for those aspects of the spectrum.

Let’s listen to a signal with an amplified low range using wave shaping and then try the same using saturation. Keep these options in mind if you’re ever looking for an alternative to compression when mixing.

Watch the video to learn more >

Traditional Compression Settings for Various Instruments

There’s a lot to run through in this section, so let’s try to do it quickly, but if you have any questions, leave them in the comments, and we’ll get back to you.

For vocals: 4:1 ratio with a softer knee, 20ms attack, 50ms release, lookahead of 2ms if available, lower the threshold until you have 6-8dB of attenuation, and use make-up gain to compensate for the changes. These settings capture the vocal quickly with the attack, lookahead, and softer knee combination, retain intelligibility with the 50ms release, and then bring up quieter details with make-up gain.

For Kick: 4:1 to 8:1 ratio with harder knee, 35ms attack, and a release no longer than the next kick instance, which will change from song to song. Lower the threshold or increase the input until you achieve 4-8dB of attenuation, depending on the genre, and compensate with makeup gain. The 35ms attack will let enough of the transient through while introducing distortion that compliments the hit. The release ensures that other kick hits don’t get over-compressed.

For Snare: Similar settings as the kick but with a release no longer than 50ms. The same concepts apply but try different compressor types to accent the highs. I like an 1176 emulation, but I encourage you to experiment.

For Bass: 2:1 to 4:1 ratios work well, with about 4-6dB of attenuation depending on the performance and genre. For a smoother bass, try a higher threshold with a softer knee, a quicker attack, and a longer release of about a half-note.

For a punchier bass, try a lower threshold with a hard knee, a moderate attack of about 50ms, and a shorter release time.

For BGVs: Start with a de-esser or MB compressor and heavily attenuate any sibilant ranges. Then, use settings similar to your lead vocal but with more attenuation overall. The goal is to control these enough to blend in with the lead, letting the lead vocal provide consonant and sibilant info.

Watch the video to learn more >

Controlling Spectral Response with Parallel Compression

As I mentioned earlier, compression’s main purpose is to bring forward quieter details. If we do this to specific frequency ranges, we have a fantastic way to equalize a signal - much more effectively than EQ alone.

For example, say I want my bass to have aggressive high frequencies. I could EQ it, or I could do the following.

Step 1: Create a send from the bass track and set it to unity.Step 2: Insert a linear phase EQ on the corresponding aux track and isolate the range I want to affect.Step 3: Heavily compress and use makeup gainStep 4: Blend the dense range with the original signal via the aux track’s channel fader.

This has helped me out more times than I can remember.

Say the mids of a vocal are lacking in detail - I’ll follow the same process.

Or maybe the high end of the snare is being buried in the mix - again, this will work so much better than simply equalizing the snare.

Let’s test this out and compare equalizing the ranges of these instruments with the parallel method I just described. Let me know if you prefer one way over the other.

Watch the video to learn more >

Controlling Creative Effects with Parallel Compression

Effects like reverb, delay, chorusing, etc, are hard to fit into a mix. Either these temporal effects overpower the entire mix, which happens a lot with reverb, or they get buried and don’t pop through when they need to.

If we use parallel compression to enhance the quieter details of these effects, we can then automate them when needed.

So say I want some delay on the vocal that’s subtle during the majority of the track, but at certain moments I want the delay to stand out. I’ll set up a send, use the desired effect, compress it, and then automate the aux track’s fader.

Or, I want the reverb to be incredibly dense. I can combine dense reflection reverb settings with compression that brings forward all details that would otherwise be masked.

Parallel compression really allows us to shape the ADSR and the overall level of detail of our effects in a way that isn’t possible with most, if not all, temporal effects.

Let’s take a listen to it, and notice how much detail parallel compression can pull out of effects.

Watch the video to learn more >