Parallel compression

Parallel compression can add energy.¹ It produces a very similar effect to Upward compression.² Hugh Robjohns breaks down the differences and dubs parallel compression “uplift compression.”³

One approach is to use a fast attack time and boost highs and lows (the latter takes advantage of the Fletcher-Munson curve). This makes kick and cymbals, for instance, consistent. I’ve heard this called New York compression.

For a transparent approach, as outlined by Bob Katz in his book Mastering Audio, set a low threshold, low ratio (about 2.5:1), fast attack, medium release (250-350ms), peak-sensing, 20ms lookahead.⁴

For a tone-shaping approach, also outlined by Bob, set the threshold around the Action range (going for at most 5-7dB of reduction), use a relatively soft knee to avoid artifacts, use a medium attack (about 125ms) to let the transients through—we’re only focusing on low-level stuff, RMS-sensing.⁵

Slower attack times may increase the transient information when blended with the dry signal.⁶ When parallel compressing for “punch” like this, make sure the release is not too quick, otherwise everything else will come up.⁷ VCA compression is good for parallel compression. In general, Feedback compression works well in this case. Negative compression can be useful too for just bringing out more transient.⁸ Molot GE can do this, as can DC8C3 and DC1A.

Parallel compression lets you add more body without increasing the peak level. Use this when a client asks for more drums.⁹ Sara Carter uses two parallel compression tracks as well as a parallel distortion track.¹⁰

Parallel compression is also useful for Controlling harsh high end. A Nonlinear process usually works on the low end first (see Use EQ before saturation).