Around a year ago I came across an enlightening post by Bob Macc on the consequences of treating your tracks with aggressive filters. You should go check it out yourself, but the big idea is that using steep filters causes phase distortion that reaches all the way up and down the frequency spectrum. Phase distortion removes clarity and impact from your mix, so in short, it’s not good. With that in mind, I’ve come up with some good practice points for filtering your audio.
First, use gentle filter slopes. 6db/octave is my go-to for LPFs, and 12db/octave is my go-to for HPFs. This lessens the impact of phase distortion, but most importantly, it sounds more natural and there’s less risk of accidentally cutting out some good stuff.
Second, use a Bessel curve. Bessel filters are linear phase, which means there’s near-zero phase distortion when they’re implemented correctly. While some EQs have linear phase DSP options, remember that if you’re using steep filter slopes, linear phase will cause distortion too; it’s the combination of gentle slopes and Bessel curves that’ll preserve your audio.
This advice is in line with the Hippocratic philosophy of mixing, e.g. “do no harm.” I’m mixing mostly acoustic music, hut it doesn’t matter if you’re mixing orchestral, folk, punk, or EDM; phase distortion from filters is rarely something you want. Following the two points above will save you a lot of mix clarity and let you put the distortion where you want it.
I want to take on the mindset that recording is additive.
Here’s what I mean; when musicians are shopping around for a studio, it’s common for the engineer to mention that such-and-such gear was used by such-and-such artist. They do this because it works; people love to hear it. There’s a mythology around famous gear that usually sounds like this: If you sing through Sinatra’s mic, you’ll sound like Sinatra. If you’re mixing through a Neve, you’ll sound like Steely Dan. It might not be explicit, but it’s often implied that the sound comes from the gear. Selling a studio based on its gear is a problem unto itself, but this all stems from the idea that recording is additive; that every piece of gear you record through is going to add something to the sound until your music comes out the other end sounding like a finished product.
But it’s actually the opposite. The microphone, the first item in the signal chain, is essentially a filter. It’s rolling off the frequencies at both ends of the audible spectrum, and adding resonance and noise. The next item is the microphone preamp; a “clean” preamp is going to do its best to preserve all the information in the original signal, while a “colored” preamp is going to lose some information, usually transients & high frequencies. If you’re tracking through a compressor, you’re decreasing dynamic range. If you’re recording to tape, you’re adding noise, losing more transient impact, and further reducing the frequency range. If you’re recording to digital, you’ll need high-quality conversion to translate the analog signal without degradation. Continue reading “Audio Myths: Gear is Additive”