Understanding EQ: Hi and Lo-pass filters (video)
A “pass” filter affects all energy above or below the target frequency, often referred to as the “cut-off frequency,” by removing that frequency content in a continuously decreasing downward slope. Pass filters are extremely valuable tools in mixing because they allow us to”bracket” the frequency energy of a given signal in the mix. We can use hi-pass filters to remove low-frequency mud or rumble from the recording process, or lo-pass filters to shave off hi-frequency content, pulling the signal back in the mix.
While most intermediate mixers are familiar with using hi-pass filters to remove mud and clear out unwanted frequency content in the low-end of a mix, many are unaware at how valuable low pass filters can be at smoothing out your mix’s frequency response and creating clarity and contrast amongst tracks. Unlike an analog mixer or analog tape, a DAW’s mixer is 100% frequency linear in it’s summing process, meaning that no frequencies are altered or “soaked up in the circuitry” during the summing process. Combine this with the fact that many modern virtual instruments and loop collections have an insane amount of high frequency extension (have you heard some of the “paint peeling” top end of plug-in synth presets lately?) and a mix can gather a boat load of fatiguing top end very quickly.
Aside from creating context (or lack there of) issues between the focal tracks in a mix (e.g. if everything has a ton of top end, the mix can lack a sense of depth of focus on the elements that you want to have pop out), I find that this build up of ultra high frequency content tends to do a few nasty things. First, all those high frequencies, if left un-checked, ram into the D/A’s anti aliasing filter all at the same time, and depending the quality of your D/A converter, can result in a nasty haze in the top end that muddles the stereo image and becomes very fatiguing to listen to over time. Second, and most important for a lot of the commercial work I am doing, I find that feeding a mix with way too much un-checked top end into mp3 (or any other compression) algorithms tends to force the algorithm to sacrifice important mid band bits in an attempt to retain a top end you can barely hear. In other words, mixes that aren’t insanely bright will tend to fair better under heavy audio compression.
Now I am not saying go and filter every track in your mix down to 10K, but I do find that subtle lo-pass filtering at very high frequencies with transparent EQ filters (e.g. filtering 6dB/oct at 12K, 15K, 20K, etc on certain non-focal tracks) can help give me a more analog feel and ensure that my mix sounds better when run through youtube’s compression algorithm 50 times.
To learn more about hi and low pass filters, check out this free clip from my Foundations of Audio: EQ and Filters course or watch the whole course at here at lynda.com