Previous Next

Production Tip #3 - Filter Every Channel

First off I want to say I’m glad these tutorials have been getting good feedback on Twitter and Facebook and I’m very happy you guys are finding them helpful!  Also, my new original track featuring Silver Medallion is nearing completion and will be mixed by the end of this coming week.  If you missed it I put up a brief preview of the drop which you can hear at : American Girl Preview

Now, onto the mixing stuff!  Today I want to pass along a piece of mixing advice that I got a while ago but didn’t really apply to my work until fairly recently, and it has made a world of difference.  That advice is simple- filter (high and lowpass) every track in your mix.

Some of you may already do this, and some people may disagree with this advice, but I have found it to produce excellent results when you apply it across a whole mix track by track.  The principle is basically this: especially with soft-synths, there is a lot of unnecessary frequency information in a track- most AU/VST synths produce a lot of high frequencies that go well beyond 20,000hz which is the upper limit of human hearing.  The problem is, these frequencies (even the ones that are not audible) eat up headroom in your mix.   You will also find that when using typical chord voicings, synth leads will produce frequencies below 150hz, which begins to interfere with your bass section.  When you have these frequencies building up across every channel going into the master, that’s a lot of headroom you’re losing.  

Additionally, it becomes more than a headroom issue for dance music.  If you’ve ever been to a club and heard a poorly mixed track, you know it’s not fun.  One of the worst things you can have is a track which is searing and overly bright in the high frequencies, because on a main room sound system this will get harsh to people’s ears very quickly.  Low frequencies are a big issue too.  The key to a great low end is not every track in the mix having bass in it- the key is that you have a very focused and powerful bass section and your chords/vocals/synths sit above it without excessive frequency overlap.  I personally try to highpass every non bass instrument around 100hz or so, unless it really removes power from the sound or makes it sound weird.  Even in that case, I will usually try to reconsider the arrangement and layering of sounds, because a well produced track should rarely have chords/leads/vocals that are interfering with the kick and bass- I’ll talk more about how this applies to songwriting and chord voicing in a future tutorial.

When applied across a whole mix, this technique will greatly reduce frequency buildup in the very low and very high areas of your spectrum.  Frequency buildup is what happens when a certain part of the frequency spectrum is overpopulated by numerous sounds in a mix- remember, the goal when mixing is to put each sound into its own frequency space, because otherwise we get buildup and masking.  

Below is an example of this principle at work using a basic chord track sequenced with Sylenth 1.  This is a basic chord progression (vi-I-V-IV) played in C.  You’ll notice in the untreated sound that it has a lot of power, which might be nice for a breakdown where there’s less competing instruments, but let’s consider plugging this sound into a drop or busy section of the mix:

Although this sound doesn’t have a ton of extra bass information, it has loads of high end buzz which will not help us attain a nice mix and will eat up headroom.  Now I’ll use the DMG EQuality EQ (Analogue Phase Mode) to strip out the lows and highs:

This aggressively takes out some of the bass, which in this case is a good thing because if we’re putting this instrument in a drop or busy section it will have a strong bassline and kick below it.  I’ve cut the highs above 20k to save headroom and remove a bit of harshness, and as the mix goes on I could easily roll off the highs some more as the sound is still a bit bright.  The big thing to understand here is that this is just one channel of our mix.  If our drop has 20 sounds, between drums, fx, bass, leads, vocals, whatever, competing for space, imagine how much of a difference you get if each channel is filtered so it’s at least somewhat isolated in the frequency spectrum.  Dealing with full on overlap in instruments and sounds is a much bigger topic, but this begins to address the issue.

One further application of this is to things like sound effects (sweeps, white noise, etc), hi hats, cymbals, and lots of instruments or sounds that have a very specific frequency range.  You can be even more aggressive than I was here when filtering such sounds: I often highpass hi hats or cymbals well above 200hz (depending on the sample, you can sometimes lop off everything below 2,000hz and still have a solid sample), and again, the more unnecessary stuff you cut out, the more easily every sound and track can gel together for a wholesome mix.  Sound effects like risers, sweeps, and the like are also very good targets for cutting highs or lows depending on the sound- it’s all about asking yourself what’s the point of this sound in the mix (like, ok, this sub impact is here to accentuate a drop off on the low end) and cutting everything in the sound that doesn’t contribute to that goal.

Also, experiment with different EQ’s/Filters as each has their own sound and sometimes will just work better for a certain instrument.  I like EQuality a lot, as well as FabFilter Volcano and Satson Channel by Sonimus for more gentle filtering- Satson has by far the most analogue and incredible sounding filters I’ve ever heard in a plugin.

You can hear the untreated sound, DMG filtered sound, and the Satson version here:

blog comments powered by Disqus

5 notes

  1. asquall reblogged this from dave-edwards
  2. dave-edwards posted this
Back to Top

Vanity by Pixel Union