Continuing with drum mixing, we’re on to the next step in organizing and “tightening up” your drum tracks. Be sure you get the free “cleansweep” plugin from brainworx before they decide to start charging for it.
In this installment, we focused on placing filters on most of the audio tracks. Unfortunately, the audio for the video (as part of uploading to youtube) lost some low and high frequency clarity. Depending on how you are listening (please don’t say “on my laptop speakers”) you may not clearly hear the benefits of “cleaning up” your audio with proper filtering. Don’t skip this step when mixing! Set yourself up so you can hear this when working on your mixes, and compare the filtered/bypassed audio often.
The audio accuracy that digital recording provides is a double-edged sword. On one side, the even frequency response, low noise floor, and low distortion figures provide us with a clean slate for recording music, and allow us to create EQ curves and funky distortion as we see (hear!) fit for our productions. On the other side, we need to be aware of the limitations of our studio monitoring (and hearing) to give us the whole story about what is going on in our mixes, especially in the very high- and low-frequency ranges.
Monitoring in a home studio is always tricky, and ghostly room resonances may float around your mix position, tricking you into hearing things that aren’t there. You should treat your room (see the first blog post for effective, inexpensive solutions), lest you be haunted by grumbling bass notes, disappearing low notes on guitars and synths, and terrifying midrange that doesn’t translate well in other listening environments.
Beyond the acoustics of the room, the unforgiving clarity of digital recording may be capturing more than just the sound source you are aiming for with your microphone. Here’s where the filter is to your audio tracks as those Ghostbusters backpacks were for the NYC undead. You need to decide, track for track, what range of your audio signal has information that you want in the mix, and what can go. I’m of the opinion that one of those Brainworx “Cleansweep” plugins belongs on every track of a home recording. Of course if you hear that it is taking something away from your sound, remove it. There are no definite rules. That said, here is a ballpark figure for what you can filter from a variety of sound sources:
– Bass drum: Cut below 30Hz to avoid subsonic problems that can steal energy and clarity from your mix. If there is snare or cymbal bleed, I like to also cut above 6k or so, and use a nice clean sample to bring back the “tick – tick” clarity of a kick drum, if needed. Keep in mind, overheads and room mics also catch a clear “picture” of the kick.
– Snare drum: Low cut frequency settings depend on the size of the snare. In most cases, there isn’t much below 80Hz. You can also trigger a low frequency sine wave to add “oomph” to a snare, while still filtering it clear of kick drum bleed that might be out-of-phase with the kick, robbing it of clarity.
– Overheads: To filter or not? For home-recorded drums, a gentle roll-off from 500Hz on down will help reduce boxy/muddy room resonances. Definitely compare by bypassing. If you went with a simple drum setup (kick/snare/overheads only) then drastically rolling off lows won’t work; you will lose the body of the toms. Filtering highs is not necessary, unless the cymbals really sound like trash can lids. In that case, notching some 1kHz-2kHz is probably a better way to take out the trash.
– Toms: Determine the fundamental of a tom, and filter everything below it. Be especially aware of floor toms, which can add a lot of low frequency energy to the mix. When mixing for vinyl, be careful not to pan the floor tom too wide, or the needle may skip.
– Guitars: The low E-string on a guitar has a fundamental frequency of around 80Hz. That means that anything below that is “just noise”. HOWEVER – heavy, distorted rhythm guitars may need some of that low end “chunk” that happens in speakers when palm muting, and may be below the fundamental frequency of the chord being played. It’s still a safe bet to cut off below 50-60Hz to keep the subsonic mud out of the picture. In the upper frequencies, guitar amps are not going to produce much energy above 8kHz. Above 8k, there is more likely to be amp hiss, cymbals bleeding into the mics, etc., so set the high filter for 8kHz or so. Clean, D.I. guitars may have some chime-like high tones you don’t want to lose, so listen and compare.
– Bass: The low E on a bass guitar is an octave lower than standard guitar at about 40Hz. That’s pretty low, but a string clunking a pickup can give us a “whump” down to 20Hz and below, so filter below the fundamental. If the low string is tuned down, take that into account. Same goes for guitars, of course. And, as with guitars, the highs from 8-10kHz on up are more likely noise than music, but listen and compare.
– Synths: With the wide variety of sounds they can produce, you have to decide on a case-by-case basis. Understanding music is important! Check to the part the synth plays. What’s lowest note in there? You could filter below that frequency, but there may be a sort of “sub octave” effect on the sound. Take this into account. Most of all – you guessed it – listen and compare.
– Vocals: Once again, go by the song arrangement. Johnny Cash hit some notes that were a lot lower than the standard range you’d expect, so saying “filter below 100Hz” would be wrong for a singer that hits a low E note with an 80Hz fundamental. In general, you may want to just stick to rolling off 75Hz and below. Female vocals likely have a higher range, so you could filter higher. A live vocal mic is a magnet for mechanical noise, pops, and rumbles, so low filtering really keeps your most prominent track clean.
– Acoustic instruments, including Guitar, banjos, mandolin, fiddle, etc.: There are a variety of stringed instruments that are gaining popularity in modern song arrangements, and you’re likely to find some in a mix you are working on. Foot thumping, passing traffic, and assorted rumble often sneak into acoustic instrument recordings, so set your filter, again, below the lowest fundamental frequency of the instrument. Sometimes it’s cool to capture the foot-tapping of a performer if it is in time with the tune, so mic that foot when tracking. If they’re out of time or inconsistent, then filter away.
– Piano: Recording a real piano is beyond the scope of this post, but you can safely filter lows from 40Hz on down to keep out mechanical noise from sustain pedals, etc. Just be sure you’re not cutting out the body of a low note that ends the tune!
That should give you a good start, and gives you some insight into how you can approach using filters. This is the practical side. Of course, you can use filters for effect as well, cutting lows and highs from drums, for example, in an intro or bridge, making them “small sounding”, and increasing the impact when the song starts or kicks back in.
One last round of info – this is all good stuff to know! – you should keep in mind that filters can also “color” the sound. The brainworx “Cleansweep” is very neutral sounding, IMHO, and that is because of it’s gentle 6dB/octave rolloff characteristic. This means that if you set the filter for 100Hz, 100Hz will be down 6dB, 50Hz down 12dB, and 25Hz down 18dB. This gentle slope sounds very natural, having the effect of “tightening” the low end overall.
A typical “hard” filter might cut 12- 18-, or even 24dB per octave. An additional characteristic of this type of “hard” filter is a “bump” around the cutoff frequency. So if you cut 24dB per octave at 80Hz, you may also be boosting a few dB at 80Hz. This can sound just right on percussive sounds, but may sound unnatural on something like vocals. If you have serious low frequency problems of a kick drum, for example (I did in the mixes for The Soundmills), then combining a 6dB/octave filter that starts rolling off at 80Hz with a hard filter that cuts everything below 30Hz may be the ticket.
One last time, that sage advice: Listen and compare.