Battle of the Multi-Bands Part II
Last time we focused on multiband compression as a corrective measure used to fix problematic guitar tracks as opposed to viewing it as a viable creative tool to enhance your guitar tone. Being able to compress designated frequency bands separately that then blend into one sound can get a little confusing. Old school audio engineers often contend that spectral dynamic processing is dangerous in the wrong hands but, I think a little recording irreverence is badly needed these days. What happened to the kind of spirit that led to overdriving Telefunken V-72s, or pressing all the ratio buttons in on an 1176? You can’t fumble upon the good stuff without breaking the rules once in a while…having said that, it may be helpful to review a few fundamentals before heading up your own recording rebellion.
Sweeping it Under the Rug
“Know thy frequency spectrum” has got to be one of the Ten Commandments for recording. For those who don’t have a handle on what goes on between 20 HZ and 20 KHZ this will be a good way for you to get better acquainted with the sonic workspace we all utilize. With a good orientation around the spectrum, you’ll not only have a much easier time with the whole concept of multi-band compression, but this may sharpen your ear for so many other recording endeavors. Many DAWs have a variable frequency tone generator or oscillator that you can sweep continuously throughout the entire audible spectrum. By sweeping through very slowly and noting the basic low/low-mid/hi-mid/high ranges you’ll start to develop an ear for what frequency areas are responsible for something to sound harsh, muddy, loud or bright. Pay particular attention to the mid range where much of the pertinent guitar tonality lives. Initially it may be difficult to correlate what you’ve learned from the variable tone generator with an actually guitar signal. Some exploratory equalizing can, in a very similar manner, help you isolate a particular frequency area within a guitar track in question. Send the signal into a fully parametric eq (that gives you the ability to adjust the bandwidth in addition to the frequency and level). Select a very narrow bandwidth, boost an extreme amount of db (around 12) and sweep through the spectrum once again this time listening for an exaggerated response from the tonality of the guitar track. From this you can pin point the frequencies that you may want to highlight with some spectral processing later.
Another prerequisite for by-the-book multi-band processing is a basic understanding of compression. It’s surprising how many engineers have a misguided notion of what a compressor actually does. In order for you to get the most out of spectral dynamic processing, it’s good to know precisely what is meant by ratio, threshold, attack, and release. Here’s a crash course: the ratio setting asks the question: how much compression do you want? The threshold allows you to determine the level you want the compression to start. The attack adjusts how quickly you want that compression to kick in once it has crossed that threshold. The release refers to how long that compression is applied after the signal falls below that threshold. There may be a disconnect between knowing the theory and being able to apply it seamlessly, but the more that this functionality becomes second nature, the more effective it will be to apply these same parameters to different frequency regions of the same source guitar track.
When armed with your refresher course of the frequency spectrum and compression, creative multi-band compression is definitely less intimidating. I look at this type of treatment as an alternative to just eqing the guitar for tonal enhancement rather than a level containment device to controlling overall peaks. Hypothetically, as a result of your exploratory eq sweep, you may determine the frequency range that is most responsible for an electric guitar’s warm tone is 200HZ-600HZ. You may find that this warmth isn’t always consistent from chord to chord. This aspect of its tonality could be solidified with some moderate compression and gain make up that would be applied only to that frequency range. You may choose to use this technique for another frequency band in the upper midrange to smooth out a little of the bite. Bear in mind that a slower attack time in this area will preserve the natural attack of the guitar while emphasizing some of character that guitar possesses in that range. A little dynamic equalization like this can go a long way. It’s definitely not a surgical kind of processing as multi-band compressors’ frequency zones are governed by crossover points that slope fairly gradually.
Other Spectral Suggestions
Here are a few multi-band tricks you can try the next time you’re in mixing mode. It seems that for singers, A & R guys and family members of singers, the vocal is never loud enough. Even in a sparse acoustic guitar and vocal arrangement, it can be difficult to place that vocal to satisfy everyone short of giving them an A cappella mix. Multi-band compression between 900HZ-4K on the acoustic guitar set to enact gain reduction keyed off of the vocal can really help in this scenario. You’ll be effectively ducking that frequency range of the guitar that competes with the intelligibility of that vocal only when it’s needed.
Everybody wants big, thick guitar tones, only for a lot of that body to be carved away with a high pass filter in the mix. Rolling off everything below 120HZ may kill off all the unnecessary sub-sonics and rumble, but at what cost to the tonality of the guitar? Multi-band limiting can be a viable alternative to using the roll off filter or radical shelving EQ. Set a frequency band between 20Hz and 120Hz and have some extreme gain reduction occur when anything offensive crosses the threshold. Essentially, it will act like a high pass filter only when there is reason for one to be there.
Columbo vs. Michelangelo
Multi-band processing is undoubtedly a powerful and creative tool for audio processing. Applying this to your guitar tracks can give you more of what you love about those tracks in the first place. Whether you’re tracking down the sonic culprit or sculpting a unique tonal presentation out of a raw guitar performance, a good grip on spectral processing can certainly give you another way for your guitar tracks to sit comfortably in your mix.