One of the consistent problems faced by the home recordist is how to get big-bucks sound out of low-bucks equipment. Although there are many pieces of equipment that can help make for better quality recordings, most of it is useless unless you have a clear-cut idea of what you're looking for, and then utilize the equipment to build that specific sound. Sometimes, just patching things around looking for something interesting can yield good results — and I'm certainly not discounting sudden chance inspirations — but you can certainly handle things much faster in the studio if you know what effect you're trying to achieve BEFORE you start plugging in the patch cords.

Some might argue that saving time is not really a consideration in a home studio, since you are not having to worry about the clock and you're not paying a costly studio bill. This is true; but the more efficiently you can work in the studio, the smoother the flow, and the lower the frustration level. The less time you take committing inspiration to tape, the better your odds of the inspiration not floating away as you twiddle knobs.

So, how do you evolve a clear-cut plan of what you want to do? Chances are you already have some kind of composition, or some kind of compositional idea, that you want to put into taped form. It probably also has variations between soft and loud parts, and between the instrumentations used to implement specific passages. All of this variety helps to keep the listener aware of the music, and interested in the general flow — but there is one often overlooked aspect of composition that can really make a difference in the overall sound. Since this concept doesn't have a name, we'll call it Frequency Balancing.

AMPLITUDE BALANCING

The basic idea of a mix is to balance the amplitudes of the various instruments used in a composition. Percussion, rhythmic instruments, and melodic instruments need to occupy their proper positions in a mix or the ear may tend to be drawn in a direction not intended by the composer. Subtly boosting an instrument can draw the ear to it; everybody reading this is probably familiar with boosting an instrumental passage, or vocal, just a little bit to increase the emphasis in a certain part of a piece. However, what's often missed is frequency balancing. Most boards have some kind of equalization facility, which can vary from simple to complex... it all depends on the board. Engineers will use this EQ to get a "correct sound", but I've never seen too much analysis of what tonal qualities make for a good "sound". So, let's see what we can do about quantifying things a bit better.

OTHER MECHANISMS FOR FREQUENCY BALANCING

In the old days, EQ was the only way to change tonal quality. With the advent of synthesized sound, however, all the rules have changed. Whereas EQ can only add or take away what is already present in an instrument, a synthesized sound can change the sound of the instrument itself, which can then be further modified by EQ if necessary.

What does this mean? Let's say you wanted to use a trumpet for a particular lead line, but that the sound was too bright. So, you take some top end off with some EQ and add a little lower midrange to smooth things out a bit, but the characteristic bite of the trumpet is still there... the EQ can modify the bite, but not remove it.

Now let's say you play synthesizer and you want a trumpet part... so you patch up your synthesizer to make a trumpet sound. Say you decide this sound is too bright. You can always change the filter response, of course, but this is still a change in EQ. A more convenient solution is to simply use a different waveform, with less harmonics, and perhaps even add some attack time to soften things a bit. Thanks to the synthesizer, you can now make radical changes in tonal quality before you even start to think about equalization.

The implications of this are perhaps obvious, and perhaps not... but they are certainly far-reaching. Not enough bottom on a sound? Don't boost the bass, add a tracking oscillator an octave below. More bite? Use a more complex waveform.

APPLYING FREQUENCY BALANCING TO A COMPOSITION

In my own pieces, I try to aim for a pretty wide-range sound; I like to have something occupying the bass slot, the various midrange slots, and the high frequency slots. However, having all these frequencies present at all times would be just as monotonous as playing at top volume all the time; varying the frequency content throughout a piece adds variety and interest to the flow.

For example, if I'm reaching for a peak in a solo, I'll start off with a more muted tone, and then as the solo builds to a climax, add some highs — or possibly kick in a distortion device to increase the harmonic content. In the meantime, suppose this solo is happening over a drum pattern that features a prominent bass drum. Since the bass drum tends to keep a listener "grounded" to the low frequencies, removing it for just a few bars towards the climax of the solo not only focuses maximum attention on the solo, but also temporarily disorients the listener by letting the rhythm float — thus making the listener more susceptible to the solo's climax. Starting off with a midrangy sound, then increasing the highs, then removing the bottom, and putting it back again later to signal the end of the solo peak can be though of as a frequency response progression as opposed to a chord progression, or a progression of dynamics in a piece.

By the way, this technique is not really limited to any one form of music. Even the very avant-garde types of music contain elements of rhythm, "soloes", and backgrounds; these are just as likely candidates for frequency shaping as, say, a band of acoustic instruments.

Another example: Let's say you have a piece that uses a string synthesizer more or less throughout, but that during the solo you want a very bright lead instrument over the string part. In terms of conventional amplitude mixing, you would therefore drop the string part back to avoid crowding the solo instrument. But wait — what's being crowded? Well, between the brightness of the average string synthesizer and the deliberately bright voice chosen for the solo sound, I'd say it's the high frequencies where the collision is occurring. So instead, try cutting off the highs on the string synthesizer during the solo, but leave the level up. The frequency response will be better balanced, and the solo instrument will stand out more. In fact, it will probably stand out enough so the listener won't notice the slightly duller string sound. What's more, when the solo ends and the strings return to their normal sound, there is a welcome change of pace that adds variety to a composition.

OPINIONS AND RECOMMENDATIONS

It seems to me that conflict appeals to the human mind, whether it's the physical conflict that athletes enjoy or the mental puzzles that engineers enjoy. In music, there are several conflicting forces which, if perfectly balanced, form "good" music. Some of the conflicts involve the steady repetition of patterns and rhythm vs. the ever-changing melody lines of solo instruments (a case of regularity versus independence), freedom versus structure (improvising around a fixed chord or rhythmic pattern), and dynamics — loud versus soft. Having these conflicts appear to be NOT in conflict, by working them against/with each other, is in some ways the essence of music. Harmony, for example, takes several different notes, yet makes them work together to form a more pleasing whole.

So, we need to avoid excessive conflict in terms of frequency response, and yet some conflict seems almost mandatory to keep the level of excitement up. Here are some specific situations I've run across to illustrate some points about frequency balancing. Naturally, these are not "rules"; any situation demands a unique treatment.

RHYTHMIC SOUNDS

Drums are an excellent example of a frequency balanced instrument. You have a bass drum for low frequencies, tom toms and the like for midrange, and cymbals for midrange to high frequency sounds. A drummer punches out not just a pattern, but an ever-changing frequency response progression. Here we already see the makings of a trend: lower frequencies are used for the rhythmic, steady underpinnings of a piece of music; midrange sounds (snare, toms) are used for continuity in the piece, and cymbals add an intense emphasis to selected parts. A cymbal crash, with its rich harmonic structure, is going to get a listener much more excited than a single hit on the tom.

GUITAR AND BASS

Guitar follows a similar pattern. The bass (rhythm) pickup is pretty pure in terms of waveform. If the guitarist wants to put the rhythm a little more up front (such as R&B type material), there will be a change to the treble (more harmonics) pickup. When it's time for a solo, what happens? The fingers gravitate towards the high frequency end of the neck, and sometimes a fuzz tone will be added to up the high frequency content. Again, the sine waves provide the rhythm, midrange gives continuity, and the higher frequencies are used to excite and arouse. (It's interesting to note that some background music sound companies roll off the high end to keep listeners functioning on a nice, steady, droning level.)

Bass becomes more noticeable if you increase the treble content; in fact, it seems anything with increased treble content becomes more noticeable.

From the above, we can draw a couple of conclusions. With a song, for example, if you want to mostly emphasize a vocal, use fairly simple waveforms for the various rhythm instruments and cut the vocal with a little brightness. If the vocal disappears to make room for a solo, make sure the solo is also a little bright so that the listener doesn't feel something's missing when the solo hits. Or, make the solo instrument a little less bright than the vocal at the beginning, and increase it towards the end for that good old tension/release effect. Remember, when I say brightness this doesn't necessarily mean EQ; it can also mean initial waveform.

High frequencies should be used sparingly. In some respects they're your ace card for drawing people's attention; it takes much less volume for a 2000 Hz tone to attract your ear than a 50 Hz tone. Also, remember that people hear best in the midrange region, so changes in bass and treble content will have to be less subtle towards response extremes than towards the midrange.

The ear tends to require fairly frequent changes in sound to keep it interested. If you use a fuzz all the way through a piece, it will lose its effect; the same is true with high frequencies. Don't start off too bright, or you'll have no place to go from there, any more than you would start a piece at a consistent maximum volume and stay there.

SUMMING UP

Above all, don't let the frequency response of a piece stay constant throughout the piece, or you'll have something less interesting than it could be. An ideal instrument for measuring frequency response information is a spectrum analyzer, but as you know they are EXPENSIVE and if you can afford one, you probably have a 16 track and are not even bothering to read this article... much less have any interest in home built gear!

However, as luck would have it I've devised a cheap little gizmo to help you out while mixing that monitors the frequency content of a piece of music. It's simple, but it's so inexpensive that it's really worth your consideration as something to add to your studio. Besides, it's fun to look at (the blinky light syndrome once again raises its LED head...). We're running out of space this month, so watch next issue for how to string together some filters and LEDs to help you monitor the frequency balance of a piece of music.

Copyright © 1979 by Craig Anderton