The San Diego Troubadour

Get the Flash Player to see this message.

  

The Zen of Recording

Putting the Squeeze On: A Compression Primer

Of all the tools we use in our daily recording existence, none comes into play on the average session more than the venerable compressor. How ironic it is, then, that this is one of the most mysterious and misused instruments we employ (besides the accordion, of course).

With proper knowledge of the process, compression has the ability to transform decent-sounding home recordings into very professional tracks. In the wrong hands, it can render all your hard work into a well-intentioned, unusable mess similar to my sister's Friday night spaghetti, but even less tasty.

There are a few different stages that comprise the overall compression effect. These work in concert with one another, making the process a collaboration of sorts. Complicating matters further is the fact that not all compressors are designed with the same number or types of stages. Some have two knobs, while others have more than ten! Generally speaking, there are about seven on average: Input Gain, Threshold, Ratio, Slope, Attack, Release, and Output Gain.

Input Gain is simple: The level of the signal coming into the unit. As with most devices, you'll mostly want to have the loudest signal possible without distortion, including the loudest "peaks."

Threshold represents the level at which the effect begins to do its work. For instance, let's assume you've set the level perfectly, so that only your loudest peaks are hitting 0 dB (the maximum without clipping or distortion). Setting the threshold to -9 dB means that once something goes above that set level, the compressor begins to affect the signal, while everything below it remains unchanged.

Ratio controls what or, more specifically, how much of what happens next. If a compression ratio is set to 3:1, that means that for every 3dB of volume increase above the threshold, the perceived increase will be only 1 dB. The compressor is essentially turning down the loudest parts of the signal. This ratio is a very common and usable one and is a very good place to start when familiarizing oneself with the concept. Lower ratios are generally used to "thicken" a signal by bringing the sharper attack transients (like the picking of a guitar string) slightly closer in volume to the rest of the signal (like the sustaining strings). In the guitar example, 3:1 might work quite well, whereas denser material (like full mixes) would probably fare better with ratios of 1.5:1 or 2:1. Vocalists can be very inconsistent with their volumes, in which case you may need to use ratios between 4:1 and 5:1. Higher ratios like 10:1 and infinity:1 are commonly known as "limiting." This means that anything over the threshold is going to be the same volume. Kind of like hitting a ceiling. This is helpful for not letting something get past absolute 0, as in digital recording. Limiting is often used in the final stages of mastering to give mixes the hottest level possible. Overuse of this ratio by setting your ratio too low can result in a thin, squeezed, unexciting and dynamic free signal, and is one of the most common mistakes that amateur recordists make. The general rule of thumb is that you don't want to really hear the effect, you just want to reap the benefits of it, i.e. a more controlled dynamic range.

Slope can make things slightly more complicated. There are two types: hard knee or soft knee. Soft knee can best be described as a gradual increase of the ratio. As the signal gets farther above the threshold, the more the ratio increases. In other words, if you've set your ratio at 4:1 with a soft knee, with a threshold of -12 dB, your actual compression of a signal at -9 db might be 2:1 and a signal at -6 dB may be compressed at a ratio of 3:1. This gives a more "invisible" quality to the effect, which many listeners prefer. Hard knee is a stricter type, working more "to the letter," as it were.

Attack specifies the amount of time it takes for the compressor to start working. Faster times indicate that it will start working immediately, while slower attack times can play a crucial role in making things sound "punchy." A bass drum compressed with a fast attack can make it much more controllable but a bit dull. By lengthening the attack time, the "click" of the beater and the exciting "thwack" of the attack can be allowed to slip through, while the overall loudness can still be managed. Toms, electric guitars, and bass all tend to sound better with slower attacks.

Release defines the amount of time it takes the compressor to "let go" of the signal once it has gone back below the threshold level again. Too fast a release of a highly compressed signal will cause a pumping or "breathing" effect, as the perceived volume is turned back up too soon. Too slow, and you may lose some clarity, as the compressor may still be turning the volume down when the next note hits. Again, you're generally looking for a more natural sound, so working with these two parameters is very helpful in achieving sonic bliss.

Output Gain (sometimes called Makeup Gain) is a way to correct the loss of volume introduced by the compressor's inherent gain reduction, but here's the good part: because the loud peaks that normally poke out have been controlled, you can turn it up louder.

Proper use of compression means your vocals can sit higher in the mix without riding levels as much. Same with bass. For snares with a lot of "ghosted" grace notes, compressing only the harder back beats will be like turning up the softer parts, making them come more alive. Drummers love this. Well?that, nachos, and shiny things.

Sven-Erik Seaholm is an independent

producer owns and operates Kitsch and Sync Production (www.kaspro.com) and performs solo (www.svensongs.com) as well as with The Wild Truth (www.thewildtruth.com). He also enjoys squeezing his lovely wife, Gaily.