First, let’s talk about "generic compressors and limiters" before progressing to Waves specifics. All compressors automatically turn down the volume when the sound gets too loud. You determine the "too loud" point by adjusting the THRESHOLD. Some compressors have an INPUT GAIN and some have an OUTPUT or MAKE-UP GAIN (same thing).

The Waves L1 output is automatically adjusted and doesn't need a conventional MAKE-UP GAIN. We use the MAKE-UP gain to turn the volume up again after the compressor has turned it down. That is one of the greatest advantages of compressors - the compressor only turns down the loudest notes and allows the rest of the music to be turned up without distortion. It also smoothes and holds back the most dynamic passages. This is the price. An INPUT GAIN is rare except on old analog compressors and usually means there is no THRESHOLD control. You turn the input up until you see and hear it compressing (as you want) then, turn up the OUTPUT GAIN until you are driving the tape machine or console at a good level.

Back to top

Limiting
Some units are designed to prevent the sound from getting louder than a certain level. We call these units LIMITERS.

Ratio
We often just want the loudest notes tamed, not broken. This is where compressors come in and with them a RATIO control. All the RATIO control does is set how quiet we want the loudest stuff. If RATIO is set to 1:1, then nothing really happens. Might as well hit BYPASS. At the other extreme, let’s say a ratio of 20:1, then we are probably limiting rather than compressing. In fact the sound would have to get 20dB over the threshold before the compressor lets even 1dB more out. That’s what the 20:1 means. If the THRESHOLD is set to 2:1, then it only requires the sound to get 2dB hotter to let 1dB more out. 2:1 is a basic easy compressor setting. It tames the dynamics without killing them and allows you to turn up the general level with the MAKE-UP GAIN. Turning the RATIO control from 1:1 and up is like dialing in the amount of reduction - much like setting a more sensitive THRESHOLD. Except not quite – the THRESHOLD determines where to start compressing, or how loud the input has to be before anything can happen. The RATIO controls how much to pull down. Both have a dramatic effect on the amount of compression and how audible the effect will be. Every control except MAKE-UP GAIN seems to influence the compression depth and amount of dB’s showing on the meters.

Attack
ATTACK and RELEASE are often the least understood controls. While these affect the depth of compression, we adjust them for other reasons and then use the THRESHOLD and RATIO to compensate. I mentioned that compressors grab the loudest notes. If you are compressing a typical pop music mix, those loudest notes will be the drums, maybe some important vocals or maybe a cool popping bass bridge. Compressing them won’t sound so good. What you really wanted was just a hot mix because it sounded a bit thin in places.

Here is where the ATTACK control helps (or should). If the ATTACK is set fast then it will "see" and "react" to every sound over the threshold, even the shortest. These transients don't give us much impression of "loudness" but will trigger the compressor. As you turn the ATTACK from fast to slow you seem to see less and less compression but really the compressor hears less drums and short duration sounds but still has those long notes. Visually it might be compared to sun-glasses that shield certain rays out. Some say "that is the goal, reduce those transients". But then we hear "those were the drums" from the musicians! The ATTACK control lets you set the compressor to avoid messing up the mix. If it is set fast, say below 30ms, it will react to the drums and reduce based on them. If set longer, 50ms to 200ms then it won't see the drums (much) and the compression will react to the overall loudness of the mix – which is generally preferable. The most common mistake engineers make using compressors is to set the attack too fast and lose the mix or highs and presence. Those transients contain a lot of highs. If transients are reduced, then so are a chunk of highs. You should be able to adjust how loud the drums are "featured" by where the ATTACK is set. The only drawback is that a slow attack will allow some transients to get through. This is why "true" limiters don't have an ATTACK control. We are depending on them to not pass transients. Can we limit or have a real fast compression and still have powerful drums? The RELEASE might help.

Release
The RELEASE sets how fast the level rises AFTER it was pulled down by the compressor. The next quiet sound should rise like a mini-crescendo. How fast or slow it rises depends on the RELEASE. Let’s say both the ATTACK and RELEASE are set fast. A drum triggers compression, i.e. the volume comes down, reduces the drum and then brings the volume back up before the next note. It may be louder but we lost too many dynamics and it doesn’t sound right. If the RELEASE is set slow, after the volume is pulled down - it tends to stay down and stay level. After that first hit the mix tends to be stable. As a quiet section is entered, less compression will take place and the sound gets gradually louder. The bad news with a long release is that we can't seem to obtain as hot a signal as we hoped. The optimum release depends on music and personal taste.

Pumping
I bet as a kid you tried playing with the volume control of the radio to the music. The compressor does that - except like a dumb machine. The results may remind you of your childhood, may be worse, may be "wow" or may be simply "disappointing". When you can hear the volume jumping around it is called PUMPING - sometimes this is great, sometimes it’s fashionable (1993 in the US) but it is usually better avoided because it annoys most listeners. Generally caused by a fast ATTACK, a RELEASE roughly in sync with the tempo, a hot drum mix, and over 6 dB of compression or limiting. Perhaps the easiest way to get a compressor to "pump" is to break all the rules, use the worst settings and make the compression as audible as possible - the sound you are looking for won't be far from that setting.

Modulation
There can be other artifacts generated by compressors. Most limiters sound questionable with fast attacks and releases. Some might add an ugly distortion called modulation. Modulation is when high notes are "squeezed" by the waveform of the low notes. It sounds like a watered down ring modulator along with erratic volume changes. Compressors are supposed to respond to the "apparent volume" and not to the waveform. Waves limiters and compressors use some inventive algorithms to avoid modulation and they allow faster settings than would normally be possible.

Overshoot
A different effect you might get using Waves and some hardware limiters is something called "Overshoot". Limiters usually just prevent a sound from going above some level, but with "overshoot" a louder sound produces less output – unleveled. Waves allows this by having ratio controls that continue to the other side of infinity (sounds like science fiction to me). Useful? Not likely. Another variation of overshoot that has less use is caused by a slow gain control element (common in Opto based limiters) and fast sidechain settings.

Back to top

Crunch
One of the more useful side-effects of some favored compressors is a quality commonly referred to as "Crunch". A few hardware examples are Urie 1176's and old Neve Limiters. It is a few flavors of distortion partially caused by the gain reduction element and often by partially clipping. It can usually be found with the output boosted just around the clip point. The L1 tends to crunch nicely if you drop the THRESHOLD low enough for 6dB or more limiting. You can do some serious creative "crunching" driving one L1 into another L1. The depth of each THRESHOLD is important as this creates the character of the "crunch". It sounds vaguely similar to some kind of tube distortion because the waveform is being shaped smoothly - not hard clipped. Also like tube amps, the second stage is distorting the crunch of the first stage. It's worth checking this trick out using the Renaissance Compressor. Turn up the MAKE-UP GAIN so that the output meter is "in the red". We will remind you again, but consider yourself warned - shooting for ever louder, hotter, raunchier sounds this way (with only a compressor) may be regrettable. As a texture creating tool on individual tracks it can be amazing. It tends to invoke "power" to everything from drums to sound effects. It helps bridge the gap between "sterile digital" and "warm analog" that adverts promote but is a gross generalization. Like most effects, it is best not to use it everywhere or too often because it can become boring and a crutch.

Back to top

Vocals
Spoken dialog and vocals can be the most challenging to compress. Acoustic instruments, like the piano or acoustic guitar also demand careful settings because the listener has a pretty good idea what these should sound like. It is easy for the audience to spot when it sounds wrong even if they can't say why. We hear real dialog every day and almost everybody hears guitars from time to time. Some producers want to be able to hear everything all the time (we hope the arrangement is good). Without the compression half the dialog may be difficult to hear. No compression and you need serious automation and time to get a great mix because some parts want to jump out here and there or disappear just as often. How do most engineers use compression on these kinds of things?

First on the list is VOCALS. The usual goal is to record a vocal that is neither too loud (in places) nor too quiet, retains some dynamics and emotion, sounds natural, and radio ready, etc. To achieve this we use light to medium compression. The key is not so much the settings on the knobs or sliders but how the GAIN REDUCTION METER looks. What you want to see is the meter going from zero to -3 typically and maybe to -6 on drastic peaks. We normally use a low RATIO like 2:1. Less, like1.5:1, will contain reasonable dynamics and be OK and 3:1 will definitely restrain dynamics. Set the threshold to get those few dB of compression on the meter when the loud phrases are sung. You won't need fast attacks because voices are not really percussion instruments (except for that ex-boss). RELEASE is the wild card. The best release setting is usually between 250ms (1/4 second) and 1000ms (1 second), but it really depends on the material. Aim for smooth. Assume the singer doesn't want somebody messing around with their performance. (Mess it up later, in the mix, when the singer is not there!) Listen to the articulation details. Things to watch out for are "headphone leakage" changes and even "feedback" if the phones are loud and compression deep. A good rule of thumb is that when the GAIN REDUCTION METER is moving a lot and quickly the compression will tend to be audible. Most of the time this is undesirable and a sign of inexperienced engineers. These techniques work for most natural sounds where you want to keep it natural sounding. The word is gentle.

Bass
How about BASS. Less rules here. A typical problem is an uneven bass guitar or player. Some notes loud, some not and emphasis in the wrong place. Here we could use a stronger RATIO like 4:1 or 10:1. Again watch the GAIN REDUCTION METER as you adjust THRESHOLD. You can get away with more dBs because who knows what the bass "should" sound like! Here, ATTACK can be used to fine tune the "pick" or first part of the note. This is also important when you have a "slapper" or "popper". A fast ATTACK is needed to grab those peaks and a slow attack can be used to "feature" the initial part of the note. The RELEASE is typically set moderate – fast enough to be ready for a quiet note, but not too fast. If the release is too fast, then it can boost the "between" notes noise. Watch out for making the bass just a "low drone" with no space between notes and no decay to the notes. If you use a bass amp and DI box, then a lot of variations are possible. If one compresses these signals individually, it may be like changing the relative balance on a note by note basis. Dangerous. You might want to consider "linking" the two compressors or mixing the two signals now and compressing that signal for a more consistent sound. Usually a combination of EQ, compression and creative miking is needed to get that bass sound you are hoping for. The most important elements are the bass player, then the instrument, then the mics & DIs. Compression will help but it is near the bottom of the list. Maxxbass by Waves is one more tool that will help create killer bottom while not pushing some low notes into the subwoofer damage zone.

Back to top

Drums
I was hoping you wouldn't ask. Throw away the rules. No matter what you do it will be an effect. The trick is making it a desired effect. There are a lot of different "tricks", we will describe four to give you some ideas.

1) Kick & Snare. You can "fool" a compressor into giving more snap, stick or beater by using a moderately slow attack and a release set to return to zero between beats. Use the ATTACK to set when in the note the compressor pulls down. A slow attack makes it late and lets the initial transient through. The THRESHOLD and RATIO depend on the consistency of the drummer. We suggest, that with less consistency, you use less of a RATIO because you are working in a "fooled" compressor mode and the amount of snap should be consistent, or if not, at least musically valid. The C1+ would be our first choice because it also has the noise gate, and EQ in the sidechain and a few other very creative options. If you don't need all that, the Renaissance Compressor will be the easiest, and the L1 is just too fast for this technique.

2) For that Zeppelin ambiance, limit the room mics - pretty drastically. Fast time settings, deep ratios, deep thresholds. Here the L1 is a good choice. Of course, if the drummer is a wimp, and the room is dead, it won't be nearly as much fun. It might work on overheads if you seem to be getting a good "kit" sound from them. Here that fast attack may be the wrong choice and you should use the C1+. You don't want to lose the snare. It also works best if the drummer uses smaller and thinner cymbals than they use for concerts.

3) Compressing a sub-mix of the drums possibly with the drum reverbs. A gentle approach on the sub-mix can lock in the sound of a kit rather than separate drums and effects. It also requires a variable attack, so the L1 is out but the Renaissance Compressor should work. The compression will affect the balance of the drums in relation to each other and the reverb so it may be easier to mix "into" the compressor rather than apply it as a last step. In practice, it means some interplay between setting the basic drum sounds and mix and setting the right overall compression. It is not easy and if you decide it's not working you almost have to start from scratch.

4) Using the compressor for distortion. Often a little or a lot of clipping or crunch is awesome on drums. It sounds more like increased upper mids than distortion and affects the balance between the attack and the ring of the note in useful ways. It even helps get some authentic vintage character on too clinical drums. The usual way is to "crank" the INPUT or OUTPUT (MAKE-UP) GAINS to drive the compressor or the next device (such as tape) into clipping. It can be done with or without compression. If you use some compression, then the usual goal is to make the distortion consistent because we are using it, like EQ, for a color change. The Renaissance Compressor is a great choice because it allows a great excess of MAKE-UP gain to be added which tends to clip the output. Two Renaissance Compressors do even more and allow more variations in drive color.

Compressing the mix
How about compressing the MIX? It's funny how many mess this up when it's so easy. First rule - consider how many more times it will be compressed by the mastering and/or broadcast chains. The common belief that it may be wiser to apply a lot of compression so they don't mess it up later is false. "More compressors" is just "more compressors" and the best approach is just a great mix that can handle a bit of squash and still sound great. Besides, trust the mastering engineer to have a great compressor, great monitors and daily experience of compressing just the right amount for the radio and CD pressing. If the mix is for TV or Radio, then there is a good chance you have a producer demanding "as hot as you can get it, and then a bit" and you know how much it gets limited by the station. Again, the best approach is the mix, but you rarely have time to "finesse". For broadcast and broadcast schedules, it is easier to mix into the compressor. In other words patch in the compressor before the mix starts and use a typical setting from your daily use. It helps to watch the GR meter, because it is all too easy to keep pushing up some fader and "for some reason" it never is quite loud enough. The L1+ is good for "loud" and fast, the Renaissance Compressor is typically more subtle and the C1+ may have more options than you really need. For big music mixes, we usually suggest the normal technique - that some gentle compression be patched and applied as a last step. To preserve the mix, use as slow attacks and releases as you can get away with. For more intensity use moderate to fast attacks and releases. For Netcast mixes you can take more liberties. This is the last stop before "data compression" which is an entirely different concept compared to "audio compression". For the net, feel free to run the mix through the Renaissance Compressor for gentle compression and dynamic control. Then use the L1 with appropriate IDR to optimize (not necessarily maximize) the final levels. Rather than run each mix with the same Renaissance Compressor & L1 settings, adjust each song, so that all songs end up at similar apparent levels (close your eyes) or at least "flow" well from one song to the next. And that was the last big hint - never forget to listen to & judge what you just did. No one cares what settings or theories you favor - they just want the mix to sound 'right" and appropriate to the emotional content of the music and the sound be in fashion or perhaps setting a new one.

Oh yeah, we started out by saying how mixes get messed up. Most people make a common error based on some rumor or some line in a magazine. They expect miracles from a compressor and when a little doesn't sound amazing then a lot must be better. Wrong. You were closer with "a little" but now it needs finesse. If the attack is too fast you lose the drums. It’s that simple. With the release set to look like a metronome, bouncing with every beat, it will probably sound pumpy and amateurish. Try slower attacks and moderate releases. With too deep of a ratio and/or threshold, you lose the dynamics and space. In other words the technique is to compress gently and musically. The closer you get to brick-wall limiting the more likely it will ruin your mix. Limiting is possible but must be used with care and as little as 2 or 3dB unless you are aiming for that squashed effect. When the schedule is tight and maximum loudness with intensity are important, it may be best to mix "into" the compressor. With a complex mix, it will generally be less difficult by compressing as a last step. For a killer professional mix, the most important thing is the basic mix. Choose your tools wisely and don't make mistakes. The compressor won't turn lame into lovely but a good compressor can make a good thing better. "Taste" is the key.

De-essing
This falls into the category of compression but has a bit of an EQ twist to it. We are only compressing based on the energy in a narrow bit of the highs where esses are strong. This is typically between 4k and 8 kHz. The best way to de-ess is prevention so that you won't need a de-esser. This is easy but probably not cheap. Use a great mic (hopefully tube condenser), a great mic pre (with minimal HF distortion), EQ higher frequencies or avoid boosting that 4K to 8K area and listen for problem sibilance as you are recording and with the experience that shows you generally add more highs in the mix. If the problem is a gap in the singers teeth rather than a gap in the budget, then a bit of cotton in the gap may help if the singer can tolerate it. If you have to use a de-esser, do it in the mix stage and patch it in after EQ and compressors. Avoid de-essing the mix or forcing the mastering engineer to. It is much harder to de-ess if the vocal is surrounded by hi-hats, cymbals, snares, guitars etc. It is easiest to use a dedicated de-esser and it hopefully does what you need. The goal is to reduce esses, not remove them. You should aim for "natural" - not the esses are gone. The THRESHOLD and RATIO depend on the song and the most natural effect. The ATTACK can usually be slow (say "stop" and "top"- notice how slow "the ess in ‘stop’ starts). Usually between 70 and 200 ms is OK. Faster than this and the de-esser reacts to other consonants or other sounds. The release can be a similar number. It is not to critical but again aim for "natural". Some de-essers compress the entire signal and some just reduce the highs only. The C1+ allows both options. We suggest trying SPLIT mode first so that you only reduce the highs. If you have to de-ess a mix we have two things to say - good luck and try the SIDECHAIN mode. For mixes the ATTACK will be critical. You may get a dB or so of de-essing even with no esses so remember to use a bit of MAKE-UP to prevent overall loss of highs. If you want to live dangerously, you might de-ess more and boost the Make-Up Gain more, getting ever brighter and yet not making your ears bleed. Before you get excited, take a listen and judge whether it sounds like some kid or machine is playing around with the treble control of a stereo. Maybe too much, maybe magic.