What Is A Broadcast Quality Recording? Part 9: Mixing To Broadcast
By Jerry Flattum - 02/03/2005 - 06:49 AM EST
What Is A Broadcast Quality Recording?
Part 9: Mixing to Broadcast
Dynamics in Mixing
The more tracks recorded, the greater chance for loss of dynamics.
But, in recording samples for instance, the problem might not be in the mix. For instance, the problem is a boomy bass. Instead of EQ’ing the bass to eliminate the boominess, simply switch the current bass sample used with another more percussive bass sample. The same technique applies to an analog bass signal as well (switching bass guitars).
Dynamic subtleties of musical instruments and individual voices are lost in a complex mix. Even in vocal harmonies, the harmonies themselves drown out many of the qualities of each individual voice. However, in harmonies, sometimes you want all the voices to sound as one.
The sound of fingers sliding along the strings of an acoustic guitar playing quietly is considerably different than the sound of a metal guitar played at maximum volume. Each guitar requires an entirely different processing. Different mics, different levels of EQ, even different mixing boards are used to capture the special sounds of instruments.
Using compressors, limiters, expanders, gates and other f/x is considerably challenging, but makes the difference between an amateur mix and a master quality recording. Mastering dynamics is subtle, unlike using a wah-wah or vocoder f/x. And, mastering is as much about how a mix “feels” as it is about how it “sounds.”
A technique to get perspective and separation within a mix is to create a separate space for each instrument and/or group of instruments. This is largely done through reverb and effects. For instance, dedicate a single reverb unit to the drums. No other instruments are sent through this effect. This creates a space for the drums. The sound of the snare is often used to find the right reverb for the entire drumset, usually a bright reverb with a short reverb time.
The goal is to create a series of sub-mixes. One submix might be the drums and bass (with a focus on processing the low end). Each submix can have its own effect.
Most mixers will say mixing and mastering is not possible without equalization. It’s used during all phases of the recording process. Individual tracks, submixes, stereo mixdowns, and finally the masters, are all EQ’d separately. Mastering "EQ's" equalize all the frequencies, from lows to highs. The process should bring out the dynamics, not bury them through over-compression in the desire to make a recording louder (commonly done in pop and rock recordings).
EQ’ing is a question of “tonal balance.” EQ adds or subtracts specific frequency ranges (low to midrange to high). Parametric EQ is made up of several bands with each band acting as a filter. Every sound and every aspect of the recording process has an equalization curve, e.g., the curve shaped by the positioning of the bands. Each band has separate controls: frequency (where the center of the band is placed), bandwidth (represented by the letter Q representing a range of frequencies), and gain (volume).
Usually, the wider the band, the less gain is needed, or, the higher the Q, the narrower the band of frequencies (Narrow Filter). In turn, a Broad Filter affects a broader range of frequencies.
Sometimes the bands of an EQ are referred to as filters: low-bass filter, band-pass filter (a range of frequencies) and high-pass filter. There are dozens of other kinds of filters. See: http://en.wikipedia.org/wiki/Electronic_filter
Clearly, EQ offers tremendous control over the shaping of individual tracks and the overall sound of a recording. But playing with frequency ranges is a delicate operation. Changes in one range of frequencies generate changes in another range. Plus, the ears can play tricks. What you hear one day is not what you hear the next.
The range of human hearing is estimated at 20-20,000 Hz (hertz).
The use of such descriptors as “muddy” or “shrill” directly relate to a specific range of frequencies. This illustrates how the ear is used to evaluate the sound while technology is used to alter it. EQ can be used to “roll off” noise in the bass, add “presence” to the high end and eliminate “pops” and “hisses” in the vocals. Spectrum Analyzers are often used to show graphically the shape of an EQ curve. There are a number of descriptors used to describe the goals in a mix and the ultimate sound after the mix. The mix might sound “punchy,” or “smooth.” A goal might be to capture the “groove” or the “ambience.”
The better job of using reverb to create a sense of space on individual tracks will mean less use of reverb on the final mix. The use of reverb in mastering, then, can be like adding an overall finish; a “shine.” There are many types of reverbs: plates, springs, reverse and gated. Too much reverb results in a “wash.” The different reverbs have different sounds. Some sound like rooms (bathroom, concert hall). Others sound “lush” or “flat.” Sometimes reverbs are used to create special effects, but this is generally applied to individual tracks and not in mastering.
How sound bounces off walls (like a padded cell) or reverberates (bathroom) is the study of acoustics.
Reverb units offer as much intricate control over how reverb is used as EQ units do when equalizing frequencies. In fact, from EQ to Reverb, Flanging to Sonic Maximization, each effect is a study in acoustics by itself.
A mixer can apply processing to individual bands or frequency regions of the mix using multiband f/x. For instance, an engineer can use compression on just the dynamics of the bass region, or widen the stereo image of the midrange. Or, the engineer can filter the bass with an equalizer, route the filtered output of the equalizer through a compressor, then mix the output of the compressor back into the mix. This can be tricky since bands of a multiband f/x have crossover points.
The signal is split or filtered into different bands using either analog or digital filters, or both. Then, multiband f/x are applied independently on at least four separate bands. Each band represents a region of the mix (low, mid-low, mid-high, high). Plus, bands have cutoffs allowing for even greater focus. The output of these bands can be muted to hear what is going on in the other bands.
Exciters are used to add sparkle or presence. Different exciters do different things like waveshaping, distortion or using delay. A small dose of distortion or “analog tube modeling” can create harmonics adding sparkle to the mix. Boosting the highs through EQ most likely will not produce the same sound. EQ brings out individual harmonics, but exciters create additional harmonics.
A similar comparison to how exciters work is found in the limitations of panning versus a dedicated stereo widening f/x. As always, it’s the ears that hear the difference. Stereo widening subtracts the left and right channels from each other, so the “signals” in those channels are decreased. Similar signals found in both channels are spaced in the middle, creating a “wider” sound. It’s a delicate balance since widening either the left, right or middle of the field will affect the other fields. Also, panning instruments while stereo widening can result in loss or gain. A multiband stereo widener allows greater control of individual frequency regions.
Hotter Is Not Better: Compressors
Some musicians think mastering is simply making a recording hotter, or louder. Many mixing and mastering engineers complain that there is an overuse of compression in current commercially released recordings. Compression is used to make the overall recording louder, under a misguided belief that "louder is better." The complaint is largely about the loss of dynamics that occurs when too much compression is used.
Sometimes a loudness maximizer is used instead of a compressor/limiter/expander. Used effectively, loudness maximization or compression can add fullness and depth to the mix.
Normalization is used to find the highest peak in a mix, and then adjust the level of the entire mix so the highest peak is at 0 dB (before clipping occurs). It appears compression is turning down the level of the mix by compressing peaks. By limiting the peaks above a certain point (Threshold) the rest of the mix can be brought up to the O dB level. Finer tuning is possible with Attack and Release adjustments, controlling volume before and after the Threshold.
Compressors are used to maintain volume levels and dynamic ranges of either the entire mix, or applied separately to vocals, drums (or other instrument) and/or the main stereo out. As with all other f/x processing, compression must be “automated” across the mix or individual track to compensate for rising/falling vocal and instrument levels.
Limiters are considered another form of compressors, generally used to set compression ratios for the highs in a mix or signals above the Threshold. Expanders are also compressors used to control the lows of a mix or the signals below the Threshold. Expanders can also bring out the soft parts of a mix, like a soft attack on a drum. Or, it can be used as a noise gate bringing out signals above or below the Threshold.
A combination compressor/limiter/expander allows greater control various levels in the mix. A single band compressor (combination limiter/compressor/expander) can be used to apply dynamics processing to the entire mix or the entire range of frequencies. Or, you can set the compressor/limiter/expander settings for a band of frequencies, as with any multiband f/x.
So, mixing is very much about controlling diverse levels at the peak 0 dB Threshold without loss of dynamics.
A multiband compressor can get more bass out of a mix compared to just boosting the low end with EQ. It affects the dynamics, not just the frequencies. It can also control unwanted noises like vocal pops or bass drum foot pedals that click or thump too much.
Automation is critical since a mix is not static, but changes over time. What happens in a bridge or solo section requires a different mix than in the verses or choruses. Certain parts in a song may need to be highlighted, like a horn “stab” or short guitar “riff.”
Recording drums, for instance, often involves an overhead microphone and individual microphones for each part of the drum kit. Balancing the sounds captured by the various microphones takes place over time, with attention paid to dynamics (a boomy kick drum, crashy cymbal or “crack” of a snare).
A mix is as much about left and right as it is top and bottom. And again, there is interplay in these regions over the span of the song. Also, it is not just sound but energy being mixed. A bass drum puts out tremendous energy while a hi-hat has little bottom.
Submixes can be EQ’d to make them sound apart or tight. Even though instruments and submixes have their own space, the idea is to have it all sound together as a whole. Most automation functions have history lists. This allows comparisons between alternative mixes.
Some mixers use meters, others don’t. Meters are useful for seeing level changes, distortion and other elements of a mix when the ear alone misses something. Meters also allow for more accurate monitoring of changes taking place across time (comparing verse to chorus or track to track.
A mastering house is where the recording, after it's been mixed, is further processed for replication and commercial release. The recording studio provides a mix not a master.
A mastering facility is concerned with the overall quality of a CD and its contents, not an individual recording of a song. The "mixing" engineer must ensure the recording has reached maximum quality of sound before a mastering house can work its magic.
A mastering facility ensures continuity of volume for all the cuts on a CD. Songs are placed in a certain order. Packages are assembled (artwork, case, shrinkrap, lyrics). Adequate space is added between songs (3-4 seconds). EQ and f/x is applied if missed in the mix.
Mastering is a remix of the mix. But, if the recording or mix is poor, the problem might not be fixable through mastering. No doubt there are engineers who feel they can fix anything, and probably can, given the staggering array of processing equipment available.
Mastering vs. Manufacturing
A mastering house might not be where the CD is manufactured, but it may be where the "glass master" is prepared. The glass master is then sent to the manufacturer where copies are made, the package is assembled, and the CD is readied for shipment.
What Happens At Radio
The difference between a mastered song and an unmastered one really becomes apparent when the two are played back-to-back over the radio. Radio stations further compress songs in ways that poorly recorded or mastered recordings sound even worse.
Live Reproduction of Studio Made Recordings
Sometimes performers replicate in a live performance precisely what they recorded in the studio. Others prefer audiences to hear something special not heard in a song or on a CD. One strategy is designed to increase the sale of a song; the other to drive audiences out to live shows.
It is a rare talent for someone to be able to hear a song in its raw state, and know the song has the potential to be a hit with the right arrangement and right artist recording it. To be ultimately successful, the recording of the song must meet the highest standards of audio quality.
Who records a song makes all the difference in the world as to the success of that song. Once an artist records a song, many other artists choose not to cover it. They prefer to release an original song. In many cases, we’ll never know what a song sounds like performed by someone other than the original artist. It’s hard to say if many of the most popular recordings so cherished by millions would have been successful if recorded by someone other than the original artist.
And when it comes to judging songs and artists, did A&R execs know that Shania Twain’s songs, co-authored with her producer/husband Mutt Lange, would outsell virtually every other country artist currently on the charts? Are Twain’s songs “great songs?” Are they better than someone else’s songs? Are her songs so great they would’ve achieved the same success if recorded by another artist? Does the world love Shania Twain or her songs? How well are her songs recorded, mixed and mastered?
Or is it the alchemical mix of artist, song and recording that makes a song a hit?
Beyond Broadcast Quality
The world of independently produced music carries many dilemmas for songwriters and artists, not just in the quality of the recording, but also in how it is marketed and distributed.
Can an independently produced recording get sufficient radio airplay? Can it compete on the Internet? Can it be distributed in large quantities across a global retail landscape?
Going the inde route means competing with commercially released recordings made by the best artists, producers, engineers, musicians, arrangers, recorded in the finest studios, with the most technologically advanced equipment available. They must produce a package rivaling those made by the best graphic designers using the best materials available.
If the goal is a million-seller, an independent must figure out a way to market and distribute to retail outlets and radio stations on a global scale. Administering copyright means forming alliances with ASCAP, BMI, HFA, etc. And it means negotiating fees for mechanical licenses, performance rights and sync rights.
Can a broadcast quality recording be produced in a home studio environment? I’m not sure yet.
[ Current Articles | Archives ]