Mastering is one of the essential processes that must be carried out on any audio product that will be released professionally in the audio market or for picture (video, film, or DVD). In this series of articles we'll look closely at the different guidelines for understanding the world of audio mastering.
We say product: often, people involved in sound production, whether musicians or sound engineers, aren't aware that they're creating a product to be sold in the marketplace; they think they're simply dealing with artists, doing a purely creative job with no prospect of it ever being sold.
If we go to any supermarket, we'll see, for example, the dairy section, the cleaning products, the canned goods... Everything we find there meets various kinds of standards that are essential for it to be sold. Take a yogurt. Aspects such as its weight, preservation, packaging, labeling, and so on are obvious, but the manufacturer has also taken into account its consistency, texture, color, fat, acidity, pH, lactic acid bacteria present, sweeteners, preservatives, the health regulations to be met, and countless other aspects so that the product is marketable and, by the way, outsells the competition. That last aspect seems not to interest the music “creative sector” in the slightest, although that's a subject we can take up in the future.
In the same way, an audio product has to meet certain standards to be marketable. In this case, the characteristics of volume, EQ, compression, and audible range may be the most obvious, but no less important are its loudness levels, peak levels, normalization, jitter, DC offset, ISRC/EAN codes, and so on. And all of this, in turn, also makes up a creative field of its own, that of mastering, just as the container or packaging is an essential part of the product, beyond its contents.
For some time now, the mastering process has attracted unusual interest from musicians and engineers. With computers now omnipresent throughout the recording process, anyone at home with a PC or Mac can have a complete studio in which to create a product from start to finish. Yet when the end seems to arrive, they compare it with the competition they hope to rival (their favorite band) and it turns out that it doesn't come close. It's as if our grandma, with those delicious pastries she bakes, wanted to compete with a big commercial bakery for shelf space in the supermarket! So the musician/engineer fixes their interest on mastering as if it were a magical process, beyond the PC or Mac they own, that will save their recording by lifting it out of demo territory and turning it into a product. However, that won't be possible if the steps previously taken in tracking and/or mixing weren't the right ones.
From Creation to Recording
If we have the chance to be part of the creative phase, let's keep the following elements in mind: the concept of the song, the melody, the rhythm, the lyrics, the arrangements, the instrumentation, the song structure, and the performance. In addition, the audience it's aimed at, its possible and certain competitors, and how it will be distributed. If we have this under control, we're on the right track.
From Tracking to the Master (from Langley Park to Memphis, as Prefab Sprout put it). Steps Before Mastering.
We must give each phase of a recording's production the importance it deserves. Recording organization, tracking, recording, editing, premixing, mixing, and mastering are different processes, all of them chained together. Mistakes made in some of them will have to be minimized in later phases. If, on the other hand, each phase is done well, the last step, mastering, will serve to enhance the final product. How did we get to the master? How did we finish the mix? How was the recording made? Were the sound sources the most suitable ones? Where does one end and the next begin? We can go “upstream,” thinking through the different production phases of our project to identify possible anomalies. Yet we're always left wondering why, then, there are professionals who devote themselves exclusively to mastering.
Phases of a Production
Each of the phases of a production has its own characteristics and, at the same time, its limitations. In very simplified form, they are:
a. Choosing the source. If it's a musician, choosing the most suitable one or the most suitable performance. Later editing doesn't fix everything. A good musician and a great performance do.
b. Choosing the instrument: guitars, basses, drums, vocals, keyboards, samples, loops... always the best one for what we're after. Careful, because there are still people who don't know that drums need to be tuned!
c. Choosing the most suitable mic setup or line(s): a whole world in itself; whether by mic or line, the signal must flow properly.
d. Preamps or mixing consoles. They're not all the same. Quality always makes the difference. Signal routing.
e. Operating systems. A/D conversion
f. Editing: proper treatment of the material before it's mixed, in terms of timing or pitch. Take selection and comping.
g. Mixing: compression, EQ, effects, whether in software or hardware or a mix of both on each track. Stereo mix.
h. Mastering. And now, what?
Once a) and b) have been done correctly, we move on to points c) and d). Inappropriate gain settings will compromise this step. Low gain keeps the signal floating in the noise floor. High gain, constantly sitting in the yellow zone of our LEDs, means we'll have to use compression almost exclusively as limiting. We should be conservative at this point, leaving dynamic headroom so that steps f) and g) can do their work properly.
Point e): we won't go into detail on this point. Still, let's be aware that there are devices with a huge difference in quality and price which, as a result, will contribute vital characteristics to the sound. In an article by Rogers Nichols on the A/D conversion of the Panasonic SV3700 DAT, this great engineer described how the difference in the behavior of a single byte (!!!) changed the sound and made it distinctive of that unit.... Are we capable of identifying such microscopic differences?
Point f): improper editing, such as not cutting material at zero-crossings or not using crossfades, will produce digital errors that end up printed on the various tracks.
Point g). Let's remember that:
● Heavy compression will leave our sound without dynamics and therefore lifeless. Light compression can produce digital peaks.
● EQ changes the level of the track and our perception of it; therefore, it will also change how the compressor acts. How we should EQ, what we should boost or cut, is vital on every track.
● Faders, even in software, have a logarithmic character, and so the way we perceive level changes at different frequencies can substantially alter our impression of the mix. This can be even more critical when working with subgroups.
● The choice of plug-ins is also decisive. Plug-ins are there to be used properly, not collected. Fewer plug-ins means less signal processing. Let's use quality plug-ins.
● Stereo mix. In almost every DAW, there's a bottleneck at this stage. Bounce, or play back and record onto another track and/or medium? Does it sound the same? We can blame the processor, the operating system, the software we use, or the player for the stereo mix, but it's common for there to be a difference. A good mix, after a bounce to stereo, is most of the time reduced to its bare minimum.
And from here on, we hope whoever masters it for us will perform the miracle of the loaves and fishes. That's hard, with the variables mentioned above, and those were only the minimal ones. For all these reasons, we must give each of the steps described the importance it deserves so that we can reach mastering in the best possible shape.
Meter, Meter, Meter
Metering is a vital element in mastering. We can measure in several formats. The most commonly used are:
● L-R Peak metering: measures the signal's peaks, fast attacks that can make our mix generate clipping errors even if RMS metering doesn't show them.
● RMS or VU metering: average measurement of the signal.
● PPM, or Peak Program Meter, metering. This metering is usually found on high-end consoles such as SSLs. Basically, a VU meter specialized in peaks. Which one should we heed? As a general rule, it's preferable to always keep the RMS reading at hand, often shown on a classic VU meter. It has a response time of around 300 ms, very close to how our ear responds. Peak metering is merely a tattletale that we've exceeded digital 0. We should calibrate our equipment, regardless of which phase we're working in. Keep in mind that -14 dBFS (Full Scale measurement in digital terms) corresponds to 0 dB on an analog VU. If the material we're working on doesn't involve heavy compression (folk, jazz), we can calibrate as -20 dBFS = 0 dB VU.
With a PPM meter (with a typical response time of 12 ms), we can eliminate the fastest peaks through compression. Keep in mind that peaks lasting between 10 and 12 ms are, as a general rule, inaudible and therefore removable.
● Pan metering: how the signal is distributed in Peak or RMS mode across the horizontal plane. It gives us an idea of the L-R balance of our stereo master.
Listen, Listen, Listen
What is the smallest change we can perceive in level, compression, the application of an effect, or the relationship between several instruments? In the different phases mentioned above, we can make noticeable changes but, as we move forward, they'll become smaller and smaller. Are we able to hear them? Although it has to do with the quality of our equipment, our listening skills must be trained to perceive differences as small as 0.2 dB in the various adjustments we make.
As part of the process of appreciating different styles and eras, we can use various software tools to visualize existing mixes. BIAS Peak or Steinberg's WaveLab are very interesting programs. Try opening music in the styles you like most and least with them. Remember that we'll learn even more from all those styles we don't normally listen to. Note the style and the year the production was made. Check how you perceive the most powerful and the quietest passages. Look at the waveform on screen. Does what you hear match what you see? Look at its frequency content. Can you draw conclusions? We'll continue from here in the next installment (Mastering, Part II).
