Mastering

by Paul Stubblebine (Published by Mix Magazine)

A little over ten years ago, when the CD format was introduced in this country, there was a lot of talk both inside and outside of the mastering profession about how that would change fundamentally the nature of mastering. There seemed to be a supposition that most of the mastering engineer's efforts were directed at optimizing, or compensating for, the LP medium, and that without the limitations of that medium there would be no more call for the engineer to treat the sound in any way. I recall one letter to the editor in a consumer audio magazine suggesting that "all those knob-twiddling mastering engineers" would soon enter forced retirement.
Looking at mastering now, with ten years of CD mastering under our belts, I don't perceive that the prediction of the first paragraph has come true, but I am intrigued by the way that a couple of other trends have quietly been expanding the way we think about our craft.

Fundamentally, I think I'm doing the same job I was doing ten years ago, and I think most mastering engineers would agree. Although I spend a lot less time carving a groove in a piece of lacquer, and a lot more time in front of a computer screen, the rest of my job can be described exactly the same way: I am taking the artist's stereo mixdown and trying to put its best foot forward, with an EQ nudge here, a level trim there, some help on a fade, a nip and a tuck, whatever this particular project needs to make it come across best. And of course I am providing a master in the form that makes it easiest for the manufacturing plant.

I still do EQ touchups on a large portion of the projects that come through. I don't feel that I'm EQ'ing to compensate for the CD medium, and in retrospect I'd have to say that very very little of the EQ'ing I did in LP mastering was to compensate for the medium. What for, then? The answer is always "to make the musical message come across better," but beyond that there are a couple of interesting observations. Over the years one of the most consistent reasons for EQ has been to get it back to what the producer and artist thought they had in the control room.

There may be many factors that contribute to getting fooled by what you hear in the control room, among them: fatigue, mood-altering experiences, and just being too close to your own project to hear objectively. Still, I believe the biggest factor is that the sound in control rooms varies all over the map. After ten years of digital recording and/or recording for a digital end product, what has the digital revolution done for control room monitoring? As far as I can tell, it has increased most people's sensitivity to hiss, hum, and buzz in the electronic system, but it hasn't revolutionized the control room monitoring situation in any way. Of course, every studio owner will tell you that " the monitors in my studio really work, they tell you exactly what's on the tape, you'll have no surprises when you take it out of here, the monitors translate perfecly to every system on the planet, etc." If the phrases sound familiar, it's because every studio owner is required to memorize them before opening his doors, just like Tim Robbins working on his cliches in the movie Bull Durham. I'll even tell you the same thing about the monitors I work on. Just a quick reality check: we can't all be right.

Actually, one of the most significant trends of the last ten years is that more and more of our recording and mixing is taking place outside of what we traditionally thought of as studios. That's right, I'm talking about the trend to home studios and (buzzword alert) project studios. I'm not arguing against this trend particularly (and I'm sure the trend is going to continue with complete indifference to whether I like it or not anyway.) I'm merely pointing out that it increases the variability of the conditions under which the tapes were monitored before they got to the mastering room.

Parallel to that trend over the last ten years we have seen an increase in the end-user listening environments that we have to consider. Back then, we concerned ourselves primarily with how the record would sound in the "average" living room and how it would sound on the radio. Now a large part of the audience listens on Walkman-style headphones. A substantial part of our audience is listening in a car. We all hope that our song will eventually be part of a movie soundtrack and be heard in large theaters on six-track surround systems. For many musical genres, the way the video clip sounds coming out of a single television speaker has a big effect on its ultimate success. The "average" living room was probably always a myth, but nowadays we have to assume that a home system could be anything from a boombox to a rack system to a high end audiophile setup to a surround-sound home theater. The audio processing used in radio is so extensive now that it's anybody's guess what your record will sound like on any particular station.

If we just look at those wildly different listening situations and ignore for the moment the rest of the places our record might be heard (elevators, supermarkets, third-generation cassette in an aerobics studio, etc.) it still leaves us with a large question: what is it about a good recording that makes it sound good under all those different conditions? I submit that it is still true that a good recording does sound good in all those places. Of course it won't sound as good on a boombox as it does on a great pair of speakers, but it will sound noticeably better on that boombox than a mediocre recording will, and the essence of the track will come through.
So what gives a recording that quality? Let's mention the non-engineering part first: a good arrangement, where the parts are not fighting each other for space. A good song, that lets a good arrangement fall into place easily. And great players, who get great sounds out of their instruments. If you've got all that going for you, even a rudimentary engineering job is going to produce a decent recording. In fact, in such a case a less-is-more approach might be best. However, there is still the question: What makes a great engineering job great, or a mix great?

You think I have an answer for that? I wish. I've been pondering that for years (on and off, not 24 hours a day,) and I've only got a couple of insights worth sharing. One is that an engineer must have a feel for the music s/he's working on, because the feel of the music is what the recording must convey. You have to be able to recognize which instruments are really creating the groove, and build your mix around them. Additionally, I'd like to pass along an insight I gleaned from Tom Flye, an engineer with lot of successful recordings to his credit. He said that when he's balancing a mix, he compares it to a standard he has in his mind, and that standard he described as a symphony orchestra. He uses that standard in mixing many kinds of music that on the surface have no similarity to symphonic music: rock, country, R&B. I think he was referring to the balance of bass to midrange to treble instruments, the balance of main melody lines to secondary lines, and the scale, or "bigness," of the overall presentation. Whether you use a symphony as your reference is not the point. You have to have a mental image you're shooting for, or your result will never rise above the ordinary.

As long as Mix magazine has lent me a soapbox, I'd like to point out one of the great paradoxes in mastering currently. While on the one hand the technical folks are working hard on 20 bit converters, dither schemes and Slicker Bullbleep Marketing in order to give us more dynamic range in the CD format, in the day-today reality of mastering we are trying to squeeze everything into the top two or three dB of the 90-plus dB we've already got. Yes, friends, the level competition we remember so fondly from LP days is still with us. There are producers who are obsessed with having their record louder than everbody else's. Since we all have the same absolute limits to work within, the inevitable result is compression, limiting, more compression, and more limiting. I offer one bit of unsolicited advice: if you're that worried about whether your record is going to be a hit, go write a better song.

In closing I'd just like to say that, although it should go without saying, all the above is just one person's opinion. If you've got a different opinion and yours is better than mine, try to express yours better than I did and put it in a letter to the editor. I'd like nothing better than to see letters every month from people in our industry examining what we're doing and suggesting ways to do it better.