by Craig Anderton
Fig. 1. A classical music file in Steinberg’s WaveLab 7. Processors include EQ and stereo width expansion; note the spectrometer window toward the top. LIKE MUSIC RECORDING AND VIDEO PRODUCTION, MASTERING—THE FINAL stage between mixing and distribution—has migrated to the desktop. As good as the tools have gotten, one reason why people still go to professional mastering engineers is to exploit their ears and experience, as mastering is a specialized art. It’s also good to employ fresh ears that haven’t been immersed in the project all along. Nonetheless, if you define mastering as “making the mix sound better prior to distribution,” why not give it a try? If you don’t like the results, you can always take your mix to a pro.
Most in-the-box mastering employs dedicated audio editing software like Sony Sound Forge, Steinberg WaveLab (Figure 1), Adobe Audition, BIAS Peak, and the like. These often include their own mastering-quality plug-ins, but you can also use third-party plug-ins from companies like Waves, Universal Audio, PSP, Voxengo, and many others. iZotope Ozone 5 is a suite of plug-ins and diagnostic tools (Figure 2) that’s a one-stop shop for mastering processors; IK’s T-Racks is a software mastering suite that can also add a vintage fl air. Many mastering engineers swear by Har-Bal, a stand-alone program with sophisticated EQ, limiting, and stereo separation options.
Increasingly, it’s also possible to master right in your DAW. Magix Samplitude has always emphasized this, and PreSonus Studio One Pro has separate tracking/mixing and mastering pages (Figure 3), with excellent album assembly options— if you edit a mix, the version in the mastering page will update automatically.
Some DAWs also include mastering plugins (like MOTU’s MasterWorks and Cakewalk’s Linear Phase EQ and multiband compressor), and provide at least primitive CD-burning capabilities. Nor are you limited to plug-ins— provided you have the spare outs and ins on your audio interface, almost all DAWs and audio editors let you send a signal to external hardware processors, which can appear in the program like plug-ins.
Th e process always starts by listening to the mix multiple times—get to know it really well before you start tweaking. But enough generalities. Here are some specific tips.
Fig. 2. iZotope Ozone 5 Advanced’s Meter Bridge includes several mastering-oriented diagnostic tools. Treat your room and use decent speakers. Accurate mastering requires accurate monitoring because you want a transportable mix that sounds good on all kinds of listening systems. For example, if you cut low end because your speakers’ bass is exaggerated by their being in the corners of an untreated room, your mix will sound fine in your studio, but won’t have enough bass when played back on iPod ear buds. For more on dealing with your room, see our feature on page 52.
Th e big three: EQ, dynamics, and everything else. I consider mastering about 85 percent EQ, ten percent dynamics, and five percent everything else (e.g., stereo enhancers and harmonic exciters). I’ve received acoustic projects recorded in a room with resonances, and used EQ to create narrow, shallow notches to take them out. EQ can emphasize or reduce the impact of individual instruments, like tame an overly-loud hi-hat without affecting other sounds, or boost a kick for a big dance mix. It can help compensate for mistakes caused by mixing in a room with bad acoustics, and add definition and sparkle. You can sometimes even minimize “digital brittleness” by adding a steep, high-cut filter starting around 15 to 18kHz.
Don’t squash. It’s ironic that even though digital playback media doesn’t have significant dynamic range limits, people act as if it does. Excessive compression or limiting may sound dramatic at first, but eventually gets fatiguing. Th e trick is finding the sweet spot between a level that’s competitive with other cuts in a playlist, yet retains dynamics. I like maximizer plug-ins with a continuously variable threshold, as I can close my eyes, move the fader back and forth, then park it in that sweet spot. Another trick is “micro-mastering.” If most of a mix’s peaks reach –5dB but a couple dozen go to –1dB, I’ll normalize just the half-cycle of those couple dozen peaks to –5dB, then normalize everything back up to –1dB. This gives a 4dB increase in level with no perceptible dynamics squash.
Don’t master while you mix. It’s technically possible to insert mastering-type plug-ins into the final master bus when mixing for “one-stop mastering,” but I advise creating a mix without any bus processors, then mastering it as a separate entity. You’ll approach the mix with fresh ears, and your original mix will always be available if you need to re-master it.
Fig. 3. PreSonus Studio One Pro has a page dedicated to mastering, which can include effects for individual tracks as well as master effects for all tracks. Downscale your changes. Half a dB of EQ added to a drum track isn’t going to make much difference, but on a final mix, you’re adding it to every instrument. If you’re new to mastering, whenever you make a change, immediately cut it in half. For example, if a 3dB boost at 4kHz sounds good, change it to 1.5dB and let your ears acclimate to the sound before you decide you need more.
Remember Newton. For every action, there’s an equal and opposite reaction. If you boost the treble, for example, you’re also de-emphasizing the bass.
Beware of inter-sample distortion. When the digital-to-analog convertor in your audio interface reconstructs a signal, it may exceed 0dB despite what the meters show. That’s because most meters measure the level of individual samples, but an analog waveform drawn from those samples may create an arc that extends above zero. SSL offers the free X-ISM metering plug-in that can alert you to inter-sample distortion.
Craig Anderton has mastered hundreds of tracks in the past few years, and is Executive Editor of Electronic Musician magazine.