By Craig Anderton
Lately, there’s been considerable controversy about mixing “inside the box” (ITB)—the process where all your processing, fader moves, and automation are done in the digital domain, inside your computer. In theory, ITB shouldn’t have any problems. But some insist that using analog summing junctions (or a “real” console) for mixing delivers superior sound quality.
What’s the truth? I believe that analog and digital, being different technologies, do have different characteristic sounds—so it’s not surprising some people might prefer one over the other. However, while I don’t buy the extreme view that ITB mixing sounds just plain bad, doing a good ITB mix involves some techniques that aren’t relevant with analog. Such as . . .
Realize that recording resolution and audio engine resolution are different. Recording resolutions higher than 24 bits are fictional, due to the limitations of A/D technology. But your sequencer’s audio engine needs far greater resolution.
This is because a 24-bit piece of audio might sound fine by itself. But when you apply a change to the signal (level, normalization, EQ, anything), multiplying or dividing that 24-bit data will likely produce a result that can’t be expressed with only 24 bits. Unless there’s enough resolution to handle these calculations, roundoffs occur—and they’re cumulative, which can possibly lead to an unpleasant sort of “fuzziness.” As a result, your audio engine’s resolution should always be considerably higher than that of your recording resolution.
Fig. 1: You can enable a 64-bit, double-precision audio engine in Cakewalk Sonar. If disabled, any 64-bit files are converted to 32 bits on playback.
Today’s sequencers use 32-bit floating point and higher resolutions (all the way up to 64-bit; see Fig. 1), but many earlier sequencers did not. If you’re mixing ITB with a sequencer that’s a few years old, upgrading may include an improved audio engine. Note that engine resolution is independent of your operating system; for example, you can use a 64-bit audio engine with a 32-bit operating system, or a 32-bit audio engine with a 64-bit operating system.
Because modern audio engines have so much headroom, it’s almost impossible to get distortion just by mixing channels together. Still, many engineers recommend keeping the master fader close to 0 and adjusting gain within individual channels to prevent overloads at the master out, rather than keeping the channel faders high and reducing the master gain to bring the levels down. Part of this is because the master output will eventually feed actual hardware, which is susceptible to overload and therefore, distortion; but some also feel that it’s possible to “stress” audio engines, which adversely affects the sound.
Analog consoles (and analog gear in general) rarely had response down to DC, due to the use of coupling capacitors to avoid transferring DC offsets from one stage to the next. But digital technology can create and reproduce subsonic signals, which has the potential to take up bandwidth and reduce headroom—I’ve measured some audio interfaces that go down to 5Hz, which is beyond the range that most speakers can reproduce anyway.
Fig. 2: Waves' LinEQ Lowband plug-in, used here with Acoustica's Mixcraft 6, is designed specifically to trim frequencies with a very sharp cutoff.
You can emulate the effect of coupling capacitors by inserting a steep low-cut filter in each channel (or at least at the overall output, but each channel is better). Set the filter frequency as high as possible, consistent with retaining a full bass sound (see Fig. 2). For example, a guitar note doesn’t go much below 90Hz, so you can set a sharp cutoff starting at 60Hz; this will tighten up the sound by getting rid of possible low-frequency sounds that have nothing to do with guitar.
Remove DC offset from your tracks before you start to mix; some DAWs have “remove DC offset” as part of their DSP menus (Fig. 3). As with subsonics, DC offset reduces headroom. For the full story on DC offset, see the article DC Offset: The Case of the Missing Headroom.
Fig. 3: Studio One Pro from PreSonus includes the Mixtool plug-in, which can be inserted in a track and set to block DC offset.
This applies to both recording or mixing. Digital metering does not necessarily show the true peak signal level, as it measures the samples themselves; interpolation may result in higher values than that of the samples themselves, leading to what’s called Inter-Sample Distortion. So, leave a few dB of breathing room for the cleanest sound. (This is less of an issue with higher sample rates, so you might consider “spending” the extra bandwidth to go for 96kHz—and you might also hear better sound quality with plug-ins, particularly distortion-oriented ones like amp simulators.)
Solid State Logic offers a free meter plug-in that shows Inter-Sample Distortion, and PreSonus’s Studio One Pro 2 DAW has meters that can be switched to indicate Inter-Sample Distortion. For more information about differences between analog and digital metering, check out the article Everything You Wanted to Know about Digital Metering.
Those EQ plug-ins that come with your host sequencer may be convenient, and these days, probably sound pretty good too. However, specialty plug-ins made for mastering may sound better—although they’ll take a bigger hit from your CPU—and give a smoother, more “analog” sound.
Although some people think dithering doesn’t matter (and frankly, for music with a limited dynamic range, it pretty much doesn’t), dithering can indeed help some digital mixes sound better (see Fig. 4).
However, there are two issues that will affect how you apply dithering. Your host may dither automatically, which you don’t want to do if your mix will be mastered later with a mastering program; and, you’ll often have a choice of dithering algorithms. Their sonic differences may not be obvious, but they can have an almost subconscious influence.
Fig. 4: Steinberg Cubase includes Apogee's UV22 High Resolution dithering plug-in; here it's inserted in the final master bus.
To evaluate the sound, take a high quality recording of material like a piano chord that decays to nothingness, copy the track, and apply different types of dithering. Cut off the note attack until it’s decayed to an extremely low level, where the dithering comes into play. Normalize each example, and you’ll hear the difference quite clearly.
Craig Anderton is Editor Emeritus of Harmony Central and Executive Editor of Electronic Musician magazine. He has played on, mixed, or produced over 20 major label releases (as well as mastered over a hundred tracks for various musicians), and written over a thousand articles for magazines like Guitar Player, Keyboard, Sound on Sound (UK), and Sound + Recording (Germany). He has also lectured on technology and the arts in 38 states, 10 countries, and three languages.