Thinking Outside the Box, Part 1: Start on the Right Terms
In this first instalment we bust a few analogue vs digital myths on our way to baking a digital cake that tastes pure analogue.
Tutorial: Dax Liniere
For a long time, there was a clear and undeniable winner of the analogue vs digital, OTB (outside-the-box) vs ITB (in-the-box) debate. Digital technology just could not match the euphonics (those tonal harmonics we find so pleasing to the ear) we’d grown used to in the analogue domain.
But as the understanding of digital signal processing has advanced and processing power became faster and more affordable, ITB has found firm and equal footing with what was once only achievable in the analogue domain.
A FIRE WITHIN
This series of articles is not intended to throw fuel on the fire, but help you blaze your own trail. While the rest are busy arguing, we can work to better ourselves at our craft and get out-of-the-box results from within it.
To really understand how to get a rich, full sound in-the-box, we need to know what’s really happening to our signals as they pass through analogue consoles and outboard gear.
You often hear people say that songs mixed on an analogue console have more ‘depth’, ‘width’ and ‘punch’ than ITB mixes. As a science-minded person who understands electronics, I’ve always found some of these descriptions to be a little dubious.
DEPTH — DOWN TO THE FLOOR
Analogue consoles, as much as we love them, are imperfect. They have background noise, present as hiss, and although a professional console will have an acceptably low noise-floor, it still has more noise than well-designed digital audio workstation (DAW) software.
The term ‘depth’ describes how far into the soundfield you can hear, or the contrast between the closest and farthest sounds. The impression of distance or depth is caused by the psycho-acoustic properties of volume and delay, hence reverb. As the tail of a reverb decays, it becomes exponentially quieter until, at some point, it drops below the noise floor, becoming masked.
It’s true that the human brain is capable of discerning sounds below a constant, steady-state noise floor, but it also stands to reason that we could hear ‘deeper’ into the soundfield with less noise present.
WIDTH — A LEAKY DEFINITION
‘Width’ is even easier to define and can be used to explain away another one of the myths surrounding analogue consoles.
Analogue equipment has an affliction called crosstalk, which is where signal from one signal path ‘leaks’ into adjacent signal paths. This can occur when one or more tracks on a printed circuit board are in close proximity. In a DAW or digital mixer, crosstalk cannot naturally occur since each ‘signal path’ is a separate stream of data.
In a stereo mixing scenario, a mono sound that’s panned centre is obviously not coming out of the centre, since there is no centre speaker. It’s created in what we call the phantom centre. This is simply the psycho-acoustic phenomenon where an identical sound of the same volume and phase arrives at both ears at the same time. This gives the impression that it originates from directly in front of us.
When you pan a sound hard left, none of that signal is routed to the right channel and vice-versa. If you adjust a pan control away from hard left, you are directing some of that sound to the right channel. The closer to centre pan, the less volume difference there is between left and right channels, until you reach the centre position where both channels receive the same signal and it appears to originate from between the stereo speakers.
If you have a pair of non-identical sounds, panning them less than hard left and right will result in the soundfield becoming narrower. Crosstalk obviously affects width, therefore claiming a mix done on an analogue console intrinsically has more ‘width’ than an ITB mix is complete nonsense.
Although a professional console will have an acceptably low noise-floor, it still has more noise than well-designed DAW software
PUNCH — CAPACITY FOR ATTACK
While the terms ‘depth’ and ‘width’ are either being used incorrectly, or relate more closely to the skill of the person who can afford to use a large-format console, we still have to address the term ‘punch’. A ‘punchy’ sound can be described as having a strong attack (without being too sharp or biting) and where the attack of the signal is quite even, despite fluctuations in input signal.
There are many parts of an analogue console where distortions and colourations are induced on the signals passing through them. The most significant are transformers and other reactive components such as capacitors.
A transformer uses two coils of wire — a ‘primary’ and a ‘secondary’ — wound in close proximity around a metal core. There is no electrical contact between the individual coils or the metal core — the signal is induced into the secondary coil by the primary coil via electromagnetism. When the coils are driven too hard, i.e. by a large signal, the transformer will saturate or ‘soft-clip’ and introduce harmonic distortion to the signal. In the right quantities, this can be quite pleasing and musical, but as with anything, it can be easily overdone.
Being a magnetic device, the sound of the saturation is somewhat like the sound of analogue tape when it is pushed too hard. Different types of transformers have different saturation characteristics, mostly depending on the type of metal used in the core (to my ear, the cleanest is nickel, followed by iron, with steel core being the most coloured).
By their nature, low frequencies have more energy than high frequencies. It’s the kind of energy found in a kick drum signal that can saturate a transformer, causing excitation of low frequencies which produces easily audible upper harmonics.
Capacitors are often used in consoles to ‘decouple’ one circuit from another (eg. the mic preamp from the EQ). In many circuit designs the audio output signal contains a portion of the DC power supply voltage and it is not desirable to pass that on to the next circuit. Capacitors are used to make that bridge, passing the audio signal voltage but blocking the power supply voltage. In other words, decoupling the DC component of the signal or removing DC bias.
Capacitors are reactive components, meaning their behaviour changes with frequency. They are capable of altering a signal’s phase and frequency response, and can therefore impart their own subtle tonal colouration on the signal. Some analogue audio circuits use many decoupling capacitors in each channel strip, collectively contributing significantly to the overall sound of the console. (Transformers are also reactive devices and can alter a signal’s phase and frequency response, in addition to introducing saturation.)
The important thing to remember is that every time your signal passes through another circuit stage, you introduce another layer of distortion and/or colouration.
It’s this short-duration saturation of transients that increases the apparent ‘punch’ of a sound. Since this soft clipping is due to the input level reaching a finite ceiling, the resulting harmonics are produced at a consistent level.
BANDWIDTH: FLAT TO THE LIMIT
Another difference between OTB and ITB is bandwidth. While console manufacturers and DAW developers all strive to give their products the widest and flattest frequency response, in reality, the limitations of physics come into play in both worlds. Any digitally sampled signal has an upper limit on the frequencies that can be captured and reproduced. Harry Nyquist, in his sampling theorem, states this is half the sample rate, which for people working at 44.1k, gives a limit of just over 22kHz. Analogue consoles, on the other hand, extend to at least 30kHz. Rupert Neve believes consoles should be ‘clean’ up to 100kHz. Whether those ‘inaudible’ high frequencies are significant is an entirely different argument, but the only way to match this ITB is to run your system at greater than a 192k sampling rate.
On the other end of the spectrum, a properly-designed DAW has a frequency range that extends flat down to 0Hz. Compared to even the highest spec’d analogue consoles which roll off around 15-20Hz, there’s a fair bit more going on down there, not to mention the phase shift caused by that roll-off.
Though most DAW developers have accomplished what console designers set out to achieve, is that actually useful? Unless you’re using your DAW for scientific purposes, the answer is a resounding no.
AT THE CROSSROADS
Of course, sonic differences aren’t the only thing setting OTB and ITB apart: ergonomics and workflow play a large part in the experience of mixing a song. A hardware console is undoubtedly a more physical, more tactile way to mix. Even a control surface can’t offer the same level of physicality due to the reduced number of controls. I quite enjoy balancing a mix spread across console faders; it’s a great way to experiment with different combinations far quicker than is possible with a mouse alone.
But I’ve observed something interesting. For most ‘A-list’ engineers if the balance of a mix doesn’t feel right… swish, all the faders go back down and they’ll try a different tack.
But when mixing ITB, people seem less likely to start over from scratch, which is odd, because it’s even easier to achieve in a DAW. Just save a copy and try something completely different, if you’re not happy with the road you ventured down, a few clicks and you’re back at the crossroads.
Another difference from ‘the old ways’ is that now we have a computer screen to steal our attention. Whereas before, there were no visual cues to pre-empt, and less distractions from our primary task of listening. Thankfully, there are solutions. Basically, what we want to do is blank our computer monitors quickly and easily when we’re doing our critical listening [which some would argue is all the time — Ed]. On a PC, this is easily achieved by setting your screensaver to a blank screen. Then use a free program like ShutdownTray to trigger the screensaver with a quick keyboard shortcut. Mac users have it even easier with Hot Corners. Just pick your corner and set it to ‘Put Display To Sleep’.
THE CHOICE IS YOURS
Working in the analogue domain imposes other limitations, but often, those limitations can be used to our advantage. OTB, the number of mixable tracks is limited by the available console inputs and D/A converter channels, and the amount of outboard processors is limited by budget and available space. ITB, our track and processor count is only limited by the computer’s available system resources. Buy a plug-in and you can use as many instances as your system will bear.
When you’re working on a production ITB, don’t leave all of your decisions until the mix phase. While it seems logical that no-one would want to paint themselves into a corner, leaving options open can be an enabler of indecisiveness. Limitations can challenge you and make you grow. They can force you to think in a different way in order to solve a problem to reach your end goal. Quite simply, practice the act of committing to decisions early in the production process. Sure you might make some mistakes, but without mistakes, how can we grow?
As mastering engineer Sean Diggins says, “We are in a time-based industry and analogue is a time-soak.”
OTB mixdowns are restricted to realtime; stems are more difficult to produce; and recalls take time and leave room for error. The sound of hardware can change due to heat, power and atmospheric variations, plus deterioration through ageing. While reliability issues can be mostly mitigated through meticulous maintenance, this requires money, time and expertise. Every mixdown, revision, stem and repair takes time. Every extra ‘instance’ of outboard gear costs money, compared to a plug-in which is bought once and inserted many times. To be fair, plug-ins can also break compatibility with system upgrades or lose support completely (as in the case of the fantastic Voxengo Marquis compressor). You also never have to worry about copy protection with outboard gear.
Are the perks of analogue mixing worth the drawbacks, or vice versa? Ultimately, that’s for you to decide. To me, there is a clear-cut winner, and that is ITB. It has everything going for it — instant recall, full automation, ‘unlimited’ processors (restricted only by your CPU power), no noise and minimal maintenance, editing capabilities that open up new sonic and production possibilities, faster-than-realtime bounces and a great sound that can be relied upon to give the same results every time. But remember, to get the most in the box, we must think outside it.