David Shapton on the complexities of audio production and how you should always consider the end user and their acoustic environment.
I'm not even slightly religious, but I really love cathedrals. They're made on a scale that makes it hard to imagine how they could ever have been conceived, never mind actually built. In a very small way, cathedrals make me feel the same way I do about the universe: how can you even start to understand its "bigness"?
That sheer enormity bestows some unique qualities on the spaces within those giant buildings. Something about such immense human-made objects makes you feel different when you're inside them. You definitely sense a human kinship with the tens of thousands of workers who built the giant structures all those years ago.
And there's something else that you sense when you're inside them. It's the sound.
Quite recently, I was inside Lincoln Cathedral (the tallest building in the world from 1311 to 1548 when its spire collapsed. These things are old! Ed.), and I had a revelation. You won't be surprised to know it wasn't a religious one. It was that the sound inside that space was different for everyone. No two people were hearing the same sound in the same place at the same time. Of course, people's experiences were related. It would be hard to disagree that a gorgeous-sounding choir was singing at one end of the aisle. But, each audible pressure perturbation was different for each ear and for everyone in the audience. That's always true, but the sheer consistency of the sound - the integrity of it - made me realise that the audible sensation was not caused by a monolithic block of moving air but an almost infinitely subtle succession of molecular-scale causal events.
Remarkably, while we can never capture the whole experience, we can get usefully close to it with modern recording techniques, and we will get better at it in the near future.
It's hard to overstate the sheer complexity of the audible experience in any space, not just a cathedral. Night-time in a rain forest, the dawn chorus in a spring meadow, the hustle of a road junction in Los Angeles, or the bustle of a spice market in Istanbul: these are as complex and as complete as their associated visual presentations. A Dummy Head binaural recording of any of these locations can sound utterly convincing and, I would argue, can be as evocative as a high-resolution video recording.
This is all to make a very simple point: audio matters. Getting it right can magnify a film's authenticity, making it feel convincing and complete. Getting it wrong can be a disaster. So, no pressure then.
My most basic advice is to think about what you're trying to achieve and check everything at every stage. It's all too easy to gloss over audio.
When you're starting out on any audio project, think ‘minimalism’. Often, the least you can get away with is the most convincing. Less is definitely more. Think of it more in terms of capturing essence rather than complexity. One example of this is pared-back acoustic recordings of pop classics: they often reveal more about a song than a full-on remix version.
Whatever you're trying to record, there are many things to consider, but two good ones to start with are signal-to-noise ratio and intelligibility.
Signal to noise ratio is the proportion of sound you want vs sound you don't want. You don't want hiss, air conditioning, traffic noise or the sound of someone brushing against a microphone cable. You can minimise environmental noise (caused by humans or nature) by getting as close to the microphone as possible without distortion. Processors called limiters can help (and some hand-held recorders have them built-in), but if you can manage without them, it will be a more natural recording. Some recorders now have "32-bit" audio. This seemingly pointless theoretical dynamic range could comfortably capture the Big Bang, but which, in practice, means you don't have to set the recording level.
Intelligibility is partly a function of signal-to-noise ratio, but many other things can affect listeners' ability to understand what is being said. It is often a victim of surround and immersive formats, typically when playback settings are wrong. If the dialogue is mixed for a centre channel, and there is no centre speaker, then you are likely to lose speech in the music and effects.
But it is also worth thinking about people whose hearing might not be perfect or who might be watching in noisy environments. That probably includes a lot of people over 55 and every ex-band member (I fall into both of those categories). The first time I saw Interstellar was on a 747 across the Atlantic. Lovely planes, though these are, they're not the quietest. You could charitably say it was like being surrounded by vacuum cleaners. I couldn't resolve a single word of Matthew McConaughey's gravelly dialogue.
Surprisingly, one of the most common reasons for getting recording wrong is loudspeakers, not microphones. If you mix or simply judge your recordings with loudspeakers that aren't representative of the devices your audience has, then you're likely to imprint your recording when you mix or master it with the opposite of the sound of your monitors.
That sounds counter-intuitive, but the reason is simple: if your loudspeakers sound very bright (i.e. they emphasise the treble), then your instinct is to EQ the sound to compensate by reducing the high frequencies. This will make it sound OK to you from your set-up, but if played back on speakers that don't boost the treble, it will sound dull. The same applies to bass. If your monitors make the bass sound too loud, you'll EQ it to a lower level, which means that on "normal" or "average" speakers, it will sound like there aren't enough lower frequencies.
Which brings me to the final topic. It doesn't matter if you've got the best loudspeakers in the world; if your room acoustics are bad, everything will sound wrong. Resonance can easily crop up at all frequencies (but especially the lower ones) and give you an inaccurate impression of the overall sound. Fixing this to perfection isn't easy, but what you can always do as a stopgap is fill the room with sound-absorbent materials - curtains, carpets, soft toys: anything that's likely to stop sound bouncing around. Beyond that, you can add acoustic panels and even call experts in to "tune" your space to as near neutrality as possible.
If none of that is possible, look for a pair of "neutral" sounding headphones like the Sony MDR-MV1s I reviewed recently (below). Some people frown at the idea of mixing on phones, but because it eliminates the room's sound, it is a perfect place to start.