Do broadcast loudness meters matter in mixing and mastering?

We’ve all heard about the new loudness standards by now – official, internationally agreed methods of measuring ‘loudness’ and a set of guidelines for broadcast – even legally binding requirements, in US TV.

But none of this matters for mixing or mastering, right? There are no rules or restrictions on how loud a CD can or should be mastered, and there never will be. The new ITU regulations only apply to broadcast, so we don’t need to worry about them… right?

Well – yes and no. The new standards and regulations are going to have an impact on the way the music we work on is heard in future, whether we like it or not – and if we don’t pay attention it could seem to suffer as a result. Here’s why.

To cut a long story short, the world is changing. Loudness standardisation is going to be everywhere before long. Situations where people hear music at the exact level we mix or master at will be the exception, not the rule.

In fact this is already the case in many situations. Loudness regulation has been in place on radio and TV for decades, it’s just that from now on it will be using LUFS instead of peak level to set the standards. Online platforms like iTunes Radio and Spotify already make sure we hear everything at similar loudness by default, and I’m sure the same will be true for YouTube and everywhere else before long. It makes sense – loudness has always been the number one reason for complaints about audio. That’s why the CALM Act was introduced in the US, to regulate the loudness of adverts, because people found them so annoying. And even when people do hear music at the exact level it was mixed or mastered at – played straight from a CD, for example – the first thing they do is adjust the playback volume to whatever they’re comfortable with. So the actual ‘loudness’ of a digital recording is becoming less and less relevant in the real world.

But even if we take it as read that the levels people listen to our music at are going to be messed with, one way or another, who cares? There’s nothing we can do about it, so we may as well just ignore it and worry about more important issues, right?

Wrong. Now, and in future, our music’s loudness is going to be measured and controlled by computer algorithms, and this affects the way it sounds in comparison to everything else. So we need to measure it and listen to the result.

And this is where things get interesting. For example, if two songs were played side-by-side direct from CD, one measuring an ‘integrated’ loudness of -8 LUFS, and the other a more conservative -12 LUFS, the first song would sound 4dB louder – although in reality the end listener would probably turn it down. But on radio, or TV, or iTunes Radio, Spotify, or any other loudness-matched replay scenario, both songs will be played back at the same perceived level. Neither will be significantly louder than the other. So how does that affect how they sound in other ways, compared to each other? The only way to know for sure is to try it ourselves and listen, and that’s why it’s essential to start getting familiar with the new LUFS loudness meters.

We can make some broad predictions though, just by thinking about the numbers. On TV, where the standard requires an overall playback loudness of -23 LUFS (to allow for very dynamic material like film soundtracks) our -12 LUFS song will be turned down by 11dB, whereas the -8 LUFS song will be reduced in level by a whopping 15dB (!) – meaning our more conservatively mastered -12 LUFS song has the advantage of peaking up to 4dB higher than the ‘louder’ example. It has 4dB more ‘crest factor’, or peak-to-loudness ratio (PLR).

That’s 4dB less limiting, or compression, or clipping, with all the sonic advantages that can bring – or even more, if we choose to mix and master at even lower levels. And, perhaps most importantly, the extra 4dB of extra crush in the ‘louder’ -8 LUFS master has no benefit, because the playback loudness is standardised. In fact, more dynamic, higher PLR material often sounds noticeably better when loudness is balanced. And all this applies in exactly the same way on Spotify or iTunes Radio, even though they use different algorithms and higher reference levels – in fact it applies in any situation where playback loudness is standardised.

Testing

If you’re interested in experimenting with loudness metering and comparisons for yourself, the key is to match the overall integrated loudness measurements of each song to hear how they will stand up against each other – I recommend adjusting them all to a nominal reference level like -16 or -23 LUFS. More and more DAWs have LUFS loudness metering built-in these days, but there are also plenty of great plug-ins available, too. TC Electronic has been instrumental in much of the research underlying the new units and standards – its LM2 and LM6 Radar meters were among the first and best. I also really like the Nugen Audio VisLM, and MeterPlugs’ LCast, both of which are firmly targeted at pro users. All of these plug-ins feature a loudness history graphic display, which is a feature I find invaluable.

Just to be clear, though, no one is telling us or our clients how loud we should mix or master the music we work on. We can still make things as dense and aggressive as we like. But we need to understand how those decisions affect the way it sounds when loudness is standardised. The truth is that in future there will be no advantage to mixing or mastering at really high levels, and therefore no need to be ‘competitive’ – and if we go too far, the music could suffer in comparison to more dynamic material. And actually that’s great news, because it means we can stop worrying about the so-called war, and concentrate on finding the perfect balance between loudness and dynamics instead – doing what’s right for the music. Which at the end of the day, is what we’re all in this crazy business for in the first place, isn’t it?

Ian Shepherd is a mastering engineer at Mastering Media and runs the Production Advice website (www.productionadvice.co.uk). He is the founder of Dynamic Range Day, and has developed the Perception plug-in with MeterPlugs, to help people find the loudness sweet spot for their music.