Don't miss the advanced techniques at the end of the mastering tutorial :)

Limiters and RMS

 

The RMS level of a piece of music can only really go so far before you loose the punch and begin to create an unpleasant squashed effect. Always give it space to breathe. The point at which this occurs is different from one piece of music to another and is quite dependent on the quality of the mix - the better the mix the more RMS can be obtained before sounding unpleasant (refer to the advanced section for further discussion on how the quality of a mix dictates how much RMS can be obtained, and discussion about what 'the quality of the mix' actually means).

After the compression stages in the master chain, the likelihood is you will have increased the RMS by a certain amount. Whatever room is left for any more RMS, if any, is achieved by the limiter. It's as if the limiter fills in the gaps at the end of the processing with RMS energy, and then seals it off. It's usually the case that the less the limiter is required to do to achieve the desired loudness, the more punchy is the result. With this idea that the limiter fills in any space left with energy, imagine now that we have just applied a low cut to our track and removed some very low end frequencies such as 30 to 40Hz. This removal of perhaps unnecessary RMS energy, will have effectively created some space for the limiter to fill in with RMS energy, and it does this by concentrating more across the remaining frequencies. Now let's imagine we have just rolled off the top a little, perhaps above 17kHz. This will effectively create more space for the limiter to replenish with RMS energy, and again will apply the energy to the remaining frequencies. Bear in mind that the top end doesn't carry anything like as much RMS energy as the lows.

Can you see how more apparent loudness can be obtained by concentrating the RMS energy into just the most relevant area of the spectrum?

We know from the chapter on EQ that the midrange carries the greatest responsibility, as in a lot of cases it is mainly midrange that will be heard. Well this is how we can ensure there is plenty of energy in the mids. If the track has a problematic huge amount of low end sub, so low that it would not even be heard on average speakers, then the limiter will not be able to give as much power to the rest of the spectrum. That is unless we correct/balance the low end by applying a shelf or a low cut (or multi-band compression - discussed soon). But don't forget that a healthy amount of low can be put to good use with your compressor by fattening and helping with translation. Just be aware that the lows require the most RMS to be heard and so they take most of it up. The more low end, the less RMS can be applied elsewhere by the limiter and therefore lower perceived loudness on certain music playing systems.

In most cases, a low cut up to around 30 to 40Hz will still allow you to drive the single band compressor with the kick and bass, but will allow you to push that tiny bit harder with the limiter before it sounds bad.

Limiters are very easy to dial in, but with ease brings carelessness. This leads on to a very important point...

Instant loudness, instant improvement? The brain's response to an increase in the loudness of music is usually positive – you turn up the radio when your favourite song comes on, not down. It's only really the transition from quiet to loud that gives us the feeling of the music sounding better. Over time loudness can cause fatigue and be rather unpleasant, even unbearable. When you start to apply the limiter, you may at first feel you have improved matters by increasing the apparent loudness. Be sure that you are really reacting to the limiter's positive effects and not just responding to everything sounding louder than it did before.

A good way to ensure you're feelings are true is to compensate for the increased perception of loudness created by the limiter by reducing the output level to your monitors by roughly the same perceived amount. When I start mastering, I set the level of my monitors to a comfortable amount using the output control on my interface. As I proceed through the mastering processes, I constantly alter the output level to my monitors so the perceived level is always roughly the same as when I started out.

The fact is, pushing the limiter too hard is bad. The fast attack is extremely damaging to the transients, and it's the transients that contain the punch. There is a well known term that resonates throughout the mastering world... 'the loudness war'.

Do not fall victim to the belief that if one song appears louder than another, for some reason it's better. If you turn up the quieter of the two, you will probably discover that the quieter one packs more punch as it hasn't been squashed. Has your hi-fi got a volume control? Has your friend's hi-fi? You can always turn it up. We want a suitable amount of loudness to give a sense of body and weight. We also need to ensure all the details are audible amongst the distraction and background noise of the consumer's environment. Anything after that and you should be questioning the need for any more RMS.

All that being said, there are ways to achieve high RMS levels without destroying the mix. The advanced section reveals how great loudness can be obtained whilst still sounding spacious and dynamic. Here's how a master chain may look when seeking high RMS levels...

Also, in the advanced section is a dedicated chapter about the master chain and how to arrange your processors to achieve certain results.

David Eley - TGM Audio

Next page... After the Limiter?