Fatness and RMS

Why does compression give fatness? I believe there are two reasons for this, one is quite technical and relates to the subject of RMS levels, the other is down to the way the compressor moves in response to the music.

Reason 1. RMS

We know from the previous chapter about EQ that the RMS level is closely proportional to our perception of loudness. Having set up your compressor using the technique discussed, you will increase the RMS level of the track without increasing the peaks. This increased perception of loudness sounds like we've fattened up the track. So what's happening here, why do RMS levels relate to loudness and why does compression increase the RMS level?

Imagine a single sound wave on a graph, any shape or form. The peak level is the very tip of the sound wave, the highest point that the wave has reached. One might think it would make sense that the higher this point is, the louder we shall perceive the sound - not necessarily.

RMS stands for 'root mean square' and is a way to measure the magnitude of varying quantities. It is the square root of the mean of the squares of the values. In simpler terms, our sensitivity to loudness is proportional to the average value of the various peak levels present in a given time. The RMS is a value given to the energy, or loudness of audio over time.

Imagine how a waveform looks inside your DAW. You can amount the RMS value to the actual amount of visual surface area that the waveform covers. Two audio files could peak at the same level but have massively different RMS values - this would appear obvious when looking at the two waveforms in an audio editor.

In the picture above: Two audio files of the same music track, the top one has not experienced mastering, the bottom one has. Below the waveforms are two level metres, one for each track. The dark blue indicates peak levels, the light blue indicates RMS. When comparing one against the other, the difference between the peak levels is much less than the difference between the RMS. The second track's RMS level has been significantly increased following the mastering process.

When compressing an audio track, we only reduce the level above a given threshold. It's as if we're squashing down just the tops of the sound waves leaving everything beneath untouched. Then, with this increased available headroom above the now lowered peaks, we can raise the overall volume of the track (makeup gain), so the peaks reach back up to their original level. Only now many more peaks will be reaching this level than before. Can you imagine how this increases the visual surface area when viewing the compressed waves in an audio editor? The result is an increase in RMS level with no increase in peak levels. It gives you a sense of the track being inflated almost, rather than just being louder. It has got fatter!

It's important to understand that this process is happening over the changing levels of different passages in the music, and not in relation to individual sound waves. The actual peak of each individual sound wave cannot be reduced without reducing the rest of the wave below it. Unless you had a super fast compressor that can attack and release as quick as a single sound wave appears and disappears. There is actually a type of processor that does this naturally - there's a dedicated chapter about in the advanced section.

Our ears' sensitivity to RMS rolls off as we get closer to the bottom of our hearing range, so to hear the bass regions at the same volume as the rest, more energy is required. Low frequency sound waves contain more energy than the rest. For this reason, low frequency audio will force the RMS level metres to jump higher in response to what you are hearing when compared to the rest of the spectrum. You need to consider this in mastering. The bass end of the music will always require more of the limited amount of available energy.

The idea of there being a limited amount of energy will be discussed shortly.

Reason 2. Compressors movement

When referring to fatness in this way, I am referring to the fattening of a certain element in a mix, which in turn will give an overall perception of fatness. In particular the kick drum, which is most effective in dance and electronic styles, although this can be perceived in any genre.

When a kick drum is very prominent in a mix, it can really fatten up the track if the style is something like dub step, dnb, or some other dance style like various forms of house.

You should now have a good understanding of how the compressor responds to one sound being louder than any other – it basically dips in volume. So what will happen every time the kick passes through the compressor? We get a dip in the overall level, and with the correct compressor settings, can be achieved in a flattering way, mostly for the instance of each kick drum. The clever thing is, we don't perceive this as a reduction of the kick itself, just the music around it. The kick, although being the main thing which is triggering the compressor, appears to be untouched by the reduction in level. The effect is as if the music is literally ducking out of the way for every kick. Think back to earlier in this chapter where we said that sometimes the pumping effect of a compressor can be desirable, well this is it. When set correctly, this can really fatten up your music. Remember how we said having an extended bottom end can actually help a track translate to a little kitchen radio, or some inferior laptop speakers? Well this is what I meant. By use of compression, the kick drum projects its imprint across the whole mix - right across the whole frequency spectrum. The way the compressor dips everything for the kick gives us a feel for the actual weight and solidity of the kick – it makes it sound fat.

The reason why this effect is most obvious on the kick is down to the RMS energy. The kick contains the most as it will likely extend down lower than anything else in the mix, and the lower the frequency the more energy is required for us to hear it. So the compressor will respond more to the kick than any other part of the spectrum. Try to imagine how the kick drum is made up of frequencies that extend right up into the high mids. Because the music can literally be forced aside for every kick, you can actually get a feel for the weight of the kick without the need to hear its bottom end. So it's possible to have a fat low end be perceived on a speaker with no low end - magic.

This effect happens best when the bottom end of the kick extends down nice and low, lower than the bass line or anything else in the mix. That way the kick doesn't necessarily have to be mixed high in the mix to achieve the effect. There will be enough RMS energy in the bottom end to get those compressor needles moving.

This audio effect is true to a whole variety of music. It doesn't just have to be dance music with a pumping kick. The effect can be slight yet still effective in all the same ways. And it's not just the kick drum, bass-lines can project their imprint across the whole spectrum too, helping with translation to inferior speakers.

By now you should be starting to realise the enormous power of your compressor when shaping a track to sound fantastic in whatever environment it is finally played in.

David Eley - TGM Audio

Next page... Limiters