Brook Preloader

Advertisements are too loud!

We all know what it’s like to have an advertisement come on tv and it seems way way louder than the show we were watching (at least in the days before Netflix when free-to-air tv wasn’t slowly dying). Surely there be some rule about how loud a commercial can be?! There is indeed, but advertising companies found a sneaky way around it by making the audio SEEM louder than they are. Let’s have a look at how they achieved this, and what has been put in place in an attempt to preventing this practice from continuing.

Compression is evil! Or is it?

I was at a family gathering and in conversation about my work in audio a relative asked me “you don’t use compression, do you!?” I answered that of course, I do. It’s a common audio processing tool that I use on a daily basis for a variety of reasons. I was a little dumbfounded that my use of compression was met with a sense of disdain, so I asked what compression meant to them? The answer I received was filled with some interesting misconceptions that I would like to clarify in this post. According to my relative:

Compression is something you drop onto to music to make it really loud.
Compression is bad because it hurts your ears.
It’s why tv commercials are so loud.
These comments aren’t entirely false, but they aren’t entirely true either. Let’s have a crash course in how compression works.

Compression Crash Course.

Compression is about squashing audio. Below is a waveform of music that is uncompressed. Notice that there are really big peaks. These are the loud parts. In between those are quieter parts. This is the dynamic range; The range between the quietest and loudest parts.
By applying compression, you are telling the compressor:

“Hey compressor, when the audio gets too loud could you please turn just those bit down?”

And the compressor says:

“No problem. How loud is too loud and how much do you want me to turn it down?”

It’s a bit more complex than that, but you get the point. So now the waveform looks like this:

Hang on a minute. If compression is used to make things seem louder, then why are using it to make parts of the audio quieter? Good question attentive reader! We have indeed made the audio quieter. So, let’s turn the whole thing up. Now the waveform looks like this:
By squashing the loud parts down, we have made the dynamic range smaller. So, the parts that were quiet before, are louder in relation to the loud parts. On average, the whole audio file is louder.

If you played the uncompressed audio next to the compressed audio, the compressed would sound louder, even though their loudest peaks are the same volume. This practice began small but escalated in what is referred to as “the loudness wars.”

The loudness wars.

Let’s go all the way back to the 1940’s when the loudness wars supposedly began…

Band A releases their song and has it played on the radio. Band B does the same, but their song is slightly louder. After hearing the songs played back to back, audiences feel that the Band B’s song has more energy, clarity, and captures their attention. So, of course, when Band A records their next song they want it to be louder than Band B. You can see where this is going. A back and forth between each band getting louder and louder. But there comes a time when the volume knob doesn’t go any higher. That’s where compression comes in.

Compression wasn’t designed as a tool to make audio seem louder. Not at all. You’ve probably heard of the term “mastering” before. It was originally the process of preparing audio to be cut into vinyl records. Compression was used to control the depth and breadth of the grooves in the record.

With the invention of magnetic tape in the 1940’s, it was possible to have much greater control over the mixing and mastering of audio. This lead to greater advances in audio recording quality.

In the 1990’s the loudness war was starting to get ridiculous. Songs were being released with the intention of sounding as loud possible with the cost being dynamic range. The waveforms of these recording looked like bricks!

Almost no dynamic range at all. The limit of loudness had finally been reached. Recordings were being criticised for being “over-compressed.” Listeners were realising that recordings with appropriate dynamic range sounded better. Dynamic range was finally valued again!

In advertising though, dynamic range definitely wasn’t valued above loudness. A new kind of measuring standard was needed. Measuring the peaks wasn’t enough, we needed to measure the average or how we perceived the loudness.

Perceived Loudness guidelines.

Yay! A way to measure perceived loudness called “LKFS.” It stands for “Loudness (L), K-Weighted (K), relative to Full Scale (FS).” I’m not going to go into the specifics of how it works but you can find our more here , but let’s discuss the name. Because a few different terms get thrown around for the same thing.

LKFS was the original name and it is still used but has been criticised for not following scientific naming conventions. It was suggested that LUFS (loudness units relative to full scale) should be used instead. Plus, that’s a lot easier to say out loud. You may also see LUFS indicated by the symbol LK. All of these mean the same thing though; a way of measuring how loud audio seems.

Different platforms and countries have different standards when it comes to LUFS, and these standards and constantly being refined. It’s important to check what the standards are before releasing your content. You don’t want to exceed the guidelines, and you also don’t want your audio to be dramatically quieter than your competition (they’re still going to be as loud as they can within the constraints).