Squished Down and Turned Up: How a Humble Process Has Changed How We Listen to Music

Compression is the process of reducing the dynamic difference between the loudest and the quietest parts of an audio sample. Photo by Mattieu A

By Nicole Bouwkamp

Have you ever listened to music for a while before suddenly feeling exhausted? Or having to turn your favorite song off because you just needed some silence? Have you driven a long way listening to the radio only to have your ears become sore and sounds muted? Ear fatigue, often felt as tiredness and a soreness, loss of sensitivity, discomfort of the ears, is caused by prolonged exposure to sound.

Thanks to the trends of music and listening environments today, ear fatigue can be experienced anywhere at any time. You just need to turn on the radio and listen for a while before you feel it or listen to music on headphones from a streaming service while in a crowd. Today’s music is part of the equation of experiencing ear fatigue. More specifically, a tool used to create music and broadcast it online and on the radio: compression.

Compression is the process of reducing the dynamic difference between the loudest and the quietest parts of an audio sample – the loud material gets quieter, and the quiet material gets louder. This is why a song may be described as punchy or having presence. The frequencies of the recorded sound are at naturally varying levels, and when compression brings up the softer frequencies and brings down the louder frequencies, the result is a more present and punchier sound (or as I like to say, a beefy sound).

Compression is often used when recording drums. Drums are the biggest producers of transient sounds, meaning that they are a loud sound with lots of attack, but decay in sound very quickly to where there is very little sound beyond that first attack. With compression, the attack is brought down in volume while the sound left after the attack is brought up. Frequencies that are naturally dynamically different are brought closer together, and you get a beefier recorded drum hit.

So, you hear everything better. That would be good with music, right?

The thing is, everything in music isn’t meant to be heard evenly all the time. One of the glories of music is the dynamic range and nuances within it, the little hidden gems of musical ideas that you discover after listening to a song multiple times, or the rise and fall of moments that can evoke emotions of triumph or despair. If there is a part of the music that grows from soft and intricate to loud and powerful, you need to actually (not) fully hear everything in relation to each other.

With compression, everything is louder, and we tend to lose the dynamic range of the music. The small nuances become more prominent and muddy the main melodies and harmonies, the rise and fall of dynamics becomes flatter, and “imperfect” playing is homogenized. This trend has been growing for nearly 30 years now, and no music is safe.

This isn’t to say that compression is bad by any means, it can actually be vital in the recording process to achieve a cleaner signal from a particularly temperamental drum, or to even out the sound from a singer who is not familiar with distancing the mic properly when they sing. Compression when cleaning the recorded sounds in the mixing process can be useful for achieving a better balanced song in the end, but I prefer to control the volume manually.

I will work harder to control the overall dynamics if it means I can keep the more natural dynamic sound of the instruments throughout. However, my ideas on how music should sound are my opinion, I will admit, and the opinion contrary to mine follows the idea of slapping compression on all the instruments for the entire song to get a more even dynamic range. This method has been steadily ruining how we listen to music for decades.

CD Loudness Wars – Yes and The Mars Volta

This growing trend is often referred to as the loudness wars. To make a song stand out from its competition, many artists, producers and labels all worked to make a loud and punchy track that would dominate sonically wherever it was heard.

To be honest, the loudness wars were an issue even in the days of the 45s. The push to get music noticed by sheer volume is an understandable goal, but ultimately a disappointing aspect of music to be focused on.

I looked at two songs that are similar in genre but are a few decades apart – Yes’ “Siberian Khatru” from their 1972 album Close to the Edge, and The Mars Volta’s “Empty Vessels Make The Loudest Sound” from their 2012 album Noctourniquet. I ended up using the Steven Wilson remix for “Siberian Khatru,” to best highlight the differences, just a note to everyone keeping track at home.

The top track is Yes, and the bottom track is The Mars Volta. Notice the differences?

What you are seeing is the visual representation of music in terms of amplitude, better known as value. As time passes on the horizonal axis, sound is visualized by the way of volume. The fine lines spiking up and down are representative of transient sounds with a lot of attack. The light blue form within the dark blue form is representative of “root mean square” levels, commonly used to measure the electric current, or voltage, that is used to produce the sound.

With Yes, you see and hear a lot of dynamic contrast between the quiet moments and the loud moments. You can even see the louder transient hits in the music that literally pop out from the rest of the music. With The Mars Volta, its looks a lot more like one wall of sound. You can even see where there might have been transient peaks in the music, but compression was applied to squish the sound together dynamically and then turned up to have one loud song. There is a lot going on when listening to it as well.

Personally, it can be hard for me to follow or really enjoy. That doesn’t mean I don’t like what I hear in terms of musical content, but it all sounds so even and loud that I find it exhausting to listen to. If everything sounds the same, how are we supposed to follow the important stuff?

Here is a nifty little tool I found online that measures the dynamic range of a song. I put in the same files I put into Audacity to see the waveforms, and the dynamic difference is really glaring.

The dynamic range comparison between Yes (left) and The Mars Volta (right)

There is a 10 decibel difference between Yes and The Mars Volta. 10 decibels translate to Yes having 10 times the range in dynamics compared to The Mars Volta. There is a lot more space for the music to live in with 10 times the dynamic range.

Queensrÿche – ‘Operation: Mindcrime’ Original Release Versus Remaster

Remember when I said that no music is safe from compression, and this epidemic trend of listening to music? Well, nearly every song that has been remastered in the (conservatively) 20 years has been treated with the squished down and turned up compression treatment.

I decided to test this out by looking at the same song from progressive metal band Queensrÿche, one being the original release and the other being the 2003 EMI remaster, which according to a forum I found was the biggest tragedy for one poster. Let’s take a look at the compression treatment on Queensrÿche’s 1988 track “Operation: Mindcrime

The top track is the original release of “Operation: Mindcrime” while the bottom track is the 2003 EMI Remaster.

Yikes, there is a lot more blue in the bottom track compared to the top. Care to guess which one that is? That’s right, the bottom track is the 2003 EMI remaster, and it hurts to listen to. Any dynamic depth is gone compared to the original, and while you may be able to hear the quiet material in the beginning better in the remaster, it is not worth the sacrifice of having the entire track distort to a single flat sound.

Just turn the volume up for the beginning on your stereo or phone if you want to really hear the beginning, don’t subject yourself to lower quality music to be lazy. Not only is everything compressed to have a flatter and more squished sound, it is then turned way up to where the signal starts to distort, creating a loud and unpleasant song to listen to.

Here is the dynamic range meters for the two tracks as well.

The dynamic range comparison between the original release (left) and the 2003 EMI Remaster (right).

The remaster has half the dynamic range in decibels, which roughly translates to half the perceived dynamic range.

This compression and dynamic range reduction can be tricky to hear definitively sometimes, especially without a comparison at the ready. Without knowing the song really well, its easy to get caught up in the content without much critical analysis into the quality.

The best way to feel how much compression is squishing the music is to see how you feel after listening to the album. If you could listen to it again pretty soon, it probably has lower compression and a higher dynamic range. If your ears are exhausted, and even if you liked the album you would have to wait a while to listen to it again, it probably has lots of compression and dynamic range reduction. Trust your ears.

Compression can be added to any stage of the album process, from recording, mixing or mastering. These days, compression is often added at every stage due to the lack of communication and understanding of what the track sounded like at the beginning and the intentions for the track.

The trend to squeeze the maximum volume from a song is prominent, which leads to compression at every stage as well. However, once compression has been applied and saved, that dynamic range is lost forever. Let’s hope the original tapes and hard drives of all this music are hidden somewhere.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s