Let’s pretend classical music is your jam! And your favorite concerts in town. Even better their here for the whole week and you’re hitting all the shows! 

The first night the orchestra is full. Not a missing member the sound pitch and volume they are able to hit impresses even the most cynical classical music snob, and blows you out of your chair.

The next night 25 of the 200 band members are unable to perform in the concert so you go to the same show the next night, and while all of the instruments in the orchestra are present there are just less of them. The concert still sounds impressive, but it’s just missing the something the first night had. 

The next night nearly half the band misses the show, again all instruments present so the concert song still sound the same, but you can really start to tell what’s missing now. 

This example could go on, but the  point we are trying to illustrate with our fake orchestra is that with a greater dynamic range [the more performers / instruments the conductor has to work with] is the deeper your music can sound. Of course this is assuming that the dynamic range [bit depth] is utilized properly.

When a song is lossy compressed  bits of information that are deemed unnecessary for the audio tracks reproduction are deleted**. The question that you have to ask when compressing an audio file is how much information can you remove and still have it sound as good as the original song? In this example each band member would represent a piece of the bit rate, and the compression would be the removal of each member of the orchestra systematically. 

**See chapter 2 for more on compression and how it works.