In ripping some of my classical music, I’ve re-discovered a simple principle of sound recording: if you want the quiet bits to sound quiet, make ’em LOUDER.
This is exemplified in a comparison of my two “O Fortuna” recordings. The one I first purchased back in 1994 is a standard classical recording in which the engineers try to faithfully capture the exact sounds of the orchestra and chorus. It’s disappointing because after the powerful opening chorus the levels drop to where you have to turn the volume up to hear what comes next. Then you blow out your speakers when the punchy bits come back.
The second is one I bought on iTunes, and it’s the “OperaBabes” recording. It’s classical music with Pop/Rock production sense applied. The opening is like a thunderclap, but when the ladies and chorus back off to hushed tones, you as the listener get yanked all the way onto the stage so you can hear them. In short, while the overal levels, as measured in decibels, stay within a much smaller range than they do in the first recording (up towards the TOP of the 90dB range of a CD), it SOUNDS like there’s more dynamic range. This is because you can HEAR the whispers.
There are obviously multiple schools of thought on audio engineering. Some audiophiles will tell you that the first method (leave the levels alone, and use the ENTIRE 90dB range of the CD format if the music has that kind of range in it) is better, because it’s somehow “purer.” Me, I’m about accessibility. Audio recordings are an illusion to begin with. A good audio engineer is a master illusionist, and will convince me, when I close my eyes, that I’m right there.