Actually, mastering decreases dynamic range, largely to increase apparent volume (apparent volume is RMS, not peak-to-peak. To be very vague, the RMS of 0 to -1

Actually, mastering decreases dynamic range, largely to increase apparent volume (apparent volume is RMS, not peak-to-peak. To be very vague, the RMS of 0 to -10 is -5, and the RMS of 0 to -20 is -10. The more dynamic range, the less apparent volume). The intent of the mixing engineer and mastering engineer is to make a recording that sounds good everywhere, not just on studio monitors. When I mix records (I've produced several), I check every "final" mix on cheap earbuds and in my bad car stereo. Sounding good on my very expensive Tannoy studio monitors is meaningless if it doesn't sound good in someone else's car. For that matter, most pop records are mixed on subjectively awful speakers. The most popular pro mixing speaker is the Yamaha NS-10. They sound shrill, piercing, and nasty, with a sharp eq spike right in the 2khz range where ears are most sensitive. If you can make a mix sound good on those, it should sound good anywhere! A lot of mixers also rely on Auratones, which are tiny 5" drivers, often used in mono. No lows, no highs. They're incredibly honest about levels.