Human ears are most sensitive to frequencies around 2,000 to 5,000 cycles per second. That’s the most useful range for hearing human speech.
But music can encompass sounds down to about 20 cycles and up to 20,000. We just can’t hear the highs and lows as well as the middle. Strangely, the louder the music, the better we hear the lowest and highest sounds. In fact, we get the fullest spectrum of sound as close as possible to the threshold of pain.
This is most noticeable with low notes. It's really hard to hear the bass at low volume, it sounds a lot more impressive when you turn the music up. This effect is known as the Fletcher-Munson curve, or the equal-loudness contour.
That's a problem for music makers. Recordings sound best at the volume at which they were mixed. Play a CD too soft and you’ll lose the lows and highs, too loud and it’ll sound hyped and unnatural.
It’s also a problem for listeners. We want to hear every detail, but we don't want to have to turn our stereos up too high. Some stereos let you account for the Fletcher-Munson curve with a “loudness” knob. Cheaper systems boost the extreme frequencies whether you like it or not-- one of the reasons a bad stereo can become fatiguing.
I don’t think there’s any way to solve the loudness problem. Listening to live acoustic music is the only way to make sure no one is fiddling around with your loudness contour.