I believe we have forgotten the fundamentals of music.
Sound is three-dimensional, and embodied. We have not one but two ears for a reason. Sure, stereo headphones will produce stereo but not three-dimensional sound. The stereophonic sound remains the same even if you move your head to face any direction. Not so with live performances or speakers spaced apart.
While it can be argued that an all-in-one unit player/speaker is located in a space it does not produce stereophonic sound very well. Firstly, the speakers are too close together. Whatever stereo sound you have it may as well be a single speaker. Secondly, today’s digital streaming music is less detailed and dynamic than the original recordings. The combination of these make for a flat and unenjoyable music experience.
It is a mistake to think playing music loud will make it more dynamic or detailed. If the data is missing detail (and therefore dynamic) no matter how good your equipment is, it will output only what it is input into it.
This means there are at least four factors which will determine how good the music will sound. These are 1) the recording equipment; 2) the amount of data preserved; 3) the processing unit or amp; and 4) the speakers.
I have a recording of Rachmaninoff playing his own Piano Concerto No. 2 but because it was recorded in the early 20th century the amount of detail is low. Similarly, the standard digitally streamed music is of a lower sample rate as well as bitrate. Both of these impact on what can be reproduced. Today while I have a decent stereo in the form of the Kenwood K-515 it still produces, in the eyes of an audiophile, well under the kind of sound that can be considered optimal or perfect. But then again, I am not spending thousands but hundreds of dollars here. For the price I paid that is good enough for me.
Buying a DSD recording of Norah Jones’s Don’t Know Why and playing that on the stereo made it so wonderful that I must say that it was value for money.