I'll use an illustration I've discussed before. My younger son has two Masters in Music, one in performance. I have listened to countless hours of practice by him and also his brother, who plays french horn. My son with music degrees plays trombone. He worked very hard to develop a very nice tone.
Interestingly enough, if you stand beside him, the tone doesn't sound good. It's edgy, almost nasty. However, if you hear him in a concert hall where you are sitting 10 rows back and there's 30 or 40 feet between you and him, the tone is sublime. Karl is smart enough and talented enough to adjust the tone so it is right for the listener in the correct environment and not pleasing to his ear or someone standing nearby.
That's the problem with some modern recordings. I've got a great recording of Anne Sophie Mutter playing encore pieces. The sound has a lot of "resolution", down to being able to hear the bow on the string in some quiet passages. It's similar to what I hear when I play violin, but not like an audience member would hear. I find the super resolution, probably from her being close miked, distracting and unrealistic. It's worse on my systems that are very accurate and revealing.
The bottom line is a prefer a more "laid back" sound to one that is "super resolving". I listen to enjoy the music, not revel in how my system can resolve gnat farts. There is a happy medium between accurate reproduction (super resolution) and laid back reproduction that lacks some nuance.
