Yesterday I was listening to music through a newly-installed set of loudspeakers in my desktop system. Some tracks sounded much better than others. While varying technical quality is a given for all recorded music, some recordings are simply sonically superior to others. Why is that?
The first major fidelity-killer is how a recording was "tracked" or initially recorded. Simple, purist, recording and microphone techniques have, to my ears, always yielded the best-sounding recordings, regardless of all other factors. One stereo pair of microphones, placed carefully and recorded directly onto the final format, be it analog or digital, is the easiest way to get a convincing phase-correct recording of a live musical event. As microphones are added it becomes increasingly difficult to get a phase-coherent recording that doesn't confuse the listener's ear-brain.
In a studio environment it is possible to duplicate the live event. It is also possible to add microphones, additional tracks, and effects, which can help or hurt the final sonics. When the musicians, engineers, and producer are all on the same page with similar technical and aesthetic goals a studio recording can be stunningly good. An example would be the album Toy Matinee by Toy Matinee. Producer and band member Pat Leonard cared, and it shows.
One of the most startling attributes of a well-recorded album is that it will sound at least good on almost ANY playback system - in your car, on your ancient transistor radio - it doesn't matter...that Toy Matinee album I mentioned earlier was, in the final mastering stages, monitored using Yamaha NS-10s (with the Kleenex mod) because if it sounded good on those things, it would sound good on anything....
Mastering covers a wide variety of actions that can be taken before a recording is deemed fit for commercial release. On some recordings "mastering" can be as simple as merely matching the relative volume levels between tracks. On other recordings Mastering can involve many processes - harmonic adjustments, time alignments, adding effects, adding limiting, dynamic range alterations, transcoding, and for LPs tweaking the signal so it will fit optimally into the LP's grooves.
Again, depending on the skill, time allowed, intent, and how much anyone cares have significant impacts on the success or failure of the mastering process. Over-processing, as we have all experienced with some pop releases, especially during the period when " the loudness wars" made it imperative the quiet parts of a recording be almost as loud as the loud parts (which were also boosted to the max) so dynamic ranges of as little as 6 dB, were not uncommon and at times seemed to be the standard.
While some of this wretched excess has been reduced, still finding commercial pop recordings with more than 10 dB difference between the loudest and softest parts can be a challenge. But for music to sound anywhere like what it should sound like, it needs dynamic acuity, which is lost when limiting and compression are used poorly.
High resolution...If we think of digital formats in the same way as luggage, it's always better to have a bag that's larger than what you need to put into it. Same with formats - if you have to reduce the amount of musical information or data density to fit into a playback format, you have lowered the final fidelity. Conversely, if you put ten oranges into a large bag, you still have ten oranges when you get home. Same with formats - if you have a 44.1/16 bit recording and you put it into a 192/24 format "bag," you still only have the data from that 44.1/16 recording, but it's in a larger container, now.
So, while having a nice big sample and high bit-rate recording is always better than listening to a truncated lossy one, it does not ensure that the recording contained will be better, only that no data was lost. If the original recording was poorly recorded and mastered a higher-rez version will not improve its sonics in any way...except in a case where a DAC consistently performs with a lower error rate when digital data is upsampled to a higher rate. Some might call such a DAC "broken" while other would call it "superior at higher bit rates."
As to whether an original analog recording can ever be considered "High Definition" or "High Rez" is a contentious issue. Mark Waldrep, principal at AIX Recordings, insists that no analog-sourced recording can ever be true high definition. But some transfer techniques, such as Plangent Processes can produce digital transfers from analog tape that have far lower IM and time distortions than the original and produce sonics that far exceed the original. So, while strictly not high-resolution (by Mark's terms), these kinds of distortion-reducing transfers are certainly a gray area in terms of potential ultimate overall fidelity.
Why It Matters
A lot of verbiage has been expended by PR folks to entice audiophiles and music lovers to embrace high-resolution music. A major focus has been on the idea of exporting the experience that the musicians had in the studio to a listener's home environment. That is a lame way to sell high-resolution, in my humble opinion. The point should be that when recorded music is delivered in a high-resolution envelope it is easier to comprehend. "Kiss the sky" is far less likely to morph into "Kiss that guy" because the artists' musical, emotional, and lyrical intent is clearer, and the listener can understand it better when it is presented in high-resolution.