Yes, the S/N is 40 db higher. I was pointing out that the analog at least does not suffer from sounding like it was reconstituted from chunks, no matter how small. I prefer audio sounding fluid, continuous and smooth. I will always prefer the sample offering the most data available, even with extraneous noises added. My brain does a fine job of signal processing, but then I live in a
metro signal area.
The distortion in weak FM signals is annoying enough that for many people, I'll go along with you that the FM HD is probably better sounding.
As an engineer, it is my job to assess technology and applications.
I have a much harder time accepting something which causes me to wonder where in the airchain the under-biased transistor is.
If you can't give me full 20khz resolution full time, I do not want to hear higher frequencies bursting forth in fits and starts.
Nothing in the real world sounds like that, except music which started out in digital to begin with.
Do you prefer a high quality silver photograph, or a medium resolution digital ink-jet photo of a large format Ansel Adams print?
Remember, the digital has no degradation, but the analog has 10,000 times more data.
Would you make the same choice if you were publicly displaying the print?
Would you present the cheaper, easier choice, or pay the money for a full continuous gray -scale silver print?
I argue that quick and easy digital could be good enough, were we willing to move enough data.
If we're not ready to move that much data, back off until it's at least equal to the previous tech.
No one is served by relaxing standards and telling the customers to get used to the degraded result.
And to those who can't hear the degradation, have you ever tried to watch a movie where the projector wasn't focused, and everybody in the theatre starts hollering?
Back in the days of human projectionists you had to hire someone who could tell the difference and properly focus the projector.
Once we get to point that everyone thinks radio should sound like a keychain pocket memo recorder, this won't be a problem.
But right now, too many people know radio need not sound gravelly, zingy, swirly or zizzly.
Such by-products only occur as transients in overtones in normally produced audio, and in those instances, they define sharpness
and high energy events. Having these these strange harmonics added to instrument voices which normally have no such harmonics
is NOT high fidelity.
32k and 48k streams are adequate to convey the data in a 300-3000 hz landline because there isn't anything that can happen any faster that they can describe it. Once you start trying to feed high frequency data to a slow stream, the lack of resolution is immediately evident to those who can tell what they're listening to. The only way to make it acceptable is to limit the high-frequency decoded result back down to a level where the sinewaves aren't audibly stair-stepped.
Speaking of shortwave, I far prefer the BBC (at least what it used to sound like) on SW to their net feed.
Same reason.