OK, I admit that yes, there aren't separate sideband carriers for each
frequency, and I do know better. IF you are listening to AM BC on a
high fidelity wideband receiver, the IBOC splatter is 45kHz wide, even
though the bands only extend 30 Khz. But compression and Fourier
transform equations mathematically proving audio response to any given
frequency response glosses over the fact that you are no longer sending
the *whole* waveform, and only plotting points on the curve. Taking
the simple CD audio example, if we sample at a 44 khz rate, we can only
measure a 22 khz sinewave twice per cycle. Is the intended waveform to
be reproduced a sine wave, sawtooth, or what? We can't tell anymore,
because we have not measured enough points on the wavefrom. Did we
measure the wave at maximum peak, somewhere in the middle, or as it was
crossing the zero line? We have to assume if it's audio that it's a
sinewave, but this ignores the non-sinusoidal characteristics of
real-life audio. This is why CDs (assuming the original audio exceeded
22 khz) end up sounding like they all got recorded through one
particular make and model of microphone. A very good microphone, but
one that added its own characteristics to the result. Just the same as
taking a large-format negative photograph which could be blown up to
billboard size with good resolution loses all that resolution when
reproduced in a magazine with an 80, 135 or 175 line screen
reproduction. All those dots, small as they may be, don't add up to as
much information as there was with the billions of little silver halide
grains in the original negative. I was trained as a radio engineer,
and decided NOT to work in the business back in 1982, as I saw this
train wreck coming. Not to mention the removal of variety from
programming, limited playlists, etc. The real reason radio must be
digitized is that radio is an ART, and art is not understood by
accountants. Art always cost money, so it must be turned into a
commodity to turn a profit. Too many people could not grasp the art,
tune the antenna, calculate complex impedances on a Smith chart, etc.
They should have gotten out of the business if they could not
understand RF. The same thing happened with cars. For all the
digitalization of engine controls, what we got was technicians who
plug in a computer to tell them what's wrong. In the 70's and 80's its
equivalent was mechanics who sprayed new parts at a problem until it
went away, perhaps never knowing or caring where the true problem had
been. Before that, you had mechanics who UNDERSTOOD how it all worked,
what made it work, and why it was founded in the laws of physics. If
you have current flowing in the primary circuit, and the points opened,
and the condenser was good, you would have a collapsing magnetic field
within the coil, and the secondary circuit would have a SPARK! It was
necessary for a distributor to point to the right wire, etc for the
engine to run , among other things, but the laws of physics always
worked.
It was comforting to know the parts did what they did inherently,
without a central processor.
I have a lot more faith in the laws of physics than I have in
computers, and analogy of making a silk purse out of a pig's ear is
very apt for IBOC. Noisiest ol' pig's ear I've ever heard. Sure as
heck ain't radio.