[Sonnova]

I was just watching "Wired Science" on PBS. They just did a "shoot-out"
between digital and analog sound. I'm not going to argue with the result,
because they so screwed-up the way that they explained what they were doing
to the TV audience as to make the entire thing worthless.

First they introduced a recording engineer who's Chicago studio is all
analog. He maintains that analog is better than digital (without defining
WHAT he means by "digital"). Then they interviewed a recording engineer that
thought that digital was better than analog (again without explaining WHAT
kind of digital: 16/44.1, 24/96, DSD, MP3 whatever). Then they thoroughly
confused the issue by interchangeably using the terms MP3 and digital
recording- as if they were one and the same. Then they picked two other
recording engineers and two musicians to listen to a cut from those same
musicians' latest recording. Sometimes they were listening to analog,
sometimes digital, and they held up paddles with the words "digital" and
"analog" written on them, to show whenever they thought they heard a
difference. The cut they played was contiguous with no breaks to indicate
when or if the media had changed (how did they do THAT without editing the
two together onto the same medium??!). In the end, the two musicians chose
correctly 53% of the time, and the two recording engineers chose correctly
55% of the time. In other words, essentially, statistically, no better than
blind chance. The conclusion that the TV show producers came to was that
digital is indistinguishable from analog.

This "test" basically just confuses the issue. They say that they were
testing the widely held belief that analog sounds better than digital. But
what they don't differentiate between is PCM digital CD vs MP3. The
impression that I was left with is that they were saying that an analog
master is statistically indistinguishable from an MP3 digital simply because
they made no effort to differentiate between MP3 and RedBook PCM and never
said what the listening "panel" was actually listening too, or the
circumstances under which the "listening test" was conducted. "Wired
Science"? Bogus science is more like it.