William Sommerwerck <
grizzle...@comcast.net> wrote:
>I think I stated the real problem, which has to be addressed, but no one is
>paying attention to.
>
>People prefer what they're accustomed to. Though early digital had real
>problems, the strongest objection was that it didn't sound like analog. This
>wasn't a major problem for classical engineers and listeners, because fidelity
>was important to them (some of them, anyhow). For genres in which fidelity
>wasn't so important, the removal of pleasing colorations was disturbing.
This is part of the problem, and it has always been part of the problem
since the move from acoustic recording to electrical.
But, it's not _all_ of the problem. Because a couple of things came long
with the digital world.
First of all, some engineers were used to pretty much doing the same thing
over and over again, using the same procedures every time, rather than very
carefully fine-tuning their setup to meet the needs of the music and the
performance. These engineers got into big trouble very fast because when
digital systems came in, the optimal procedures changed. This was not a
huge game-changer, but it took some getting used to and in the process some
well-known big names turned out to have real trouble adapting.
Secondly, a lot of the early digital stuff really did just sound awful.
There are a lot of things that you could blame for the problems, from
bad reconstruction and anti-aliasing filters (or totally missing ones!)
to systems that leaked correlated noise from the digital section into the
analogue grounds. I'm not going to point any fingers here because I was
responsible for some pretty crappy-sounding stuff myself at the time. The
implementation took a few years to come up to the level of the theory.
Thirdly, a lot of people are listening to perceptually-encoded systems
which are severely compromised and then tarring all digital systems with
that same brush. This is like saying that analogue recording is no good
because cassettes sound awful (which arguments I have heard in this very
newsgroup I might add).
>This problem isn't new. In a 1958 issue of "Tape Recording", a writer told of
>his problems in getting engineers to switch from mastering on wax (and other
>recording techniques). They liked what they were familiar with, and could not
>"hear" the improvements.
That's true, and that's always going to be the case.
>Messrs. Burnett, Scholz, Young, etc, will never be convinced of digital's
>greater accuracy, because their recordings don't have to be "accurate".
>Rather, they need to shown how to get the results they want.
There are two things here. First of all there is production, and indeed
production systems may not need to be accurate, their purposes is to get
the sound that the producer wants or is appropriate (which may be an accurate
one, but may not be). Secondly there is distribution, and the distribution
format needs to be one that will present the sound as heard by the producers
most accurately in the listener's home.
The problem in the digital world of 2013 is that we now have spectacularly
accurate production systems but the distribution is now almost uniformly
being done on lossily-compressed files. Everything is upside-down.