There are some items, like watches, cars, wines and
hi fi audio gear where people with enough money (more money than sense, some
would say) spend inordinate amounts for little or no real gain. A $250,000 Rolls-Royce
is no doubt better constructed than a $25,000 Chevrolet, but is it really 10
times better, especially on a slow daily commute? A $50,000 Patek Philippe watch
certainly has more status than a $50 Timex, but it doesn’t really tell time any
better – certainly not $49,950 better.
And as several researchers have shown, even people who fancy themselves wine
experts often can’t really distinguish $10/bottle wines from $200/bottle wines
in blind tests. In fact in one published
test experts confidently rated two glasses of wine quite differently in a blind
test, although both glasses were in fact poured from the same bottle.
The world of hi
fidelity music has always had roughly three levels – (1) the cheap boom-box level
for casual listening of popular music, (2) a mid-level where a reasonable living-room
system might cost $1000-$3000 including speakers, receiver/amplifier, and cd
player. Above that (3) is a stratospheric
level where people obsess about things like $7000 audio cables (Pear ANJOU Speaker Cables), $200,000 hand-built
amplifiers (Dynaudio Arbiter
monoblocks), $40,000 speaker sets (Waterfall
Audio Niagara Platinum Speakers) and $8000 headphones (Final Audio Muramasa
VIII). I have no doubt that to a very
experienced audio engineer with decades of experience in the studio these may possibly
sound a bit better, but are they really $20,000-$300,000 better for the average
listener?
I am thinking about this right now because my Logitech
Duo wireless music devices just died. They were discontinued several years ago
and support/repair is almost nonexistent, so I had to replace them. In the
process of researching on the web what was available these days (I went with
Sonos devices) I discovered - typical for the web - numerous highly emotional
and acrimonious debates on such subjects as whether it was worth buying expensive
external DACs (digital to analogue converters) for the Sonos Connect, and whether
16-bit digital files (CD quality) were adequate for “real” audiophiles or
whether one had to move up to 24-bit files, which are not supported by the Sonos
system.
Of course many of these people are absolutely convinced
they can hear differences with more expensive equipment (especially expensive equipment
they have just bought) but few have really done blind A/B tests, and if they
did many would no doubt be surprised and dismayed to find out that they really couldn’t
reliably tell which was the output from the more expensive system.
In fact human hearing isn’t really all that
good compared to many other creatures. At our best when we were young we may have heard
a range from about 10 Hz to 20k Hz, and by the time we are old enough to afford
expensive audio equipment, and to have lived with loud city noises, rock bands
and music around us for several decades, our hearing is much worse, especially
at the upper end.
The spirited debate about 16-bit vs 24-bit music
files is largely based on a faulty understanding of the physics involved.
Despite the prevalent myth that more bits somehow gives more “accurate” music,
it really only gives more dynamic range (loud to soft), and 16 bits already encompasses
the entire loud-soft range (up to about 96 db, about the sound standing next to
a loud gas lawn mower.) that the human ear can comfortably listen to. A 24-bit
range just pushes the upper end to about 144db, which is about the ear-damaging
sound level standing right next to a 155mm howitzer when it is fired. For studio work where the digital files are
going to be merged and manipulated, it may well be worth working in 24-bit
formats to avoid introducing rounding errors, but 16-bit final output is more
than enough dynamic range for any meaningful music listening.
What does actually make a difference in the “accuracy”
of digital music is the sampling rate. There is an arcane thing called the “Nyquist-Shannon
sampling theorem” that says that to accurately digitally record a wavelength of
frequency x one needs to sample at least
2x times the frequency plus about 10%. CDs sample at 44,0000 samples per
second, which happens to be 20k Hz times 2 plus 10%, meaning that (not by
accident) standard CDs already sample at
a rate capable of accurately recording up to the maximum 20k Hz frequency that a young person can
hear, and far better than most of us older folks can hear.
The upshot of all of this is that, once again, as
with cars and wines and watches, a lot of people are apparently being sold very
expensive stuff that they really don’t need for their purposes with appeals to “status”
and “more expensive is better”. I
suppose this is a natural feature of any economy. As the old saying goes ‘There's a sucker born every minute”, except
that it is now probably down to every 10 seconds or less.
If I had enabled comments
on this blog I would no doubt be swamped with emotional and outraged flames
from people who have sunk a lot of money into their expensive audio systems (or
cars or watches or wine collections) and desperately need to believe their
money was well spent. As a classical music lover with normal adult ears, I find
my rather modest mid-level system (320kbs MP3 files or internet streaming
stations fed through the Sonos wireless system to a
$500 Denon receiver and $100 AR-6
speakers) more than adequate, and in fact I really can’t hear the difference
between 320kbs files and 160kbs files, or even some 128kbs files, so even my
setup is probably overkill.
Posts
