> Your theoretical numbers kind for remind me of a sign I once saw. "If you can't
> dazzle them with brilliance. Baffle them with Bull Shit".
What I showed you aren't bullshit numbers, but proven theories regarding
signal processing. If you can represent the full range, the "warmth" of the
audio is irrelevant. An audio signal is merely a sequence of sound
intensities. That's all your ear detects. When you hear a sound, it's
just a membrane moving to varying sound pressures, translating that motion
via a small bone over to another membrane that rests at the opening to your
cochlea. That smaller membrane moves, and pushes a fluid through the sealed
cochlea. Small hairs inside the cochlea then translate those varying motions
into electrochemical signals that travel into your brain for processing. As
long as you generate a sound pressure that is continuous, and a comparison of
the output with technical means (such as an oscilloscope) is the same, the
sound pressures getting passed to the cochlea are exactly the same, regardless
of media.
> Seriously, what it really comes down to is not numbers, but rather what the human eye
> can see and hear.
Exactly. And what I was pointing out is that with a sufficiently high bit
width and sample rate, your output from a digital medium can be absolutely
indistinguishable from an analog signal. I agree that a CD can never sound
as good as a live performance, but I *do* believe that a CD sounds much
better than an LP. Ever listen to Holst's "The Planets" on vinyl? Ever
hear it on CD? Ever hear it live played by the London Symphony? Sure, the
symphony sounds the best. The LP blows. The CD is the closest we can
currently get to being there.
> I included sight because my nephew who was a video major in college once made a point
> about video that has always stuck with me. I realize that video isn't necessarily a
> digital media. But it is an electronic one. As opposed to film which is a "true"
> analog reproduction.
Actually, it's not. Film is actually "digital" in nature. A movie that you
see in the theater is a "sample" of the scene taken every 1/30 (or 1/24,
depending on format and media) of a second. This is no different than
sampling a sound pressure every 1/44100 of a second. Still, you don't hear
folks coming out of the woodwork saying that they're only going to go to see
plays because movies don't reproduce the full quality of the work. That's
because designers, after trying 10 frames per second (back with Charlie
Chaplin and such) found that 30 frames per second was enough to fool the
human eye into believing that it was seeing an accurate reproduction of
reality. Audio engineers just need to find the same thing.
However, what I'm talking about is reproducing the reality of the sound.
Just because the sound coming out of your speakers is the same one that came
from the instrument *doesn't* mean that it sounds any better, however.
I guess that's what the problem is here... when I listen to vinyl, I hear a
certain warmth of audio and depth. However, this warmth and depth was NEVER
in the original performance... it's merely the artifact of the medium. Folks
who maintain that tapes and vinyl give better sound are locking into the
artifacts that you get when you have a medium that doesn't fully reproduce
reality. I guess that's okay, if you're in it for a subjective playback. I
mean, music is indeed a subjective medium in the first place.
Seriously, though... the other day, I found a copy of the Kraftwerk album
Ralf and Florian in CD format in a friend's collection. I listened to it.
It sounded *exactly* the same as the copy I had in vinyl. Why? Because the
CD was mastered from the original vinyl recording. Now, this brings up a key
point. Just because it's on CD (or another digital medium), doesn't mean
that it won't have that warmth and depth that so many audiophiles will claim
is the advantage of vinyl. If you like that sort of warmth, master out to
vinyl, and then record it back onto CD, and you'll have the same effect.
You never "lose" the warmth of a recording when going to CD, you merely never
have that warmth introduced into the recording at all. I suppose that folks
could design a filter to simulate a vinyl mastering, and apply it prior to
the pressing of the CD, and everyone would have the same "warmth" on CD that's
so loved by vinylphiles.
> major. He said know, "because as good as you can make video, all you have to do is
> double the size of the film and you will once again have better quality on film.
You're not comparing the final output. You're talking about one frame at a
time. When you talk about increasing the size of the film, you're talking
about making your "samples" bigger. So, let's say we go with something like
a 64 bit sample? That's the same thing as increasing our film from 35mm to
70mm. You're still only playing back at 24 frames per second (or 44.1kHz, in
the audio analogy), but you can represent more detail in each sample. Let's
keep looking at this video analogy, because it works very well.
Let's say that you're a worm. (You're not, just work with me here.) Your
vision is very simple. You can only discern 2 states, visually. "Light" or
"Dark". If you're watching a film in a movie theater, the film can be
infinitesimally small, because it just needs to relay a very tiny amount of
information. As humans, we've got a retina that can discern a LOT more than
just "light" and "dark". However, we *do* have a finite number of light
sensing cells (rods and cones) on our retinas. That means that if we were
using a digital technique to display the movie, we'd need no more than that
number of pixels to represent a frame. That's because each cell can only
view a certain chunk of the screen, and by having that many "chunks", you'll
be using your retina to its maximum potential. Each cell also has a certain
range of possible values that it can send back to your brain. Your brain is
a digital system... impulses sent along the axons of your neurons are either
"on" or "off"... there's no signal... just a series of pulses. If we make
each pixel have enough possible values such that it fills the entire range
of each retinal cell, then we've got a digital movie that's completely
indistinguishable from reality (although, it *would* take up a lot of space).
Coming back to our audio model, if we can find the range of possible values
for every hair cell in your cochlea, then we can create a wide enough sample
to approximate all possible sound levels. Find a high enough sampling rate
(which is quite possible via theoretical means in signal processing), and
you have a signal which is 100% indistinguishable from reality.
But, as I stated before, it seems that the difference between analog and
digital technologies in the ears of someone like yourself has nothing to
do with how well you can replicate reality, but a subjective analysis of
some unquantifiable artifacts introduced into the recording at the media
conversion stage.
> Enjoy the weekend (I guess it's already Sunday where our list members from down under
> live).
Thanks. =) I will, I'm sure... I'm celebrating my mother's birthday tomorrow,
and getting to spend some well deserved time with my girlfriend after a long
week of work. You have a good weekend too...
I hope that my response this time around didn't seem too much like the last
one, with so many numbers that it seemed that I was pulling the wool over
your eyes... I've had this discussion numerous times before, with folks in
both the music and audio engineering fields, digital signal engineers, and
audiophiles. Academia is a great place to find groups full of folks like
the above, all working in close concert with one another. =)
/Andrew
-----------------------------------------------------------------
To stop getting this list send a message containing just the word
"unsubscribe" to [EMAIL PROTECTED]
