Digital versus Analog Recording
The Internet has a long standing debate on whether digital recording, or analog recording, is inherently better. I will not give a primer here on how digital works, other than to say that both digital and analog recording technologies are based on different methods for capturing an approximation of the sound.
I bring up the subject because people will debate endlessly over which makes the
best recordings, analog or digital? And yet, they forget that no matter
how the music got to the CD, once it is on the CD it is digital anyway! And,
we can all point to examples of excellent recordings done either way.
So, what is going on here? If the final result is a great sounding CD,
and the CD is digital, then the difference must be in how much of the original sound
ends up on the CD, and not the method used to get it there. And the key
lies in the fact that any recording method is just an approximation of the
original sound.
I say "approximation", because no technology is able to perfectly capture a
sound. There are limitations inherent in both analog and digital technology.
Analog recording uses magnetic particles imbedded in a plastic tape. The
sound is captured by applying a varying magnetic field as the tape moves past
the record head. The sound modulates the magnetic field, and the particles
on the tape hold the magnetic strength applied at the instant the particle
passes the head.
Digital recording uses an Analog to Digital converter (AD) to approximate the
sound at a given point in time, and store the value as a number. The size
of the number is the "bit depth", often 16 bits or 24 bits. The given
point in time is the frequency, such as 44.1khz used for CDs. So, the
sound is a series of approximations, in the case of a CD, 44100 of them per
second.
In the case of Analog, the best we can do for "bit depth" is to offer an
estimate of at least 32 bits because it is essentially infinite---the magnetic
particles will store any value of the continuously varying signal. The
equivalent of "sample rate" is the size of the particles, and how fast they move
across the record head. If we assume one micron particles (a conservative
estimate), at 15ips tape speed, we are looking at an equivalent sample rate of
around 384k! And I am being quite conservative here. In reality, it
is probably much higher.
Why does this matter. Well the more numbers you have, the more information
you can store! The more information you can store, the closer will be your
approximation to the original sound.
This addresses another topic that is often debated. For example, does 96k
sampling really sound better than 48k? From what we learned above, the
answer is an emphatic "yes"! And the reason is that there is a lot more of
the original sound represented in the the 96k data.
You might be thinking, "Hold on. Didn't you tell us that analog tape is
like 384k sample rate?". And, yes, I did, so, in one respect, analog can
hold a lot more information than current digital systems. However, analog
has its weaknesses too. One is much lower dynamic range, typically around
60 db, compared with 96db for 16bit digital. And this lower
dynamic range limits the amount of information analog can hold despite its high
equivalent sampling rate. The other analog weakness is
non-linearity. Digital is completely linear---what you hear is what you
get. But the magnetic particles on analog tape can over-saturate,
basically stuffing more sound into the upper levels than at lower levels.
This "down-side" is actually used to advantage because it is the property that
gives analog tape its built-in few dbs of compression, a desirable feature in
many cases.
Amazing recordings have been done in analog despite the limitations because
knowledgeable engineers have managed the recording process from start to finish
in such a way as to preserve the quality of the sound.
And, since we know there are amazing recordings done digitally from start to
finish, it is obviously possible in the digital realm to manage the chain of processing to deliver
great sound to the final product.
So which is better? I think that is the wrong question. The question
should be, regardless of which technology is used to convey the sound from
performance to CD, how well was that whole process managed?
It all comes down to data management! Whether you are recording to analog
tape or through digital converters, it boils down to how well you manage to
maintain the integrity of the sound through all the various processing steps.
And, so, engineers spend their life trying
to maximize the amount of information that survives from the studio recording
session to the final CD. That is what counts in obtaining the best
recordings possible.
|
Recording Tips
1) First and foremost, capture a good performance---never think "I can fix it in the mix!"
2) Choose the right mic and the right placement to get the sound you want---not EQ.
3) If you must use EQ, use subtractive EQ! Remember, taking away the low end boosts
the high end when you compensate upward with the level control. I read a great quote recently.
An engineer said, "cut if you want it to sound better, boost if you want it to sound different".
4) Use omni mics. They sound better---more open and natural. They have no proximity
effect, so you can still achieve separation by moving the mic in much closer than you
are used to with cardioid mics.
5) Any cable has resistance and capacitance which negatively affect the signal,
so always use the shortest cable possible for a given situation.
6) Phantom Power is, at best, a necessary evil. The circuitry required to phantom
power microphones, either capacitors or transformers, degrades the low level mic
signal, and introduces noise and ripple from the power supply. If you must use
phantom power, at least use batteries! This one is difficult to achieve because we
seldom have access to the phantom power circuitry. But, if you are looking for that
last bit of improvement, this really does help.
7) All things being equal, a large room sounds better than a small room.
8) The smaller the room, the more you need sound treatment, which usually means absorption (bass traps, blankets, curtains, carpets, rugs, etc.).
9) Record as far from walls as possible, but not equal distance from any two walls (unless you have very good sound treatment).
10) A walk-in closet full of clothes makes a great vocal booth, although almost totally absorptive.
|
Electronics Tips
1) The ideal is a piece of wire with gain.
2) Every part in a circuit, regardless of quality, alters the sound for the worse,
so to get as close as possible to the ideal---strive always to use the fewest
components possible.
3) DC coupling sounds better---capacitors create filtering and phase problems,
avoid them when possible.
|
|
